Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Greater than the sum of its parts : issues in the diagnosis and management of individuals co-infected… Braitstein, Paula 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-994430.pdf [ 11.36MB ]
Metadata
JSON: 831-1.0099819.json
JSON-LD: 831-1.0099819-ld.json
RDF/XML (Pretty): 831-1.0099819-rdf.xml
RDF/JSON: 831-1.0099819-rdf.json
Turtle: 831-1.0099819-turtle.txt
N-Triples: 831-1.0099819-rdf-ntriples.txt
Original Record: 831-1.0099819-source.json
Full Text
831-1.0099819-fulltext.txt
Citation
831-1.0099819.ris

Full Text

GREATER THAN THE S U M OF ITS PARTS: ISSUES IN THE DIAGNOSIS AND M A N A G E M E N T OF INDIVIDUALS CO-INFECTED WITH H I V AND HEPATITIS C INITIATING ANTIRETROVIRAL THERAPY  by P A U L A BRAITSTEIN B.A. McGill University, 1991 M.A. Simon Fraser University, 1998 M.Sc. University of British Columbia, 2001  A THESIS SUBMITTED I N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Health Care and Epidemiology^ We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A January, 2005  © Paula Braitstein, 2005  ABSTRACT  Objectives:  To determine the prevalence of hepatitis C co-infection among previously  antiretroviral naive HIV-infected individuals initiating antiretroviral treatment (ART) i n a population-based program; to describe the effect of H C V co-infection on immunologic response to A R T ; adherence to ART; and the effect of H C V co-infection on n o n accidental mortality among this population. Methods:  British Columbia's H I V / A I D S D r u g Treatment Program (DTP) provides  antiretroviral therapy to all eligible H I V positive persons i n British Columbia free of charge. Data were d r a w n from a nested cohort w i t h i n the DTP of previously ART naive individuals whose first ever ART was a triple-combination and w h o began treatment between August 1996 - July 2000. Retrospective H C V testing was performed on stored plasma samples from before and, i n some cases, after ART initiation. Results:  Of the 1136 individuals whose stored baseline plasma samples were tested,  606  (51%) tested H C V antibody-positive (Ab+) and 580 (49%) were H C V antibody-negative. A t baseline, 179 (30%) of the antibody positive results were H C V R N A negative. Of these, 118 samples were re-tested for H C V R N A on a sample taken 6-12 months postART initiation; 24 (20%) were n o w found to be H C V R N A positive. I n this study population, people w i t h positive H C V antibodies have an attenuated absolute CD4 response to ART, but a preserved CD4 fraction response; increased biochemical markers of hepatic i n j u r y , and are independently less likely to adhere to antiretroviral medication, after controlling for biochemical markers of hepatic injury and injection drug use. Finally, H C V antibody positive adults are nearly three times more likely to die a non-accidental death, after controlling for age, gender, injection d r u g use, and adherence to ART, and are twice as likely to die an HIV-related death. Conclusions:  Due to shared routes of transmission and the infectiousness of the  hepatitis C virus, there is a very high prevalence of H C V i n this population-based cohort of HIV-infected individuals. H C V adversely affects the effectiveness of ART, including immunologic response and ability to adhere. This has significant implications for the survival of H I V / H C V co-infected patients.  ii  TABLE OF CONTENTS  Abstract  ii  Table of Contents  iii  List of Tables  vi  List of Figures  viiii  Acknowledgements  xi  Dedication  xii  Chapter 1: Introduction  1  1.1 H I V / H C V Co-infection Epidemiology  1  1.2 Current Issues  2  1.3 Outstanding Questions and Study Justification  6  1.4 Study Objectives and Thesis Organization  7  1.5 Summary  10  1.6 References  11  Chapter 2: Background  23  2.1 Forward: "Special Considerations i n the Initiation and Management of Antiretroviral Therapy i n H I V / H C V Co-infected Populations"  23  2.2 Introduction  24  2.3 Virologic and Immunologic Response to ART  26  2.4 Safety of Antiretroviral Agents i n H I V / H C V Co-infection  28  2.5 Immune Suppression, Immune Restoration, and Hepatic Injury  33  2.6 Conclusions and Recommendations  38  2.7 References  41  .'  Chapter 3 : Study Setting, Overview of Methods, and Limitations  61  3.1 Study Setting  61  3.2 Study Population  62  3.3 Other Sources of Data  63  i. Laboratory Data  63  ii. Mortality Data  64  iii. Adherence Measure  64  3.4 Statistical Methods  65  3.5 Strengths and Limitations  66  3.6 Summary  69  3.7 References  70 73  Chapter 4: H C V Diagnostic Issues in HIV Co-Infection 4.1 Forward: "The Effect of Antiretroviral Therapy on the Detectability of H C V R N A Before and After Initiating H I V Treatment i n a Population-Based Cohort of HIV-infected Individuals"  73  4.2 Introduction  74  4.3 Methods  76  4.4 Results  79  4.5 Discussion and Conclusions  81  4.6 References  86  Chapter 5: Response to Antiretroviral Therapy Among H I V / H C V Co-infected  94  5.1 Forward:"Attenuated Absolute but Preserved CD4 Fraction Response to Antiretroviral Therapy Among H I V / H C V Co-infected A d u l t s "  94  5.2 Introduction  95  5.3 Methods  97  5.4 Results  101  5.5 Discussion and Conclusions  103  5.6 References  107  Chapter 6: Safety and Tolerability of Antiretroviral Therapy in H I V / H C V CoInfected Individuals  121  6.1 Forward: "Adherence to Antiretroviral Therapy and H C V - H I V Coinfection"  121  6.2 Introduction  122  6.3 Methods  123  6.4 Results  128  6.5 Discussion and Conclusions  131  6.6 References  134  C h a p t e r 7: M o r t a l i t y a n d H I V / H C V C o - I n f e c t i o n  144  7.1 Forward: "Non-Accidental M o r t a l i t y and Causes of Death A m o n g Previously A n t i r e t r o v i r a l Naive H I V / H C V Co-infected v s . H I V Monoinfected Adults i n a Population-Based H I V Treatment Program  144  7.2 Introduction  145  7.3 Methods  145  7.4 Results  146  7.5 Discussion and Conclusions  147  7.6 References  150  C h a p t e r 8: S u m m a r y , R e c o m m e n d a t i o n s ,  Further Research, and Conclusions  156  8.1 Summary of Findings  156  8.2 Unique Contribution, Impact, and Implications  160  8.3 Recommendations  164  8.4 Future Research  167  8.5 Conclusions  169  8.6 References  171  A p p e n d i x 1: S t a t e m e n t o f A u t h o r s h i p  176  A p p e n d i x 2: C I H R P r o t o c o l  177  A p p e n d i x 3: C e r t i f i c a t e s o f E t h i c a l A p p r o v a l  193  Appendix 3.1 H C V Testing  194  Appendix 3.2 Linkage w i t h Providence Laboratory  196  A p p e n d i x 4: E x e c u t i v e  S u m m a r y : R o a d m a p f o r I m p r o v i n g A c c e s s to C a r e a n d  Treatment for P e o p l e L i v i n g w i t h H I V a n d Hepatitis C Co-infection  197  v  LIST OF TABLES Table 1.1 Prevalence of H I V / H C V Co-Infection f r o m Selected Studies and Populations  Table 1.2  20  Summary of Results of 3 Recent Randomized Clinical Trials of  Pegylated-Interferon + Ribavirin Treatment in H I V / H C V C-o-infected Patients  22  Table 2.1 Summary of Evidence For and Against Controversial Hypotheses  53  Table 2.2 S u m m a r y  of available literature  regarding hepatotoxicity  of  antiretroviral agents i n H I V / H C V co-infected individuals  56  Table 2.3 Special Considerations and their rationale regarding the initiation and management of antiretroviral therapy i n H I V / H C V co-infected individuals  58  Table 4.1 Univariate and Multivariate Factors Associated w i t h Baseline H C V Antibody Prevalence  90  Table 4.2 Univariate and Multivariate Factors Associated w i t h Baseline H C V Antibody Positive and H C V R N A Discordance at Baseline  91  Table 4.3 Factors Associated w i t h H C V R N A Emergence Following Antiretroviral Initiation  Table 5.1.1 Unadjusted and adjusted factors associated w i t h an absolute CD4  92  increase of at least 75 c e l l s / m m  3  111  Table 5.1.2 Unadjusted and adjusted factors associated w i t h an CD4% increase of at least 10%  112  Table 5.3 M i x e d Effects Model Results of Impact of H C V Co-infection on CD4 increase Over 48 Weeks of ART i n Both Cohorts, Overall and A m o n g Those >95% Adherent  113  Table 5.4 Absolute and Fraction CD4 Responses to Initiation of Antiretroviral Therapy A m o n g a Population-Based Program of HCV-positive and HCV-negative Individuals, Stratified by Baseline CD4 Count  114  Table 6.1 Characteristics i n First Year of Therapy A m o n g HCV-Positive and HCVNegative I n d i v i d u a l s i n the British Columbia H I V / A I D S D r u g Treatment Program.  138  Table 6.2 Factors Associated w i t h Adherence to Antiretroviral Medication  139  Table 6.3.1 Logistic Regression Analysis of Factors Associated w i t h Adherence to Antiretrovirals (continuous liver injury markers)  140  Table 6.3.2 Logistic Regression Analysis of Factors Associated w i t h Adherence to Antiretrovirals (dichotomous liver injury markers)  141  Table 6.4.1 Median Results of Biochemical Markers of Liver Injury by the Four Possible Categories of H C V and I D U status  Table 6.4.2  142  Logistic Regression Analysis of Factors Associated w i t h Adherence  W i t h Four-Level H C V / I D U Variable  143  Table 6.4.3 Logistic Regression Analysis of Factors Associated w i t h Adherence W i t h Four-Level H C V / I D U Variable, Adjusted for Liver Injury  143  Table 7.1 Unadjusted and Adjusted Cox Models of Predictors of Non-Accidental Mortality  153  Table 7.2. Causes of Death A m o n g Previously Treatment Naive H I V Mono- and H I V / H C V Co-Infected I n d i v i d u a l s i n a Population-Based H I V Treatment Program, Accidental Deaths Included  154  LIST OF FIGURES Figure 2.1 Overlap Between the M o r b i d i t y Associated w i t h Hepatitis C and Toxicity Associated w i t h Antiretrovirals  60  Figure 4.1 Overview of Study Design and Results  93  Figure 5.1.A Kaplan-Meier Analysis of the Effect of Hepatitis C on time to Absolute CD4 Increase of >75 cells/mm  115  3  Figure 5.1.B Kaplan-Meier Analysis of the Effect of Hepatitis C on time to Absolute CD4 Increase of >75 c e l l s / m m A m o n g Individuals >95% Adherent to 3  Antiretrovirals  115  Figure 5.1.C Kaplan-Meier Analysis of the Effect of Hepatitis C on time to CD4 Fraction Increase of >10%  116  Figure 5.1.D Kaplan-Meier Analysis of the Effect of Hepatitis C on time to CD4 Fraction Increase of >10% A m o n g Individuals >95% Adherent to Antiretrovirals. .116  Figure 5.2.A Adjusted Average Number of Absolute CD4 cells Gained Over 48 Weeks of Antiretroviral Treatment  117  Figure 5.2.B Adjusted Average Number of Absolute CD4 cells Gained Over 48 Weeks of A n t i r e t r o v i r a l Treatment A m o n g I n d i v i d u a l s Antiretrovirals  >95% Adherent to 118  F i g u r e 5.2.C  Adjusted Average Increase of CD4 Fraction Gained Over 48 Weeks of  Antiretroviral Treatment  F i g u r e 5.2.D  119  Adjusted Average Increase of CD4 Fraction Gained Over 48 Weeks of  Antiretroviral Treatment Among Individuals >95% Adherent to Antiretrovirals. ..120  Figure  7.1.A.  Kaplan-Meier product l i m i t estimates of cumulative Survival  Following Initiation of Antiretroviral Therapy by H C V Serostatus  Figure  7.1.B.  155  Kaplan-Meier product l i m i t estimates of cumulative Survival  Following Initiation of Antiretroviral Therapy A m o n g Individuals >95% Adherent to Treatment by H C V Serostatus  155  ACKNOWLEDGEMENTS  First, to Tante Rachel, for putting her life on the line so that I could have one. Second, to my mother and father, for giving me opportunities they never had.  I thank the members, staff, and volunteers of the British Columbia Persons w i t h AIDS Society who have inspired and taught me for so many years. I am indebted to the staff of the H I V / A I D S D r u g Treatment Program, the St. Paul's AIDS Program, the UBC V i r o l o g y Laboratory, the BC Center for Excellence i n H I V / A I D S , the V i r o l o g y Lab there, and the clinic staff of the John Ruedy Immunodeficiency Clinic for their support, both technical and moral. I n particular, I must acknowledge Benita Yip, whose patience and expertise has been invaluable to the timely completion of this project.  I especially thank M y r n a Reginaldo, Liz Ferris,  Jennifer Adachi, Thinh Le, Peter Vann, Kelly Hsu, Val Montessori, Anita Palepu, Chris Sherlock, Linda Merrick, Brian Wynhoven, and Richard Harrigan for their support. A n honourable m e n t i o n must go to Melanie Rusch w h o w i l l soon be w r i t i n g an acknowledgements page of her own. I thank m y committee members, Dr. Bob H o g g , Dr. Julio Montaner, Dr. M a r t i n Schechter, and Dr. Michael V. O'Shaughnessy, for seeing both the scientist and the activist i n me, and one way or another, fostering both.  To Laurel Slaney i n the  Department of Health Care and Epidemiology at UBC, thank-you, for all your moral and material support. Funding of this degree was very generously provided through Doctoral Fellowships by the Michael Smith Foundation for Health Research and the Canadian Institutes of Health Research, an operating grant by CIHR, and salary support from the BC Persons w i t h AIDS Society and the BC Center for Excellence i n H I V / A I D S .  I am profoundly grateful to m y friends, who support me and help me keep it all i n perspective. I especially thank Glen Hillson, for, among other things, allowing me to learn from his unfortunate example. A n d to A l i Karatas, for being there i n the end.  xi  DEDICATION  This dissertation is dedicated to the memory of Glen Edward Hillson, my friend, brother, and teacher.  Of course.  CHAPTER 1: INTRODUCTION  1.1  T H E EPIDEMIOLOGY OF H I V AND HEPATITIS C CO-INFECTION  The Acquired Immune Deficiency Syndrome (AIDS) was first recognized i n 1981 and the H u m a n Immunodeficiency Virus (HIV) first isolated i n 1983. While H I V occurred i n isolated individuals much earlier, even as long ago as the late 1950's, it d i d not emerge i n epidemic proportions u n t i l the 1980's due to a complex interaction of social, geopolitical, and biological factors [1]. N o w , however, the disease is present i n over 200 countries world-wide and is continuing to spread. Since its recognition, over 20 million people have died f r o m AIDS. In 2003, approximately five million people became newly infected w i t h H I V and three m i l l i o n died f r o m AIDS. Over 42 million people are currently l i v i n g w i t h H I V . The H I V epidemic disproportionately affects developing countries, women, youth, and marginalized populations such as aboriginal people, those w h o are incarcerated, and the poor [2]. Since the approval of the first drugs for treatment of H I V infection i n 1987 by the US Food and D r u g A d m i n i s t r a t i o n (FDA), over twenty different drugs have been approved for use. I n 1996, the use of triple-combination ' H i g h l y Active Antiretroviral Therapy' ( H A ART) heralded a new era i n H I V / A I D S . Since H A A R T ' s inception, mortality and morbidity due to human immunodeficiency virus (HIV) have declined dramatically [3, 4]. It is estimated that as much as 3% of the w o r l d ' s population is chronically infected w i t h the hepatitis C virus (HCV) [5], including an estimated 2% of people from the United States [6], and 1 % of Canadians [7]. Chronic hepatitis C is a major cause of chronic liver disease and death throughout the w o r l d [8]. As survival has increased for 1  patients w i t h H I V infection w i t h access to H A A R T , liver disease has emerged as a leading cause of morbidity and mortality among them [9-11]. Numerous published studies have established that the prevalence of H C V coinfection i n HIV-infected populations is very high (see Table 1.1) [12-29]. It is estimated that on average, approximately 30% of HIV-positive people i n high-income countries are co-infected w i t h H C V [30, 31]. However, because i n some populations including injection d r u g users, prison inmates, and hemophiliacs, the prevalence of H C V is so widespread, the prevalence of H I V / H C V co-infection ranges from 50-99% [13,14, 21]. H C V is prevalent i n 1-5% of the w o r l d population, w i t h Africa having the highest prevalence at 5% [32]. I n a meta-analysis of 25 African cohorts, the relative risk for H C V infection for persons w i t h H I V was 1.52 [22]. Reports from a diverse array of countries including Greece [20], Thailand [19], India [33], Russia [34], and China [14], all indicate that H I V / H C V co-infection is both common and entrenched, particularly where the primary H I V risk factor i n co-infected patients is injection drug use or blood transfusion.  1.2 CURRENT ISSUES  Treatment of Chronic Hepatitis C Infection The current standard treatment for H C V infected patients is pegylated interferon (PEG-IFN) combined w i t h ribavirin, a nucleoside analogue. The goal of H C V treatment is to induce viral clearance and reduce or reverse the progression of hepatic fibrosis [8]. Even if viral clearance cannot be achieved, interferon therapy has been shown to reverse or significantly slow the progression of fibrosis and limit the potential for development of hepatocellular carcinoma [35, 36].  2  I n H C V mono-infected patients; sustained viral response (SVR) rates of up to 88% have been achieved w i t h the use of pegylated interferon i n combination w i t h ribavirin [37]. SVR is defined as the absence of H C V - R N A in peripheral blood 24 weeks after the termination of therapy. Despite the high prevalence of H I V - H C V co-infection, the safety and efficacy of peg-interferon combination treatments designed for mono-infected H C V patients have only recently been evaluated for use w i t h co-infected patients i n randomized clinical trials (RCT) (Table 1.2). Results from three multicentre RCT's involving co-infected patients have recently shown that the efficacy of pegylated interferon is decreased to approximately 40% overall, in comparison to its efficacy i n mono-infected patients, approximately 55% overall, although no head-to-head studies have been performed [8, 37, 38]. H C V genotype 1 significantly reduces the probability of a successful treatment outcome [8], and accounts for 70-75% of all H C V infections in the United States [39] and Canada [40]. Genotype 1 is also predominant in H I V / H C V co-infected populations [41, 42]. H C V treatment effectiveness in HIV-positive individuals may also be limited by the prohibitory cost of g r o w t h factors (used to enable continued full-dose ribavirin, a significant predictor of treatment success [38], and treatment guidelines which require people to abandon treatment by week twelve if their H C V R N A has not declined by a m i n i m u m of 2 logs (in spite of evidence of altered H C V viral dynamics post-treatment in the presence of H I V [43, 44]). Given these limitations, H C V treatment can be expected to have a favorable impact on HCV-related morbidity and mortality in perhaps 20% of H I V / H C V co-infected individuals [31,41,45].  3  Influence of H C V on HIV disease progression The question of the impact of H C V on the natural history of H I V disease remains of considerable interest. Prior to the availability of combination antiretroviral therapy (pre-1996), increased rates of H I V disease progression largely obscured any effects of H C V disease on H I V [46]. Since then, Greub et al. (2000) reported from the Swiss Cohort Study that H C V infection was associated w i t h a 70% increased probability i n progressing to a new AIDS-defining event or death [47]. Some have confirmed his findings since [48-50], while others refute them [46, 51].  There are many ways that  H C V could impact H I V disease progression that may not be captured b y current studies. For example, liver disease may preclude the use of antiretroviral therapy because of hepatotoxicity. Many co-infected individuals, because of sociodemographics or uncontrolled addiction, simply never access antiretroviral therapy and die of AIDS [52]. It is also possible that H C V has a direct influence on H I V infection, although the mechanism of action is as yet unknown [30].  Influence of HIV on H C V disease progression Numerous authors have demonstrated that co-infection w i t h H I V increases the relative risk of developing cirrhosis and that the rate of hepatic fibrosis is more rapid i n H I V co-infected patients [31, 53-56]. I n a study comparing rates of fibrosis progression in H I V / H C V co-infected patients and H C V mono-infected patients, Benhamou et al. (1999) found higher fibrosis scores, higher necro-inflammatory activity and more rapid progression to cirrhosis i n co-infected patients [53]. T w o large studies comparing disease progression i n co-infected and mono-infected patients showed that cirrhosis developed i n 15-25% H C V / H I V co-infected patients w i t h i n 10-15 years compared w i t h 3-6% of those patients w h o were H I V negative [57, 58]. A meta-analysis by Graham et al. (2001) examined the results of 8 different cohort studies and concluded that there is a  two to three-fold increase i n the rate of fibrosis progression i n H I V / H C V infected patients [59]. Other predictors of fibrosis progression include age over 35 years, alcohol consumption of over 50 g / d a y , and a CD4+ T cell count of <500 cells/mm3 [31].  Immune Suppression In 1999, the US Public Health Service/Infectious Diseases Society of American amended their HIV-guidelines to include H C V as an opportunistic pathogen among HIV-positive people. I n spite of this, the mechanisms or determinants involved i n the increased severity of H C V i n co-infected persons are not w e l l understood. I t is n o w evident that the impact of H I V on the course of H C V related liver disease is particularly marked as CD4 cell counts decline and immunodeficiency increases. A l o w baseline CD4 cell count is correlated w i t h increased progression of fibrosis, higher incidence of cirrhosis and hepatocellular carcinoma among H I V / H C V co-infected persons [53, 60, 61]. Typically a CD4 count less than 200 c e l l s / m m is considered to be " l o w " ; however, 3  two independent studies have found the severity of liver fibrosis i n chronic hepatitis C infection to be independently associated w i t h a CD4 count lower than 500 cells/mm , 3  after controlling for H I V infection, age, duration of H C V infection, and alcohol abuse [55, 62]. Although the mechanism(s) for the increased pathogenicity of H C V i n H I V coinfected patients are not entirely clear, i t is generally accepted that H I V and hepatitis C both contribute to dysregulation of the cellular immune response, which is itself the mechanism that causes fibrosis [38, 63]. Thomas (2002) has suggested that H I V selectively infects the CD4+ lymphocytes that have been activated against the H C V infection - thus reducing the cellular response to H C V infection and leading to the persistence of H C V i n hepatocytes [64].  5  1.3 OUTSTANDING QUESTIONS AND STUDY JUSTIFICATION  A l t h o u g h there are numerous investigations i n the literature regarding the prevalence of H I V / H C V co-infection, there is no evidence i n the literature of a population-based sample having been examined. Further, there is only very limited data about the prevalence of H I V / H C V co-infection in British Columbia [13, 29, 65]; all of this data is d r a w n from the same cohort of injection drug users. There are also numerous unresolved questions regarding the accurate diagnosis of H C V in H I V co-infected individuals. For example, there are case reports of both spontanenous H C V R N A clearance [66, 67] and spontaneous H C V R N A emergence upon treatment initiation [68]. There are references i n the literature to HIV-positive individuals testing false-antibody negative for H C V [69, 70], and while 15-25% of H C V mono-infected individuals will, spontaneously clear H C V R N A without treatment, this issue has not been well characterized in HIV-infected populations [71, 72]. Of the 1388 individuals w h o first initiated antiretroviral treatment in British Columbia between August 1996 and July 2000, and whose first ever ART was a triplecombination therapy, at the outset of this project only 538 people (39%) had any documented H C V serology. Those for w h o m we had data were more likely to be male (41% vs. 30%, p=0.002), and were more likely to have an HIV-experienced physician (42% vs. 30%, p<0.001) [73]. It was also confirmed through a preliminary diagnostic study that approximately 5% of HIV-positive individuals w h o have no antibodies to H C V w i l l have H C V R N A present [73]. Thus, a study regarding diagnostic issues in HIV-co-infected patients initiating antiretroviral therapy was conceived of, funded (by a CIHR Operating Grant, $93,000), and implemented (Appendix 2) as part of this doctoral project.  6  The majority of people co-infected w i t h H I V / H C V w i l l never fully benefit from current H C V therapy, pegylated interferon combined w i t h ribavirin.  I n order to  maximize the effectiveness of ART, and to mitigate H C V disease progression, tailored H I V and antiretroviral management is criticial. Outstanding antiretroviral-related questions of concern to clinicians and patients i n the setting of H I V / H C V co-infection include when to start antiretroviral therapy, and what to start w i t h . Central to the antiretroviral management of H I V / H C V co-infected patients is being able to predict the immunologic response to ART, and immunologic response to ART i n H I V / H C V coinfected patients is not w e l l understood. The safety and tolerability of antiretroviral agents i n H I V / H C V co-infected individuals has not been f u l l y characterized, and requires further study. Adherence to medication i n A R T treated patients is a critical determinant of outcome, and factors associated w i t h non-adherence i n H I V / H C V coinfected populations need greater elucidation. The impact of H C V disease on mortality among individuals i n the era of highly active antiretroviral therapy also needs further study.  1.4 STUDY OBJECTIVES AND THESIS ORGANIZATION  The overall a i m of this thesis is to address some of the outstanding questions regarding H I V / H C V co-infected individuals initiating antiretroviral therapy from a population health perspective. This thesis w i l l explore 4 primary objectives:  O b j e c t i v e 1: T o m e a s u r e t h e b a s e l i n e p r e v a l e n c e o f H e p a t i t i s C i n f e c t i o n a m o n g H I V positive individuals initiating H i g h l y Active Antiretroviral T h e r a p y ( H A A R T ) p o p u l a t i o n - b a s e d H I V treatment cohort.  I n Chapter  4,  in a  I describe the results of the  retrospective testing of stored baseline samples from the H O M E R Cohort. I describe the  7  baseline prevalence of hepatitis C co-infection among this population using antibodydata, and sociodemographic characteristics (age, gender, median income) associated w i t h testing antibody positive for HCV.  I n addition, I describe the prevalence of  discordant responses, including h o w sociodemographics and immune status are associated w i t h antibody and R N A discordant responses. M y primary hypotheses were 1) that there w o u l d be a Hepatitis C prevalence of 30% among this population (defined as having positive antibodies to HCV) at the time of ART initiation; 2) that at baseline, 10-15% of those w i t h positive H C V antibodies w o u l d have no evidence of H C V R N A ; and 3) that i n 25% of these cases, H C V R N A w o u l d emerge 6-12 months following ART initiation.  Objective 2: To characterize the impact of the hepatitis C virus (HCV) on CD4 response to antiretroviral treatment (ART) over the first 48 weeks of therapy in previously ART-nai've HIV-infected individuals. Specifically, i n this chapter I examine i m m u n o l o g i c response to A R T using b o t h absolute and CD4 fraction responses. I n addition, I describe the role of baseline absolute CD4 count on these responses. M y primary hypotheses were that HCV-positive individuals w o u l d have a smaller CD4 recovery over the first 48 weeks of ART using both absolute and fraction outcomes, and that this blunted response w o u l d be most pronounced among those initiating treatment w i t h <200 CD4 cells/mm . 3  Objective 3: To describe the effect of hepatitis C (HCV) co-infection on adherence to antiretroviral treatment (ART) during the first year of therapy in a population-based HIV/AIDS drug treatment program. The primary hypothesis was that while H C V coinfected individuals w o u l d be less likely to adhere to their ART, there w o u l d be a strong and significant interaction between H C V seropositivity and any biochemical 8  marker of liver injury, indicating that increased hepatic injury i n HCV-co-infected patients w o u l d be a significant factor i n this non-adherence. I also sought to elucidate the relative effects of hepatitis C infection and any history of injection drug use on adherence.  O b j e c t i v e 4: T o d e s c r i b e t h e e f f e c t o f H C V s e r o s t a t u s o n t h e r i s k o f n o n - a c c i d e n t a l mortality i n a p o p u l a t i o n - b a s e d treatment p r o g r a m of p r e v i o u s l y A R T naive patients, and  to d e s c r i b e c a u s e s o f d e a t h a m o n g t h i s p o p u l a t i o n .  People w h o are co-infected  w i t h H I V and H C V can have multiple high risk factors for mortality, including drug use (associated w i t h accidents, overdoses, suicides), H I V disease, or end-stage liver disease secondary to H C V infection. This chapter sought to describe whether H C V serostatus was an independent predictor of non-accidental mortality, and to describe causes of death i n this population. Chapter 7's primary hypothesis was that people w h o are co-infected w i t h H C V w o u l d have an independently higher risk of nonaccidental mortality.  This thesis is d i v i d e d into 8 chapters. This first chapter provides some preliminary background and context, and outlines the study's justification and objectives. Chapter 2 is a review paper currently i n press w i t h the journal AIDS that provides a summary and analysis of the research to date i n the area of antiretroviral use in the context of H I V / H C V co-infection. Chapter 3 provides an overview of the study setting, the methods employed, and the limitations of m y approach.  Chapters 4  through 7 are based o n four research papers that are currently, or w i l l be shortly, submitted for publication. Finally, Chapter 8 provides a discussion of the findings,  9  outlines the unique contribution of the work, makes recommendations, and outlines areas for future research.  1.5 SUMMARY  In summary, H I V and hepatitis C co-infection is a relatively recent phenomenon w i t h wide-ranging clinical and public health implications. Although great strides have been made i n the clinical management of H I V disease, major challenges have emerged, including co-morbidities such as hepatitis C. Very high H C V prevalence in high H I V risk communities, faster H C V disease progression, poorer response to H C V treatment, complications related to H I V management, plus the frequent complexities of uncontrolled addictions, poverty, and psychiatric issues, all combine to make H I V / H C V co-infection a very complex domain for clinicians and other care providers. As H I V treatment and management is becoming increasingly accessible to communities most at risk of H I V - H C V co-infection, the issue of hepatitis C is causing H I V clinics and health care providers to integrate H C V issues into their practices.  This dissertation  addresses several key issues related to the H I V management of patients who are coinfected w i t h H I V and H C V , and has implications for patients, physicians, and policy makers.  1 0  1.6 REFERENCES  1.  Garrett, L., Hatari: Vinidogodogo (Danger: A Very Little Thing), i n The Coming Plague: Newly Emerging Diseases in a World Out Of Balance, L. Garrett, Editor. 1994, Farrar, Straus and Giroux: U.S. p. 281-389.  2.  UNAIDS., 2004 Report on the Global AIDS Epidemic. 2004, United Nations: Geneva.  3.  Hogg, R., et al., Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA, 1998. 279(6): p. 450-4.  4.  Palella, F.J., Jr., et al., Declining morbidity and mortality among patients ivith advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N e w England Journal of Medicine., 1998. 338(13): p. 853-60.  5.  Lavanchy, D., Global surveillance and control of hepatitis C. Journal of V i r a l Hepatitis, 1999. 6: p. 35-47.  6.  A n o n y m o u s , National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C 2002 (June 10-12, 2002). Gastroenterology., 2002.123(6): p. 2082-99.  7.  Health Canada., Epidemiology of the transmission of blood-borne pathogens. 1997, Laboratory Centre for Disease Control.  8.  Poynard, T., et al., Viral hepatitis C. Lancet, 2003. 362(9401): p. 2095-100.  9.  Soriano, V., et al., Impact of chronic liver disease due to hepatitis viruses as cause of hospital admission and death in HIV-infected  drug users. European Journal of  Epidemiology., 1999.15(1): p. 1-4. 10.  Martin-Carbonero, L., et al., Increasing impact of chronic viral hepatitis on hospital admissions and mortality among HIV-infected patients. AIDS Research & H u m a n Retroviruses., 2001.17(16): p. 1467-71.  1 1  11.  Bica, I., et al., Increasing Mortality Due to End-Stage Liver Disease in Patients with Human Immunodeficiency Virus Infection. Clinical Infectious Diseases (CID), 2001. 32(Feb 1): p. 492-497.  12.  Staples, CJ., D. Rimland, and D. Dudas, Hepatitis C in the HIV Atlanta Veterans' Affairs Cohort Study. Clinical Infectious Diseases, 2000. 29(1): p. 409-10.  13.  Patrick, D.M., et a l , Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001.165(7): p. 889-95.  14.  Zhang, C , et al., High prevalence of HIV-1 and hepatitis C virus coinfection among injection drug users in the southeastern region of Yunnan, China. J Acquir Immune Defic Syndr, 2002. 29(2): p. 191-6.  15.  Sherman, K.E., et al., Hepatitis C Virus prevalence among patients infected with Human Immunodeficiency  Virus: a cross-sectional analysis of the US adult AIDS  Clinical Trials Group. Clinical Infectious Diseases, 2002. 34(6): p. 831-7. 16.  Sulkowski, M.S. and D.L. Thomas, Hepatitis C in the HIV-infected patient. Clin Liver Dis, 2003. 7(1): p. 179-94.  17.  A m i n , J., et al., HIV and hepatitis C coinfection within the CAESAR study. H I V Med, 2004. 5(3): p. 174-9.  18.  Bowker, S.L., et al., Human immunodeficiency virus (HIV) and hepatitis C virus (HCV) in a Northern Alberta population. Can J Public Health, 2004. 95(3): p. 188-92.  19.  Chanbancherd, P., et al., High frequency ofHIV-1 and hepatitis C co-infection among young Thai men: evidence for a changing pattern of HIV transmission in Thailand. Southeast Asian J Trop M e d Public Health, 2003. 34(3): p. 580-2.  20.  Dimitrakopoulos, A., et al., The prevalence of hepatitis B and C in HIV-positive Greek patients: relationship to survival of deceased AIDS patients. J Infect, 2000. 40(2): p. 127-31.  1  2  21.  Filippini, P., et al., Hepatitis viruses and HIV infection in the Naples area. Infez Med, 2003.11(3): p. 139-45.  22.  Gisselquist, D., L. Perrin, and S.F. M i n k i n , Parallel and overlapping HIV and bloodborne hepatitis epidemics in Africa. Int J STD AIDS, 2004.15(3): p. 145-52.  23.  Hammer, G.P., et al., Low incidence and prevalence of hepatitis C virus infection among sexually active non-intravenous drug-using adults, San Francisco, 1997-2000. Sex Transm Dis, 2003. 30(12): p. 919-24.  24.  Pallas, J.R., et al., Coinfections by HIV, hepatitis B and hepatitis C in imprisoned injecting drug users. Eur J Epidemiol, 1999.15(8): p. 699-704.  25.  Segurado, A . C . , et al., Hepatitis C virus coinfection in a cohort of HIV-infected individuals from Santos, Brazil: seroprevalence and associated factors. AIDS Patient Care STDS, 2004.18(3): p. 135-43.  26.  Staples, C.T., Jr., D. Rimland, and D. Dudas, Hepatitis C in the HIV (human immunodeficiency virus) Atlanta V.A. (Veterans Affairs Medical Center) Cohort Study (HAVACS):  27.  the effect of coinfection on survival. Clin Infect Dis, 1999. 29(1): p. 150-4.  Copeland, L., et al., Changing patterns in causes of death in a cohort of injecting drug users, 1980-2001. Arch Intern Med, 2004.164(11): p. 1214-20.  28.  Hall, C.S., et al., Hepatitis C virus infection in San Francisco's HIV-infected urban poor. J Gen Intern M e d , 2004.19(4): p. 357-65.  29.  Miller, C.L., et al., The Future Face of Coinfection: Prevalence and Incidence of HIV and Hepatitis C Virus Coinfection Among Young Injection Drug Users. J Acquir Immune Defic Syndr, 2004. 36(2): p. 743-749.  30.  Rockstroh, J.K. and U. Spengler, HIV and hepatitis C virus co-infection. Lancet Infect Dis, 2004. 4(7): p. 437-44.  31.  Soriano, V., et al., Care of patients zuith hepatitis C and HIV co-infection. AIDS, 2004. 18(1): p. 1-12.  32.  World Health Organization., Hepatitis C, i n Fact sheet N°l 64. October 2000.  33.  Bhattacharya, S., et al., Co-infection immunodeficiency  with hepatitis  virus among patients with sexually  C virus  and human  transmitted  diseases in  Pondicherry, South India. Indian J Pathol Microbiol, 2003. 46(3): p. 495-7. 34.  Abdala, N., et al., Estimating the prevalence of syringe-borne and sexually transmitted diseases among injection drug users in St Petersburg, Russia. Int J STD AIDS, 2003. 14(10): p. 697-703.  35.  Nishiguchi, S., et al., Prevention of hepatocellular carcinoma in patients with chronic active hepatitis C and cirrhosis. Lancet, 2001. 357: p. 196-7.  36.  Poynard, T., et al., Impact of pegylated interferon alfa 2b and ribavirin on liver fibrosis in patients with chronic hepatitis C. Gastroenterology, 2002.122(5): p. 1303-1313.  37.  Flamm, S., Chronic Hepatitis C Virus Infection. JAMA, 2003. 289(18): p. 2413-2417.  38.  Torriani, F.J., et al., Peginterferon Alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J M e d , 2004. 351(5): p. 438-50.  39.  A n o n y m o u s , Recommendations from the National Institutes of Health consensus development conference statement: management of hepatitis C: 2002. Hepatology., 2002. 36(5): p. 1039.  40.  Anonymous, Canadian Consensus Conference on the Management of Viral Hepatitis. Canadian Journal of Gastroenterology, 2000.14(Suppl B): p. 5B-20B.  41.  Perez-Olmeda, M., et al., Virological characteristics of hepatitis C virus infection in HIV-infected individuals with chronic hepatitis C: implications for treatment. Aids, 2002.16(3): p. 493-5.  42.  Serfaty, L., et al., Impact of early-untreated HIV infection on chronic hepatitis C in intravenous drug users: a case-control study. AIDS, 2001.15: p. 2011-2016.  14  43.  Ballesteros, A . L . , et al., Early HCV dynamics on Peg-interferon and ribavirin in HIV/HCV  co-infection: indications for the investigation of new treatment approaches.  Aids, 2004.18(1): p. 59-66. 44.  Torriani, F., et al., Hepatitis C Virus (HCV) and Human Immunodeficiency (HIV) Dynamics during HCV Treatment in HCV/HIV  Virus  Coinfection. Journal of  Infectious Diseases, 2003.188(15 Nov): p. 1498-507. 45.  Perez-Olmeda, M . , et al., Pegylated IFN-alpha2b plus ribavirin as therapy for chronic hepatitis C in HIV-infected patients. Aids., 2003.17(7): p. 1023-8.  46.  Sulkowski, M.S., et al., Hepatitis C and progression of HIV disease. Jama, 2002. 288(2): p. 199-206.  47.  Greub, G., et al., Clinical  progression, survival,  and immune recovery  during  antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. 48.  Klein, M., R. Lalonde, and S. Suissa, The Impact of Hepatitis C Virus Coinfection on HIV Progression Before and After Highly Active Antiretroviral Therapy. JAIDS, 2003. 33: p. 365-72.  49.  Daar, E . , et al., Hepatitis C Virus Load is Associated ivith Human Immunodeficiency Virus Type 1 Disease Progression in Hemophiliacs. Journal of Infectious Diseases, 2001.183: p. 589-595.  50.  De Luca, A., et al., Coinfection with hepatitis viruses and outcome of initial antiretroviral regimens in previously naive HIV-infected subjects. Arch Intern M e d , 2002.162(18): p. 2125-32.  51.  L a w , W.P., et al., Impact of viral hepatitis co-infection on response to antiretroviral therapy and HIV disease progression in the HIV-NAT 1169-77.  cohort. Aids, 2004. 18(8): p.  52.  Wood, E., et al., Prevalence and correlates of untreated human immunodeficiency virus type 1 infection among persons who have died in the era of modern antiretroviral therapy. Journal of Infectious Diseases, 2003.188: p. 1164-70.  53.  Benhamou, Y., et al., Liver fibrosis progression in human immunodeficiency virus and hepatitis C virus coinfected patients. The Multivirc Group. Hepatology., 1999. 30(4): p. 1054-8.  54.  Einav, S. and M.J. Koziel, Immunopathogenesis  of hepatitis C virus in the  immunosuppressed host. Transplant Infectious Disease., 2002. 4(2): p. 85-92. 55.  Martin-Carbonero, L., et al., Incidence and predictors of severe liver fibrosis in human immunodeficiency  virus-infected  patients with chronic hepatitis C: a European  collaborative study. Clin Infect Dis, 2004. 38(1): p. 128-33. Epub 2003 Dec 8. 56.  Dieterich, D., Hepatitis C virus and human immunodeficiency virus: clinical issues in coinfection. American Journal of Medicine, 1999.107(6B): p. 79S - 84S.  57.  Sanchez-Quijano, A., et al., Influence of human immunodeficiency virus type 1 infection on the natural course of parenterally acquired hepatitis C. Eur J C l i n Microbial Infect Dis, 1995.14: p. 949-53.  58.  Soto, B., et al., Human immunodeficiency virus infection modifies the natural history of chronic parenterally-acquired hepatitis C with an unusually rapid progression to cirrhosis. J Hepatol, 1997. 26: p. 1-5.  59.  Graham, C , et al., Influence of Human Immundeficiency Virus Infection on the Course of Hepatitis C Virus Infection: A Meta-Analysis. Clinical Infectious Diseases, 2001. 33: p. 562-569.  60.  Rockstroh, J., et al., Immunosuppression may lead to progression of hepatitis C virusassociated liver disease in hemophiliacs coinfected with HIV. American Journal of Gastroenterology, 1996. 91(12).  61.  D i Martino, V., et al., The influence of human immunodeficiency virus coinfection on chronic hepatitis C in injection drug users: a long-term retrospective cohort study. Hepatology, 2001. 34(6): p. 1193-9.  62.  Puoti, M., et al., Liver fibrosis progression is related to CD4 cell depletion in patients coinfected with hepatitis C virus and human immunodeficiency virus. J Infect Dis, 2001. 183(1): p. 134-7.  63.  Lee, W. and D. Dieterich, Challenges in the Management of HIV and Hepatitis C Virus Co-infection. Drugs, 2004. 64(7): p. 693-700.  64.  Thomas, D.L., Hepatitis C and human immunodeficiency virus infection. Hepatology., 2002. 36(5 Suppl 1): p. S201-9.  65.  Strathdee, S., et al., Needle Exchange is not Enough: lessons from the Vancouver Injecting Drug Users Study. AIDS, 1997.11(8): p. F59-65.  66.  Fialaire, P., et al., Sustained disappearance of hepatitis C viremia in patients receiving protease inhibitor treatment for human immunodeficiency virus infection. Journal of Infectious Diseases., 1999.180(2): p. 574-5.  67.  Yokozaki, S., et al., Immunologic dynamics in hemophiliac patients infected with hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. Blood, 2000. 96(13): p. 4293-9.  68.  Cooper, C. and D. Cameron, Interpretation of undetectable hepatitis C virus RNA levels in HIV-hepatitis C virus co-infection. AIDS, 2004.18(2).  69.  Cribier, B., et al., High hepatitis C viraemia and impaired antibody response in patients coinfected with HIV. Aids, 1995. 9(10): p. 1131-6.  70.  George, S.L., et al., Hepatitis C virus viremia in HIV-infected individuals with negative HCV antibody tests. J Acquir Immune Defic Syndr, 2002. 31(2): p. 154-62.  71.  Piasecki, B., et al., Influence of alcohol use, race, and viral coinfections on spontaneous HCV clearance in a US Veteran population. Hepatology, 2004. 40: p. 892-899.  72.  Mehta, S.H., et al., Protection against persistence of hepatitis C. Lancet, 2002. 359(9316): p. 1478-83.  73.  Braitstein, P., et al. Dangerous Oversights: A comparison of HCV-RNA  testing vs.  HCV-antibody testing among HIV-infected individuals, i n Canadian Association for HIV Research. 2001. Toronto, Canada. 74.  Paris, R., et al., The association between hepatitis C virus and HIV-1 in preparatory cohorts for HIV vaccine trials in Thailand. AIDS, 2003.17(9): p. 1363-7.  