UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Virologic studies in pediatric HIV-1 infection Binns, Christine Ruth 2001

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

Item Metadata

Download

Media
831-ubc_2001-0324.pdf [ 5.62MB ]
Metadata
JSON: 831-1.0090070.json
JSON-LD: 831-1.0090070-ld.json
RDF/XML (Pretty): 831-1.0090070-rdf.xml
RDF/JSON: 831-1.0090070-rdf.json
Turtle: 831-1.0090070-turtle.txt
N-Triples: 831-1.0090070-rdf-ntriples.txt
Original Record: 831-1.0090070-source.json
Full Text
831-1.0090070-fulltext.txt
Citation
831-1.0090070.ris

Full Text

V I R O L O G I C S T U D I E S I N P E D I A T R I C H I V - 1 I N F E C T I O N By Christine Ruth Binns B. Sc. (Hon.), The University of Ottawa, 1993 A THESIS S U B M I T T E D IN PARTIAL F U L F I L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF i GRADUATE STUDIES Experimental Medicine Program We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH C O L U M B I A July 2001 © Christine Ruth Binns, 2001 U B C Special Collections - Thesis Authorisation Form Page 1 of 1 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g of t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the head of my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department of L ^ O C ^ U X A ; \ X M ! TSTX/JS) \lzA A f X A Q Q The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada http://www.library.ubc.ca/spcoll/thesauth.html 9/7/2001 Abstract The proportion o f HIV-1-infected children is increasing in parallel with that o f women o f child-bearing age. There are currently over 200 HIV-1-infected Canadian children. This study was undertaken with the main objective being to evaluate several cl inical and laboratory measures in this population to gain a better understanding o f their disease status and their risk for progressing to A I D S . These measures included viral load testing, drug resistance testing, C D 4 + T cell levels, and viral phenotype testing. It was hypothesized that a group o f cl inical and/or laboratory tests exist which would be optimal markers by which to assess a pediatric HIV-1 patient's current status as well as the risk of disease progression over time. Secondary objectives included to determine the prevalence o f resistance to zidovudine ( Z D V ) in HIV-1-infected Canadian children and to determine the association o f drug resistance with the other cl inical and laboratory variables evaluated. HIV-1-infected pediatric patients were recruited from seven healthcare centers across Canada. Plasma and peripheral blood mononuclear cells ( P B M C s ) were isolated from whole blood samples donated by these patients. Plasma viral load determinations were performed. P B M C s were cocultured to generate viral stocks from which viral phenotype and standardized Z D V resistance testing assays were performed. P B M C s were also used to set up quantitative micrococultures which were used as the basis for our rapid phenotypic resistance assay and for determining cell-associated viral loads. Data was available for a total o f 86 patients. 23/78 (30%) H I V - 1 positive viral isolates were o f the SI phenotype. 14/60 (23%) isolates showed resistance to Z D V according to the rapid phenotypic resistance assay. Median cell-associated viral load was 125 IU/10 6 cells and median plasma viral load was 25 000 copies/ml. 26/53 (49%) isolates showed resistance to Z D V according to the genotypic resistance assay. A positive correlation was observed between cell-associated and plasma viral load (r=0.37, p=0.004). Plasma viral load and the presence o f encephalopathy were also highly correlated (p=0.023), as were plasma viral load and perinatal transmission (p=0.004). In contrast, cell-associated viral load tended to be higher in children without encephalopathy and who had acquired H I V - 1 non-perinatally (highly significant (p=0.001)). 35/45 (78%) o f the samples on which comparative analysis o f genotypic and phenotypic ii resistance was performed produced concordant results. Phenotypic resistance, genotypic resistance, and SI phenotype were found to be correlated with lower CD4 cell counts and percents. According to the results obtained in this study, the prevalence of Z D V resistance among HIV-1-infected Canadian children is high. It is surprising that more significant associations between different parameters were not found in establishing this "viral inventory". Associations may become more clear as prospective data are obtained. 111 Table of Contents Abstract i i Table of Contents .' iv Lis t o f Tables vi List o f Figures v i i i Acknowledgements ix I N T R O D U C T I O N ; 1 Epidemiology o f Pediatric HIV-1 Infection 1 Epidemiology of Pediatric HIV-1 Infection in Canada 2 Pathogenesis o f Pediatric H I V - 1 Infection 3 Cl in ica l Course o f Pediatric H I V - 1 Infection 6 Bacterial Infections 9 Vi ra l Infections : 10 Fungal Infections 12 Pneumocystis carinii Pneumonia (PCP) 13 Other Parasitic Infections 14 Lymphoid Interstitial Pneumonitis 15 Neoplasms 16 Central Nervous System and Developmental Complications 16 Otorhinolaryngologic Complications 17 Cardiovascular Complications • 18 Gastrointestinal and Nutritional Complications 19 Nephrologic Complications 19 Hematologic Complications 20 Cl inical Interventions 21 Zidovudine 21 Didanosine 23 Zalcitabine 23 Combination Therapy 24 Early vs. Late Fetal Transmission 25 Factors Associated Wi th Risk o f Transmission 28 Diagnosis o f HIV-1 Infection in Children 32 Antibodies to HIV-1 33 In Vi t ro Ant ibody Production Assays 34 p24 Antigen Assay 35 HIV-1 Culture 35 Polymerase Chain Reaction 36 Surrogate Tests-Immunologic Parameters 37 Viro logic Hypothesis o f HIV-1 39 Antiretroviral Drug Resistance 43 Zidovudine Resistance 43 Resistance Testing Methodologies 45 Research Hypothesis 46 Specific Objectives 46 M A T E R I A L S A N D M E T H O D S 47 Patient Population : 47 Specimen Collect ion 47 Isolation o f Lymphocytes 47 Qualitative Macrococulture Assay 48 Quantitative Micrococulture Assay 49 Rapid Quantitative Culture-Based Zidovudine Resistance Assay 50 Standard HIV-1 Drug Susceptibility Assay 51 HIV-1 Vi ra l Phenotype Assay (HIV-1 Syncytium-Inducing Assay) 55 iv Plasma V i r a l Load Assay ( N A S B A ® ) 56 A . Nucleic A c i d Isolation 57 B . Amplif icat ion 59 C . Hybridization/Detection 61 Vi ra l Isolate Sequencing 65 Cl in ica l Data 65 Statistical Methods 65 R E S U L T S 66 Patient Population 66 Qualitative Macrococultures 66 Cell-Associated V i r a l Load (Quantitative Micrococultures) 66 Vi ra l Phenotype 67 Plasma V i r a l Load 67 Standard Phenotypic Resistance Testing ( A C T G ) 68 Quantitative Culture-Based Z D V Resistance Assay 68 Vi ra l Isolate Sequencing (Genotypic Resistance) 69 C D 4 C e l l Counts 69 Mode o f Transmission 70 Presence of Symptoms 70 Encephalopathy 70 C O R R E L A T I V E A N A L Y S E S , 71 Plasma Vi ra l Load and Cell-Associated V i r a l Load 71 Vi ra l Load and V i r a l Phenotype 71 Plasma V i r a l Load and C D 4 C e l l Count 72 Plasma V i r a l Load and Presence o f Symptoms 72 Plasma V i r a l Load and Encephalopathy 72 Plasma V i r a l Load and Mode o f Transmission 72 Cell-Associated V i r a l Load and C D 4 C e l l Count 73 Cell-Associated V i r a l Load and Presence o f Symptoms 73 Cell-Associated Vi ra l Load and Presence of Encephalopathy 73 Cell-Associated V i r a l Load and Mode of Transmission 74 Comparative Resistance: Phenotype vs. Genotype 74 Comparative Resistance: Genotype vs. Rapid Phenotype vs. Standard Phenotype ( A C T G ) 75 Genotypic Resistance and Other Variables 76 Phenotypic Resistance and Other Variables 77 Phenotypic Resistance and Plasma Vi ra l Load 78 Phenotypic Resistance and Cell-Associated Vi ra l Load 79 V i r a l Phenotype and C D 4 C e l l Count 79 Vi ra l Phenotype and C D 4 Percent 80 D I S C U S S I O N 81 Conduct o f Study 81 Cell-Associated V i r a l L o a d 81 Plasma Vi ra l Load 81 Phenotypic Resistance 82 Genotypic Resistance 83 Prevalence o f Zidovudine Resistance 83 Vi ra l Phenotype 84 C O N C L U S I O N 85 R E F E R E N C E S 86 v Lis t of Tables Table 1 Cumulative Number o f Canadian Perinatally HIV-1-Exposed Infants Page 3 Table 2 Common V i r a l Pathogens in HIV-1-Infected Children Page 11 Table 3 Proposed Laboratory-Based Definition of Early vs. Late HIV-1 Infection Page 25 Table 4 Sensitivity o f Early Diagnostic Tests for HIV-1 Infection in Infants Page 37 Table 5 Study Population Page 66 Table 6 Qualitative Macrococultures Page 66 Table 7 Descriptive Characteristics o f Cell-Associated V i r a l Load ( IU/10 6 cells) Page 67 Table 8 Vi ra l Phenotype (SI/NSI) Determinations Page 67 Table 9 Descriptive Characteristics o f Plasma V i r a l Load (copies/ml) Page 68 Table 10 Characteristics o f Z D V Resistance Using Rapid Phenotypic Resistance Testing Page 68 Table 11 Characteristics o f Z D V Resistance Using Genotypic Analysis Page 69 Table 12 Descriptive Characteristics o f C D 4 C e l l Counts (cells/mm 3) and C D 4 Percent Page 69 Table 13 Relationship Between Plasma Vi ra l Load (copies/ml) and V i r a l Phenotype (SI/NSI) Page 71 Table 14 Relationship Between Cell-Associated Vi ra l Load ( IU/10 6 cells) and V i r a l Phenotype (SI/NSI) Page 71 Table 15 Relationship Between Plasma V i r a l Load (copies/ml) and Presence of Symptoms Page 72 Table 16 Relationship Between Plasma V i r a l Load (copies/ml) and Presence o f Encephalopathy Page 72 Table 17 Relationship Between Plasma Vi ra l Load (copies/ml) and Mode o f Transmission Page 73 Table 18 Relationship Between Cell-Associated V i r a l Load ( IU/10 6 cells) and Presence o f Symptoms Page 73 Table 19 Relationship Between Cell-Associated V i r a l Load ( IU/10 6 cells) and Presence o f Encephalopathy Page 74 Table 20 Relationship Between Cell-Associated Vi ra l Load ( IU/10 6 cells) and Mode o f Transmission Page 74 Table 21 Relationship Between Genotypic and Phenotypic Resistance Page 75 Table 22 Comparison o f Genotypic, Rapid and Standard Phenotypic Resistance Page 76 Table 23 Relationship Between C D 4 C e l l Count (cells/mm 3) and Genotypic Resistance Page 76 Table 24 Relationship Between C D 4 Percent and Genotypic Resistance Page 7 6 Table 25 Relationship Between Genotypic Resistance and Presence o f Symptoms Page 77 Table 26 Relationship Between Genotypic Resistance and Presence o f Encephalopathy Page 77 Table 27 Relationship Between Genotypic Resistance and Mode of Transmission Page 77 Table 28 Relationship Between Phenotypic Resistance and C D 4 C e l l Count (cells/mm 3 ) Page 77 Table 29 Relationship Between Phenotypic Resistance and C D 4 Percent Page 78 Table 30 Relationship Between Phenotypic Resistance and Presence o f Symptoms Page 78 Table 31 Relationship Between Phenotypic Resistance and Presence of Encephalopathy Page 78 Table 32 Relationship Between Phenotypic Resistance and Mode o f Transmission Page 78 Table 33 Relationship Between Plasma Vi ra l Load (copies/ml) and Phenotypic Resistance Page 79 Table 34 Relationship Between Cell-Associated V i r a l Load (IU/10 f > cells) and Phenotypic Resistance Page 79 Table 35 Relationship Between V i r a l Phenotype (SI/NSI) and C D 4 C e l l Count (cells/mm 3 ) Page 80 Table 36 Relationship Between V i r a l Phenotype (SI/SNI) and C D 4 Percent Page 80 v i i Lis t of Figures Figure 1 Reported Number o f Infants Exposed to H I V in utero and the Number with Confirmed H I V Infection Figure 2 HIV-1 Figure 3 The Pattern o f HIV-1 Infection in vivo Figure 4 24-Wel l Plate Format for Rapid Quantitative Z D V Resistance Assay Figure 5 96-Wel l Plate Format for H I V - 1 V i r a l Stock Titration Figure 6 96-Wel l Plate Format for Z D V Susceptibility Testing Figure 7 96-Wel l Plate Format for the M T - 2 HIV-1 Vi ra l Phenotype Assay Figure 8 Schematic presentation-Nucleic A c i d Isolation Process Figure 9 Amplif icat ion Figure 10 Schematic Representation o f the Hybridization Format Figure 11 Electrochemiluminescence Process vui Acknowledgements There are several people whom 1 would like to thank, as this has been a tremendously long journey. Firstly, I would l ike to thank my supervisor, Dr. Brian Conway, for all o f your support and encouragement. Thank you for giving me the opportunity to further my academic career. Y o u r confidence in me never faltered, even when mine did. Thank you for sticking with me for such a long time and for not giving up on me. Y o u never missed an opportunity to acknowledge accomplishments, no matter how small, and always supported new endeavors. Y o u and Ho l ly made me feel like family. Secondly, I would like to thank all the members of the Conway lab, both past and present. I learned a great deal from you al l . We shared some real ups and downs, and some o f us a move across the country. Thank you for all the times you helped me out. I always enjoyed your companionship and w i l l remember our time together fondly. Thirdly, I would like to thank all o f my family and friends who helped me see this thesis from conception to completion. There are so many. M u m and Dad, Lesley and Roger, you never doubted that I could do it. Thank you for the encouragement, the "constant reassurance", and gentle pushes when 1 needed them. Y o u were never too busy to talk. Thank you especially Sarah for always having a sympathetic ear and an open door. Finally, 1 would like to thank my husband Gavin , who has endured it a l l . Y o u r patience, above al l , has been truly phenomenal. Y o u never let me quit and never once complained. Without your unwavering love, understanding and support I would not have finished. Thank you for lending your computer expertise and typing skills. Y o u mean more to me than you wi l l ever know. ix I N T R O D U C T I O N Epidemiology of Pediatric HIV-1 Infection The spread of HIV-1 throughout the world could be thought o f as the sum o f several epidemics in different populations. These populations include men having sex with men, intravenous drug users, and heterosexual men and women to name the major groups, most o f which include women and children. The dynamics of these epidemics vary from continent to continent, country to country and even within different regions of a given country. However, none are discrete and they are all in a dynamic equilibrium with one another. Neither are they discriminatory, as HIV-1 can infect anyone. The first cases o f A I D S in children were reported in 1982 (1-3), one year after the first case was described in adults. In the nineteen years since then, HIV-1 infection has had a significant impact on the health o f children worldwide. In 1983, H I V - 1 was determined to be the cause o f the syndrome. Since then, the growth o f the A I D S epidemic in children has paralleled that o f infection in women, with more than 70% o f female cases o f A I D S diagnosed during the childbearing years (13-39 years) (10). As o f November 20, 1996, the number o f cumulative A I D S cases in adults and children reported to the W o r l d Health Organization since the onset o f the pandemic was 1,544,067(4) and by December 1998, the total number o f A I D S deaths since the beginning o f the pandemic was 13.9 mi l l ion (330). B y the end o f 1998 the W o r l d Health Organization had estimated that 13.8 mi l l ion women and 1.2 mi l l ion children were infected with HIV-1 worldwide (330). In total, it is estimated that the number o f persons l iv ing with H I V -1/AIDS (including all asymptomatic individuals) at the end of 2000 was 36.1 mi l l ion . Today, almost all children (85% in the United States and the vast majority worldwide) who acquire HIV-1 do so from their infected mothers, and this is termed vertical transmission (8). B y the end o f 2000, the W o r l d Health Organization had determined that 13.3 mi l l ion women and children had died from HIV-1-related disease, and that 16.4 mil l ion women and 1.4 mil l ion children were l iv ing with H I V / A I D S (73). Originally, HIV-1 disease in North Amer ica was believed to be confined to men having sex with men. The spread of HIV-1 by heterosexual contact and intravenous drug use has increased the prevalence of 1 infection among sexually active women o f childbearing age. Epidemiological surveys in Europe and North America show that o f women delivering infants between 1989-1996, 0.15-0.24% were infected with H I V -1. In sub-Saharan Afr ica , the proportion o f pregnant women receiving antenatal care who are infected is as high as 20-30% (74). Many epidemiological studies have attempted to evaluate the rate o f vertical transmission o f HIV-1 infection. In Afr ica , reported transmission rates have ranged from 28-52% of infants born to infected mothers (62). In Europe and North America, transmission rates have been lower, ranging from 10 to 39% (63,67,109,1 10). Epidemiology of Pediatric HIV-1 Infection in Canada The first A I D S case diagnosed in Canada was in 1982 (331). B y December 4, 1986, the total number o f A I D S cases reported to the Laboratory Centre for Disease Control ( L C D C ) at Health Canada was 830 and 7690 by the end o f 1991. U p to the end o f December 1999, a cumulative number o f 16 913 A I D S cases in Canada have been reported to L C D C ; 196 o f those cases were among children less than 15 years old. The total number of positive H I V tests among Canadian children under the age of 15 between November 1985 and December 1999 was 664. A Health Canada surveillance report published in A p r i l o f 1999 reported that between 1984 and 1998, there had been 218 confirmed cases o f pediatric HIV-1 infection in Canada, and 924 perinatally HIV- l - exposed infants (5). Figure 1 depicts the reported number o f infants born to HIV-1 positive mothers and the number of infants with confirmed HIV-1 infection (1988-1997) (332). Figure 1: Reported Number of Infants Exposed to HIV-1 in Uteroand the Number with Confirmed HIV-1 Infection 100 80 60 Q3 40 20 n MM 1 ill k L 1 1 1 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Year n Number of infants HIV-l-exposed (n=676) • Number of infants confirmed HIV-1-positive (n=216) Source: Bureau of H I V / A I D S , S T D , a n d T B , L C D C , Health Canada, M a y 1999. 2 Table 1 shows the cumulative number o f Canadian perinatally HIV-1-exposed infants, by geographic region and current status, 1984-1998 (5). Table 1: Cumulative Number o f Canadian Perinatal HIV-1-Exposed Infants Region Exposed Indeterminate' Confirmed Infected Asymptomatic Confirmed Infected Symptomatic2 Sero-reverted Died of AIDS Died/ Other Unknown Total n % British Columbia 3 1 32 102 6 2 0 146 15.8 Alberta Prairie/ Territories3 7 12 0 0 10 13 18 48 1 4 1 0 0 0 37 114 4.0 8.3 Ontario Quebec Atlantic Provinces4 23 17 3 18 26 0 51 63 4 138 198 9 33 59 4 5 9 0 3 1 0 271 373 20 29.3 40.4 2.2 Total % 65 7.0 45 ' 4.9 173 18.7 513 55.5 107 11.6 17 1.8 4 0.4 924 100.0 100.0 includes mild, moderate, and severe combined. 3Includes Saskatchewan, Manitoba, and Northwest Territories. 4lncludes Newfoundland, and Nova Scotia. Source: Health Canada. HIV and AIDS in Canada: Surveillance report to December 31. 1998. Division of HIV/AIDS surveillance, Bureau of HIV/AIDS, STD, and TB, LCDC, HPB, Health Canada, April 1999. The majority of perinatally infected children in Canada acquire HIV-1 from mothers who are either injection drug users or sexual partners o f injection drug users (9). A startling trend in Canada is that the proportion o f A I D S cases in women is increasing. U p to 1990, 6.2% o f A I D S cases in Canada occurred in women. Between 1990 and 1995, this rose to 6.9%, and in 1996, it was 9.5%. A s this proportion increases, so may the proportion o f infants who become perinatally infected. Whi le the majority o f A I D S cases in women are still related to heterosexual transmission, the contribution o f injection drug users to the total is rising (6.5% before 1990, 19.5% between 1990 and 1995, and 2 5 % in 1996) (6). Pathogenesis of Pediatric HIV-1 Infection Human immunodeficiency virus type 1 (HIV-1) , the virus which causes the majority of cases of acquired immune deficiency syndrome (AIDS) , is a nononcogenic, cytopathic retrovirus. It belongs to the Ientivirus family. Being an R N A virus, H I V - 1 possesses the ability to synthesize a D N A provirus from a template o f genomic viral R N A . It accomplishes this using a virally encoded D N A polymerase called reverse transcriptase. 3 Figure 2: HIV-1 Figure 2 shows a schematic diagram o f H I V - 1 . A mature HIV-1 vi r ion is slightly more than 100 tim in diameter. When viewed by electron microscopy, it appears as a dense cylindrical core surrounded by a l ipid envelope. The vir ion core contains structural proteins, the R N A genome and viral ly encoded enzymes. These enzymes include reverse transcriptase and integrase, which are needed in the early stages of viral replication. The R N A genome o f H I V - 1 is - 1 0 kilobase (kb) pairs long and is comprised o f the gag, pol, and env genes characteristic o f most retroviruses, as wel l as two flanking long terminal repeat ( L T R ) sequences. It also contains at least six additional genes (tat, rev, nef, vpu, vpr, vij), which are needed for the coordination o f viral gene expression and replication. Reverse transcriptase, protease and integrase are encoded by the pol gene, while the structural proteins of the vir ion core are encoded by the gag gene. The external and transmembrane envelope glycoproteins are encoded by the env gene. Early in the A I D S epidemic, a dramatic decline in the absolute number o f circulating C D 4 + T cells and in the ratio o f C D 4 + T to C D 8 + T cells was observed in adult A I D S patients (25, 26). The isolation o f HIV-1 from the C D 4 + T lymphocytes in the circulation of patients with A I D S (27-29), prompted investigators to propose the C D 4 + T cell as the principal target o f HIV-1 infection in vivo. H I V - 1 was subsequently shown to infect and replicate in C D 4 + T cells in culture and to cause a rapid and profound cytopathic effect in 4 these cells (30). The C D 4 molecule on the surface of the C D 4 + T cells is the high-affinity receptor by which HIV-1 preferentially infects the cells (31,32). The life cycle of HIV-1 begins when it comes into contact with and binds to a C D 4 molecule on a C D 4 + T cell. This binding is mediated by the g p l 2 0 surface glycoprotein o f H I V - 1 , a cleavage product o f the gp 160 precursor protein, the transmembrane glycoprotein (gp41) being the other cleavage product. As either a trimer or a tetramer o f gpl20/gp41, these proteins are incorporated into the outer l ip id bilayer o f the vir ion. Several specific sites on the g p l 2 0 molecule have been determined to be associated with C D 4 binding. The affinity of gp 120 for C D 4 is greater than that o f the M H C class II molecule (31,32). The binding of the g p l 2 0 and the C D 4 molecule triggers a series o f events which results in the fusion o f the virus and cell membranes at least partially mediated by the binding of a specific domain o f g p l 2 0 to a chemokine receptor. Cocch i et al. demonstrated that the beta-chemokines, R A N T E S , MIP-1 alpha, and M l P - l b e t a suppressed the replication o f primary clinical isolates o f HIV-1 in vitro (339), which led to the identification o f chemokine receptors as HIV-1 coreceptors. Cellular chemokine receptors are seven-transmembrane, G-protein-coupled receptors, which transduce chemokine binding into intracellular signals. A l l HIV-1 stt "ains use either the C C R 5 or C X C R 4 coreceptors; some use both. Sequences within the third variable region (V3) o f the HIV-1 envelope gp l20 seem to influence coreceptor usage. The virus nucleocapsid core is then introduced into the host cell cytoplasm, and reverse transcription occurs, the genomic viral R N A being converted to proviral D N A . This is accomplished through the coordinated activities o f the H I V - 1 reverse transcriptase and a second viral ly encoded enzyme, ribonuclease H , which serves to degrade the viral R N A template. A second strand of D N A is then synthesized, producing a double-stranded D N A copy o f the original HIV-1 R N A genome. The newly synthesized linear viral D N A is then translocated into the cell 's nucleus, where it is integrated into the host's D N A , using the viral ly encoded integrase enzyme. In resting cells, the newly synthesized H I V - 1 D N A may remain in an unintegrated circular form in the cytoplasm, until it is degraded or translocated to the nucleus fol lowing host cell activation. Once integrated, the proviral D N A may remain latent or be immediately transcribed into new viral R N A and proteins, through a complex interaction o f host cell derived and virally-mediated events. A detailed discussion of these events may be found in Bio logy and Molecular Biology o f H I V , by Ruth I. Connor and David D . Ho (333). 5 Clinical Course of Pediatric HIV-1 Infection "The common denominator o f infection with HIV-1 in children and adults is a profound immunosuppression, rendering the host susceptible to the development o f various opportunistic infections and neoplasms (7)." Many o f the cl inical manifestations o f HIV-1 disease are similar in adults and children. However, H I V - 1 infection has other effects, which may be more profound in or confined to infants and children as organ systems such as the central nervous system are still not yet fully developed. The course o f infection in vivo is controlled by complex interactions between the virus and the host's immune system. Fo l lowing initial infection with H I V - 1 , many individuals experience a period of cl inical latency, an asymptomatic phase with may last up to 10 years or more (334). In contrast to older children and adults, infants infected with HIV-1 in utero or during the perinatal period (during mid to late gestation or within 28 days o f birth) have a relatively short asymptomatic interval before they develop symptomatic disease. This being said, two distinct c l in ical courses are recognized. About one fifth o f maternally infected children have a rapid decline in C D 4 + T lymphocytes, resulting in an early diagnosis o f A I D S and death, whereas most have a form o f disease which progresses more slowly and is probably similar to that observed in adults (75,45,46,85,24). Retrospective studies o f perinatally infected children suggest that the median time from infection to the onset o f cl inical symptoms was 5-10 months (9,23,24), and that the mean incubation from birth to A I D S has been estimated as 4.1 months for rapid progressors and 6.1 years for slow progressors (46). The longest asymptomatic period reported in these studies was 7.3 years. It has been estimated that - 2 0 % o f HIV-1 -infected infants w i l l develop A I D S during the first year o f life (46), and >90% o f infected infants can be expected to develop HIV-1-related symptoms by age 12-18 months (67,75,24). Defining the potential factors which contribute to long-term survival in children can assist in understanding the mechanisms o f HIV-1 disease in this population. A number o f anecdotal reports or studies with small data sets have been published (38,49,220). In a larger study o f 182 children who survived longer than 5 years, De Martino and the Italian Registry for H I V - 1 Infection in Children found that a substantial number of children do survive after early childhood and that severe disease, low C D 4 + T cell count and p24 antigenemia do not necessarily preclude such long-term survival (221). 6 The exact reason for the bimodal distribution o f disease progression remains controversial, although it has been suggested that it may be associated with the timing o f infection during pregnancy and delivery. Blanche et al has shown that in maternally HIV-1-infected infants, the rate o f disease progression varies directly with the severity o f the disease in the mother at the time o f delivery (227). It has also been hypothesized that HIV-1 infection may progress at different rates based on the tissue distribution and maturity of C D 4 + T cells at the time o f infection, specifically, disease may progress more rapidly in infants i f the timing o f transmission in utero coincides with the period o f rapid expansion o f C D 4 immunocompetent cells in the fetus (41). This expansion would allow for the spread o f H I V - 1 into target cells, whose normal migration between the bone marrow, spleen, and thymus would allow for HIV-1 to diffuse throughout the body. The fact that the immune system in the infant is immature may also hinder its ability to respond effectively to restrict the spread of the virus. Thus, activation of HIV-1 infection in the perinatally infected infant could result in a much more rapid and severe destruction o f immune function. Infection o f immature target cells, such as noncirculating thymocytes or hematopoietic progenitor cells, may initially restrict levels o f H I V - 1 in tissues and peripheral circulation (21). However, i f immunologically naive cells are exposed to antigens or immune cytokines in the first few months o f life, this may activate those cells, which would in turn increase virus replication, thereby adding to the virus burden in plasma and circulating mononuclear cells (22). Addit ional factors, including the strain of virus passed from mother to infant, may also contribute to viral pathogenesis and disease progression in infected neonates and children. The hallmark of immune deficiency in pediatric HIV-1 infection, as in adult infection, is the progressive decrease o f C D 4 + T lymphocytes. However, a simple quantitative decrease o f this cell population does not sufficiently explain all irregularities in immune responsiveness known in pediatric H I V - 1 infection. Almost immediately fol lowing infection in adults (independent o f the route o f infection), a continual loss of CD4+ T lymphocytes usually in conjunction with a rise in C D 8 + T lymphocytes is seen, resulting in the reversal o f the ratio of C D 4 + to C D 8 + T cells in the peripheral blood. The most reliable predictor o f clinically relevant immune deficiency is the C D 4 + T lymphocyte count itself, with most opportunistic infections occurring when it decreases below 200-300 cells/mm 3 . 