75.  Puig-Basagoiti, F., et al., Prevalence and route of transmission of infection with a novel DNA virus (TTV), hepatitis C virus, and hepatitis G virus in patients infected with HIV. J Acquir Immune Defic Syndr, 2000. 23(1): p. 89-94.  76.  Lodenyo, PL, et al., Hepatitis B and C virus infections and liver function in AIDS patients at Chris Hani Baragwanath Hospital, Johannesburg. East A f r M e d J, 2000. 77(1): p. 13-5.  77.  Maida, M., et al., Prevalence of hepatitis C infection in Malawi and lack of association with sexually transmitted diseases. Eur J Epidemiol, 2000.16(12): p. 1183-4.  78.  Rouet, F., et al., HBV and HCV prevalence and viraemia in HIV-positive and HTVnegative pregnant women in Abidjan, Cote d'lvoire: the ANRS 1236 study. J M e d Virol, 2004. 74(1): p. 34-40.  79.  Ayele, W., et al., Higher prevalence of anti-HCV  antibodies among HIV-positive  compared to HTV-negative inhabitants of Addis Ababa, Ethiopia. J M e d Virol, 2002. 68(1): p. 12-7. 80.  Chung, R.T., et al., Peginterferon Alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected persons. N Engl J Med, 2004. 351(5): p. 451-9.  81.  Perrone, C , et al. Pinal results of ANRS HC02-RIBAVIC:  a randomized controlled  study of pegylated interferon alpha-2b plus ribavirin vs interferon alpha-lb plus ribavirin 1  8  for the initial treatment of chronic hepatitis C in HIV-infected patients, i n 1 Conference on Retroviruses and Opportunistic Infections.  2004. S a n F r a n c i s c o , C A .  Table 1.1. Prevalence of H C V , HIV, and H I V / H C V Co-infection in Selected Countries Location  China Thailand  4.07 5.60  HIV Prevalence (%) in Country 0.06 0.90  India  1.85  0.47  Brazil  2.60  0.37  Russia Spain Italy  2.00 0.74 0.48  0.60 0.34 0.24  Switzerland  0.24  0.18  Canada  0.15  0.18  Heterosexual sex, M S M , I D U Heterosexual sex, M S M , I D U IDU IDU Heterosexual sex, I D U Foreign citizens, heterosexual sex, IDU, M S M IDU, M S M  USA  1.80  0.32  MSM, IDU  South Africa  1.70  11.7  Heterosexual sex  3.9 [77]  7.30  Heterosexual sex  13% HIV+ male sugar workers (verses 10% HIV-) [77]  Cameroon  12.50  3.44  Heterosexual sex  Ethiopia  0.18  2.07  Heterosexual sex  9.0% HIV+ pregnant women (verses 8.0 % HIVpregnant women) [78] 4.5% HIV+ women in community survey [79]  Malawi  ro G  Main HIV transmission routes  H C V Prevalence in HIV+ Study Groups  HCV Prevalence (%) in Country  IDU IDU  99.3% HIV+ persons [14] 51% HIV+ men identified at screening for potential HIV vaccine trial (verses 8.3% ifHIV-) [74] 49.5% of HIV-infected young M S M and 2.2% among uninfected men [19]. 21.4% HIV+ persons in hospital (verses 3.5% ifHIV-) [33] 36.2% HIV+ persons (84.8% HIV+ with IDU) [25] 91% HIV+ with IDU [34] 42% HIV+ persons [75] 54% HIV+ (80% HIV+ with IDU) [21] 37.2% HIV+ persons on ART (87.7% of cohort with IDU) [47] 62.4% HIV+ cohort (91% of cohort with IDU) [18] 95% HIV+ with IDU [13] 16% HIV+persons (73% HIV+ with IDU or MSM) [15] 1% Hospitalized AIDS persons (2000) [76]  20  ART - antiretroviral therapy I D U - injection drug use M S M - men who have sex w i t h men H I V country prevalence calculated by dividing total number of adults and children w i t h H I V by country population; data from U N A I D S / W H O Global H I V / A I D S Online Database H C V country prevalence from W o r l d Health Organization Weekly Epidemiological Record No. 46,14 November.  2  1  Table 1.2.  Summary of Results of Three Recent Randomized Clinical Trials of  Pegylated Inteferon + Ribavirin Treatment in H I V / H C V Co-infected Patients [38, 80, 81]. APRICOT  A C T G 5071  RIBAVIC  40 81 79 6.7 61 540 16  44 85 47 6.2 74 470 37  40 74 n/a 5.9 58 514 39  A C T G 5071  RIBAVIC  860 289 No Yes 31  133 66 No No 12  412 205 Yes No 42  IFN+RBV  PEG-IFN  PEG-IFN+RBV  APRICOT overall % Genotype 1 % Genotype 2 / 3 %  12 7 20  20 14 36  40 29 62  A C T G 5071 overall % Genotype 1 % Genotype 2 / 3 %  12 6 33  -  27 14 73  RIBAVIC overall % Genotype 1 % Genotype 2 / 3 %  18 5  -  26 11 43  Baseline Parameters Mean age (years) Male gender (%) White race (%) H C V Viral load (log IU) Genotype 1 (%) Mean CD4 cell count (cells/mm ) Bridging fibrosis/cirrhosis (%) 3  Study Characteristics  APRICOT  Total number of patients Number on PEG-IFN + RBV Used RBV dose escalation Used haematologic growth factors Treatment Discontinuation (%) Virologic Response Rates  -  22  CHAPTER 2: BACKGROUND  2.1 F O R W A R D  The literature review contained in this chapter has appeared i n A I D S as: Braitstein P, Palepu A , Dieterich D, Benhamou Y , Montaner J. "Special Considerations in the Initiation and Management of Antiretroviral Therapy A m o n g H I V / H e p a t i t i s C Co-Infected Patients".  A I D S . 2004 Nov 19;18(17):2221-34.  This review was conducted by Paula Braitstein w h o also prepared the first draft.  A  number of experts were then asked to provide feedback and be co-authors on the manuscript.  Two other reviews related to the topic of H I V / H C V co-infection were conducted but are not included in this dissertation. They are:  Braitstein P. " H I V and Orthotopic Liver Transplantation". British Columbia Medical Journal [1].  Braitstein P, D r u m m o n d A , Palepu A , Montessori V . "Treatment and Care Issues for People Co-Infected W i t h H I V and Hepatitis C " in: Framing V i r a l Hepatitis, Raymond Schinazi, Editor (invited, under review).  23  2.2 INTRODUCTION  The use of highly active antiretroviral therapy (HAART) has led to dramatic reductions i n HIV-related morbidity and mortality in HIV-infected individuals [2-4]. However, parallel w i t h this success has been the subsequent emergence of co-morbidities, such as viral hepatitis.  The Hepatitis C Virus (HCV) i n particular, because of its h i g h  prevalence in HIV-positive populations (up to 90% in some [5-8]) and its greatly increased pathogenicity in the setting of H I V [7, 9], is now a leading cause of death of HIV-positive individuals [10-12].  A number of reviews have recently been published regarding the care and treatment of H I V / H C V co-infected persons [7, 9, 13]. These reviews have primarily focussed on the treatment and management of hepatitis C in HIV-co-infected individuals, but have not generally examined issues related to H I V management in these patients. Unfortunately, a large proportion of people co-infected w i t h H I V / H C V w i l l never fully benefit from H C V therapy. Many individuals w i l l not be able to access therapy, either because of contraindications to treatment, such as psychiatric co-morbidity (including depression) or anemia, or because they are otherwise considered ineligible (e.g. are considered 'nonresponders' to previous treatment, continue to use illicit drugs or alcohol, etc.). Of the individuals w h o do access treatment, recent data suggest that the probability of a sustained virologic response (SVR) may be significantly less i n HIV-co-infected individuals (SVR: 40% overall in HIV-positive, vs. 55% in HIV-negative, and among genotype 1, 29% i n HIV-positive, vs. 45% i n HIV-negative [14-16]).  Genotype 1  accounts for 70-75% of all H C V infections in the United States [17] and Canada [18], and is also predominant i n H I V / H C V co-infected populations [19, 20]. H C V treatment effectiveness is limited by a number of factors including ineligibility for treatment, the prohibitory cost of g r o w t h factors to enable continued full-dose ribavirin, treatment 24  guidelines which force people to abandon treatment by week twelve if their H C V R N A has not declined by a m i n i m u m of 2 logs (in spite of evidence of altered H C V viral dynamics post-treatment in the presence of H I V [21, 22]), the predominant prevalence of genotype 1, and the negative influence of H I V infection. Given the above limitations, H C V treatment can be expected to have a favorable impact on HCV-related morbidity and mortality i n perhaps 20% of H I V / H C V co-infected individuals [9, 19, 23]. Therefore, optimum H I V and antiretroviral management are critical both for controlling H I V , and for mitigating HCV.  Key antiretroviral-related questions of concern to clinicians and patients in the setting of H I V / H C V co-infection include the best time to initiate antiretroviral therapy. Factors to consider are CD4 count, H I V viral load, hepatic inflammation and function, and the anticipated consequences of immune restoration in terms of auto-inflammatory responses and elevation of H C V viral loads. Other important questions relate to the safety and tolerability of these agents in H I V / H C V co-infected individuals, including liver enzyme elevations, as well as metabolic and mitochondrial toxicities, which are themselves linked to H C V infection.  The objectives of this review are therefore to summarize the available evidence regarding 1) the impact of hepatitis C on the virologic and immunologic response to antiretroviral therapy, including what is k n o w n regarding treatment interruptions; 2) the safety of antiretroviral agents in co-infected individuals; and 3) the relationship between immune suppression, immune restoration and hepatic injury.  This review involved computerized, English-language literature searches of M E D L I N E and PubMed databases (January 1985 to M a y 2004) for published studies in humans 25  that examined H I V and hepatitis C. Keywords for the search included H I V , AIDS, h u m a n immunodeficiency virus, acquired i m m u n e deficiency syndrome, H C V , hepatitis C, hepatitis, immune deficiency, immune restoration, toxicity, diabetes, mitochondria, and metabolism.  The bibliographies of selected articles were also  searched for pertinent studies.  2.3  IMPACT  OF H C V  ON  THE VIROLOGIC AND  IMMUNOLOGIC  RESPONSE  TO  ANTIRETROVIRAL THERAPY  A number of studies have examined the impact of H C V on the virologic and immunologic response to antiretroviral therapy. While most have found that there is no impact of H C V on the H I V virologic response to antiretroviral therapy [12, 24, 25], there are mixed reports regarding immunologic response. Table 1 summarizes the conflicting evidence.  Greub et al. (2000) defined an immunologic response to treatment as an increase of at least 50 CD4 cells/mm3, and found that H I V / H C V co-infection was associated w i t h a smaller CD4 recovery [12]. Zala et al (2004) found that while 86% of HCV-negative individuals had a CD4 increase of at least 75 cells/mm3 at 48 weeks, only 64% of HCVpositives d i d .  Further, w h i l e it took a median of 17 weeks for HCV-negative  individuals to achieve an increase of at least 75 cells/mm3, HCV-positive individuals took a median of 29 weeks [26]. I n a population-based cohort of previously treatmentnaive individuals, these data were confirmed using mixed effects models, showing that while HCV-negative individuals gained an average of 36 cells per year after adjustment for confounders, H I V / H C V co-infected individuals lost on average 5 cells over a 12 month period [27]. Although Klein et al. (2003) found no difference in the mean CD4 26  cell count 24 months post-initiation of antiretroviral therapy, the HCV-positive individuals had a significantly reduced probability of achieving a CD4 increase of at least 50 cells/mm3 (Hazard Ratio 0.48, 95% confidence interval 0.23-0.97, p=0.04), after adjustment for baseline CD4, viral load, previous nucleoside experience, and duration of H I V infection [24]. A number of reports have found no difference in CD4 increase by H C V serostatus [24, 25, 28, 29].  The lack of consistency i n results regarding immunologic response among H I V / H C V co-infected persons on H A A R T may be due to a number of factors. It can take up to 24 months of treatment for a complete CD4 response to occur [30], beyond the time frame of many studies. H o w a CD4 increase is defined is also another key factor (e.g. 'time to' CD4 increase of 50 cells vs. 100 cells vs. using all intra-subject CD4 measures), and most of the reports which found no difference in CD4 response d i d not clearly define what actually constituted a CD4 response. A third issue may be that a blunted response may mostly occur in those H I V / H C V co-infected individuals w i t h lower baseline CD4 counts. For example, in Klein's study where no difference in mean CD4 response was found, the baseline CD4 was w e l l above 200 among both HCV-positive and H C V negative subjects [24], whereas the baseline HCV-positive CD4 count i n Greub's landmark study was 172 cells [12].  The lack of consistency may also due to  inconsistencies in measures of, or accounting for, adherence to ART.  Antiretroviral treatment interruptions are frequent among persons co-infected w i t h H I V / H C V , and injection d r u g use has been identified as an important predictor of treatment interruptions among H I V / H C V co-infected persons [12, 31, 32]. However, the majority of injection d r u g users are co-infected w i t h hepatitis C. Therefore, it is difficult to k n o w whether the reduced levels of adherence and increased rates of 27  treatment discontinuation seen in this population are due to lifestyle issues commonly associated w i t h injection drug use, or whether they are related to the increased toxicity experienced by many co-infected patients.  Only a few studies have examined H I V / H C V co-infection as an independent factor in switching or interrupting H A A R T [25, 33, 34]. M e l v i n et al. reported antiretroviral discontinuation rates due to hepatic toxicity of more than two-fold in HCV-co-infected individuals compared to H I V mono-infected persons [25]. A m o n g a population of 465 previously antiretroviral naive individuals, H C V seropositivity was associated w i t h an adjusted 40% increased risk of discontinuing or changing initial H A A R T regimens w i t h i n the first year of treatment [33]. The authors indicate that whether this was due to histological damage, reduced adherence, or increased hepatotoxicity is not clear, because hepatic cirrhosis was also i n d e p e n d e n t l y  associated w i t h  HAART  discontinuation (AOR 2.1, 95% CI: 1.1-3.8) [33]. Aceti (2002) reported that 83% of those who discontinued antiretrovirals because of hepatotoxicity were H C V co-infected [28], whereas D ' A r m i n i o Monforte et al. (2000) found no impact of H C V on treatment interruptions due to toxicity [34].  2.4 SAFETY OF ANTIRETROVIRAL AGENTS IN H I V / H C V CO-INFECTED INDIVIDUALS  Antiretroviral agents are associated w i t h a diverse array of short- and long-term toxicities, ranging in severity from benign to life-threatening [30]. There have been a number of reviews published addressing hepatotoxicity [35-37], b u t few  that  specifically focus on the role of H I V / H C V co-infection [9, 37]. Although hepatotoxicity is most commonly defined as an increase in liver enzymes, there are several other types of liver-related toxicities which are both associated w i t h antiretroviral therapy, and 28  independently the result of hepatitis C virus infection (see Figure 1). These include mitochondrial dysfunction, including peripheral neuropathy, metabolic changes such as insulin resistance and diabetes mellitus, and hepatic steatosis. Thus the potential for overlapping pathology is significant among co-infected i n d i v i d u a l s  receiving  antiretrovirals, and while these interactions have not been well characterized, some data are available and are described below.  Elevated Liver Enzymes Hepatitis C i n the absence of H I V infection independently causes elevated liver enzymes [16, 38]. There is an emerging consensus that H C V infection increases the risk of developing elevated liver enzymes by 2-3 fold i n HIV-co-infected individuals receiving antiretrovirals [39-42], and H C V accounts for the majority (over 75%) of severe hepatotoxicity in most studies (albeit that the prevalence of H C V co-infection i n a geographical area w i l l have an impact on this figure) [40, 42-47] (see Table 2). H I V / H C V co-infected individuals may also be more likely to experience the clinical symptoms associated w i t h highly elevated liver enzymes, such as malaise, asthenia, nausea/vomiting, fever, jaundice, and decompensation of previous ascites [41]. Nunez (2001) found that while most study participants who developed liver enzyme elevations experienced pure cytolysis (68/222), of the 6 participants w h o developed cytolysis and cholestasis, all but one were HCV-positive [47].  It is notable that others have found little or no association between elevated liver enzymes and hepatitis C among those co-infected w i t h H I V [48-50]. A meta-analysis of this question w o u l d be helpful. Some of the relatively l o w rates of severely elevated A L T may be due to small numbers [50], or selection bias [29]. Recently, Becker (2004) compared liver toxicity across the A T H E N A (Amsterdam), Collaborations i n H I V 29  Outcomes Research US (CHORUS), Italian Cohort of Naive for Antiretrovirals (ICONA), and Target Cohorts [35]. In spite of differences in the prevalence of viral hepatitis and definitions of hepatotoxicity across the cohorts, the authors concluded that there was a low overall prevalence of hepatotoxicity in these cohorts (<6.5%), but that in each cohort the risk of developing hepatotoxicity was associated w i t h viral hepatitis, and in 3 of the 4 cohorts, also w i t h an elevated baseline A L T . Although there was no consistent association found between hepatotoxicity and a particular drug or drug class, ritonavir and the recent use of nevirapine (during the first 12 weeks) were significantly associated w i t h hepatotoxicity [35].  Furthermore, female sex was independently  associated w i t h hepatotoxicity in the A T H E N A cohort [35, 51], while age over 60 years was a factor in the Target cohort [35]. Also, H C V genotype 3 has been specifically associated w i t h the development of acute liver enzyme elevations in H I V / H C V coinfected populations [52, 53]. Genotype 3 is also strongly predictive of hepatic steatosis [54, 55], underscoring the importance of k n o w i n g patients' H C V genotype prior to initiating antiretroviral treatment.  H I V / H C V co-infected individuals are more susceptible to the hepatotoxicity associated w i t h certain drugs, particularly nevirapine [30, 35, 40, 41] and full dose ritonavir [28, 35, 48, 56]. A n important and as yet unanswered question is whether small doses of ritonavir produce less hepatotoxicity than full-dose ritonavir, and whether even low doses should be avoided in co-infected individuals. I n a small, retrospective analysis of HCV-co-infected participants enrolled i n the Abbott 863 trial comparing nelfinavir to lopinavir/ ritonavir, there was a statistically significant increase in mean A L T levels i n the nelfinavir, but not the l o p i n a v i r / r i t o n a v i r , arm at 24 weeks. By week 48, both groups mean A L T level was back to baseline [57]. I n contrast, Aceti (2002) found that saquinavir boosted w i t h ritonavir, but not saquinavir alone, was associated w i t h both 30  overall and severe hepatotoxicity i n co-infected patients [28]. A recent Canadian study also found that independent predictors of grade 3/4 elevations i n A L T among 202 HBV and/or  H C V co-infected  individuals  were  older  age and current  use of  lopinavir/ritonavir [58].  Mitochondrial Toxicity The pathogenesis of H C V is not fully understood [16], but is believed to be at least i n part due to intra-hepatic mitochondrial damage [59-61]. Mitochondrial toxicity has also been w i d e l y associated w i t h nucleoside analog use, particularly stavudine (d4T), didanosine (ddl), zidovudine (AZT) and the combination of ribavirin plus d d l [37, 6264].  Inhibition of mitochondrial D N A synthesis can lead to hyperlactatemia, lactic acidosis and even death [65-68]. It is now well established that using d d l or d4T simultaneously w i t h r i b a v i r i n is contraindicated because of the risk of fatal lactic acidosis [9]. Independent risk factors for the development of symptomatic lactic acidosis are H C V or HBV co-infection, being female, liver disease, pregnancy, and obesity [37].  Mitochondrial damage is also associated w i t h fatty liver [37, 69], and is k n o w n to greatly accelerate fibrosis [37, 70], particularly among people infected w i t h H C V genotype 3 [54, 55].  Although not life-threatening, peripheral neuropathy is a painful and often progressive symptom w i t h substantial impact on quality of life. This is a well-documented side effect of d4T, d d l , and ddC [30], and is attributed to nucleoside-induced mitochondrial toxicity [71]. What is less k n o w n is that peripheral neuropathy is independently related 31  to H C V infection, possibly through its effects on mitochondria [72-74]. H C V infection is often associated w i t h cryoglobulinemia and peripheral neuropathy is a comparatively common complication of cryoglobulinaemia associated w i t h H C V infection and is thought to be attributable to nerve ischaemia [74]. H C V R N A has been detected in nerve tissue suggesting a possible direct role of H C V [72]. A nerve conduction study showed abnormal findings in 77% of investigated HCV-positive patients [73].  Metabolic Changes Hepatic steatosis is linked to insulin resistance, which itself can be caused by both H I V protease inhibitors and H C V infection [37, 65, 70, 75, 76]. Duong et al. (2001) found more insulin resistance in a group of 29 H I V / H C V co-infected people compared to 76 H I V mono-infected controls, but had rates comparable to a group of 121 H C V monoinfected controls [77]. Of note, all the HIV-infected persons were on H A A R T [77].  Similarly, H C V is linked w i t h the development of type 2 diabetes [78-80]. I n a large retrospective cohort of over 40,000 US HIV-positive veterans, H C V co-infected people had a nearly two-fold increased risk of developing diabetes mellitus, controlling for age, race, and history of drug/alcohol use [81]. Mehta et al. (2003) found that both H C V coinfection and protease inhibitor use were independently associated w i t h developing hyperglycemia among 1230 individuals initiating antiretroviral therapy, adjusting for age and baseline glucose level [82].  Liver fat content may be higher in those w i t h lipodystrophy, although the relationship to H C V infection has not been examined [75]. There is some evidence to suggest that coinfected patients may be more vulnerable to the symptoms associated  with  lipodystrophy syndrome, including lipoatrophy [71]. For example, Duong et al. (2001) 32  observed more frequent lipoatrophy in H I V / H C V co-infected patients (41% vs. 14%, p=0.003), and found that peripheral fat wasting was independently associated w i t h H I V / H C V co-infection [77].  Surprisingly, a number of reports indicate that H I V / H C V co-infected individuals have more favorable profiles in terms of total cholesterol, low-density lipoprotein cholesterol (LDL) and triglyceride plasma levels, compared to H I V mono-infected people [77, 83, 84]).  The u n d e r l y i n g mechanism remains u n k n o w n but may represent impaired  synthesis of cholesterol in the liver [84].  I n summary, there are several potential adverse interactions between antiretroviral toxicities and hepatitis C infection. More research is needed to understand these interactions; to k n o w whether their combined effects are additive or synergistic; to improve our understanding of the pathogenesis of H C V infection and the mechanisms of d r u g toxicities; and to optimize the clinical management of co-infected patients receiving antiretrovirals.  More research is also needed to determine the safety and  tolerability of specific antiretrovirals including low doses of ritonavir in H I V / H C V coinfected individuals, using either existing observational data or through clinical trials.  2.5 IMMUNE SUPPRESSION, IMMUNE RESTORATION, AND HEPATIC INJURY  The relationship between antiretroviral use, immune suppression, immune restoration, and hepatic injury is complex and not well characterized.  W i t h or without antiretroviral use, low baseline CD4 count is strongly associated w i t h progression of fibrosis, cirrhosis, and hepatocellular carcinoma among H I V / H C V co33  infected persons [85-88]. While most of these studies consider a ' l o w ' CD4 count to be below 200 cells/mm3, Puoti et al. (2001) found that the severity of liver fibrosis in chronic hepatitis C infection was independently associated w i t h a CD4 count of less than 500 cells/mm3 at time of biopsy, after controlling for H I V infection, age, duration of H C V infection, and alcohol abuse [89]. There are virtually no studies that contradict the finding that immunodeficiency is associated w i t h the progression of hepatic disease.  Perhaps because of their role in preserving functional immunity, a number of studies have found a favorable impact of antiretrovirals in H I V / H C V co-infected persons in terms of fibrosis/cirrhosis progression and liver-related mortality. Not using protease inhibitor therapy was strongly associated w i t h faster progression to cirrhosis among 182 H I V / H C V co-infected patients [90]. It is important to note that these findings may be limited by a possible selection bias whereby those individuals w i t h more advanced liver disease may have been less likely to have been prescribed protease inhibitors because of their k n o w n hepatotoxicity. [90]. Qurishi et al. (2003) concluded that antiretroviral use significantly reduced long-term liver-related mortality among 285 H I V / H C V coinfected patients [91]. However, although the investigators showed decreasing rates of liver-related mortality between 1990 and 2002, their study population consisted predominantly of hemophiliacs infected during the early to mid-1980's. Thus it is likely that the majority of patients i n their cohort died of other causes (e.g. their blood disorders, H I V infection, etc.) before dying of liver disease, since effective antiretroviral therapy only became available in the m i d to late 1990's. A third study [92] concluded that antiretrovirals d i d not influence the progression of fibrosis, but the study was cross-sectional, small (n=42), and based on only one biopsy, the timing of which was not clear.  34  Individuals who are more immunocompromised when they initiate H A A R T are at risk for experiencing Immune Reconstitution and Inflammatory Syndrome (IRIS) [30, 93]. In 1999, the US Public Health Service/Infectious Diseases Society of American amended their HIV-guidelines to include H C V as an opportunistic pathogen among HIV-positive people. Chronic H C V is not the consequence of the direct destruction of hepatic cells by the virus. Rather, it results from an intermediate immune response that is large enough to induce hepatic cell destruction and fibrosis but not enough to eradicate the virus from its reservoirs [15]. It is believed that hepatic injury resulting from immune recovery is the result of a similar mechanism [94]. As the immune system recovers, the cytotoxic T-cell response becomes more efficient at lysing infected hepatocytes, releasing H C V R N A from the hepatocytes killed and increasing plasma levels of H C V R N A [94]. Increased lysis of infected hepatocytes upon immune reconstitution may also cause an increase i n A L T levels [28]. The increased H C V replication then further damages liver cells through apoptosis and other means such as cytokine disruption [89, 95]. Hepatic fibrosis itself is a w o u n d healing response to ongoing liver injury [89], and H C V is considered an immune-mediated disease [96].  The potential for immune restoration to enhance liver disease progression is supported by data regarding both increased viral replication, and fibrosis progression. Perhaps because of immune dysregulation, H I V / H C V co-infected individuals have higher baseline H C V viral loads compared to H C V mono-infected individuals [8, 20, 97, 98] and, w i t h one exception [50], antiretroviral initiation has been found to significantly enhance H C V replication [29,46, 89,94,99-101].  Numerous studies [29, 57, 98] have found a relationship between baseline immune function, as measured by CD4 counts, and the magnitude of the increase in H C V RNA. 35  I n patients starting f r o m a more immunocompromised state (baseline CD4 <350 cells/mm3)  ;  H C V R N A levels increased w i t h i n the first 16 weeks  following  antiretroviral therapy initiation and remained high throughout the 48 weeks of therapy followed. I n individuals w i t h baseline CD4's >350 cells, significant increases in H C V R N A were found w i t h i n the first 16 weeks, but these increases were transient and the H C V viral load returned to baseline levels w i t h i n the following 16 week period [29]. Sherman (2004) found that only patients w i t h baseline CD4 count below 100 cells/mm3 had significant increases in their H C V R N A (>0.5 log) at week 24 and 48 [57].  There are several case series and reports i n the literature regarding the impact of antiretroviral-related immune reconstitution on the development of hepatic fibrosis, cirrhosis, and end-stage liver disease [99, 101-104]. Although this is a difficult question to study because of the need for repeat liver biopsies in patients prior to and following antiretroviral initiation, these reports suggest significant potential for severe hepatic deterioration parallel to an immune response among those individuals w h o are more immunocompromised on treatment initiation [102, 104]. Benhamou (2001) found that a baseline CD4 count of less than 200 cells/mm3 was independently associated w i t h a nearly three-fold increase i n the risk of developing cirrhosis following initiation of antiretroviral therapy [105].  Elevated A L T / A S T levels are common in H I V / H C V co-infected individuals, and may be related to the increased destruction of infected hepatocytes that accompanies immune restoration. Sulkowski et al. found that a CD4 increase of more than 50 cells was associated w i t h severe hepatotoxicity (defined as a grade 3 or 4 change in A S T / A L T ) [40, 106].  However, others have f o u n d only weak or no associations  between baseline CD4 counts, CD4 recovery, and the development of elevated liver 36  enzymes, in spite of high levels of the latter among co-infected patients and evidence of other types of immune reconstitution syndrome [28,47,94].  While immune recovery may have deleterious consequences i n many cases, it is noteworthy that there have been two case reports of spontaneous clearance of H C V R N A following H A A R T initiation, presumably related to immune reconstitution [8, 107]. Younger age at time of H A A R T initiation (<20 years of age) may improve the possibility of this outcome [8].  In spite of the widespread use of H A A R T , H C V is a leading cause of death among HIVinfected individuals [10-12].  The use of antiretroviral therapy and the subsequent  preservation of functional i m m u n i t y may have protective effects in preventing or slowing the progression of hepatic injury, especially among those individuals w h o initiate treatment while their immune system is still intact. However, the data suggest that immune restoration due to antiretroviral therapy among those individuals w i t h immunodeficiency may have a deleterious impact on the progression of H C V disease. The issue has not been w e l l studied. However, current data indicate that those coinfected individuals who are not eligible for H C V treatment might consider initiating H A A R T at higher CD4 levels than their H I V mono-infected counterparts to mitigate the potential immune reconstitutive effects of antiretroviral therapy. Treating H C V early in both the H I V and H C V disease process may have the dual beneficial effect of maximizing the chance of achieving a sustained H I V virologic response [9, 14], and maximize the benefits of future antiretroviral therapy. More prospective research is needed to q u a n t i f y the impact of antiretroviral-related immune restoration on progressive histologic hepatic i n j u r y , and to elucidate the  immune-mediated  components of the two viruses combined. As there is also evidence that antiretroviral 37  use is associated w i t h H C V R N A replication in previously H C V R N A undetectable individuals, and the development of H C V antibodies following antiretroviral initiation [108], patients should be tested for the presence of H C V R N A and antibodies before and after initiating treatment [102,108].  2.6 CONCLUSIONS & RECOMMENDATIONS  Currently available data convey two key messages related specifically to the use of antiretroviral therapy in H I V / H C V co-infected individuals: 1) that both immune suppression and restoration can contribute to the onset and acceleration of HCV-related liver disease; and 2) that the morbidity associated w i t h H C V infection, such as insulin resistance, diabetes, mitochondrial dysfunction, and liver inflammation are all also associated toxicities of antiretroviral therapy w h i c h together may be at least additive. Whether the apparent aggravation of liver injury and H C V disease progression among those co-infected is due to immune restoration, antiretroviral-related hepatotoxicity, or enhanced H C V replication is not clear, although it is likely due to interactions between the three. Specific recommendations based on the available evidence for the initiation and management of antiretroviral therapy in H I V / H C V co-infected individuals are summarized in Table 3.  Hepatitis C treatment has evolved to a degree where sustained virologic clearance is possible for some patients who are eligible for treatment, including up to 70% of those HIV-co-infected w i t h genotypes 2 / 3 , but fewer than 30% of those H I V co-infected w i t h genotype 1, the dominant genotype among H I V / H C V co-infected populations [14,16]. Thus while H C V treatment should be considered for most H I V / H C V co-infected individuals, only a relatively small proportion of them can expect to eliminate their 38  H C V infection. If H C V treatment is not an option, then H I V treatment becomes central to the health management of H I V / H C V co-infected patients.  Unfortunately,  insufficient research has thus far been conducted to indicate when the most appropriate time is to initiate ART in dually infected patients. Similarly, the relative effectiveness of various available regimens in H I V / H C V co-infected patients is not fully established.  The data regarding baseline CD4, CD4 response, and progression of hepatic injury, especially liver fibrosis/cirrhosis are intriguing. Clearly this is an area that requires more prospective research to confirm the data to date, and to elucidate the mechanisms by which immune deterioration and restoration are linked to the progression of fibrosis and cirrhosis.  For example, if the pathogenetic mechanism of hepatitis C is  immunologic cytotoxicity, w h y do people w i t h the lowest i m m u n i t y have the worst hepatic disease? The 2002 International AIDS Society treatment guidelines indicate that H A A R T treatment in people w i t h CD4 counts over 200 c e l l s / m m 3 should be individualized, considering the rate of decline of the CD4 count, H I V viral load, patient interest and potential to adhere, and i n d i v i d u a l risks of toxicity and d r u g - d r u g interactions [30]. The data presented in this review suggest that H C V status should also be considered a factor in that decision. This may be particularly important in men, and in those w h o acquired H C V at a later age, since both are important predictors of liver fibrosis progression in HIV-co-infected individuals [20, 90,109].  There is significant potential for overlap between HCV-attributable morbidity and antiretroviral-associated toxicity, and the clinical research to-date suggests this is resulting i n elevated rates of mitochondrial and metabolic adverse events.  More  research is needed to understand whether these interactions are additive or synergistic, to elucidate the pathogenesis of H C V infection, and to develop better clinical 39  monitoring and responses to these emerging issues. Further research is also needed to clarify the role that this increased m o r b i d i t y has on antiretroviral treatment discontinuation, and its subsequent impact on mortality.  I n conclusion, there are numerous burgeoning and maturing epidemics of H I V / H C V co-infection around the w o r l d , and if epidemiological co-infection projections are accurate, both H I V and H C V diseases can be expected to progress, in the absence of treatment, w i t h i n seven to fifteen years f o l l o w i n g infection [110].  A significant  proportion of these individuals are current drug users and H C V treatment may never be an option for many. H I V treatment, however, is becoming ever more accessible to drug-using populations and others and may be the only treatment option available. This review suggests that managing the H I V aspects of H I V / H C V co-infection is a complex therapeutic domain, and one in which the ultimate outcome for patients is held in the balance.  40  2.7 REFERENCES  1.  Braitstein,  P.,  HIV infection and liver transplantation: A review of the literature.  B r i t i s h C o l u m b i a M e d i c a l J o u r n a l , 2004. 46(1): p . 14-20. 2.  H o g g , R . , et a l . ,  Improved survival among HIV-infected patients after initiation of  triple-drug antiretroviral regimens. C M A J :  Canadian Medical Association Journal,  1999.160(5): p . 659-65. 3.  H o g g , R . , et a l . ,  Improved survival among HIV-infected individuals following initiation  of antiretroviral therapy. 4.  P a l e l l a , F . J . , J r . , et a l . ,  J A M A , 1998. 279(6): p . 450-4.  Declining morbidity and mortality among patients with advanced  human immunodeficiency virus infection. HIV Outpatient Study Investigators. N e w E n g l a n d J o u r n a l o f M e d i c i n e . , 1998. 338(13): p . 8 5 3 - 6 0 . 5.  H a l l , C , et a l . ,  Hepatitis C Virus Infection in San Francisco's HIV-infected Urban  Poor. High Prevalence but Low Treatment Rates. J o u r n a l  of G e n e r a l  Internal  M e d i c i n e , 2 0 0 4 . 1 9 ( 4 ) : p . 357-365. 6.  P a t r i c k , D . M . , et a l . ,  Incidence of hepatitis C virus infection among injection drug users  during an outbreak of HIV infection. C m a j , 7.  S o r i a n o , V . , et a l . ,  2 0 0 1 . 1 6 5 ( 7 ) : p . 889-95.  Care of patients with chronic hepatitis C and HIV co-infection:  recommendations from the HIV-HCV International Panel. A i d s , 8.  Y o k o z a k i , S., et a l . ,  2 0 0 2 . 1 6 ( 6 ) : p . 813-28.  Immunologic dynamics in hemophiliac patients infected with  hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. B l o o d , 2 0 0 0 . 96(13): p . 4293-9. 9.  S o r i a n o , V . , et a l . ,  Care of patients with hepatitis C and HIV co-infection. A I D S ,  2004.  18(1): p . 1-12. 10.  C a c o u b , P . , et a l . , Mortality  among human immunodeficiency virus-infected patients  with cirrhosis or hepatocellular carcinoma due to hepatitis C virus in French departments 4 1  of internal medicine/infectious  diseases, in 1995 and 1997. Clinical Infectious  Diseases, 2001. 32: p. 1207-1214. 11.  Bica, I., et al., Increasing Mortality Due to End-Stage Liver Disease in Patients with Human Immunodeficiency Virus Infection. Clinical Infectious Diseases (CID), 2001. 32(Feb 1): p. 492-497.  12.  Greub, G., et al., Clinical  progression, survival,  and immune recovery  during  antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. 13.  Sulkowski, M.S. and D.L. Thomas, Hepatitis C in the HIV-infected patient. Clin Liver Dis, 2003. 7(1): p. 179-94.  14.  Torriani, F.J., et al., Peginterferon Alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med, 2004. 351(5): p. 438-50.  15.  Poynard, T., et al., Viral hepatitis C. Lancet, 2003. 362(9401): p. 2095-100.  16.  Flamm, S., Chronic Hepatitis C Virus Infection. JAMA, 2003. 289(18): p. 2413-2417.  17.  A n o n y m o u s , Recommendations from the National Institutes of Health consensus development conference statement: management of hepatitis C: 2002. Hepatology., 2002. 36(5): p. 1039.  18.  Anonymous, Canadian Consensus Conference on the Management of Viral Hepatitis. Canadian Journal of Gastroenterology, 2000.14(Suppl B): p. 5B-20B.  19.  Perez-Olmeda, M., et al., Virological characteristics of hepatitis C virus infection in HIV-infected individuals with chronic hepatitis C: implications for treatment. Aids, 2002.16(3): p. 493-5.  20.  Serfaty, L., et al., Impact of early-untreated HIV infection on chronic hepatitis C in intravenous drug users: a case-control study. AIDS, 2001.15: p. 2011-2016.  4 2  21.  B a l l e s t e r o s , A . L . , et a l . ,  HIV/HCV  Early HCV dynamics on Peg-interferon and ribavirin in  co-infection: indications for the investigation of new treatment approaches.  A i d s , 2 0 0 4 . 1 8 ( 1 ) : p . 59-66. 22.  T o r r i a n i , F . , et a l . ,  Hepatitis C Virus (HCV) and Human Immunodeficiency  (HIV) Dynamics during HCV Treatment in HCV/HIV  Coinfection.  Virus  Journal  of  I n f e c t i o u s D i s e a s e s , 2 0 0 3 . 1 8 8 ( 1 5 N o v ) : p . 1498-507. 23.  P e r e z - O l m e d a , M . , et a l . , Pegylated  hepatitis C in HIV-infected patients. 24.  IFN-alpha2b plus ribavirin as therapy for chronic  A i d s . , 2 0 0 3 . 1 7 ( 7 ) : p . 1023-8.  K l e i n , M . , R . L a l o n d e , a n d S. S u i s s a ,  The Impact of Hepatitis C Virus Coinfection on  HIV Progression Before and After Highly Active Antiretroviral Therapy.  J A I D S , 2003.  33: p . 365-72. 25.  M e l v i n , D . , et a l . ,  The impact of co-infection with hepatitis C virus and HIV on the  tolerability of antiretroviral therapy. 26.  Z a l a , C , et a l .  ofHAART  The impact of the hepatitis C virus on CD4-T cell response post-initiation  among patients enrolled in clinical trials, i n 11th Conference on Retroviruses  and Opportunistic Infections. 27.  A I D S , 2000. 14(4): p . 4 6 3 - 4 6 5 .  B r a i t s t e i n , P . , et a l .  2004. S a n F r a n c i s c o , C A .  Impact of the Hepatitis C virus on CD4 response post initiation of  Highly Active Antiretroviral  Therapy Among a Population-Based HIV Treatment  Cohort, i n International AIDS Society Conference on HIV Pathogenesis and Treatment. 2003. P a r i s , F r a n c e : A n t i v i r a l T h e r a p y . 28.  A c e t i , A . , et a l . ,  Hepatotoxicity development during antiretroviral therapy containing  protease inhibitors in patients with HIV: the role of hepatitis B and C virus infection. J A c q u i r I m m u n e D e f i c S y n d r , 2002. 29(1): p . 41-8. 29.  C h u n g , R . , et a l . ,  Immune recovery is associated with persistent rise in hepatitis C virus  RNA, infrequent liver flares, and is not impaired by hepatitis C virus in co-infected subjects.  A I D S , 2 0 0 2 . 1 6 ( 1 4 ) : p . 1915-23.  30.  Yeni, P.G., et al., Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society-USA Panel. Jama, 2002. 288(2): p. 222-35.  31.  Dieleman, J., et al., Determinants of recurrent toxicity-driven switches of highly active antiretroviral therapy. The ATHENA  32.  Cohort. AIDS, 2002.16(5): p. 737-745.  Wood, E., et al., Elevated rates of antiretroviral treatment discontinuation among HIVinfected injection drug users: implications for drug policy and public  health.  International Journal of D r u g Policy, 2004.15: p. 133-138. 33.  Ripamonti, D., et al., Hepatitis C infection increases the risk of the modification of first highly active antiretroviral therapy in HIV-infected patients. AIDS, 2004. 18(2): p. 33436.  34.  d'Arminio Monforte, A., et al., Insights into the reasons for discontinuation of the first highly active antiretroviral therapy (HAART) regimen in a cohort of antiretroviral naive patients. AIDS, 2000.14(5): p. 499-507.  35.  Becker, S., Liver toxicity in epidemiological cohorts. Clin Infect Dis, 2004. 38(Suppl 2): p. S49-55.  36.  Kontorinis, N . and D. Dieterich, Hepatotoxicity of antiretroviral therapy. AIDS Rev, 2003. 5(1): p. 36-43.  37.  Dieterich, D., Managing antiretroviral-associated liver disease. J Acquir Immune Defic Syndr, 2003. 34(Suppl 1): p. S34-9.  38.  Trimoulet, P., et al., Intrahepatic HCV RNA loads in 37 HIV-HCV  co-infected patients  with controlled HIV infection. J Med Virol, 2002. 67(2): p. 143-151. 39.  