7 Neonatal HIV-1 infection is also characterized by a continuous decline in C D 4 + T cells (1,2,42-44). Newborns have a much higher percentage of circulating C D 4 + T lymphocytes than adults and also a higher percentage of naive compared with memory T lymphocytes. This persists over the first two years of life (132). Thereafter, the percentage and absolute numbers o f C D 4 + T cells in infants begin to approximate those o f adults. The rate of C D 4 + T cell decline is often much more rapid in neonatal infection than in adults, with the expected reversal o f the C D 4 / C D 8 ratio. A s in adult infection, C D 4 + T cell reduction in neonates correlates with cl inical disease as demonstrated by opportunistic infections and other symptoms. In multiple studies, loss o f C D 4 + T cells or reversal o f the C D 4 / C D 8 ratio has correlated with circulating viral load, loss o f mitogenic responsiveness, and development o f opportunistic infections, with the exception that opportunistic infections occur in H I V - 1 -infected neonates at much higher C D 4 + T cell counts than in HIV-1-infected adults. In one study, 8 o f 22 infants diagnosed with Pneumocystis carinii pneumonia had >450 C D 4 + T cells (47). B cell abnormalities have a major impact on immune deficiency in pediatric H I V - 1 infection, whether directly or as a consequence o f C D 4 + T cell dysfunction. Increased absolute numbers o f circulating B cells are present in HIV-1-infected infants. Hypergammaglobulinemia, especially o f the IgG or IgA class, is commonly the earliest laboratory abnormality of HIV-1 infection and is normally detectable by the age of 6 months (69,75). It is often used as an indicator o f HIV-1 infection during the first few months o f life. Levels o f IgG increase with the age o f the child and also with the stage of H I V - 1 infection, the highest levels being present in older children and when infection is symptomatic (44). Abnormalities in B cell function cause susceptibility to bacterial infections, particularly those caused by Streptococcus pneumoniae, H. influenzae, Salmonella species, Staphylococcus aureus and Escherichia coli. Repeated infections caused by these agents are considered AIDS-def ining illnesses in children (48). Other related immunologic abnormalities include decreased serum levels o f insulin-like growth factor-1 and growth hormone (49), or increased levels o f p 2 microglobulin and neopterin. The most common and specific laboratory findings associated with HIV-1 infection in infants are a decreased C D 4 + T cell count, an inverted C D 4 / C D 8 + T cell ratio, and polyclonal hypergammaglobulinemia. These findings are not only highly indicative of HIV-1 infection but also give us a quantitative measure of immune deficiency. A n y o f these findings along with any unexplained 8 marrow dysfunction (anemia, leukopenia, thrombocytopenia) should prompt further investigations to rule out HIV-1 infection. Many physical findings are associated with HIV-1 infection, mostly as a result o f secondary opportunistic infections and malignancies. Diffuse lymphadenopathy is the most common condition directly related to HIV-1 itself. Hepatosplenomegaly, parotid enlargement, and tonsillar hypertrophy may relate to this chronic lymphoid stimulation. In one prospective study, at least one of these findings was present in 73% of HIV-1 -infected infants before their first birthday (75). Parotitis, due to lymphocytic infiltration o f the parotids is present in 10-43% o f cases (130). Early symptoms of HIV-1 infection may be nonspecific, but their persistence and severity should raise suspicion (131). A common early presentation is recurrent oral candidiasis with failure to thrive. HIV-1 infection o f the central nervous system (CNS) causes several distinct neurological syndromes, including subacute encephalitis, vacuolar myelopathy, aseptic meningitis, and peripheral neuropathy. Significant neurodevelopmental changes also occur, with failure to reach specific milestones often noted very early on (49). It has been shown that macrophages within the brain and C N S are primary targets for HIV-1 infection. Detection o f H I V - 1 D N A in brain tissue (222,223) as wel l as the isolation o f infectious virus from the brain and cerebrospinal fluid (13,14) strongly suggests that H I V - 1 may be a causal factor in the subacute encephalitis associated with A I D S . Complications o f HIV-1 infection can be classified according to their etiology. Opportunistic infections are the primary sources o f morbidity and mortality in HIV-1-infected children. Fortunately, many o f these infections can be prevented or effectively treated. Bacterial Infections HIV-1-infected children are at risk for various types of bacterial infections. This may be caused by cellular immunodeficiency (Listeria monocytogenes, Salmonella species), neutropenia, frequent use o f indwelling devices (Staphylococcus species, Pseudomonas species), and inadequate humoral responses to encapsulated organisms such as o f Streptococcus pneumoniae and Haemophilus influenzae. Serious recurrent bacterial infections (two or more in a 2-year-span) constitute an A l D S - d e f i n i n g illness in children (224). Serious infections, which occur more frequently in HIV-1-infected children, as compared with uninfected children, include pneumonia, bacteremia, sinusitis, meningitis, urinary tract infection, and soft 9 tissue infection (133,48,134). Most o f these infections are caused by common childhood organisms, particularly S. pneumoniae (133). Gram-negative infections are frequently terminal, and often follow prior hospitalization and extensive antibiotic therapy (131). The use of intravenous gammaglobulin (1V1G) for the prevention o f bacterial infections has been controversial, mainly due to the lack of adequately controlled studies. A multicentre, randomized, double-blind, placebo-controlled trial o f the efficacy o f monthly I V I G versus albumin (placebo) infusions in 372 symptomatic HIV-1-infected children showed that I V I G was effective in prolonging the time to the development of a serious bacterial infection, and decreased the number o f serious bacterial infections and the number o f hospital admissions for children with C D 4 + T lymphocyte counts o f at least 200 cel ls /mm 3 (133). Two other trials have achieved similar results (225,226), although survival rates were unchanged. Overall , the use of I V I G is only recommended in selected groups of HIV-1-infected children. Mycobacterial disease is becoming increasingly important in HIV-1-infected children. Infections due to Mycobacterium avium complex ( M A C ) occur in severely immunocompromised patients (i.e. low C D 4 + T lymphocyte counts). These organisms (M. avium, M. intracellular) are found widely in the environment. Infections are usually disseminated and fever, diarrhea, and wasting are the most common manifestations. Wasting is the most serious problem and thought to be the most important factor which causes the life expectancy in patients with HIV-1 and M A C infections to be shortened (135). The organism may be isolated from feces, blood, bone marrow, sputum, and gastric lavage fluid. Improvement in many patients can be achieved with multidrug regimens (three or more drugs) i f given for prolonged periods o f time as early as possible in the course o f the disease (135). The effect o f treatment on survival is not known. Problems which are difficult to control with therapy are diarrhea, weight loss, and anemia. Agents used in the treatment of M A C infections are rifampin, amikacin, ethambutol, ciprofloxacin, and clofazimine (an agent commonly used for Mycobacterium leprae infections), and less commonly, cycloserine and ethionamide. Viral Infections Most viral pathogens reported in HIV-1-infected children are members o f the herpes-virus family. Table 2 shows the most common viral pathogens in HIV-1-infected children and their cl inical presentations (10). 10 Cytomegalovirus ( C M V ) infection occurs commonly and can affect life expectancy (140). The most important clinical manifestations associated with C M V infection are hepatitis, pneumonitis, esophagitis, encephalopathy, retinitis, and colitis. Dissemination with multiorgan involvement can occur. C M V colitis with and without hemorrhage in both HIV-1-infected adults and infants has been described (141). C M V encephalitis in infants with H I V - 1 infection has been documented. However, the diagnosis o f C M V encephalitis can be difficult because o f concurrent HIV-1 involvement of the central nervous system. C M V retinitis, common in adults, occurs less frequently in HIV-1-infected children (142). Treatment with ganciclovir is recommended, especially in patients with localized disease (i.e. retinitis, colitis), but renal and liver function and hematologic parameters should be monitored in light o f known drug toxicity. Concomitant treatment with zidovudine could be hazardous due to potential synergistic bone marrow toxicity. Table 2: Common V i r a l Pathogens in HIV-1-Infected Children Vi ra l Agent Cl in ica l Presentations Cytomegalovirus ( C M V ) Hepatitis, pneumonitis, retinitis, colitis, esophagitis, encephalitis Herpes simplex virus ( H S V ) types 1 and Recurrent or progressive mucocutaneous disease, may 2 disseminate Varicella-zoster virus M a y become progressive, recurrent, or chronic with hyperkeratotic lesions Zoster: may appear shortly after varicella Epstein-Barr virus ( E B V ) Lymphocytic interstitial pneumonitis (?), parotitis (?), lymphomas, hairy leukoplakia Respiratory syncytial virus ( R S V ) Higher incidence of pneumonitis, wheezing uncommon Measles Interstitial pneumonitis, rash may be absent Molluscum contagiosum virus Aggressive and longer cl inical course, larger lesions Papillomavirus Higher prevalence o f large condyloma acuminatum lesions, recurrence common Herpes simplex virus ( H S V ) also affects HIV-1-infected children and usually presents as recurrent episodes of gingivostomatitis or perineal ulcerations (3). These infections can become chronic and at times disfiguring. Treatment with topical, oral, or intravenous antivirals such as acyclovir is often necessary to control the infection. Severe primary chickenpox (caused by varicella zoster virus or V Z V ) or recurrent infection in a dermatomal distribution may be seen in some children. Recurrences may happen shortly (weeks/months) after varicella (143). A s these represent an unusual complication in young children, their presence can be considered a potential indicator o f immunodeficiency. Zoster infection may become 11 chronic and be associated with hyperkeratotic changes (143,144). Because o f the risk o f serious disease, the use of varicella zoster immune globulin ( V Z I G ) is indicated in HIV-1-infected children exposed to active cases o f chickenpox. In treatment of active disease acyclovir remains the drug o f choice for H S V and V Z V infections, but drug resistance is reported in HIV-1-infected adults. In these cases, foscarnet could be used since some favorable responses to this drug in these circumstances have been reported (145,146). Experience with foscarnet in children is limited. Vidarabine has not been shown to be effective in the treatment of herpetic infections in HIV-1-infected patients (145,146). Epstein-Barr virus ( E B V ) infection in HIV-1-infected children is suspected to play a role in the pathogenesis o f lymphoid interstitial pneumonitis (LIP) and certain lymphomas, though it's cl inical manifestations are not thoroughly understood. E B V is also implicated in hairy leukoplakia, an oral lesion quite prevalent in HIV-1-infected adults but uncommon in children. Increased morbidity and mortality from common respiratory viruses have been reported in children with HIV-1 infection. Symptoms associated with respiratory syncytial virus ( R S V ) infection include pneumonia, relative absence o f wheezing, and prolonged viral carriage. Mortali ty may be higher (20%) than in non-infected children (147). I f discovered early in patients with respiratory complaints, therapy with ribavirin might be instituted. Adenovirus and parainfluenza 3 virus may also have a more virulent course in these children (148,149). Fatal measles has occurred in HIV-1-infected children (150). Children exposed to measles should receive I V I G although the efficacy o f this treatment is unknown. Prevention o f measles (even though it is a live vaccine) is essential and immunization is recommended even in symptomatic HIV-1-infected children. Fungal Infections Mucosal infections with Candida species occur at some time in the majority o f children with HIV-1 infection. Oral candidiasis appears as pseudomembranous creamy plaques, as erythema on the hard palate or dorsum of the tongue (atrophic), or as an angular cheilitis. The plaques are easy to remove and an erythematous base remains. Cultures and biopsies are seldom required for diagnosis. These lesions usually respond to topical nystatin or clotrimazole troches. For patients who do not respond to the initial treatment, a short course o f oral ketoconazole may be beneficial. Recurrence is common, especially i f the patient is treated with antibacterial drugs, and may prove very difficult to control in some patients. These patients 12 may require chronic therapy. Candida esophagitis also occurs with or without concurrent oral involvement. C l in ica l symptoms o f esophagitis include pain on swallowing, poor appetite, weight loss, and vomiting. Candida esophagitis usually responds rapidly to oral ketoconazole. Intravenous amphotericin B is administered when treatment with ketoconazole fails. Disseminated candidiasis has been observed, usually in the hospitalized chi ld as a complication o f infected indwelling devices and antibiotic use (151). Secondary prophylaxis for thrush is frequently used. Most patients w i l l do well treated with topical agents like nystatin (Mycostatin) suspension or clotrimazole troches. Some patients w i l l require ketoconazole. Infections with other fungal pathogens are uncommon in pediatric patients (151). Pneumocystis carinii Pneumonia (PCP) P. carinii is a wel l -known cause o f pneumonia in malnourished infants and in immunocompromised hosts (136). P C P is the most frequent and lethal opportunistic infection in HIV-1-infected children under one year o f age (136), representing 39% o f AIDS-def in ing illness in children reported to the C D C through 1990. It is often the presenting illness associated with HIV-1 infection in infants (137,138). Most cases o f P C P occur between 3 and 6 months of age, and 10% to 20% of HIV-1-infected infants may acquire it (231). Approximately one third o f these infants die within 2 months o f a P C P diagnosis (231). The fact that it is a primary infection as opposed to a reactivation infection, may explain the different pathophysiology compared to in adults. Mortality from P C P is high in infants (often because the diagnosis is delayed, as HIV-1 infection may not have been previously suspected). In patients requiring respiratory support, acute respiratory distress syndrome is a frequent complication, further worsening the prognosis (139). Trimethoprim-sulfamethoxazole combination ( T M P - S M X ) and pentamidine are the two most commonly used drugs to treat P C P . Due to its safety and efficacy profiles, T M P - S M X is the first-line drug for treatment in children and pentamidine is reserved for those patients who are hypersensitive to T M P - S M X or who are considered treatment failures. There are few adequate data regarding the use o f steroids as an adjuvant therapy for P C P in children. McLaughl in et al describe greatly improved survival in children infected with H I V - 1 who have P C P -related respiratory failure when treated with adjunctive corticosteroids (228). One other previous study also 13 showed benefit (230). Unfortunately, no recommendations regarding the use o f steroids in children younger than 12 years o f age were made by the National Institutes of Health-University o f California consensus panel on the use of corticosteroids as adjunctive therapy for Pneumocystis carinii in 1990 (229). Because of the high mortality associated with P C P and since both primary and secondary prophylaxis has been shown to be effective in other immunocompromised hosts (i.e. children with leukemia) and in adults with HIV-1 infection, guidelines for P C P prophylaxis in HIV-1-infected or -exposed children have since been developed (232), based on age-adjusted C D 4 + T lymphocyte counts, to an equivalent o f 200 cel ls /mm 3 in adults (132,137,138). Thea et al have published a study reporting evidence that primary antimicrobial P C P prophylaxis is highly effective in decreasing the frequency o f P C P and early death in infants with perinatal H I V - 1 infection (233) Other Parasitic Infections Parasitic pathogens seen in association with HIV-1 infection in children are numerous and varied. Congenital toxoplasmosis with C N S involvement has occurred in HIV-1-infected children whose mother probably acquired toxoplasmosis prior to conception (152-154). In immunosuppressed HIV-1-infected women, chronic parasitemia may occur more commonly, which places their children at risk for vertical transmission of toxoplasmosis. Congenital Toxoplasma encephalitis with discrete nodules or abscesses can occur. Congenital toxoplasmosis should be ruled out in HIV- l - in fec ted young infants with neurologic involvement since it is amenable to therapy. Congenital toxoplasmosis can be difficult to diagnose because of a weak or absent antibody response in the child. Treatment with pyrimethamine and sulfadiazine with folinic acid is the approach o f choice when congenital toxoplasmosis is strongly suspected or documented. Toxoplasmosis in older children may have a clinical presentation similar to that in adults (i.e. Toxoplasma cerebral abscess). Infection in these cases wi l l usually be caused by ingesting raw or poorly cooked meat or being exposed to cat feces. Therapy is the same as described above. Diarrhea caused by the intestinal protozoans Cryptosporidium and Isospora belli is wel l recognized in HIV-l - infec ted patients. In immunocompetent individuals, cryptosporidiosis produces a self-limiting illness but in HIV- l - in fec ted patients it can produce a severe secretory diarrhea with massive fluid losses, much like cholera. Treatment is mainly supportive. Maintenance of fluid and electrolyte balance and 14 central hyperalimentation are the most important. Therapy of cryptosporidiosis is much more problematic (155). Diarrhea due to hospora usually responds well to T M P - S M X therapy. Lymphoid Interstitial Pneumonitis Lymphoid interstitial pneumonitis/pulmonary lymphoid hyperplasia complex (LIP) is one of the most common pulmonary conditions associated with HIV-1 infection in children (45,156) and is a criterion for diagnosis o f A I D S in children less than 13 years old (224). LIP is not seen in adults. Its onset is usually insidious and may not be cl inical ly evident in the early phases (3,156). Coughing is the most frequent symptom. Listening to the chest usually reveals normal findings but wheezing may be present. Clubbing occurs frequently, over time. The typical radiographic picture shows symmetric bilateral reticulonodular interstitial infiltrates and hilar adenopathy. W i t h progression of disease, the nodules enlarge and coalesce. Histopathology usually shows a mononuclear interstitial infiltrate mostly composed o f immunoblasts, plasma cells, and C D 8 + T lymphocytes. A peribronchiolar component and lymphoid nodule formation are sometimes seen and may represent a more advanced stage o f the disease (156). These lesions may remain static over long periods or progress to capillary block with arterial desaturation and chronic lung disease. In a small number of patients, L IP resolves spontaneously. Definitive diagnosis is made by lung biopsy. Potential causes of interstitial infiltrates other than LIP , such as tuberculosis, P C P , and C M V infection should be excluded and a trial o f antibiotics should be administered when the diagnosis o f LIP is considered in a symptomatic patient. In those who lack respiratory symptoms, the diagnosis is based on radiographic findings alone. Bacterial infection is a frequent complication but P C P is seldom seen. The prognosis is usually good when L I P is the first symptom in patients with HIV-1 disease (45). The pathogenesis o f this disorder is not well understood. L I P is the pulmonary manifestation o f a systemic process of immune activation, which includes parotid enlargement, generalized lymphadenopathy, and hypergammaglobulinemia as the most obvious manifestations (156). The management of most patients with L I P is limited to observation and monitoring o f oxygen saturation. Some patients require supplemental oxygen intermittently (overnight) or on a continuous basis. Anecdotal reports o f children with progressive L I P with oxygen desaturation treated with steroids suggest a beneficial effect. 15 Neoplasms Malignancies have been reported to the C D C as the A I D S defining illness in about 2% o f pediatric patients (197) and in a number o f patients, it was the initial HIV-1-associated illness. C N S lymphoma is found in up to 6% of adult patients but only a few cases have been reported in HIV- l - in fec t ed children (157,176,178). Burkit t 's Lymphoma/B-cel l leukemia, Hodgkin ' s disease, non-Hodgkin 's lymphoma, and pulmonary leiomyosarcoma have all occurred in children (197-200). Kaposi ' s sarcoma is rarely cl inically evident in children (201) and usually presents as diffuse lymphadenopathy. The incidence o f secondary malignancies in HIV-l - infec ted children wi l l most l ikely increase since treatment with antiretroviral agents wi l l prolong life, and because o f the relentless increase in the number of new HIV-1 infections. Many patients respond well to chemotherapy, especially when the malignancy is the presenting illness and the overall condition of the patient is good (198,200). Certain tumors like C N S lymphomas may be radiosensitive (157,176). HIV-1 infection should not be a contraindication for chemotherapy and radiotherapy. Better policies for the management o f these patients are needed, especially regarding el igibi l i ty for novel chemotherapy protocols, which often exclude patients with underlying immune disease. Another way o f reviewing the conditions associated with HIV-1 infection in children is according to the affected system, as the patient would present for medical evaluation. Central Nervous System and Developmental Complications C N S abnormalities are common and have been reported in 50% to 90% o f children with HIV-1 infection (157-159). In children, the most important problem is the development o f an encephalopathy, which may be the presenting feature o f HIV-1 infection for some patients (160). There is evidence suggesting that A I D S encephalopathy is a consequence o f H I V - 1 infection o f the C N S (158,161,162,170). H I V - 1 nucleotide sequences have been isolated from the brains o f both adults and children with encephalopathy. HIV-1 has also been cultured from the cerebrospinal fluid (14). Intrathecal synthesis o f specific HIV-1 antibody has also been demonstrated (158,162,169). A t least one report has suggested that tumor necrosis factor may mediate myelin damage (163,171). The severity o f encephalopathy in infants may be compounded by environmental considerations, including maternal use of il l ici t drugs during pregnancy. 16 Since HIV-1 infection occurs during the development o f the C N S , the associated encephalopathy has some features which are unique to children. In particular, loss o f developmental milestones, and diffuse bilateral pyramidal tract signs are reported. Acquired microcephaly, ataxia, seizures, and other movement disorders are also seen (168). Radiologic findings in H I V - 1 encephalopathy include calcification o f basal ganglia vessels, usually visible on computed tomography (CT) scan (164), and cerebral atrophy (on C T scan or magnetic resonance imaging ( M R I ) of the brain). Cerebrospinal fluid analysis may show mild pleocytosis or elevated protein levels. Recent studies have documented that more subtle neurodevelopmental deficiencies may occur in asymptomatic and in mi ld ly i l l HIV-1-infected children. In one (165), children with transfusion-acquired HIV-1 infection during the neonatal period had diminished school achievement and performed poorly in tasks requiring motor speed, visual scanning, and cognitive abilities when compared to matched uninfected controls. In another (166), language capabilities at 18 to 30 months were impaired in asymptomatic H I V -1-infected children born to substance-abusing mothers, compared to matched uninfected controls. Although the pathogenesis o f these abnormalities is unclear, HIV-1-infected children should be monitored routinely for neurodevelopmental deficits and referred for early intervention. Further studies are required to define the full spectrum o f neurodevelopmental and cognitive defects in this population. Finally, in a minority o f cases, a more progressive encephalopathy has been described, with a median survival o f less than one year (45). Pathologic findings described include corticospinal tract changes (167), demyelination with or without axonal damage (with evidence of H I V - 1 in brain tissue), severe white matter degeneration and inflammatory cell infiltrates with macrophages, microglial nodules with and without multinucleated giant cells, and various forms of vascular disease (158,171). Therapy with zidovudine may stabilize or reverse the progression o f this encephalopathy (172-174). In one study, brain growth, cognitive improvements, and improvement o f auditory brainstem responses were observed (175). Otorhinolaryngologic Complications Otorhinolaryngologic problems are common in HIV-1-infected children, but seldom life-threatening. Otitis media and sinusitis are common complications and may be acute, recurrent or chronic. The prevalence o f pathogens typifies that o f immunocompetent hosts. In otitis media, pneumococcus, Moraxella catharrhalis, and H. influenza prevail. Initial treatment should be the same as that for an 17 immunocompetent host. Unfortunately, symptomatic children do appear to have more therapeutic failures despite adequate antimicrobial therapy (179). The child with recurrent ear infections should be considered for antibiotic prophylaxis or ventilation tubes, as appropriate. Chronic suppurative infections require aggressive drainage and broad-spectrum intravenous antibiotic therapy that includes coverage for Pseudomonas. Patients wi th sinusitis usually require 14 to 21 days o f antibiotic therapy. Pneumonia may develop as a complication o f recurrent sinusitis and should be borne in mind i f appropriate symptoms develop. Nearly all patients develop lymphadenopathy, especially in the cervical region. L y m p h nodes are usually not tender, firm and freely moveable. When malignancy or opportunistic infections are suspected, a biopsy and/or aspiration are warranted. Otherwise, the nodes need only be observed. Bacterial adenitis occasionally occurs and is often found in association with L I P and elevated immunoglobulin levels. In these cases, the nodes enlarge quickly and are usually tender. This typically responds well to antibiotics. As described above, primary, recurrent, or chronic H S V infections o f the oral cavity are also seen. These lesions can progress and affect adjacent areas and may require oral or intravenous acyclovir therapy. V Z V may affect one or more branches o f the trigeminal nerve, causing acute and severe pain. Cardiovascular Complications Cardiac complications associated with H I V - 1 infection are increasingly recognized (182,183,188). These abnormalities are often minor and asymptomatic. In a prospective study o f 88 symptomatic H I V - l -infected children who underwent electrocardiographic and echocardiographic evaluations as part o f a zidovudine study, more than ha l f had abnormalities but only 14% o f patients required medical intervention for cardiac reasons (188). The prevalence of cardiovascular abnormalities varies among different studies. Wi th noninvasive methods, from 50% to more than 90% o f HIV- l - in fec ted children showed some kind of cardiovascular abnormality during the course of HIV-1 infection (182,184,188). The most common findings from echocardiographs are ventricular dysfunction and pericardial effusion. Pericardial effusion is found in about one third of children (182,183). Obvious signs o f cardiac dysfunction usually occur later in the disease (45). The presence of cardiac dysfunction may be masked by pre-existing clinical conditions such as hepatosplenomegaly, fever, pulmonary infections, and anemia 18 (182,183). Patients typically die o f noncardiac causes (183), but deaths attributed to heart failure have been reported (182). Pathologic examinations usually show enlarged hearts with hypertrophy and/or dilatation (182). Mononuclear infiltrates are seen in the myocardium (with or without necrosis) (182). Evidence of opportunistic pathogens has rarely been found (182,183,185). The HIV-1 genome has been isolated within the myocardium in adults and children using in situ hybridization (186,187), suggesting a direct role for the virus in cardiac disease. Other possible indirect mechanisms include autoimmunity and cytokine-mediated disease. There is no apparent association between the presence o f cardiac abnormalities and immunologic abnormalities (188). Patients with significant cardiac dysfunction usually respond well to medical intervention. Treatments include diuretics, digitalization, and fluid restriction and should be started when cl inical ly significant cardiac dysfunction is diagnosed (183). Zidovudine given over a 6-month period does not appear to alter the course o f cardiac disease (188). Anemia , hypoxemia, and nutritional deficiencies are potentially aggravating factors. The lack of adequate knowledge o f the pathophysiology o f cardiovascular disease limits capabilities to prevent or revert it in HIV-1-infected children. Gastrointestinal and Nutritional Complications The gastrointestinal tract is commonly involved in children infected with H I V - 1 . This frequently causes nutritional deficiencies and failure to thrive. Poor nutrition may be the result o f insufficient caloric intake due to infectious complications o f the oral cavity and esophagus, gastrointestinal malabsorption, and chronic or recurrent gastrointestinal or systemic infections. Acute and chronic diarrhea in HIV-1-infected children are the most common presenting conditions, many of which are treatable, especially i f a specific infectious etiology is identified. Nephrologic Complications Although data on renal disease in children with HIV-1 infection are limited, the condition does occur in this population (181). Based on published studies, 30-55% of HIV-1-infected children w i l l have renal disease or urinary and electrolyte abnormalities at some time during the course o f infection (234), and 3-29% wi l l have more definite evidence o f renal disease (181,234). In children, cl inical manifestations include proteinuria, edema, intermittent hematuria, metabolic acidosis, and hyponatremia. Persistent 19 proteinuria is a common early finding. Cl in ica l ly evident renal involvement is a late finding. In three studies, the mean age of onset o f progressive renal disease was between 35 and 39 months (range, 13-90 months) (45,181,234). Ultrasound examination usually shows enlarged and hyperechoic kidneys but these findings are non-specific and are similar to those o f acute glomerulonephritis. Pathologic examination o f kidney tissue obtained by percutaneous biopsy from children with clinically evident renal disease showed several types o f lesions, including focal glomerulosclerosis, mesangial hyperplasia, focal necrotizing glomerulonephritis, and minimal-change disease (181). Renal failure was associated with focal glomerulosclerosis but was not considered to be the direct cause o f death. Irrespective of the histology type and the level o f renal function the outcome was poor once cl inical ly evident nephropathy was diagnosed. Some of the medications used to treat HIV-related conditions cause renal toxicity and this requires careful observation. Results o f specific management of end-stage renal disease are discouraging. There is minimal experience with dialysis in pediatric patients with HIV-1-infection. Experience with peritoneal dialysis is very scanty but may be preferable to hemodialysis. Generally speaking, these patients are very poor candidates for renal transplantation. In many instances, supportive management is the best that can be offered to patients with renal failure. The need for renal biopsy should be considered on an individual basis since specific therapy might be used in certain glomerulopathies. Hematologic Complications Anemia, leukopenia, and thrombocytopenia are common in the course o f H I V - 1 infection in children. Microcytic/hypochromic anemia is most common. This anemia may have an iron deficiency component, which may be treated with an oral iron supplement. Anemia frequently complicates zidovudine therapy in about 20% o f children, and is typically macrocytotic. Autoimmune hemolytic anemia has also been described and is part o f the spectrum of autoimmune phenomena affecting these patients. Treatment with steroids may elicit some response (189). Neutropenia is seen in about half o f children undergoing zidovudine therapy and often requires dose reduction or temporary suspension of therapy (173). Thrombocytopenia occurs in about 13% o f HIV- l - in fec ted children (190) and may be the presenting condition (191). Significant bleeding may occur, including C N S hemorrhage. Bone marrow examination 20 usually uncovers normal to increased numbers of megakaryocytes and the majority o f patients have increased levels o f antiplatelet antibodies and circulating immune complexes. It appears that both shortened platelet life span and suppressed platelet production occur in HIV-1-infected patients. Zidovudine produces a temporary elevation in platelet counts in most HIV-1-infected adults (192-194), and children (195); however, many treatment failures in children have been reported (196). Some patients respond to high-dose 1VIG infusions (190), but this treatment rarely results in a sustained remission. Many patients with cl inical ly relevant thrombocytopenia w i l l require steroids and eventually splenectomy (190) in order to achieve a sustained remission. Clinical Interventions Significant advances have been made in the use of antiretroviral agents in HIV-1-infected children, building on therapeutic advances made in adult medicine. In this context, the developmental impact o f A I D S in children and the different clinical manifestations and altered pharmacokinetics o f drugs in the pediatric population have been considered. Most o f the new information comes from the use o f the dideoxynucleosides zidovudine (3'-azido-3'-deoxythymidine, Z D V , azidothymidine, A Z T , Retrovir), 2' ,3'-dideoxyinosine (didanosine, ddl , Videx) , and 2' ,3 '-dideoxycytidine (ddC, Zalcitabine). Zidovudine The first effective drug made available for the treatment of A I D S was the dideoxynucleoside zidovudine ( Z D V ) . It is a thymidine analogue whose principal mode of action is to block HIV-1 replication (235,237). Z D V is phosphorylated by cellular enzymes into a 5'-triphosphate form which interferes with the viral RNA-dependent D N A polymerase reverse transcriptase and chain elongation o f the viral D N A , resulting in inhibition of viral replication (235, 237). Z D V therapy in patients with advanced H I V - 1 disease prolongs survival, decreases the frequency and severity of opportunistic infections, improves neurologic function, transiently improves C D 4 + T-lymphocyte counts, and decreases the serum concentration o f HIV-1 antigen (236,193,238-242). Anemia and neutropenia are the most frequent adverse reactions associated with Z D V therapy (243). In a double-blind placebo-controlled study of adults with A I D S and advanced AIDS-related complex, patients treated with Z D V had improved survival and decreased incidence of opportunistic infections (236). Further studies of this kind, in subjects with asymptomatic (245) and mildly symptomatic 21 HIV-1 disease and less than 500 C D 4 + T lymphocytes/mm 3 also showed that Z D V therapy delayed progression o f HIV-1 disease and produced little toxicity (244). Z D V has shown beneficial effects in children and the Food and Drug Administration approved it in M a y o f 1990 for pediatric use. Its pharmacokinetics (202) and adverse effects (172,173) in children are very similar to those in adults. In a multicentre phase II study from the A I D S C l in i ca l Trials Group involving 88 children, commonly noted adverse effects were neutropenia (48%) and anemia (23%). In some cases, these complications reversed spontaneously but most required a dose reduction or discontinuation of the drug. Usually the dose is reduces to a half dose when the absolute neutrophil count is lower than 750 cells/mm 3 and discontinued when lower than 500 cel ls /mm 3 until the marrow function recovers. Some patients required blood transfusion. Reported beneficial effects o f Z D V in children have been weight gain, decreases in size o f the liver and spleen, lowering o f the total IgG and I g M toward more normal values, decreases in the amount o f circulating p24 antigen in serum and cerebrospinal fluid, transient elevation in the numbers o f C D 4 + T lymphocytes (average 12 weeks), decreased frequency of virus isolation, and cl inical (cognitive function) and radiologic improvement (brain growth) o f HIV-1-associated encephalopathy (172,173,175,17,203). HIV-1-associated thrombocytopenia may improve at least temporarily with Z D V therapy, as reported in several studies in adults (192-194,17). There are no large studies in children but anecdotal evidence supports the same conclusions. Failure o f Z D V to improve platelet counts in HIV- l - in fec ted children has been reported (196). A phase 1 evaluation o f the safety, tolerability, and pharmacokinetics o f Z D V administered to infants exposed at birth to HIV-1 was conducted. A total o f 32 symptom-free infants were enrolled before 3 months of age. Infants under 2 weeks o f age were administered oral doses o f 2mg/kg every 6 hours and older infants were given 3mg/kg every 6 hours. It was found that Z D V at those doses in that age group was well tolerated, although some anemia (62.5%) and neutropenia (28.1%) were present, which generally resolved spontaneously (246). Z D V at a dose o f 180mg/m 2/dose every 6 hours is the recommended regimen for symptomatic children. Complete blood cell counts should be monitored closely throughout therapy and the dose adjusted accordingly. 22 Didanosine Didanosine (ddl, Videx) (204-206) is an alternative nucleoside analog, which can be administered to children who cannot take Z D V , either because o f intolerance or disease progression while on Z D V . A phase I/I I, dose-ranging study by Butler et al. (204) investigated the cl inical virological and immunological responses of 43 children, 16 o f whom had been previously treated with Z D V and 27 o f whom had not. In this study, 30% o f the children showed an increase in C D 4 + T cell counts. Weight gain exceeded 10% in 30% of the children and a reduction in lymphadenopathy and hepatosplenomegaly in 63 and 83%, respectively. Improvement in the parameters investigated was seen in both previously ZDV-treated and ZDV-nai 've patients. Other beneficial effects reported were decreases in p24 antigen levels and resolution of thrombocytopenia. Stabilization of neurologic disease has been noted as wel l , and the drug penetrates the cerebrospinal fluid (206). Preliminary data indicate that the drug is well tolerated. Another advantage o f the drug is its longer dosing interval, being given at 200mg/m 2/day in two divided doses. Side effects include pancreatitis and neutropenia in small numbers o f patients. Retinal changes have been observed in some didanosine-treated children. The significance of this finding is still unknown. Zalcitabine Experience with zalcitabine (ddC) is limited. One of the original trials ( A I D S Cl in ica l Trials Group Study 138) studied the efficacy and safety o f zalcitabine monotherapy in 170 children with symptomatic H I V - 1 infection in whom therapy with Z D V had failed (247). The children were randomly put on one of two treatment regimens: 0.005mg ddC/kg every 8 hours (n=86) or 0 . 0 l m g ddC/kg every 8 hours (n=84). U p to 18% of the children had previously shown intolerance of Z D V , while the remainder had developed progressive disease while on Z D V therapy. The efficacy endpoints investigated included C D 4 + T cell counts, p24 antigen levels, weight changes and survival. Baseline C D 4 + T cell counts were low in this group of children, with a median count o f about 50 cells/mm 3 . The median baseline C D 4 + T cell count was 61 cel ls /mm 3 for the high-dose group and 44 cel ls /mm 3 for the low-dose group. C D 4 + T cell count in both groups initially decreased from baseline until around 12 weeks in the low-dose group and until around 24 weeks in the high-dose group, then began to increase again toward baseline. B y 36 weeks, the count had returned towards baseline values in the group taking 0.005 mg/kg, but not in the group taking 0.01 23 mg/kg. A 50% decrease in p24 antigen levels occurred in 65% of children taking low-dose ddC and in 57% of those taking the high dose. Over the 36-week period o f treatment, almost all o f the children showed an improvement in growth rate. Peripheral neuropathy did occur, but it was less common in those children than in adults receiving ddC, with only eight children out o f 170 developing this condition. These overall findings indicate that ddC is generally wel l tolerated in this group. Preliminary data on ddC indicate that neutropenia and pancreatitis with associated abnormalities in calcium and phosphate metabolism are the dose-limiting toxicities. St i l l , cl inical and laboratory measures have improved in some children during therapy with ddC. Combination Therapy A study by Pizzo et al. (248) investigated the combination of ddC and Z D V in children previously treated with ddC therapy for 8 weeks. 25 patients started out with the ddC monotherapy. 13 of them were then switched to an alternating schedule o f one week of ddC followed by three weeks o f Z D V for 12-18 months. O f the 13 patients, 11 gained weight and more than half exhibited an increase o f more than 10% in CD4+ T cell counts and the C D 4 : C D 8 ratio. Addit ionally, four patients suffering from encephalopathy all improved on the combination therapy. Husson et al. (249) investigated the combination of varying doses o f didanosine with Z D V in a phase I/I I study. 68 children who had not been treated previously or had developed intolerance to Z D V were enrolled. Eight dose combinations were investigated in the previously untreated children. A reduced dose of Z D V was used in children who had previously shown hematological intolerance to this drug. With the combination therapy, the geometric mean titres o f cultured virus showed a highly significant decrease from baseline after 12-24 weeks o f therapy, both in plasma and in peripheral blood mononuclear cells. There was a trend towards an increased weight for age in many o f the children; 49% gained more than 10% of their baseline weight within 24 weeks. Addit ional ly, there was a statistically significant increase in the mean C D 4 + T cell count from baseline to 24 weeks. This effect was predominantly seen in the ZDV -naive patients. A pharmacokinetic evaluation of the combination of Z D V and didanosine was performed as part o f this study (250). The area under the plasma concentration-time curve ( A U C ) remained unchanged when the 24 drugs were administered in combination as compared to when the drugs were administered singly. However, there was a greater interpatient and intrapatient variability with didanosine than with Z D V . Recently, Englund et al. (251) have suggested that in symptomatic HIV-1 -infected children, treatment with either didanosine alone or Z D V plus didanosine was more effective that treatment with Z D V alone. The efficacy of didanosine alone was similar to that o f the combination therapy and was associated with less hematologic toxicity. Over the past several years, significant progress has been made in antiretroviral therapy, both in children and adults. The standard of care has become the use of three or more agents, including members o f new drug classes, such as non-nucleoside reverse transcriptase inhibitors and protease inhibitors. These are reviewed in (338). The focus o f our work, as summarized in this thesis, has been based on children treated with nucleoside analogues, as extensively reviewed in this section. Early vs. Late Fetal Transmission Vertical transmission o f HIV-1 from an infected mother to her child now accounts for virtually all new cases of HIV-1 infection in infants and children in North America and Europe. HIV-1 can be transmitted before, during, or after delivery. Table 3 provides a laboratory-based definition o f early versus late transmission (329). Early (in utero) H I V - 1 infection Positive H I V - 1 culture or P C R within 48h o f birth" Late (intrapartum) H I V - 1 infection Negative H I V - 1 culture, P C R , o r p 2 4 antigen wi thin one week o f life and Positive HIV-1 culture, P C R , or p24 antigen between age 7 and 90 days 0 Positive cord blood sample must be confirmed with sample from peripheral blood obtained wi thin 48 h o f birth. Second confirmatory sample obtained outside neonatal period should be positive by H I V culture o f P C R . 0 Infant must not be breast feeding. Data are limited, however, about the relative proportion and efficiency o f transmission during the intrauterine, intrapartum, or postpartum periods. The clinical observation that 20-30% o f infected infants appear to develop rapid, early onset o f A I D S during the first few months of life (and have presumably been infected for a longer period o f time) provided initial indirect evidence for intrauterine transmission. The development of easily performed, reliable tests for the detection of HIV-1 proteins and nucleic acids in placental tissue and aborted fetal organs has provided more direct evidence for transmission during gestation. 25 Apparently, intrauterine transmission can occur during each trimester o f pregnancy. HIV-1 has been identified by culture or polymerase chain reaction (PCR) in fetal tissue obtained from therapeutic abortions as early as 10 weeks gestation (58-61,35,75,39). HIV-1 has also been detected in placental tissue from HIV-1-infected women as early as 8 weeks gestation by techniques including ultrastructural examination, virus culture, immunocytochemistry, and in situ hybridization (34,77-79). Al though contamination with maternal blood could potentially confound studies of placental or fetal tissue obtained from abortuses, many investigators have demonstrated the ability o f certain placenta-derived cells to support HIV-1 replication in vitro (33,80,81). Several investigators have detected virus in fetal tissue from abortuses obtained during the second trimester (286). Proviral sequences have been detected by P C R in fetal organs from .abortions performed between 16 and 24 weeks gestation (35,76,93). Although tissue from 33 fetuses between 16 and 24 weeks gestation were analyzed, potential maternal blood contamination could be excluded in only nine (35). HIV-1 proviral D N A sequences were detected in fetal thymus (6/8), spleen (8/9), and peripheral blood (5/9) sampled from these nine fetuses. Although all nine fetuses had evidence o f HIV-1 proviral D N A in one or more sites, all fetal organ samples were negative for HIV-1 by both virus isolation and p24 antigen testing. Lyman et al. (82) used P C R to detect proviral D N A in fetal central nervous system tissue from second trimester abortions. Proviral sequences were detected in 8/23 (30%) fetal organs. Maury et al. (33) have reported that placental tissue from both first trimester and term placentas expresses C D 4 and can be infected by H I V - 1 in vitro. Other studies have failed to detect either HIV-1 in fetuses or in most infected infants at birth, suggesting that mother to infant transmission may also occur at or near the time o f delivery (36). It is now felt that the majority of transmission events occur late in pregnancy or at parturition (37,22). T w i n studies that show a greater chance o f the firstborn twin o f an HIV- l - in fec ted mother becoming infected compared with the second born twin suggest that transmission o f HIV-1 infection may occur perinatally (37). A n updated analysis o f 92 twin pairs showed that 30 pairs were discordant for infection. In 23 o f 30 discordant twin pairs, the firstborn twin was infected, whereas in seven twin pairs the second born was infected (p=0.004). Therefore the presenting twin had a 3-fold greater risk of infection than the second born twin. Although this relationship was not influenced by the mode o f delivery, risk of transmission was higher overall in 26 twins delivered vaginally than those delivered operatively. N o data were available, however, regarding duration of ruptured membranes, length of labor, or indication for cesarean delivery. These data suggest that the presenting twin would have a prolonged exposure to infected blood and cervical secretions in the genital tract during the later stages o f pregnancy and delivery. If direct mucocutaneous exposure to virus during delivery or ascending infection during labor are significant factors in HIV-1 transmission, preventive strategies aimed at reducing virus inoculum by methods such as vir icidal lavage o f the birth canal or immediate surface decontamination o f the newborn might be utilized. Although an increased risk o f vertical transmission of HIV-1 has also been associated with prematurity (55,62,97), vaginal delivery (37,22,63,64,97), maternal viral load levels (287), and advanced immunodeficiency in the mother in some studies (55,62,63,66), such relationships have not been found by other investigators (37,67-69). There have been conflicting results from studies o f singleton births regarding the influence o f mode of delivery on transmission rates. Several investigators have reported similar infection rates regardless o f mode of delivery (67,69,83). The European Collaborative Study found that cesarean delivery tended to decrease the risk o f perinatal infection but only for emergency and not elective cesarean deliveries (63). This trend in relative risk reduction with cesarean delivery has also been reported by others, although reported reductions in transmission are marginal (64). HIV-1 can also be transmitted postnatally through breastfeeding. The role o f breastfeeding in postnatal transmission is clearly established for women who become infected after delivery, but the role of breastfeeding in women infected before delivery is uncertain (62,63,68,69,72). Although both free virus and proviral D N A have been found in breast milk (86,87), demonstration of breast milk transmission has been epidemiologically complex. Several studies have shown an increased risk o f postpartum transmission by breastfeeding among women with primary HIV-1 infection during the peripartum period (67,87,72,88,89,90, 62). The infant, therefore, could be potentially exposed to secretions or cells containing a high viral inoculum. Both high (67,62) and low (69) rates of HIV-1 transmission in breastfed infants have been reported. For example, the European Collaborative Study group reported a 2-fold increase in the risk o f infection among breastfed infants (31 vs. 14%), but only 38 of 828 children evaluated were breastfed (63). A meta-analysis of published prospective studies estimated the additional attributable risk of transmission posed by 27 breastfeeding from mothers with established H I V - 1 infection before pregnancy was 16% (95% confidence interval 8-25%) and was 26% (95% confidence interval 14-39%) from mothers who develop primary H I V -1 infection postpartum (63). Because of the risk of transmission and the availability o f safe alternatives for infant nutrition, breastfeeding is not recommended in Canada for HIV-1-infected mothers. Factors Associated With Risk of Transmission Maternal factors influencing the rate of HIV-1 transmission from mother to child are incompletely defined. Indirect measures o f increased viral burden, C D 4 + T cell count, maternal disease stage, and absence of maternal antibody to specific HIV-1 epitopes have been associated with an increased risk o f transmission. Addit ionally, characteristics o f the maternal virus strains may also influence transmissibility. Advanced disease stage, either during pregnancy or within months of giving birth, has not often been associated with increased rates o f vertical transmission in many studies (62,63,91,92). The presence o f HIV-1-related symptoms combined with decreasing C D 4 + T cell number was also associated with increased vertical transmission (94,62,108). A trend toward higher transmission for women with A I D S (31%) vs. asymptomatic women (14%) was found for the 615 women evaluated by the European Collaborative Study (94). A significant risk for transmission was found i f the women were stratified by C D 4 + T cell number (400/ul (19%), <700/ul (22%), and >700/ul (6%)) or C D 4 / C D 8 ratios (>0.6 (12%) and <0.6 (24%)). Several studies have noted a correlation between p24 antigenemia and risk o f vertical transmission (75,62). High p24 antigen coupled with low C D 4 + T cell number or advanced disease stage has been linked with elevated transmission risk in several studies (95). Several investigators have attempted to define the biologic and genetic features o f the virus associated with perinatal transmission. In one study, the biological characteristic o f the virus isolated from three transmitting and four non-transmitting mothers and their infants were evaluated (96). The virus isolated from the non-transmitting mothers grew more slowly and to lower levels than the isolates taken from mothers who transmitted the virus to their infants. The HIV-1 genome is characterized by a high degree o f genetic variability. The complex mixture of variants which exist in an infected individual are the result o f competition and selection in response to 28 immunologic pressure for change, and to alterations in cell tropism and replication efficiency among the variants (98). In an initial study to investigate the role o f selection in perinatal HIV-1 transmission, the distribution o f distinguishable genotypes transmitted between mother and child were analyzed (53). Comparisons of virus sequences from the three transmission pairs showed that specific sequences were highly conserved between each mother and her infant and that the infant's prevalent virus sequence was derived from a single variant present in its mother. Furthermore, the infants' virus sequences were less diverse than those of their mothers. Sequence sets from additional mothers and their infants have confirmed the observation that the infant's virus sequences are less diverse (96). This relatively narrow distribution is compatible with random transmission o f a limited number o f virions during gestation. Because o f genetic evolution, these variants may be a minor form in the mother postpartum but represent a prevalent form found during gestation. Alternatively, there may be selection of an antigenically distinct variant in the mother that escapes a critical immune surveillance mechanism. Specific biological characteristics o f the transmitted virus such as differences in cell tropism or replicative capacity may also be important. HIV-1-specif ic humoral and cellular immune responses, while ultimately unsuccessful, inhibit virus replication and spread after infection and may be important in determining long-term disease outcome (99). Maternal HIV-1-specif ic immune response may be involved in preventing, or possibly enhancing, vertical transmission. Several investigators have proposed the level and specificity o f maternal anti-HIV-1 specific antibody may be important in determining transmission. Antibodies to the hypervariable domain o f the gp l20 (the V 3 region) have HIV-1 neutralizing activity in vitro. Mutations in this region have been associated with changes in cell tropism and neutralization escape (98). Several early studies have suggested that HIV- l - in fec ted pregnant women with high antibody titres to conserved portions o f the V 3 hypervariable loop and/or high avidity-high affinity antibody against the principal neutralizing domain o f the V 3 loop may have a lower rate o f HIV-1 transmission to their infants (55,56,66). However, these investigators evaluated antibody to V 3 peptides that encompassed different areas o f the V 3 loop and more recent studies could not replicate these associations (71,100). Other factors, such as the maternal cytotoxic immune response to H I V - 1 , may be important in reducing transmission. 29 Additionally, enhancement o f the maternal humoral and/or cellular immune response to HIV-1 through passive or active immunization, or both, may also be helpful. If protective virus epitopes that induce antibody with broad neutralizing capacity can be identified, passive immunization o f the mother and/or infant with one or more monoclonal antibodies or active immunization with a subunit HIV-1 vaccine may provide optimal preventive interventions. If, however, there is selective vertical transmission o f a maternal virus neutralization-escape variant, a polyvalent hyperimmune HIV-1 globulin preparation or vaccine w i l l be necessary. Fetal cell susceptibility to HIV-1 infection could vary by gestational age (possibly because o f developmental differences in C D 4 expression), and different fetal organ systems could vary in susceptibility to infection. Immature thymic cells have been shown to be readily infected with HIV-1 (101), and neonatal and cord blood macrophages have been found to be more susceptible to infection by HIV-1 isolates than adult macrophages. If certain fetal cells or organ systems were particularly susceptible to HIV-1 infection, virus could infect these tissues (i.e. the thymus or central nervous system), escaping detection in peripheral blood samples obtained during the neonatal period. Infection of fetal stem cells may be more immunologically devastating than infection o f more mature cells because o f resulting stem cell destruction or dysfunctional cellular maturation, perhaps resulting in the more rapid disease course observed in perinatal HIV-1 infection when compared with HIV-1 infection in adults. The potential role o f the maturing immunologic capabilities o f the fetus and fetal response to HIV-1 infection has not yet been evaluated; however, an immature immune system may be less able to restrict HIV-1 replication. A relative deficiency o f circulating HIV-1 gag-specific cytotoxic T lymphocytes has been described in infants acquiring H I V - 1 infection during gestation when compared with HIV- l - in fec ted adults (102). It has been theorized that HIV-1 infection of early precursor thymic cells could lead to immunologic tolerance, inhibiting the ability to mount an effective immune response owing to perception of HIV-1 antigen as " s e l f (103). Prevention and/or treatment o f fetal infection may prove difficult. Indirect therapy to the fetus could be provided by transplacental passage o f maternal antiretroviral therapy. However, toxicity to the developing placenta and fetus is a concern. Passive immunization of the fetus with neutralizing antibody could be 30 accomplished through transplacental active transport o f antibody exogenously administered to the mother after the second trimester o f pregnancy, or induced in the mother by active immunization. This approach should be prioritized in research settings. Intensive exposure o f the thin skin or mucosal surfaces o f the fetus or newborn to maternal secretions during birth or through swal lowing o f infected amniotic fluid could provide a significant dose o f virus. Modification of obstetrical practices could influence virus transmission occurring during the intrapartum period. If intrapartum transmission of HIV-1 occurs primarily through direct exposure o f the infant to cel l -associated or free virus in genital secretions, cesarean section performed before labor might be expected to reduce the risk o f transmission. Addit ional ly, virucidal cleansing o f the birth canal before vaginal delivery and immediate surface decontamination of the infant by washing may provide more practical strategies to reduce transmission. However, i f maternal-fetal blood exchange at the time o f delivery is a significant source of virus, such measures might prove less beneficial. In the absence o f prospective, controlled evaluations of operative delivery, current obstetric guidelines do not recommend cesarean section in H I V -l-infected pregnant women other than for standard obstetric indications. Provision o f antiviral therapy to the infant before intense virus exposure during delivery could reduce the risk o f infection. Transplacental passage o f antiviral drugs provided to the mother during labor could provide systemic antiviral activity in the infant at the time o f exposure during delivery. Maintenance o f antiviral activity in the infant for a period after delivery through short-term administration o f an antiretroviral agent might be desirable. Connor and the Pediatric A I D S Cl in i ca l Trials Group Protocol 076 Study Group published a study in 1994 which indicated that in pregnant women with mildly symptomatic HIV-1 disease and no prior treatment with antiretroviral therapy during their pregnancy, a regimen consisting o f Z D V given ante partum and intra partum to the mother and to the newborn for six weeks reduced the risk o f maternal-infant HIV-1 transmission by approximately 67% (290). The role of the newborn immune response to HIV-1 in averting transmission is unclear. Although one report noted reduced cellular immunity in HIV- l - in fec ted infants (102), other researchers have noted relatively normal cell-mediated and humoral immune responses in HIV- l - in fec t ed infants during the first 2 years o f life, with subsequent weakening (104). The presence o f HIV-1-specif ic antibodies mediating 31 cellular cytotoxicity, neutralization and syncytium inhibition has correlated with slower disease progression in infected infants in several reports but obviously did not prevent transmission (105,106). Several factors potentially affect the transmission o f HIV-1 from mother to fetus. These include 1) the placental barrier against HIV-1 transmission, 2) transfer o f HIV- l - in fec ted maternal leukocytes during pregnancy or during labor, in part influenced by placental pathology associated with A I D S , 3) qualitative and quantitative characteristics o f the maternal virus, and 4) the nature of the maternal immune response against the virus. Addit ional virologic characteristics may influence in utero transmission o f H I V - 1 . It was noted that minor genetically distinct subsets o f HTV-1 were selectively transmitted from infected mothers to their fetuses (53). Such observations imply the existence o f either selective transmission o f specific HIV-1 strains which can cross the placenta or HIV-1 strains with enhanced ability to infect the fetus. The phenomenon of selective cellular tropism o f HIV-1 is well established. Thus it is conceivable that certain strains may have particular tropism for the placenta, and are subsequently passed on to the fetus. The influence o f specific anti-HIV-1 immune responses in determining the rate o f transmission o f the virus to the fetus is not clearly understood. Maternal g p l 2 0 antibodies have been reported to reduce transmission in several reports (37,66) but not in others (71). L o w maternal anti-gpl20 antibody titres have been associated with an increased risk for vertical transmission of HIV-1 (55). Initial studies (56) suggested a role for antibodies to the principal neutralizing epitope (V3) of HIV-1 in the transplacental transmission o f the virus, but subsequent work shows no correlation (57). Diagnosis of HIV-1 Infection in Children Early diagnosis o f HIV-1 infection in infants is essential to identify those patients who might benefit from early antiviral therapy, prophylactic treatment for opportunistic infections, along with aggressive treatment of bacterial infections, growth and development disorders, and psychosocial problems. Differentiating infected from uninfected infants w i l l eliminate unnecessary procedures and therapy in those who are not, in fact, infected with HTV-1 . Some early diagnostic tests may also provide information which helps establish the timing of transmission from mother to infant. 