Nelson, M . and G. Matthews, British HIV Association (BHIVA) guidelines for treatment and management of HIV and hepatitis C coinfection. 2003, BHIVA.  44  40.  Sulkowski, M.S., et al., Hepatotoxicity  associated with nevirapine or efavirenz-  containing antiretroviral therapy: role of hepatitis C and B infections. Hepatology, 2002. 35(1): p. 182-9. 41.  Martinez, E., et al., Hepatotoxicity in HIV-1 infected patients receiing nevirapinecontaining antiretroviral therapy. AIDS, 2001.15: p. 1261-68.  42.  den Brinker, M., et al., Hepatitis B and C virus co-infection and the risk for hepatotoxicity of highly active antiretroviral therapy in HIV-1 infection. AIDS, 2000.14: p. 2895-2902.  43.  Saves, M., et al., Hepatitis B or Hepatitis C Virus Infection is a Risk Factor for Severe Hepatic Cytolysis after Initiation of a Protease Inhibitor-Containing Regimen in Human Immunodeficiency  Virus-Infected  Patients.  Antimicrobial  Agents  and  Chemotherapy, 2000. 44(12): p. 3451-3455. 44.  Saves, M., et al., Severe hepatic cytolysis: incidence and risk factors in patients treated by antiretroviral combinations Aquitaine Cohort, France, 1996-1998. AIDS, 1999. 13(17): p. F115-F121.  45.  Rodriguez-Rosado, R., J. Garcia-Samaniego, and V. Soriano, Hepatotoxicity after introduction of highly active antiretroviral therapy. Aids., 1998.12(10): p. 1256.  46.  Michelet, C , et al., Differential effect of ritonavir and indinavir on immune response to hepatitis C virus in HIV-1 infected patients. AIDS, 1999.13(14): p. 1995-1997.  47.  Nunez, M., et al., Risk factors for severe hepatic injury after introduction of highly active antiretroviral therapy. J Acquir Immune Defic Syndr, 2001. 27(5): p. 426-31.  48.  Sulkowski, M., et al., Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA, 2000. 283(1): p. 74-80.  49.  Palmon, R., et al., Lack of hepatotoxicity associated with nonnucleoside reverse transcriptase inhibitors. J Acquir Immune Defic Syndr, 2002. 29(4): p. 340-5. 45  50.  Zylberberg, H . , et al., Tritherapy for Human Immunodeficiency Virus Infection Does Not Modify Replication of Hepatitis C Virus in Coinfected Subjects. Clinical Infectious Diseases, 1998. 26(May): p. 1104-06.  51.  Wit, F., et al., Incidence of and risk factors for severe hepatotoxicity associated with antiretroviral combination therapy. Journal of Infectious Diseases, 2002. 186: p. 2331.  52.  Livry, C , et al., Acute liver enzyme elevations in HIV-l-infected patients. H I V C l i n Trials, 2003. 4(6): p. 400-10.  53.  Nunez, M . , et al., Role of hepatitis C virus genotype in the development of severe transaminase elevation after the introduction of antiretroviral therapy. JAIDS Journal of Acquired Immune Deficiency Syndromes, 2002. 30(1): p. 65-8.  54.  Rubbia-Brandt, L., et al., Steatosis affects chronic hepatitis C progression in a genotype specific way. Gut, 2004. 53(3): p. 406-12.  55.  Sharma, P., et al., Hepatic steatosis in hepatitis C virus genotype 3 infection: does it correlate with body mass index, fibrosis, and HCV risk factors? D i g Dis Sci, 2004. 49(1): p. 25-9.  56.  Picard, O., O. Rosmorduc, and J. Cabane, Hepatotoxicity associated with ritonavir. Annals of Internal Medicine, 1998.129(8): p. 670-1.  57.  Sherman, K., et al. Evaluation of HCV RNA and Liver Injury in HCV'/HIV Coinfected Patients  Initiating  Lopinavir/r  or Nelfinavir-based  Therapy, i n Conference  on  Retroviruses and Opportunistic Infections. 2004. San Francisco, C A . 58.  Chihrin, S., et al. Exposure to lopinavir/r is a risk factor for grade 3/4 elevations of ALT in HIV and hepatitis B and/or hepatitis C co-infected patients, i n Canadian Association for HIV Research. 2004. Montreal, PQ.: Canadian Journal of Infectious Diseases.  46  59.  Barbaro, R., G . D i Lorenzo, and A.e.a. Asti, Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: Ultrastructural  and biochemical  findings.  American Journal of Gastroenterology, 1999. 94: p. 2198-205. 60.  O k u d a , M . , et al., Mitochondrial  injury, oxidative stress, and antioxidant gene  expression are induced by hepatitis C virus core protein. Gastroenterology, 2002. 122(2): p. 366-75. 61.  Rust, C . and G.J. Gores, Does hepatitis C cause liver injury by pathways associated with mitochondrial dysfunction? A m J Gastroenterol, 1999. 94(8): p. 2003-5.  62.  Butt, A . A . , Fatal lactic acidosis and pancreatitis associated with ribavirin and didanosine therapy. AIDS Read, 2003.13(7): p. 344-8.  63.  Walker, U . A . , et al., Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology, 2004. 39(2): p. 311-7.  64.  Lafeuillade, A . , G . Hittinger, and S. Chadapaud, Increased mitochondrial toxicity with ribavirin in HIV/HCV  65.  coinfection. Lancet, 2001. 357(9252): p. 280-1.  Carr, A . , et al., A syndrome of peripheral lipodystrophy, hyperlipidemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS, 1998.12: p. F51-F58.  66.  Cote, H . C . , et al., Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients.[comment]. N e w England Journal of Medicine., 2002. 346(11): p. 811-20.  67.  Montaner, J.S., et al., Mitochondrial toxicity in the era of HAART: evaluating venous lactate and peripheral blood mitochondrial DNA in HIV-infected  patients taking  antiretroviral therapy. J Acquir Immune Defic Syndr, 2003. 34(Suppl 1): p. S85-90. 68.  A d i n o f i , L . , et al., Steatosis accelerates the progression of liver damage of chronic hepatitis C patient and correlates with specific HCV genotype and visceral obesity. Hepatology, 2001. 33: p. 1358-1364.  47  69.  M i l l e r , K . , et a l . ,  Lactic acidosis and hepatic steatosis associated with use of stavudine:  report offour cases. 70.  A n n a l s o f I n t e r n a l M e d i c i n e , 2 0 0 0 . 1 3 3 ( 3 ) : p . 192-196.  L o n a r d o , A . , et a l . ,  Steatosis and hepatitis C virus: mechanisms and significance for  hepatic and extrahepatic disease. 71.  H e a t h , K . V . , et a l . ,  G a s t r o e n t e r o l o g y , 2 0 0 4 . 1 2 6 ( 2 ) : p . 586-97.  Emerging drug toxicities of highly active antiretroviral therapy for  human immunodeficiency virus (HIV) infection. C u r r  D r u g T a r g e t s , 2 0 0 3 . 4(1): p . 13-  22. 72.  D e M a r t i n o , L . , et a l . ,  Viral RNA in nerve tissues of patients with hepatitis C infection  and peripheral neuropathy. M u s c l e 73.  74.  I j i c h i , T . , et a l . ,  Peripheral neuropathy in Japanese patients with hepatitis C virus  infection. I n t e r n  M e d , 2 0 0 3 . 42(5): p . 3 9 4 - 9 .  N e m n i , R . , et a l . ,  cryoglobulinemia. 75.  N e r v e , 2 0 0 3 . 27(1): p . 102-4.  Peripheral neuropathy in hepatitis C virus infection with and without J N e u r o l N e u r o s u r g P s y c h i a t r y , 2 0 0 3 . 74(9): p . 1267-71.  S u t i n e n , J . , et a l . , Increased fat  accumulation in the liver in HIV-infected patients with  antiretroviral therapy-associated lipodystrophy. A I D S , 76.  M a e n o , T . , et a l . , Mechanisms  2 0 0 2 . 1 6 : p . 2183-93.  of increased insulin resistance in non-cirrhotic patients  with chronic hepatitis C virus infection.  J Gastroenterol  H e p a t o l , 2 0 0 3 . 18(12): p .  1358-63. 77.  D u o n g , M . , et a l . , Association  infection in HIV-Hepatitis  between insulin resistance and hepatitis C virus chronic  C co-infected patients undergoing antiretroviral therapy.  J o u r n a l o f A c q u i r e d I m m u n e D e f i c i e n c y S y n d r o m e s , 2 0 0 1 . 27: p . 2 4 5 - 2 5 0 . 78.  W a n g , C . S . , et a l . ,  Community-based study of hepatitis C virus infection and type 2  diabetes: an association affected by age and hepatitis severity status. A  m J Epidemiol,  2 0 0 3 . 1 5 8 ( 1 2 ) : p . 1154-60. 79.  Weinman,  S . A . a n d L . M . Belalcazar,  G a s t r o e n t e r o l o g y , 2 0 0 4 . 1 2 6 ( 3 ) : p . 917-9.  Hepatitis C: a metabolic liver disease.  80.  Ratziu, V . , J.B. Trabut, and T. Poynard, Fat, diabetes, and liver injury in chronic hepatitis C. Curr Gastroenterol Rep, 2004. 6(1): p. 22-9.  81.  Butt, A . , S. Fultz, and A . Justice. Hepatitis C increases the risk of DM in HIV infected veterans in care, i n Conference on Retroviruses and Opportunistic Infections. 2003. Boston, M A .  82.  Mehta, S.H., et al., The effect of HAART and HCV infection on the development of hyperglycemia among HIV-infected persons. J A c q u i r Immune Defic Syndr, 2003. 33(5): p. 577-84.  83.  Patroni, A . , et al., Effect of highly active antiretroviral therapy (HAART) and hepatitis C Co-infection on hyperlipidemia in HIV-infected patients: a retrospective longitudinal study. H I V C l i n Trials, 2002. 3(6): p. 451-61.  84.  Torti, C , et al., Influence of hepatitis C virus coinfection on lipid abnormalities in HIVpositive patients after highly active antiretroviral therapy. J A c q u i r Immune Defic Syndr, 2002. 29(3): p. 315-7.  85.  Benhamou, Y., et al., Liver fibrosis progression in human immunodeficiency virus and hepatitis C virus coinfected patients. The Multivirc Group. Hepatology., 1999. 30(4): p. 1054-8.  86.  Garcia-Samaniego, J., et al., Hepatocellular carcinoma in HIV-infected patients with chronic hepatitis C.[comment]. American Journal of Gastroenterology., 2001. 96(1): p. 179-83.  87.  D i Martino, V . , et al., The influence of human immunodeficiency virus coinfection on chronic hepatitis C in injection drug users: a long-term retrospective cohort study. Hepatology, 2001. 34(6): p. 1193-9.  88.  Rockstroh, J., et al., Immunosuppression may lead to progression of hepatitis C virusassociated liver disease in hemophiliacs coinfected with HIV. A m e r i c a n Journal of Gastroenterology, 1996. 91(12).  89.  Puoti, M et al., Liver fibrosis progression is related to CD4 cell depletion in patients v  coinfected with hepatitis C virus and human immunodeficiency virus. J Infect Dis, 2001. 183(1): p. 134-7. 90.  Benhamou, Y . , et al., Factors affecting liver fibrosis in human immunodeficiency virus and hepatitis c virus coinfected patients: impact of protease inhibitor therapy. Hepatology, 2001. 34: p. 283-287.  91.  Qurishi, N . , et al., Effect of antiretroviral therapy on liver-related mortality in patients with HIV and hepatitis C virus co-infection. Lancet, 2003. 362: p. 1708-13.  92.  Martinez-Sierra, C , et al., Progression of Chronic Hepatitis C to Liver Fibrosis and Cirrhosis in Patients Coinfected with Hepatitis C Virus and Human Immunodeficiency Virus. Clinical Infectious Diseases, 2003. 36: p. 491-8.  93.  De Simone, J., R. Pomerantz, and T. Babinchak, Inflammatory Reactions in HIV-1Infected Persons After Initiation of Highly Active Antiretroviral Therapy. Annals of Internal Medicine, 2000.133: p. 447-454.  94.  Martin-Carbonero, L . , et al., Liver injury after beginning antiretroviral therapy in HIV/hepatitis C virus co-infected patients is not related to immune reconstitution. A i d s , 2002.16(10): p. 1423-5.  95.  Einav, S. and M . J . K o z i e l , Immunopathogenesis of hepatitis C virus in the immunosuppressed host. Transplant Infectious Disease., 2002. 4(2): p. 85-92.  96.  Blackard, J., et al., Hepatitis C Virus (HCV) Diversity in Subjects Initiating Highly Active Antiretroviral  HIV-HCV-Co-Infected  Therapy. Journal of Infectious  Diseases, 2004.189(15 Apr): p. 1472-81. 97.  Soriano, V . , et al., Mortality due to chronic viral liver disease among patients infected with human immunodeficiency virus. Clinical Infectious Diseases, 1995. 33(10): p. 1793-5.  50  98.  Cribier, B., et al., High hepatitis C viraemia and impaired antibody response in patients coinfected with HIV. A i d s , 1995. 9(10): p. 1131-6.  99.  Vento, S., et al., Enhancement of hepatitis C virus replication and liver damage in HIVcoinfected patients on antiretroviral combination therapy. A i d s , 1998.12(1): p. 116-7.  100.  Eyster, M . , et al., Increasing hepatitis C virus RNA levels in hemophiliac patients with human immunodeficiency virus infection. Clinical Infectious Diseases, 1994. 84: p. 1020-3.  101.  Ragni, M . and F. Bontempo, Increase in Hepatitis C Virus Load in Hemophiliacs during Treatment with Highly Active Antiretroviral Therapy. Journal of Infectious Diseases, 1999.180: p. 2027-9.  102.  John, M . , J. Flexman, and M . A . French, Hepatitis C virus-associated hepatitis following treatment of HIV-infected patients with HIV protease inhibitors: an immune restoration disease? A i d s , 1998.12(17): p. 2289-93.  103.  Karras, A . , et al., Severe anoxic hepatic necrosis in an HIV-1 -hepatitis C virus coinfected patient starting antiretroviral triple combination therapy. A I D S , 1998.12(8): p. 827-29.  104.  Zylberberg, PL, et al., Rapidly evolving hepatitis C virus-related cirrhosis in a human immunodeficiency virus-infected patient receiving triple antiretroviral therapy. Clinical Infectious Diseases., 1998. 27(5): p. 1255-8.  105.  De Bona, M . , et al., The impact of liver disease and medical complications on quality of life and psychological distress before and after liver transplantation. Journal of Hepatology, 2000. 33(4): p. 609-15.  106.  Sulkowski, M . , et al., Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B infection. J A M A , 2000. 283(1): p. 74-80.  5 1  107.  Fialaire, P et al., Sustained disappearance of hepatitis C viremia in patients receiving v  protease inhibitor treatment for human immunodeficiency virus infection. Journal of Infectious Diseases., 1999.180(2): p. 574-5. 108.  Cooper, C . and D . Cameron, Interpretation of undetectable hepatitis C virus RNA levels in HIV-hepatitis C virus co-infection. AIDS, 2004.18(2).  109.  Anonymous, National Institutes of Health Consensus Development  Conference  Statement: Management of hepatitis C 2002 (June 10-12, 2002). Gastroenterology., 2002.123(6): p. 2082-99. 110.  Graham, C , et al., Influence of Human Immundeficiency Virus Infection on the Course of Hepatitis C Virus Infection: A Meta-Analysis. Clinical Infectious Diseases, 2001. 33: p. 562-569.  111.  Monforte A d e , A . , et al., Low frequency of severe hepatotoxicity and association with HCV coinfection in HIV-positive patients treated with HAART. J A c q u i r Immune Defic Syndr, 2001. 28(2): p. 114-23.  112.  Puoti, M . , et al., Liver damage and kinetics of hepatitis C virus and human immunodeficiency virus replication during the early phases of combination antiretroviral treatment. Journal of Infectious Diseases, 2000.181(June): p. 2033-2036.  113.  John, P.R. and P.J. Thuluvath, Outcome of liver transplantation in patients with diabetes mellitus: a case-control study. Hepatology., 2001. 34(5): p. 889-95.  52  Table 2.1. Summary of Evidence For and Against Controversial Hypotheses Hypothesis: HCV Causes a Blunted Immune Response For Against Z a l a et a l . (2004): S m a l l e r Klein et al. (2003): no difference in mean proportion of H C V + had > 75 cell C D 4 count 24 m o n t h s p o s t - A R T increase, and took longer to achieve initiation[24] in those who d i d [26] Braitstein et al. (2003): Multivariate C h u n g et al. (2002): no difference i n m i x e d effects m o d e l s suggest mean C D 4 cell increase at weeks 16 or smaller average gain of CD4's over 48[29] 18 months [27] K l e i n et a l . (2003): H C V + h a d A c e t i et a l . (2002): no significant reduced hazard of achieving > 50 differences i n C D 4 cell count increases cells / m m 3 [24] after 6,12, or 24 months of therapy[28] Greub et al. (2000): H C V + h a d M e l v i n et al. (2000): no significant longer time to C D 4 increase of > 50 difference i n change from baseline of cells/mm3[12] CD4 lymphocyte count[25] Hypothesis: HCV co-infection leads to higher toxicity-related antiretroviral discontinuation For Against M e l v i n et al. (2000): H C V - p o s i t i v e D ' A r m i n i o Monforte et al. (2000): no twice as likely to discontinue due to impact of H C V o n treatment severe hepatotoxicity [25] interruptions due to toxicity[34] R i p a m o n t i et a l . (2004): H C V positive had 40% increased risk of discontinuing or changing A R T i n first year of treatment[33] Aceti et al. (2002): 83% of those who d i s c o n t i n u e d A R T because of hepatotoxicity were H C V positive[28] Hypothesis: HCV-seropositivity leacs to increased hepatotoxicity as measured by AST and/'or\LT / (also see Table 1) For ( H C V + vs. H C V - ) Against ( H C V + vs. H C V - ) L i v r y et al. (2003): H C V + odds Palmon et al. (2002): 0% vs. 2.3% [49] ratio: 3.4 (vs. H C V - OR: 1.0) [52] Aceti et al. (2002): 5% vs. 1% [28] S u l k o w s k i et al. (2000): 30% vs. 30% (ritonavir combinations) and 8% vs. 5% (non-ritonavir combinations) [106] Sulkowski et al. (2002): 17% vs. 7% [40] Wit et al. (2002): 34% vs. 23% [51] Nunez et al. (2001): 16% vs. 5% [47] Monforte et al. (2001): 11% vs. 7% [111] Den Brinker et al. (2000): 33% vs. 12% [42] 53  S a v e s et a l . (2000): O d d s R a t i o v s . 1.0  8.0  [43]  M i c h e l e t et a l . (1999): 5 7 % v s .  11%  [46] S a v e s et a l . (1999): 10% v s . <1% Rodriguez-Rosaldo 2 1 % v s . 7.4%  et  al.  [44]  (1998):  [45]  Hypothesis: Antiretroviral therapy increases HCV replication For  Against  C h u n g et a l . (2002): A m o n g CD4  <350  cells/mm3,  baseline  H C V R N A  i n c r e a s e d 0.43 l o g at w e e k 16 a n d 0.59 l o g at w e e k 48; a m o n g b a s e l i n e  Z y l b e r b e r g et a l . (1998): n o s i g n i f i c a n t c h a n g e i n H C V R N A after 9 of A R T  months  [50]  CD4  >350, H C V R N A i n c r e a s e d 0.26 l o g at w e e k 16 a n d 0.10 l o g at w e e k 4 8 Martin-Carbonero  et  al.  [29]  (2002):  increase i n H C V R N A f r o m  baseline  5.71 l o g s t o 5.88 l o g s (p=0.01) [94] P u o t i et a l . (2000): f r o m b a s e l i n e o f 7 l o g , H C V R N A i n c r e a s e d to 7.33 l o g at d a y 14 a n d 7.29 l o g at d a y 21 Michelet  et  increases  b e t w e e n 0.22  log  al.  depending  [112]  (1999): H C V R N A on  log and  treatment  0.61  group  [46] Ragni  et  al.  (1999):  H C V  i n c r e a s e d f r o m 141 X 105 b a s e l i n e to 198 X 105 48 a n d 2 2 7 X 105  R N A  Eq/ml  at  E q / m l at w e e k  E q / m l at w e e k  96  p o s t - A R T i n i t i a t i o n [101]  Hypothesis: That immune restoration , a deleterious impact on progression has c>f HCV-related \tepatic fibrosis For  Against  Patients initiating A R T f r o m a m o r e  Finding  immunocompromised  inhibitors  state  have  g r e a t e r i n c r e a s e s i n H C V R N A [29,  that  that  was  not  using  associated  protease  with  faster  p r o g r e s s i o n to c i r r h o s i s [90]  57, 98] Case  reports  indicating fibrosis,  and  rapid  case  series  evolution  cirrhosis, a n d  of  end-stage  F i n d i n g that a n t i r e t r o v i r a l use  reduced  long-term liver-related mortality a m o n g co-infected h e m o p h i l i a c patients  [91]  liver disease u p o n A R T initiation [99,101,103,104,113] L o w b a s e l i n e C D 4 c o u n t is s t r o n g l y  C a s e reports of s p o n t a n e o u s  associated  of H C V R N A f o l l o w i n g a n t i r e t r o v i r a l  fibrosis,  with  progression  cirrhosis,  of and  clearance  i n i t i a t i o n [8,107]  h e p a t o c e l l u l a r c a r c i n o m a [85-88, 90] F i n d i n g t h a t C D 4 i n c r e a s e o f > 50  Other findings suggesting weak or  cells w a s  association b e t w e e n baseline C D 4 , C D 4  associated  with a  grade  no  54  3/4 change i n A S T / A L T [40,106]  recovery, a n d the developm ent of elevated liver enzymes [28,47,94]  5 5  T a b l e 2.2. S u m m a r y o f a v a i l a b l e l i t e r a t u r e r e g a r d i n g h e p a t o t o x i c i t y agents i n H I V / H C V co-infected  Reference &  Hepato-  Sample Size  L i v r y 2003  individuals.  Overa  toxicity  11  Defined  Rate  ALT  21%  [52]  >=2.5X  n=239  U L N  of antiretroviral  HCV+  HCV-  Predictive  Rate  Rate  Factors  Odds  Odds  H C V ,  H B V ,  r a t i o : 3.4  r a t i o : 1.0  antiretroviral therapy, C D C Stage  C ,  baseline A S T A c e t i 2002  A L T  [28]  U L N  >5X  3.2%  n=1325  1%  5%  HCV,  unadjuste  ritonavir,  d  among  co-  infected  only,  6.8,  odds: 4.6-  10.0  w i t h  or  without saquinavir, alcohol, CD4 to Martin-  Severe  Carbonero  >=5XULN  14%  14%  no  response  HAART  NNRTI's  -  (undifferentia  2 0 0 2 [94]  ted)  n=42 Sulkowski  G r a d e 3 or  2 0 0 2 [40]  4  n=568  from baseline  8%  Pi's  Pi's  inhibitors  1.1%  0%  2.3%  -  6.3%  34%  23%  Baseline A L T ,  change  15.6%  17%  7%  (nevir  21%  for  apine)  E F V  +  6% E F V  H C V , for +  H B V ,  nevirapine, protease  (efavi renz) Palmon  G r a d e 3 or  2002 [49]  4  n=272  change  from baseline  Wit  2002  [51] n = 5 6 0  Grade  4  (lOXULN  ritonavir,  or  nevirapine,  >200U  above  saquinavir,  baseline)  H C V ,  H B V ,  female  sex,  recent discontinuati on  of  first  3TC, ever  HAART Nunez  A L T  2001 [47]  U L N  n=222  >=3.5X  >5X or  9%  16%  5%  Alcohol abuse, age,  older H C V 56  baseline  infection, d d l , ritonavir Baseline A L T , H C V , HbsAg, previous nonHAART therapy, A Z T Baseline A L T , HBV, H C V  Monforte 2001 [111] n=1255  A L T >=200 IU/L  4.5%  11%  7%  Den Brinker 2000 [42] n=394  ALT or AST >=5XULN and absolute increase of >100U/1 Severe (grade 3 / 4 AST/ALT)  18%  33% (45% HbsAg+)  12%  10%  Ritonavir combo: 30% Nonritonavir combo: 8% OR: 8.0  Ritonavir combo: 30% Nonritonavir combo: 5% OR: 1.0  HBV, H C V  57% ritonavir, 10% indinavir  11%  ritonavir  Baseline A L T , HBV, H C V infection HCV  Sulkowski 2000 [106] n=298  Saves 2000 [43] n=1047 Michelet 1999 [46] n=146  Severe: ALT >=5XULN Mean A L T rise  5%  Saves 1999 [44] n=1997 Rodriguez -Rosado 1998 [45] n=187  A L T >=5X ULN  2%  10%  <1%  AST or ALT >2fold from baseline  14%  21%  7.4%  Ritonavir, C D 4 increase >50 cells  57  Table 2.3. Special Considerations and their rationale regarding the initiation and management of antiretroviral therapy in H I V / H C V co-infected individuals.  Special Consideration Rationale Before Starting Antiretroviral Therapy: Treat the H C V if possible, even • r e d u c e t o x i c i t y of antiretroviral without indicators of H C V disease therapy progression, p r i o r to i n i t i a t i n g • reduce probability of liver disease antiretroviral therapy and at h i g h progression CD4 counts. • m a x i m i z e probability of successful H C V treatment outcome Genotype the H C V  • genotype 3 is prone to severely elevated liver enzymes and development of fatty liver • genotype 3 has favorable H C V treatment profile  When to Start Antiretroviral Therapy: C o n s i d e r i n i t i a t i n g antiretroviral • even m i l d immune suppression may therapy at higher C D 4 counts (e.g. enhance H C V replication and accelerate >350 c e l l s / m m ) , p a r t i c u l a r l y liver disease a m o n g m e n a n d those w h o • m o d e r a t e a n d severe immune acquired H C V at an older age. suppression are strongly associated with accelerated liver disease • i m m u n e reconstitution may cause enhanced H C V replication and histologic damage through an autoinflammatory response • evidence may suggest a b l u n t e d i m m u n e response among co-infected initiating H A A R T When Starting /antiretroviral Therapy: A v o i d using antiretroviral drugs or • hepatitis C i n d e p e n d e n t l y causes combinations of drugs k n o w n to mitochondrial damage; cause mitochondrial toxicity (i.e. • clinical manifestations of d d l , d4T), and monitor closely for m i t o c h o n d r i a l damage include fatty signs of lactic acidemia/acidosis, liver (especially i n those w i t h genotype particularly a m o n g w o m e n and 3 and / o r lipodystrophy), peripheral older individuals. neuropathy, l i p o a t r o p h y , lactic acidemia A v o i d concomitant use of A Z T and • both A Z T and r i b a v i r i n can cause ribavirin, and concomitant use of profound anemia d d l or d4T and ribavirin • d d l , d4T, and ribavirin can all cause lactic a c i d o s i s , e s p e c i a l l y w h e n combined A v o i d using nevirapine and full • both agents are strongly associated dose ritonavir and be aware of the w i t h severe hepatic toxicity 3  58  lack of data regarding booster doses of ritonavir i n co-infected patients. A v o i d concomitant use of hepatitis C treatment. Consider baseline liver biopsy.  • overlapping and additive toxicities are very difficult to tolerate • to k n o w whether liver disease postHAART i n i t i a t i o n is stable or progressing After Starting Antiretroviral Therapy: If patient tested H C V a n t i b o d y • Antibodies to H C V may only appear negative p r i o r to a n t i r e t r o v i r a l w i t h an immune response therapy initiation, retest w i t h i n 6-12 • L i m i t e d evidence suggests the remonths; if patient had a negative emergence of H C V R N A u p o n H C V R N A test prior to H A A R T H A A R T initiation initiation, retest within 6-12 months. Rely on a variety of markers of liver • A L T levels are a poor predictor of inflammation and disease, including liver disease ultrasound and biopsy, to assess H C V disease progression. Monitor co-infected patients closely • associated with mitochondrial for fatty liver, particularly among damage, i n s u l i n resistance, a n d lipodystrophy those w i t h H C V genotype 3. Monitor co-infected patients closely for development of insulin resistance and diabetes, and encourage proactive preventive measures (diet and exercise).  • insulin resistance and diabetes are l i n k e d to both H C V infection and antiretroviral use  59  FIGURE  2.1.  OVERLAP  BETWEEN  MORBIDITY ASSOCIATED  WITH  HEPATITIS  C  A N D  TOXICITY ASSOCIATED W I T H ANTIRETROVIRALS  Antiretrovirals  Hepatitis C  Mitochondrial toxicity  Metabolic Abnormalities dislipidemia insulin resistance diabetes  hyperlactatemia  Age & Biological Sex  lactic acidosis/ acidosis peripheral  neuropathy lipoatrophy, lipodystrophy  lipoa trophy,  lipodystrophy  Hepatic Injury trartsaminitis tatty liver fibrosis / cirrhosis end stage liver disease  60  CHAPTER 3: STUDY SETTING, OVERVIEW OF METHODS, AND LIMITATIONS 3.1  STUDY SETTING: T H E BRITISH COLUMBIA PROGRAM A N DTHE  HOMER  HIV/AIDS  DRUG TREATMENT  COHORT  Antiretroviral medications have been centrally distributed at no cost to eligible HIV-infected individuals since 1986. In October 1992, the distribution of antiretroviral agents became the responsibility of the H I V / A I D S D r u g Treatment Program of the British C o l u m b i a Centre for Excellence i n H I V / A I D S .  This  antiretroviral d r u g distribution program remains the only free source of antiretroviral medication i n this Canadian province (and is a unique program i n Canada). The Centre's H I V / A I D S Drug Treatment program has received ethical approval from the University of British Columbia Ethics Review Committee at its St. Paul's Hospital site, and the program conforms w i t h the province's Freedom of Information and Protection of Privacy Act. The Centre distributes antiretroviral drugs based on specific guidelines generated by the Therapeutic Guidelines Committee [1]. These guidelines have been and continue to be consistent w i t h those treatment guidelines published by the International A I D S Society [2-5]. The Centre's guidelines recommend that C D 4 and plasma H I V - 1 R N A levels be monitored at baseline, at 4 weeks after starting antiretroviral therapy and every three months thereafter. Plasma viral loads are measured using the A m p l i c o r H I V - 1 M o n i t o r ™ (Roche Diagnostics Branchburg, NJ).  6 1  A l l classes of federally licensed antiretroviral drugs are currently available through the program, including all nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors (NNRTI), and protease inhibitors (PI).  Tenofovir, atazanavir, and enfuvirtide are also available.  Eligibility for antiretroviral medication has remained consistent w i t h current international recommendations [5]. Individuals are automatically entered into the D T P w h e n they are first prescribed  any a n t i r e t r o v i r a l agent.  A t D T P entry,  sociodemographic  information is captured, and the participant's complete history (if any) of antiretroviral use, C D 4 cell count, plasma v i r a l load, and disease stage are recorded at entry and w i t h each subsequent physician visit,.  Typically, patients  are followed-up at 3-month intervals, at which time prescriptions are renewed or altered based on treatment success and other clinical factors.  Blood drawn for  the purposes of C D 4 and viral load testing is stored for each i n d i v i d u a l at baseline and at each follow-up visit for future research activities related to H I V disease.  3.2  STUDY POPULATION  The H A A R T Observational Medical Evaluation and Research ( H O M E R ) study is a nested cohort w i t h i n the B C Centre for Excellence's H I V / A I D S D r u g Treatment Program.  It includes a l l previously A R T naive i n d i v i d u a l s w h o initiated  antiretroviral therapy w i t h a triple-drug regimen consisting of either two nucleoside reverse transcriptase inhibitors (NRTI) plus either a protease inhibitor  62  (PI), or a non-nucleoside reverse transcriptase inhibitors ( N N R T I ) , between August, 1996 and July, 2000. The data used i n these analyses are based on individuals from the H O M E R Cohort, for w h o m there was H C V serological data.  H O M E R consists of 1388 participants.  There were available samples w i t h  sufficient volume for testing for 1257 of them, and unambiguous H C V antibody results were obtained for 1186. Therefore, there were 1186 individuals available for analysis.  3.3 O T H E R D A T A  SOURCES  Laboratory Data: H C V antibody and H C V R N A testing were conducted on samples stored at baseline (0 to - 6 months p r e - A R T initiation) as part of treatment recipients participation i n the H I V / A I D S D r u g Treatment Program. A l l antibody and R N A testing was conducted at the U B C Virology Laboratory, located i n St. Paul's Hospital (Appendix 3.1). C D 4 and plasma H I V viral load data are collected routinely as patients receive their regular monitoring, as part of the approved data collection component of the H I V / A I D S D r u g Treatment Program. The biochemical markers of hepatic injury (alanine aminotransferase  ( A L T ) , aspartate aminotransferase  (AST),  albumin, and International Normalized Ratio (INR)) were obtained through a U B C Ethics approved linkage with the Providence Health Laboratory (Appendix 3.2).  63  Mortality: A l l deaths from H I V / A I D S i n the province were determined through a B C Centre for Excellence i n H I V / A I D S ' s initiated linkage w i t h the province's death registry housed at the British Columbia Ministry of Health's Vital Statistics Agency i n Victoria. A file was acquired from the British Columbia Ministry of Health's Vital Statistics Agency, which queried their database to extract records for all deaths that contained an ICD-9 (1995-1999) or ICD-10 (2000-2001) code indicating that H I V met the criteria of an underlying or associated cause of death. V i t a l Statistics defines a death as " H I V underlying" if the attending physician a n d / o r coroner determines that the death was directly due to H I V infection. Alternatively, Vital Statistics defines a death as " H I V associated" if the attending physician a n d / o r coroner determines that H I V infection was only a condition that contributed to the individual's underlying cause of death. Persons who died of accidental causes—such as illicit d r u g overdoses, suicides, and accidents—are not routinely tested for H I V i n the province, and unless an individual is experiencing morbidity due to confirmed H I V infection, their death should not be coded as associated with H I V [19].  Adherence to Antiretrovirals: The measure of adherence used by the H I V / A I D S Drug Treatment Program is a continuous variable based on the ratio of time that medication dispensed w o u l d last as a proportion of follow-up time.  This  calculation is restricted to each patient's first year on therapy to avoid reverse causation that could occur among patients who cease antiretroviral therapy after they have become too sick to take medication [6]. This measure is a strong predictor of H I V virologic response [7] and survival [8, 9]. For the purposes of the present analyses, adherence was a binary variable where 1 was being greater  64  than 95% adherent (meaning having refilled at least 95% of their prescriptions during the year), and 0 less than 95% adherent.  This is based on the rationale  that maximal virologic suppression requires greater than 95% adherence to antiretroviral medication [9,10].  3.4 S T A T I S T I C A L M E T H O D S  Standard statistical techniques were used i n each study. In Chapters 4, 5, 6 and 7, parametric and non-parametric tests were used, some i n cross- sectional analysis and some i n both cross-sectional and longitudinal analyses. For the cross sectional analyses, such as baseline characteristics, categorical explanatory variables were analyzed using Pearson's Chi-square test a n d continuous variables were analyzed using the Wilcoxon rank s u m test. Fisher's exact test was used for 2X2 contingency tables i n w h i c h any of the expected cell frequencies was less than 5.  Multivariate methods were used in Chapters 4-7 to account for potential confounding effects. Chapters 4 and 6 used multivariate logistic regression; chapter 5 used a combination of mixed effects models and C o x Proportional Hazards methods to describe the impact of H C V co-infection on C D 4 response to A R T . Chapter 7 used standard survival analysis techniques. Cumulative event rates were derived using Kaplan-Meier methods, and s u r v i v a l curves were compared between groups w i t h the log-rank test. C o x proportional hazards regression was used to calculate univariate and adjusted relative hazards and  65  95% confidence intervals (CIs). In each case, the assumption of proportional hazards was validated by inspection of log (-log [survival function]) estimates against log time plots.  A l l statistical analyses were performed using S A S software version 6.2 (SAS, Cary, N C ) . A l l tests of significance were two-sided, w i t h a p-value of less than 0.05 indicating that an association was statistically significant. A more detailed description of the specific statistical methods e m p l o y e d i n each manuscript can be found i n the relevant chapters.  3.5  STRENGTHS A N D LIMITATIONS  Each study that makes up this thesis has specific strengths and limitations that w i l l be addressed i n detail i n each chapter. Nevertheless, some of these issues are common through many or all of the studies, and should be discussed at the outset.  A key first strength is that the data are from a setting where all H I V / AIDS care, antiretrovirals, and laboratory monitoring are available free of charge, and where previous studies have s h o w n that v i r t u a l l y a l l patients acquire antiretrovirals through a centralized source [10]. A s such, the interpretability of the findings presented here are less compromised by selection factors that may bias other cohort studies. The H O M E R cohort may be considered representative of those i n d i v i d u a l s c o m i n g forward for H I V treatment, but w i l l not be representative of all HIV-infected individuals in British C o l u m b i a [11]. This is especially pertinent i n the context of H I V / H C V co-infection because the people  66  most likely to have died of H I V i n BC without accessing antiretrovirals, may also have been more likely to have been co-infected w i t h H C V [11].  The measure of adherence used i n these analyses has been validated as a predictor of virologic suppression and mortality [7, 9, 12], and the fact that we were able to account for adherence to A R T i n chapters 5-7 significantly strengthens the interpretability of our findings. However, it should be noted that this is quite a crude measure, and is a proxy for actually swallowing pills. Further, though generally referred to as 'adherence', conceptually it includes people who don't adhere because of lifestyle or other, and those people w h o can't adhere, because of toxicity or co-morbidity.  H a v i n g H C V serology data for nearly the entire H O M E R cohort for chapters 5-7 enabled these analyses to have sufficient power, and to be conducted i n a population-based setting. These data are nonetheless limited by the fact that the testing was conducted on baseline samples, and may not reflect a patient's current H C V status.  For example, they may have acquired H C V  infection since they initiated A R T , or they may have been effectively treated for H C V since the time of testing. A related strength of this study was that we were able to re-test most antibody p o s i t i v e / R N A negative discordant samples following treatment initiation.  Objectives 5, 6, and 7 all use H C V serological data to define H C V coinfection. This could potentially lead to a misclassification bias by counting as 'HCV-negative' those 5% who have no positive antibodies but do have detectable H C V R N A [13, 14]. Misclassification could also occur i n counting as ' H C V -  67  positive' those individuals w i t h positive antibodies, but no R N A . A t the outset of this project, reports i n the literature indicated that a smaller proportion of H I V co-infected spontaneously clear H C V (resulting i n positive antibodies but no R N A ) compared to HIV-uninfected, although good studies examining this question are still lacking [15, 16]. A s w i l l be seen i n Chapter 4, a very high proportion of individuals w i t h positive H C V antibodies i n this cohort had no detectable H C V R N A at baseline (30%) or 6-12 months following H I V treatment initiation (25%), resulting i n a concern that estimates based on H C V serology w o u l d be biased. P C R testing was conducted on all baseline antibody-positive samples, and 10% of the antibody-negatives.  Therefore i n order to assess if there was a  possible misclassification bias because of the h i g h p r o p o r t i o n of antibodypositive / P C R negative discordant results i n the study, among individuals w i t h positive H C V antibodies, I performed a survival analysis comparing those w i t h detectable and undetectable H C V R N A . In unadjusted and adjusted C o x models, there were no differences detected, suggesting less of a potential for bias. There is justification to define ' H C V co-infected' based primarily on H C V serology. The absence of H C V R N A could be due to several factors, one of which is spontaneous clearance. Most of the samples tested were quite old (7-9 years), and h a d already been repeatedly frozen a n d t h a w e d .  Further,  undetectability does not indicate that there is no v i r u s present:  the  A m p l i P r e p / C O B A S system used i n this study for qualitatively detecting H C V R N A has a lower limit of detection of 100 c o p i e s / m l [17]. There is literature to  68  suggest that i n H I V co-infected patients, there is poor correlation between levels of H C V R N A i n the liver and i n the plasma, showing l o w levels i n the plasma but high levels i n the liver [18]. H I V has recently been found to facilitate H C V infection of macrophages [19], possibly creating a reservoir for H C V . In sum, lack of detectability of the virus may not mean the virus is not there. Therefore, basing analyses on H C V serological data is considered an acceptable approach.  3.6  SUMMARY  In summary, the methods employed i n this thesis are diverse and rely on a number of data sources. Through a retrospective diagnostic testing study and the statistical evaluation of quantitative data, these studies have endeavoured to describe some of the impact of H C V i n HIV-infected i n d i v i d u a l s initiating antiretroviral treatment. While each of the approaches applied i n the various sub-studies has i n d i v i d u a l strengths and weaknesses, together the combined studies provide insight into some of the possible consequences of H I V / H C V coinfection i n the H A A R T era. The unique contribution that these studies have made w i l l be described i n detail i n Chapter 8.  69  3.7 REFERENCES  1.  B C Center for Excellence  in HIV/AIDS.,  Therapeutic Guidelines for the  Treatment of HIV/AIDS and Related Conditions. 2.  C a r p e n t e r , C , et a l . , Antiretroviral  1999,2001: V a n c o u v e r .  therapy for HIV infection in 1996.  JAMA,  Therapy for HIV Infection in 1997.  JAMA,  1996. 276: p . 146-154. 3.  C a r p e n t e r , C , et a l . , Antiretroviral 1997. 277: p . 1962-1969.  4.  Carpenter,  C ,  et  al.,  Antiretroviral  therapy in adults.  recommendations of the International AIDS Society-USA Panel.  Updated  J A M A , 2000.  283(3): p . 3 8 1 - 9 1 . 5.  Y e n i , P . G . , et a l . ,  Antiretroviral treatment for adult HIV infection in 2002:  updated recommendations of the International AIDS Society-USA Panel. J a m a , 2002. 288(2): p . 222-35. 6.  S c h e c h t e r , M . , et a l . ,  Higher socioeconomic status is associated with slower  progression of HIV infection independent of access to health care. J  Clin  E p i d e m i o l , 1994. 47(1): p . 59-67. 7.  L o w - B e e r , S., et a l . ,  Adherence to triple therapy and viral load response.  J  A c q u i r I m m u n e D e f i c S y n d r , 2000. 23(4): p . 3 6 0 - 1 . 8.  W o o d , E . , et a l . ,  Is there a baseline CD4 cell count that precludes a survival  response to modern antiretroviral therapy? A i d s , 9.  H o g g , R . S . , et a l . ,  2 0 0 3 . 1 7 ( 5 ) : p . 711-20.  Intermittent use of triple-combination therapy is predictive of  mortality at baseline and after 1 year offollow-up.  A i d s , 2 0 0 2 . 1 6 ( 7 ) : p . 1051-8.  70  10.  S t r a t h d e e , S., et a l . ,  drug users. 11.  Wood,  Barriers to use of free antiretroviral therapy in injection  J A M A , 1998. 280: p . 547-9.  E . , et  Prevalence  al.,  and correlates  of untreated  human  immunodeficiency virus type 1 infection among persons who have died in the era of modern antiretroviral therapy.  Journal of Infectious Diseases, 2003.188: p.  1164-70. 12.  W o o d , E . , et a l . ,  Effect of medication adherence on survival of HIV-infected  adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 x 10(9) cells/L. 13.  A n n I n t e r n M e d , 2 0 0 3 . 1 3 9 ( 1 0 ) : p . 810-6.  Hepatitis C virus viremia in HIV-infected individuals with  G e o r g e , S . L . , et a l . ,  negative HCV antibody tests.  J A c q u i r I m m u n e D e f i c S y n d r , 2 0 0 2 . 31(2): p .  154-62. 14.  B r a i t s t e i n , P . , et a l .  Dangerous Oversights: A comparison ofHCV-RNA  vs. HCV-antibody  testing among HIV-infected  Association for HIV Research. 15.  Sulkowski,  M . S . , et a l ,  individuals,  testing  i n Canadian  2001. T o r o n t o , C a n a d a .  Hepatitis C and progression of HIV disease.  Jama,  2002. 288(2): p . 199-206. 16.  Piasecki,  B . , et a l . ,  Influence of alcohol use, race, and viral coinfections on  spontaneous HCV clearance in a US Veteran population.  H e p a t o l o g y , 2004. 40:  p . 892-899. 17.  Krajden, M . ,  Hepatitis C virus diagnosis and testing.  C a n a d i a n Journal of  P u b l i c H e a l t h . R e v u e C a n a d i e n n e d e S a n t e P u b l i q u e , 2 0 0 0 . 9 1 ( S u p p l 1): p . S34-9, S36-42.  7 1  18.  Furione, M . , et al., Dissociation of serum and liver hepatitis C virus RNA levels in patients coinfected with human immunodeficiency virus and treated with antiretroviral drugs. J C l i n Microbiol, 2004. 42(7): p. 3012-6.  19.  Laskus,  T.,  et  al.,  Human  immunodeficiency  virus  facilitates  infection/replication of hepatitis C virus in native human macrophages. Blood, 2004.103(10): p. 3854-9.  72  CHAPTER 4 DIAGNOSTIC ISSUES AMONG H I V / H C V CO-INFECTED INDIVIDUALS  4.1  FORWARD  This chapter presents the findings of the CIHR-funded study, "Establishing the True Prevalence of Hepatitis C A m o n g HIV-infected Individuals Initiating Antiretroviral Therapy" (Appendix 2). It is under review w i t h Hepatology as:  P. Braitstein, M . Krajden, C . Sherlock, B. Y i p , R. G a l l i , P.R. Harrigan, M . T . Schechter, J.S.G. Montaner, R.S. Hogg. "Antiretroviral Therapy and the Detectability of H C V R N A Before and After Initiating Treatment i n a Population-Based Cohort of HIV-infected Individuals".  73  4.2  INTRODUCTION  It is estimated that as m u c h as 3% of the w o r l d ' s population is infected w i t h the Hepatitis C virus ( H C V ) [1], including an estimated 2% of people from the United States [2], and 1% of Canadians [3]. A s of December, 2002, the United Nations had estimated that there were 42 m i l l i o n people living w i t h H I V / A I D S i n the w o r l d [4], including approximately 40,000 i n Canada [5].  Due to shared routes of transmission such as  receiving contaminated blood products and injection drug use, the prevalence of coinfection by both H I V and Hepatitis C is considerable, particularly i n some populations. In Canada and elsewhere i n the Developed W o r l d i n c l u d i n g the United States, it is estimated that as many as 30% of all individuals who have H I V also have Hepatitis C [6-8].  However, these estimates are drawn from clinic-based, clinical trial based, or  population-specific (e.g. injection drug users) samples.  In British C o l u m b i a , a l l  estimates of H I V / H C V co-infection are d r a w n from an urban cohort of injection drug users [7, 9, 10]. End-stage liver disease has become a leading cause of death i n H I V infected i n d i v i d u a l s and is threatening gains made by h i g h l y active antiretroviral therapy, particularly among H C V / H I V co-infected individuals [11-13].  The accurate diagnosis of H C V i n HIV-infected individuals is complicated by several factors. Due to a variety of factors including immune suppression, approximately 5% of HIV-infected patients w i l l test false antibody-negative for H C V [14, 15].  Further,  among those not infected w i t h H I V , approximately 15-20% can expect to spontaneously clear H C V R N A through a host immune response [16]. However, the probability of spontaneous H C V R N A clearance among H I V co-infected i n d i v i d u a l s has been reported to be much lower, at 5-10%, particularly among those w i t h lower C D 4 counts  74  [6, 17, 18].  Factors found to be associated w i t h spontaneous clearance include  significant alcohol use, race, and H I V co-infection [19].  Immune reconstitution may play an additional complicating factor i n the diagnosis of hepatitis C among HIV-infected patients. There have been two case reports of spontaneous clearance of H C V R N A following H A A R T initiation, presumably related to immune reconstitution [20, 21]. A n d of much greater concern, a recent but very small study found that of 50 H I V and H C V seropositive individuals assessed, H C V R N A was not detectable at baseline i n 10 of them. However following antiretroviral initiation, 4 (25%) of these subsequently became H C V R N A positive [22].  Hepatitis C treatments are evolving and offer hope to many H I V / H C V co-infected patients [23]. For those individuals for w h o m H C V treatment w i l l not be effective, there are several issues related to their H I V management and lifestyle that can be modified to reduce the probability of disease progression and maximize their therapeutic options [24].  The timely and accurate diagnosis of H C V i n HIV-positive individuals is of the  utmost clinical importance, and understanding the impact of antiretroviral therapy on that diagnosis is imperative.  Therefore, the objectives of the present study were to measure the baseline (pre-ART) prevalence of Hepatitis C antibodies among H I V - p o s i t i v e i n d i v i d u a l s initiating antiretroviral therapy i n a population-based H I V treatment cohort, the prevalence of discordant positive antibody/negative R N A results at baseline, and to describe factors associated w i t h each. Secondly, we sought to measure the impact of antiretroviral therapy on the detectability of H C V R N A among those individuals w h o were H C V R N A negative but antibody positive at baseline using a sample taken six to twelve 75  months following treatment initiation, and to describe factors associated w i t h newly detectable H C V R N A .  4.3  METHODS  The HIV/AIDS Drug Treatment Program Antiretroviral medications have been centrally distributed at no cost to eligible H I V infected individuals since 1986.  In October 1992, the distribution of antiretroviral  agents became the responsibility of the H I V / A I D S D r u g Treatment Program of the British C o l u m b i a Center for Excellence i n H I V / A I D S .  This antiretroviral d r u g  distribution program remains the only free source of antiretroviral medication i n the province.  The Centre's H I V / A I D S D r u g Treatment program has received ethical  approval from the University of British Columbia Ethics Review Committee at its St. Paul's H o s p i t a l site, and the program conforms w i t h the province's Freedom of Information and Protection of Privacy Act.  The Centre distributes antiretroviral drugs based on specific guidelines generated by the therapeutic guidelines committee, and have remained consistent w i t h international standards [25-28]. Plasma v i r a l loads were measured using the A m p l i c o r H I V - 1 Monitor™ (Roche Diagnostics Branchburg, NJ). A l l three classes of federally licensed antiretroviral drugs are currently available through the program, i n c l u d i n g all nucleoside and nucleotide reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors, and protease inhibitors.  Individuals are automatically entered into the D T P when they are first prescribed any antiretroviral agent.  A t D T P entry and w i t h each subsequent physician visit, the  76  participant's complete history (if any) of antiretroviral use, C D 4 cell count, plasma viral load, and disease stage are recorded. Typically, patients are followed-up at 3-month intervals, at w h i c h time prescriptions are renewed or altered based on treatment success and other clinical factors.  Blood drawn for the purposes of C D 4 and viral load testing  is stored for each individual at each follow-up visit for future research activities related to H I V disease.  Study Population: The H A A R T Observational M e d i c a l Evaluation and Research ( H O M E R ) study is a nested cohort w i t h i n the B C Centre for Excellence's H I V / A I D S D r u g Treatment Program. It includes all previously A R T naive individuals w h o initiated antiretroviral therapy w i t h a t r i p l e - d r u g regimen consisting of either two nucleoside reverse transcriptase inhibitors (NRTI) plus either a protease inhibitor (PI), or a non-nucleoside reverse transcriptase inhibitors (NNRTI), between August, 1996 and July, 2000. The data used i n these analyses are based on individuals from the H O M E R Cohort.  Study Design: Frozen, archived samples of b l o o d plasma are available to the D r u g Treatment Program. The first sample taken, that from within 6 months prior to H A A R T initiation, was tested for H C V antibodies (Ab). A l l individuals w i t h positive antibodies were tested for the presence of H C V R N A , and due to the prohibitive cost of P C R testing, only a random sample of 10% of those w i t h negative H C V antibodies were tested using P C R methods. Individuals w i t h positive-antibody and negative P C R results at baseline  77  were re-tested using a sample from 6 - 1 2 months post-ART initiation to determine H C V R N A detectability following treatment initiation.  HCV Screening: Blood samples are collected and stored frozen at -20C until time of processing. Plasma is separated within 6 hours of collection by centrifuging at 800-1600xg for 15 minutes at room temperature.  Some 0.025 u / L of sample was aliquoted from the stored sample.  Certified negative human plasma was added to the aliquot, and then used for testing.  Blood samples were tested for H C V antibody using the Abbott primary screen. This assay has a sensitivity greater than 99% and a specificity of 99%, i n immune-competent patients [29]. H C V R N A was detected using the Roche Cobas H C V A m p l i P r e p / C O B A S A m p l i c o r test. The A m p l i P r e p / C O B A S A m p l i c o r Hepatitis C V i r u s test (v.2.0) is a qualitative nucleic acid amplification test for the detection of H C V i n clinical specimens of human plasma.  Sample preparation is automated using the C O B A S AmpliPrep  Instrument, and amplification and detection are automated using the C O B A S Amplicor Analyzer. This is a qualitative assay with a lower level of detection of H C V R N A of 100 copies per ml. The specificity for this assay exceeds 98% [29].  A l l testing was completed at the University of British Columbia Virology Laboratory at St Paul's Hospital, and ethics approval was obtained prior to testing.  Statistical Methods: Baseline sociodemographic and clinical factors associated w i t h H C V antibody-positive prevalence and antibody-positive/PCR-negative results at baseline were performed using parametric and non-parametric methods. Student's t-test was used for normally 78  distributed continuous variables, and the W i l c o x o n Rank S u m test for non-normal continuous variables. Categorical data were analyzed using Pearson's Chi-Square test. Fisher's Exact Test was used for contingency tables i n w h i c h 25% or more of the expected cell frequencies were less than 5. A s this was a descriptive analysis, factors independently associated w i t h each outcome were assessed using forward stepwise multivariate logistic regression. The multivariate models were constructed by entering those variables w h i c h were bivariately statistically significant (p<0.05) or if they were considered potential confounders.  Independent variables measured were gender, age at baseline, whether they had any history of injection drug use, C D 4 count at baseline, H I V l o g  10  plasma R N A at baseline,  whether they had an A I D S diagnosis at baseline, and type of A R T initiated (protease inhibitor based or non-nucleoside reverse transcriptase inhibitor based). Ethnicity was not examined because ethnicity data is not routinely collected i n the DTP.  4.4  RESULTS  Of 1388 eligible individuals i n the H O M E R Cohort, 1257 samples were both available and had sufficient quantity for testing. Of these, 1186 samples had clear antibody results, including 606 (51%) antibody positive, and 580 (49%) antibody negative (Figure 1). Factors associated w i t h H C V positive antibody prevalence are summarized in Table 1 and were male sex (78% vs. 93%, p<0.001), any history of injection drug use (47% vs. 6%, p<0.001), an A I D S diagnosis at baseline (11% vs. 15%, p=0.028), and C D 4 fraction at baseline (19% vs. 16%, p<0.001). N o t associated w i t h H C V antibody prevalence were  79  median age at baseline, absolute baseline C D 4 count, baseline H I V R N A , or type of A R T initiated.  A m o n g those w i t h positive antibody test results, 605 samples were tested for H C V R N A at baseline (one clotted), and 425 (70%) were positive, while 179 (30%) were negative. A m o n g those w i t h negative antibody results, 58 were tested for H C V R N A . One clotted and was therefore not included; 56 (98%) were R N A negative, while 1 (2%) was H C V R N A positive. Factors associated w i t h antibody p o s i t i v e / R N A negative discordancy at baseline were any history of injection drug use (32% vs. 53%, p<0.001), having an AIDS diagnosis at baseline (17% vs. 8%, p=0.002), median baseline C D 4 (230 vs. 290, p=0.001), but not male sex, median age at baseline, or median baseline loglO H I V R N A .  In  multivariate logistic regression (Table 2), factors independently associated with baseline antibody-positive/RNA-negative discordancy were any I D U (Adjusted Odds Ratio, A O R 0.45, 95% confidence interval, CI: 0.31 - 0.65, p<0.001), having an A I D S diagnosis at baseline ( A O R 1.86, 95% CI: 1.05 - 3.28, p=0.033), and absolute baseline C D 4 count (per 100 cells) ( A O R 0.89, 95% CI: 0.81 - 0.98, p=0.018).  There were 118/179 discordant patients at baseline for w h o m a sample taken 6-12 months post-ART initiation was available. Of these, 94 (80%) remained H C V R N A negative, while 24 (20%) became H C V R N A positive. Compared to patients whose H C V R N A remained undetectable, factors associated w i t h new H C V R N A detectability following ART-initiation were median baseline C D 4 (70 vs. 230, p=0.040) and median log  10  H I V R N A at baseline  (5.5 vs. 5.0, p=0.018), but not male sex (67% vs. 78%,  p=0.226), any history of I D U (42% vs. 29%, p=0.235), having an A I D S diagnosis at baseline (13% vs. 18%, p=0.761), or median age at baseline (35.8 vs. 38.0, p=0.124). A s s u m m a r i z e d i n Table 3, i n multivariate logistic regression, the o n l y  factor 80  independently associated w i t h newly detectable H C V R N A i n those undetectable at baseline was baseline plasma l o g H I V R N A (AOR 2.80, 95% CI: 1.08 - 7.24, p=0.034). 10  4.5  DISCUSSION  O u r data indicate a very h i g h prevalence of H C V antibodies at baseline i n this population-based cohort, at 51%. These participants were less likely to be male, more likely to have any history of injection drug, use, and had a higher C D 4 fraction at baseline.  O u r study also found a very h i g h p r o p o r t i o n of discordant antibody  p o s i t i v e / P C R negative results at baseline (30%). Factors independently associated w i t h this baseline discordancy include injection drug use, baseline C D 4 count, and having an A I D S diagnosis at baseline. O u r data further suggest that H C V R N A may become detectable i n 20% of individuals w h o at baseline were H C V R N A negative, following A R T initiation, and that baseline H I V R N A may play an important role i n this change.  Our data have several critical implications. One, because the study population is a population-based sample, the prevalence of H C V we have found is the most accurate indication of the prevalence of H C V / H I V co-infection i n British Columbia, Canada, and is, to our knowledge, the only population-based measure reported i n the literature. Our finding is higher than other reported averages [30, 31], and may i n part reflect the large injection drug using population i n British Columbia. However, previous work done by our center suggests that 30% of people d y i n g HIV-related deaths never access antiretroviral therapy and are therefore not captured i n our cohort. These individuals are more likely to have sociodemographic profiles consistent w i t h BC's injection drug using population and therefore may be more likely to be co-infected w i t h H C V [32]. 81  Therefore, it may be expected that our estimate of 51% is a conservative reflection of the true prevalence of H C V among HIV-infected patients i n this province.  Two, although factors associated with H C V antibody-positive prevalence are consistent with others reported i n the literature [6, 33, 34], the majority of individuals with positive H C V antibodies i n our study have no reported history of injection drug use. This may indicate that H C V is a problem that goes w e l l beyond the traditional populations considered at h i g h risk of acquiring H C V (i.e. injection d r u g users).  H C V is  approximately 10 times more infectious than H I V [30], and can be spread through minute amounts of blood that may be present i n intranasal drug-using equipment, improperly sterilized tattoo equipment, or certain types of sexual behavior. Our study underlines the importance of H C V testing i n all HIV-infected patients.  The third major implication is that having undetectable H C V R N A may not mean that the virus is cleared from these patients. Being undetectable may be due to a variety of factors, including the age of our samples, or the fact that some of them were frozen and thawed a number of times prior to this study. The lower limit of detectability for the assay used i n this study is 100 copies/ml [29].  However, the prevalence of discordant positive antibodies/undetectable R N A at baseline in our study (30%) is significantly higher than other reports which suggest that only 5-10% of H I V co-infected may spontaneously clear H C V R N A [6, 17, 18]. If these results are really due to spontaneous clearance, they may be so high because of the high prevalence of n o n - I D U i n our study. M a n y individuals, particularly men who have sex with men over the age of 40 or 50, may have experimented w i t h injection drugs i n the 82  1970's or 1980's, contracted H C V infection prior to contracting H I V infection, and were thus more likely to have spontaneously cleared the H C V infection.  A previous very small report indicated that 4/10 baseline H C V R N A negative H I V infected patients developed detectable H C V R N A following A R T initiation [22]. O u r study confirms that at least 20% of HIV-infected individuals may develop detectable H C V R N A following A R T initiation, and further suggests that this may be related to plasma H I V R N A levels. This may be an indication of direct viral-viral interactions, or it may be related to the presence of H C V R N A reservoirs, particularly i n H I V coinfected patients [35]. Laskus et al. (2004) recently found that H C V was more likely to replicate i n monocytes/macrophages i n H I V co-infected patients [36]. The association in our data w i t h baseline H I V R N A may be related to the enhanced expression of H C V R N A from monocytes following treatment initiation.  Other data indicate several  mechanisms by w h i c h suppression of H I V replication and a reconstituted immune system influence H C V viral diversity i n H I V co-infected patients [37]. Controlled clinical trials prospectively examining some of these questions w o u l d be very useful.  This investigation has several important strengths. population-based cohort, and therefore  The study p o p u l a t i o n is a  has the advantage  of being b r o a d l y  representative of the population of people living w i t h H I V and Hepatitis C who have initiated antiretroviral therapy i n the H A A R T era, in the Developed W o r l d (in spite of the limitation referred to earlier). Two, because stored samples were available from both before and after treatment initiation, we were able to examine the effect of A R T on the prevalence of H C V i n this HIV-infected population, and to investigate factors associated w i t h antibody p o s i t i v e / R N A negative discordancy and post-treatment H C V R N A detectability. Three, effective H C V treatment only became available very recently, 83  and all samples tested were from a time period where the effect of H C V treatment, particularly i n HIV-co-infected patients, w o u l d be negligible, and is therefore not a significant potential confounder i n this analysis. Fourth, all the testing was done using the same assays, i n the same laboratory, eliminating any bias due to differences i n assays or laboratories.  There may also be limitations to this analysis. Due to cost constraints, we were not able to test the entire group of nearly 4000 individuals currently receiving antiretroviral therapy i n the province of British Columbia.  W e hoped that selecting previously  antiretroviral naive individuals w h o had initiated treatment w i t h a triple-combination w o u l d enable us to control for the potential confounding effects of extensive pretreatment. A l s o due to cost constraints, we were only able to test 10% of antibodynegative results. However, the probability of false-antibody negative results has been well established elsewhere [14] and was not a primary objective of our analysis. Third, the age of the samples and the repeated thawing of them may have compromised the integrity of sample.  H o w e v e r , samples that were not clearly antibody positive or  negative were excluded from all future analysis, as were samples that were not clearly either P C R positive or negative. Four, our measure of injection drug use is a composite using data available either from physician-report or patient self-report. There is likely some under-reporting of injection drug use, particularly among men who have sex w i t h men whose experience of using injection drugs may have been twenty or more years ago, and was only recreational i n nature. Finally, the sub-study examining H C V R N A detectability post-treatment had a small number of people w i t h the outcome of interest, limiting the interpretability of factors associated w i t h experiencing it.  84  Although screening for hepatitis C among HIV-infected persons is considered standard of care in British Columbia, many physicians do not routinely test their patients because they perceive them to be at low-risk. Routine screening involves only antibody testing, w i t h P C R testing only i n exceptional circumstances.  O u r data suggest a much higher  prevalence of H C V than previously expected, and underscore the importance of testing all HIV-infected i n d i v i d u a l s for the presence of H C V antibodies and R N A prior to initiating antiretroviral therapy.  This w o u l d enable appropriate therapeutic and  lifestyle management choices to be made, ultimately maximizing both H I V and H C V treatment options.  Patients w h o have no evidence of H C V antibodies or R N A at  baseline, should be re-tested following A R T initiation so that the optimum medical management of these patients can occur.  In summary, H C V co-infection is an important and clinically very relevant co-morbidity in H I V . It is w i d e l y prevalent among HIV-infected populations, and may be higher than expected because of its infectiousness. There are numerous options for the H I V and H C V treatment and management of these patients, but i n order to effect these choices, an accurate and timely diagnosis must be made, recognizing that antiretroviral therapy and its subsequent effects on immune reconstitution may alter that diagnosis.  85  4.6  1.  REFERENCES  Lavanchy, D . ,  Global surveillance and control of hepatitis C.  Journal  of  Viral  H e p a t i t i s , 1999. 6: p . 35-47. 2.  Anonymous,  National Institutes of Health Consensus Development Conference  Statement: Management of hepatitis C 2002 (June 10-12, 2002).  Gastroenterology.,  2002.123(6): p . 2 0 8 2 - 9 9 . 3.  Health Canada.,  Epidemiology of the transmission of blood-borne pathogens. 1 9 9 7 ,  Laboratory Centre for Disease Control. 4.  AIDS epidemic update.  UNAIDS., HIV/AIDS:  5.  Health  2002, Joint  United  Nations Programme  on  Geneva.  Canada.,  HIV/AIDS  Epi Update.  2000, L a b o r a t o r y C e n t r e f o r D i s e a s e  Control. 6.  Sherman,  K . E . , et a l . ,  Hepatitis C Virus prevalence among patients infected with  Human Immunodeficiency  Virus: a cross-sectional analysis of the US adult AIDS  Clinical Trials Group. C l i n i c a l 7.  P a t r i c k , D . M . , et a l . ,  I n f e c t i o u s D i s e a s e s , 2 0 0 2 . 34(6): p . 831-7.  Incidence of hepatitis C virus infection among injection drug users  during an outbreak of HIV infection. C m a j , 8.  Bonacini, M . a n d M . Puoti,  2001.165(7): p . 889-95.  Hepatitis C in patients with human immunodeficiency  virus infection: diagnosis, natural history, meta-analysis of sexual and vertical transmission, and therapeutic issues. 9.  S t r a t h d e e , S., et a l . ,  Needle Exchange is not Enough: lessons from the Vancouver  Injecting Drug Users Study. A I D S , 10.  M i l l e r , C . L . , et a l ,  A r c h I n t e r n M e d , 2000.160(22): p . 3365-73.  1997.11(8): p . F 5 9 - 6 5 .  The Future Face of Coinfection: Prevalence and Incidence of HIV  and Hepatitis C Virus Coinfection Among Young Injection Drug Users.  J Acquir  I m m u n e D e f i c S y n d r , 2004. 36(2): p . 743-749.  86  11.  B i c a , I., et a l . ,  Increasing Mortality Due to End-Stage Liver Disease in Patients with  Human Immunodeficiency Virus Infection.  C l i n i c a l I n f e c t i o u s D i s e a s e s ( C I D ) , 2001.  3 2 ( F e b 1): p . 4 9 2 - 4 9 7 . 12.  C a c o u b , P . , et a l . , Mortality  among human immunodeficiency virus-infected patients  with cirrhosis or hepatocellular carcinoma due to hepatitis C virus in French departments, of internal medicine/infectious  diseases, in 1995 and 1997.  Clinical Infectious  D i s e a s e s , 2 0 0 1 . 32: p . 1207-1214. 13.  M o n g a , H . K . , et a l . , Hepatitis  C virus infection-related morbidity and mortality among  patients with human immunodeficiency virus infection.  C l i n I n f e c t D i s , 2 0 0 1 . 33(2): p .  240-7. 14.  15.  G e o r g e , S . L . , et a l . ,  Hepatitis C virus viremia in HIV-infected individuals with negative  HCV antibody tests.  J A c q u i r I m m u n e D e f i c S y n d r , 2002. 31(2): p . 154-62.  B r a i t s t e i n , P . , et a l .  Dangerous Oversights: A comparison of HCV-RNA  HCV-antibody  testing among HIV-infected individuals, i n Canadian Association for  HIV Research.  2001. T o r o n t o ,  Canada.  Chronic Hepatitis C Virus Infection. J A M A ,  16.  F l a m m , S.,  17.  V i l l a n o , S . A . , et a l . ,  Mehta,  2 0 0 3 . 289(18): p . 2413-2417.  Persistence of viremia and the importance of long-term follow-up  after acute hepatitis C infection. H e p a t o l o g y , 18.  testing vs.  S . H . , et a l . ,  1999. 29(3): p . 9 0 8 - 1 4 .  Protection against persistence of hepatitis C.  Lancet,  2002.  359(9316): p . 1478-83. 19.  P i a s e c k i , B . , et a l . , Influence  of alcohol use, race, and viral coinfections on spontaneous  HCV clearance in a US Veteran population. H e p a t o l o g y , 20.  F i a l a i r e , P . , et a l . ,  2 0 0 4 . 40: p . 892-899.  Sustained disappearance of hepatitis C viremia in patients receiving  protease inhibitor treatment for human immunodeficiency virus infection.  Journal of  I n f e c t i o u s D i s e a s e s . , 1 9 9 9 . 1 8 0 ( 2 ) : p . 574-5.  87  21.  Yokozaki,  S., et a l . ,  Immunologic dynamics in hemophiliac patients infected with  hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. B l o o d , 2 0 0 0 . 96(13): p . 4293-9. 22.  Cooper, C . and D . Cameron,  Interpretation of undetectable hepatitis C virus RNA  levels in HYV-hepatitis C virus co-infection. 23.  T o r r i a n i , F . J . , et a l . ,  Peginterferon Alfa-2a plus ribavirin for chronic hepatitis C virus  infection in HIV-infected patients. 24.  Braitstein,  P . , et a l . ,  Antiretroviral  A I D S , 2004.18(2).  N E n g l J M e d , 2004. 351(5): p . 438-50.  Special Considerations in the Initiation and Management of  Therapy Among HIV and Hepatitis C Co-Infected Individuals.  AIDS,  2004. i n p r e s s . 25.  C a r p e n t e r , C , et a l . , Antiretroviral  therapy for HIV infection in 1996. J A M A ,  1996.  Therapy for HIV Infection in 1997. J A M A ,  1997.  276: p . 146-154. 26.  C a r p e n t e r , C , et a l . , Antiretroviral 277: p . 1962-1969.  27.  C a r p e n t e r , C , et a l ,  Antiretroviral therapy in adults. Updated recommendations of the  International AIDS Society-USA Panel. 28.  Y e n i , P . G . , et a l . ,  J A M A , 2 0 0 0 . 283(3): p . 3 8 1 - 9 1 .  Antiretroviral treatment for adult HIV infection in 2002: updated  recommendations of the International AIDS Society-USA Panel. J a m a ,  2002. 288(2): p .  222-35. 29.  Krajden, M . ,  Hepatitis C virus diagnosis and testing.  C a n a d i a n Journal of Public  H e a l t h . R e v u e C a n a d i e n n e d e S a n t e P u b l i q u e , 2 0 0 0 . 9 1 ( S u p p l 1): p . S 3 4 - 9 , S36-42. 30.  Sulkowski,  M.S. and D . L . Thomas,  Hepatitis C in the HIV-infected patient.  Clin  L i v e r D i s , 2 0 0 3 . 7(1): p . 179-94. 31.  Rockstroh,  J.K. a n d U . Spengler,  HIV and hepatitis C virus co-infection.  Lancet  I n f e c t D i s , 2004. 4(7): p . 437-44.  88  32.  W o o d , E . , et a l . , Prevalence and  correlates of untreated human immunodeficiency virus  type 1 infection among persons who have died in the era of modern antiretroviral therapy. J o u r n a l o f I n f e c t i o u s D i s e a s e s , 2 0 0 3 . 1 8 8 : p . 1164-70. 33.  A l t e r , M . , et a l . ,  through 1994. 34.  The prevalence of hepatitis C virus infection in the United States, 1988  N e w E n g l a n d J o u r n a l o f M e d i c i n e , 1999. 341: p . 556-62.  H a l l , C , et a l . ,  Hepatitis C Virus Infection in San Francisco's HIV-infected Urban  Poor. High Prevalence but Low Treatment Rates.  J o u r n a l of G e n e r a l  Internal  M e d i c i n e , 2 0 0 4 . 1 9 ( 4 ) : p . 357-365. 35.  F u r i o n e , M . , et a l . ,  Dissociation of serum and liver hepatitis C virus RNA levels in  patients coinfected with human immunodeficiency virus and treated with antiretroviral drugs. J C l i n 36.  M i c r o b i o l , 2004. 42(7): p . 3012-6.  L a s k u s , T . , et a l . ,  Human immunodeficiency virus facilitates infection/replication of  hepatitis C virus in native human macrophages. 37.  B l a c k a r d , J . , et a l . ,  B l o o d , 2 0 0 4 . 1 0 3 ( 1 0 ) : p . 3854-9.  Hepatitis C Virus (HCV) Diversity in  Subjects Initiating Highly Active Antiretroviral  Therapy.  HIV-HCV-Co-Infected J o u r n a l of Infectious  D i s e a s e s , 2 0 0 4 . 1 8 9 ( 1 5 A p r ) : p . 1472-81.  89  Table 4.1. Baseline factors associated with H C V seroprevalence among 1186 HIVinfected individuals initiating antiretroviral therapy  HCV-positive n=606  HCV-negative n=580  p-value  Male sex  473(78%)  541 (93%)  <0.001  Any IDU*  285 (47%)  36(6%)  <0.001  37.8 (32.2-44.0)  36.8 (32.0-43.8)  0.505  65(11%)  87(15%)  0.028  280(130-430)  270 (130-420)  0.562  19 (11-27)  16(9-24)  <0.001  Log HIV R N A (median, IQR)  5.0(4.6-5.0)  5.0(4.6-5.0)  0.994  Started with a PI  417(69%)  388 (67%)  0.480  Median age (IQR**) AIDS diagnosis Absolute CD4 (median, IQR) CD4 Fraction (%) 10  * I D U = injection drug use ** IQR = interquartile range  90  T a b l e 4.2.  L o g i s t i c regression analysis of factors associated w i t h positive a n t i b o d y  b u t n e g a t i v e H C V R N A r e s u l t s at b a s e l i n e (n=179)  Unadjusted Odds  p-value  (95% C I ) M a l e sex  0.88  1.13  0.876  0.43  (yes v s . n o )  (0.30 - 0.62)  A I D S diagnosis (yesvs.no) Absolute C D 4 ( p e r 100 c e l l s ) Log  1 0  H I VR N A  2.24  0.257  <0.001  0.93  1.10  0.385  0.45  <0.001  (0.31 - 0.65) 0.003  1.86  0.033  (1.05-3.28) 0.001  (0.79 - 0.94)  (0.72-1.20)  0.201  (0.89-1.37)  (1.33-3.79) 0.87  0.75 (0.48-1.17)  (0.92-1.34) Any IDU  p-value  (95% C I )  (0.58-1.33) A g e ( p e r 10 y r s )  Adjusted Odds  0.89  0.018  (0.81 - 0.98) 0.594  0.85  0.214  (0.65-1.10)  91  T a b l e 4. 3. L o g i s t i c r e g r e s s i o n a n a l y s i s o f f a c t o r s a s s o c i a t e d w i t h u n d e t e c t a b l e H C V R N A at b a s e l i n e b u t d e t e c t a b l e H C V R N A 6 - 1 2 m o n t h s p o s t - A R T i n i t i a t i o n a m o n g H I V - i n f e c t e d p a t i e n t s (n=118)  Unadjusted Odds  p-value  0.55  0.239  (0.21 - 1 . 4 6 ) A g e ( p e r 10 y r s )  0.62  (yes v s . n o ) AIDS  diagnosis  (yes v s . n o ) Absolute C D 4 ( p e r 100 c e l l s ) Log  1 0  H I VR N A  0.55  0.093  0.239  0.505  2.44  0.60  0.132  1.89  0.246  0.26  0.086  (0.06-1.21) 0.137  0.77  0.144  (0.54-1.09)  (0.62-1.07)  (1.09-5.50)  0.246  (0.65 - 5.50)  (0.17-2.39) 0.81  0.50  (0.31 - 1 . 1 7 )  (0.21 - 1 . 4 6 ) 0.64  p-value  (0.15.-1.66)  (0.35-1.08) Any IDU  Odds  (95% C I )  (95% C I ) M a l e sex  Adjusted  0.031  2.80  0.034  (1.08-7.24)  92  Figure 4.1 Summary of Study Design and Results  S t o r e d baseline p l a s m a samples ( H O M E R ) n=1388 S a m p l e s available and sufficient v o l u m e for testing n=1257  H C V antibody testing ( A b b o t t p r i m a r y screen) n=1257 Results available n = l 186  H C V - A b pos N=606  P C R testing o n all available samples n= 605 P C R - p o s : 425 (70%) P C R - n e g : 179 (30%)  TJ  H C V A b neg N=580  P C R testing o n r a n d o m 10% n=58 P C R - p o s : 1 (2%) P C R - n e g : 56 (97%)  6-12 months post-treatment: N = 1 1 8 A b + / P C R at'-aseline with an available & sufficient sample P C R - p o s : 24 (20%) P C R - n e g : 94 (80%)  CHAPTER 5: RESPONSE TO ANTIRETROVIRAL THERAPY AMONG H I V / H C V CO-LNFECTED ADULTS 5.1 FORWARD This chapter is currently under review w i t h the Journal of Infectious Disease as: Braitstein P, Zala C, Yip B, Moore D , H o g g RS, Montaner.  "Attenuated Absolute but  Preserved C D 4 Fraction Response to Antiretroviral Therapy A m o n g H I V / H C V C o Infected Adults i n a Population-Based Antiretroviral Treatment Program'.  94  5.2 I N T R O D U C T I O N  The  use  of  Highly  Active  Antiretroviral  Therapy  (HAART)  has  led  to  dramatic  r e d u c t i o n s i n H I V - r e l a t e d m o r b i d i t y a n d m o r t a l i t y i n H I V - i n f e c t e d i n d i v i d u a l s [1-3]. However,  parallel  morbidities,  such  with as  a p p r o x i m a t e l y 30%  this  viral  of  success  hepatitis.  has  been  The  hepatitis  HIV-positive people  the  subsequent C  i n the  virus  emergence  (HCV)  Developed  is  World,  of  co-  present  and  in  in  some  p o p u l a t i o n s s u c h as i n j e c t i o n d r u g u s e r s a n d h e m o p h i l i a c s , n e a r l y a l l t h o s e w h o a r e H I V i n f e c t e d a r e c o - i n f e c t e d w i t h H C V [4-7].  Immunologic  and  virologic  response  to  antiretroviral therapy  populations have been previously examined.  in H C V  co-infected  M o s t s t u d i e s h a v e f o u n d t h a t t h e r e is n o  i m p a c t o f H C V o n t h e H I V v i r o l o g i c r e s p o n s e t o a n t i r e t r o v i r a l s [8-11].  However,  there  are conflicting r e p o r t s i n the literature r e g a r d i n g i m m u n o l o g i c response.  G r e u b et a l .  found  response  that H C V infection  independently  predicted  a  poorer  immune  a n t i r e t r o v i r a l t h e r a p y , d e f i n e d as t i m e t o a n i n c r e a s e o f at l e a s t 50 c e l l s / m m 3 contrast, several others  have  f o u n d little o r n o  difference  i n C D 4 increase  to  [8].  In  by H C V  serostatus [9,11,12].  The  data  inconsistent  regarding for  immunologic  several  reasons.  response It c a n t a k e  to  antiretroviral  u p to  24  months  treatment of  treatment  c o m p l e t e C D 4 r e s p o n s e to o c c u r [13], b e y o n d t h e t i m e f r a m e o f m a n y s t u d i e s . CD4  increase  become  sequestered  consistency [14,15].  is m e a s u r e d  a n d defined  secondary  to  is a n o t h e r  H C V  key  infection,  b e t w e e n , a n d i n t e r p r e t a b i l i t y of, a b s o l u t e  factor.  Indeed,  potentially  may  a n d fraction T-cell  for  a  How a  T-cells  impacting  be  on  may the  populations  I m m u n e d y s r e g u l a t i o n f r o m b o t h H I V a n d H C V i n f e c t i o n s m a y also l e a d to a n  impaired i m m u n o l o g i c response  primarily among  those w i t h  a lower  baseline  CD4 95  count. For example, i n two studies that found no difference i n C D 4 response, the baseline C D 4 was well above 200 among both HCV-positive and HCV-negative subjects  [9,11],  whereas the baseline HCV-positive C D 4 count i n the Swiss Cohort was  172 cells [8].  The question of immunologic response to antiretrovirals among those H I V / H C V coinfected is critical because of its implications for when to start antiretroviral treatment in these patients. HCV-related hepatic fibrosis progression i n HIV-infected individuals is strongly associated w i t h a weakened immune system [16], and delaying antiretrovirals for too long may result i n more rapid progression of H C V disease - particularly if patients' immunologic recovery will only be partial. Although treatments for hepatitis C are becoming more effective, unfavorable genotype [17, 18], and concomitant anemia, depression  [16, 19],  and H I V infection  [20-22]  mean that only between  20-30%  of H I V / H C V co-infected patients can expect to successfully complete H C V treatments. Antiretroviral therapy is thus paramount to the health management of co-infected individuals.  Therefore the primary objective of this study was to examine immunologic response to the first 48 weeks of antiretroviral treatment among previously antiretroviral naive HIV/HCV  co-infected and H I V mono-infected individuals i n a population-based  H I V / A I D S drug treatment program, examining both absolute C D 4 and C D 4 fraction outcomes. O u r secondary objective was to describe the contribution of baseline C D 4 count to this response.  96  5.3  METHODS  Data Source: The British Columbia HIV/AIDS Drug Treatment Program Antiretroviral medications have been centrally distributed at no cost to eligible H I V infected individuals since 1986. In October 1992, the distribution of antiretroviral agents became the responsibility of the H I V / A I D S D r u g Treatment Program of the British Columbia Center for Excellence in H I V / A I D S .  This antiretroviral drug distribution  program remains the only free source of antiretroviral medication in this Canadian province (and is a unique program in Canada).  The Centre's  HIV/AIDS Drug  Treatment program has received ethical approval from the University of British Columbia Ethics Review Committee at its St. Paul's Hospital site, and the program conforms w i t h the province's Freedom of Information and Protection of Privacy Act.  The Center distributes antiretroviral drugs based on specific guidelines generated b y the Therapeutic Guidelines Committee [23]. These guidelines have been and continue to be consistent w i t h those treatment guidelines published by the International AIDS Society [13, 24-26]. The Centre's guidelines recommend that C D 4 and plasma HIV-1 R N A levels be monitored at baseline, at 4 weeks after starting antiretroviral therapy and every three months thereafter.  Plasma viral loads are measured  using the  Amplicor HIV-1 Monitor™ (Roche Diagnostics Branchburg, NJ). A l l classes of federally licensed antiretroviral drugs are currently available through the program,  including all nucleoside  nucleoside reverse transcriptase  reverse  transcriptase  inhibitors (NRTI),  non-  inhibitors (NNRTI), and protease inhibitors (PI).  Tenofovir, Atazanavir, and Enfuvirtide are also available. Eligibility for antiretroviral medication has remained consistent w i t h current international recommendations [13].  97  Individuals are automatically entered into the D T P when they are first prescribed any antiretroviral agent.  A t D T P entry and with each subsequent physician visit, the  participant's history (if any) of antiretroviral use, C D 4 cell count, and plasma viral load are recorded. Typically, patients are followed-up at 3-month intervals, at which time prescriptions are renewed or altered based on treatment success and other clinical factors.  Blood d r a w n for the purposes of C D 4 and viral load testing is stored for each  individual at each follow-up visit for future research activities related to H I V disease.  Study Population The H A A R T Observational Medical Evaluation and Research (HOMER) study is a nested cohort within the B C Center  for Excellence's H I V / A I D S D r u g Treatment  Program. It includes all previously A R T naive individuals who initiated antiretroviral therapy with a triple-drug regimen consisting of either two  nucleoside  reverse  transcriptase inhibitors (NRTI) plus either a protease inhibitor (PI), or a non-nucleoside reverse transcriptase inhibitors (NNRTI), between August, 1996 and July, 2000. The data used in these analyses are based on individuals from the H O M E R Cohort, for w h o m there was H C V serological data.  Outcome Measures The primary outcome measure for this analysis was C D 4 response, measured using both absolute C D 4 counts and CD4's as a percentage of the total lymphocyte count (referred to throughout this manuscript as C D 4 fraction), comparing H I V monoinfected individuals to H I V / H C V  co-infected individuals.  Events were  defined,  respectively, as achieving an absolute C D 4 increase of >75 cells or a C D 4 fraction increase of at least 10% within the first 48 weeks of treatment.  Both events were 98  defined a priori as clinically significant changes. Secondly, we examined both absolute and C D 4 fraction measures as continuous outcomes throughout the first 48 weeks of A R T . Participants were only included i n the repeated measures analysis if they had at least three C D 4 (absolute and fraction) and plasma viral load results during the 48 weeks, including a baseline measure.  Baseline was defined as the most recent CD4 test taken within 180 days prior to starting antiretrovirals.  Baseline characteristics examined were gender, age, baseline C D 4  (absolute and percent) and H I V l o g plasma viral load (pVL), and whether patients had 10  an AIDS diagnosis. These were analyzed comparing HCV-antibody positive individuals to H C V antibody-negative individuals.  Our definition of adherence to antiretroviral medication is based on the proportion of time that medication dispensed would last over the first year of follow-up.  This  calculation was restricted to each patient's first year on therapy to avoid the reverse causation that could occur among patients who cease antiretroviral therapy after they have become too sick to take medication. We have previously demonstrated how this estimate strongly predicts virologic response [27] and survival [28, 29].  For the  purposes of our analyses, adherence was a binary variable where 1 was being greater than 95% adherent (meaning prescriptions dispensed w o u l d last throughout the first year of follow-up), and 0 being less than 95% adherent.  Statistical Analysis  99  H C V positive and H C V negative participants were compared using both parametric and distribution-free methods. Categorical data were analyzed using Pearson's % test. 2  Fisher's exact test was used for contingency tables in which 25% or more of the expected cell frequencies were less than 5. Continuous variables were analyzed using the Wilcoxon rank-sum test. Kaplan-Meier analysis was used to calculate cumulative rates of a C D 4 increase of >75 cells or a C D 4 fraction increase of at least 10%.  In  secondary analyses, we stratified the population by baseline C D 4 count (<200 cells, <350 cells, > 350 cells). Cox proportional hazards regression was used to calculate univariate and adjusted relative hazards and 95% confidence intervals (CI's). Possible confounders adjusted for were gender  (male vs. female), age at baseline (continuous), AIDS  diagnosis at baseline (yes vs. no), baseline C D 4 and loglO-transformed plasma HIV-1 R N A levels (continuous), and adherence to antiretroviral therapy in the first year of treatment (>95% vs. <95%). The assumption of proportional hazards was validated b y inspection of log (-log{survival function))estimates against log time plots.  Multivariate mixed effects models were constructed modeling C D 4 slope (absolute and percent) over the first 48 weeks of A R T , primarily for the whole population and secondarily stratified by baseline C D 4 count (<200 cells, <350 cells, > 350 cells). Independent fixed effects examined included H C V antibody status (positive vs. negative), time (months), gender  (male vs. female), age at baseline, and loglO-  transformed plasma HIV-1 R N A levels (continuous) over time. The only random effect was patient. Correlation between repeated measures was assumed constant. The data were analyzed using SAS software, Version 6.12. A l l reported p values are two-tailed.  5.4 R E S U L T S  100  There were 1186 patients available for this analysis, including 606 (51%) who were H C V antibody positive, and 580 (49%) who were H C V antibody negative. characteristics  comparing  HCV-positive  versus  HCV-negative  Baseline  individuals  are  summarized i n Table 4.1. (page 90). There were no statistical differences at baseline in their gender, median age, absolute C D 4 count, log plasma H I V R N A , or type of A R T initiated. However, HCV-positive individuals were less likely to have had an AIDS diagnosis at baseline (11% vs. 15%, p=0.028), and i n spite of the absolute C D 4 count, had a significantly higher baseline CD4 percent (19% vs. 16%, p<0.001).  Figure 5.1 displays the results of the Kaplan-Meier analysis examining the effect of H C V on time to C D 4 absolute (p<0.001) and fraction events (p=0.021). W h e n stratified b y baseline C D 4 count, consistent associations were found with each respective measure, (Absolute event: baseline C D 4 <200 cells/mm3: p<0.001, <350: p<0.001, and >350: p=0.006; Fraction event: baseline C D 4 <200: p=0.408, C D 4 <350: p=0.703, C D 4 >350: p=0.001).  