32 It is often difficult to diagnose HIV-1 infection in infancy, requiring a positive culture or direct viral detection assay results. These tests may be falsely negative during the first few weeks o f life in approximately half o f those infants who are subsequently shown to be infected. The presence o f A I D S -defining symptoms and a declining C D 4 + T cell count provides strong evidence for infection during infancy but it would be best to make the diagnosis at an earlier point in time to optimize the results o f medical intervention. During the first 12-18 months o f life, serologic testing o f a chi ld born to an H I V - l -infected mother w i l l not be useful in determining the infection status of that infant because o f the presence of maternal antibodies in the infant's circulation, acquired passively from the mother in utero. In infants younger than 18 months, tests for IgG antibody to HIV-1 in the circulation do not differentiate between infant and maternal antibody. This is quite important, as only 13-35% o f infants w i l l actually be infected. A more detailed discussion o f diagnostic modalities and maternal/fetal H I V - 1 transmission follows. Antibodies to HIV-1 Serologic tests for HIV-1-specif ic antibodies are the mainstay o f the laboratory diagnosis o f H I V - 1 infection in adults, children suspected o f acquiring HIV-1 through nonperinatal routes (i.e., blood or blood products), and children with suspected perinatally acquired infection who are older than 18 months. IgG antibodies are generally detectable within 4-12 weeks after exposure (111) and throughout the entire . course o f the disease; however, patterns o f antibody production to certain viral proteins may change over time. The most common antibody screening tests are enzyme-linked immunosorbent assays ( E L I S A ) used in conjunction with confirmatory tests such as the Western blot or immunofluorescence assay ( IFA) to confirm the specificity o f E L I S A reactions. Studies have shown that virtually 100% o f infants born to seropositive mothers w i l l test antibody positive at birth, but only 20-30% w i l l be infected. Those who are uninfected lose maternal antibody usually between 6 and 12 months of age, but a small proportion may retain maternal antibody for up to 18 months (63,67,112,279). A positive antibody test alone identifies perinatally-exposed infants. Careful follow-up and management is warranted for those cases. Cl in ica l evaluation with repeated testing over at least the first 2 years of life has been the primary means of establishing the diagnosis in these infants. Infants who become antibody 33 negative (serorevert) and remain well are generally considered uninfected, although there have been a few reports o f asymptomatic infants who have seroreverted but have positive virus tests such as culture or P C R (113). In addition, rare reports have described perinatally infected children who had lost maternal antibody and later seroconverted (114,115). Repeat Western blot testing over the first year o f life and comparison o f the band pattern over time can occasionally identify an infected infant. The infant's band pattern at birth is usually identical to the mother's pattern, since most o f the detectable antibody is o f maternal origin. The appearance of new bands on postnatal samples that were not present in the birth sample indicates de novo production o f antibody by the infant (116). Because maternal I g M and IgA antibodies do not cross the placenta, the presence o f these HIV-1-specif ic antibodies can be used to indicate the presence o f HIV-1 infection in infants. Several studies indicate that HIV-1 -specific IgA assays can detect most HIV- l - in fec ted infants by the age o f 6 months but are generally negative in infected infants younger than 3 months, presumably because HIV-1-specif ic IgA antibody is not yet produced in sufficient quantity (52,117,280). Less success has been obtained using HIV-1 -specific I g M assays, presumably because production o f this antibody type in infants is more transient and at lower serum levels (65). In Vitro Antibody Production Assays Another potential diagnostic test is the in vitro antibody production assay ( I V A P ) . This assay detects the presence o f HIV-1-specif ic antibody-producing B lymphocytes in the infant, which indicates that the infant's immune system has been stimulated by HIV-1 infection. The standard I V A P assays require 7-10 days to complete. Two methods for this technique have been described. In one method, P B M C s are separated from whole blood, carefully washed to remove plasma, placed in medium, and stimulated to produce antibody with either pokeweed mitogen (127) or Epstein-Barr virus (50). HIV-1-sensit ized B cells w i l l produce antibody that is released into the culture supernatant, and can be detected using standard methods. There are some important limitations to this assay. False-positive tests in uninfected infants have been reported in the first 2 months o f life. These spurious results might result from the detection o f maternal B lymphocytes that are producing antibody but may not harbor the virus. Addit ional ly, in the presence o f 34 abundant maternal IgG in the infant's serum, false-positive tests may result from maternal anti-HIV-1 antibody that has adhered to infant B cells in the culture. This method is not currently used in clinical practice. p24 Antigen Assay The standard p24 antigen assay has been a helpful but limited method for diagnosing pediatric HIV-1 infection. Studies of infants born to HIV- l - in fec ted mothers have found very few infants to be antigen positive early in the course o f infection, because of the apparent low levels o f antigen in the first month of life and the presence o f excess maternal antibody which complexes to any free p24 antigen that is present (112,118). Despite these limitations, the antigen test can be helpful in certain settings. Hypogammaglobulinemic infants with HIV-1 infection can have positive antigen tests (119). Recent studies have shown that modification o f the standard p24 antigen assay by acidification o f the sample to dissociate the immune complexes can increase the sensitivity o f the assay to detect HIV-1 infection in infants (122,123). The immune complex dissociation p24 antigen assay has been shown to be highly sensitive and specific for diagnosis o f HIV-1 infection in infants, although samples taken in the first week o f life have been somewhat problematic in that false-positive and false-negative tests have been observed. HIV-1 Culture Virus culture is one o f the most sensitive techniques for detecting HIV-1 infection in infants and is used extensively in research and clinical settings. Micrococultures in 12-, 24-, and 96-well plates are equally sensitive and specific compared with standard flask methods and require far less blood (18). HIV-1 culture is not useful as a rapid diagnostic test, because cultures typically take 7-28 days or more to complete. The standard technique involves coculturing patients' P B M C s which have been isolated from whole blood with phytohemagglutinin-stimulated P B M C feeder cells from healthy uninfected donors. A t least 2x10° patient cells should be cocultivated with an equal number o f donor cells. The cocultured cells should be stimulated with T cell growth factor (interleukin-2), which enhances viral growth. Cultures must be supplemented every 3-5 days with fresh feeder cells and monitored at least weekly for the presence of virus by measuring i f HIV-1 is present in culture supernatants (124,125). 35 Techniques which specifically quantitate the amount o f HIV-1 growing in culture have also been developed. The technique involves coculturing serial dilutions o f patient P B M C s (separated from whole blood) with a constant number o f uninfected donor cells (127). Cultures are then monitored in the same way as described above. The lowest dilution o f cells required to produce a positive culture is the end point, and the titre o f HIV-1 is expressed as the tissue culture infective dose per 10 6 cells. Plasma viral loads can also be measured using serial dilutions o f plasma cultured in the same manner as above and results expressed as tissue culture infective dose per mill i l i tre plasma (127). B y these techniques, studies have found that the mean viral titres in plasma were lowest in asymptomatic HIV- l - in fec t ed patients and higher in symptomatic patients (127). Likewise, the percentage o f infected P B M C s was also lowest among asymptomatic patients and higher among symptomatic patients. This assay is more commonly used for clinical monitoring than as a diagnostic test. Polymerase Chain Reaction P C R is the most sensitive diagnostic technique for detecting H I V - 1 . D N A sequences in P B M C s can readily be detected in infants. The test only requires ~ 1ml blood and can be completed in a matter o f hours (283). Studies evaluating the use o f P C R for early diagnosis o f H I V - 1 infection have shown that -30-50% o f HIV- l - in fec ted infants w i l l test positive close to the time of birth (112,22). This percentage increases to nearly 100% by age 1-3 months. These sensitivities are comparable with those o f virus culture (125). P C R can also be used to measure active virus replication by detecting viral R N A in plasma (22). Several studies indicate that infants who lose maternal antibody and remain healthy are uninfected based on negative virus cultures and P C R (126). Some laboratories have reported rare seroreverting children who w i l l occasionally test positive by P C R on a single specimen, but negative on subsequent specimens (126, 118,281,282). The reasons why these infants tested positive on one occasion are unclear. A mix-up in specimens or other laboratory errors is one possible explanation (282). Alternatively, transient P C R positivity may represent 1) true infection that has cleared, 2) persistent infection that is below the level detectable by the test, or 3) maternal blood cell contamination. Continued evaluation o f these children is important. 36 Surrogate Tests-Immunologic Parameters Tests which measure immunologic abnormalities most commonly associated with HIV-1 infection can also be useful for the diagnosis o f infection in infants. According to the C D C classification system for HIV-1 disease in children, the combination o f both humoral and cellular immunodeficiency in a symptomatic infant born to an HIV- l - in fec ted mother is diagnostic o f HIV-1 infection in the absence of specific diagnostic tests (224). The classic immunologic abnormalities include low T helper (CD4+) lymphocyte counts (adjusted for age), elevated T supressor (CD8+) counts (particularly early in life), reversed C D 4 : C D 8 ratio; depressed lymphocyte responses to mitogens in vitro, strikingly elevated immunoglobulin levels (most commonly IgG) or (rarely) hypogammaglobulinemia, and decreased specific antibody responses. C D 4 + T cell counts are used to monitor immunosuppression and disease progression over time. The most important factor which must be considered when choosing assays for the diagnosis o f perinatal HIV-1 infection is the age o f the infant at the time o f testing. Although the precise kinetics o f viral replication and antibody production in infants remain to be defined, several studies indicate that virus load is probably lowest at the time of birth in most infants and increases during the first few months o f life (125,22,112). IgA antibody production appears to also be low during the first few months of life (117,52). I V A P assays may yield false-positive results in the first 1-2 months o f life. Thus, in the neonatal period, most studies have shown that the greatest percentage of infected infants are detected using sensitive virologic assays such as P C R or virus culture (284). Several assays can detect infection by age 3-6 months (128). Even the most sensitive assays can detect no more than 50% o f infected infants around the time of birth. This is l ikely because the viral load is extremely low, virus is suppressed or sequestered in other tissues, or infection has only recently been transmitted during labor. Table 4. summarizes the relative sensitivity o f early diagnostic tests for HIV-1 infection in infants from birth to 6 months (335). Table 4: Sensitivity o f Early Diagnostic Tests for HIV-1 Infection in Infants Time to Detection Laboratory Method" 1 week 2-4 weeks 1-2 months 3-6 months >6 months HIV-1 IgA antibody testing ( E L I S A ) <10% 10-30% 20-50% 50-80% 70-90% E L I S P O T > 9 5 o / o > 9 5 o / o I V A P >95% >95% HIV-1 p24 antigen 10-25% 20-50% 30-60% 30-50% 20-40% P C R 30-50% 50% 70-90% >95% >95% V i r a l culture 30-50% 50% 70-90% >95% >95Q/0 a H I V - l IgA antibody testing (ELISA/Weste rn blot) not useful for detection o f HIV-1 infection. 37 The stage of disease also affects the l ikelihood that a given assay w i l l detect the presence o f HIV-1 infection. A s disease progresses, the virus load both in the plasma and cells increases (127,19). Some severely immunosuppressed adults and children with end-stage disease may lose antibody but this is highly unlikely to occur in very young infants. Some perinatally infected infants may be hypogammaglobulinemic and w i l l not produce antibody and this should be considered in appropriate settings. However, the t iming o f transmission may also affect the l ikelihood o f early detection o f infection in perinatally infected infants. Theoretically, infants infected in utero should be positive at birth, whereas those infected late in pregnancy and during the intrapartum or postpartum periods may not become positive until sometime after birth. One study evaluating P C R , virus culture and plasma p24 antigen tests for diagnosis o f HIV-1 infection in children under the age of 6 months found that P C R and culture were comparable in sensitivity, detecting 90% of all positive specimens. Both assays found HIV-1 in only half o f infected newborns, suggesting that this fraction o f children was infected during gestation, and the rest were infected at or near the time of delivery. Plasma p24 antigen was detected in 75% o f all samples tested but in only half o f infected children during the first 2 months of life and 88% o f samples from children during the next 4 months (285). With respect to immunologic assays, low C D 4 count and high Ig (particularly IgG) levels in a child with physical signs of HIV-1 infection (diffuse lymphadenopathy, hepatosplenomegaly) and a possible source of exposure to HIV-1 are very suggestive of HIV-1 infection. pVmicroglobul in and neopterin levels have also been suggested as potentially valuable markers for diagnosis o f HIV-1 infection in children (288,289) but they are too non-specific for widespread use. Prenatal diagnostic techniques available today, such as ultrasound, amniocentesis, chorionic vil lus sampling ( C V S ) , cordocentesis, fetal blood sampling, and embryoscopy may allow physicians to identify fetal HIV-1 infection early in gestation. This would allow HIV-1 seropositive women to better decide whether to carry or terminate their pregnancies. Furthermore, as therapeutic agents become available for use in pregnancy, the H I V - 1 status of the fetus may be used in deciding i f and when to start therapy and in assessing responses to therapy. There are, however, several important questions which w i l l need to be 38 answered before prenatal H I V - 1 diagnosis w i l l be conducted routinely. It must first be determined that an HIV-l - infec ted fetus can be detected with a high degree o f accuracy. The risk o f both missing the diagnosis in an infected fetus and making the diagnosis o f fetal infection because o f maternal contamination o f fetal specimens must be greatly reduced. Finally, it is crucial that the techniques used have virtually no risk of infecting an otherwise healthy fetus. Although the technology may exist to satisfactorily achieve these goals, the tests we have today do not provide any safe or reliable prenatal diagnostic methods for H I V - 1 infection o f the fetus. Virologic Hypothesis of HIV-1 Initial infection in older children and adults can be asymptomatic or result in an acute, self-limiting mononucleosis-like syndrome occurring 2-4 weeks after infection and resolving within 1-2 weeks (11,12). Both the acute HIV-1 syndrome and asymptomatic infection are followed by production o f anti-HIV-1 antibodies, termed seroconversion, which usually appear 6-12 weeks after infection. High levels o f infectious HIV-1 have been demonstrated in both the plasma and peripheral blood mononuclear cells ( P B M C ) (15, 16) o f infected individuals during primary infection, suggesting that the initial viral burden may be rather high prior to the development of an immune response to the virus. In fact, such a response does develop, leading to the restriction o f viral replication and the establishment o f chronic infection (16). In general terms, lower levels o f HIV-1 are found in the plasma and P B M C s o f asymptomatic or mildly symptomatic children whereas severely symptomatic children may have a virus burden comparable to that found in adults (17-19). However, in studies by Saag et al. (20), viral levels in the plasma o f each o f five children infected in the perinatal period did not appear to relate to immune disease or duration o f infection. This is in marked contrast to findings in adult patients, where levels o f plasma viremia decrease rapidly after seroconversion and remain relatively low throughout the asymptomatic phase o f infection (324). Thus, significant differences in the virology of HIV-1 may be present in infected adults, as compared to children. In addition, a shift has occurred in the accepted thinking about how HIV-1 coexists with its human host. Previously, it was thought that fol lowing seroconversion, a period of latency was entered, during which time little viral replication occurred. F i g 3 shows a generalized picture o f the pattern o f HIV-1 infection in 39 vivo according to this model. Fo l lowing primary infection, an extended period exists which lasts 2-15 or more years and is characterized by low but steady viral loads in the circulation (252). This time seems to be a period o f relatively little activity, however, recent data (summarized below) have shown this to be a time o f extensive viral replication and cell k i l l ing . Recognition o f this phenomenon has allowed a major shift in our understanding o f the virus-host interaction. In HIV-1 infection, we now have to think o f A I D S as being the end product o f damage accumulated during the entire course o f infection, even while the patient seems healthy, rather than a "threshold" phenomenon developing in isolation as the host's immune response to HIV-1 suddenly collapses. Relative amounts CD4 cells * Virus load ' 1 \ V/.^.^*- response i / 1 ~~ iL 2-10 Weeks 2-15 Years Fig. 3: The pattern of HIV infection in vivo. Shown is the typi-cal picture observed. There is considerable variation from patient to patient (252). The fact that genetic diversity accumulates very rapidly in HIV-1 populations growing in an individual wai the first clue that the cl inical ly latent period might be a time of rapid ongoing virus replication (253). The only explanation for how so much diversity could accumulate is that viral turnover continues during the time of relative equilibrium. W h y then isn't the virus load in an infected individual constantly increasing? 40 The steady state concept was conceived to explain this phenomenon (254). In this case, the rate o f infection of new cells equals the rate o f death of infected cells. The rate o f cell replacement equals the rate of death of infected plus uninfected cells. The amount o f circulating virus equals the balance between the rate of virus release into the blood and its rate of clearance from the blood, and the rate o f virus release is proportional to the number o f infected virus-producing cells (255). This new steady state concept is o f paramount importance. In its simplest form, the steady state can be visualized as a collection o f cells with similar kinetic rates o f infection, virus production, and cell death. For every cell infected, one infected cell dies, each infected cell makes on average the same amount of virus, and for each vir ion released into the blood, one vir ion is removed from the blood. Real life is, o f course, much more complicated that this. Examining numbers of infected cells or the amount o f virus in the blood (256,257) as a snapshot, does not disclose the underlying dynamics. In order to understand the dynamics, the system has to be disturbed and the response to that disturbance measured. Introducing antiretroviral therapy is one way to disturb the system. In studies performed on viral production in such systems, the number o f virus-producing cells declines with kinetics which reflect their natural half-lives as well as the clearance rate of free virus. Several papers were published based on these studies (258-260), leading to the generation of mathematical models to summarize viral dynamics in the infected host. The half-life o f the circulating virus population in the steady state is 1-1.5 days. The population o f wild-type viruses is decreased to as little as 1 % o f its original level fol lowing ~ 1 week. If any viral replication is allowed to persist in the presence o f drugs, almost al l genomes present in the circulation after ~ 2 weeks contain mutants which are resistant to one or more of the drugs being used, and viral load approaches pre-treatment levels. Another group of more refined studies (261) has helped us expand on these conclusions. The overall average replication cycle is approximately 2 days. This means that 180 generations pass per year, and that 1000 generations pass in 5-6 years. Less than 1% o f the virus in blood at any time comes from cells infected more than 2 weeks previously. This means that at most, only a minor contribution comes from latently or chronically infected cells. The kinetic parameters seem to be independent o f the clinical state. In addition, the steady state is quite vigorous over a wide range o f virus replication efficiencies. 41 There are controls in place which maintain the steady state. The balance between immune response and virus replication, between immune response and antigenic variation (262), and the availability o f target cells (255,263) is somehow maintained. Understanding and modeling the balancing forces which keep HIV-1 infection under control for such a long time in spite o f the rapid dynamics o f infection is a research aim of major importance. Recently, the significance o f plasma viral load in relation to this new model o f viral dynamics has been elucidated. It has been clearly shown that there is an excellent correlation between the steady state level o f virus in blood and the probability o f disease progression. Mellors et al published an elegant study which showed that higher viral loads present at the time a patient is diagnosed predict much more rapid disease progression (264). They also showed that plasma viral load was a strong predictor o f rapid progression independent o f C D 4 T cell count (265). These data have been confirmed in patients receiving antiretroviral therapy, with a 50% reduction in HIV-1 R N A level being associated with a 27% decrease in the adjusted risk of disease progression (266). Addit ional evidence to support this has been presented by Jurriaans et al. (273), Henrard et al . (274), and Katzenstein (325). It also appears that the level o f plasma HIV-1 viral load also predicts the progression rate of disease in children (267-272). Shearer et al. (276) also showed that the level o f plasma H I V - 1 viral load in infants predicted risk for disease progression and have demonstrated that in perinatally infected infants, HIV-1 R N A levels are high and decline more slowly than adults during the first two years o f life. Infants with very high viral loads are at increased risk for a rapid progression o f disease, which suggests that early treatment with antiretroviral agents may be indicated in the first few months o f live to reduce this risk. Vi ra l dynamics as they are now understood have significant implications in terms o f current therapeutic strategies. Unt i l recently, all antiviral therapy showed a similar response in viral load, which was not sustained, due to the potency of the agents and regimens being used. More recently, therapeutic regimens have been described which give rise to greater viral load suppression (277,278). If a treatment can be found which suppresses viral replication even more efficiently (perhaps completely), and sustains that suppression, it may lead to a long-term remission and the delay in immune disease progression, perhaps indefinitely. Although such a treatment does not yet exist, we all l ive in anticipation o f its development. 42 Antiretroviral Drug Resistance A n important feature o f retroviral reverse transcriptases is that they do not have the ability to edit errors in transcription which occur during nucleic acid replication. Based on newly understood HIV-1 viral dynamics, the high rate o f viral turnover in HIV-1 infection indicates that each single-point mutation may arise 10 4 to 10 5 times each day (301). The selective pressure exerted by antiviral therapy promotes the emergence o f strains o f virus which are resistant to the drug being used. Such resistant isolates have been found in clinical samples from patients receiving all o f the currently approved nucleoside analogue reverse transcriptase inhibitors. For several o f these agents, the emergence o f drug-resistant strains has been associated with loss o f antiretroviral activity and disease progression (302). There are several factors, including particular properties o f the virus, host, and drug, which combined determine the rapidity with which drug resistance develops, and thus the duration o f cl inical benefit for each individual. Faster development of resistance is associated with low C D 4 + T cell counts, more advanced HIV-1 disease and high plasma viral loads. The influence of the syncytium-inducing phenotype is currently the subject o f some debate (303). The development o f resistance is also dependent on the combination o f drugs which the patient is receiving. The pressure on the virus to develop resistance is proportional to the potency of the drug combination in the patient. Combination therapy is extremely important, because none o f the currently available drugs has the ability to completely suppress viral load when administered as a monotherapy. Thus, the goal o f current therapies is to find the correct combination of drugs which have the ability to achieve suppression o f virus replication (viral load) to a level which delays the emergence o f resistance. Zidovudine Resistance Z D V was the first nucleoside analog to be introduced to treat patients with HIV-1 infection. A s previously noted, it was shown to prolong survival, decrease the frequency and severity o f opportunistic infections, improve C D 4 + T lymphocyte counts and decrease serum HIV-1 p24 antigen concentration and the titre o f infectious H I V - 1 in plasma in patients with advanced HIV-1 infection (236,193,127). It was also shown to delay disease progression in patients with mildly symptomatic H I V - 1 infection and asymptomatic H I V - l -infected individuals with peripheral blood C D 4 + T lymphocyte counts < 5 0 0 x l 0 c cell//xl (244,245,312). However, most o f these beneficial effects were of limited duration (236,193,31 1). With in two years o f the 43 introduction o f Z D V , strains o f HIV-1 with 100-fold decreased susceptibility to this agent were isolated from patients who had been on prolonged therapy (215,313,3 14). In these patients, the median drug concentration required to reduce viral replication by 50% ( IC 5 0 ) increased from an average o f 0.003umol/l before therapy to 3.0umol/l after 18 months of treatment (298) (phenotypic resistance defined as ICSo> 1 umol/1). ZDV-resistant HIV-1 can be detected as early as 6 months after the beginning o f therapy and about 50% of patients have resistant isolates after 2 years (298). There were originally four mutations identified in the HIV-1 reverse transcriptase sequence which were associated with Z D V resistance. These were: D 6 7 N , K 7 0 R , T 2 1 5 Y / F , and K 2 1 9 Q (215). A fifth mutation associated with Z D V resistance, M 4 1 L , was subsequently described (304). The order in which these mutations appeared was also subsequently elucidated. In the typical patient, there is a transient appearance of the K.70R mutation; its disappearance is paralleled by the appearance o f the T215 Y / F mutation. After prolonged therapy the K 7 0 R mutation may reappear in parallel to the D 6 7 N mutation (305). Over time, high-grade phenotypic resistance develops with all five mutations being present i f Z D V therapy is continued. The most reliable and durable marker o f disease progression is the presence of the mutation at codon 215. The presence o f Z D V resistant mutations in patient isolates are associated with C D 4 + T cell decline and clinical failure on Z D V monotherapy, and decreased clinical responses to subsequent therapy with other nucleoside-containing drugs or drug combinations (306,307). In the A I D S Cl in ica l Trials Group ( A C T G ) 116B/117 trial, high level phenotypic resistance to Z D V at baseline was an independent prediction of disease progression or death, with a relative hazard ratio of 1.93 (213). The use of Z D V is now so widespread that transmission o f already Z D V resistant virus has occurred by sexual, percutaneous (in a healthcare worker) and maternal-infant routes (308-310). Transmission o f such an isolate from one child to another has also been documented (319). Few studies have been published on Z D V resistance in children. In 1993, a study o f 19 children showed that individuals with Z D V resistant strains had worse clinical outcomes than children whose viruses remained susceptible, as determined by a 50% decline in absolute C D 4 + T cell counts after one year o f treatment, failure to thrive, or death. There is an urgent need to confirm these findings in larger, more varied pediatric patient populations. 44 Resistance Testing Methodologies Antiretroviral drug susceptibility assays are divided into two groups, genotypic and phenotypic. The first involves isolating R N A or D N A from biological materials to determine the presence or absence of specific mutations associated with drug resistance (215-218). The second involves growing HIV-1 in vitro and measuring its replication characteristics in the presence o f a given drug. In a prototypic method, H I V - l -infected peripheral b lood mononuclear cells ( P B M C s ) , with seronegative phytohemagglutinin-stimulated donor P B M C s to obtain an H I V - 1 stock are incubated. The virus stock is then titrated for viral infectivity (50% tissue culture infective dose) by use of serial fourfold virus dilutions in donor P B M C s . A standardized inoculum of 4,000 50% tissue culture infective doses per 10 c cells is used in a second step of the procedure to acutely infect seronegative donor P B M C s in a 7-day microtitre plate assay with triplicate wells containing Z D V concentrations ranging from 0 to 5.0 u M . The 50% inhibitory concentration ( I C 5 0 of Z D V ) is then determined (212-214,124,127). A n isolate is deemed sensitive i f the I C 5 0 is < 0.2 u M and deemed highly resistant i f the 1 C 5 0 is >1.0 u M . A third method is a variation o f a quantitative micrococulture outlined in the A C T G Vi ro logy Manual for HIV-1 Laboratories (124). The assay is less expensive and time consuming than the standard phenotypic resistance assay. The infected P B M C s from patients are isolated and directly cultured with P H A -stimulated negative donor P B M C s (322). Six 5-fold serial dilutions o f P B M C s were prepared in duplicate and to one duplicate was added Z D V to a final concentration o f 5 u M . The cultures are maintained for 14 days, with new drug and P B M C s added on day 7. P24 antigen levels in the supernatants are then determined. A reduction in viral replication in the row of wells containing Z D V compared to the parallel row o f drug-free control wells was taken as a crude measure o f Z D V susceptibility. This assay can be adapted to test susceptibility to other antiretrovirals, al lowing for the direct evaluation o f drug susceptibility over 14 days, minimizing the selective effects o f in vitro passage that may be problematic in the macro-culture methodology. The use of both genotypic and phenotypic drug resistance testing may provide us with complimentary information and it would be useful to evaluate both methods in parallel to establish the usefulness o f the information they generate. 