Adherence, as expected, had a significant on the outcomes. After restricting the KaplanMeier analysis to those >95% adherent, H C V has a much lesser but  nonetheless  significant effect on absolute C D 4 response (p=0.024), but no longer had an effect o n C D 4 fraction response (p=0.729).  These results again remained consistent across the  baseline C D 4 strata (absolute C D 4 response: baseline C D 4 <200: p=0.301, C D 4 <350: p=0. 0.037, C D 4 >350: p=0.341; fraction C D 4 response: baseline C D 4 <200: p=0.613, C D 4 <350: p=0. 0.306, C D 4 >350: p=0.070).  As summarized in Tables 5.1.1. and 5.1.2., in multivariate C o x Regression models, controlling for age at baseline, gender, having an AIDS diagnosis at baseline, baseline 101  C D 4 and plasma log viral load, as well as adherence over the first 12 months of ART, H C V seropositivity remained predictive of being less likely to achieve >75 cells (overall Adjusted H a z a r d Ratio, A H R : 0.84, 95%CI: 0.72-0.97, p=0.021), but there was no effect of H C V on the probability of achieving a C D 4 fraction increase of >10% ( A H R : 0.89, 95%CI: 0.70-1.14, p=0.369). The effect of baseline C D 4 on these estimates was limited (absolute outcome: baseline C D 4 <200 A H R : 0.81, 95%CI: 0.65-1.00; C D 4 <350 A H R : 0.82, 95%CI: 0.67-0.99, p=0.036; C D 4 >350 A H R : 0.86, 95%CI: 0.67-1.10, p=0.248; fraction outcome: baseline C D 4 <200 A H R : 0.68, 95%CI: 0.46-1.01), p=0.129; C D 4 <350 A H R : 0.77, 95%CI: 0.56-1.05, p=0.056; C D 4 >350 A H R : 0.97, 95%CI: 0.62-1.50, p=0.874).  The effects of examining both absolute and fraction outcomes as continuous measures are summarized i n Table 5.2 and 5.3, and Figure 5.2.  The median (IQR, interquartile  range), unadjusted, absolute C D 4 change among the H C V co-infected was 60 (-10, 170) compared to 140 (50, 230) among the non-co-infected (p<0.001).  This significant  contrast remained across the three baseline C D 4 strata (Table 5.3). Similarly, while the H C V co-infected gained a median C D 4 fraction of 4% (-1, 8) over 48 weeks, the H C V negative group gained a median of 6% (2, 9) (p<0.001), and this difference  was  consistent regardless of baseline C D 4 (Table 5.3).  Figures 5.2 displays the effect of H C V serostatus on adjusted C D 4 change over 48 weeks of A R T using the mixed effects analyses, with the actual model results provided in Table 5.2.  In the absolute C D 4 analysis, controlling for time (months), age at  baseline, gender, having an AIDS diagnosis at baseline, and plasma log viral load response (as a repeated measure) over the 48 week period, although both H C V positive and HCV-negative individuals had statistically significant increases over time, the H C V co-infected individuals only gained an average of 17 cells, compared to the 85 102  cells that the non-co-infected gained over 48 weeks of treatment. Even after restricting the analysis to those individuals w h o were greater than 95% adherent, those who were H C V positive on average gained 40 cells, compared to the 105 cells among the H C V negative. This strong difference remained across the baseline C D 4 strata (Table 5.3).  The multivariate mixed effect C D 4 fraction analysis shows similar results, whereby the H C V co-infected saw an average increase of only 0.96%, compared to the H C V negatives i n w h o m their average C D 4 percent increased by 2.5%. Both increases were statistically significant. After restricting to those w h o were at least 95% adherent, where the HCV-positive individuals gained on average 2.5%, the HCV-negative gained 3.4%. However, the H C V positive individuals have a statistically significant higher baseline CD4 fraction, while there was no difference in the baseline absolute C D 4 between the two groups.  5.5  DISCUSSION A N D CONCLUSIONS  Our results suggest  that H C V has a strong influence on the C D 4 response  antiretroviral therapy over the first 48 weeks of treatment.  to  Baseline C D 4 count does  not significantly alter the magnitude of response in either absolute or fraction analysis. However, it must be noted that, as can be seen from Table 4.1, the baseline fraction of H C V co-infected individuals in our cohort is significantly higher than the non-coinfected (p<0.001).  This results i n a much higher starting point i n the mixed effects  analyses, and in spite of the smaller average gain, the H C V co-infected still have a higher average C D 4 fraction than the HCV-negative by the end of 48 weeks of treatment.  This is also probably w h y in adjusted Cox analysis, H C V had no effect on 103  the time to a C D 4 fraction increase of at least 10% of the C D 4 percent, but did negatively affect the time to an absolute C D 4 event (>75 cells).  Our findings do support those of others [8, 30]. A n Italian cohort used linear regression and found that HCV-positive (HbsAg-) patients had 34 cells fewer than HCV-negative patients, and in C o x multiple regression found a reduced time to achieving a C D 4 increase of 100 cells or else reaching the 500 cell/mm3 threshold (p=0.01) [30]. These results are quite similar to Greub et al.'s, although they defined their event as an increase of at least 50 cells [8]. Zala et al. (2004) have also reported a blunted C D 4 response using mixed effects measures in a nested cohort of several clinical trials populations [32]. Our data are in contrast to those studies that found little or no impact of H C V on C D 4 response to treatment [9-12, 33]. In a university-based cohort at Johns Hopkins Hospital H I V Clinic, Sulkowski et al. (2002) found no difference in the proportion of H C V co-infected vs. non- in gaining either 50 or 100 cells, after 1, 2, and 3 years of A R T . Their cohort also found that H C V co-infected individuals had a statistically much higher C D 4 fraction at baseline than the HCV-negative patients, but found no evidence of altered effects of H C V on either absolute C D 4 or the fraction [11]. A recent study from the Thai H I V - N A T Cohort found that mean increases were significantly lower at week 4, but by week 48 the difference disappeared.  They also  found that H C V co-infection was not associated with a hazard rate increase of at least 100 cells [34].  Our analyses have several advantages and strengths. First, our sample is drawn from a substantially sized population-based program, making it more generalizable than clinicbased studies, cohort data, or clinical trials populations.  Secondly, we were able to  account for the confounding effect of adherence, both through our adherence measure, 104  as well as controlling for log viral load response over time i n mixed effects analysis. Third, by using both repeated measures and survival analyses, and both absolute and fraction outcomes, we were able to elucidate some of the complexities in analyzing immunologic response to treatment in H I V / H C V co-infection.  The results of this study could be affected b y several potential limitations. One, our measure of adherence is a proxy measure; however, we have previously shown this measure to be highly predictive of both virologic response [27] and survival [28, 29]. Two, the number of analyses conducted may predispose to a type 1 error occurring. However,  the  data  are  relatively consistent  across  baseline  C D 4 strata,  and  absolute/fraction outcomes respectively, so we feel confident the effects of this occurring are limited. Three, our center has previously shown that individuals who are not treated for H I V infection in spite of treatment eligibility are substantially different sociodemographically from those who receive treatment, and they are more likely to be co-infected with H C V [35].  However, this bias may only be diluting the effect  because increasing numbers of H C V co-infected patients w o u l d increase our power to detect differences.  The damage caused by chronic H C V infection is not believed to be the consequence of direct destruction of hepatic cells by the virus, but from an intermediate immune response that is large enough to induce hepatic cell destruction and fibrosis but not enough to eradicate the virus from its reservoirs [21]. A blunted immune response in H I V / H C V co-infected individuals may be due to the non-specific immune stimulation driven by chronic H C V infection, or it may be that infection of immune cells by H C V could favour C D 4 T-cell depletion [36].  Our data support the hypothesis that H C V is  related to a smaller C D 4 recovery in both absolute and fraction terms.  Our data also  105  raise the potentially significant issue of C D 4 sequestration among HCV-co-infected patients. They indicate the importance for physicians to measure both absolute and fraction C D 4 measures among their H I V / H C V co-infected patients, and suggest that all future analyses related to hepatitis C and immunologic function should examine both absolute and C D 4 fraction outcomes.  In conclusion, immunologic response to ART among H I V / H C V co-infected adults is a complex issue. Our data suggest that immunologic indicators, in this case absolute C D 4 and C D 4 fraction, may be strongly affected by the presence of HCV. I n our population, while there was no difference between baseline absolute C D 4 counts, there was a marked difference in absolute C D 4 response to ART. I n contrast, while there was a significant difference at baseline in C D 4 fraction depending on H C V serostatus, the effect of H C V on fraction immune response was much more limited. Baseline C D 4 count appears, in our study, to have had minimal effect.  106  5.6  1.  REFERENCES  Hogg, R., et al., Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. C M A J : Canadian Medical Association Journal, 1999.160(5): p. 659-65.  2.  Hogg, R., et al., Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. J A M A , 1998. 279(6): p. 450-4.  3.  Palella, F.J., Jr., et al., Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. NewEngland Journal of Medicine., 1998. 338(13): p. 853-60.  4.  Hall, C , et al., Hepatitis C Virus Infection in San Francisco's HIV-infected Urban Poor. High Prevalence but Low Treatment Rates. Journal of General Internal Medicine, 2004.19(4): p. 357-365.  5.  Patrick, D . M . , et al., Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001.165(7): p. 889-95.  6.  Soriano, V . , et al., Care of patients with chronic hepatitis C and HIV co-infection: recommendations from the HIV-HCV  7.  International Panel. A i d s , 2002.16(6): p. 813-28.  Yokozaki, S., et a l , Immunologic dynamics in hemophiliac patients infected with hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. Blood, 2000. 96(13): p. 4293-9.  8.  Greub, G , et al., Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5.  9.  Klein, M . , R. Lalonde, and S. Suissa, The Impact of Hepatitis C Virus Coinfection on HIV Progression Before and After Highly Active Antiretroviral Therapy. JAIDS, 2003. 33: p. 365-72.  107  10.  Melvin, D., et al., The impact of co-infection with hepatitis C virus and HIV on the tolerability of antiretroviral therapy. AIDS, 2000.14(4): p. 463-465.  11.  Sulkowski, M.S., et al., Hepatitis C and progression of HIV disease. Jama, 2002. 288(2): p. 199-206.  12.  C h u n g , R., et al., Immune recovery is associated with persistent rise in hepatitis C virus RNA, infrequent liver flares, and is not impaired by hepatitis C virus in coinfected subjects. AIDS, 2002.16(14): p. 1915-23.  13.  Yeni, P.G., et al., Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society-USA Panel. Jama, 2002. 288(2): p. 222-35.  14.  Hilleman, M.R., Strategies and mechanisms for host and pathogen survival in acute and persistent viral infections. Proc Natl Acad Sci U S A , 2004.  15.  Neau, D., et al., T-lymphocyte populations in hepatitis C and HIV co-infected patients treated with interferon-alfa-2a and ribavirin. H I V M e d , 2003. 4(2): p. 120-6.  16.  Soriano, V . , et al., Care of patients with hepatitis C and HIV co-infection. AIDS, 2004. 18(1): p. 1-12.  17.  Perez-Olmeda, M . , et al., Virological characteristics of hepatitis C virus infection in HIV-infected individuals with chronic hepatitis C: implications for treatment. Aids, 2002.16(3): p. 493-5.  18.  Serfaty, L . , et al., Impact of early-untreated HIV infection on chronic hepatitis C in intravenous drug users: a case-control study. AIDS, 2001.15: p. 2011-2016.  19.  Sulkowski, M.S. and D . L . Thomas, Hepatitis C in the HIV-infected patient. C l i n Liver Dis, 2003. 7(1): p. 179-94.  20.  Torriani, F.J., et a l , Peginterferon Alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J M e d , 2004. 351(5): p. 438-50.  21.  Poynard, T., et al., Viral hepatitis C. Lancet, 2003. 362(9401): p. 2095-100.  108  22.  Flamm, S., Chronic Hepatitis C Virus Infection. J A M A , 2003. 289(18): p. 2413-2417.  23.  H I V / A I D S , B.C. Centre for Excellence in H I V / A I D S . , Therapeutic Guidelines for the Treatment of HIV/AIDS and Related Conditions. 1999, 2001: Vancouver.  24.  Carpenter, C , et al., Antiretroviral therapy for HIV infection in 1996. J A M A , 1996. 276: p. 146-154.  25.  Carpenter, C , et al., Antiretroviral Therapy for HIV Infection in 1997. J A M A , 1997. 277: p. 1962-1969.  26.  Carpenter, C , et al., Antiretroviral therapy in adults. Updated recommendations of the International AIDS Society-USA Panel. J A M A , 2000. 283(3): p. 381-91.  27.  Low-Beer, S., et al., Adherence to triple therapy and viral load response. J Acquir Immune Defic Syndr, 2000. 23(4): p. 360-1.  28.  Wood, E., et al., 7s there a baseline CD4 cell count that precludes a survival response to modern antiretroviral therapy? Aids, 2003.17(5): p. 711-20.  29.  H o g g , R.S., et al., Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year of follow-up. A i d s , 2002.16(7): p. 1051-8.  30.  De Luca, A . , et al., Coinfection with hepatitis viruses and outcome of initial antiretroviral regimens in previously naive HIV-infected subjects. A r c h Intern M e d , 2002.162(18): p. 2125-32.  31.  Lincoln, D., K . Petoumenos, and G.J. Dore, HIV/HBV  and HIV/HCV  coinfection,  and outcomes following highly active antiretroviral therapy. H I V M e d , 2003. 4(3): p. 241-9. 32.  Zala, C , et al. The impact of the hepatitis C virus on CD4-T cell response post-initiation of HAART among patients enrolled in clinical trials, i n 11th Conference on Retroviruses and Opportunistic Infections. 2004. San Francisco, C A .  109  33.  Aceti, A . , et al., Hepatotoxicity development during antiretroviral therapy containing protease inhibitors in patients with HIV: the role of hepatitis B and C virus infection. J Acquir Immune Defic Syndr, 2002. 29(1): p. 41-8.  34.  Law, W.P., et al., Impact of viral hepatitis co-infection on response to antiretroviral therapy and HIV disease progression in the HIV-NAT  cohort. Aids, 2004. 18(8): p.  1169-77. 35.  Wood, E., et al., Prevalence and correlates of untreated human immunodeficiency virus type 1 infection among persons who have died in the era of modern antiretroviral therapy. Journal of Infectious Diseases, 2003.188: p. 1164-70.  36.  Soriano, V . , et al., Consensus conference on chronic viral hepatitis and HIV infection: updated Spanish recommendations. J Viral Hepat, 2004.11(1): p. 2-17.  110  Table 5.1.1. Unadjusted and adjusted factors associated w i t h an absolute C D 4 increase of at least 75 cells/mm3 Unadjusted Hazard 95% Adherent (yes vs. no)  2.15(1.84-2.51)  Baseline l o g H I V v i r a l load (per log) 1.26(1.07-1.49)  P-Value (95% CI)  Adjusted Hazard  P-Value (95% CI)  <0.001  1.99 (1.69-2.35)  <0.001  0.006  1.27(1.07-1.50)  0.006  H C V Serostatus (Positive vs. Negative)  0.68 (0.59-0.78)  <0.001  0.84(0.72-0.97)  0.021  Gender (male vs. female)  1.70(1.34-2.17)  0.594  0.88 (0.70-1.10)  0.248  Age (per 10 yr increase)  1.06 (0.98-1.15)  0.129  0.99 (0.91-1.07)  0.744  Baseline C D 4 (per 100 cells)  0.98(0.95-1.01)  0.266  0.99 (0.96-1.02)  0.476  A I D S Diagnosis at Baseline (yesvs.no) 1.06(0.86-1.31)  0.594  0.88 (0.70-1.10)  0.248  1 1 1  Table 5.1.2. Unadjusted and adjusted factors associated w i t h a C D 4 % increase of at least 1 0 % Unadjusted Hazard (95% CI)  P-Value  Adjusted Hazard ( 9 5 % CI)  <0.001  1.81 (1.32-2.49)  <0.001  <0.001  2.00 (1.51-2.64)  <0.001  H C V Serostatus 0.76 (0.61-0.96) (Positive vs. Negative)  0.021  0.89 (0.70-1.14)  0.369  Baseline C D 4 (per 100 cells)  1.03 (0.98-1.09)  0.194  1.05 (0.99-1.10)  0.099  A I D S Diagnosis at Baseline (yesvs.no) 0.85 (0.60-1.22)  0.379  0.81 (0.56-1.18)  0.280  1.16(0.80-1.70)  0.442  0.91 (0.61-1.36)  0.648  Age 0.95 (0.84-1.08) (per 10 yr increase)  0.437  0.92(0.81-1.05)  0.217  Baseline log H I V v i r a l load (per log) 1.72 (1.26-2.34) 9 5 % Adherent (yes vs. no)  Gender (male vs. female)  2.00(1.53-2.58)  P-Value  112  Table 5.2. M i x e d Effects M o d e l Results of Impact of H C V Co-infection on C D 4 increase Over 48 Weeks of A R T i n Both Cohorts, O v e r a l l and A m o n g Those >95% Adherent Effect  Estimate  A) Absolute C D 4 Increase Overall ( n=893) 577.31 Intercept H C V status at time 0 3.31 6.63 H C V - C D 4 increase per 4 wks -4.81 H C V + C D 4 increase per 4 wks -1.01 Age 31.16 Male -203.97 AIDS at baseline -50.69 LoglO V i r a l load over time  Std Error  P-Value  39.04 15.27 0.55 0.73 0.82 23.68 21.74 2.22  <0.001 0.828 <0.001 <0.001 0.219 0.188 <0.001 <0.001  B) Absolute C D 4 Increase A m o n g Those >95% Adherent 49.64 543.91 Intercept 19.45 -12.46 H C V status at time 0 0.62 8.93 H C V - increase/4 wks 0.89 -3.91 H C V + increase/4 w k s 0.99 -0.58 Age 34.56 22.57 Male 24.44 -206.12 AIDS at baseline 2.79 -44.33 LoglO Viral load over time  <0.001 0.522 <0.001 <0.001 0.560 0.514 <0.001 <0.001  C) C D 4 % Increase O v e r a l l Intercept H C V status at time 0 H C V - increase/4 wks H C V + increase/4 wks Age Male AIDS at baseline LoglO V i r a l load over time  1.76 0.69 0.02 0.03 0.04 1.11 0.99 0.08  <0.001 0.002 <0.001 <0.001 0.071 0.458 <0.001 <0.001  2.21 0.87 0.02 0.03 0.04 1.54 1.11 0.10  <0.001 0.017 <0.001 0.039 0.176 0.775 <0.001 <0.001  33.03 2.16 0.22 -0.14 -0.07 -0.82 -9.56 -2.39  D) C D 4 % Increase A m o n g Those >95% Adherent 31.78 Intercept H C V status at time 0 2.09 0.28 H C V - increase/4 wks -0.07 H C V + increase/4 wks -0.06 Age -0.44 Male -10.01 AIDS at baseline -2.27 LoglO V i r a l load over time  113  T a b l e 5.3. A b s o l u t e a n d F r a c t i o n C D 4 R e s p o n s e s t o I n i t i a t i o n o f A n t i r e t r o v i r a l T h e r a p y A m o n g a P o p u l a t i o n - B a s e d P r o g r a m of H C V - p o s i t i v e a n d H C V - n e g a t i v e I n d i v i d u a l s , Stratified b y B a s e l i n e C D 4 C o u n t  CD4 Response  Overall  Baseline CD4 <200  Baseline CD4 <350  Baseline CD4 >350  N=341  N=197  60 (0, 160) 140 (40, 210) <0.001  30 (-80, 170) 160 (60, 290) <0.001  37 cells (p<0.001) 84 cells (p<0.001)  -15 cells (p<0.001) 88 cells (p<0.001)  * Among >95% Adherent Only: Adjusted average number of CD4's gained over 48 weeks HCV+ 40 cells (p<0.001) 54 cells (p=0.022) 56 cells (p=0.001) HCV105 cells (p<0.001) 91 cells (p<0.001) 105 cells (p<0.001)  8 cells (p=0.002) 108 cells (p<0.001)  ABSOLUTE CD4 CHANGE N=538 N=222 Median (interquartile range) CD4 change over first 48 weeks (unadjusted) HCV+ 50 (-20, 160) 50 (0, 50) HCV140 (50, 240) 140 (50, 200) p-value <0.001 p=0.002 *QAdjusted average number of CD4's gained over 48 weeks HCV+ 17 cells (p<0.001) 40 cells (p=0.001) HCV85 cells (p<0.001) 80 cells (p<0.001)  CD4 FRACTION CHANGE N=522 N=222 Median (interquartile range) CD4% change over first 48 weeks (unadjusted) HCV+ 2 (-1.7, 7) 1 (-2, 7) HCV6 (2, 10) 5 (3, 10) p-value <0.001 p<0.001 * Adjusted average percent change over 48 weeks HCV+ 0.36% (p<0.001) HCV2.5% (p<0.001)  0.24% (p<0.001) 2.6% (p<0.001)  N=330  N=197  2 (-1, 8) 6 (2, 10) <0.001  2 (-2, 6) 6 (2, 10) <0.001  0.72% (p=0.001) 2.5% (p<0.001)  -0.48% (p<0.001) 2.6% (p<0.001)  *Among >95% Adherent Only: Adjusted average number of CD4's gained over 48 weeks HCV+ 1.9% (p=0.017) 1.6% (p=0.040) 2.0% (p=0.094) HCV3.2% (p<0.001) 3.2% (p<0.001) 3.1% (p<0.001)  1.6% (p=0.074) 3.4% (p<0.001)  * Adjusted for time (months), age, gender, AIDS diagnosis at baseline, plasma log viral load response over 48 week period. ±. Adjusted for baseline age, gender, AIDS diagnosis, baseline CD4 and HIV viral load, and adherence over first 12 months of therapy. Q See Table 3 for the models from which these 'adjusted average number 'results are derived.  114  Figure 5.1. (A-D): Kaplan-Meier Analysis of the Effect of Hepatitis C on the:  Figure  5.1.A  Time to Absolute CD4 Increase of At Least 75 cells/mm3  HCV-negative HCV-positive  Days from initiation of ART  Figure  5.I.B.  Time to Absolute CD4 Increase of At Least 75 cells/mm3 - Among >95% Adherent Only  Days from Start of ART  115  Figure  5.I.C.  Time to CD4 Percent Increase of At Least 10%  100  HCV-negative " » H C V - p o s i t i v e  0  T  1  1  1  1  1  1  1  1  1  28  56  84  112  140  168  196  224  252  280  r——r 308  336  Time from Start of A R T  Figure  5.I.D.  Time to CD4 Percent Increase of At Least 10% Among >95% Adherent Only 100  n  HCV-negative  »  HCV-positive  o  i—i 0  28  56  1 1—i 1—i i — i 1 1—r 84  112  140  168  196  224  252  280  308  336  Days from Start of A R T  116  Figure 5.2.A. Adjusted Average Number of Absolute CD4 Cells Gained Over 48 Weeks of Antiretroviral Treatment  117  Figure 5.2.B. Adjusted Average Number of Absolute CD4 Cells Gained Over 48 Weeks of Antiretroviral Treatment Among Individuals A t Least 95% Adherent  118  Figure 5.2.C. Adjusted Average Increase of CD4 Fraction Gained Over 48 Weeks of Antiretroviral Treatment  1 1 9  Figure 5.2.D. Adjusted Average Increase of C D 4 Fraction G a i n e d Over 48 Weeks of Antiretroviral Treatment A m o n g Individuals A t Least 95% Adherent to Therapy  120  CHAPTER 6: SAFETY AND TOLERABILITY OF ANTIRETROVIRAL THERAPY IN H I V / H C V CO-INFECTED PATIENTS 6.1 FORWARD  This chapter is under review with the Tournal of Acquired Immune Deficiency Syndromes as:  P. Braitstein, A . Justice, D . Bangsberg, B. Y i p , V . Alfonso, M T . Schechter, J S C Montaner, RS. Hogg. "Adherence to Antiretroviral Therapy and H I V / H C V Co-Infection".  121  6. 2 INTRODUCTION  It is estimated that on average, 30% of HIV-positive people i n N o r t h America and Europe are co-infected w i t h the Hepatitis C Virus (HCV) [ 1 , 2]. However, because i n some populations including injection drug users, prison inmates, and hemophiliacs, the prevalence of H C V is so widespread, the prevalence of H C V co-infection among H I V positive individuals in these populations ranges f r o m 50-99% [3-6].  As survival has  been increasing for patients w i t h H I V infection due to the use of H i g h l y Active Antiretroviral Therapy, liver disease has emerged as a leading cause of morbidity and mortality in HIV-infected populations [7-9]. The Hepatitis C Virus (HCV) in particular, because of its high prevalence in HIV-positive populations, and its greatly increased pathogenicity in the setting of H I V [2, 10], is now a leading cause of death of H I V positive individuals [9,11,12].  Adherence to antiretroviral therapy is paramount to its effective use [13-15]. Factors previously found to be associated w i t h adherence are physician experience [16], age [17], alcohol use and incarceration [17], toxicity [18], and injection drug use [17, 19]. Injection d r u g use has also been identified as an important predictor of treatment discontinuation [15, 20]. However, although the majority of HIV-positive injection drug users are also co-infected w i t h hepatitis C, only a few studies have examined H I V / H C V co-infection as a factor in switching or interrupting H A A R T . While most have found that there is a strong relationship between treatment interruptions and H C V coinfection [21-23], none have looked explicitly at the question of adherence.  The primary objective of the present study was to describe the effect of hepatitis C (HCV) co-infection, controlling for biochemical markers of liver injury and injection 122  drug use (IDU), on adherence to antiretroviral treatment (ART) during the first year of therapy i n a population-based H I V / A I D S d r u g treatment program.  The primary  hypothesis was that H C V serostatus w o u l d have an independent negative effect on adherence, and that this w o u l d be the result of more liver injury, as evidenced by an interaction between H C V serostatus and any biochemical marker of liver injury. The secondary hypothesis was that H C V w o u l d be more strongly negatively associated w i t h adherence than a history of injection drug use.  6 . 3 METHODS  Data Source and Study Population: The British Columbia HIV/AIDS  Drug Treatment Program (DTP)  Antiretroviral medications have been centrally distributed at no cost to eligible H I V infected individuals since 1986. I n October 1992, the distribution of antiretroviral agents became the responsibility of the H I V / A I D S D r u g Treatment Program of the British Columbia Centre for Excellence i n H I V / A I D S .  This antiretroviral d r u g  distribution program remains the only free source of antiretroviral medication in this Canadian province (and is a unique program i n Canada). The Centre's H I V / A I D S Drug Treatment program has received ethical approval f r o m the University of British Columbia Ethics Review Committee at its St. Paul's Hospital site, and the program conforms w i t h the province's Freedom of Information and Protection of Privacy Act.  The Centre distributes antiretroviral drugs based on specific guidelines generated b y the Therapeutic Guidelines Committee [24]. These guidelines have been and continue  123  to be consistent w i t h those treatment guidelines published by the International AIDS Society [25-28]. The Centre's guidelines recommend that CD4 and plasma HIV-1 R N A levels be monitored at baseline, at 4 weeks after starting antiretroviral therapy and every three months thereafter. Plasma viral loads are measured using the Amplicor HIV-1 M o n i t o r ™ (Roche Diagnostics Branchburg, NJ). A l l classes of federally licensed antiretroviral drugs are currently available t h r o u g h the p r o g r a m , including all nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors ( N N R T I ) , and protease inhibitors (PI).  Tenofovir, Atazanavir, and  Enfuvirtide are also available. Eligibility for antiretroviral medication has remained consistent w i t h current international recommendations [28].  Individuals are automatically entered into the DTP when they are first prescribed any antiretroviral agent.  A t DTP entry and w i t h each subsequent physician visit, the  participant's complete history (if any) of antiretroviral use, CD4 cell count, plasma viral load, and disease stage are recorded. Typically, patients are followed-up at 3-month intervals, at w h i c h time prescriptions are renewed or altered as necessary.  Blood  d r a w n for the purposes of CD4 and viral load testing is stored for each individual at each follow-up visit for future research activities related to H I V disease.  The H A A R T Observational Medical Evaluation and Research (HOMER) study is a nested cohort w i t h i n the BC Center for Excellence's H I V / A I D S D r u g Treatment Program. It includes all previously ART naive individuals w h o initiated antiretroviral therapy w i t h a t r i p l e - d r u g regimen consisting of either t w o nucleoside reverse transcriptase inhibitors (NRTI) plus either a protease inhibitor (PI), or a non-nucleoside reverse transcriptase inhibitors (NNRTI), between August, 1996 and July, 2000. The data 124  used i n these analyses are based on individuals f r o m the H O M E R Cohort, for w h o m there was H C V serological data.  H C V serological data was obtained t h r o u g h  retrospective testing of stored plasma samples taken just prior to treatment initiation.  The BC Centre for Excellence i n H I V / A I D S is located w i t h i n Providence Health Hospital i n Vancouver. The markers of hepatic injury were obtained through a linkage w i t h the Providence Laboratory, where a majority of HIV-infected patients i n the province receive their H I V care.  Outcome  Measures  The primary outcome measure for this analysis was adherence to antiretroviral therapy. Our definition of adherence is based on the ratio of the time that medication dispensed w o u l d last as a proportion of follow-up time. This calculation was restricted to each patient's first year o n therapy to avoid the reverse causation that could occur among patients w h o cease antiretroviral therapy after they have become too sick to take medication. We have previously demonstrated h o w this estimate strongly predicts virologic response [13] and survival [29, 30]. For the purposes of our analyses, adherence was a binary variable where 1 was being greater than 95% adherent (meaning having refilled at least 95% of their prescriptions during the year), and 0 less than 95% adherent. Baseline characteristics examined were gender (male vs. female), age (continous), absolute baseline CD4 (continuous) and H I V l o g  10  plasma R N A (pVL) (continous),  whether patients had an AIDS diagnosis (yes vs. no) or any history of injection drug use (yes vs. no), and type of therapy initiated (Pi-based vs. NNRTI-based).  125  Markers of liver injury were alanine aminotransferase (ALT) (International Units per liter ( I U / L ) ) , aspartate aminotransferase (AST) ( I U / L ) , a l b u m i n ( g r a m s / L ) , and International Normalized Ratio (INR). The analysis was restricted to the patient's first year on ART, and if a patient had more than one test result i n that year, the latest result was taken.  These markers were analyzed as continuous variables to allow for the  possibility that patients could be experiencing symptoms of liver injury w i t h o u t necessarily meeting a particular threshold of 'significant' liver injury (e.g. A L T > 2.5 times the upper l i m i t of normal). Markers of liver i n j u r y were also defined dichotomously accordingly to clinically significant thresholds: A L T > 2.5Xuln, AST > 2.5Xuln, albumin <30, and I N R > 1.3. A global measure of hepatic injury was created b y defining a patient as having any liver injury if they had at least one of the four dichotomous outcome measures. The analysis was restricted to the patient's first year on ART, and if a patient had more than one test result i n that year, the latest result was taken.  Statistical Analysis H C V positive and H C V negative participants were compared using both parametric and distribution-free methods. Categorical data were analyzed using Pearson's % test. 2  Fisher's exact test was used for contingency tables i n which 25% or more of the expected cell frequencies were less than 5. Continuous variables were analyzed using the Wilcoxon rank-sum test. Adherent versus non-adherent individuals were analyzed i n a similar fashion.  I n unadjusted logistic regression, H C V antibody-status (positive vs. negative), any history of injection d r u g use (yes vs. no), gender (male vs. female), baseline CD4 (per 126  100 cells), age at baseline (per 10 years) and whether the patient had an AIDS diagnosis at baseline (yes vs no) were separately tested for their association w i t h adherence, as were A L T , AST, albumin, and INR, as both continuous and dichotomous variables. Separate models were constructed to assess the relative effects of 1) H C V + biochemical markers of liver injury; 2) I D U + biochemical markers of liver injury; 3) H C V + I D U + biochemical markers of liver injury. The final adjusted model of factors independently associated w i t h >95% adherence used a f o r w a r d stepwise multivariate logistic regression model fit by including H C V + I D U + biochemical (continuous) markers of liver injury + baseline sociodemographic and clinical characteristics considered potential sources of confounding (gender, age, baseline CD4, and whether they had an AIDS diagnosis at baseline).  Separate modeling was conducted using the global measure of hepatic injury. This was done in order to create an interaction term between H C V co-infection and hepatic injury, based on the hypothesis that while those w i t h H C V infection w o u l d be less likely to adhere to treatment, this would be most strongly associated w i t h the combined effect of H C V infection and liver injury.  To elucidate the differential effects of injection drug use and hepatitis C, a four-level variable was constructed: 0 = no I D U / n o H C V ; 1 = I D U but no H C V ; 2 = no I D U but HCV-positive; and 3 = both I D U and HCV-positive. This variable was described using Pearson's Chi Square statistic.  These categories were also used to create separate  variables, each w i t h no I D U and no HCV as the reference, and entered into multivariate logistic regression models.  Pearson's Correlation Coefficient was used to assess  correlation between the H C V and injection drug use variables.  127  6.4 RESULTS  There were 1186 patients eligible for analysis, including 606 (51%) HCV-antibody (Ab) positive, and 580 (49%) HCV-Ab negative. As summarized in Table 4.1 (page 90), HCVpositive individuals were less likely to be male (78% vs. 93%, p<0.001), and to have an AIDS diagnosis at baseline (11% vs. 15%, p=0.028), but were more likely to have any history of I D U (47% vs. 6%, p<0.001), and had a higher fraction CD4 at baseline (19% vs. 16%, p<0.001).  There were no statistical differences (p>0.05) between HCV-positive  and HCV-negative patients at baseline in terms of age, absolute CD4 count, log plasma H I V RNA, or type of therapy initiated.  In their first year of antiretroviral therapy (Table 6.1), H I V / H C V co-infected patients were significantly less likely to be at least 95% adherent to their treatment (42% vs. 72%, p<0.001). They had a higher median A L T (52 vs. 35, p<0.001), AST (46 vs. 29, p<0.001), INR (1.1 vs. 1.0, p=0.037), and lower albumin (39 vs. 41, p<0.001). The H C V co-infected were significantly more likely to have had an A L T > 2.5X u l n in their first year of treatment (26% vs. 8%,p<0.001), an A L T > 5X u l n (4% vs. 1 % , p=0.049), an AST > 2.5Xuln (15% vs. 2%, p<0.001), or an albumin < 30 (20% vs. 11%, p=0.034). There was only a trend of a difference in the proportions of having an I N R > 1.3 (15% vs. 8%, p=0.136). Overall, the H C V co-infected were significantly more likely to have any biochemical marker of liver injury (26% vs. 8%, p<0.001).  As summarized i n Table 6.2, adherent patients were more likely to be male (91% vs. 78%, p<0.001), to have had an AIDS diagnosis at baseline (16% vs. 8%, p<0.001), and to be slightly older at baseline (median 37.6 vs. 36.7, p=0.001).  They were less likely to  have any history of I D U (21% vs. 35%, p<0.001), to be H C V Ab-positive (38% vs. 69%, p<0.001), and to have any biochemical marker of liver injury (12% vs. 25%, p<0.001), 128  including A L T > 5X uln ( 1 % vs. 4%, p=0.019), and an albumin result of < 30 (8% vs. 24%, p<0.001). Adherent patients had significantly lower median AST (31 vs. 37, p<0.001) and higher albumin (41 vs. 38, p<0.001), but there was no difference in their median A L T or INR.  Tables 6.3 summarize the results of the logistic regression analysis of factors independently associated w i t h being at least 95% adherent to ART.  As seen in Table  6.3.1, which presents the results using continuous measures of liver injury, the adjusted odds ratio (AOR) and 95% confidence interval (95%CI) for H C V infection remained significant in the final model (AOR 0.39, 95%CI: 0.22-0.73, p=0.003), as d i d albumin (AOR 1.07, 95%CI: 1.03-1.11, p=0.002), and male sex (AOR 2.42, 95%CI: 1.18-4.96, p=0.016). Even w h e n only adjusting for biochemical markers of liver injury but not H C V co-infection, I D U became non-significant (AOR 0.67, 95%CI: 0.41-1.14).  As seen in Table 6.3.2, i n the final model, both H C V co-infection (AOR: 0.35, 95% CI: 0.24 - 0.51, p<0.001) and liver injury (AOR 0.47, 95%CI: 0.23 - 0.94) were negatively associated w i t h adherence, while male sex remained positively associated (AOR 2.58, 95%CI: 1.59-4.18, p<0.001). The interaction term of HCV*liver injury fell out of the final model (AOR 1.42, 95%CI: 0.62-3.25, p=0.412). Interestingly, although I D U became nonsignificant in the final model (AOR: 0.80, 95%CI: 0.55-1.16, p=0.243), it d i d remain significant after adjusting only for liver injury, but not other factors such as HCV.  We then endeavoured to separate the effects of injection drug use and hepatitis C. Of the 544 individuals w i t h no reported history of I D U and negative H C V serology, 400 (74%) were at least 95% adherent, and 144 (26%) were not. Of the 36 individuals who had a reported history of I D U but negative H C V serology, 20 (55%) were adherent, and 129  16 (45%) were not (p=0.019). Of the 321 individuals w i t h no history of I D U but w h o had positive H C V antibodies, 134 (42%) were adherent, while 187 (58%) were not (p<0.001). Finally, of the 285 individuals w i t h both a reported history of I D U and H C V infection, 121 (42%) were adherent, while the remaining 164 (58%) were not (p<0.001). Pearson's Correlation Coefficient between any injection d r u g use and hepatitis C serostatus was 0.46.  Tables 6.4 summarize the results of the analyses aimed at further separating the effects of H C V and injection d r u g use.  As seen i n Table 6.4.1, w i t h o u t exception, the  individuals w h o have a history of injection drug use but no H C V infection have the most favorable results regarding biochemical markers of liver injury, while those w i t h positive H C V antibodies have the worst. In logistic regression analysis only controlling for baseline sociodemographic and clinical factors (Table 6.4.2), while the I D U + / H C V group are less likely to be at least 95% adherent to medication (AOR 0.47, 95%CI: 0.230.93, p=0.030), those w i t h H C V infection w i t h or w i t h o u t I D U history are even less likely to be adherent ( I D U - / H C V + AOR: 0.27, 95%CI: 0.20 - 0.36, p<0.001; I D U + / H C V + AOR: 0.30, 95%CI: 0.22 - 0.42, p<0.001). When also controlling for biochemical markers of liver injury as continuous measures, those w i t h a history of I D U but no H C V are not independently less likely to adhere to treatment (AOR: 0.19, 95%CI: 0.03 - 1.32, p=0.126), while those w i t h H C V infection (with or without IDU) remain significantly less likely to be adherent ( I D U - / H C V + A O R : 0.34, 95%CI: 0.18 - 0.64, p<0.001; I D U + / H C V + AOR: 0.44, 95%CI: 0.23 - 0.87, p=0.018).  130  6.5 DISCUSSION  I n summary, our data suggest that H C V co-infection, biochemical markers of liver injury (notably albumin), male sex, and to a lesser extent age, are those factors most strongly associated w i t h adherence to antiretroviral therapy. While our findings do not suggest an interaction between H C V and liver injury, they do support the hypothesis that H C V is more strongly associated w i t h adherence than a history of injection drug use.  Our data support those of others who have found that H C V infection is associated w i t h treatment discontinuations and treatment interruptions.  M e l v i n et al. reported  antiretroviral discontinuation rates due to hepatic toxicity of more than two-fold i n HCV-co-infected individuals compared to H I V mono-infected persons [22]. Among a population of 465 previously antiretroviral naive individuals, H C V seropositivity was associated w i t h an adjusted 40% increased risk of discontinuing or changing initial H A A R T regimens w i t h i n the first year of treatment [21]. The authors indicate that whether this was due to histological damage, reduced adherence, or increased hepatotoxicity is not clear, because hepatic cirrhosis was also independently associated w i t h H A A R T discontinuation (AOR 2.1, 95% CI: 1.1-3.8) [21]. Aceti (2002) reported that 83% of those w h o discontinued antiretrovirals because of hepatotoxicity were H C V coinfected [31]. Of note, D'Arminio Monforte et al. (2000) found no impact of H C V on treatment interruptions due to toxicity [23].  There are a number of important strengths to our analysis. One, the data are d r a w n from a population-based program, and are therefore more broadly generalizable than other study populations.  Secondly, we have endeavoured to untangle the rather 131  complex relationship between H C V infection, injection d r u g use, and hepatic injury, enabling a much deeper understanding of adherence issues as they relate particularly to H C V infection and injection drug use. T h i r d , to our knowledge, we are the first to explicitly examine the question of hepatitis C co-infection and adherence to antiretroviral therapy, and to ask the question about the combined effect of H C V coinfection and hepatic injury. Finally, more effective hepatitis C treatment (i.e. pegylated interferon combined w i t h ribavirin) only became available i n British Columbia i n June, 2003, and as this analysis was restricted to patients first year on H I V treatment, the effect of H C V treatment on our results is negligible.  There may also be potential limitations to our analysis. One, the markers of liver injury are only present for a m a x i m u m of 72% of the cohort, and this may limit the results. However, we feel that it is likely the distribution of missing values w o u l d remain proportionate across the adherence and liver injury categories, and that having data from more patients w o u l d strengthen the results. Two, the measures of liver injury are biochemical, and A L T and AST levels in particular may not be reliable measures of liver inflammation or injury, as others have shown that patients w i t h normal A L T levels may have hepatic scarring [32].  Furthermore, there are many symptoms of H C V disease  and ART hepatotoxicity which could impact on adherence, without being captured by these measures, including nausea, malaise, etc. This may be the reason w h y H C V infection itself remains associated w i t h adherence i n adjusted models.  A third  limitation to our study is that our measure of injection d r u g use is a composite one, drawing upon both physician and patient self-report data. The fact that only 47% of those w i t h H C V infection have any reported history of I D U suggests that there may be underreporting of this variable. However, as H C V is highly infectious and individuals at risk for H I V infection may also have multiple risk factors for H C V (e.g. certain sexual 132  practices, tattooing, intranasal cocaine use), this underestimation may be minimal. Further, the I D U measure is based on any history of I D U , which may include those who used injection drugs recreationally many years ago, but w h o are otherwise not typical of those I D U generally considered at high-risk of non-adherence (i.e. actively streetinvolved and actively using injection drugs).  A final limitation of this analysis is that  we were not able to account for the effects of either alcohol use or depression or depressive symptoms on adherence.  Our data have important implications for the conduct of clinical research regarding HIV/hepatitis C co-infection and injection drug users. Accurate estimates of effects due to I D U may be difficult to obtain without adjusting for the effects of H C V infection, and vice versa. More research is urgently needed to elucidate the mechanisms involved in precluding or preventing H I V / H C V co-infected individuals and those w i t h a history of injection drug use from maximally adhering to their antiretroviral therapy.  I n conclusion, our data suggest that H C V infection, albumin (as a marker of hepatic injury), and male sex, are independently associated w i t h adherence, and after adjusting for these other factors, injection drug use is not. Further, although H C V co-infected individuals are more likely to have any biochemical evidence of liver injury, the combined effect of H C V and liver injury in this analysis was negligible, although both HCV and liver injury were both independent factors.  133  6.6 REFERENCES  1.  Rockstroh, J.K. and U. Spengler, HIV and hepatitis C virus co-infection. Lancet Infect Dis, 2004. 4(7): p. 437-44.  2.  Soriano, V., et al., Care of patients with hepatitis C and HIV co-infection. AIDS, 2004. 18(1): p. 1-12.  3.  Filippini, P., et al., Hepatitis viruses and HIV infection in the Naples area. Infez Med, 2003.11(3): p. 139-45.  4.  Patrick, D.M., et al., Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001.165(7): p. 889-95.  