45 Research Hypothesis Many clinical and laboratory measures exist which produce information to al low physicians and care givers to have a better understanding o f an HIV- l - in fec ted chi ld 's disease status, and their risk o f progressing to A I D S . These measures include viral load testing, drug resistance testing, monitoring o f C D 4 + T cell levels, viral phenotype testing (as some phenotypes grow more rapidly and aggressively than others (296, 297, 303, 321, 325)), along with regular evaluation for cl inical disease progression. There are many practical reasons why it is impossible to perform every test on every patient at every point in time. Therefore, it is imperative, to make the best use of time and resources, that the most accurate and informative tests (singly or in combination) be found to monitor the disease state o f children infected with H I V - 1 . In this study we hypothesized that an algorithm comprising a group o f cl inical and/or laboratory tests exist which would be the best markers by which to assess a pediatric HIV-1 patient's current disease status as well as the risk of disease progression over time. Specific Objectives 1. To determine the prevalence o f resistance to Z D V in a cross-section o f HIV- l - in fec t ed Canadian children and to compare different resistance testing methodologies 2. To determine the association o f drug resistance with clinical and laboratory variables including: a. Cell-associated viral load b. Plasma viral load c. V i r a l phenotype d. C D 4 + T cell count and C D 4 + T cell percent e. Presence or absence o f HIV-1-related symptoms f. Presence or absence o f HIV-1 -related encephalopathy g. Mode of acquisition o f H I V - 1 46 MATERIALS AND METHODS Patient Population A total o f 103 patients were recruited from seven centres across Canada: Izaac Walter K i l l a n Hospital in Halifax, St. Joseph's Health Centre in London, St. Justine's Hospital in Montreal , The Children 's Hospital in Montreal, The Chi ldren 's Hospital o f Eastern Ontario ( C H E O ) in Ottawa, The Hospital for Sick Children in Toronto, and British Columbia Children's Hospital in Vancouver. A l l patients were know to be infected with H I V - 1 and between the ages of 3 months and 18 years o f age at the time of enrollment. Specimen Collection Approximately 10.0ml o f anticoagulated whole blood was collected in heparinized tubes from each patient at their respective clinics. The samples were then packed and transported in sa f -T-paks® by either air or ground transport according to the Canadian Transportation o f Dangerous Goods Regulations ( T D G Regulations) to our laboratory within 24 hours o f the sample being drawn. The samples from C H E O were processed on the day the specimen was drawn, within six hours o f sample collection. Isolation of Lymphocytes Isolation of lymphocytes was performed in a B S L - 3 containment facility in a laminar flow hood (18,19,124). Upon receipt o f a specimen, the tube was inverted eight times and centrifuged at 400 x g for 10 minutes. A total o f 2.0-3.0ml plasma was removed and placed into 1.5ml screw cap tubes in 1.0ml aliquots and frozen at -70°C to be used later for plasma viremia measurements. The volume of removed plasma was then replaced with Phosphate Buffered Saline (PBS) at room temperature and the contents o f the tube was mixed. The blood was then layered slowly onto F ico l l -Hypaque® solution (Pharmacia, Uppsala, Sweden or Piscataway, N e w Jersey, U S A ) (blood volume to F ico l l volume ratio 4:3) in a 15 or 50ml polystyrene centrifuge tube and centrifuged at 400 x g for 30 minutes at 2 2 ° C without a brake. The remaining supernatant was removed and discarded and the layer o f peripheral blood mononuclear cells ( P B M C s ) was then carefully removed, then mixed with 40.0ml l x P B S at room temperature and centrifuged at 400 x g for 15 minutes. The P B S was removed and the pellet o f cells was then resuspended 47 in 10.0ml P B S , 50ul o f the suspension was removed for the enumeration o f cells (by Trypan Blue dye exclusion) and the remainder was centrifuged at 60-100 x g for ten minutes. After the spin, the supernatant was discarded and the cells were resuspended in l x P B S or R-3 medium [RPMI-1640 ( G I B C O Laboratories, Grand Island, New York , U S A ) with 10% heat-inactivated fetal bovine serum; 2 m M L-glutamine; 5% interleukin-2 (IL-2; Boehringer Mannheim, Brussels, Belgium); penicillin (100units/ml)/streptomycin (lOOug/ml); l O m M H E P E S Buffer] to a concentration of 1.0-3.0x10 6cells/ml, depending on the number of cells obtained. For storage, a minimum of 1 .OxlO 6 cells were saved in 1.5ml screw cap tubes (1.0-3.0x10 6cells/tube). The tubes were centrifuged, the supernatant removed, and the remaining pellets were stored at -70°C to be used at a later date for proviral load measurement or sequencing. Qualitative Macrococulture Assay From the P B M C s isolated, a separate aliquot o f 6.0x10 6cells was removed after the final centrifugation above and transferred to a fresh 15ml polystyrene centrifuge tube. These patient cells were used to prepare a qualitative P B M C macrococulture assay, according to the A I D S Cl in ica l Trials Group ( A C T G ) Vi ro logy Manual, September 1994 edition (124,207,208). The concentration o f the cells was adjusted to 2.0x10 6cells/ml in R-3 medium. One to three day old, HIV-1-negative, PHA-st imulated donor P B M C s were sedimented in a 15ml polystyrene centrifuge tube at 400 x g for 10 minutes at 22°C. The cells were then resuspended in R-3 medium and enumerated by Trypan Blue dye exclusion. The cells were adjusted to a concentration of 2.0x10 6 cells/ml. 6 .OxlO 6 donor P B M C s were required for each sample (3.0ml cells in R-3 medium at 2.0x10 6 cells/ml). The donor cells were combined with the 6.0x10 6 patient cells (3.0ml cells in R-3 medium at 2.0x10 ( >cells/ml) in a 25cm 3 flask (total volume: 6.0ml). The flask was then incubated on its side in a humidified chamber at 5% C 0 2 and 37°C for 21 days. On days 3, 10, and 17, the flask was removed from the incubator, examined under the light microscope for fungal contamination and set upright in a laminar flow hood for 20 minutes. Approximately 3.0ml o f supernatant was removed from the top of the culture and frozen in 1.0ml aliquots in 1.5ml screw cap tubes at -70°C. The volume of supernatant removed was replaced with fresh R-3 medium. The suspension was mixed and the flask was then returned to the incubator. 48 On days 7 and 14, the flasks were removed from the incubator, examined under the light microscope and set upright in a laminar flow hood for 20 minutes. Approximately 3.0ml o f supernatant was removed from the top of the culture and again frozen in 1.0ml aliquots in 1.5ml screw cap tubes at -70°C. The volume of supernatant removed was replace with 1-3 day old, HIV-1-negative, PHA-st imulated donor P B M C s which had been centrifuged at 400 x g for 10 minutes at 22°C, enumerated, and resuspended in R-3 medium to 2x l0 6 ce l l s /ml . On day 21 the culture was terminated. Aga in , 3.0ml o f supernatant was removed and frozen as above. A l l o f the collected frozen supernatants were later thawed and H I V - 1 p24 antigen levels were measured using commercial p24 assay kits (Organon Teknika B V , Boxtel , The Netherlands or Coulter, Hialeah, Florida, U S A ) , as a measure o f ongoing viral replication. From these stocks, viral phenotype assays and conventional HIV-1 drug susceptibility assays were performed, as described below. Quantitative Micrococulture Assay A final cellular aliquot (minimum of 2.7x10 C isolated patient P B M C s ) was then used to set up a quantitative P B M C micrococulture assay (124). The isolated P B M C s from each sample were enumerated and diluted to a concentration o f 1.0xl0 6 cells/ml in a sterile 15ml conical centrifuge tube, which was labeled tube A . To each o f five additional tubes, labeled B through F, was added 2.4ml R-3 medium. From tube A , 0.6ml o f suspended patient P B M C s were removed, transferred to tube B , and mixed. A fresh pipette was then used to remove 0.6ml of cell suspension from tube B and to transfer it to tube C. After thorough mixing, the procedure was repeated for the next four dilutions. The final dilutions in tubes A through F contained 1x10°, 2 x l 0 5 , 4 x l 0 4 , 8 x l 0 3 , 1.6x103, and 320 patient P B M C s / m l , respectively. Then, 1-3 day old, HIV-1-negative, PHA-st imulated donor P B M C s were then centrifuged at 400 x g for 10 minutes at 20°C , resuspended in R-3 medium, enumerated and diluted to a concentration o f 1.25xl0 6 cells/ml. 0.8 ml o f the donor P B M C suspension was then pipetted into six wells o f two consecutive rows of a Costar (Cambridge, Massachusetts, U S A ) 24-well micrococulture plate. Thereafter, 1.0ml of patient cells from the 15ml tubes marked A through F was added to the corresponding wells o f the micrococulture plate (see figure 4), with duplicate cultures established for each patient cell concentration. 0.2ml of R-3 medium was then added to the top row of wells. The culture was then monitored as described 49 above, with the viral load being calculated based on the highest dilution (i.e. least number o f patient cells) yielding a positive result. This result was expressed as infectious units per mi l l ion cells (IU/10 6 cells). Rapid Quantitative Culture-Based Zidovudine Resistance Assay The above described quantitative micrococultures which were set up from each patient specimen formed the basis o f our novel assay for the determination of viral susceptibility. Our resistance assay was thus based on the A C T G Quantitative Micrococulture Assay with some modifications (124,219). To the bottom row of wells in the micrococulture plate containing patient and donor P B M C s was added 0.2ml o f 5 0 u M zidovudine (Glaxo Wellcome, Research Triangle Park, North Carolina, U S A ) . The final concentration o f donor P B M C s was l x l 0 6 c e l l s / m l and the final concentration of Z D V was 5 u M . The plate was then covered, taped around the sides with masking tape and placed in a 5% C 0 2 humidified incubator at 37 °C. Figure 4: 24-Wel l Plate Format for '. P B S P B S P B S P B S P B S P B S cont. A cont. B cont. C cont. D cont. E cont. F 5 u M A 5 u M . B 5 u M C 5 u M D 5 u M E 5 u M F P B S P B S P B S P B S P B S P B S On day 7, 1,0ml o f the culture medium was removed from the surface o f each well in both the control and ZDV-conta in ing rows and stored at -70°C in sterile 1.5ml screw cap tubes. The removed supernatant was then replaced with 0.8ml of fresh R-3 medium containing 6.25x10 5 , 1-3 day old seronegative, P H A -stimulated donor P B M C s . To each well o f the top row was then added 0.2ml R-3 medium and to each well o f the second row was added 0.2ml 5 0 u M Z D V . The final concentration of added P B M C s was 0.5x 10 6cells/ml and the final concentration of drug in the second row of wells was 5 u M . The assay was terminated on day 14. Aga in , from each wel l was removed 1.0ml o f supernatant which was frozen at -70°C. V i r a l replication was evaluated by measuring viral p24 antigen levels (Organon Teknika or Coulter) in culture supernatants saved on days 7 and 14. A reduction in viral replication (and thus viral 50 titre) in the row of wells containing Z D V compared to the parallel row o f drug-free control wells was taken as a crude measure of drug susceptibility. Standard HIV-1 Drug Susceptibility Assay Using the viral stocks generated from the qualitative P B M C macrococulture assay, a standard HIV-1 drug susceptibility assay was performed on selected patient samples according to standardized protocols (127,212-214) to confirm the results obtained using our novel quantitative culture-based Z D V resistance assay. Prior to performing the actual susceptibility assay, the infectious titre o f each patient isolate was determined. This was accomplished by titrating the virus-containing supernatant stock in a 96 well flat-bottom tissue culture plate. Seven serial four-fold dilutions, ranging from 1:16 through 1:65,536 were performed in triplicate for each patient. For this purpose, 1-3 day old, HIV-1-negative, PHA-stimulated donor P B M C s were sedimented at 400 x g for 10 minutes at 22°C. The supernatant was removed and discarded. The P B M C s were resuspended in R-3 medium, and enumerated. The suspension was adjusted to 4x l0 6 ce l l s /ml and returned to the incubator until needed. Then, 200ul o f l x P B S was added to all wells in rows A , B , F, G , and H and to the wells in columns 1, 2, and 10 to 12 in rows C to E with a multichannel pipettor (see Figure 5). 150j.il o f R-3 medium was then added to each wel l o f columns 4 to 9 in rows C to E with a multichannel pipettor. The virus stock to be tested was then rapidly thawed in a water bath at 37°C until only a small crystal o f ice remained. The sample was then diluted 1:12 in R-3 medium (0.1ml of virus stock was added to 1.1ml R-3 medium) in a fresh sterile 1.5ml screw cap tube, and 200ul o f diluted stock was added to each wel l o f column 3 in rows C to E . Tojdilute, 50uf was then transferred from column 3 to column 4 in rows C to E , using a multichannel pipettor. The pipette tips were then changed and the contents o f column 4 were mixed, and 50ul o f sample from column 4 was then transferred to column 5. A similar procedure was employed for column 6-9, and 50ul o f solution was removed from column 9 after mix ing and was discarded. Final ly, 50ui o f donor P B M C s (at 4x l0 6 ce l l s /ml ) were added to the wells o f columns 3 to 9, rows C to E , moving from right to left. The plate was then covered, taped around the sides with masking tape, and incubated at 37°C and 5% C 0 2 in a humidified chamber. The final dilutions o f the original viral stock in columns 3 to 9 were: 4"2, 4"3, 4"4, 4"5, 4"6, 4"7, and 4"8 cells/ml as shown in Figure 5. 51 Figure 5: 96-Wel l Plate Format for HIV-1 V i r a l Stock Titration 1 2 3 4 5 6 7 8 9 10 11 12 A P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S B P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S C P B S P B S 4"2 4"3 4"4 4"5 4"6 4"7 4"8 P B S P B S P B S D P B S P B S 4"2 4"3 4"4 4"5 4- 6 4"7 4"8 P B S P B S P B S E P B S P B S 4-2 4-3 4"4 4"5 4-« 4' 7 4"8 P B S P B S P B S F P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S G P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S H P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S On day 4, the cells in the wells in rows C to E were resuspended with a multichannel pipettor. M o v i n g from right to left across the plate, the cells were resuspended and 125y.il o f suspension was removed and discarded, and 150ul o f fresh R-3 medium was added back to each well o f rows C to E , again moving from right to left. The plate was then recovered, retaped, and returned to the incubator. On day 7 the HIV-1 titration assay was terminated and the supernatants were tested for HIV-1 p24 antigen levels according to the manufacturer's instructions using the Organon Teknika or Coulter assay kits. This was accomplished by transferring lOOul o f supernatant from the titration plate to the wells o f an H I V - 1 p24 antigen plate which contained 100f.il o f R-3 medium and 20ul o f the manufacturer's disruption buffer containing Triton X - 1 0 0 . The p24 antigen assay was then performed and a well was scored positive i f the HIV-1 p24 antigen level was >50pg/ml. The 50% tissue culture infectious dose ( T C I D 5 0 ) was then calculated using the Spearman-Karber method. Once the T C I D 5 0 was determined, the standard antiretroviral drug susceptibility testing assay for Z D V was performed. A series o f 2 X working solutions of Z D V were prepared by diluting 50ul o f 1 m M stock Z D V in 4.95ml R-3 medium to obtain a concentration of l O u M Z D V . The l O u M stock was then further diluted to 2 u M Z D V by adding 1,0ml o f l O u M Z D V to 4.0ml o f R-3 medium. Further dilutions were made in a similar fashion to yield 2 X stock solutions of Z D V at concentrations of lO .OuM, 2 . 0 u M , 0 . 2 u M , 0 .02uM, and 0 .002uM. The Z D V susceptibility assay was performed in a sterile flat-bottomed 96-wel l plate (see figure 6). 200ul of l x P B S was added to all wells in rows A , B , G , and H with a multi-channel pipettor and in a similar way to the wells o f columns 1, 2, and 9 through 12 in rows C to F. Then, lOOul o f R-3 medium was added to 52 the wells in rows C through F in column 3 and lOOul o f 2 X Z D V working solution in ascending concentration was added to the wells in rows C through F in columns 4 to 8. Thereafter, 1-3 day old, HIV-1-negative, PHA-stimulated P B M C s were sedimented at 400 x g for 10 minutes at 22°C. The supernatant was removed and the cells resuspended in R-3 medium and enumerated. The cells were adjusted to a concentration of 4x l0 6 ce l l s /ml . To each o f two sterile 15ml conical centrifuge tubes was added 1.0ml of the donor cell suspension, and 1.0ml of R-3 medium was added to one o f the tubes, bringing the final concentration o f cells to 2xl0 f , cel ls /ml . The cap was secured and the tube placed in a humidified chamber at 37°C and 5% CG"2- The cells in the other tube were sedimented at 400 x g for 10 minutes and the supernatant was removed. The chosen viral stock (titrated previously) was then rapidly thawed in a water bath at 37°C until only a small crystal o f ice remained. The required amount o f stock was then added to the tube containing the sedimented P B M C s to make the final concentration o f virus 4000 T C I D 5 0 / m l (final volume was kept to <1.0ml). The suspension was mixed gently and incubated at 37°C in the humidified chamber for one hour. Fol lowing the incubation, the donor cells containing the virus stock were diluted to a final volume o f 2.0ml in R-3 medium. 100f.il o f the infected cells were then added to each well o f columns 3 through 8 in rows C to E, and 100f.il o f the uninfected donor P B M C s in the conical tube from the incubator were added to each well o f columns 3 through 8 in row F. The final volume in each well was 200f.il and the final concentration o f Z D V in each wel l in columns 3 through 8 were 0, 0.001, 0.01, 0.1, 1.0, and 5.0uJVl Z D V as shown in Figure 6. The plate was then sealed and incubated at 37°C and 5% C 0 2 in a humidified chamber. Figure 6: 96-Wel l Plate Format for Z D V 1 2 3 4 5 6 7 8 9 10 11 12 A P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S B P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S C P B S P B S 0 0.001 0.01 0.1 1.0 5.0 P B S P B S P B S P B S D P B S P B S 0 0.001 0.01 0.1 1.0 5.0 P B S P B S P B S P B S E P B S P B S 0 0.001 0.01 0.1 1.0 5.0 P B S P B S P B S P B S F P B S P B S 0 0.001 0.01 0.1 1.0 5.0 P B S P B S P B S P B S G P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S H P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S 53 On day 4, the plate was removed from the incubator and examined under a microscope for any H I V - 1 -induced cytopathic effect. 2 X Z D V working solutions were made up as on the day the experiment was set up, and 0.5ml o f the working solutions were added to 0.5ml R-3 medium to give I X working solutions. M o v i n g from left to right across the plate, the cell suspension in the wells in rows C to F and columns 3 to 8 was mixed and 125ul was removed and discarded. 150ul o f the I X Z D V working solutions were then added back to the appropriate wells. The plate was then sealed and replaced in the humidified 37°C incubator. On day 7, the H I V - 1 drug susceptibility assay was terminated and an HIV-1 p24 antigen evaluation was performed on the culture supernatants using the Organon Teknika p24 antigen assay kit. Dilutions were prepared for the wells in rows C to E , columns 3 to 8 o f the susceptibility assay plate as follows: a. T o the first six wells o f rows C to H in a 96-well flat-bottomed polystyrene plate (dilution plate), 2 0 5 o f R-3 medium was added followed by 25ul o f lysis buffer, using a multichannel pipettor. b. 20ul o f supernatant from each well in rows C , D and E in columns 3 to 8 o f the drug susceptibility assay were then transferred to their respective rows in the dilution plate using an 8-channel pipettor (1:12.5 dilution). 20ul o f cell suspension was then transferred from row C to row F, row D to row G , and row E to row H in the dilution plate fol lowing mixing using the 8-channel pipettor (1:156 dilution). The plate was then placed in a low density polyethylene bag to prevent drying of the sample by sublimation. The plate was then stored at -70°C and the H I V - 1 p24 antigen assay was run within 72 hours. Prior to the determination o f the p24 antigen levels, the dilution plate was thawed at ambient temperature, and 20ul o f solution from each wel l in rows G to H was transferred to the appropriate wells o f the HIV-1 p24 antigen assay kit plate which already contained 180ul o f R-3 medium (1:1560 final dilution). For some samples, a range o f dilutions had to be tested until the untreated control cultures could be accurately quantified on the linear portion o f the calibration curve o f the assay. The 50% inhibitory concentration (IC50) w a s then calculated, as described in the above titration experiments. A n I C 5 u value of <0 .2uM was 54 considered to be an indication that the patient's virus was sensitive to Z D V (327) and an I C 5 0 value o f >1 .OuM was considered to be an indication that the patient's virus was resistant to Z D V . HIV-1 Viral Phenotype Assay (HIV-1 Syncytium-Inducing Assay) This in vitro assay using M T - 2 cells was used to detect syncytium-inducing (SI) variants o f HIV-1 (124,209,317). The assay was performed in duplicate (for each viral isolate) in 96 wel l flat-bottomed tissue culture plates using the cell-free viral stocks generated from the previously grown qualitative macrococultures and quantitative micrococultures, frozen at the time o f harvesting and not thawed until needed for this assay. A l l stocks used were previously determined to be positive for H I V - 1 using the Organon Teknika p24 antigen detection kit. Prior to their use in this assay, M T - 2 cells were maintained in culture at a concentration o f l -2x l0°ce l l s /ml in R-10 medium [ R P M I 1640, 10% heat inactivated Fetal Bovine Serum, penici l l in (100units/ml)/streptomycin (100ng/ml), 2 m M L-glutamine, l O m M Hepes Buffer], The cultures were split 1:10 every 3-4 days. On day 0 o f the assay, the M T - 2 cells were harvested and centrifuged at 400 x g for 10 minutes. The supernatant was removed and the cells were enumerated. The cell concentration was then adjusted to 3 .4xl0 5 ce l l s /ml in R-10 medium. The configuration of the test plate is shown in figure 7, with 200(.il o f l x P B S added to all the wells in rows A , D, and H , and to the wells in columns 1, 3 through 10, and 12 in row G , to the wells in columns 1 and 12 in rows B , C , and F, and to the wells in 1,2, 11, and 12 in row E . Thereafter, 15OJLLI o f the M T - 2 cell suspension was added to each duplicate sample well (numbered wells in figure 7), and to the positive and negative control wells (POS and N E G wells in figure 7), and 50ul o f HIV- l - in fec ted cell-free supernatant (to be tested) was added to the duplicate sample wells, al lowing for the evaluation o f 16 patient samples/plate. The positive control consisted o f 50ul o f a known SI virus stock while the negative control was 50f.il o f R-10 culture medium. 55 Figure 7: 96-Wel l Plate Format for the M T - 2 H I V - 1 Vi ra l Phenotype Assay 1 2 3 4 5 6 7 8 9 10 11 12 A P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S B P B S N E G 1 • 2 3 4 5 6 7 8 N E G P B S C P B S P O S 1 2 3 4 5 6 7 8 P O S P B S D P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S E P B S P B S 9 10 11 12 13 14 15 16 P B S P B S F P B S N E G 9 10 11 12 13 14 15 16 N E G P B S G P B S P O S P B S P B S P B S P B S P B S P B S P B S P B S POS P B S H P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S P B S On days 3, 6, 9, and 12, each sample well was examined for syncytium formation. When syncytia were observed, the day o f positivity was noted. Fo l lowing examination o f the wells, the cells in each well were resuspended and 130ul o f cell suspension was removed and replaced with 150|il o f R-10 medium. O n day 14 the assay was terminated. If no syncytia were observed on day 14, the isolates in question were deemed to be nonsyncytium-inducing (NSI). Plasma Viral Load Assay ( N A S B A ® ) The N A S B A Assay by Organon Teknika was used to detect plasma viral R N A levels in each patient . (210,211,287). The technique is used to amplify specific R N A targets quickly using small volumes of source material o f different types. The basic principles o f the assay involve the isolation and purification of R N A from the plasma sample to be evaluated, the amplification of specific target sequences in the H I V -1 genome and the detection o f the amplified product using chemiluminescent probes. The quantity of amplified product is measured by the quantitative reading system and calculations based on the relative amounts o f the wild-type R N A and three internal standards determines the original amount o f R N A in the sample. The substrate for the assay was a plasma sample taken from each patient, removed at the time o f P B M C isolation and immediately frozen at -70°C in 1.0ml aliquots. The procedure itself has three technical components (as shown in Figures 8-11) (336): A . Nucleic A c i d Isolation B . Amplif icat ion C. Hybridization/Detection 56 A. Nucleic Acid Isolation The isolation procedure is based on the binding of nucleic acid to silica particles. These silica particles remain bound to the nucleic acid through a series o f washing steps. The nucleic acid is then liberated from the silica by a small volume o f elution buffer. This is known as the " B o o m " procedure (328). The guanidine thiocyanate ( G u S C N ) lysis buffer tubes, the tubes containing patient plasma samples, and the G u S C N wash buffer, were thawed at 37°C in an incubator. The elution buffer (1 m M Tris at pH 8.5) was thawed at room temperature and the Q-sphere (containing three R N A internal standards ( Q A (10°), Q B (10 5), and Q c (10 4 copies))) and si l ica were stored at 4 ° C until needed. The R N A internal standards differ from the sample (wild-type) R N A and from each other by a 20-nucleotide randomized sequence with the same nucleotide composition. The standards and sample are together in the same tube, and this design o f internal standards results in equal efficiency o f isolation and amplification. When warmed, the lysis buffer tubes were vortexed and then "quick spun" for ten seconds to remove any liquid from the cap. In a laminar flow hood, the plasma samples were vortexed and then 100|il o f plasma was added to each lysis buffer tube. The lysis buffer tubes were then vortexed, quick spun and left to sit at room temperature and the tubes with the remaining plasma were returned to the -70°C freezer. In a separate clean (free o f viral R N A ) laminar flow hood, the Q-sphere was reconstituted with 220ul o f elution buffer and vortexed, and 25j_il o f the suspension was added to 225\x\ o f elution buffer in a separate 1.5ml screw cap tube and vortexed. O n a bench top in the isolation area, 20ul o f calibrator was added to each lysis buffer tube, which was then vortexed and quick spun. The silica was then vortexed and 50p.l was added to each lysis buffer tube. The silica has a high density and thus had to be vortexed after every second tube. The lysis buffer tubes were then left for 10 minutes at room temperature and were vortexed every two minutes. During the 10 minutes, two 50ml conical polystyrene tubes, one containing ~30 ml 70% ethanol and the other containing ~12ml acetone were prepared. The lysis buffer tubes were then quick spun and the lysis buffer was removed. lOOOul o f wash buffer was then added to each tube. After a vortex and a quick spin, the wash was removed and another lOOOul o f wash buffer was added to the tubes. The vortex and quick spin were repeated and the washes were repeated (two wash buffer washes, two 70% ethanol washes and one acetone wash total). Fol lowing the final wash with acetone, the tubes were spun 57 for exactly ten seconds to pellet the silica. A l l the acetone was removed and the tubes were placed in a block heater at 56°C without their caps on, but covered with a paper towel, to dry. The caps were then replaced and 50ul o f elution buffer was added to each tube. After vortexing, the tubes were placed in the heating block at 56°C with their caps on for ten minutes, vortexing once again after 5 minutes. During the incubation time, ten new 1,5ml screw cap tubes were obtained and labeled for each sample. Fol lowing the incubation, the lysis buffer tubes were spun for 2 minutes at maximum speed in the microcentrifuge, and 5f.il o f supernatant was removed and transferred to the fresh screw cap tubes. The lysis buffer tubes were then stored at -70°C and the 5p.l samples o f purified viral R N A were frozen at -70°C until they were needed for the amplification step of the procedure. 58 Fig. 8: Schematic presentation Nucleic Acid Isolation process Sample + Lysis buffer 1 Nucleic Acid Proteins Lipids Purified Nucleic Acid B. Amplification The amplification reaction is based on the simultaneous action of three enzymes: Av ian Myelobastosis Virus reverse transcriptase (RT) , Escherichia C o l i RNase H and Phage T 7 R N A polymerase. The steps of the reaction take place within one reaction tube at a constant temperature (41°C) . Basically, the amplification process begins with the annealing of Primer 1 (P| (5 '-59 AATTCTAATACGACTCACTATAGGGA7TGCCTCTCTGCATCATTA-3')) (337), which contains a 3 ' terminal sequence which is complementary to the target sequence and a 5' terminal sequence which is recognized by T 7 R N A polymerase (in italics). The target sequence is a highly conserved region o f the gag gene in the HIV-1 genome. The R T then extends P, to synthesize a copy o f c D N A . R N A s e H then degrades the R N A strand o f the resulting R N A : D N A hybrid. This allows primer 2 (P 2 (5 ' -A G C A T T G G G A C C A G C G G C T A - 3 ' ) ) (337) to anneal to the single-stranded c D N A . Fo l lowing annealing, P 2 is extended with R T to form a double-stranded D N A containing a double stranded T 7 promoter. The T 7 R N A polymerase recognizes the promoter and generates multiple copies o f a single-stranded R N A product. Each copy serves as a template for a repetition of the process described above, however, the annealing o f primers is reversed due to the anti-sense R N A . The amplification reaction occurs over a 90 minute time period at a constant temperature, resulting in the accumulation of single-stranded R N A target. The procedure was begun in the clean laminar flow hood. The enzyme mix, containing the three enzymes and primer diluent (dimethylsulfoxide ( D M S O ) in water) were thawed. First, 45ul o f enzyme diluent was added to the enzyme tube, which was shaken lightly. It was then set aside at room temperature for at least 15 minutes and no longer than 45 minutes before use. Light shaking was applied during this time to ensure adequate resuspension o f the enzyme. Fol lowing this, 120ul o f primer diluent was added to the primer tube, which was then vortexed vigorously to ensure complete resuspension o f the primers. O n a clean bench top, lOu l o f primer mix was then added to each screw cap tube containing the 5ul aliquot, which had previously been removed from the -70°C freezer and thawed. The tubes were then placed in a 65°C heating block for 6 minutes, then transferred to a 41°C heating block for 6 minutes, fol lowing which, 5uJ of enzyme solution was added to each tube. One tube was removed from the heating block at a time. Fol lowing the addition o f the enzyme, the tube was flicked several times to ensure that the mixture was well homogenized. The tube was then replaced in the heating block. After the enzyme was added to the last tube, the tubes were then heated for an additional 5 minutes. Fo l lowing the incubation, the tubes were transferred to a 41°C water bath away from the clean area and away from the isolation area and incubated for 90 minutes. Fol lowing this amplification step, the tubes were frozen at -70°C and transported to the hybridization/detection area. 60 Fig . 9: Amplif icat ion ssRNA 5' 5'-3-Primer 1 :|". dsDNA 5' 100 "C, Ice Primer 1 I 3" RT 5' 5'-3'-3" •3' 5' 5" RT 5'-100 C, Ice • 3" I 5" RNase H Primer 2 3' 5' 3'- •5" 5" 3" 5"-^3' RT 5" T7 RNA pol 5" 5" | Primer 2 5'—3' 5" RT 5' 3 3 T ^ w w 5 " 3 1 / V V V W | Pnmer2 * ° 3 - / W T W V 5 1 I \ 3 ' 5" T7 RNA pol T7 RNA pol 5'—3' RT 5 ' - 3' 3'fVwwj, Amplification Phase 5'_. 3' •-• :3" 5" RT RNase H 5" 3' 5 ' - 3' Primer 1 3" 5' C. Hybridization/Detection The quantitation o f the amplified products is performed using the N A S B A Quantitative Reading System ( N A S B A Q R System). It is an automated system which utilizes electrochemiluminescence technology ( E C L ) . This means that a chemical reaction takes place in an electromagnetic field and a light emission is produced. The hybridization format consists o f a magnetic bead-oligo which is coated with streptavidin. 61 To it is bound a biotin-oligo which binds to a specific sequence on the target R N A . In addition, four ruthenium (Ru) labeled probes are used, one for each internal standard and one for the sample R N A . The probes bind to the R N A and are part o f the chemical reaction that takes place during the detection. In a dead air box, the hybridization mixtures were prepared according to the manufacturer's instructions. In the hybridization room, 41 culture tubes were obtained and arranged into three rows o f ten tubes and one row of eleven tubes. Wi th a repeat pipettor, using a 0.5 ml tip, 20ul o f w i ld type ( W T ) hybridization mixture was added to each tube in the row with 11 tubes. To each o f the first row o f 10 tubes was added 20ul o f Qa hybridization mixture. The additions were repeated for the next ten tubes with the Qb mixture and again with the Qc mixture in the last row of tubes. Qa, Qb, and Qc differ from each other only by a small, randomized amplification fragment and each hybridization mixture contains probes specific for one internal standard. Ten eppendorf tubes were then obtained and labeled with the same identifier as the sample amplification tubes. To each tube was added lOOul o f detection diluent, and 5ul o f amplified product. The tubes were then vortexed and quick spun. For each sample, 5ul o f diluted product was added to the first column of four culture tubes containing the W T , Qa, Qb, and Qc mixtures, 5ul o f detection diluent was added to the Assay Negative Tube (the 41st tube). The tubes were then covered with adhesive tape and shaken lightly, and placed in a water bath at 4 1 ° C for 30 minutes, shaken every ten minutes. 62 Fig. 10: Schematic Representation of the Hybridization Format Bead oligo Streptavidin Biotin-oligo Magnetic particle Probe oligo Diluted amplificate ^yV\A/ v /V\AA^ 30' 41 °C ^A/VVWVVV^ NASBA QR System Fol lowing the hybridization, 300| i l o f Assay solution (containing 0 . 1 M free Tripropylamine (TPA)) was added to each tube and to a separate clean culture tube using a repeat pipettor and a 2.5ml syringe. The detection carousel was then loaded with the culture tubes; beginning with the reference tube, the assay 63 negative tube and then the sample tubes, and placed into the detection machine for reading. A t this point, each sample tube contains a buffer solution with free T P A and R N A product l inked to a magnetic bead and to a ruthenium-labeled probe. Figure 11 illustrates the chemical reaction which then occurs in the quantitative reading system. Upon activation o f the magnet, the magnetic beads are pulled against an anode by the magnetic field. A specific voltage is applied at the anode, which initiates an oxidation reaction of both the T P A and the Ru. Both molecules release an electron and become positively charged. Upon oxidation, the T P A releases a hydrogen atom (deprotonation) and becomes an unstable, highly reducing intermediate. A reaction then takes place between the T P A intermediate and the oxidized Ru . The Ru is elevated to an excited and unstable state. When the excited Ru falls back to its base state, a light photon is emitted at 620 nm. The light emission is then measured by a photo multiplier tube ( P M T ) . The amount o f light emitted is proportional to the amount o f product in each tube. Calculations based on the relative amounts o f the four products reveal the original amount o f wild-type R N A in the sample. Plasma viral load results were expressed as copies/ml of plasma. Fig. 11: Electrochemiluminescence Process Electrode TPA TPA TPA TPA TPA H +, TPA" + o TPA^ JPA ^ Ru 2 + V „ 3+ „ 2+ Ru Ru 620 nm Magnet Electrode 64 Viral Isolate Sequencing The pol gene o f selected HIV-1 isolates was sequenced to determine i f Z D V resistance mutations were present at codons 215 and 219, to confirm the phenotypic results obtained using our novel quantitative culture-based Z D V susceptibility assay (215-218). The sequencing was performed in collaboration with the laboratory o f Dr. Sharon Cassol (Brit ish Columbia Centre o f Excellence in H I V - 1 / A I D S ) and was based on stored frozen pellets o f P B M C s isolated from each patient at the time o f initial sample processing. PCR-based sequencing was completed on D N A isolated from these cell pellets. The procedure involved two separate P C R reactions, and two sets o f nested primers as previously described. The amplified D N A was purified. The P C R product was run on an agarose gel and the resulting bands containing the appropriate D N A fragments was removed and purified using Centri-Sep columns. The D N A was then sequenced using an A B I D N A automated sequencer and the results reported as wi ld type or mutant for each codon of interest (215 and 219). A n isolate was considered genotypically resistant i f a characteristic mutation was present at one or other codon. Clinical Data For each patient, the fol lowing data were available: presence or absence o f symptoms, presence or absence of encephalopathy, mode o f transmission, and prior antiretroviral therapy (including any antiretroviral therapy received by the mother during pregnancy). This information, along with the C D 4 + T cell count and percent, was integrated into our analysis o f the results. Statistical Methods Descriptive statistics were performed for quantitative measurements o f plasma viral load and cel l -associated viral titre. Comparisons o f these measurements with phenotype and Z D V resistance assay results were carried out using the Mann-Whitney test. Bivariate analysis o f quantitative measurements o f viral load and cell-associated viral load were carried out using Spearman's rank correlation coefficient. Bivariate relationships between qualitative, dichotomous variables were evaluated using Pearson's chi-square test. A l l p-values are two-sided. 65 RESULTS Patient Population A t least one virologic parameter was evaluated in a total o f 86 patients as shown in Table 5. 12 patients from Ottawa, 14 from Vancouver, 10 from Montreal Children 's , 6 from London, 1 from Halifax, 12 from St. Justine's, and 31 from Toronto were assessed. Table 5: Study Population C I T Y SITE N U M B E R Toronto Hospital for Sick Children 31 Vancouver B C Children's Hospital 14 Ottawa Children's Hospital o f East 12 Montreal Saint Justine's Hospital 12 Montreal Montreal Children's Hospital 10 London St. Joseph's Health Centre 6 Halifax Izaac Walter K i l l a n Hospital 1 Qualitative Macrococultures Qualitative macrococultures were performed for 74 patients (see Table 6) who were enrolled in the study and viral isolates were successfully generated in 60 (81%) study participants. Table 6: Qualitative Macrococultures Culture Number Percent Positive 60 81 Negative 14 !9 Cell-Associated Viral Load (Quantitative IVIicrococultures) In general, these were set up in parallel with the qualitative macrococultures, with the highest dilution of patient cells yielding a positive culture taken as a measure of the number of infectious viral units per mil l ion cells (5, 25, 125, 625, 3125, or 15625 IU/10 6 cells, respectively, based on the dilutions we have used). Table 7 shows the descriptive characteristics o f the cell-associated viral loads obtained in 60 pediatric patients in which the assay was completed. The minimum load was 5 I U / 1 0 6 cells and the 66 maximum load was 15625 I U / 1 0 6 cells. The mean titre was 543 + 2094 I U / 1 0 6 cells, the median being 125 IU/10 6 cells. Categorically, 25% of the samples had loads below 25 IU/10 6 cells, and 75% below 125 IU/10 6 cells. A total o f 13 samples yielded indeterminate results, 8 with loads below 5 IU/10 6 cells, and 5 others with apparently positive results, but an inconsistent pattern o f results on serial cell dilutions. Consequently, these latter 5 values were not felt to be interpretable. Table 7; Descriptive Characteristics o f Cell-Associated Vi ra l Load ( IU/10 6 cells) Cell-Associated Viral Load IU/106 Cells Logio Number o f Samples 60 60 M i n i m u m 5 0.70 M a x i m u m 15625 4.19 Mean 543 1.85 Standard Deviation 2094 0.80 Coefficient o f Variat ion 385% 43% 25th Percentile 25 1.40 Median 125 2.10 75th Percentile 125 2.10 Viral Phenotype Vira l phenotype (Sl /NSI) determinations were completed in 78 patients (from whom viral isolates were generated on qualitative or quantitative cultures) with 23 (30%) containing the syncytium inducing (SI) variant o f the virus and 55 (70%) the non-syncytium inducing (NSI) variant (Table 8). Table 8: V i r a l Phenotype (SI/NSI) Determinations Phenotype Number Percent N S I 55 70 SI 23 30 Total 78 100 Plasma Viral Load Plasma viral load testing was completed on a total o f 85 patients. Table 9 shows the descriptive characteristics o f the viral load results obtained. The lower limit o f detection o f this assay was 1 000 copies/ml of plasma, with any sample testing negative considered to have 1 000 copies/ml for the purposes of analysis. The maximum viral load value obtained was 5 100 000 copies/ml. The mean and standard 67 deviation were found to be 214 580 and 682 570 copies/ml, respectively, with 25% o f the samples having viral load values below 8 500 and 75% of the samples had viral load values below 76 000 copies/ml. The median value was 25,000 copies/ml Table 9: Descriptive Characteristics o f Plasma Vi ra l Load (copies /ml) Viral Load Copies/ml Logio Number o f Samples 85 85 M i n i m u m 1 000 2.00 M a x i m u m 5 100 000 5.71 Mean 214 580 3.44 Standard Deviation 682 570 0.83 Coefficient o f Variation 318% 24% 25th Percentile 8 500 2.92 Median 25 000 3.40 75th Percentile 76 000 3.88 Standard Phenotypic Resistance Testing ( A C T G ) Standard phenotypic resistance testing was completed on a total o f nine patient samples. 5/9 (55%) of the samples were determined to be sensitive to Z D V and 4/9 (45%) resistant. Quantitative Culture-Based Z D V Resistance Assay Definitive results using our pilot assay were generated in 60 patients, with 46 (77%) found to be sensitive to Z D V and 14 (23%) resistant. In an additional 14 samples, indeterminate results were obtained, due to the viral load being too low or too high, preventing us from appreciating a difference in viral replication in the presence and absence o f the drug. The results are summarized in Table 10. Table 10: Characteristics o f Z D V Resistance Using Rapid Phenotypic Resistance Testing Z D V Resistance Number Percent Sensitive 46 62 Resistant 14 19 Indeterminate 14 19 Total 74 100 68 Viral Isolate Sequencing (Genotypic Resistance) As mentioned in Materials and Methods, patient P B M C s at time of isolation were pelleted and frozen to be used later for viral isolate sequencing. The R T fragments o f the viral genome were amplified and sequenced to establish the presence or absence o f Z D V resistance-conferring mutations at codons 215 and 219 specifically. 53 samples were completed in all . Table 11 demonstrates the characteristics o f genotypic resistance in those patients assessed examining codon 215 only. For further analyses, those samples that were in a transition state, were considered resistant. Table 11: Characteristics o f Z D V Resistance Using Genotypic Analysis Z D V Resistance Number Percent W i l d Type (Sensitive) 27 51 Mutant (Resistant) 22 42 Transition State 4 7 Total 53 100 CD4 Cell Counts Table 12 shows the descriptive characteristics o f the C D 4 + T cell counts obtained for 70 patients at the time of their initial study evaluation. Values ranged from 0-2,402 C D 4 cel ls /mm 3 (mean 493 ± 563). The median value was 329 cel ls /mm 3 with 25% below 63 and 75% below 661 cel ls /mm 3 . Table 12: Descriptive Characteristics o f C D 4 Ce l l Counts (cells/mm 3) and C D 4 Percent CD4 Cell Count Cells/mm 3 Percent Number o f Samples 70 69 M i n i m u m 0 0.0 M a x i m u m 2 402 60.6 Mean 493 18.9 Standard Deviation 563 14.0 Coefficient o f Variation 114% 74% 25th percentile 63 5.0 Median 329 19.2 75th percentile 661 29.2 69 Mode of Transmission The manner in which the virus was contracted (mode of transmission) was assessed in 86 patients. O f the 69 patients for whom this information was available, 47 (68%) acquired the virus perinatally, and 22 (32%) acquired the virus non-perinatally, usually through blood transfusions. Presence of Symptoms This information was available in 69 patients, 57 (83%) o f whom were exhibiting symptoms at time o f sample collection, the remaining 12 (17%) being asymptomatic. Encephalopathy This information was available in 56 patients, only 8 (14%) o f which showed any cl inical evidence of H1V-1-associated encephalopathy. 70 CORRELATIVE ANALYSES Plasma Viral Load and Cell-Associated Viral Load We compared the plasma and cell-associated viral loads in the circulation to determine i f a relationship existed between these two variables, which measure slightly different stores o f H I V - 1 . The two were positively correlated (r=0.37, p=0.004) although not absolutely so. Viral Load and Viral Phenotype The relationship between plasma or cell-associated viral load and phenotype was also explored. Tables 13 and 14 show that there was no correlation between these variables. Plasma Viral Load Phenotype vuai rnenoiype ( S I / I N ; p-value* Descriptive Measure NSI(n=54) SI(n=23) 0.956 Copies/ml Copies/ml M i n i m u m 1 000 1 000 Max imum 5 100 000 450 000 Mean 299 760 84 990 Standard Deviation 842 590 135 070 25th percentile 8 700 10 000 Median 25 000 34 000 75th percentile 91 000 97 000 *Based on Mann-Whitney test Table 14: Relationship Between Cell-Associated V i r a l Load (IU/10 I 7TT. I 77. 77 : I • ^ Cell-Assoc. Viral Load Descriptive Measure Min imum Max imum Mean Standard Deviation 25th percentile Median 75th percentile NSI(n=42) Phenotype SI(n=18) IU/10" cells 5 3125 218 503 25 125 125 * Based on Mann-Whitney test IU/10 6 cells 5 15625 1302 3707 25 125 125 cells) and V i r a l Phenotype (SI/NSI) p-value* 0.410 71 Plasma Viral Load and CD4 Cell Count The relationships between plasma viral load and absolute C D 4 cell count and percentage were assessed in 69 patients. There were no significant relationships found between these variables (data not shown). Plasma Viral Load and Presence of Symptoms The relationship between plasma viral load and the presence or absence o f symptoms was examined. A total o f 68 samples were assessed. Table 15 shows the relationship between these two variables. N o i significant relationship was found. Table 15: Relationship Between Plasma V i r a l Load (copies/ml) and Presence o f Sympt | —— —— -— - I — , Plasma Viral Load Descriptive Measure Min imum Max imum Mean Standard Deviation 25th percentile Median 75th percentile Presence of Symptoms No (n=12) Copies/ml 1 000 2 000 000 215 240 576 690 5 250 10 500 31 500 Yes (n=56) Copies/ml 1 000 5 100 000 252 460 786710 630 27 000 103 500 p-value* 0.281 *Based on Mann-Whitney test Plasma Viral Load and Encephalopathy Among the symptoms, it is interesting to note that the presence of encephalopathy was highly correlated with a higher plasma viral load. (See Table 16) Table 16: Relationship Between Plasma Vi ra l Load (copies/ml) and Presence of Encephalopathy Encephalopathy present absent n mean median p-value 8 887 870 74 000 0.023 48 76 760 16 000 Plasma Viral Load and Mode of Transmission The relationship between plasma viral load and mode o f transmission (perinatal or non-perinatal) was also explored (Table 17). In 68 patients, there was a significant correlation between these two variables (p=0.004), with a significantly higher plasma viral load in children who acquired their infection perinatally. 72 Plasma Viral Load Mode of Transmission p-value* Descriptive Measure Perinatal (n=46) Non-perinatal (n=22) 0.004 Copies/ml Copies/ml Min imum 1 000 1 000 Maximum 5 100 000 97 000 Mean 352 310 19 820 Standard Deviation 895 440 26 720 25th percentile 10 000 3 500 Median 34 000 11 350 75th percentile 210 000 20 000 *Based on Mann-Whitney test Cell-Associated Viral Load and CD4 Cell Count The relationship between cell-associated viral load and absolute C D 4 cell count and percentage were assessed in 54 patients. There were no significant relationships between these variables. Cell-Associated Viral Load and Presence of Symptoms This analysis was performed in parallel with that o f viral load and symptoms of HIV-1 infection. In this case no significant association was found. (See Table 18) Table 18: Relationship Between Cell-Associated V i r a l Load (IU/10 Cell-Assoc. Viral Load Presence of Symptoms p-value* Descriptive Measure No (n=9) Yes (n=46) 0.706 IU/10 6 cells IU/10 6 cells Min imum 5 5 Max imum 3 125 15 625 Mean 467 606 Standard Deviation 1 015 2 354 25th percentile 25 25 Median 125 125 75th percentile 125 125 *Based on Mann-Whitney test Cell-Associated Viral Load and Presence of Encephalopathy Among the symptoms, children without encephalopathy tended to have a slightly higher cell-associated viral load (Table 19), although this association did not quite reach statistical significance. It is interesting 73 to note that this relationship is inverted compared to the previously described relationship between pi, viral load and encephalopathy. Table 19: Relationship Between Cell-Associated V i r a l Load ( IU/10 6 cells) and Presence o f Encephalopathy Cell-Assoc. Viral Load Presence of Encephalopathy p-value* Descriptive Measure No (n=38) Yes (n=6) 0.068 IU/10 6 cells IU/10 6 cells Min imum 5 5 Max imum 15 625 3 125 Mean 688 775 Standard Deviation 2 586 1 181 25th percentile 25 125 Median 25 375 75th percentile 125 625 *Based on Mann-Whitney test Cell-Associated Viral Load and Mode of Transmission In contrast, a significantly higher cell-associated viral load was measured in children with non-perinatal acquisition of HIV-1 infection (Table 20). This finding also differs somewhat from the demonstrated correlation between plasma viral load and perinatal HIV-1 infection. Table 20: Relationship Between Cell-Associated Vi ra l Load ( IU/10 6 cells) and Mode o f Transmi, Cell-Assoc. Viral Load Mode of Transmission p-value* Descriptive Measure Perinatal (n=39) Non-perinatal (n=16) 0.001 IU/10 6 cells IU/10 6 cells M i n i m u m 5 5 Max imum 3 125 15 625 Mean 411 1 004 Standard Deviation 823 3 899 25th percentile 25 5 Median 125 25 75th percentile 625 25 *Based on Mann-Whitney test Comparative Resistance: Phenotype vs. Genotype We compared the prevalence o f Z D V resistance obtained using our rapid phenotypic resistance testing method and those obtained by formal sequencing (genotypic resistance). Table 21 shows the relationship between these two variables. A n isolate was deemed genotypically resistant to Z D V i f it carried the typical mutation at codon 215. It should be noted that a mutation, which confers resistance to a drug, would be 74 detectable before the virus population in a patient becomes sufficiently resistant to the drug for it to be noticed phenotypically. That may be the case with the 8 samples, which were found to have mutant codons at position 215 yet were found to be sensitive by our rapid phenotypic resistance assay. In the case o f the two samples which were found to be resistant by our assay, but were w i ld type at codon 215, mutations at other codons may have been present which would have also conferred resistance to Z D V . It is known, in fact, that the two samples were mutant at codon 219. Thus, o f 45 evaluable isolates, 35 (78%) produced concordant results, with 8/10 mismatches possibly attributable to the presence o f genotypic resistance in advance of the development o f phenotypic resistance. Table 21: Relationship Between Genotypic and Phenotypic resistance Phenotype Sensitive Resistant Wild Type (Sensitive) Genotype Mutant (Resistant) 25 2 8 10 Comparative Resistance: Genotype vs. Rapid Phenotype vs. Standard Phenotype ( A C T G ) Standard phenotypic resistance assays according to the A I D S Cl in ica l Trials Group ( A C T G ) protocols were performed on selected samples to confirm the results obtained from our pilot assay. In total, assays were completed on 9 patients. Table 22 shows a comparison o f the results obtained for those samples on which all three resistance assays were completed. Looking at the results from the rapid and standard phenotypic testing only, the first four results were completely concurrent between the two phenotypic assays. The fifth sample was indeterminate by our assay, susceptible by the A C T G assay, but shown to be in a transition state between wi ld type and mutant at codon 215 by genotypic analysis. The sixth sample, also in a transition state, was sensitive by our assay and resistant by the A C T G assay. The last three samples showed concurrence between the genotypic assay and A C T G assay, but were indeterminate or sensitive by the rapid pilot assay. 75 Table 22: Comparison of Genotypic, Rapid and Standard Phenotypic Resistance dotlntiini/t A f <•.«»». I"> 'J r»I. . • , I ~Z. 1 Genotypic Assay Rapid Phenotypic Assay Standard Phenotypic Assay f A C T G ) W T M U TS TS M U M U M U S S S S I s I s I 0.088 0.144 0.171 0.173 0.163 2.65 4.35 >10 >10 Genotypic Resistance and Other Variables The relationship between genotypic Z D V resistance and absolute C D 4 cell counts, C D 4 cell percentages, symptoms, encephalopathy and the mode of transmission were evaluated (Tables 23-27). C D 4 cell counts and percentages were lower in patients with resistant isolates. N o other correlations were identified. Table 23: Relationship Between C D 4 Ce l l Count (cells/mm 3) and Genotypic Resistance CD4 Cell Count Descriptive Measure Min imum Maximum Mean Standard Deviation 25th percentile Median 75th percentile Genotypic Resistance Sensitive (n=25) 1 900 798 574 400 661 1 080 * Based on Mann-Whitney test Resistant (n=24) 1 806 278 450 35 75 326 p-value* <0.001 Table 24. Relationship Between C D 4 Percent and Genotypic Resistance Percent CD4 I Genotypic Resistance Descriptive Measure Minimum M a x i Mean Standard Deviation 25th percentile Median 75th percentile Sensitive (n=25) 60.6 26.0 13.3 21.1 29 32.1 * Based on Mann-Whitney test Resistant (n=24) p-value* 0.30 42.0 12.2 11.5 2.3 7.0 21.0 <0.001 76 Table 25. Relationship Between Genotypic Resistance and Presence o f Symptoms * Based on chi-squared test (uncorrected) Genotypic Resistance Presence of Symptoms Sensitive Resistant p-value* No 7 (29%) 2 (8%) 0.056 Yes 17(71%) 23 (92%) Total 24(100%) 25 (100%) Table 26. Relationship Between Genotypic Resistance and Presence of Encephalopathy Genotypic Resistance Encephalopathy Sensitive Resistant p-value* present 2 (11%) 5 (23%) 0.301 absent 17(89%) 17 (77%) Total 19(100%) 22 (100%) Based on chi-squared test (uncorrected) Table 27. Relationship Between Genotypic Resistance and Mode of Transmission Genotypic Resistance Mode of Transmission Sensitive Resistant p-value* Perinatal 19(76%) 18 (72%) 0.747 Non-perinatal 6 (24%) 7 (28%) Total 25 (100%) 25 (100%) Based on chi-squared test (uncorrected) Phenotypic Resistance and Other Variables Similar analyses were performed based on the phenotypic resistance data, using the rapid pilot (Tables 28-32). Similar results were obtained. Table 28. Relationship Between Phenotypic Resistance and C D 4 Ce l l Count (cells/mm 3) CD4 Cell Count Phenotypic Resistance Descriptive Measure Sensitive (n=43) Resistant (n=l 1) p-value* Minimum 0 10 0.009 Maximum 2 402 600 Mean 639 145 181 Standard Deviation 633 25th percentile 80 14 Median 489 69 75th percentile 912 240 * Based on Mann-Whitney test 77 Table 29. Relationship Between Phenotypic Resistance and C D 4 Percent Percent CD4 Descriptive Measure M i n i m u m Max imum Mean Standard Deviation 25th percentile Median 75th percentile Phenotypic Resistance Sensitive (n=43) 0.00 60.0 21.5 14.7 8.0 24.1 32.0 Resistant (n=ll) 1.7 23.0 10.7 8.6 2.3 6.0 19.4 p-value* 0.026 * Based on Mann-Whitney test Table 30. Relationship Between Phenotypic Resistance and Presence o f Symptoms Phenotypic Resistance Presence of Symptoms Sensitive Resistant p-value* N o 9 (21%) 0 (0%) 0.083 Yes 34 (79%) 12(100%) Total 43 (100%) 12 (100%) * Based on chi-squared test (uncorrected) Table 31. Relationship Between Phenotypic Resistance and Presence o f Encephalopathy Phenotypic Resistance Encephalopathy Sensitive Resistant p-value* Present 4 (12%) 29 (88%) 2(18%) 9 (82%) 0.612 Absent Total 33 (100%) 11 (100%) * Based on chi-squared test (uncorrected) Table 32. Relationship Between Phenotypic Resistance and Mode o f Transmission Phenotypic Resistance Mode of Transmission Sensitive Resistant p-value* Perinatal 30 (70%) 9 (75%) 0.724 Non-perinatal 13 (30%) 3 (25%) Total 43 (100%) 12 (100%) * Based on chi-squared test (uncorrected) Phenotypic Resistance and Plasma Viral Load We also examined whether any relationship existed between phenotypic resistance and plasma or cell associated viral load (Tables 33 and 34). N o such relationships were identified. 78 Table 33. Relationship Between Plasma Vi ra l Load (copies/ml) and Phenotvnic Resistance m. \ri i • i , . . . I 1 1 J r Plasma Viral Load Descriptive Measure Minimum Maximum Mean Standard Deviation 25th percentile Median 75th percentile Z D V Resistance Sensitive (n=45) Copies/ml 000 2 400 000 240 390 562 870 6 400 23 000 110 000 Resistant (n=14) Copies/ml 100 5 100 000 439 820 346 610 20 000 33 000 76 000 p-value* 0.487 *Based on Mann-Whitney test Phenotypic Resistance and Cell-Associated Viral Load A comparison between cell-associated viral load and rapid phenotypic Z D V resistance was also performed. A total of 56 samples were compared. Table 34 shows the comparison between these two variables. N o correlation was found. Cell-Assoc. Viral Load Z D V Resistance p-value* Descriptive Measure Sensitive (n=46) Resistant (n=14) 0.248 IU/10 6 cells IU/10 6 cells M i n i m u m 5 5 Max imum 15 625 3 125 Mean 584 409 Standard Deviation 2 356 817 25th percentile 25 25 Median 125 125 75th percentile 125 625 cells) and Phenotypic Resistance Viral Phenotype and CD4 Cell Count Finally, we were able to demonstrate a clear association between viral phenotype and absolute C D 4 cell count and percentage (Tables 35 and 36), but with no other clinical variables (data not shown). 79 Table 35. Relationship Between Vi ra l Phenotype (SI/NSI) and C D 4 Ce l l Counts (cells/mm 3 CD4 Cell Count Descriptive Measure Min imum Maximum Mean Standard Deviation 25th percentile Median 75th percentile * Based on Mann-Whitney test NSI (n=46) Viral Phenotype 2 402 657 620 102 509 912 SI (n=21) 0 825 169 206 56 85 189 p-value* 0.001 Viral Phenotype and CD4 Percent As with C D 4 cell count, C D 4 cell percent was compared with the viral phenotype (SI/NSI) o f patient samples to discover relationships that may have existed. A s with C D 4 cell count, a relationship was found between C D 4 cell percent and viral phenotype. A total o f 66 samples had data available on both these variables. See Table 36. Table 36. Relationship Between V i r a l Phenotype (SI/NSI) and C D 4 Percent Percent CD4 Viral Phenotype Descriptive Measure NSI (n=46) SI (n=20) p-value* Min imum 0.3 0.00 <0.001 Maximum 60.6 30.0 Mean 22.7 10.1 Standard Deviation 14.1 9.2 Coefficient o f Variation 62% 92% 25th percentile 11.2 2.7 Median • 24.5 6.4 75th percentile 32.2 18.5 *Based on Mann-Whitney test 80 DISCUSSION Conduct of Study This was an extended cross-sectional study, performed in all tertiary care institutions in Canada providing care for HIV- l - in fec ted children. A s the vast majority o f such children are fol lowed in one o f these institutions, this can be taken as a very representative evaluation o f the status o f pediatric H I V infection in our country between 1993-95. Standardized case report forms were designed to ensure a uniform method for the collection o f cl inical data. In addition, all o f the blood samples were provided to the virology research laboratory in a blinded fashion, and the results o f all assays were completed and the results tabulated individually before any analysis was undertaken. We are confident that this methodology has led to the generation o f a very reliable and objective data set, increasing the power o f the results we have generated. Cell-Associated Viral Load Cell-associated viral load was determined from the quantitative micrococultures that were set up according to standard protocols. These assays were completed in 60 patients. Cell-associated viral load was highly significantly correlated with plasma viral load ( r 2 =0.37, p=0.004). Sei et al also found this association (r2=0.86, p=0.0025) (323) in 16 patients with symptomatic HIV-1 infection. In our study, cell-associated viral load was also significantly correlated with mode of transmission (p=0.001), higher fol lowing perinatal infection. A trend toward viral load being associated with the presence o f encephalopathy was seen, (p value approaching statistical significance), suggesting an association between viral load and this most severe complication of pediatric HIV-1 infection. Plasma Viral Load Plasma viral load was measured using the Organon Teknika N A S B A assay. This assay was chosen because it only required a small volume (100^1) o f plasma, a particularly favorable characteristic in pediatric studies. Duplicate samples were not required because, according to the manufacturer, good precision (<0.3 logs) is produced by the use o f the three internal R N A calibrators (328). The threshold of 81 detection o f this assay was 1000 copies/ml o f plasma. Overall , this assay generated results in one day and 10 samples could be run in one assay. The R N A isolation procedure was a bit tedious, but the remainder o f the assay was highly mechanized and efficient, especially since the amplification segment o f the assay was isothermal. A total o f 85 samples were evaluated for plasma viral load, generating results highly correlated with the cell-associated viral loads. However, a highly significant correlation was found between plasma viral load and mode o f transmission (p=0.004), favouring higher loads in non-perinatal infection. Saag et al. (20) also found that each o f five children infected with HIV-1 in utero or during the perinatal period had detectable plasma viremia, independent o f C D 4 + cell count, duration of infection, or cl inical state. However, the 4 children infected by HIV-1 at older ages had detectable plasma viremia less frequently. The reasons for this association merit further exploration, but may relate to treatment effects not controlled for in our current analysis. A s such, it may be that patients receiving antiviral drugs may have lower plasma viral loads. I f this were more frequent in patients with perinatal acquisition o f infection, this could explain our results. Cell-associated viral load is much less susceptible to treatment effects and the true relationship between viral load and mode o f acquisition may be towards higher loads fol lowing perinatal infection, as shown in the cellular assays. This is relatively easy to explain, as such patients would have been infected longer and there may be an association between total body viral burden and the duration of infection. The consistency o f the pathologic significance of viral burden may best be appreciated by examining the relationship o f plasma and cell-associated viral load with encephalopathy, a serious virologically-related complication o f HIV-1 infection in children. Phenotypic Resistance Phenotypic resistance to Z D V was not found to be associated with plasma viral load, cell-associated viral load, phenotype, and mode o f transmission or presence o f encephalopathy. However, a strong correlation was seen between Z D V resistance and lower absolute CD4+ T cell count (p=0.009) and percent (p=0.026). This trend has been demonstrated on numerous occasions in the past, and speaks o f the more rapid acquisition o f Z D V resistance in patients with advanced immune disease, first described by Richman et al. in 1991 (217). 