5.  Zhang, C , et al., High prevalence of HIV-1 and hepatitis C virus coinfection among injection drug users in the southeastern region of Yunnan, China. J Acquir Immune Defic Syndr, 2002. 29(2): p. 191-6.  6.  Sherman, K.E., et al., Hepatitis C Virus prevalence among patients infected with Human Immunodeficiency Virus: a cross-sectional analysis of the US adult AIDS Clinical Trials Group. Clinical Infectious Diseases, 2002. 34(6): p. 831-7.  7.  Soriano, V., et al., Impact of chronic liver disease due to hepatitis viruses as cause of hospital admission and death in HIV-infected drug users. European Journal of Epidemiology., 1999.15(1): p. 1-4.  8.  Martin-Carbonero, L., et al., Increasing impact of chronic viral hepatitis on hospital admissions and mortality among HIV-infected patients. AIDS Research & H u m a n Retroviruses., 2001.17(16): p. 1467-71.  9.  Bica, I., et al., Increasing Mortality Due to End-Stage Liver Disease in Patients with Human Immunodeficiency Virus Infection. Clinical Infectious Diseases (CID), 2001. 32(Feb 1): p. 492-497.  134  10.  Soriano, V., et al., Care of patients with chronic hepatitis C and HIV co-infection: recommendations from the HIV-HCV  11.  International Panel. Aids, 2002.16(6): p. 813-28.  Cacoub, P., et al., Mortality among human immunodeficiency virus-infected patients with cirrhosis or hepatocellular carcinoma due to hepatitis C virus in French departments of internal medicine/infectious diseases, in 1995 and 1997. Clinical Infectious Diseases, 2001. 32: p. 1207-1214.  12.  Greub, G., et al., Clinical  progression, survival, and immune recovery during  antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. 13.  Low-Beer, S., et al., Adherence to triple therapy and viral load response. J Acquir Immune Defic Syndr, 2000. 23(4): p. 360-1.  14.  Wood, E., et al., Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 x 10(9) cells/L. A n n Intern Med, 2003.139(10): p. 810-6.  15.  W o o d , E., et al., Adherence and plasma HIV RNA responses to highly active antiretroviral therapy among HIV-1 infected injection drug users. Cmaj, 2003.169(7): p. 656-61.  16.  Delgado, J., et al., Highly active antiretroviral therapy: physician experience and enhanced adherence to prescription refill. A n t i v i r Ther, 2003. 8(5): p. 471-8.  17.  O'Connell, J.M., et al., Age, adherence and injection drug use predict virological suppression among men and women enrolled in a population-based antiretroviral drug treatment programme. A n t i v i r Ther, 2003. 8(6): p. 569-76.  18.  Heath, K.V., et al., Intentional nonadherence due to adverse symptoms associated with antiretroviral therapy. J Acquir Immune Defic Syndr, 2002. 31(2): p. 211-7.  135  19.  Palepu, A., et al., Uptake and adherence to highly active antiretroviral therapy among HIV-infected  people with alcohol and other substance use problems: the impact of  substance abuse treatment. Addiction, 2004. 99(3): p. 361-8. 20.  Wood, E., et al., Adherence to antiretroviral therapy and CD4 T-cell count responses among HIV-infected injection drug users. Antivir Ther, 2004. 9(2): p. 229-35.  21.  Ripamonti, D., et al., Hepatitis C infection increases the risk of the modification of first highly active antiretroviral therapy in HIV-infected patients. AIDS, 2004.18(2): p. 33436.  22.  M e l v i n , D., et al., The impact of co-infection with hepatitis C virus and HIV on the tolerability of antiretroviral therapy. AIDS, 2000.14(4): p. 463-465.  23.  d A r m i n i o Monforte, A., et al., Insights into the reasons for discontinuation of the first highly active antiretroviral therapy (HAART) regimen in a cohort of antiretroviral naive patients. AIDS, 2000.14(5): p. 499-507.  24.  BC Centre for Excellence i n H I V / A I D S . , Therapeutic Guidelines for the Treatment of HIV/AIDS and Related Conditions. 1999, 2001: Vancouver.  25.  Carpenter, C , et al., Antiretroviral therapy for HIV infection in 1996. JAMA, 1996. 276: p. 146-154.  26.  Carpenter, C , et al., Antiretroviral Therapy for HIV Infection in 1997. J A M A , 1997. 277: p. 1962-1969.  27.  Carpenter, C , et al., Antiretroviral therapy in adults. Updated recommendations of the International AIDS Society-USA Panel. JAMA, 2000. 283(3): p. 381-91.  28.  Yeni, P.G., et al., Antiretroviral  treatment for adult HIV infection in 2002: updated  recommendations of the International AIDS Society-USA Panel. Jama, 2002. 288(2): p. 222-35. 29.  Wood, E., et al., Is there a baseline CD4 cell count that precludes a survival response to modern antiretroviral therapy? Aids, 2003.17(5): p. 711-20. 136  30.  H o g g , R.S., et al., Intermittent use of triple-combination  therapy is predictive of  mortality at baseline and after 1 year of follow-up. Aids, 2002.16(7): p. 1051-8. 31.  Aceti, A., et al., Hepatotoxicity development during antiretroviral therapy containing protease inhibitors in patients with HIV: the role of hepatitis B and C virus infection. J Acquir Immune Defic Syndr, 2002. 29(1): p. 41-8.  32.  Pradat, P., et al., Predictive value of ALT levels for histologic findings in chronic hepatitis C: A European Collaborative Study. Hepatology, 2002. 36: p. 973-977.  137  Table 6.1. Characteristics in First Year of ART Among HCV-positive and H C V negative Individuals in the British Columbia HIV/AIDS Drug Treatment Program  HCV-positive N=606  HCV-negative n=580  p-value  >95% Adherent  255 (42%)  420 (72%)  <0.001  AST > 2.5Xuln (n=846)  59 (15%)  9 (2%)  <0.001  A L T > 2.5Xuln (n=630)  42 (13%)  9 (3%)  <0.001  A L T > 5.0Xuln (n=630)  13 (4%)  4 (1%)  0.049  Albumin <30 (n=323)  39 (20%)  13 (11%)  0.034  INR >1.3 (n=260)  24 (15%)  8 (8%)  0.136  Any liver injury (n=852) 126 (32%)  39 (9%)  <0.001  ALT (median, IQR)  52 (32 - 89)  35 (27 - 51)  <0.001  AST (median, IQR)  46 (30 - 80)  29 (23 - 36.5)  <0.001  ALB (median, IQR)  39 (33 - 42)  41 (37 - 43)  <0.001  INR (median, IQR)  1.1 (0.9-1.2)  1.0 (0.9-1.1)  0.037  *ART: antiretroviral therapy  138  Table 6.2. Factors associated with Adherence to Antiretroviral Medication  Adherent n=675  Non-Adherent n=511  P-value  614 (91%)  400 (78%)  <0.001  37.6 (32.5 - 44.6)  36.7(31.1-42.9)  0.001  141 (21%)  180 (35%)  <0.001  HCV-Antibody positive 255 (38%)  351 (69%)  <0.001  AIDS diagnosis at baseline 109 (16%)  43 (8%)  <0.001  Any Liver Injury (n=852) 71 (14%)  94 (27%)  <0.001  AST > 2.5Xuln (n=846)  27 (5%)  41 (12%)  <0.001  A L T > 2Xuln (n=630)  23 (6%)  28 (10%)  0.070  ALT > 5Xuln(n=630)  5 (1%)  12 (4%)  0.019  Albumin < 30 (n=323)  14 (8%)  38 (24%)  <0.001  INR >1.3 (n=260)  13 (10%)  19 (14%)  0.368  A L T (median, IQR)  39 (29 - 60)  42 (29 - 76)  0.215  AST (median, IQR)  31 (25 - 44)  37 (27 - 67)  <0.001  ALB (median, IQR)  41 (37 - 43)  38 (31 - 41)  <0.001  INR (median, IQR)  1.1 (1.0-1.1)  1.1 (1.0-1.2)  0.280  Male Gender Age at baseline (median, IQR) Any IDU  Table 6.3.1: Logistic Regression Analysis of Factors Associated with Being At Least 95% Adherent to Antiretroviral Therapy in First Year of Treatment (n=1186)  Unadjusted  Model 1  Model 2  Model 3  Model 4  Hepatitis C positive 0.27 (0.19-0.37) <0.001 p-value  0.36 (0.21-0.62) <0.001  -  0.36 (0.20-0.64) <0.001  0.39 (0.22-0.73) 0.003  Any History of IDU 0.46 (0.33-0.66) <0.001 p-value  -  0.67 (0.41-1.14) 0.139  1.05 (0.59-1.86) 0.881  1.09 (0.60-1.98) 0.772  ALT (continuous) p-value  0.997 (0.995-0.999) 0.010  1.00 (0.99-1.01) 0.972  1.00 (0.99-1.01) 0.909  1.00 (0.99-1.01) 0.979  1.00 (0.995-1.01) 0.857  AST (continuous) p-value  0.994 (0.991-0.996) <0.001  0.998 (0.99-1.01) 0.592  0.997 (0.990-1.00) 0.356  0.998 (0.99-1.01) 0.587  1.00 (0.99-1.01) 0.595  1.06 (1.02-1.10) <0.001  1.07(1.03-1.11) <0.001  1.06 (1.02-1.10) 0.002  1.07 (1.03-1.11) 0.002  _  -  -  2.42 (1.18-4.96) 0.016  Albumin (continuous)1.08 (1.04-1.12) <0.001 p-value Male Sex p-value  3.90 (2.40-6.40) <0.001  Age (per 10 yr increase)1.27 (1.11-1.45) p-value <0.001  1.32 (0.95-1.85) 0.097  AIDS at baseline p-value  1.90 (1.20-3.10) 0.012  1.14 (0.56-2.35) 0.718  CD4 (per 100 cells) p-value  0.96 (0.90-1.00) 0.302  0.91 (0.80-1.04) 0.166  -tr O  140  Table 6.3.2 Logistic Regression Analysis of Factors Associated with Being At Least 95% Adherent to Antiretroviral Therapy in First Year of Treatment (n=1186)  Unadjusted Hepatitis C positive 0.27 (0.19-0.37) <0.001 p-value Any IDU p-value  0.46 (0.33-0.66) <0.001  Any Liver Injury p-value  0.40 (0.27-0.60) <0.001  HCV*Liver Injury p-value  0.30 (0.19-0.48) <0.001  Male Sex ~ p-value  3.90 (2.40-6.40) <0.001  Age- (per 10 yr increase)1.27 (1.11-1.45) <0.001 p-value  Model 1  Model 3  Model 4  Model 5  0.31 (0.21-0.44) <0.001  0.31 (0.21 - 0.45) <0.001  0.35 (0.24-0.51) <0.001  0.82 (0.56-1.22) 0.328  0.75(0.52-1.07) 0.80 (0.55 -1.16) 0.115 0.243  0.56 (0.36-0.86) 0.42 (0.28-0.64) 0.55 (0.36-0.85) 0.008 <0.001 0.007  0.54 (0.24-1.25) 0.47 (0.23 - 0.94) 0.150 0.033  0.28 (0.20-0.39) <0.001 -  -  -  AIDS at baseline ~ 1.90 (1.20-3.10) 0.012 p-value CD4 (per 100 cells) ~ 0.96 (0.90-1.00) 0.302 p-value  Model 2 -  0.47 (0.33-0.66) <0.001  -  -  -  -  -  2.58 (1.59-4.18) <0.001  -  -  -  1.27 (1.06-1.52) 0.009  -  1.47 (0.90-2.42) 0.126  -  0.95 (0.88-1.02) 0.158  -  _  -  1.03 (0.39-2.72) 1.42 (0.62-3.25) 0.955 0.412  141  Table 6.4.1. Median results (interquartile range) of biochemical markers of liver injury by the four possible categories of H C V and IDU status Median (interquartile range)  IDU-/HCVn=544  IDU+/HCVn=36  IDU-/HCV+ n=321  IDU+/HCV+ n=285  p-value  ALT  35 (27--52)  31.5 (27 -42)  49 (33 --94)  53 (31 --88)  <0.001  AST  29 (23 --36)  28 (22 - 39)  46 (30 --81)  46 (29 --78)  <0.001  Albumin  41 (37 --43)  41.5 (40 -43)  39 (32 --42)  38 (33 --42)  0.006  INR  1.1 (1.0--1.1)  1.0 (1.0--1.0)  1.1 (1.0--1.2)  1.1 (1.0--1.2)  0.053  "or  142  Table 6.4.2. Logistic Regression Analysis of Factors Associated with Being At Least 95% Adherent to Antiretroviral Medications, Controlling for Gender, Age at Baseline, AIDS Diagnosis at Baseline, and Baseline CD4  Unadjusted Odds  p-value  Adjusted Odds  p-value  (95% CI)  (95% CI) IDU-/HCV-  -  -  -  -  IDU+/HCVIDU-/HCV+ IDU+/HCV+  0.45 (0.23-0.89) 0.26 (0.19-0.35) 0.27 (0.20-0.36)  0.022 <0.001 <0.001  0.47 (0.23-0.93) 0.27 (0.20-0.36) 0.30 (0.22-0.42)  0.030 <0.001 <0.001  Table 6.4.3. Logistic Regression Analysis of Factors Associated with Being At Least 95% Adherent to Antiretroviral Medications, Controlling for AST, A L T , Albumin, Gender, Age at Baseline, AIDS Diagnosis at Baseline, and Baseline CD4  Unadjusted Odds  p-value  IDU+/HCVIDU-/HCV+ IDU+/HCV+  0.45 (0.23-0.89) 0.26 (0.19-0.35) 0.27 (0.20-0.36)  p-value  (95% CI)  (95% CI) IDU-/HCV-  Adjusted Odds  0.022 <0.001 <0.001  -  -  0.19 (0.03-1.32) 0.34 (0.18-0.64) 0.44 (0.23-0.87)  0.126 0.001 0.018  143  CHAPTER 7: MORTALITY AND H I V / H C V C O - l N F E C T I O N  7.1 FORWARD  This chapter is currently has been conditionally accepted by the Canadian Medical Association journal as: Braitstein P, Yip B, Montessori V, Moore D, Montaner, JSG, Hogg RS. "Non-Accidental Mortality and Causes of Death Among Previously Antiretroviral Naive H I V / H C V CoInfected vs. H I V Mono-Infected Patients i n a Population-Based Program".  144  7.2 INTRODUCTION  Hepatitis C is w i d e l y prevalent among HIV-infected individuals [1]. While several authors have previously found that liver disease has become a leading cause of death of HIV-positive individuals [2-4], debate continues as to the effect of the hepatitis C virus (HCV) on H I V disease progression, as measured either by a new AIDS-defining illness, CD4 decline, or HIV-related mortality [4-9].  M o r t a l i t y i n this population can be  strongly confounded by adherence to antiretroviral therapy (ART), injection drug use, and previous ART use.  During the 1990's, Vancouver, British Columbia, had an explosive epidemic of H I V and H C V infection among the city's 10,000 injection d r u g users, and over 30% of this population is co-infected w i t h H I V and H C V [10, 11]. Here we report on the effect of H C V serostatus on the risk of non-accidental mortality i n a population-based H I V treatment program of previously ART naive patients, adjusting for adherence to ART and injection drug use, and describe broad causes of non-accidental death among these individuals.  7.3 METHODS  The British Columbia H I V / A I D S D r u g Treatment Program, previously described i n detail elsewhere, [12] is the only source of free antiretrovirals i n BC, and follows therapeutic guidelines developed consistent w i t h international standards [13]. Data for this analysis is d r a w n f r o m those previously antiretroviral therapy (ART) naive individuals w h o initiated ART w i t h a triple-drug regimen consisting of either t w o nucleoside reverse transcriptase inhibitors (NRTI) plus either a protease inhibitor (PI), 145  or a non-nucleoside reverse transcriptase inhibitor (NNRTI). This analysis is restricted to those for w h o m there was H C V serological data, and w h o first initiated antiretroviral therapy between August 1996 and July, 2000. The data are censored at June 30, 2003. Mortality data were obtained through a linkage w i t h British Columbia Vital Statistics, and ICD9/10 codes were used to evaluate the underlying cause of death. Accidental deaths were excluded from all survival analyses.  Our definition of adherence is based on the ratio of the time that medication dispensed w o u l d last as a proportion of follow-up time, and has previously been validated as a strong predictor of both virologic response [14J and survival [15, 16].  Statistical  analyses included parametric and non-parametric methods and standard survival analysis techniques including Kaplan-Meier and Cox Proportional Hazards methods.  7.4 RESULTS  There were 1186 individuals eligible for analysis, including 606 (51%) HCV-antibody positive, and 580 (49%) H C V - a n t i b o d y negative.  Baseline characteristics are  summarized i n Table 4.1 (page 90). HCV-positive individuals were less likely to be male (78% vs. 93%, p<0.001), to have had an AIDS diagnosis at baseline (11% vs. 15%, p=0.028), and to have a lower median CD4 fraction (0.19 vs. 0.16, p<0.001). There was no statistical difference in their absolute baseline CD4 count, baseline log H I V viral load, or whether they initiated ART w i t h a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor.  146  There were 163 non-accidental deaths during the study period, including 118 (72%) among the H C V - p o s i t i v e i n d i v i d u a l s , and 45 (28%) among the HCV-negatives (p<0.001).  Figure 7.1.1 displays the results of Kaplan Meier analysis, i n which H C V  serostatus is strongly associated w i t h time to non-accidental death (p<0.001), even after restricting the analysis to those w h o were over 95% adherent to their antiretroviral medication (Figure 7.1.2) i n the first year of treatment (p<0.001). Table 7.1 summarizes the unadjusted and adjusted hazards of a non-accidental death. After controlling for age at baseline, gender, having any history of injection drug use, baseline CD4 and log H I V viral load, having an AIDS diagnosis at baseline, and adherence to antiretrovirals, H C V serostatus remained strongly predictive of experiencing non-accidental mortality in this population (Adjusted Hazard 2.79, 95% confidence interval: 1.91-4.09, p<0.001).  Table 7.2. describes the causes of death in this population, including accidental deaths. H I V / H C V co-infected individuals were statistically more likely to die an HIV-related cause of death (13% vs. 6%, p<0.001), and appeared more likely to die of accidental causes (24% vs. 13%), liver-related (4% vs. 2%) and u n k n o w n causes (6% vs. 0%).  7.5 DISCUSSION  These data strongly support the hypothesis that H C V seropositivity is an independent predictor of non-accidental mortality in this previously ART-nai've HIV-infected population w h o initiated ART w i t h a triple-combination.  Our data suggest that  H I V / H C V co-infected patients are more likely to die of HIV-related, liver-related and unknown causes, compared to their H I V mono-infected counterparts.  147  Several authors have found that H C V co-infection has an adverse impact on H I V disease progression and HIV-related mortality [4, 5>17]. Indeed, there are a number of ways in which H C V could impact on HIV-related survival and mortality. Liver injury and increased A R T toxicity may preclude H I V / H C V co-infected patients f r o m tolerating antiretroviral therapy [18, 19]. Our center has previously shown that 30% of individuals i n British Columbia d y i n g of HIV-related causes have never accessed antiretroviral therapy, and these patients sociodemographic profile is consistent w i t h individuals expected to be co-infected w i t h H C V i n this province [20]. I n addition, because the progression of H C V disease is exacerbated i n the setting of H I V , it is possible that the underlying cause of death may be noted as HIV-related.  There are several key strengths to this analysis. The data are population-based, and are therefore more generalizeable than other studies. Two, by adjusting for adherence to antiretroviral medications and a history of injection drug use, we were able to account for two potentially very important confounding factors. Three, because the analysis is restricted to those individuals who initiated ART since 1996, our analysis is not subject to the k i n d of survivorship bias inherent in those analysis which analyze the survival of H I V / H C V co-infected individuals before and after 1996. Four, by excluding accidental deaths, we remove those deaths due to overdose, suicide, violence, and other causes of death that may be more strongly associated w i t h H C V co-infection because of the lifestyle associated w i t h injection drug use.  There may also be potential limitations to this analysis. One, the H C V data are based only on serology, and have not been PCR confirmed. Two, our measure of adherence is restricted to the first year of therapy. However, this was done explicitly to avoid the  148  possible reverse causation that may result f r o m those individuals w h o become less adherent to their ART because they are too sick to take the medications.  I n summary, these data suggest that hepatitis C positive antibody status is independently predictive of non-accidental mortality in this population-based program of HIV-infected individuals receiving antiretroviral therapy, even after controlling for adherence to ART and injection drug use, and these deaths are largely HIV-related. ICD codes are notoriously problematic indicators of causes of death, particularly for diseases like hepatitis C or syndromes like AIDS. However, further w o r k is urgently needed to fully characterize the mechanisms responsible for the increased mortality observed i n H I V and H C V co-infected patients.  149  7.6 REFERENCES  1.  Rockstroh, J.K. and U. Spengler, HIV and hepatitis C virus co-infection. Lancet Infect Dis, 2004. 4(7): p. 437-44.  2.  Cacoub, P., et al., Mortality among human immunodeficiency virus-infected patients ivith cirrhosis or hepatocellular carcinoma due to hepatitis C virus in French departments of internal medicine/infectious  diseases, in 1995 and 1997. Clinical Infectious  Diseases, 2001. 32: p. 1207-1214. 3.  Bica, I., et al., Increasing Mortality Due to End-Stage Liver Disease in Patients with Human Immunodeficiency Virus Infection. Clinical Infectious Diseases (CID), 2001. 32(Feb 1): p. 492-497.  4.  Greub, G., et al., Clinical  progression, survival,  and immune recovery during  antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. 5.  Daar, E., et al., Hepatitis C Virus Load is Associated with Human Immunodeficiency Virus Type 1 Disease Progression in Hemophiliacs. Journal of Infectious Diseases, 2001.183: p. 589-595.  6.  Sulkowski, M.S., et al., Hepatitis C and progression of HIV disease. Jama, 2002. 288(2): p. 199-206.  7.  Cooper, C.L., Natural history of HIV and HCV coinfection. J Int Assoc Physicians AIDS Care (Chic 111), 2003. 2(4): p. 147-51.  8.  Law, W.P., et al., Impact of viral hepatitis co-infection on response to antiretroviral therapy and HIV disease progression in the HIV-NAT  cohort. Aids, 2004. 18(8): p.  1169-77. 9.  Qurishi, N., et al., Effect of antiretroviral therapy on liver-related mortality in patients with HIV and hepatitis C virus co-infection. Lancet, 2003. 362: p. 1708-13. 150  10.  Strathdee, S., et al., Needle Exchange is not Enough: lessons from the Vancouver Injecting Drug Users Study. AIDS, 1997.11(8): p. F59-65.  11.  Patrick, D.M., et al., Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001.165(7): p. 889-95.  12.  Hogg, R., et al., Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. CMAJ: Canadian Medical Association Journal, 1999. 160(5): p. 659-65.  13.  Yeni, P.G., et al., Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society-USA Panel. Jama, 2002. 288(2): p. 222-35.  14.  Low-Beer, S., et al., Adherence to triple therapy and viral load response. J Acquir Immune Defic Syndr, 2000. 23(4): p. 360-1.  15.  Wood, E., et al., Is there a baseline CD4 cell count that precludes a survival response to modern antiretroviral therapy? Aids, 2003.17(5): p. 711-20.  16.  H o g g , R.S., et al., Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year offollow-up. Aids, 2002.16(7): p. 1051-8.  17.  Klein, M., R. Lalonde, and S. Suissa, The Impact of Hepatitis C Virus Coinfection on HIV Progression Before and After Highly Active Antiretroviral Therapy. JAIDS, 2003. 33: p. 365-72.  18.  M e l v i n , D., et al., The impact of co-infection with hepatitis C virus and HIV on the tolerability of antiretroviral therapy. AIDS, 2000.14(4): p. 463-465.  19.  Ripamonti, D., et al., Hepatitis C infection increases the risk of the modification offirst highly active antiretroviral therapy in HIV-infected patients. AIDS, 2004.18(2): p. 33436.  151  Wood, E., et al., Prevalence and correlates of untreated human immunodeficiency virus type 1 infection among persons who have died in the era of modern antiretroviral therapy. Journal of Infectious Diseases, 2003.188: p. 1164-70.  152  Table 7.1. Unadjusted and Adjusted Cox Models of Predictors of Non-Accidental Mortality  Unadjusted Hazard p-value (95% C.I.)  Adjusted Hazard (95% C.I.)  p-value  H C V serostatus 2.81 (positive vs. negative) (1.99 - 3.96  <0.001  2.79 (1.91-4.09)  <0.001  Age at baseline (per 10 years)  1.37 (1.17-1.59)  <0.001  1.33 (1.13-1.58)  <0.001  Adherence (>95% vs. <95%)  0.39 (0.29 - 0.54)  <0.001  0.40 (0.29 - 0.57)  <0.001  Any IDU (yes vs. no)  0.96 (0.68-1.36)  0.831  0.59 (0.41 - 0.85)  0.005  Baseline CD4 (per 100 cells)  0.73 (0.67-0.80)  <0.001  0.74 (0.67-0.82)  <0.001  Baseline log viral load 2.25 (per log increase) (1.37-3.71)  0.002  1.70 (1.06 - 2.73)  0.028  AIDS at baseline (yes vs. no)  1.66 (1.12-2.45)  0.011  1.10 (0.72 -1.68)  0.670  Male gender (yes vs. no)  0.77 (0.51-1.16)  0.208  1.02 (0.66-1.56)  0.940  153  Table 7.2. Causes of Death Among Previously Treatment Naive HIV Mono- and H I V / H C V Co-infected Individuals in a Population-Based HIV Treatment Program (July 1, 2000 - June 30, 2003)  Cause  Accidental HIV-related* Liver-related Cancer -Other U n k n o w n Cause  HIV 52/580 (9%)  HIV/HCV 156/606 (26%)  7 (1.2%) 35 (6%) 1 (0.2%) 2 (0.3%) 7 (1.2%) 0 (0%)  38 (6%) 79 (13%) 7 (1%) 3 (0.5%) 20 (3%) 9 (1%)  * p for HIV-related cause between H I V / H C V co-infected and H I V monoinfected is <0.001  154  Figure 7.1. Probability of Survival by H C V Serostatus Among HIV-infected Individuals Initiating Antiretroviral Therapy (accidental deaths excluded)  Figure la. Probability of Survival by H C V Serostatus 100  </)  80  —•  HCV-negative HCV-positive  Months from Starting ART  A t Months: 0 N H C V - n e g : 580 N H C V - p o s : 606  12 547 553  24 526 504  36 502 464  48 365 327  60 245 196  72 109 62  Figure l b . Probability of Survival by H C V Serostatus Among Those More than 95% Adherent to their A R T  100  O  HCV-negative  *HCV-positive  Months from Start of A R T  A t months: N HCV-neg: N HCV-pos:  0 420 255  12 407 231  24 397 216  36 384 203  48 283 138  60 189 85  72 86 29  155  CHAPTER 8: SUMMARY, RECOMMENDATIONS, FURTHER RESEARCH, AND CONCLUSIONS  8.1 SUMMARY OF STUDY FINDINGS  A t the beginning of this doctoral project, a review of available research on the issue of antiretroviral therapy in H I V / H C V co-infected patients was undertaken [1]. It became apparent that while w o r k in this area has not been extensive, what research is available suggests there are numerous outstanding questions related to H I V / H C V coinfection, including issues regarding the antiretroviral management of H I V / H C V coinfected patients. Indeed, this literature review identified several possibilities of how antiretroviral interventions could be tailored i n order to maximize H I V treatment success, and mitigate H C V disease progression.  These possibilities may include  initiating antiretroviral therapy earlier i n H I V disease progression, choosing drugs which are less hepatotoxic and avoiding certain combinations of these agents, and close monitoring of metabolic and mitochondrial abnormalities because of the potential overlap between ART toxicity and H C V morbidity.  This review also identified several areas r e q u i r i n g further research that subsequently formed the core of this doctoral project. These questions include the population-level prevalence of H C V in the HIV-infected population of British Columbia receiving antiretroviral treatment, and the potential impact of immune reconstitution on the accurate diagnosis of H I V / H C V co-infected individuals, the effect of H C V on the immunologic response to ART, issues regarding the safety and tolerability of antiretroviral agents i n H I V / H C V co-infected patients, and ultimately, the effect of H C V on mortality in HIV-positive individuals receiving antiretroviral therapy.  156  I n Chapter 4, I describe the findings of a C I H R funded study investigating hepatitis C diagnostic issues i n H I V infected individuals initiating antiretroviral therapy.  This research provides among the first population-based estimates of  H I V / H C V co-infection in the Developed World. The findings of this research indicate that the prevalence of H C V seropositivity in HIV-infected individuals i n British Columbia is much higher than previously believed, at 51%. A l t h o u g h sample quality may have been a limitation, a very high proportion (30%) of the H C V antibody positive patients had no detectable R N A at baseline. These findings are much higher than other reports of possible H C V R N A spontaneous clearance in HIV-infected populations [2, 3]. Further, our findings suggest that 20% of previously H C V R N A negative individuals w i l l develop detectable H C V R N A following ART initiation, and confirm a previous pilot report [4]. Even after taking into consideration these additional 24 patients who became H C V R N A detectable post-antiretroviral treatment, 25% of this population remained H C V R N A undetectable.  These data have important implications regarding the accurate diagnosis and management of H C V i n HIV-infected patients. They underscore the importance of testing all HIV-infected individuals for H C V , using the HCV-antibody test, prior to starting antiretroviral therapy. I n spite of the cost of PCR testing, patients should be tested for the presence of H C V R N A as w e l l in order to make informed therapeutic decisions (e.g. w h e n and whether to treat H C V , alcohol reduction, diet and lifestyle changes, etc.). If PCR negative, then patients can have peace of m i n d and yet know the absence of H C V R N A may not be permanent and that they should be re-tested f o l l o w i n g A R T i n i t i a t i o n or at regular intervals.  These data suggest that an  undetectable H C V R N A result does not necessarily mean an H I V co-infected individual has spontaneously cleared the virus.  However, if PCR positive, this knowledge  provides crucial health information. A treatment plan can be developed, including the possible effects of H C V genotype. H C V genotype has important implications for the  157  outcome of antiviral treatment [5], on the development of hepatic steatosis [6, 7], and on the development of acute transaminitis in H I V / H C V co-infected patients [8, 9]. Testing prior to ART can enable H C V treatment prior to taking ART [10], as well as tailored ART if H C V treatment is not an option or has not been effective [1]. Our data and others [2] suggest that the diagnosis of chronic H C V infection can be altered by a significant change in immune function [11, 12]. Therefore, H C V testing should be redone following antiretroviral treatment initiation.  A n important reason w h y knowing a patient's H C V status prior to initiation of ART is described i n Chapter 5, whereby some of the possible effects of H C V on the immunologic response to ART are demonstrated. The issue of immunologic response to treatment is a complex domain, and one requiring further study, particularly at the basic science level to understand the immunologic interactions between H I V and HCV. I n this study, I showed that while H C V co-infected and H I V mono-infected adults have statistically the same baseline absolute CD4 count, the H C V co-infected have a profoundly altered absolute CD4 response to treatment. Further, the H C V co-infected have a significantly higher CD4 fraction at baseline, and perhaps because of this or because of other factors, do not have a significantly altered CD4 fraction response. This may be due to physiologic factors related to H C V infection, including the possible sequestration of CD4 cells in the l y m p h nodes or other organs and cells as a result of H C V infection [13]. These data suggest that baseline absolute CD4 count does not significantly alter the impact of either absolute or fraction CD4 response.  Fundamental to the success of antiretroviral therapy i n any population is adherence [14]. I n Chapter 6, I question whether the effect of injection drug use on adherence is real, given that the vast majority of these individuals are co-infected w i t h HCV. Indeed, these data suggest that H C V infection is one of the strongest predictors of adherence to antiretroviral therapy.  I n multivariate logistic regression, H C V  serostatus at baseline is associated w i t h a 60% reduced probability of being at least 95% 158  adherent, while in the same model, injection drug use has no independent effect. After creating a 4-level variable intended to disentangle some of the confounding effects of I D U and H C V , these findings suggest that H C V has a stronger effect on the probability of adherence than does injection drug use. These data should be interpreted cautiously, as the measure of I D U in this cohort is a composite one and has not been validated.  I also hypothesized that hepatitis C w o u l d be associated w i t h poor adherence through the mechanism of increased hepatic injury, and tested this hypothesis by creating an interaction term between H C V and having any biochemical marker of liver injury. Although liver injury remains independently associated w i t h poor adherence after adjustment for H C V infection (particularly a l o w albumin level), the interaction term is not significant, suggesting that the decreased adherence among H C V co-infected patients is not due to more biochemical liver injury among them. The interpretation of these findings should be cautioned by noting that many individuals may experience increased toxicity through symptoms such as nausea, malaise, etc., that may reduce adherence, but not be captured through these dichotomized biochemical markers. Further research in these areas is urgently needed. The combined effects of under-diagnosed H C V infection, impaired immune response to ART, increased liver injury and poorer adherence to ART are culminated i n Chapter 7. Chapter 7 shows the effect of H C V serostatus at baseline on non-accidental mortality in this population of HIV-infected patients receiving antiretrovirals. In short, being HCV-positive is associated w i t h a 3 times greater risk of non-accidental death, after adjusting for adherence to antiretroviral therapy and injection drug use (and other relevant factors such as age and gender).  People w i t h H C V are more likely to die an  HIV-related death, and appear more likely to die of liver-related, accidental, and u n k n o w n causes.  There is debate regarding the effect of H C V on the progression of  H I V disease and HIV-related mortality in the era of highly active antiretroviral therapy [15, 16]. A l t h o u g h this study was not designed to answer this particular question, it 159  does suggest that H C V has a strong and independent effect on the non-accidental mortality of individuals receiving antiretroviral therapy, highlighting the need for better characterization of the causes of death in this population.  Together, the data presented here demonstrate that hepatitis C co-infection is an important clinical and public health problem in British Columbia. H C V co-infection poses substantial challenges to patients, physicians, and policy-makers. H I V / H C V coinfection is a very complex therapeutic domain, and the people i n w h o m both infections are more likely to co-exist are more likely to be marginalized socioeconomically [17,18]. These individuals may be at higher risk of concomitant mental illness and addictions [19-21], both complicating factors in the clinical setting. Although H C V infection can and should be treated, the majority of H I V coinfected persons w i l l not f u l l y benefit f r o m H C V treatment, and this places extraordinary burdens on the H I V management of these patients. This dissertation is a broad look at some of the implications of H C V co-infection on antiretroviral management, and as importantly, provides the basis and a framework for more extensive H I V / H C V co-infection research i n the future.  8.2 UNIQUE CONTRIBUTION, IMPACT, AND IMPLICATIONS  This thesis makes several unique contributions to the expansion of knowledge regarding H I V / H C V co-infection. First, I initiated and prepared the review that is presented in Chapter 2, and w i l l be published in the international medical journal AIDS. This work provides a comprehensive summary of the research in the area of H I V / H C V co-infection and antiretroviral therapy to date, and identifies several opportunities for improving the antiretroviral management of co-infected patients. This review paper contains recommendations for further areas of research and intervention. To ensure that these recommendations were relevant and comprehensive, I invited international  160  experts to provide feedback and suggestions on these recommendations and to be coauthors on this paper. When published, the review w i l l be available on Medline and w i l l be a comprehensive and thought-provoking resource for all those interested i n this area of H I V / H C V co-infection treatment and research.  Second, m y w o r k n o w enables the characterization of hepatitis C w i t h i n the HOMER cohort, and allows for the characterization of the natural history of H I V / H C V co-infection in the context of antiretroviral therapy. As a direct result of this thesis, there now exists a population-based cohort of H C V co-infected and H C V un-infected previously treatment naive individuals who initiated ART w i t h a triple-combination. There is n o w data on H C V antibody status at baseline for the entire population, qualitative H C V R N A data on those individuals w h o tested antibody positive at baseline, plus 10% of those individuals who tested antibody negative at baseline. The formation of this cohort w i l l have major implications on the capacity to lead H I V / H C V co-infection research in British Columbia for years to come. T h i r d , this study provides among the first population-based measures of H I V / H C V co-infection prevalence in the w o r l d . It suggests that H C V co-infection is prevalent in over half of HIV-infected individuals w h o first initiated H I V treatment between August 1996 and July 2000 in British Columbia, and is probably a conservative estimate of H C V co-infection in British Columbia because of the increasing number of injection drug users and others accessing ART since 2000. The high prevalence of HCV, particularly among those w i t h no reported history of injection drug use, suggests that there may be significant under-reporting of H C V infection secondary to under-testing of individuals not generally perceived to be at risk for HCV. I n addition, this study has shown that initiation of antiretroviral treatment may be related to the development of detectable H C V R N A i n those patients in w h o m it was undetectable prior to starting treatment.  161  Fourth, I have added substantially to the literature regarding immunologic response to A R T i n H I V / H C V co-infected adults.  A question of debate among  researchers and clinicians, I have highlighted some of the complexity of this question by providing novel information regarding the difference i n absolute and CD4 fraction measures of immunologic function and response, showing that H C V has an important impact on both baseline immune function and immunologic response to ART. As a result of this work, I provide a basis for multiple other research questions and a possible framework for understanding some of the conflicting results currently existing in the literature on this question. Further, through this w o r k I have identified that H I V / H C V co-infected individuals immunologic function and response should be measured through both outcomes. Although not found to be an important contributing factor, I tested the novel hypothesis that baseline CD4 w o u l d be associated w i t h the magnitude of immune response, and found that baseline CD4 has little or no effect.  Fifth, I have significantly elucidated our understanding of adherence issues i n this population. I tested the hypothesis that hepatitis C w o u l d be a stronger factor associated w i t h adherence compared to injection drug use, and I examined whether the effect of H C V was related to increased liver injury among these patients. I have shown that the question of adherence in H I V / H C V co-infected patients is complex, and is the result of the interplay between H C V infection, more liver injury (and therefore less tolerability), and a history of injection drug use. This research is the first to specifically examine the effect of H C V on adherence, and is the first to examine both H C V and injection drug use together, w i t h an attempt to separate the effects. This research opens several new lines of i n q u i r y regarding adherence, hepatitis C, and toxicity, and challenges assumptions and previous data regarding adherence among injection drug users [22j. Due i n part to the success of this effort, the BC Center for Excellence in H I V / A I D S has initiated discussions w i t h Providence Health Care to obtain real time access to safety laboratory profiles for all H I V / A I D S D r u g Treatment Program patients.  162  Finally, the study I present i n Chapter 7 is among the first to describe the profound effect of H C V serostatus on non-accidental mortality i n a population-based program of individuals receiving antiretroviral therapy.  It provides substantial  justification for further research in the domain of H I V / H C V co-infection, and provides support for initiatives aimed at increasing access to H I V / H C V treatment, care, and support. These data have important implications for health care planners and providers. There are an estimated 15,000 HIV-positive individuals i n the province of British Columbia [23]. Only 4500 are or have ever received antiretroviral therapy i n the province. If over 50% of these individuals are co-infected w i t h H C V , then a total of over 7500 people may be expected to come forward for treatment and care related to their H I V and / o r H C V infection, including the 606 individuals in the H O M E R cohort. Many of these patients may not be aware of their H C V infection. These individuals w i l l have multiple needs, and w i l l pose among the most challenging patient population for health care providers.  Adequate H I V / H C V care must address the social determinants of  health, addictions, and mental health issues. Optimally, patients w o u l d be treated i n holistic models, whereby both infectious diseases are treated in consideration of each other, and in the context of improving patient's health and quality of life more broadly. It should be anticipated that these patients may require more than average health care resources [24-31].  Indeed, as the epidemic of H I V among Vancouver's injection drug users began i n the mid-1990's [18], assuming an average progression rate of 7 to 15 years for both H I V and H C V , it can be expected that the effect of injection drug use and H I V / H C V coinfection on the health of these patients has begun to take its toll, and currently published data support this [22, 25-34].  163  8.3 RECOMMENDATIONS  The strengths and limitations of the studies presented here have been outlined i n Chapter 3, and in detail as part of each study. For instance, although antiretrovirals are centrally distributed enabling population level analyses to be conducted, a significant proportion of individuals d y i n g of H I V infection i n British Columbia die w i t h o u t accessing antiretrovirals, and are probably more likely to be co-infected w i t h HCV [31]. These patients are therefore not captured by this population-based study, and suggest our prevalence finding may be a very conservative estimate.  A n important first recommendation therefore, is to expand and enhance efforts to bring people w h o are at risk for H I V a n d / o r H C V infection into contact w i t h health care. This includes addiction management, p r i m a r y care, and mental health care. Further, I recommend that criteria for accessing H C V treatment and H C V treatment guidelines in British Columbia and elsewhere be expanded, and that H I V co-infected specific guidelines in Canada be significantly elaborated, i n particular to include issues related to antiretroviral therapy. Hepatitis C is considered o be a potentially curable infection, because a sustained virologic response (no evidence of H C V R N A 6 months after completing treatment) is believed to indicate that the infection has been eradicated [5]. Appropriate side effect management and adherence supports, and treating H C V while immune function is still strong regardless of markers of hepatic injury, w i l l maximize the probability of having a successful H C V treatment outcome in H I V infected patients [10, 35]. H I V has become a chronic manageable condition for many, but because of viral integration into the host's D N A , it is doubtful that a cure can ever be found. Therefore, antiretroviral therapy is currently anticipated to be a life-long treatment requiring a long-term strategy [36]. Effectively treating H C V w i l l maximize H I V therapeutic options and prevent downstream morbidity. More research into the clinical and economic implications of treating H C V preemptively i n HIV-positive populations is needed. 164  There are a number of clinical implications arising f r o m this body of w o r k for which I propose recommendations.  M a n y of these are contained w i t h i n the initial  literature review I conducted and have co-authored w i t h leading clinician scientists i n the field [1]. Summarized in more detail in Table 2.3, they include considering initiating ART at higher absolute CD4 counts than might otherwise be done (e.g. >350 cells/mm3) and anticipating a blunted CD4 response; avoid using d r u g or combinations of drugs known to cause mitochondrial toxicity (such as d d l , d4T, a n d / o r ribavirin); avoid using nevirapine and f u l l dose ritonavir because of their propensity for severe hepatotoxicity [37]; rely on a variety of markers of liver disease, including biopsy and ultrasound, because biochemical markers of liver injury are not very reliable; and close monitoring of H I V / H C V co-infected patients for insulin resistance and diabetes because of their heightened risk [38, 39]. Current guidelines do recommend H C V antibody testing among H I V positive individuals. Based on the literature as reviewed, and as a result of findings related to this study, it is recommended that physicians and patients consider re-testing following antiretroviral initiation.  