82 Genotypic Resistance Genotypic resistance was not found to be associated with mode o f transmission or presence o f encephalopathy. However, as with phenotypic resistance, a strong correlation was found between genotypic resistance and lower absolute C D 4 + T cell count (p<0.001) and percent ( p O . O O l ) . This confirms the previously demonstrated associations with phenotypic drug resistance. Prevalence of Zidovudine Resistance In this study, using our rapid phenotypic resistance assay, the prevalence o f Z D V resistance in the population was 19% (14/74). There were 46/74 (62%) sensitive isolates while indeterminate results were obtained in 14 (19%) cases. Examining definite results only, 77% (46/60) o f the virus isolates were susceptible to Z D V and 23% (14/60) were resistant. Us ing the genotypic resistance assay, 5 1 % (27/53) of the isolates sequenced showed the presence of a mutation conferring resistance to Z D V at codon 215, and 49% (26/53) did not. The higher rate o f genotypic vs. phenotypic resistance reflects the fact that resistance mutations appear in the viral population long before the development o f full-blown phenotypic resistance. For our genotypic studies, we focused on changes at codon 215 of the reverse transcriptase gene. In previously published studies, every case where a reduction in Z D V susceptibility occurred, there were mutations in the reverse transcriptase at codon 70, codon 215 or both (216). A complete picture could have been generated by looking at changes at codon 70. However, we were not prepared to undertake this additional sequencing as part o f this pilot project. Using the formal standard phenotypic resistance assay (124), which we conducted to confirm the results o f our rapid phenotypic assay, only nine useable results were obtained. We attempted to complete the test in 29 cases. The high failure rate speaks to the complexity of the assay and its lack o f applicability to the evaluation o f small samples, such as those generated in pediatric studies. It should be noted that concurrence between the two phenotypic assays was only obtained for strains that were sensitive by the rapid pilot test we are developing. The rapid assay has several advantages. It can be adapted to test susceptibility o f HIV-1 to other antiretroviral drugs. It also takes much less time than the standard phenotypic assay. It directly cultures virus-containing patient P B M C s , immediately following their isolation from whole blood. Virus stocks do 83 not need to be generated and then titrated. With the standard phenotypic assay, virus stocks do need to be generated, which may or may not be successful, especially i f small volumes o f blood are used. Because there are so many more steps involved in the A C T G resistance assay, there is more room for error and failure. For example, 60 rapid phenotypic resistance assays could be performed, whereas only 29 titrations were successfully performed and of those 29, only 11 generated titres high enough that the second part o f the resistance assay could be performed. The fewer the steps involved, the more results can be generated more easily. A l s o , for treatment purposes, a test which can detect resistance quickly is needed, so that decisions can be made regarding changes in therapy as soon as possible. One disadvantage of the rapid resistance assay is that in the limited number o f samples we evaluated, it seemed to underestimate the prevalence o f resistance. More comparative data w i l l have to be generated, although we did detect a significant number o f resistant isolates quite apart from the comparisons and we are encouraged by the preliminary results o f Nelson (300) using an assay similar to ours. Clearly, more work is needed, but we feel that a rapid phenotypic assay w i l l become the standard o f care in the near future. Viral Phenotype Debate still exists as to the clinical usefulness of evaluating viral phenotype in HIV- l - in fec ted patients. It has been shown that patients harboring SI virus are more likely to progress to A I D S more rapidly. Nielsen et al. have also shown that development o f phenotypic and genotypic resistance was faster in patients harboring SI isolates (326). In this study, viral phenotype was determined in a total o f 78 patients. 23/78 (29%) of the patient isolates had the SI phenotype while 55/78 (71%) did not. There was no correlation between viral phenotype and plasma viral load, cell-associated viral load, phenotypic resistance, and presence of symptoms, or mode of transmission. There was also a strong correlation between viral phenotype and absolute C D 4 + T cell count (p=0.001) and percent (p<0.001), confirming previous observations. The independent value of measuring the phenotype was not demonstrated in this study and remains to be fully elucidated. 84 CONCLUSION We have successfully completed a comprehensive cross-sectional study o f pediatric H I V infection in Canada. This, in itself was a very useful exercise. It has led to the creation o f a truly national database, useful to all researchers in this field. It has also established an infrastructure that could be used for the design and implementation o f future clinical trials and other studies o f pathogenesis. We have detected a high prevalence of zidovudine resistance, a fact which w i l l have to be taken into account in the design o f future guidelines and trials. In establishing our "viral inventory", it is somewhat surprising that we were not able to demonstrate any significant associations between different parameters, except for the relationship between plasma and cell-associated viral load. It may be that the associations wi l l become more clear as prospective data are collected. Preliminary data to support this statement may be the association between viral load and mode of transmission, which may be reflective o f the higher loads seen in children that have been infected for a longer period o f time. Whatever the answer, we have successfully defined the status o f HIV- l - in fec ted Canadian children, and future students and other researchers w i l l be able to build on this to increase our insight into the ideal management of these individuals. 85 R E F E R E N C E S 1. Oleske J, Minnefor A , Cooper R, et al. Immune deficiency in children. Journal o f the American Medical Association 1983;249:2345-2349. 2. Rubinstein A , S ick l ick M , Gupta A , et al . Acquired immunodeficiency with reversed T4/T8 ratios in infants born to promiscuous and drug-addicted mothers. Journal o f the American Medica l Association 1983;249:2350-2356. 3. Scott G B , Buck B E , Letterman J G , et al. Acquired immunodeficiency syndrome in infants. N e w England Journal o f Medicine 1984;310:76-81. 4. From the Wor ld Health Organization. Acquired Immunodeficiency Syndrome ( A I D S ) - Data as o f 20 November 1996. Pediatric A I D S and HIV-1 Infection: Fetus to Adolescent 1997:8( 1 ):79-83. 5. Health Canada. H I V and A I D S in Canada: Surveillance report to December 1998. Divis ion of H I V -1/AIDS Surveillance, Bureau o f H I V - 1 / A I D S and S T D , L C D C , Health Canada, A p r i l 1999. 6. Health Canada. A I D S in Canada: Quarterly Surveillance update. Div i s ion o f H I V - 1 / A I D S Surveillance. Bureau o f H I V - 1 / A I D S and S T D , L C D C , H P B , Health Canada, M a y 1997. 7. Rosenberg Z F , Fauci A . Immunopathology and pathogenesis o f H I V - 1 infection, from: Pizzo and Wilfert eds: Pediatric A I D S . The challenge of HIV-1 infection in infants, children and adolescents 2 n d edition, Wi l l i ams and Wi lk ins , Baltimore, M D 1994. 8. Cowan M J , Hel lman D , Chudwin D, et al. Maternal transmission of acquired immunodeficiency syndrome. Pediatrics 1984;73:382-6. 9. Rogers M F , Thomas P A , Starcher ET, et al. Acquired immunodeficiency syndrome in children: report of the centres for disease control national surveillance 1982-1985. Pediatrics 1987;79:1008-1014. 10. Indacochea FJ , Scott G B . HIV-1 infection and the acquired immunodeficiency syndrome in children. Current Problems in Pediatrics 1992; 166-204. 11. Ho D D , Sarngadharan M G , Resnick L , et al. Primary human T-lymphotropic virus type III infection. Annals o f Internal Medic ine 1985;103:880-883. 12. Cooper D A , G o l d J , MacLean P, et al . Acute A I D S retrovirus infection: definition o f a cl inical illness associated with seroconversion. Lancet 1985;1:537-540. 13. Levy J A , Shimabukuro J, Hollander H , et al. Isolation o f AIDS-associated retroviruses from cerebrospinal fluid and brains o f patients with neurological symptoms. Lancet 1985;2:586-588. 14. Ho D D , Rota T R , Schooley R T , et al. Isolation o f H T L V - I I I from cerebrospinal fluid and neural tissues of patients with neurologic syndromes related to acquired immunodeficiency syndrome. N e w England Journal o f Medic ine 1985;313:1493-1497. 15. Goudsmit J, d e W o l f F, Paul D A , et al. Expression of human immunodeficiency virus antigen ( H I V - 1 -A g ) in serum and cerebrospinal fluid during acute and chronic infection. Lancet 1986;2:177-180. 16. Daar ES, M o u d g i l T, Meyer R D , et al. Transient high levels o f viremia in patients with primary HIV-1 infection. New England Journal o f Medicine 1991;324:961-964. 86 17. Srugo I, Brunell P A , Chelyapov N V , et al. Virus burden in human immunodeficiency virus type-1 infected children: relationship to disease status and effect o f antiviral therapy. Pediatrics 1991;87(6):921-925. 18. Al iment i A , Luzuriaga K , Stechenberg B , et al. Quantitation of human immunodeficiency virus in vertically infected infants and children. Journal o f Pediatrics 1991 ;119:225-229. 19. Alimenti A , O ' N e i l l M , Sull ivan J L , et al. Diagnosis o f vertical human immunodeficiency virus type 1 infection by whole blood culture. Journal o f Infectious Diseases 1992; 166:1146-1148. 20. Saag M S , Crain M J , Decker W D , et al. High-level viremia in adults and children infected with human immunodeficiency virus: relation to disease stage and C D 4 + lymphocyte levels. Journal o f Infectious Diseases 1991;164:72-80. 21. M c C u n e J M . H I V - 1 : The infective process in vivo. Cel l 1991;64:351-363. 22. Kr iv ine A , Firtion G , Cao L , et al. H I V - 1 replication during the first few weeks o f life. Lancet 1992;339:1187-1189. 23. Krasinski K , Borkowski W , Holzman R S . Prognosis o f human immunodeficiency virus infection in children and adolescents. Pediatric Infectious Diseases Journal 1989;8:216-220. 24. Tovo P - A , deMartino M , Gabiano C , et al. Prognostic factors and survival in children with perinatal HIV-1 infection. Lancet 1992;339:1249-1253. 25. Masur H , Michel i s M A , Greene J B , et al. A n outbreak o f community-acquired Pneumocystis carinii pneumonia: initial manifestation o f cellular immune dysfunction. N e w England Journal o f Medicine 1981;305:1431-1438. 26. Gottlieb M S , Schroff R, Schanker H M , et al. Pneumocystic carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: evidence o f a new acquired cellular immunodeficiency. N e w England Journal o f Medicine 1981 ;305:1425-1431. 27. Barre-Sinoussi F, Chermann J C , Rey F, et al. Isolation o f a T-lymphotropic retrovirus from a patient at risk for acquired immunodeficiency syndrome (AIDS) . Science 1983;220:868-871. 28. Levy J A , Hoffman A D , Kramer S M , et al. Isolation o f lymphocytopathic retroviruses from San Francisco patients with A I D S . Science 1984;225:840-842. 29. Ga l lo R C , Salahuddin S Z , Popovic M , et al. Frequent detection and isolation o f cytopathic retroviruses ( H T L V - I I I ) from patients with A I D S and at risk for A I D S . Science 1984;224:500-503. 30. Popovic M , Sarngadharan M G , Read E , et al. Detection, isolation, and continuous production of cytopathic retroviruses ( H T L V - I I I ) from patients with A I D S and p re -AIDS. Science 1984;224:497-500. 31. Dalgleish A G , Beverley P C , Clapham PR, et al. The C D 4 (T4) antigen is an essential component o f the receptor for the A I D S retrovirus. Nature 1984;312:763-767. 32. Klatzmann D , Champagne E , Chamaret S, et al. T-lymphocyte T4 molecule behaves as the receptor for human retrovirus L A V . Nature 1984;312:767-768. 33. Maury W , Potts B J , Rabson A B . HIV-1 infection of first-trimester and term human placental tissue: a possible mode of maternal fetal transmission. Journal o f Infectious Diseases 1989;160:583-588. 87 34. Lewis S H , Reynolds-Kohler C , Fox H E , et al. HIV-1 in trophoblastic and vi l lous hofbauer cells, and haematological precursors in eight-week fetuses. Lancet 1990;335:565-568. 35. Courgnaud V , Laure F, Brossard A , et al. Frequent and early in utero HIV-1 infection. A I D S Research and Human Retroviruses 1991;7:337-341. 36. Ehmst A , Lindgren S, Dictor M , et al. HIV-1 in pregnant women and their offspring: evidence for late transmission. Lancet 1991;338:203-207. 37. Goedert JJ, Duliege A M , Amos C L , et al. High risk of HIV-1 infection for first-born twins. The International Registry o f HIV-1-Exposed Twins. Lancet 1991;338:1471-1475. 38. K e l l y D A , Hallett R J , Saeed A , et al. Prolonged survival and late presentation o f vertically transmitted HIV-1 infection in childhood. Lancet 1987; 1:806-807. 39. Papiernik M , Brossard Y , M u l l i e z N , et al. Thymic abnormalities in fetuses aborted from human immunodeficiency virus type 1 seropositive women. Pediatrics 1992;89:297-301. 40. Epstein L G , Boucher C A , Morr ison S H , et al. Persistent human immunodeficiency virus type 1 antigenemia in children correlates with disease progression. Pediatrics 1988;82:919-924. 41. Ammann A J. Immunopathogenesis o f pediatric acquired immunodeficiency syndrome. Journal o f Perinatology 1988;8:154-159. 42. Blanche S, LeDeist F, Fischer A , et al. Longitudinal study of 18 children with perinatal L A V / H T L V I I I infection: attempt at prognostic evaluation. Journal o f Pediatrics 1986;109:965-970. 43. Pahwa S, Kaplan M , F ik r ig S, et al. Spectrum of human T-cel l lymphotropic virus type infection in children. Journal o f the American Medica l Association 1986;255:2299-2305. 44. deMartino M , Tovo P - A , G a l l i L , et al. Prognostic significance o f immunologic changes in 675 infants perinatally exposed to human immunodeficiency virus. Journal o f Pediatrics 1991;119:702-709. 45. Scott G B Hutto C , Makuch R W , et al. Survival in children with perinatally acquired immunodeficiency virus type 1 infection. N e w England Journal o f Medicine 1989;321:1791-1796. 46. Auger I, Thomas P, DeGruttola V , et al. Incubation period for pediatric A I D S patients. Nature 1988;366:575-577. 47. Leibovitz E , Rigaud M , Pollack H , et al. Pneumocystis carinii pneumonia in infants infected with the human immunodeficiency virus with more than 450 C D 4 T-lymphocytes per cubic millimeter. N e w England Journal o f Medicine 1990;323:531-533. 48. Bernstein L J , Krieger B Z , N o v i c k B , et al. Bacterial infection in the acquired immunodeficiency syndrome of children. Pediatric Infectious Diseases Journal 1985;4:472-475. 49. Lepage P, Van de Perre P, Van Viet G , et al. Cl in ica l and endocrinologic manifestations in perinatally human immunodeficiency virus type 1-infected children aged 5 years or older. American Journal o f Diseases o f Children 1991; 145:1248-1251. 50. Pahwa S, Chirmule N , Leombruno C , et al. In vitro synthesis o f human immunodeficiency virus-specific antibodies in peripheral blood lymphocytes of infants. Proc. Nat l . Acad . Sc i . U S A 1989;86:7532-7536. 88 51. Pyun K H , Ochs H D , Dufford M T W , Wedgewood R J . Perinatal infection wi th human immunodeficiency virus: specific antibody responses by the neonate. N e w England Journal o f Medicine 1987;317:611-614. 52. Quinn T C , K l ine R L , Halsey N , et al. Early diagnosis o f perinatal H I V - 1 infection by detection o f viral-specific IgA antibodies. Journal o f the American Medical Association 1991;266:3439-3442. 53. Wol insky S, W i k e C , Korber B , et al. Selective transmission o f human immunodeficiency virus type 1 variants from mothers to infants. Science 1992,255:1134-1136. 54. Ryder R, Hassig S. The epidemiology o f perinatal transmission o f H I V - 1 . A I D S 1988;2:83-89. 55. Goedert J, Mendez H , Drummond J. Mother to infant transmission o f human immunodeficiency virus type 1: association with prematurity and low anti-gpl20. Lancet 1989;2:1351-1354. 56. Devash Y , Cavel l i T, W o o d D , et al. Vertical transmission of HIV-1 is correlated with the absence of high affinity/avidity maternal antibodies to g p l 2 0 principal neutralizing domain. Proc. Nat l . Acad . Sci . U S A 1990;87:3445-3448. 57. Halsey N , Markham R, Wahren B , et al. Lack of association between maternal antibodies to V 3 loop peptides and maternal-infant HIV-1 transmission. Journal o f Acquired Immune Deficiency Syndromes 1992;5:153-157. 58. LaPointe N , Michaud J, Pekovic D, et al. Transplacental transmission o f H T L V - I I I virus. New England Journal o f Medicine 1985;312:1325-1326. 59. Jovaisas E, K o c h M A , Achafer A , e ta l . L A V / H T L V - I I I in 20-week fetus. Lancet 1985;2:1129. 60. Sprecher S, Soumenkoff G , Puissant F, et al. Vertical transmission of HIV-1 in 15-week fetus. Lancet 1986;2:288-289. 61. Mano H , Chermann J C . Fetal human immunodeficiency virus type 1 infection of different organs in the second trimester. A I D S Research and Human Retroviruses 1991;7:83-88. 62. Ryder R W , Nsa W , Hassig S E , et al. Perinatal transmission o f the human immunodeficiency virus type 1 in infants o f seropositive women in Zaire. N e w England Journal o f Medic ine 1989,320.1637-1642. 63. European Collaborative Study. Risk factors for mother-to-child transmission o f H I V - 1 . Lancet 1992;339:1007-1012. 64. Italian Multicentre Study. Epidemiology, clinical features and prognostic factors o f paediatric HIV-1 infection. Lancet 1988;2:1043-1046. 65. Weiblen B - J , Lee F K , Cooper E R , et al. Early diagnosis o f HIV-1 infection in infants by detection of IgA HIV-1 antibodies. Lancet 1990;335:988-990. 66. Rossi P, Moschese V , Brol iden P A , et al. Presence o f maternal antibodies to human immunodeficiency virus 1 envelope glycoprotein gp 120. epitopes correlates with the uninfected status o f children born to seropositive mothers. Proc. Natl . Acad . Sc i . U S A 1989;86:8055-8058. 67. Blanche S, Rouzioux C , Guihard Moscato M - L , et al. A prospective study o f infants born to women seropositive for human immunodeficiency virus type 1. New England Journal o f Medicine 1989;320:1643-1648. 89 68. K i n d C , Brandle B , Wyler C A , et al. Epidemiology o f vertically transmitted HIV-1 infection in Switzerland: results from a nation-wide prospective study. European Journal o f Pediatrics 1992;151:442-448. 69. Hutto C , Parks W , La i S, et al. A hospital-based prospective study o f perinatal infection with human immunodeficiency virus type 1. Journal o f Pediatrics 1991;118:347-353. 70. Martin N , L e v y J A , Legg H , et al. Detection o f infection with human immunodeficiency virus ( H I V -1) type 1 in infants by an anti-HIV-1 immunoglobulin A assay using recombinant proteins. Journal o f Pediatrics 1991;1 18:354-358. 71. Parekh B S , Shaffer N , Pau C - P , et al. Lack o f correlation between maternal antibodies to V 3 loop peptides o f gp 120 and perinatal H I V - 1 transmission. A I D S 1991;5:1179-1184. 72. Van de Perre P, Simonon A , Msellat i P, et al. Postnatal transmission o f human immunodeficiency virus type 1 from mother to infant: a prospective cohort study in K i g a l i , Rwanda. N e w England Journal o f Medicine 1991;325:593-598. 73. Wor ld Health Organization. A I D S Epidemic Update: December 2000. 74. Oxtoby M . Perinatally acquired human immunodeficiency virus infection. Pediatric Infectious Diseases Journal 1990;9:609-619. 75. European Collaborative Study. Children bom to women with HIV-1 infection: natural history and risk o f transmission. Lancet 1991;337:253-260. 76. Soiero R, Rubinstein A , Rasbaum W K , et al . Maternofetal transmission o f A I D S : frequency o f human immunodeficiency virus type 1 nucleic acid sequences in human fetal D N A . Journal o f Infectious Diseases 1992;166:699-703. 77. H i l l W C , Boton V , Carlson RJ . Isolation of acquired immunodeficiency syndrome virus from the placenta. American Journal o f Obstetrics and Gynecology 1987;157:10-11. 78. Chandwani S, Greco M A , Mittal K , et al. Pathology and human immunodeficiency virus expression in placentas of seropositive women. Journal o f Infectious Diseases 1991;163:1134-1138. 79. Mattern C F T , Murray K , Jensen A , et al. Localization of human immunodeficiency virus core antigen in term human placentas. Pediatrics 1992;89:207-209. 80. Douglas, G C , Fry C N , T h i r k i l l T, et al. Cell-mediated infection o f human placental trophoblast with HIV-1 in vitro. A I D S Research and Human Retroviruses 1991;7:735-740. 81. Mano H , Cherman JC. Replication o f human immunodeficiency virus type 1 in primary cultured placental cells. Research in Vi ro logy 1991;142:95-104. 82. Lyman W D , Kress Y , Kure K , et al. Detection o f HIV-1 in fetal central nervous system tissue. A I D S 1990;4:917-920. 83. Lindgren S, Anzen B , Boh l in A B , et al. HIV-1 and child-bearing cl inical outcome and aspects o f mother-to-infant transmission. A I D S 1991;5:1111-1116. 84. Bryson Y J , Luzuriaga K , Sull ivan J L , et al. Proposed definition for in utero versus intrapartum transmission o f HIV-1 [letter]. N e w England Journal o f Medicine 1992;327:1246-1247. 90 85. Blanche S, Tardieu M , Duliege A M , et al . Longitudinal study o f 94 symptomatic infants with perinatally acquired immunodeficiency virus infection: evidence for a bimodal expression of cl inical and biological symptoms. American Journal o f Diseases o f Children 1990;144:1210-1215. 86. Thiry L , Sprecher-Goldberger S, Jonckheer T, et al. Isolation o f A I D S virus from cell-free breast milk of three healthy carriers. Lancet 1985;2:891-897. 87. Zeigler J V , Cooper D A , Johnson R O , et al. Post-natal transmission of AIDS-associated retrovirus from mother to infant. Lancet 1985;1:896-897. 88. Colebunders R, Kapita B , Nekwei W , et al. Breastfeeding and transmission o f H I V - 1 [letter]. Lancet 1988;1:1487. 89. Weinbreck P, Loustaud V , Denis F, et al. Postnatal transmission of HIV-1 infection [letter]. Lancet 1988;1:482. 90. Stiehm E R , V i n k P. Transmission o f human immunodeficiency virus infection by breast feeding. Journal o f Pediatrics 1991;118:410-412. 91. Rouzioux C , Costagliola D , Burgard M , et al. T iming o f mother-to-child H I V - 1 transmission depends on maternal status. A I D S 1993;7(supplement 2):S49-S52. 92. Newel l M - L and Peckham C. R ick factors for vertical transmission of HIV-1 and early markers o f HIV-1 infection in children. A I D S 1993;7(supplement 1):S91-S97. 93. Brossard Y , A u b i n J-T, Mandelbrot L , et al. Frequency o f early in utero H I V - 1 infection: a blind D N A polymerase chain reaction study on 100 fetal thymuses. A I D S 1995;9:359-366. 94. The European Collaborative Study. Vertical transmission of HIV-1 maternal immune status and obstetric factors. A I D S 1996;10:1675-1681. 95. Scarlatti G , Lombard! V , Plebani A , et al. Polymerase chain reaction, virus isolation and antigen assay i n H l V - 1 antibody positive mothers and children. A I D S 1991;5:1173-1178. 96. Rossi P. Sienna Consensus Workshop. Maternal factors involved in mother-to-child transmission o f H I V - 1 . Journal o f Acqui red Immune Deficiency Syndromes 1992;5:1019-1029. 97. Tovo, P - A , deMartino M , Gabiano C , G a l l i L , Capello N , et al. Mode o f delivery and gestational age influence perinatal HIV-1 transmission. Journal o f Acquired Immune Deficiency Syndrome 1996;11:88-94. 98. Johnson M A , Cann A J . Molecular determination of cell tropism o f human immunodeficiency virus. Cl in ica l Infectious Diseases 1992;14:747-755. 99. Yarchoan R, Mitsuya H , Broder S. The immunology o f HIV-1 infection: implications for therapy. A I D S Research and Human Retroviruses 1992;8:1023-1031. 100. Robertson C A , M o k J Y Q , Froebel K S , et al. Maternal antibodies to gp 120 V 3 sequence do not correlate with protection against vertical transmission o f human immunodeficiency virus. Journal o f Infectious Diseases 1992;166:704-709. 101 .Schnittman S M , Donnury S M , Greenhouse JJ, et al. Evidence o f susceptibility o f intrathymic T cell precursors and their progeny carrying T-cells receptor phenotypes T C R alpha/beta+ and T C R 91 gamma/delta+ to human immunodeficiency virus infection: a mechanism for C D 4 + (T4) lymphocyte depletion. Proc. Na t l . Acad . Sc i . U S A 1990;87:7727-7731. 102. Luzuriaga K , Koup R A , Pikora C A , et al. Deficient human immunodeficiency virus type 1 specific cytotoxic T cell responses in vertically infected children. Journal o f Pediatrics 1991;119:230-236. 103. Cler ic i M , Sison A V , Berzofsky J A , Rakusan T A , Cellular immune factors associated with mother-to-infant transmission of HIV-1 A I D S 1993;7:1427-1433. 104. Borkowsky W , Rigaud M , Krasinski K , et al. Ce l l mediated and humoral immune responses in children infected with human immunodeficiency virus during the first 4 years o f life. Journal o f Pediatrics 1992;120:371-375. 105. Ljunggren K , Moschese V , Broliden P A , et al. Antibodies mediating cellular cytotoxicity and neutralization correlate with a better clinical stage in children born to human immunodeficiency virus-infected mothers. Journal o f Infectious Diseases 1990;161:198-202. 106. Brenner TJ , Dahl K E , Olson B , et al. Relation between HIV-1 syncytium inhibition antibodies and clinical outcome in children. Lancet 1991;337:1001-1005. 107. Ammann A , Bohme P, Borkowsky W , et al. Maternal factors involved in mother-to-child transmission of H I V - 1 : report o f the consensus workshop held in Sienna, Italy, Jan 17-18, 1992. Journal o f Acquired Immune Deficiency Syndromes 1992;5:1019-1029. 108. ViscarelIo R, DeGennaro N , Andiman W . Maternal C D 4 lymphocyte counts as a predictor o f perinatal transmission of HIV-1 [abstract], American Journal o f Obstetrics and Gynecology 1993;168:418. 109. Mayers M M , Davenny K , Schoenbaum E E , et al. A prospective study o f infants o f human immunodeficiency virus seropositive and seronegative women with a history o f intravenous drug use or o f intravenous drug-using sex partners in the Bronx, N e w Y o r k Ci ty . Pediatrics 1991 ;88:1248-1256. 1 lO.Gabiano C , Tovo P - A , deMartino M , et al. Mother-to-child transmission o f human immunodeficiency virus type 1: risk of infection and correlates o f transmission. Pediatrics 1992;90:369-374. 111 .Horsburgh C R , O u C - Y , Joason J, et al. Duration of human immunodeficiency virus infection before detection o f antibody. Lancet 1989;2:637-340. 112. Rogers M F , O u C - Y , Rayfield M , et al. Use of the polymerase chain reaction for early detection o f the proviral sequences o f human immunodeficiency virus in infants born to seropositive mother. New England Journal o f Medicine 1989;320:1649-1654. 113. European Collaborative Study. Mother to child transmission of HIV-1 infection. Lancet 1988;2:1039-1043. 114. Johnson JP, Na i r P, Hines S E , et al. Natural history and serologic diagnosis o f infants born to human immunodeficiency virus-infected women. American Journal o f Diseases o f Children 1989; 143:1147-1153. 115. Aui t i F, L u z i G , Mezzaroma I, Scano G , Papetti C . Delayed appearance o f HIV-1 infection in children. Lancet 1987;2:858. 116. Johnson JP, Nai r P, Alexander S. Early diagnosis o f H I V - 1 infection in the neonate [letter]. N e w England Journal o f Medicine 1987;316:273-274. 92 117. Landesman S, Weiblen B , Mendez H , et al. Cl in ica l utility o f HIV-1 - IgA immunoblot assay in the early diagnosis o f perinatal H I V - 1 infection. Journal o f the American Medica l Association 1991;266:3443-3446. 118. Borkowsky W , Krasinski K , Paul D , et al. Human immunodeficiency virus type 1 antigenemia in children. Journal o f Pediatrics 1989;114:940-945. 119. Borkowsky W , Krasinski K , Paul D , Moore T, Bebenroth D, Chandwani S, Human-immunodeficiency-virus infections in infants negative for anti-HIV-1 by enzyme-linked immunoassay. Lancet 1987;1:1168-1171. 120. Chaisson R E , A l l a i n J, Volberding P A . Significant changes in HIV-1 antigen level in the serum o f patients treated with azidothymidine. N e w England Journal o f Medicine 1986;315:1610-1611. 121 .Fahey J L , Taylor J M G Detels R, et al. The prognostic value o f cellular and serologic markers in infection with human immunodeficiency virus type 1. New England Journal o f Medicine 1990;322:166-172. 122. Palomba E , Gay V , deMartino M , Fundaro C , Perugini L , Tovo P - A . Early diagnosis o f human immunodeficiency virus infection in infants by detection o f free and complexed p24 antigen. Journal of Infectious Diseases 1992;165:394-395. 123. Miles S A , Balden E , Magpantay L , et al. Rapid serologic testing with immune-complex-dissociated HIV-1 p24 antigen for early detection o f HIV-1 infection in neonates. N e w England Journal o f Medicine 1993;328:297-302. 124. Hollinger F B , ed. A C T G Vi ro logy Manual for HIV-1 laboratories, Bethesda: National Institutes of Health September, 1994. 125. Burgard M , Mayaux M - J , Blanche S, et al. The use o f viral culture and p24 antigen testing to diagnose human immunodeficiency virus infection in neonates. N e w England Journal o f Medicine 1992;327:1192-1197. 126. Jones D S , Abrams E , Ou C - Y , et al. The lack of detectable human immunodeficiency virus (HIV-1) infection in antibody-negative children born to HIV- l - in fec ted mothers. Pediatric Infectious Diseases Journal 1993;12:222-227. 127. Ho D D , Moudg i l T, A l a m M , Quantitation o f human immunodeficiency virus type 1 in the blood of infected persons. N e w England Journal o f Medicine 1989;321:1621-1625. 128. Rossi P. Early diagnosis o f H I V - 1 infection in infants. Consensus Workshop, Siena, January, 1992. Journal o f Acquired Immune Deficiency Syndrome 1992;5:1169-1172. 129. Tersmette M , Gruters R A , d e W o l f F, et al. Evidence for a role o f virulent human immunodeficiency virus (HIV-1) variants in the pathogenesis o f acquired immunodeficiency syndromes: studies on sequential HIV-1 isolates. Journal o f Vi ro logy 1989;63:2118-2125. 130. Rubinstein A , Bernstein L . The epidemiology of pediatric acquired immunodeficiency syndrome. Cl in ica l Immunology and Immunopathology 1986;40:115-121. 131 .Falloon J , Eddy J , Wiener L , et al . Human immunodeficiency virus infection in children. Journal o f Pediatrics 1989;114:1-23. 93 132. M c K i n n e y R E , Walter E B , Tudor-Wil l iams G , et al. C D 4 + lymphocytes in children under 2 years old: establishment o f normal values in a population at risk for HIV-1 and prognostic application to infected children [abstract]. Pediatric Research 1991 ;29:178A. 133. The National Institute of C h i l d Health and Human Development Intravenous Immunoglobulin Study Group. Intravenous immune globulin for the prevention o f bacterial infections in children with symptomatic human immunodeficiency virus infection. New England Journal o f Medicine 1991;325:73-80. 134. Krasinski K , Borkowsky W , Bonk S, et al. Bacterial infections in human immunodeficiency virus-infected children. Pediatric Infectious Diseases Journal 1988;7:323-328. 135. Horsburgh C R . Mycobacterium avium complex infection in the acquired immunodeficiency syndrome. N e w England Journal o f Medic ine 1991;324:1332-1338. 136. Hughes W T . Pneumocystis carinii pneumonia: new approaches to diagnosis, treatment and prevention. Pediatric Infectious Diseases Journal 1991;10:391-399. 137. Kovacs A , Frederick T, Church J, et al. C D 4 T-lymphocytes and Pneumocystis carinii pneumonia in pediatric HIV-1 infection. Journal o f the American Medical Association 1991;265:1698-1703. 138. Connor E , Bagarazzi M , McSherry G , et al. C l in ica l and laboratory correlates o f Pneumocystis carinii pneumonia in children infected with H I V - 1 . Journal o f the American Medica l Association 1991;265:1693-1697. 139. Peterson J, Church J, Gomperts E , et al. Lymphocyte phenotype does not predict immune function in pediatric patients infected with the human immunodeficiency virus type 1. Journal o f Pediatrics 1989;1 15:944-948. 140. Frenkel L D , Gaur J G , Shah K D , et al. Cytomegalovirus infection in children with A I D S . Review of Infectious Diseases 1990; 12: S820-S826. 141 . L i m W , Kahn E , Gupta A , et al. Treatment o f cytomegalovirus enterocolitis with ganciclovir in an infant with acquired immunodeficiency syndrome. Pediatric Infectious Diseases Journal 1988;7:354-357. 142. Levin A V , Zeichner S, Duker JS, et al. Cytomegalovirus retinitis in an infant with acquired immunodeficiency syndrome. Pediatrics 1989;84:683-687. 143. Jura E , Chadwick E G , Josephs S H , et al. Varicella-zoster virus infections in children infected with human immunodeficiency virus. Pediatric Infectious Diseases Journal 1989;8:586-590. 144. Pahwa S, Biron K L i m W , et al. Continuous varicella-zoster infection associated with acyclovir resistance in a chi ld with A I D S . Journal o f the American Medica l Association 1988;260:2879-2882. 145.Safrin S, Assaykeen T, Follansbee S, et al. Foscarnet therapy for acyclovir-resistant mucocutaneous herpes simplex virus infection in 26 A I D S patients: preliminary data. Journal o f Infectious Diseases 1989;160:1078-1084. Or 1990;161:1078-1084. 146.Safrin S, Crumpacker C , Chatis P, et al. A controlled trial comparing foscarnet with vidarabine for acyclovir-resistant mucocutaneous herpes simplex in the acquired immunodeficiency syndrome. New England Journal o f Medicine 1991;325:551-555. 