When R N A is found to be positive, it should be immediately  genotyped because genotype 3 in particular is prone to severely elevated liver enzymes and development of fatty liver [6, 7, 37, 40]. I n addition, genotype also has a favourable HCV treatment profile [5, 35]. A number of important research recommendations arise f r o m this work. One, I have shown that absolute and fraction CD4 measures at baseline may be significantly different in H C V co- and H I V mono-infected adults, and that this can have important implications for measuring immune response. Thus, future research regarding clinical issues i n H I V / H C V co-infection should examine both measures of immunologic function. The question of immune response to ART could be tested i n a randomized clinical trial, w h i c h could obviate most major potential biases. A second research recommendation regards clinical research among people w i t h a history of injection  165  drug use. This research and others indicate the substantial clinical morbidity that can arise from H C V infection, ultimately resulting, in this study population, in a 3 times higher rate of non-accidental mortality.  In British Columbia and elsewhere, over 90%  of I D U are infected w i t h H C V [18]. Therefore the question of injection drug use i n clinical research cannot be properly examined when not also examining the effect of H C V disease.  The injection d r u g use variable used i n this dissertation does need  validation. Finally, t h r o u g h research conducted related to this dissertation and related activities, I authored a report entitled 'Roadmap for I m p r o v i n g Access to Care and Treatment for People Co-Infected w i t h H I V and Hepatitis C i n Canada'. This report was the result of a national symposium sponsored by the Canadian Treatment Action Council and co-chaired by myself. It brought together clinicians, scientists, industry, government, and consumers and their advocates to identify barriers to effective care and treatment among persons H I V / H C V co-infected, and strategies for overcoming them. The issues and recommendations arising from this meeting are summarized i n the Final Report, and cover issues regarding clinical care, research, community supports, federal and provincial policies, and industry initiatives.  The Executive  Summary of this report is included as A p p e n d i x 4 of this dissertation.  Key  recommendations arising from this initiative include the establishment of a national database of disease and treatment outcomes among H I V / H C V co-infected patients, establishment of national and provincial guidelines regarding H I V and organ transplantation (currently BC has the only policy i n Canada), elaborate on treatment and management guidelines for H I V / H C V co-infected, and broaden criteria for accessing H C V treatment in H I V co-infected patients.  166  8.4 FUTURE RESEARCH  While specific future research needs and opportunities are contained w i t h i n each chapter, this thesis points towards several broad areas in need of future research.  One is related to the issue of immune reconstitution and its effects on fibrosis progression and related issues. As noted in chapter 2, it is a difficult question to study because of the need for at least two liver biopsies pre- and post-ART initiation; liver biopsies are associated w i t h severe adverse events [5].  Biopsies themselves have  limited reliability, and for these reasons, other measures of hepatic disease and hepatic injury are urgently needed [5]. The effect of immune reconstitution has not been well established and represents an important concern given the prevalence of H C V infection and current treatment guidelines which recommend delaying ART [36]. This issue w o u l d be ideally tested in a controlled clinical trial, w i t h H I V mono- and H I V / H C V coinfected patients initiating H I V treatment at different CD4 thresholds, comparing immune response to treatment, and in a sub-study, effects of immune reconstitution on the progression of liver disease. As chapters 4 and 5 illustrate, there is a great deal about the pathophysiology of H I V / H C V co-infection that is not w e l l understood [41].  I n fact, although it is  increasingly understood that H C V is an immune-mediated disease, there remains more questions than answers in the field of hepatitis C chronic infection [5]. Indeed, although some factors are k n o w n to be associated w i t h spontaneous clearance and v i r a l eradication u p o n treatment (including younger age, higher CD4 counts, etc), the pathophysiology of H I V / H C V co-infection may hold clues to understanding more about chronic H C V infection specifically and human immunology generally.  More  bench science regarding H I V / H C V co-infection is needed, and immunologists, virologists and microbiologists w o r k i n g i n b o t h H I V and H C V should seek opportunities for cross-collaboration. 167  Qualitative and quantitative research is needed to elucidate issues related to toxicity of and adherence to ART i n H I V / H C V co-infected patients.  Although the  study presented i n chapter 6 found that there was not an interaction between any biochemical markers of liver injury and HCV-positivity, this does not mean that H C V co-infected patients are less able to fully adhere to treatment because of having more toxicity.  Indeed, this study supports other literature that H C V co-infected have  significantly elevated levels of transaminases and other markers of hepatic injury [37], but indicates that there is something else about having H C V leading to significantly poorer adherence levels (including symptoms related to liver disease not captured by the measures used). Our data suggest that the effect of H C V is independent of injection drug use.  I n addition to the need for more research regarding adherence issues, more  research is also needed to examine the additional burden of morbidity and toxicity that H C V confers to H I V infection. This includes the entire spectrum of potential toxicity, from mitochondrial and metabolic changes, to nausea, fatigue, and other non-life threatening symptoms that nonetheless have important implications for patients' quality of life.  The above are areas for future research that have emerged f r o m this specific doctoral project. There are, however, many other critical areas for future research. For example, natural history studies of H I V / H C V co-infection, treated and untreated, are urgently needed. Secondly, the area of H C V treatment in H I V co-infected patients also demands more attention. More research is needed into maximizing H C V treatment, including the use of red blood cell g r o w t h factors, increasing adherence, and the economic costs of delaying treatment. Understanding to what extent treatment can confer histologic improvement in spite of the continued presence of H C V R N A is crucial in this population, because their liver disease w i l l progress more rapidly. K n o w i n g the dynamics of w h y H I V co-infected patients respond less well to existing H C V treatments is critical.  168  The W o r l d Health Organization estimates that Africa has the highest prevalence of H C V globally [42]; it also has the highest prevalence of H I V [43]. However the prevalence of H C V / H I V co-infection is not established, and a subject of debate because the primary H I V risk factor in most places in Africa is heterosexual sex. However, H C V is highly infectious, and more likely to be transmitted through health care settings, circumcision, or the use of certain traditional medicines.  As antiretrovirals are  becoming more w i d e l y available throughout much of Africa and elsewhere i n lowincome settings, surveillance systems are needed to monitor emerging trends and conditions that may include H C V co-infection.  The area of primary prevention has not yet been addressed i n this dissertation, but not because of its insignificance. Preventing H I V and H C V disease transmission requires a variety of approaches, including harm reduction i n both drug use and sex, and p r o v i d i n g the means necessary so that people have options i n terms of their lifestyle, and health management. The prevention of these infectious agents cannot be accomplished without addressing underlying social determinants of health, including poverty, housing, and gender equality and the social and economic empowerment of women.  8.5  CONCLUSIONS  H I V co-infection significantly exacerbates H C V infection, leading to more fibrosis, cirrhosis, end-stage liver disease, and liver-related mortality [10, 41].  This  study has shown that H C V significantly complicates antiretroviral management because of the effects of immune suppression and restoration on H C V disease, an impaired immune response to treatment, increased toxicity of, and poorer adherence to, antiretroviral therapy, together resulting in elevated rates of non-accidental mortality i n 169  spite of accessing ART.  It is the interactions between the t w o viruses and their  associated effects and complications that results in a disease process that goes beyond the sum its respective parts. In many places in the w o r l d , including Thailand [44, 45], Russia [45], and China [46], burgeoning H I V epidemics are the result of widespread injection drug use. It is of paramount importance to know the burden of H I V / H C V co-infection globally, and to not assume infection rates simply from primary H I V transmission categories. Both H I V and H C V can be managed to some degree, but to do so effectively requires addressing them together, because they significantly affect each other. Therefore, H I V and H C V clinicians and scientists must collaborate together along w i t h bench scientists, policy makers, and people living w i t h H I V / H C V co-infection to move forward on those issues w h i c h can significantly enhance access to treatment and care, and maximize their effectiveness.  170  8.6 REFERENCES  1.  Braitstein, P., et al., Special Considerations Antiretroviral  Therapy Among  in the Initiation  HIV and Hepatitis  and Management  C Co-Infected  Individuals.  of  AIDS,  2004. in press. 2.  Sherman, K.E., et al., Hepatitis Human  Immunodeficiency  C Virus prevalence  Virus: a cross-sectional  among patients infected analysis  of the US adult  with AIDS  Clinical Trials Group. Clinical Infectious Diseases, 2002. 34(6): p. 831-7. 3.  Piasecki, B., et al., Influence of alcohol use, race, and viral coinfections on spontaneous HCV clearance in a US Veteran population. Hepatology, 2004. 40: p. 892-899.  4.  Cooper, C. and D. Cameron, Interpretation levels in HIV-hepatitis  of undetectable  hepatitis C virus  RNA  C virus co-infection. AIDS, 2004.18(2).  5.  Poynard, T., et al., Viral hepatitis C. Lancet, 2003. 362(9401): p. 2095-100.  6.  Rubbia-Brandt, L., et al., Steatosis affects chronic hepatitis C progression in a genotype specific way. Gut, 2004. 53(3): p. 406-12.  7.  Sharma, P., et al., Hepatic steatosis in hepatitis C virus genotype 3 infection: does it correlate with body mass index, fibrosis, and HCV risk factors? D i g Dis Sci, 2004. 49(1): p. 25-9.  8.  Nunez, M., et al., Role of hepatitis  C virus genotype in the development of severe  transaminase elevation after the introduction  of antiretroviral  therapy. JAIDS Journal of  Acquired Immune Deficiency Syndromes, 2002. 30(1): p. 65-8. 9.  Livry, C , et al., Acute liver enzyme elevations in HIV-1-infected  patients. H I V Clin  Trials, 2003. 4(6): p. 400-10. 10.  Soriano, V., et al., Care of patients with hepatitis C and HIV co-infection. AIDS, 2004. 18(1): p. 1-12.  17 1  11.  Furione, M., et al., Dissociation  of serum and liver hepatitis C virus RNA  patients coinfected with human immunodeficiency  virus and treated with  levels in  antiretroviral  drugs. J Clin Microbiol, 2004. 42(7): p. 3012-6. 12.  George, S.L., et al., Hepatitis C virus viremia in HIV-infected  individuals with negative  HCV antibody tests. J Acquir Immune Defic Syndr, 2002. 31(2): p. 154-62. 13.  Laskus, T., et al., Human  immunodeficiency  virus facilitates  infection/replication  of  hepatitis C virus in native human macrophages. Blood, 2004.103(10): p. 3854-9. 14.  H o g g , R.S., et al., Intermittent  use of triple-combination  therapy is predictive  of  mortality at baseline and after 1 year of follow-up. A i d s , 2002.16(7): p. 1051-8. 15.  Klein, M., R. Lalonde, and S. Suissa, The Impact of Hepatitis C Virus Coinfection on HIV Progression Before and After Highly Active Antiretroviral  Therapy. JAIDS, 2003.  33: p. 365-72. 16.  Greub, G , et al., Clinical antiretroviral  progression,  survival,  and  immune  recovery  during  therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss  HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. 17.  Miller, C.L., et al., The Future Face of Coinfection: Prevalence and Incidence of HIV and Hepatitis  C Virus Coinfection  Among  Young  Injection  Drug  Users. J Acquir  Immune Defic Syndr, 2004. 36(2): p. 743-749. 18.  Patrick, D.M., et al., Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001.165(7): p. 889-95.  19.  Braitstein, P., et al., Quality of life, depression, and fatigue among persons co-infected ivith HIV and Hepatitis  C: Outcomes from a population-based  cohort. AIDS Care, i n  press. 20.  Fleming, C.A., et al., Health-related quality of life of patients with HIV disease: impact of hepatitis C coinfection. Clin Infect Dis, 2004. 38(4): p. 572-8. Epub 2004 Jan 26.  172  21.  v o n Giesen, H.J., et al., Psychomotor  slowing  in hepatitis  C and HIV infection. J  Acquir Immune Defic Syndr, 2004. 35(2): p. 131-7. 22.  W o o d , E., et al., Adherence among HIV-infected  23.  to antiretroviral  therapy and CD4 T-cell count responses  injection drug users. A n t i v i r Ther, 2004. 9(2): p. 229-35.  Health Canada., Epidemiology  of the Transmission  of Bloodborne Pathogens. 2004,  Health Canada. 24.  Anis, A . H . , et al., Leaving hospital against medical advice among HIV-positive  patients.  CMAJ, 2002.167(6): p. 633-7. 25.  Palepu, A., et al., Substance  abuse treatment and emergency department  among a cohort of HIV-infected  utilization  persons with alcohol problems. J Subst Abuse Treat,  2003. 25(1): p. 37-42. 26.  Palepu, A., et al., The social determinants  of emergency department and hospital use by  injection drug users in Canada. J Urban Health, 1999. 76(4): p. 409-18. 27.  Palepu, A., et al., Impaired virologic response to highly active antiretroviral  therapy  associated with ongoing injection drug use. J A c q u i r I m m u n e Defic Syndr, 2003. 32(5): p. 522-6. 28.  Palepu, A., et al., Hospital utilization  and costs in a cohort of injection drug users.  Cmaj, 2001.165(4): p. 415-20. 29.  Wood, E., et al., Elevated rates of antiretroviral  treatment discontinuation  infected  for  injection  drug  users: implications  drug  policy  among  and public  HIVhealth.  International Journal of D r u g Policy, 2004.15: p. 133-138. 30.  W o o d , E., et al., Socioeconomic status, access to triple therapy, and survival from HTVdiease since 1996. AIDS, 2002.16: p. 1 - 8.  31.  Wood, E., et al., Prevalence and correlates of untreated human immunodeficiency type 1 infection among persons who have died in the era of modern antiretroviral  virus  therapy.  Journal of Infectious Diseases, 2003.188: p. 1164-70. 173  32.  W o o d , E., et al., Adherence antiretroviral  and plasma  HIV  RNA  responses to highly  active  therapy among HIV-1 infected injection drug users. Cmaj, 2003. 169(7):  p. 656-61. 33.  Palepu, A., et al., Uptake and adherence to highly active antiretroviral HIV-infected  people with alcohol and other substance use problems:  therapy among the impact of  substance abuse treatment. Addiction, 2004. 99(3): p. 361-8. 34.  Palepu, A., et al., Alcohol  Use and Incarceration  Suppression  Drug Users Starting  Among  Injection  Adversely  Affect HIV-1  Antiretroviral  RNA  Therapy. J Urban  Health, 2003. 80(4): p. 667-75. 35.  Torriani, F.J., et al., Peginterferon infection in HIV-infected  36.  patients. N Engl J M e d , 2004. 351(5): p. 438-50.  Yeni, P.G., et al., Antiretroviral recommendations  Alfa-2a plus ribavirin for chronic hepatitis C virus  treatment for adult HIV infection in 2002: updated  of the International  AIDS Society-USA  Panel. Jama, 2002. 288(2): p.  222-35. 37.  Dieterich, D., Managing  antiretroviral-associated  liver disease. J Acquir I m m u n e  Defic Syndr, 2003. 34(Suppl 1): p. S34-9. 38.  Butt, A., et al., Risk of diabetes in HIV infected veterans pre- and post-HAART  and the  role of HCV coinfection. Hepatology, 2004. 40(1). 39.  Duong, M., et al., Association infection  in HIV-Hepatitis  between insulin resistance and hepatitis C virus chronic  C co-infected patients undergoing  antiretroviral  therapy.  Journal of Acquired Immune Deficiency Syndromes, 2001. 27: p. 245-250. 40.  Miller, K., et al., Lactic acidosis and hepatic steatosis associated with use of stavudine: report offour cases. Annals of Internal Medicine, 2000.133(3): p. 192-196.  41.  Rockstroh, J.K. and U. Spengler, HIV and hepatitis  C virus co-infection.  Lancet  Infect Dis, 2004. 4(7): p. 437-44. 174  42.  W o r l d , H.O., Hepatitis C (Factsheet #164). 2000, W H O .  43.  U N A I D S , 2004 Report on the Global AIDS Epidemic. 2004, United Nations: Geneva.  44.  Chanbancherd, P., et al., High frequency of HIV-1 and hepatitis C co-infection young Thai men: evidence for a changing pattern of HIV transmission  in  among  Thailand.  Southeast Asian J Trop M e d Public Health, 2003. 34(3): p. 580-2. 45.  Abdala, N., et al., Estimating  the prevalence of syringe-borne and sexually  transmitted  diseases among injection drug users in St Petersburg, Russia. Int J STD AIDS, 2003. 14(10): p. 697-703. 46.  Zhang, C., et al., High prevalence of HIV-1 and hepatitis C virus coinfection injection drug users in the southeastern  region of Yunnan,  among  China. J Acquir Immune  Defic Syndr, 2002. 29(2): p. 191-6.  175  Appendix 1 S t a t e m e n t of A u t h o r s h (Appendix  Withdrawn)  Principal Applicant: Hogg et al.  Research Proposal  Appendix 2 Grant Submitted to the Canadian Institutes for Health Research, September 2003 Funding Applied for: $125,000 Funding Received: $93,000 Rating: 3.90 Preamble: This protocol was submitted to CIHR in March, 2003, and was given divergent reviews (one reviewer rated it Excellent, the other as requiring revision — summary score: 3.4). We have revised the protocol based on the reviewers comments. Please refer to the Responses to Previous Reviews for details of the reviews and our responses.  A. Specific Aims The overarching aim of this project is to develop an appropriate testing algorithm for hepatitis C virus (HCV) among an HIV-infected population seeking HIV treatment in a Developed World setting with universal access to health care (British Columbia, Canada). To this end, this study has the following objectives and hypotheses: Primary  Objective:  To measure the point prevalence of Hepatitis C infection among HIV-positive individuals initiating Highly Active Antiretroviral Therapy (HAART) in a population-based HIV treatment cohort. Primary  Hypothesis:  That there will be an overall Hepatitis C prevalence of 30% among this population (defined as having positive HCV antibodies and a positive HCV RNA test; or being antibody-negative but having evidence of HCV RNA; or being antibody-positive without evidence of HCV RNA) at the time of initiation of antiretroviral therapy, but a true prevalence of hepatitis C (defined only as either having positive antibodies and a positive RNA test; or being antibody-negative but having evidence of HCV RNA) of 20%. (Please see Appendix One for a Glossary of terms and acronyms.) Secondary Objectives:  To describe the sociodemographic and clinical factors associated with HCV prevalence. (Hypothesis: That male gender, older age, and a history of injection drug use will be associated with prevalent HCV cases, in both definitions of prevalent.) To measure the positive and negative predictive values of the currently used HCV-antibody test among HIV-positive individuals who have initiated HAART since 1996. To quantify the prevalence of discordant HCV-antibody and PCR test results among HIVpositive individuals. (Hypothesis: That there will be a prevalence of 5% of antibody-negative but RNA-positive discordant responses; and a prevalence of 15% of antibody-positive, but RNA-negative discordant responses.) To describe the sociodemographic and clinical factors associated with discordant test results. (Hypothesis: That compared to people who are True HCV Positives , people with antibodypositive, but RNA-negative discordant responses will be more likely to be female and to be of median younger age; and that antibody-negative but RNA-positive discordant responses 177  Principal Applicant: Hogg et al.  Research Proposal  will be more likely to have a baseline CD4 count below 200 cells/mm .) (Please see Methods section and/or Figure 1 for definitions of True H C V Positives , etc.) 3  B. Background and Rationale It is estimated that as much as 3% of the world s population is infected with the Hepatitis C virus (HCV) [1], including an estimated 2% of people from the United States [2], and 1% of Canadians [3]. As of December, 2002, the United Nations had estimated that there were 42 million people living with HIV/AIDS in the world [4], including approximately 40,000 in Canada [5]. Due to shared routes of transmission such as receiving contaminated blood products and injection dmg use, the prevalence of co-infection by both HIV and Hepatitis C is considerable, particularly in some populations. In Canada and elsewhere in the Developed World including the United States, it is estimated that as many as 30% of individuals who have HIV also have Hepatitis C [6-8]. There is an important negative synergy between HIV and Hepatitis C whereby people who are co-infected with both viruses have faster rates of liver fibrosis and development of cirrhosis compared to H C V mono-infected individuals [9-17], poorer responses to Hepatitis C treatment [8, 18-23], and overall worse survival [8, 12, 24-28]. These negative outcomes appear to be at least partially mediated through factors that are sometimes modifiable and sometimes not. These factors include CD4 count and other immunological factors [29-32], type and length of Hepatitis C treatment [18-20, 33-37], use of antiretroviral therapy for HIV infection [22, 31, 38-45], age and gender [10, 16, 46, 47]. Due to these co-factors, a consensus is developing that among HIV-infected individuals, Hepatitis C needs to be detected and managed as early as possible [17, 22, 45, 48-51]. However, several issues inhibit the timely and accurate diagnosis of Hepatitis C among HIV-infected individuals. One of these may include a misperception among some primary health care providers that only high-risk individuals (i.e. injection drug users, hemophiliacs) are at risk for Hepatitis C, and therefore do not proactively test all HIV-positive patients. Indeed, only 552 of the 1422 individuals who have initiated antiretroviral therapy in British Columbia since 1996 have documentation of H C V serology. Factors associated with not getting tested include being female, and having a physician with less HIV experience (unpublished data). A second factor preventing the timely and accurate diagnosis of hepatitis C in the HIV-infected population is the predisposition, due to an altered immunologic capacity, of HIV-infected individuals to test negative for H C V antibodies, but to have detectable H C V R N A when tested using P C R methods. Although Thio et al. (2000) found that inconsistent antibody and P C R results were primarily related to seroconversion [52], our center has confirmed other reports [23, 53] which suggest that approximately 5% of individuals who test antibody-negative may have detectable H C V R N A with no evidence of seroconversion [54]. If accurate, this would mean that at least 50 of the 1422 individuals who have started antiretroviral therapy since 1996 may have been misdiagnosed as not having Hepatitis C. PCR testing for the detection of H C V R N A is not routinely done in British Columbia and elsewhere because of the cost of the test, reinforcing the need for the development of an appropriate testing algorithm in this population.  178  Principal Applicant: Hogg et al.  Research Proposal  A third factor in the potential for misdiagnosis of Hepatitis C is the unknown rate of H C V clearance in the HIV infected population. In the H C V mono-infected population, it is estimated that approximately 15% of antibody-positive individuals will not develop chronic infection due to spontaneous clearance of H C V viremia [55,56]. It is not currently known whether the same proportion of individuals with H I V infection will test antibody-positive but not have any detectable virus. People who are infected with HIV at the time of H C V acquisition may have higher H C V viral loads during seroconversion compared to people without HIV [57], and this would suggest that HIV-infected individuals may be less likely to spontaneously clear H C V . Factors associated with spontaneous clearance in the non-HIV-infected population include younger age, and female gender [2]. Distinguishing between Overall Prevalence and True Prevalence has a number of important clinical and policy related implications. Although it is important to know who has positive antibodies for Hepatitis C, these individuals do not represent the burden of hepatitis C disease. In contrast, individuals who have evidence of H C V R N A are those who are at risk for disease progression, who may require treatment, and who are at greater risk of transmitting the virus. Although the use of Highly Active Antiretroviral Therapy (HAART) has largely transformed HIV disease into a chronic, manageable illness [40], most i f not all antiretroviral agents are primarily metabolized through the liver, and can cause varying degrees of hepatotoxicity [43, 44]. Hepatotoxic drugs are known to exacerbate H C V infection [2]. Conversely, viral coinfection can further exacerbate drug-related hepato-toxicities [58-60]. Despite the gains made by the use of antiretroviral therapy, viral co-infection with Hepatitis C has already become a leading cause of morbidity and mortality among people living with HIV [14, 59, 61-63]. These issues all accentuate the importance of early and accurate diagnosis of Hepatitis C in the HIVinfected population. In summary, although the timely and accurate diagnosis of H C V in HIV-positive individuals is of the utmost clinical importance, there are several issues which prevent a valid and reliable estimate of active H C V infection in this population. While it is generally considered that 30% of persons with H I V are co-infected with Hepatitis C, there are several reasons why these estimates may be either under or over-estimates. The burden of Hepatitis C among HIV-positive individuals is clinically significant and complex, and the accurate and timely diagnosis of these individuals is essential. C. Roles and Contributions of Investigators Robert Hogg, PhD. Principal Investigator (5% effort). Dr. Hogg is responsible for overseeing and advising on study design, interpretation and application of results, and development of publications. Paula Braitstein, PhD (Cand). Co-investigator, and graduate student whose dissertation topic is HIV/Hepatitis C co-infection (30% effort). Responsible for developing project design and implementation, in consultation with co-investigators, including all statistical analysis and development of publications. P. Richard Harrigan, PhD. Co-investigator (1% effort), virologist and head of the H I V laboratory, responsible for advising on technical components of the assays and virologic aspects of the study. Mel Krajden, M P . Co-Investigator (5% effort), responsible for advising on technical components of assays, and interpretation and implications of findings on a policy level. 179  Principal Applicant: Hogg et al.  Research Proposal  Val Montessori, M P . Co-investigator (5% effort), a physician specializing in HIV/Hepatitis C co-infection, responsible for advising on clinical aspects and implications of research questions and results. Julio S.G. Montaner, M P . Co-investigator (2% effort), physician specializing in H I V and antiretroviral therapy, responsible for advising on clinical aspects and implications of research questions and results. Michael V . O'Shaughnessy, PhP. Co-investigator (2% effort), a virologist and senior policy director, responsible for advising on both scientific and policy implications of study in design, analysis, interpretation. Chris Sherlock. M P . Co-investigator (10% effort), Pirector of lab where testing will be performed, Pr. Sherlock is responsible for leading technical aspects of assay utilization and testing of samples. Martin T. Schechter, M P . PhP. Co-investigator (2% effort), a clinical epidemiologist responsible for advising on study design and analysis.  D . Research Design and Methods Data Source: T h e H I V / A I D S D r u g Treatment P r o g r a m ( D T P )  Antiretroviral medications have been centrally distributed at no cost to eligible HIV-infected individuals since 1986. In October 1992, the distribution of antiretroviral agents became the responsibility of the H I V / A I P S Prug Treatment Program of the British Columbia Center for Excellence in HIV/AIPS. This antiretroviral drug distribution program remains the only free source of antiretroviral medication in this Canadian province (and is a unique program in Canada). The Centre's HIV/AIPS Prug Treatment program has received ethical approval from the University of British Columbia Ethics Review Committee at its St. Paul's Hospital site, and the program conforms with the province s Freedom of Information and Protection of Privacy Act. The Center distributes antiretroviral drugs based on specific guidelines generated by the Therapeutic Guidelines Committee [64]. In 1992, the HIV/AIPS Prug Treatment Program made available double combination therapy for individuals with C P 4 cell counts of 350/mm or less. In Pecember 1995, double combination therapy was made available to everyone with C P 4 cell counts of 500/mm or less. In June 1996 the Centre adopted plasma viral load driven antiretroviral therapy guidelines, consistent with those put forward by the International A I P S Society — USA[65]. In brief, antiretroviral therapy na ve individuals with plasma viral load > 100,000 copies/ mL were offered triple drug regimens (i.e. two nucleosides plus a protease inhibitor (PI) or a non-nucleoside reverse transcriptase inhibitor (NNRTI)), while those with plasma viral loads from 5,000 to 100,000 copies/mL were offered dual nucleoside therapy. The Centre guidelines were revised in July 1997 to recommend triple combination therapy for all antiretroviral na ve individuals with plasma HIV-1 R N A levels greater than 5,000 copies/mL or C P 4 cell counts below 500 cells/mm . The Centre s guidelines recommend further that plasma HIV-1 R N A levels be monitored at baseline, at 4 weeks after starting antiretroviral therapy and every three months thereafter. Plasma viral loads are measured using the Amplicor HIV-1 Monitor" (Roche Piagnostics Branchburg, NJ). 3  3  3  A l l classes of federally licensed antiretroviral drugs are currently available through the program, including all nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse 180  Principal Applicant: Hogg et al.  Research Proposal  transcriptase inhibitors (NNRTI), and protease inhibitors (PI). Tenofovir, Atazanavir, and Enfuvirtide are also available. Eligibility for antiretroviral medication has remained consistent with current international recommendations [45]. Individuals are automatically entered into the DTP when they are first prescribed any antiretroviral agent. At DTP entry and with each subsequent physician visit, the participant s complete history (if any) of antiretroviral use, CD4 cell count, plasma viral load, and disease stage are recorded. Typically, patients are followed-up at 3 month intervals, at which time prescriptions are renewed or altered based on treatment success and other clinical factors. Blood drawn for the purposes of CD4 and viral load testing is stored for each individual at each followup visit for future research activities related to HIV disease. Study Population: The population of interest for this study are the 1422 individuals whose first ever antiretroviral combination consisted of a triple-drug regimen of either two NRTI s plus a PI, or two NRTI s plus a NNRTI, and who initiated therapy between July 1996 and August, 2000. Study Design and Outcome Measures: Frozen, archived samples of blood plasma are available to the Drug Treatment Program. The first sample taken, that from just prior to H A A R T initiation, will be tested for H C V antibodies (Ab) and H C V R N A , the latter using polymerase chain reaction (PCR) technology. This will establish a baseline point prevalence of Hepatitis C for this cohort, and help to inform the degree to which immune suppression contributes to the test s predictive values, and discordant test results. For individuals with negative-antibody but positive PCR results, a sample obtained from them at least six months later will be tested to determine whether the discordant result was due to seroconversion. For individuals with positive antibodies but negative PCR results, a second test will be performed to confirm the absence of H C V replication, as recommended by the National Institutes of Health [2]. To determine sociodemographic and clinical factors associated with prevalent cases (defined in two ways to evaluate differences in associated factors depending on the definition), and having discordant test results (each of false-antibody-negative and false-antibody-positive , the following outcomes will be assessed in a cross-sectional analysis using multivariate logistic regression (see Table 1): • • •  True Prevalence (true Ab-positives plus false Ab-negatives versus true Ab-negatives) Overall prevalence (true Ab-positives plus false Ab-negatives plus false Ab-positives versus true Ab-negatives) Discordant results: a) Antibody-positive, R N A negative versus true Ab-positives b) Antibody-negative, R N A positive versus true Ab-positives  Table 1. Dependent Variables Defined Outcome True  HCV  Definition Coded as 1 true positives plus false  Definition Coded as O true negatives 181  Principal Applicant: Hogg et al. Prevalence Overall HCV Prevalence Discordant (Antibody positive, RNA negative) Discordant (Antibody negative, RNA positive)  Research Proposal  negatives true positives plus false negatives plus false positives Antibody-positive, R N A negative Antibody-negative, RNA positive  true negatives true positives  true positives  Definitions (see Figure 1):  True-negative : those individuals who test HCV negative on both antibody and PCR; True-positive : those individuals who test HCV positive on both antibody and PCR; Seroconverters : those whose first HCV test showed negative antibodies but was positive on PCR, and whose follow-up test at least six months later shows positive antibodies; these individuals will subsequently be classed as true positives. False antibody negative : those individuals whose initial and confirmatory tests showed negative antibodies and positive PCR. False antibody positive : those whose initial and confirmatory tests showed positive antibodies and negative PCR, and for whom there is no record of hepatitis C treatment prior to testing. (Please note that the term false antibody positive is used for convenience to refer to people who it is presumed have spontaneously cleared the hepatitis C virus.) I f an individual is found, upon confirmatory testing, to have both positive antibodies and a positive PCR test, they will subsequently be classified as a true-positive.  Figure 1. Antibody Test  + RNA PCR Test RNA  -  true positives  false antibody negative  false antibody positive  true negatives  182  Principal Applicant: Hogg et al.  Research Proposal  HCV Screening: Blood samples are collected and stored frozen at -20C until time of processing. Plasma is separated within 6 hours of collection by centrifuging at 800-1600xg for 15 minutes at room temperature. Each consenting individual s sample will be tested for both HCV antibodies and for the presence of HCV RNA. Blood samples will be tested for HCV antibody using the dual enzyme immunoassay (EIA) algorithm (Abbott primary screen followed by Ortho HCV test i f primary screen is positive). These third generation EIA assays have a sensitivity greater than 99% and a specificity of 99%, in immune-competent patients [2]. HCV RNA will be detected using the Roche Cobas HCV AmpliPrep/COBAS Amplicor test. The AmpliPrep/COBAS AmplicorHepatitis C Virus test (v.2.0) is a qualitative nucleic acid amplification test for the detection of HCV in clinical specimens of human plasma. Sample preparation is automated using the COBAS AmpliPrep Instrument, and amplification and detection are automated using the COBAS Amplicor Analyzer. This is a qualitative assay with a lower level of detection of HCV RNA of 100 copies per ml. The specificity for this assay exceeds 98% [2]. A l l testing will be completed at the University of British Columbia Virology Laboratory at St Paul's Hospital. The Standard Operating Procedures for these assays have been included as an appendix to this proposal (Appendix Two). Statistical Methods: Hepatitis C prevalence will be calculated at baseline for the entire cohort of 1422 individuals. Individuals will be considered true prevalent cases i f their sample is antibody-positive and PCR-positive or i f they have been found to be false negative cases. A second overall prevalence measure will be calculated to include both of these groups, in addition to individuals with positive antibodies but negative PCR results. Standard calculations of sensitivity and specificity, and proportions of false-positives and falsenegatives for the antibody tests will be performed. We will calculate the positive and negative predictive values of the HCV antibody test, stratifying by baseline CD4 count (>500 cells/mm , 200-500 cells/mm , 100-199 cells/mm , and <100 cells/mm ), enabling the development of an appropriate testing algorithm for an HIV infected population (based on the hypothesis that immune suppression is related to discordant results). 3  3  3  3  To describe factors associated with prevalent cases, and discordant test results, cross-sectional logistic regression multivariate analysis will be used. To assess bivariate associations, Student s t-test will be used for normally distributed continuous variables, and the Wilcoxon Rank Sum test for non-normal continuous variables. Pearson s Chi-Square test will be used for categorical data in which all the cells have a minimum of five observations, and i f any of the cells contain five or fewer observations then Fisher s Exact Test will be used. Factors independently associated with the outcomes will be assessed using multivariate logistic regression. The multivariate model will be built by entering those variables which are bivariately statistically significant (p<0.05) (all variables assessed in bivariate analyses will have previously been hypothesized to be of potential clinical significance), assessing potential statistical interactions and confounders as appropriate, beginning with the most significant variables first. Independent variables to be considered will be: 183  Principal Applicant: Hogg et al.  Research Proposal  Baseline sociodemographics: gender (male/female), age (continuous measure), primary HIV risk group (sexual, drug use, other) (note that ethnicity will not be examined as data on ethnicity is not routinely collected) Baseline clinical characteristics: CD4 count (continuous measure), HIV viral load (continuous measure), AIDS diagnosis (yes/no) Power for Primary Hypothesis: (see Figure 2) At an alpha of 0.05, a sample size of 1422 individuals, using a null probability of 15%, and an alternative probability of 30% (the primary hypothesis being that there will be a 30% overall prevalence of hepatitis C at baseline in this cohort) this study has over 95% power. Power for Secondary Objectives: If, as the literature suggests, it is assumed that 5% of those who test antibody negative in fact have detectable HCV RNA (this would be the secondary outcome with the least power), at an alpha of 0.05, sample 1 consisting of 50 false negatives, sample 2 consisting of 363 true positives, assuming that 25% of false negatives have a baseline CD4 below 200 cells/mm compared to 15% of true positives (please refer to secondary study hypotheses), this analysis would have over 90% power. 3  Figure 2. 1422 individuals in cohort  (Assume 30% HCV prevalence based on antibody test) •  427 people antibody-positive and 995 people antibody-negative • 15% of these Ab+ may be PCR- (15% of 427) = 64 5% of Ab- may be PCR+ (5% of 995) = 50 True positives (427-64) = 363  £ . Study Limitations Due to cost constraints, we are not able to test the entire group of nearly 4000 individuals currently receiving antiretroviral therapy in the province of British Columbia. We hope that selecting antiretroviral na ve individuals who have initiated HAART will enable us to control for the potential confounding effects of pre-treatment, while providing extensive information regarding HCV testing in individuals initiating and receiving antiretroviral therapy in the HAART-era.  184  Principal Applicant: Hogg et al.  Research Proposal  A second limitation of this study is that we have chosen not to quantify the HCV RNA or to genotype the samples. We feel that to do so would be beyond the scope of this particular project, and is considered a next step . Third, obtaining consent for each individual may present logistical difficulties, perhaps resulting in a selection bias. However, because of the important nature of the information to patients and their physicians, we anticipate that there will be a high degree of compliance. A comparison of those individuals who provide consent to those who do not will be performed to elucidate any bias occurring. There may be a response bias introduced into the study by obtaining consent from patients via their physicians, in that patients who have been lost-to-follow up in their routine clinical care may be more likely to have hepatitis C (by virtue of being more likely to be an injection drug user and therefore not consistently followed). We will have these patients baseline information, including risk group, and will endeavor to quantify the bias related to this issue, and control for it as much as possible in the analysis. Fourth, although we are endeavoring to accurately classify individuals as true cases or not, there is potential for misclassification i f a seroconversion takes longer than the period over which individuals are to be tested (6 months) (i.e. they would be counted as a discordant response rather than as a true positive). Misclassification may also occur i f individuals who had initially tested antibody-negative but PCR-positive, but who upon re-testing developed positive-antibodies but due to an activated immune response (since by this point they will have started antiretroviral therapy), rather than due to seroconversion (i.e. they would be classified as a true positive rather than discordant). A fifth and final limitation of this analysis is that ideally we would calculate the incidence of hepatitis C in this population from 1996 until the end of 2002, thereby enabling both an understanding of the incidence of HCV in this population, as well as an accurate and current point prevalence, and the ability to conduct prospective analyses. Unfortunately, the costs of PCR testing are prohibitively expensive, resulting in a modified proposal that will at least enable us to target our PCR testing in the clinical setting for all individuals. F . Study Strengths  This investigation has several important strengths. The study population is a population-based cohort, and therefore has the advantage of being broadly representative of the population of people living with HIV and Hepatitis C who have initiated antiretroviral therapy in the HAART era, in the Developed World. Thus, the study will be more generalizable than other patient populations in whom such an investigation could occur (eg. hospital cohorts, clinical trial, etc.). Second, it is an investigation that, although relatively simple and straightforward in its concept and approach, will have significant clinical implications. While the cohort will be small enough to make the project feasible, the cohort is large enough to enable powerful results. A third important strength is that while this particular investigation is cross-sectional, because the testing will be conducted on samples drawn and stored at baseline, and because the testing will not be anonymous and will eventually be linked with patients other records existing through the HIV/AIDS Drug Treatment Program, the potential exists for extensive subsequent prospective analyses to be performed regarding the impact of Hepatitis C on HIV-positive individuals. Conversely, without doing this kind of study, any further analyses conducted on coinfected individuals in our cohort will be potentially biased both through selection bias of those 185  Principal Applicant: Hogg et al.  Research Proposal  who have received a hepatitis C test, and through misclassification bias related to the HCV antibody vs. PCR status issue. Fourth, all the testing will be done using the same assays, in the same laboratory, eliminating any bias due to differences in assays or laboratories. G . Projected Outcomes  It is anticipated that this research will yield important results of direct clinical interest to patients, their physicians, and policy makers. Specifically, this project will: S quantify the prevalence of hepatitis C infection in a population-based cohort of HIV infected individuals initiating antiretroviral treatment; S contrast HCV prevalence depending on how it is defined; •S provide important information to clinicians and policy-makers regarding the burden of Hepatitis C among HIV-positive individuals initiating antiretroviral therapy; S calculate the sensitivity, specificity, positive and negative predictive values of the HCV antibody test relative to a qualitative PCR test for HCV RNA in an HIV-infected population; •S describe characteristics associated with prevalent cases, and discordant test results; •S enable the development of an appropriate HCV testing algorithm for the HIV clinical setting. H . Tentative Timetable  August 2003: January 2004: January 2004 — October 2004: February 2004 — October 2004: November 2004 — December 2004:  Obtain Ethics Approval Receive notification of funding Obtain informed consent from participants Test stored samples Analyze results and prepare abstracts/ manuscripts; Communicate results to patients through their physicians.  I. E t h i c a l a n d I n f o r m e d C o n s e n t Issues  Human Subjects Involvement:  As this will be a retrospective analysis, the involvement and enrolment of participants will be restricted to requesting their consent to test their previously stored blood samples. Human subjects will only be involved via requests, through their physicians, to sign informed consents. No new data, blood samples, or interventions of any kind will be required for this study. Sources of Material:  No new data or blood will be required of participants. Just prior to initiation of antiretroviral therapy (at baseline), participants provided blood samples for storage for future clinical research purposes. Participants will be asked, by their family physicians, to sign an informed consent allowing us to test their stored blood samples for the purposes of the study. The samples are stored by identification numbers for which only two people in the BC Center for Excellence have links to personal identifying data (Data Managers). A l l analyses, including both the testing of the blood and the statistical analyses, will be done using only the identification number, and there will not at any point be an opportunity for anyone involved in the study to know the names of the individuals being tested. Potential Risks:  186  Principal Applicant: Hogg et al.  Research Proposal  As there is no intervention intended for this study, the risks to study participants are minimized, especially physical risks. There is a psychological risk to finding out that one has Hepatitis C when one previously did not know (because of not being tested) or because one had falsely tested antibody-negative. However, the benefit of knowing whether one has hepatitis C or not would outweigh the risk associated with the knowledge, given the seriousness of having hepatitis C. The fact that the results will be communicated by patients physicians w i l l alleviate psychological risks to the patient. Patients may request a Hepatitis C antibody test from their physicians in the context of routine care. However, qualitative PCR tests are not routinely done. Recruitment and Informed Consent:  Recruitment for this study will be limited to requesting individuals consent, through their physicians, to test their previously stored blood samples. An explanatory letter to all antiretroviral prescribing physicians will be sent, with copies of the informed consent for them to present to patients. Patients will not be directly contacted by the BC Center for Excellence in HIV/AIDS. Patients will be requested by their physicians to sign the informed consent, in the context of their routine care. The informed consent is a 2-page form which explains in lay language the purpose, reason, and methods of the study, clearly articulating that only stored blood will be tested (see Appendix 3). The consent forms will then be sent back to the Data Managers of the BC Center for Excellence in HIV/AIDS, by the physician, in the same manner that antiretroviral prescriptions, requests for HIV resistance testing, and a number of other confidential requests are communicated to the BC Center for Excellence in HIV/AIDS. The Data Managers will document whether consent was given or denied, and whether the patient is lost-to-follow-up, in an Oracle Database that contains all other patient information. From this point, all data that is accessed utilizes only identification numbers, and is stripped of all personal identifying information. To date, there has not yet been a known breach of confidentiality within the BC Center for Excellence in HIV/AIDS. We are therefore reasonably confident that the system that exists for protecting patient confidentiality will be reliable. Potential Benefits of Proposed  Research  Although screening for hepatitis C among HIV-infected persons is considered standard of care in British Columbia, many physicians do not routinely test their patients. Routine screening involves only antibody testing, with qualitative PCR testing only in exceptional circumstances. Therefore, a direct benefit for patients will be the dual testing of their blood, enabling a much more accurate diagnosis of their hepatitis C status. I f positive, patients will be then be enabled to pursue appropriate treatment and management options (e.g. HCV treatment, adapting lifestyle, etc.), and i f negative, patients will have the opportunity to be appropriately counseled regarding avoiding hepatitis C transmission. This study will help raise the profile of the burden of hepatitis C in British Columbia and the rest of Canada, and will encourage patients and physicians to test all HIV-infected patients. Benefits from this study will also be conferred on a population-level. This study will provide comprehensive diagnostic data that will enable a wide variety of epidemiological and clinically relevant investigations in the future (with appropriate ethical approval for each). This research can then be applied to improved patient management. Importance of the Knowledge to be Gained  As previously discussed, the knowledge to be gained from this project will be substantial on both an individual patient-level, and on a population-level. To our knowledge, such a study has not yet been performed, particularly in an HIV-infected population, in whom there is evidence to 187  Principal Applicant: Hogg et al.  Research Proposal  suggest the results may be different. This study will contribute substantially to the field of Hepatitis C diagnostics, and may also provide insight on an epidemiological level into the interaction between HIV and Hepatitis C. Also to our knowledge, there is no other population-based HIV treatment cohort for whom extensive hepatitis C data also exists. This cohort has contributed significantly to the field of HIV clinical management [40, 66-68], and has the potential to address many of the important, and largely understudied, questions outstanding in the area of the clinical management of people who are co-infected with HIV and Hepatitis C [2].  188  Principal Applicant: Hogg et al.  Research Proposal  References Cites  1. 2.  3. 4. 5. 6.  7. 8.  9.  10.  11. 12.  13.  14. 15.  16. 17. 18. 19.  Lavanchy, D., Global surveillance and control of hepatitis C. Journal of Viral Hepatitis, 1999. 6: p. 35-47. Anonymous, National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C 2002 (June 10-12, 2002). Gastroenterology., 2002. 123(6): p. 2082-99. Canada, H., Epidemiology of the transmission of blood-borne pathogens. 1997, Laboratory Centre for Disease Control. AIDS, TJ., AIDS epidemic update, . 2002, Joint United Nations Programme on HIV/AIDS: Geneva. Canada, H., HIV/AIDSEpi Update,. 2000, Laboratory Centre for Disease Control. Sherman, K.E., et al., Hepatitis C Virus prevalence among patients infected with Human Immunodeficiency Virus: a cross-sectional analysis of the US adult AIDS Clinical Trials Group. Clinical Infectious Diseases, 2002. 34(6): p. 831-7. Patrick, D.M., et al, Incidence of hepatitis C virus infection among injection drug users during an outbreak of HIV infection. Cmaj, 2001. 165(7): p. 889-95. Bonacini, M. and M. Puoti, Hepatitis C in patients with human immunodeficiency virus infection: diagnosis, natural history, meta-analysis of sexual and vertical transmission, and therapeutic issues. Arch Intern Med, 2000. 160(22): p. 3365-73. Benhamou, Y., et al., Liver fibrosis progression in human immunodeficiency virus and hepatitis C virus coinfected patients. The Multivirc Group. Hepatology., 1999. 30(4): p. 1054-8. Benhamou, Y., et al, Factors affecting liver fibrosis in human immunodeficiency virus and hepatitis c virus coinfected patients: impact of protease inhibitor therapy. Hepatology, 2001. 34: p. 283-287. Bonacini, M., et al, Patients co-infected with HIV and hepatitis C virus demonstrate higher levels of hepatic HCV RNA. Journal of Viral Hepatitis, 1999. 6(3): p. 203-8. D i Martino, V., et al., The influence of human immunodeficiency virus coinfection on chronic hepatitis C in injection drug users: a long-term retrospective cohort study. Hepatology, 2001. 34(6): p. 1193-9. Garcia-Samaniego, J., et al., Influence of hepatitis C virus genotypes and HIV infection on histological severity of chronic hepatitis C. American Journal of Gastroenterology, 1997. 92(7): p. 1130-4. Graham, C.S., et al, Influence of human immunodeficiency virus infection on the course of hepatitis C virus infection: a meta-analysis. Clin Infect Dis, 2001. 33(4): p. 562-9. Puoti, M., et al., Liver fibrosis progression is related to CD4 cell depletion in patients coinfected with hepatitis C virus and human immunodeficiency virus. Journal of Infectious Diseases, 2000. 183(1). Rai, R., et al, Severity and correlates of liver disease in hepatitis C virus-infected injection drug users. Hepatology, 2002. 35(5): p. 1247-55. Talal, A.H., P.W. Canchis, and I. Jacobson, The HCV and HIV Coinfected Patient: What Have We Learned About Pathophysiology? Curr Gastroenterol Rep, 2002. 4(1): p. 15-22. Cotler, S.J. and D.M. Jensen, Treatment of hepatitis C virus and HIV co-infections. Clin Liver Dis, 2001. 5(4): p. 1045-61. Gaglio, P.J., et al, The effect of HIV coinfection on hepatitis C: A review. J La State Med Soc, 2001. 153(11): p. 552-8. 189  Principal Applicant: Hogg et al. 20. 21. 22.  23. 24. 25.  26.  27. 28. 29.  30. 31.  32.  33. 34.  35. 36.  37.  38. 39.  Research Proposal  Fukuda, M., et al, Predictive factors in eradicating hepatitis C virus using a relatively small dose of interferon. Journal of Gastroenterology & Hepatology, 1998. 13(4): p. 412-8. Liang, T., et al, Pathogenesis, Natural History, Treatment, and Prevention of Hepatitis C. Annals of Internal Medicine, 2000. 132(4): p. 296-305. Pol, S., A. Vallet-Pichard, and H. Fontaine, Hepatitis C and human immune deficiency coinfection at the era of highly active antiretroviral therapy. J Viral Hepat, 2002. 9(1): p. 1-8. Tedaldi, E. and P. Bean, Diagnosis, evaluation, and treatment of HIV/HCV coinfection. A m Clin Lab, 2001. 20(9): p. 26-32. Garcia-Samaniego, J., et al, Hepatocellular carcinoma in HIV-infected patients with chronic hepatitis C. Am J Gastroenterol, 2001. 96(1): p. 179-83. Greub, G., et al, Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: The Swiss HIV Cohort Study. Lancet, 2000. 356: p. 1800-1805. Silvestri, F., et al, Impact of hepatitis C virus infection on clinical features, quality of life and survival of patients with lymphoplasmacytoid lymphoma/immunocytoma. Annals of Oncology, 1998. 9(5): p. 499-504. Staples, C.J., D. Rimland, and D. Dudas, Hepatitis C in the HIV Atlanta Veterans' Affairs Cohort Study. Clinical Infectious Diseases, 2000. 29(1): p. 409-10. Yee, T.T., et al, The natural history of HCV in a cohort of haemophilic patients infected between 1961 and 1985. Gut, 2000. 47(6): p. 845-51. Chung, R., et al, Immune recovery is associated with persistant rise in hepatitis C virus RNA, infrequent liver test flares, and is not impaired by hepatitis C virus in co-infected subjects. AIDS, 2002. 16(14): p. 1915-1923. Daar, E.S., et al, Relation between HIV-1 and hepatitis C viral load in patients with hemophilia. J Acquir Immune Defic Syndr, 2001. 26(5): p. 466-72. Greub, G., et al, Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet, 2000. 356(9244): p. 1800-5. Yokozaki, S., et al, Immunologic dynamics in hemophiliac patients infected with hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. Blood, 2000. 96(13): p. 4293-9. Cozzolongo, R., R. Cuppone, and O.G. Manghisi, The Treatment of Chronic Hepatitis C not Responding to Interferon. Curr Pharm Des, 2002. 8(11): p. 967-75. Ikeda, K., et al, Long-term interferon therapy for 1 year or longer reduces the hepatocellular carcinogenesis rate in patients with liver cirrhosis caused by hepatitis C virus: a pilot study. Journal of Gastroenterology & Hepatology, 2001. 16(4): p. 406-15. Rodriguez-Rosado, R., et al, Management of hepatitis C in HIV-infected persons. Antiviral Res, 2001.52(2): p. 189-98. Schlaak, J.F., et ah, Sustained suppression of HCV replication and inflammatory activity after interleukin-2 therapy in patients with HIV/hepatitis C virus coinfection. J Acquir Immune Defic Syndr, 2002. 29(2): p. 145-8. Thibault, V., et al, Interleukin 2 treatment does not modify hepatitis B or C replication in human immunodeficiency virus-infected patients: results from a randomized control trial. Hepatology, 2002. 35(1): p. 238-9. Bruno, R., P. Sacchi, and G. Filice, Mitochondrial toxicity in HIV-HCV coinfection: It depends on the choice of antiretroviral drugs? Hepatology, 2002. 35(2): p. 500-1. Filippini, P., et al., Can HCV affect the efficacy of anti-HIV treatment? Arch Virol, 2000. 145(5): p. 937-44. 190  Principal Applicant: Hogg et al. 40.  41. 42. 43.  44.  45.  46.  47.  48. 49.  50. 51. 52. 53.  54.  55. 56. 57. 58.  Research Proposal  Hogg, R., et al, Improved survival among HIV-infected patients after initiation of tripledrug antiretroviral regimens. CMAJ: Canadian Medical Association Journal, 1999. 160(5): p. 659-65. Lafeuillade, A., G. Hittinger, and S. Chadapaud, Increased mitochondrial toxicity with ribavirin in HIV/HCV coinfection. Lancet, 2001. 357(9252): p. 280-1. Nunez, M., et al., Risk factors for severe hepatic injury after introduction of highly active antiretroviral therapy. J Acquir Immune Defic Syndr, 2001. 27(5): p. 426-31. Sulkowski, M., et al, Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA, 2000. 283(1): p. 74-80. Sulkowski, M.S., et al, Hepatotoxicity associated with nevirapine or efavirenz-containing antiretroviral therapy: role of hepatitis C and B infections. Hepatology, 2002. 35(1): p. 182-9. Yeni, P.G., et al., Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society-USA Panel. Jama, 2002. 288(2): p. 222-35. Kolk, D.P., et al, Significant closure of the human immunodeficiency virus type 1 and hepatitis C virus preseroconversion detection windows with a transcription-mediatedamplification-driven assay. Journal of Clinical Microbiology, 1761. 40(5): p. 1761-6. Hayashi, J., et al, Age-related response to interferon alfa treatment in women vs. men with chronic hepatitis C virus infection: women 39 years or less of age respond better to HCV treatment than men and women older than 40 years. Archives of Internal Medicine, 1998. 158(2): p. 177-81. Bonacini, M., Management issues in patients coinfected with hepatitis C virus and HIV. AIDS Read, 2002. 12(1): p. 19-21, 24-6. Chung, R., et al., Immune recovery is associated with persistent rise in hepatitis C virus RNA, infrequent liver flares, and is not impaired by hepatitis C virus in co-infected subjects. AIDS, 2002. 16(14): p. 1915-23. Soriano, V., et al., Care of patients with chronic hepatitis C and HIV co-infection: recommendations from the HIV-HCV International Panel. Aids, 2002. 16(6): p. 813-28. Sulkowski, M.S., et al, Hepatitis C virus infection as an opportunistic disease in persons infected with human immunodeficiency virus. Clin Infect Dis, 2000. 30 Suppl 1: p. S77-84. Thio, C.L., et al., Screening for hepatitis C virus in human immunodeficiency virus-infected individuals. Journal of Clinical Microbiology, 2000. 38(2): p. 575-7. John, M., J. Flexman, and M.A. French, Hepatitis C virus-associated hepatitis following treatment of HIV-infected patients with HIV protease inhibitors: an immune restoration disease? Aids, 1998. 12(17): p. 2289-93. Braitstein, P., et al. Dangerous Oversights: A comparison of HCV-RNA testing vs. HCVantibody testing among HIV-infected individuals, in Canadian Association for HIV Research. 2001. Toronto, Canada. Mehta, S.H., et al., Protection against persistence of hepatitis C. Lancet, 2002. 359(9316): p. 1478-83. Alter, M., et al., The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. New England Journal of Medicine, 1999. 341: p. 556-62. Beld, M., et al, Evidence that both HIV and HIV-induced immunodeficiency enhance HCV replication among HCVseroconverters. Virology., 1998. 244(2): p. 504-12. Gonzalez de Requena, D., et al., Liver toxicity caused by nevirapine. Aids, 2002. 16(2): p. 290-1. 191  Principal Applicant: Hogg et al. 59.  60.  61.  62. 63.  64. 65. 66.  67. 68.  Research Proposal  Bica, I., et al, Increasing Mortality Due to End-Stage Liver Disease in Patients with Human Immunodeficiency Virus Infection. Clinical Infectious Diseases (CID), 2001. 32(Feb 1): p. 492-497. Reisler, R., et al. Risk of Grade IV Events and Death in HIV Patients Co-Infected with Hepatitis B and/or Hepatitis C Receiving HAART. in 9th Conference on Retroviruses and Opportunistic Infections. 2002. Seattle, WA. Braitstein, P., et al. Hepatitis C is an independent predictor of mortality among a population-based treatment cohort of antiretroviral naive individuals initiating triplecombination therapy, in Canadian Association for HIV Research. 2003. Halifax: Canadian Journal of Infectious Diseases. Neff, G., D. Jayaweera, and A. Tzakis, Liver transplantation for HIV-infected patients with end-stage liver disease. Current Opinion in Liver Transplantation, 2002. 7(2): p. 114-123. Soriano, V., et al, Mortality due to chronic viral liver disease among patients infected with human immunodeficiency virus.[comment]. Clinical Infectious Diseases., 2001. 33(10): p. 1793-5. HIV/AIDS, B.C.f.E.i., Therapeutic Guidelines for the Treatment of HIV/AIDS and Related Conditions, . 1999, 2001: Vancouver. Carpenter, C , et al, Antiretroviral therapy for HIV infection in 1996. JAMA, 1996. 276: p. 146-154. Cote, H.C., et al, Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients, [comment]. New England Journal of Medicine., 2002. 346(11): p. 811-20. Hogg, R., et al, Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA, 1998. 279(6): p. 450-4. Wood, E., et al, Socioeconomic status, access to triple therapy, and survival from HIVdiease since 1996. AIDS, 2002. 16: p. 1 - 8.  192  Appendix 3 Certificates of Ethics A p p r o v a l Appendix 3.1 H C V Testing (original and amendment) Appendix 3.2 Linkage with Providence Laboratory  Appendix 4 Executive S u m m a r y : Roadmap for A d d r e s s i n g the Epidemic of H I V and Hepatitis C Co-infection i n Canada: Issues, R e c o m m e n d a t i o n s , P r i o r i t i e s , a n d N e x t Steps June 2004.  197  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada:  Issues, Recommendations, Priorities and Next Steps June, 2004  Report from the National Stakeholders Meeting on Improving Access to Care, Treatment, and Support for People Living with HIV and Hepatitis C Co -infection, Montreal, Quebec, January 2004.  Dedication  This report is dedicated in loving memory to  Glen Edward Hillson for whom knowledge about and action on these issues did not happen fast enough.  Preamble The Canadian Treatment Action Council (CTAC) is a national, non-profit, consumer-driven organization dedicated to improving the lives of people living with HIV/AIDS by promoting informed public policy and public education, and promoting awareness of issues that impact access to treatment and health care for people living with HIV/AIDS. Over the course of 2003, C T A C , with the financial support of Schering-Plough Canada and Agouron/Pfizer, sponsored a series of regional fora in Vancouver, Montreal, Toronto, and Halifax, regarding treatment and care issues in HIV and Hepatitis C co-infection. In January, 2004, these fora culminated in a multidisciplinary gathering of 50 people living with HIV and/or H C V co-infection, physicians (including general practitioners, hepatologists, and gastroenterologists), epidemiologists, and people working in community organizations, correctional settings, government, and the pharmaceutical industry (see Appendix One). The national meeting was supported by Schering-Plough, Hoffmann-LaRoche, the Anemia Institute, Agouron/Pfizer, Shire BioChem/GlaxoSmithKline, Bristol Myers Squibb, Abbott Laboratories, and Boehringer Ingleheim. The purpose of CTAC's regional fora and national meeting was to identify barriers to the appropriate treatment, care and support of people who are co-infected, and to identify mechanisms and the key players involved in moving past those barriers. The national meeting was intended to produce a report that would serve as a 'roadmap' of where we are in Canada with the epidemic of HIV/hepatitis C co -infection in terms of treatment and care issues, where we need to go, and how we can get there. This document is a summary and synthesis of that national meeting, bolstered b y other relevant information and recommendations where appropriate, and was prepared by Paula Braitstein (Board Member, Canadian Treatment Action Council (CTAC), Senior Policy Advisor on Health Promotion, BC Persons with AIDS Society (BCPWA)). C T A C would like to thank the organizers of the meeting, including Louise Binder, James Kreppner, Philip Lundrigan, Paula Braitstein, Sheena Sargent, Patrick Mclntyre, Claire Checkland, Kim Thomas, Lome Fox, Marie Prevost, Susan Redgrave, Daryle Roberts, Daryn Bond, Francoise Grothe, and Marlene Allan. C T A C would like to gratefully acknowledge the work of the notetakers, Chantale Perron and Terry Pigeon. Last but not least, the organizing committee would like to acknowledge and thank Mardie Serenity for her logistical wizardry, excellent minutes, attention to detail, and generally keeping it all together!  T a b l e o f Contents List of Acronyms Executive Summary Advocacy Priorities Next Steps 1. Summary of Recommendations by Stakeholder Federal Government Provincial Governments Correctional Services Industry Clinicians, Health Care Services, and Health Authorities Scientists Community Organizations Activists and Consumers 2. Epidemiology and Population Health Issues 3. Clinical Issues Liver Transplantation and HIV Clinical Research Needs Health Care Needs 4. Federal and Provincial Policy Issues 5. Correctional Settings 6. Prevention and Education 7. Support Needs 8. Summary of Key Recommendations  Appendix One: Appendix T w o : Appendix Three:  1 2 3 3 5 5 5 6 7 7 8 9 9 10 11 15 15 17 19 21 22 24 25  List of meeting participants Availability of Treatments by Province and Territory for Hepatitis C Letter to B C Pharmacare by coalition of B C activists regarding lack of evidence used in setting criteria in British Columbia for accessing H C V treatment.  List of Acronyms AIDS  Acquired Immune Deficiency Syndrome  ALT  Alanine aminotransferase (a liver enzyme)  APRICOT  AIDS Pegasys Ribavirin International Co-Infection Trial  ASO  AIDS Service Organization  BCPWA  British Columbia Persons with AIDS Society  CAHR  Canadian Association for HIV Research  CanFAR  Canadian Foundation for AIDS Research  CASL  Canadian Association for the Study of the Liver  CIHR  Canadian Institutes of Health Research  CSHA  Canadian Strategy on HIV/AIDS  CTAC  Canadian Treatment Action Council  CTN  Canadian HIV Trials Network  EPO  erythropoietin  FRSQ  Federation de Recherche de la Societe Quebecoise  HAART  Highly Active Antiretroviral Therapy  HCV  Hepatitis C Virus  HIV  Human Immunodeficiency Virus  LCDC  Laboratory Centre for Disease Control  MAT/DOT  Maximally Assisted Therapy/Daily Observed Therapy  MSFHR  Michael Smith Foundation for Health Research  OHTN  Ontario HIV Treatment Network  PAS A N  Prisoners' HIV/AIDS Support Action Network  VANDU  Vancouver Area Network of Drug Users  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  1  Executive S u m m a r y Hepatitis C and HIV co-infection is an important Canadian health issue that is not receiving the attention it demands. Hepatitis C affects approximately 30% of people living with HIV (and approximately 10% of those infected with hepatitis C are also infected with HIV). Co-infected individuals are more likely to be Aboriginal, young, current or former injection drug users, current or former inmates, and people who received contaminated blood or blood products in the course of their healthcare. The majority of people who are co-infected live in Montreal or Vancouver, with emerging epidemics in Ottawa, Toronto, Calgary, and Edmonton. In the presence of HIV, hepatitis C disease progression takes place 2 to 3 times faster, compared to people who only have hepatitis C. Thus, in seven to fifteen years after becoming infected, approximately 50-70% of co-infected people will begin to develop liver inflammation, and at least 20-30% will progress to liver fibrosis and cirrhosis, including end-stage liver disease. Because of the rapidity of disease progression in co-infected individuals, and their often unique constellation of social and health needs (e.g. in dealing with addiction, mental health, and HIV treatment issues), co-infected people represent a distinct population, falling between the cracks of both HIV and Hepatitis C treatment systems. HIV care providers and community services are ill-equipped, under-funded, and often lack sufficient or appropriate information on hepatitis C. Hepatitis C care providers are sometimes uninterested in HIV issues or HIV -positive people and often also require education and awareness training on HIV. Hepatitis C community services are almost non-existent, in part because of the lack of leadership by the Federal government (approximately 1% of Canadians are believed to be infected with hepatitis C, yet Canada has no national Hepatitis C Strategy or dedicated funding). The average onset of HIV symptoms in the absence of treatment is seven to ten years after infection. The onset of fibrosis or scarring of the liver due to hepati tis C in the presence of HIV is seven to fifteen years. Many individuals who are currently co-infected acquired their two viruses together, and many of them became dually infected in the 1990's. Because of this, there is urgency to address the unique issues arising from the convergence of these epidemics. The purpose of C T A C ' s regional fora and national meeting was therefore to identify barriers to the appropriate treatment, care and support of people who are co-infected, and to identify mechanisms and the key players involved in moving past these barriers. The meeting was intended to produce a report that would serve as a 'roadmap' of where we are in Canada with the epidemic of HIV/hepatitis C co-infection in terms of treatment and care issues, where we need to go, and how we can get there. Areas identified as needing particular attention were: •  Clinical issues including access to treatment, management of side effects, balancing HIV . management, psychiatric and mental health supports, transplantation, and health care delivery  •  Defining research priorities  •  Policy issues ranging from federal and provincial strategies and funding, to formulary drug coverage and transplantation  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  2  •  Prevention education, including primary and secondary prevention  •  Community support services  •  Correctional settings  Advocacy Priorities Short-term: •  Identify best practices and standards of care for HIV/HCV co-infection.  •  Advocate with provincial and private payers to broaden criteria for accessing treatment, and to cover concomitant Growth Factors, if necessary.  •  Update treatment and management guidelines for co-infection, including H C V treatment issues, HIV treatment issues, side effect management, transplantation, nutritional issues, and psychiatric issues (expand upon what is contained in the C A S L 2004 viral hepatitis guidelines).  •  Disseminate treatment and management guidelines widely.  •  Work with C T N to expand their H C V co-infection 'Core' as a basis for a network of investigators in co-infection.  •  Have provincial transplant centers and the Canadian Society of Transplantation develop appropriate guidelines for the assessment and transplantation of HIV co-infected individuals.  •  Identify pharmacoeconomists who can conduct research into the cost-effectiveness of properly addressing HIV/HCV co-infection, and the cost-effectiveness of treatment, and those who are willing to work with community activists to decipher existing materials.  •  Use advocacy issues, such as lack of access to treatment, to raise public awareness and apply pressure on government through media campaigns.  Long-term: •  Develop a national observational cohort of co-infected individuals, on and off treatment.  •  Expand CTAC'.s Post-Approval Surveillance System project to incorporate H C V treatments.  Next Steps 1.  C T A C to disseminate meeting report to all meeting participants, and other relevant stakeholders who did not attend.  2. C T A C to organize a national meeting of key clinicians, scientists, and consumers, to develop a research agenda for distribution to all research funding bodies (including the pharmaceutical industry, CIHR, O H T N , and CTN). 3. National state-of-the-art co-infection treatment and management guidelines should be developed and published.  Roadmap tor Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  3  4. CTAC's Co-Infection sub-committee to identify individuals interested in working together to develop workplan priorities to continue this work, including following-up on priorities highlighted in this report. 5.  Advocate with the federal government for the immediate establishment of an on -going National Hepatitis C Strategy, appropriately resourced (as per the Canadian AIDS Society document entitled "A National Hepatitis C Strategy in Canada: A Discussion Paper", www.cdnaids.caV  6. PAS A N to circulate updated information regarding co-infection in correctional settings. 7.  HepCure B C to conduct a survey on the availability of H C V treatment across the country, and disseminate findings (see Appendix Two).  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  4  1. S u m m a r y o f R e c o m m e n d a t i o n s b y Stakeholder Federal Government •  Fund and implement an ongoing national hepatitis C stra tegy that significantly incorporates HIV co-infection.  •  Incorporate priorities regarding co-infected individuals into the CSHA, with extra dollars attached.  •  Devote more money to researching clinical aspects of co-infection, including natural history and pathogenesis issues.  •  Create a network of databases for sharing data regarding treatment outcomes among co infected individuals, and a network of physicians and researchers focussing on co-infection.  •  Establish Centers for Excellence in Hepatitis C with expertise in HIV co-infection.  •  Provide sufficient financial resources to allow integrated and specialized clinics to operate.  •  Immediately implement and fund more harm reduction and addiction treatment services.  •  Fund AIDS Service Organizations and other community-based organizations to provide resources and materials to co-infected individuals.  •  Pressure the Canadian Society for Transplantation to develop appropriate guidelines for transplantation in HIV-infected individuals.  •  The Ministerial Council on HIV/AIDS should advise the federal HIV/AIDS Division to incorporate H C V co-infection into the revised CSHA, with the recommendation of attaching not currently allocated dollars.  •  Include significant participation from both HIV and hepatitis C groups in the membersh ip of all federal government committees addressing either HIV or hepatitis C issues.  •  Classify hepatitis C as an AIDS defining illness, and classify addiction as a disability (where it is not already).  •  Support, fund, and implement general and targeted education campaigns aimed at increasing the number of people getting tested for both viruses, and at decreasing the stigma associated with having them. Health Canada should lead and fund these initiatives, in collaboration with grassroots organizations and the provincial government.  Provincial Governments •  Expand eligibility criteria for accessing and remaining on hepatitis C treatment for as long as patient and physician believe it necessary and appropriate.  •  Recognize and accept that the cost of growth factors is part of the cost of H C V treatment; then negotiate with the companies that make hepatitis C treatments, and those that make the growth factors, to enable the combined usage at a reduced cost.  •  Develop provincial Hepatitis C Strategies, with devoted money, paying particular attention to addressing clinical management and community support issues.  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  5  •  Identify what specialized services, in terms of H C V treatment and HIV co-infection, are available in Canada, and where they are available.  •  Regularly revise treatment and management guidelines based on current evidence, and disseminate widely to physicians and patients.  •  Ensure that each province has appropriate guidelines for liver transplantation in HIV-infected individuals.  •  Develop Centers for Excellence in Hepatitis C with expertise in HIV co-infection.  •  Provide sufficient financial resources to allow integrated and specialized clinics for coinfection to operate.  •  Immediately fund and implement more harm reduction and addiction treatment services.  •  Support, fund, and implement general and targeted education campaigns aimed at increasing the number of people getting tested for both viruses, and at decreasing the stigma associated with having them. Health Canada should lead and fund these initiatives, in collaboration with grassroots organizations and the provincial government.  •  Include significant participation from both HIV and hepatitis C groups in the membership of all provincial government committees addressing either HIV or hepatitis C issues.  •  Classify addiction as a disability (where it is not already).  •  Fund AIDS Service Organizations and community-based organizations to provide resources and materials to co-infected individuals.  Correctional Services •  Immediately implement recommendations from existing reports regarding safe drug use and tattooing, methadone treatment, addiction treatment, harm reduction, and unhindered access to knowledgeable care providers and specialists.  •  Enhance collaboration between existing clinics and hepatitis, HIV, and infectious disease specialists.  •  Integrate other health care modalities into all clinics, and move toward a holistic and patient centered model of care.  •  Provide opportunities to see patients in health care settings daily or weekly to assist them with receiving and tolerating their treatments (e.g. daily observed therapy, maximally assisted therapy), and provide adequate nutritional and mental health supports.  •  Develop new, or adapt existing, drop-in day clinics to help patients receive and tolerate their treatments.  •  Immediately implement more harm reduction and addiction treatment services.  •  Launch general and targeted education campaigns aimed at increasing the number of people getting tested for both viruses, and at decreasing the stigma associated with being infected.  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  6  Industry •  Make products, including both H C V treatments such as pegylated interferon and also supportive therapies such as erythropoeitin (EPO), more accessible through price reduction and other means.  •  Work with public payers to expand eligibility criteria for accessing and remaining on hepatitis C treatment for as long as patient and physician believe it necessary and appropriate.  •  Conduct pharmacoeconomic studies to show the cost-effectiveness of supportive treatments such as EPO, and early treatment of H C V infection.  •  Participate in the development of a HIV/HCV co-infection research agenda, and integrate these priorities into their drug development plans.  •  Devote more money to researching clinical aspects of co-infection, including natural history, pathogenesis issues, H C V treatment, and HIV treatment.  •  Always provide an expanded access program for new drugs for hepatitis C, with designated spaces reserved for HIV co-infected individuals.  •  Include, and separately analyze, H C V co-infected individuals in research for HIV and HIVrelated products.  •  Support community initiatives in the areas of harm reduction, addiction treatment, and poverty reduction.  •  Support the development of drop-in centers for co-infected persons, including explicit supports for co-infected people on treatment.  Clinicians, Health Care Services, and Health Authorities •  Use clinical authority to advocate that third party payers expand their eligibility criteria for accessing and remaining on hepatitis C treatment for as long as patient and phys ician believe it necessary and appropriate.  •  Encourage public payers and pharmaceutical manufacturers to negotiate the costs of H C V treatment, including growth factors.  •  Identify best practices and standards of care elsewhere in the world for the treatment and management of HIV/HCV co-infection.  •  Identify what specialized services, in terms of H C V treatment and HIV co-infection, are available in Canada, and where they are available.  •  Regularly revise treatment and management guidelines based on current evidenc e, in consultation with consumers, and disseminate widely to other physicians and patients.  •  Refer HIV-positive patients to transplant centers for assessment, even if the transplant center does not have an HIV infection policy, and not wait until the patient has decompensated cirrhosis to refer them.  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  7  •  Pressure organ transplant centers to develop appropriate policies and guidelines for assessing and performing liver transplants on people living with HIV. Clinicians and surgeons working in transplant centers should be proactive in developing these policies and guidelines.  •  Encourage and support existing HIV clinics in working more collaboratively with hepatitis experts.  •  Hepatitis experts should become more proactive in learning about HIV and collaborating with HIV experts.  •  Move all clinics toward integrating other health care modalities, and toward a holistic and patient-centered model of care.  •  Develop Centers for Excellence in Hepatitis C with expertise in HIV co-infection.  •  Provide opportunities to see patients in health care settings daily or weekly to assist them with receiving and tolerating their treatments (e.g. daily observed therapy, maximally assisted therapy).  •  Develop or adapt drop-in day clinics to help patients receive and tolerate their treatments.  •  Implement more harm reduction and addiction treatment services.  •  Develop Continuing Medical Education programs specifically to train physicians on coinfection, and develop and offer more training to front-line workers.  •  Develop mentorship and training programs in co-infection for physicians and researchers.  •  Support efforts to classify hepatitis C as an AIDS defining illness, and addiction as a disability.  •  Develop more drop-in centers for persons co-infected with HIV and H C V , including explicit supports for people on treatment.  Scientists •  Conduct pharmaeconomic studies to show the cost-effectiveness of early H C V treatment and the use of supportive therapies such as EPO.  •  Regularly revise treatment and management guidelines based on current evidence, in collaboration with consumers, and disseminate widely to physicians and patients.  •  Develop an HIV/HCV research agenda in order to identify research priorities across the Four Pillars Drug Strategy (harm reduction, prevention, treatment, enforcement).  •  The Canadian Association for HIV Research (CAHR) should write a letter to CIHR, O H T N , C T N , and other research institutions such as CanFAR, to advocate for more money to be devoted to researching various aspects of co-infection, including natural history and pathogenesis issues.  •  Develop Centers for Excellence in Hepatitis C with expertise in HIV co-infection.  •  The CIHR Advisory Committee on HIV/AIDS and the Federal Ministerial Council on HIV/AIDS should recommend the same thing to these research organizations.  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  8  •  The Canadian HIV Trials Network should create a network of databases for sharing (anonymized) data regarding treatment outcomes among co-infected, and a network of physicians and researchers focussing on co-infection.  •  Research funding bodies such as CIHR, CanFAR, the C T N , and the O H T N should solicit research proposals specific to the issue of HCV/HIV co-infection.  Community Organizations •  Develop and implement more harm reduction services.  •  Develop targeted education campaigns aimed at increasing the number of people getting tested for both viruses, and at decreasing the stigma associated with having them.  •  Lobby for funding dedicated to HCV/HIV co-infection to be built into the Canadian Strategy on HIV/AIDS, with devoted materials and resources developed as a result.  •  Make hepatitis C co-infection a priority in organizational workplans.  •  Encourage participation of both HIV-positive and hepatitis C-positive individuals on organizational committees.  •  Provide resources and materials to HCV/HIV co-infected individuals about co-infection.  •  Develop peer-driven networks and groups to foster mutual support and collective action.  •  Develop more drop-in centers for persons co-infected, including explicit supports for people on treatment.  Activists and Consumers •  Have courage, be tenacious, and know your stuff.  •  Familiarize yourself with the recommendations for all stakeholders, select the issue(s) that is/are of highest personal importance and/or interest, and actively work towards the achievement of the recommended actions, either as an individual or through an affiliate organization.  Roadmap for Addressing the Epidemic of HIV and Hepatitis C Co-Infection in Canada  9  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099819/manifest

Comment

Related Items