147.Chandwani S, Borkowsky W , Krasinski K , et al. Respiratory syncytial virus infection in human immunodeficiency virus-infected children. Journal o f Pediatrics 1990; 117:251-254. 94 148. Janner D , Petrie A M , Belchis D , et al. Fatal adenovirus infection in a chi ld with acquired immunodeficiency syndrome. Pediatric infectious Diseases Journal 1990;9:434-436. 149. Josephs S, K i m H W , Brandt C D , et al. Parainfluenza 3 virus and other common respiratory pathogens in children with human immunodeficiency virus infection. Pediatric Infectious Diseases Journal 1988;7:207-209. 150. Markowitz L E , Chandler F W , Roldan E O , et al. Fatal measles pneumonia without rash in a child with A I D S . Journal o f Infectious Diseases 1988;158:480-483. 151. Leibovitz E , Rigaud M , Chandwani S, et al. Disseminated fungal infections in children infected with human immunodeficiency virus. Pediatric Infectious Diseases Journal 1991;10:888-894. 152. Mitchel l C D , Er l ich SS, Mastrucci M T , et al. Congenital toxoplasmosis occurring in infants infected with acquired immunodeficiency virus 1. Pediatric Infectious Diseases Journal 1990;9:512-518. 153. Cohen-Addad N , Joshi V , Shares L , et al. Congenital acquired immunodeficiency syndrome and congenital toxoplasmosis: pathologic support for a chronology o f events. Journal o f Perinatology 1988;3:328-331. 154.Shanks G D , Redfield R R , Fisher G W . Toxoplasma encephalitis in an infant with acquired immunodeficiency syndrome. Pediatric Infectious Diseases Journal 1987;6:70-71. 155. Borowitz S M , Saulsbury FT . Treatment of chronic cryptosporidial diarrhea with orally administered human serum immune globulin. Journal o f Pediatrics 1991;118(4?):593-595. 156. Rubinstein A , M o r e c k i R, Silverman B , et al. Pulmonary disease in children with acquired immune deficiency syndrome and AIDS-related complex. Journal o f Pediatrics 1986;108:498-503. 157. Belman A L , Diamond G , Dickson D , et al. Pediatric acquired immunodeficiency syndrome: neurologic syndromes. American Journal o f Diseases o f Children 1988;142:29-35. 158. Epstien L G , Sharer L R , Oleske J M , et al. Neurologic manifestations o f human immunodeficiency virus infection in children. Pediatrics 1987;78:678-687. 159. Belman A L , Ultmann M H , Horoupian P, et al. Neurological complications in infants and children with acquired immune deficiency syndrome. Annals o f Neurology 1985;18:560-566. 160. Davis S L , Halstead C C , Levy N , et al. Acquired immune syndrome presenting as progressive infantile encephalopathy. Journal o f Pediatrics 1987; 110:884-888. 161 .Shaw G M , Harper M E , Hahn B , et al. H T L V - I I I infection in brains o f children and adults with A I D S encephalopathy. Science 1985;227:177-182. 162. Resnick L , deMarzo-Veronese F, Schupbach J, et al. Intrablood-brain barrier synthesis o f H T L V - I I I -specific IgG in patients wi th neurologic symptoms associated with A I D S or AIDS-related complex. New England Journal o f Medicine 1985;313:1498-1504. . 163. M i n t z M , Rapaport R, Oleske J M , et al. Elevated serum levels o f tumor necrosis factor are associated with progressive encephalopathy in children with acquired immunodeficiency syndrome. American Journal o f Diseases o f Children 1989;143:771-774. 164. Belman A L , Lantos G , Horoupian D, et al. A I D S : calcification of the basal ganglia in infants and children. Neurology 1986;36:1192-1199. 95 165. Cohen S E , Mundy T, Karassik B , et al . Neuropsychological functioning in human immunodeficiency virus type 1 seropositive children infected through neonatal blood transfusion. Pediatrics 1991 ;88:58-68. 166. Condini A , A x i a G , Cattelan C , et al. Development of language in 18-30 month-old HIV- l - in fec ted but not i l l children. A I D S 1991;5:735-739. 167. Dickson D W , Belman A L , K i m T S , et al. Spinal cord pathology in pediatric acquired immunodeficiency syndrome. Neurology 1989;39:227-235. 168. Epstein L G , Sharer L R , Joshi V V , et al. Progressive encephalopathy in children with acquired immune deficiency syndrome. Annals o f Neurology 1985;17:488-496. 169. Epstein L G , Goudsmit J, Paul D A , et al. Expression of human immunodeficiency virus in cerebrospinal fluid o f children with progressive encephalopathy. Annals o f Neurology 1987;21:397-401. 170. Epstein L G , Sharer L R , Goudsmit J. Neurological and neuropathological features o f human immunodeficiency virus infection in children. Anna l o f Neurology 1988;23 (supplement):S19-S23. 171.Selmaj K W , Raine C S . T N F mediates myelin and oligodendrocyte damage in vitro. Annals o f Neurology 1988;23:339-346. 172. M c K i n n e y R E , Pizzo P A , Scott G B , et al. Safety and tolerance o f intermittent intravenous and oral zidovudine therapy in human immunodeficiency virus-infected pediatric patients. Journal o f Pediatrics 1990; 116:640-647. 173. M c K i n n e y R E , Maha M A , Connor E M , et al. A multicentre trial o f oral zidovudine in children with advanced human immunodeficiency virus disease. N e w England Journal o f Medic ine 1991 ;324:1018-1025. 174. Brouwers P, Moss H , Wolters P, et al. Effect o f continuous-infusion zidovudine therapy on neuropsychologic functioning in children with symptomatic human immunodeficiency infection. Journal o f Pediatrics 1990;117:980-985. 175. Decarli C , Fugate L , Falloon J, et al. Brain growth and cognitive improvement in children with human immunodeficiency virus-induced encephalopathy after 6 months o f continuous infusion zidovudine therapy. Journal o f the Acquired Immune Deficiency Syndrome 1991;4:585-592. 176. Epstein L G , Decarlo FJ , Joshi V V , et al. Primary lymphoma o f the central nervous system in children with acquired immunodeficiency syndrome. Pediatrics 1988;82:355-363. 177. Andiman W A , Eatsman R , Mart in K , et al. Opportunistic lymphoproliferations associated with Epstein-Barr viral D N A in infants and children with A I D S . Lancet 1985;2:1390-1393. 178. Cocchi P, Calabri G , Sa lv i G , et al. AlDS-associated lymphoma o f the brain in a child. Pediatrics 1988;82:678-679. 179. Principi N , Marchisio P, Tornaghi R, et al. Acute otitis media in human immunodeficiency virus-infected children. Pediatrics 1991;88:566-571. 180. Hass A , Hyatt A C , Kattan M , et al. Hoarseness in immunocompromised children: association with invasive fungal infection. Journal o f Pediatrics 1987; 1 1:731-733. 96 181 .Strauss J, Abi tbo l C , Zil leruelo G , et al. Renal disease in children with the acquired immunodeficiency syndrome. N e w England Journal o f Medicine 1989;321:625-630. 182.Lipshultz S E , Chanock S, Sanders SP, et al. Cardiovascular manifestations o f human immunodeficiency virus infection in infants and children. American Journal o f Cardiology 1989;63:1489-1497. 183.Stewart J M , K a u l A , Gromish D S , et al. Symptomatic cardiac dysfunction in children with human immunodeficiency virus infection. American Heart Journal 1989; 117:140-144. 184.1ssenberg H J , Charytan M , Rubinstein A . Cardiac involvement in children with acquired immune deficiency [abstract]. American Heart Journal 1985; 110:710. 185. Medlock M D , T i l l e l i JT, Pearl G S . Congenital toxoplasmosis in a newborn with acquired immunodeficiency syndrome. Pediatric Infectious Diseases Journal 1990;9:129-132. 186. Lipshultz SE , Fox C H , Perez-Atayde A R , et al. Identification of human immunodeficiency virus-1 R N A and D N A in the heart o f a chi ld with cardiovascular abnormalities and congenital acquired immune deficiency syndrome. American Journal o f Cardiology 1990;66:246-250. 187. Grody W W , Cheng L , Lewis W . Infection o f the heart by the human immunodeficiency virus. American Journal o f Cardiology 1990;66:203-206. 188. K.avanaugh-McHugh A L , R u f f A J , Rowe S A , et al. Cardiac abnormalities in a multicentre interventional study o f children with symptomatic HIV-1 infection [abstract]. Pediatric Res. 1991;29:176A. 189. Chadwick E , Schmidt M L , Yogev R , Characteristics o f autoimmune hemolytic anemia in H I V - l -infected children [abstract]. Pediatric Res. 1991:29:168A. 190. Ellaurie M , Burns E R , Berstein L J , et al. Thrombocytopenia and human immunodeficiency virus in children. Pediatrics 1988;82:905-908. 191. Saulbury FT , Boyle R J , W y k o f f R F , et al. Thrombocytopenia as the presenting manifestation o f human T-lymphotropic virus type III infection in infants. Journal o f Pediatrics 1986;109:30-34. 192. Montaner JS , L e T , Fanning M , et al. The effect o f zidovudine on platelet count in H I V - 1 infected individuals. Journal o f Acqui red Immune Deficiency Syndrome 1990;3:565-570. 193. Fischl M A , Richman D D , Causey D M , et al. Prolonged zidovudine therapy in patients with A I D S and advanced AIDS-related complex. Journal o f the American Medica l Associat ion 1989;262:2405-2410. 194. The Swiss Group for C l in i ca l Studies on the Acquired Immunodeficiency Syndrome ( A I D S ) . Zidovudine for the treatment o f thrombocytopenia associated with human immunodeficiency virus. A prospective study. Annals o f Internal Medicine 1988;109:718-721. 195. Rigaud M , Leibovi tz E , Sinquee C , et al. Thrombocytopenia in children infected with human immunodeficiency virus (HIV-1) : long term follow-up and therapeutic considerations. Journal o f Acquired Immune Deficiency Syndromes 1992;5:450-455. 196. Lambert G , Arpadi S, W i z n i a A , et al. Zidovudine therapy in thrombocytopenic human immunodeficiency virus-infected children. Pediatric Infectious Diseases Journal 1991; 10:875. 97 197. Horowitz M E , Pizzo P A . Cancer in the child infected with the human immunodeficiency syndrome [editorial]. Journal o f Pediatrics 1990;116:730-731. 198. Montalvo F W , Casanova R, Clave l L A . Treatment outcome in children with malignancies associated with human immunodeficiency virus infection. Journal o f Pediatrics 1990; 116:735-738.' 199. Chadwick E G , Connor E J , Hanson I C G , et al. Tumors of smooth-muscle origin in HIV- l - in fec ted children. Journal o f the American Medical Association 1990;263:3182-3184. 200. Weinblatt M , Scimeca P, James-Herry A . Cancers in children infected with human immunodeficiency virus. Journal o f Pediatrics 1991 ;118:163. 201. Buck B E , Scott G B , Valdez-Dapena M , et al. Kaposi ' s sarcoma in two infants with acquired immune deficiency syndrome. Journal o f Pediatrics 1983;103:911-913. 202. Balis F M , Pizzo P A , Eddy J, et al. Pharmacokinetics o f zidovudine administered intravenously and orally in children with human immunodeficiency virus infection. Journal o f Pediatrics 1989; 114:880-884. 203. Pizzo P A , Eddy J, Falloon J , et al. Effect o f continuous intravenous infusion o f zidovudine ( A Z T ) in children with symptomatic HIV-1 infection. New England Journal o f Medicine 1988;319:889-896. 204. Butler K M , Husson R N , Bal is F M , et al. Dideoxyinosine in children with symptomatic human immunodeficiency virus infection. New England Journal o f Medicine 1991;324:137-144. 205. Pizzo P A . Treatment o f human immunodeficiency virus-infected infants and young children with dideoxynucleosides. American Journal o f Medicine 109;88(suppl 5B) : 16S-19S. 206.Scott G B , Knupp C A , Doyle M G , et al. A phase I trial o f dideoxyinosine (ddl) in pediatric HIV-1 infection: cerebrospinal fluid and plasma pharmacokinetics [abstract]. Pediatric Research 1991;29:184A. 207. Brooks Jackson J, Coombs R W , Sannerud K , et al. Rapid and sensitive viral culture method for human immunodeficiency virus type 1. Journal o f Cl in ica l Microbio logy 1988;26(7): 1416-1418. 208. Hollinger F B , Bremer J W , Myers L , et al. Standardization o f sensitive human immunodeficiency virus culture procedures and establishment of a multicenter quality assurance program for the A I D S clinical trials group. Journal o f Cl in ica l Microbiology 1992;30(7): 1787-1794. 209. Japour A J , Fiscus S, Arduino J M , et al. Standardized microtitre assay for determination o f syncytium-inducing phenotypes o f cl inical human immunodeficiency virus type 1 isolates. Journal o f Cl in ica l Microbiology 1994;32:292-300. 210. Compton J. Nucleic acid sequence-based amplification. Nature 1991 ;350:91-92. 21 l.Steketee R W , Abrams E J , Thea D M , et al. Early detection o f perinatal human immunodeficiency virus (HIV-1) type 1 infection using HIV-1 R N A amplification and detection. The Journal o f Infectious Diseases 1997;175:707-711. 212.Japour A J , Mayers D L , Johnson V A , et al. Standardized peripheral blood mononuclear cell culture assay for determination o f drug susceptibilities o f clinical human immunodeficiency virus type 1 isolates. Ant imicrobia l Agents and Chemotherapy 1993;37(5): 1095-1101. 98 213. D ' A q u i l a R T , Johnson V A , Welles S L , et al. Zidovudine resistance and HIV-1 disease progression during antiretroviral therapy. Annals o f Internal Medicine 1995; 122(6):401-408. 214. Japour A J , Welles S, D ' A q u i l a R T , et al. Prevalence and clinical significance o f zidovudine resistance mutations in human immunodeficiency virus isolated from patients after long-term zidovudine treatment. The Journal o f Infectious Diseases 1995;171:1 172-1179. 215. Larder B A , Kemp S D . Mul t ip le mutations in HIV-1 reverse transcriptase confer high-level resistance to zidovudine ( A Z T ) . Science 1989;246:1155-1158. 216. Larder B A , K e l l a m P, K e m p S D . Zidovudine resistance predicted by direct detection of mutations in D N A from HIV-1-infected lymphocytes. A I D S 1991 ;5(2): 137-144. 217. Richman D D , Guatelli JC , Grimes J, et al. Detection o f mutations associated with zidovudine resistance in human immunodeficiency virus by use of the polymerase chain reaction. The Journal o f Infectious Disease 1991;164:1075-1081. 218. Frenkel L M , Wagner L E , A twood S M , et al. Specific, sensitive, and raid assay for human immunodeficiency virus type 1 pol mutations associated with resistance to zidovudine and didanosine. Journal o f Cl in ica l Microbio logy 1995;33(2):342-347. 219. Nielsen K , Wei L S , S im M S , et al. Correlation of clinical progression in human immunodeficiency virus-infected children with in vitro zidovudine resistance measured by a direct quantitative peripheral blood lymphocyte assay. The Journal o f Infectious Diseases 1995;172:359-364. 220. Thomas P, Singh T, Wi l l i ams R, et al. Trends in survival for children reported with maternally transmitted acquired immunodeficiency syndrome in N e w Y o r k C i ty , 1982-1989. Pediatric Infectious Diseases Journal 1992; 11:34-39. 221. DeMartino P M , and the Italian Register for HIV-1 Infection in Children. Features of children perinatally infected with HIV-1 surviving longer than 5 years. Lancet 1994;343:191-195. 222. Wi ley C A , Schrier R D , Nelson J A , et al. Cellular localization o f human immunodeficiency virus infection within the brains o f acquired immune deficiency syndrome patients. Proc. Nat l . Acad . Sc i . U S A 1986;83:7089-7093. 223. Koenig S, Gendelman H , Orenstein J, et al. Detection o f A I D S virus in macrophages in brain tissue from A I D S patients with encephalopathy. Science 1986;233; 1089-1093. 224. Revision o f the C D C surveillance case definition for acquired immunodeficiency syndrome. M M W R 1987;36:3S-15S. 225.Spector S A , Gelber R D , McGra th N , et al. A controlled trial o f intravenous immune globulin for the prevention of serious bacterial infections in children receiving zidovudine for advanced human immunodeficiency virus infection. N e w England Journal o f Medicine 1994; 331:1181-1187. 226. Mofenson L M , M o y e J Jr., Bethel J, et al. Prophylactic intravenous immunoglobulin in H I V - l -infected children with C D 4 + counts o f 0.20x10 9 /Lor more: effect on viral , opportunistic, and bacterial infections. Journal o f the American Medica l Association 1992;268:483-488. 227. Blanche S, Mayaux M - J , Rouzioux C , et al. Relation o f the course of H I V - 1 infection in children to the severity of the disease in their mothers at delivery. N e w England Journal o f Medicine 1994;330:308-312. 99 228. McGlaugh l in G E , Virdee SS, Schleien C L , et al. Effect o f corticosteroids on survival o f children with acquired immunodeficiency syndrome and Pneumocystis carinii-rdated respiratory failure. Journal o f Pediatrics 1995;126:821-824. 229. National Institutes o f Health-University o f California Expert Panel for Corticosteroids as Adjunctive Therapy for Pneumocystis carinii. Consensus statement on the use o f corticosteroids as adjunctive therapy for Pneumocystis pneumonia in the acquired immunodeficiency syndrome. N e w England Journal o f Medicine 1990;323:1500-1504. 230.Sleasman J W , Hemenway C , K l e i n A S , et al. Corticosteroids improve survival o f children with A I D S and Pneumocystis carinii pneumonia. American Journal o f Diseases o f Children 1993;147:30-34. 231.Simonds R J , Oxtoby M J , Caldwel l B , et al. Pneumocystis carinii pneumonia among U S children with perinatally acquired H I V - 1 infection. Journal o f the American Medica l Associat ion 1993;270:470-473. 232. Centers for Disease Control . Guidelines for prophylaxis against Pneumocystis carinii pneumonia for children infected with the human immunodeficiency virus. M M W R 1991;40(RR-2):1-13. 233. Thea D M , Lambert G , Weedon J, et al. Benefit o f primary prophylaxis before 18 months o f age in reducing the incidence o f Pneumocystis carinii pneumonia and early death in a cohort o f 112 human immunodeficiency virus-infected infants. Pediatrics 1996;97(l):59-64. 234. Connor E , Gupta S, Joshi V , et al. Acqui red immunodeficiency syndrome associated renal disease in children. Journal o f Pediatrics 1988;113:39-44. 235. Mitsuya H , Broder S. Inhibition o f the in vitro infectivity and cytopathic effect o f human T-lymphotrophic virus type III/Iymphadenopathy associated virus ( H T L V - I I I / L A V ) . Proc. Natl . Acad. Sci . U S A 1986;83:1911-1915. 236. Fischl M A , Richman D D , Grieco M H , et al. The efficacy o f azidothymidine ( A Z T ) in the treatment of patients with A I D S and AIDS-related complex: a double-blind placebo-controlled trial. N e w England Journal o f Medicine 1987;317:185-191. 237. Mitsuya H , Weinhold K J , Furman P A , et al. 3 ' -Azido-3 ' -deoxythymidine ( B W A 5 0 9 U ) : an antiviral agent inhibits the infectivity and cytopathic effect o f human T-lymphotropic virus type III/lymphadenopathy-associated virus in vitro. Proc. Natl . Acad . Sci . U S A 1985;82:7096-7100. 238. Pinching A J , Helbert M , Peddle B , et al. Cl in ica l experience with zidovudine for patients with acquired immune deficiency syndrome and acquired immune deficiency syndrome-related complex. Journal o f Infection 1989;18:33-40. 239.Stambuk D , Hawkins D , Gazzard B G . Zidovudine treatment o f patients with acquired immune deficiency syndrome and acquired immune deficiency syndrome-related complex: St. Stephen's Hospital experience. Journal o f Infection 1989;18:41-51. 240.Yarchoan R, Thomas R V , Grafman J, et al. Long-term administration of 3 ' -azido-2 ' ,3 ' -dideoxythymidine to patients with AIDS-related neurologic disease. Annals o f Neurology 1988;23(suppl):S82-S87. 241 .Chaisson R E , Leuther M D , A l l a i n JP et al. Effect o f zidovudine on serum human immunodeficiency virus core antigen levels. Results from a placebo-controlled trial. Archives o f Internal Medicine 1988;148:2151-2153. 100 242.Spector S A , Kennedy C , McCutchan J A , et al. The antiviral effect o f zidovudine and ribivirin in clinical trials and the use o f p24 antigen levels as a virologic marker. Journal o f Infectious Diseases 1989;159:822-828. 243. Richman D D , Fischl M A , Grieco M H , et al. The toxicity of azidothymidine ( A Z T ) in the treatment of patients with A I D S and AIDS-related complex: a double-blind, placebo-controlled trial. N e w England Journal o f Medicine 1987;317:192-197. 244. Fischl M A , Richman D D , Hansen N , et al . The safety and efficacy o f zidovudine ( A Z T ) in the treatment o f subjects with mi ld ly symptomatic human immunodeficiency virus type 1 (HIV-1) Infection. Annals o f Internal Medicine 1990; 112:727-737. 245. Volberding P A , Lagakos S W , K o c h M A , et al. Zidovudine in asymptomatic human immunodeficiency virus infection: a controlled trial in persons with fewer than 500 CD4-posi t ive cells per cubic millimeter. N e w England Journal o f Medicine 1990;322:941-949. 246. Boucher F D , M o d l i n JF, Wel ler S, et al. Phase I evaluation of zidovudine administered to infants exposed at birth to the human immunodeficiency virus. Journal o f Pediatrics 1993;122:137-144. 247.Spector S A , Blanchard S, Connor E M , et al. Results o f a cl inical trial comparing two doses of 2 ' 3 ' -dideoxycytidine (ddC) in the treatment of children with symptomatic human immunodeficiency virus infection who were intolerant or had failed zidovudine therapy ( A C T G 138) [abstract]. Pediatric Research 1994,35:197A. 248. Pizzo Pa, Butler K , Balis F, et al. Dideoxycytidine alone and in an alternating schedule with zidovudine in children with symptomatic human immunodeficiency virus infection. Journal o f Pediatrics 1990; 117:799-808. 249. Husson R N , Muel ler B U , Farley M et al. Zidovudine and didanosine combination therapy in children with human immunodeficiency virus infection. Pediatrics 1994;93:316-322. 250. Mueller B U , Pizzo P A , Farley M , et al. Pharmacokinetic evaluation o f the combination o f zidovudine and didanosine in children with human immunodeficiency virus infection. Journal o f Pediatrics 1994;125:142-146. 251. Englund J A , Baker C J , Raskino C , et al. Zidovudine, didanosine, or both as the initial treatment for symptomatic HIV- l - in fec ted children. N e w England Journal o f Medicine. 252. Pantaleo G , Graziosi C , Fauci A S . The immunopathogenesis o f human immunodeficiency virus infection. New England Journal o f Medicine 1993;328:327-335. 253.Saag M S , Hahn B H , Gibbons J, et al. Extensive variation of human immunodeficiency virus type-1 in vivo. Nature 1988;334:440-444. 254. Coffin J M . HIV-1 replication dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 1995;267:483-488. 255. Coffin J. HIV-1 viral dynamics. A I D S 1996;10(suppl 3):S75-S84. 256. Embretson J, Supancic M , Ribas J L , et al. Massive covert infection o f helper T-lymphocytes and macrophages by HIV-1 during the incubation period o f A I D S . Nature 1993;362:359-362. 257. Pantaleo G , Graziosi C , Demarest JF, et al. HIV-1 infection is active and progressive in lymphoid tissue during the cl inical ly latent stage of disease. Nature 1993;362:35-359. 101 258. Ho D D , Neumann A U , Perelson A S , et al. Rapid turnover o f plasma virions and C D 4 lymphocytes in HIV-1 infection. Nature 1995;373:123-126. 259. Wei X , Ghosh S, Taylor M E , et al. V i r a l dynamics in human immunodeficiency virus type 1 infection. Nature 1995;373:117-122. 260.Schuurmann R, Nijhuis M , van Leeuwen R, et al. Rapid changes in human immunodeficiency virus type 1 R N A load and appearance o f drug-resistant virus populations in persons treated with lamivudine. Journal o f Infectious Diseases 1995;171:1411-1419. 261. Perelson A S , Neumann A U , Markowi tz M et al. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996;271:1582-1586. 262. Nowak M A , Anderson R M , M c L e a n A R , et al. Antigenic diversity thresholds an the development of A I D S . Science 1991;254:963-969. 263. M c L e a n A R , Frost S D W . Zidovudine and H I V - 1 : mathematical models o f within-host population dynamics. Review o f Medica l Vi ro logy 1995;5:141-147. 264. Mellors J W , Rinaldo C R J , Gupta P, et al. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 1996;272:1167-1170. 265. Mellors J W , Kingsley L A , Rinaldo C R , et al. Quantitation o f HIV-1 R N A in plasma predicts outcome after seroconversion. Annals o f Internal Medicine 1995;122(8):573-579. 266. Coombs R W , Welles S L , Hooper C , Association o f plaxma human immunodeficiency virus type q R N A level with risk o f cl inical progression in patients with advanced infection. The journal o f Infectious Diseases 1996;174:704-712. 267. De Rossi A , Masiero S, Giaquinto C , et al. Dynamics o f viral replication in infants with vertically acquired human immunodeficiency virus type 1 infection. Journal o f C l i n i c a l Investigation 1996;97:323-330. 268. Palumbo P E , K w o k S, Waters S, et al. V i r a l measurement by polymerase chain reaction-based assays in human immunodeficiency virus-infected infants. Journal o f Pediatrics 1995;126:592-595. 269 .O 'Br ien T R , Blattner W A , Waters D , et al. Serum HIV-1 R N A levels and time to development of A I D S in the Multicenter Hemophil ia Cohort Study. Journal o f the American Medica l Association 1996;276:105-110. 270.Mcintosh K , Shevitz A , Zaknun D , et al. Age- and time-related changes in extracellular viral load in children vertically infected by human immunodeficiency virus. Pediatric Infectious Diseases Journal 1996;15:1087-1091. 271 .Mofenson L M , Kore l i tz J, Meyer W A III, et al. The relationship between serum H I V - 1 R N A levels, C D 4 lymphocyte percent, and long-term mortality risk in HIV- l - in fec ted children. Journal o f Infectious Diseases (in press). 272. Abrams E J , Weedon JC , Lambert G , Steketee R. HIV-1 viral load early in life as a predictor o f disease progression in HIV- l - in fec ted infants [abstract]. In: V o l . 2. Abstracts. X I International Conference on A I D S , Vancouver, B . C . , July 7-12, 1996:25. 273. Jurriaans S, Van Gemen G , Weverl ing G J , et al. The natural history of H I V - 1 infection: virus load and virus phenotype independent determinants o f cl inical course? V i ro logy 1994;204:223-233. 102 274. Henrard D R , Phil l ips JR, Muenz L R , et al. Natural history o f H I V - 1 cell-free viremia. Journal o f the American Medical Association 1995;274:554-558. 275. H o D D . Vi ra l counts count in HIV-1 infection. Science 1996;272:1124-1125. 276.Shearer W T , Quinn T C , LaRussa P, et al. V i r a l load and disease progression in infants infected with human immunodeficiency virus type 1. New England Journal o f Medicine 1997;336( 19): 1337-1342. 277. Larder B A , Kemp S D , Harrigan PR. Potential mechanism for sustained antiretroviral efficacy o f A Z T - 3 T C combination therapy. Science 1995;269:696-699. 278. Condra J H , Schleiff W A , Blahy O M , et al. In vivo emergence of HIV-1 variants resistant to multiple protease inhibitors. Nature 1995;374:569-571. 279. Rogers M F , Ou C , Ki lbourne B , et al. Advances and problems in the diagnosis o f human immunodeficiency virus infection in infants. Pediatric Infectious Diseases Journal 1991;10:523-531. 280,Schupbach J, Wunderl i W , K i n d C , et al. Frequent detection o f H I V - l - a n d IgG-specific I g M and IgA antibodies in HIV-1-posi t ive cord-blood sera: fine analysis by western blot A I D S 1989;3:583-589. 281. De Rossi A , Ades A E , Mammano R, et al. Antigen detection, virus culture, polymerase chain reaction, and in vitro antibody production in the diagnosis o f vertically transmitted HIV-1 infection. A I D S 1991;5:15-20. 282. Bryson Y J , Pang S, W e i L S , et al. Clearance o f HIV-1 infection on a perinatally infected infant. N e w England Journal o f Medicine 1995;332:833-838. 283. Brandt C D , Rakusan T A , Josephs S H , et al. Detection of human immunodeficiency virus type 1 infection in young pediatric patients by using polymerase chain reaction and biotinylated probes. Journal o f C l in i ca l Microb io logy 1992;30(l):36-40. 284. Kr iv ine A , Yakudima A , L e M a y M , et al. A comparative study o f virus isolation, polymerase chain reaction, and antigen detection in children o f mothers infected with human immunodeficiency virus. Journal o f Pediatrics 1990;116:372-376, 285. Borkowsky W , Krasinski K , Pollack H , et al. Early diagnosis o f human immunodeficiency virus infection in children <6 months o f age: Comparison of polymerase chain reaction, culture and plasma antigen capture techniques. The Journal o f Infectious Diseases 1992; 166:616-619 286. Viscarello R R , Cul len M T , DeGrennaro N J , et al . Fetal blood sampling in human immunodeficiency virus-seropositive women before elective midtrimester termination o f pregnancy. American Journal o f Obstetrics and Gynecology 1992; 167:1075-1079. 287. Thea D M , Steketee R W , Pliner V , et al. The effect o f maternal viral load on the risk o f perinatal transmission of H I V - 1 . A I D S 1997; 11(4):437-444. 288. Chan M M , Campos J M , Josephs S, Rifai N . p 2 -microglobulin and neopterin: predictive markers for human immunodeficiency virus type 1 infection in children? Journal o f Cl in ica l Microbio logy 1990;28(10):2215-2219. 289. Fahey J L , Taylor J M G . Detels R, et al. The prognostic value o f cellular and serologic markers in infection with human immunodeficiency virus type 1. N e w England Journal o f Medicine 1990;322:166-172. 103 290. Connor E M , Sperling R S , Gelber R, et al. Reduction o f maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N e w England Journal o f Medicine 1994;331 (18): 1173-1180. 291 .Von Briesen H , Becker W B , Henco K , et al. Isolation frequency and growth properties o f HIV-1 variants: Mul t ip le simultaneous variants in a patient demonstrated by molecular cloning. Journal o f Medical Vi ro logy 1987;23:51-66. 292.Sakai K , Dewhurst S, M a X , V o l s k y D J . Differences in cytopathogenicity and host cel l range among infectious molecular clones of human immunodeficiency virus type 1 simultaneously from an individual. Journal o f V i ro logy 1988;62:4078-4085. 293.Schuitemaker H , Koot M , Kootstra N A , et al. Biological phenotype of human immunodeficiency virus type 1 clones at different stages o f infection: progression o f disease is associated with a shift from monocytotropic to T-cell-tropic virus populations. Journal o f virology 1992;66:1354-1360. 294.Schwartz S, Felber B K , Fenyo E M , et al. Rapidly and slowly replicating human immunodeficiency virus type 1 isolates can be distinguished according to target-cell tropism in T-cel l monocyte cell lines. Proc. Natl . Acad . Sci . U S A 1989;86:7200-7203. 295.Schuitemaker H , Kootstra N A , de Goede R E Y , et al. Monocytotropic human immunodeficiency virus type 1 (HIV-1) variants detectable in all stages o f HIV-1 infection lack T-cel l line tropism and syncytium-inducing ability in primary T-cel l culture. Journal o f V i ro logy 1991;65:356-363. 296.Richman D D , and Bozzette S A . The impact o f the syncytium-inducing phenotype of human immunodeficiency virus on disease progression. The Journal o f Infectious Diseases 1994; 169:968-974. 297.Spencer L T , Ogino M T , Dankner W M , and Spector S A . Cl in ica l significance of human immunodeficiency virus type 1 phenotypes in infected children. The Journal o f Infectious Diseases 1994;169:491-495. 298. Larder B A , Darby G , Richman D D . HIV-1 with reduced sensitivity to zidovudine ( A Z T ) isolated during prolonged therapy. Science 1989;243:1731-1734. 299. Richman D D . Zidovudine resistance of human immunodeficiency virus. Review of Infectious Diseases 1990; 12(suppl):5507-5512. 300. Nielsen K , W e i L S , S im M - S , et al. Correlation of clinical progression in human immunodeficiency virus-infected children with in vitro zidovudine resistance measured by a direct quantitative peripheral blood lymphocyte assay. The Journal o f Infectious Diseases 1995;172:359-364. 301. Roberts N A . Drug-resistance patterns o f saquinavir and other HIV-1 protease inhibitors. A I D S 1995; 9(suppl 2):S27-S32. 302. Johnson V A . Nucleoside Reverse Transcriptase Inhibitors and Resistance o f Human Immunodeficiency Virus Type 1. The Journal o f Infectious Diseases 1995; 171 (suppl 2):S140-S149. 303. Boucher C A , Lange J M A , Miedema F F , et al. HIV-1 biological phenotype and the development of zidovudine resistance in relation to disease progression in asymptomatic individuals during treatment. A I D S 1992;6(11): 1259-1264. 304. Kel lam P, Boucher C A , Larder B A . Fifth mutation in human immunodeficiency virus type 1 reverse transcriptase contributes to the development o f high-level resistance to zidovudine. Proc. Natl . Acad . Sci . U S A 1992;89:1934-1938. 104 305. Boucher C A B , O 'Su l l i van E , Mulder J W , et al. Ordered appearance o f zidovudine-resistant mutations during treatment of 18 human immunodeficiency virus-positive subjects. The Journal o f Infectious Diseases 1992;165:105-110. 306. Tudor-Wil l iams G , St Cla i r M H , M c K i n n e y R D , et al. HIV-1 sensitivity to zidovudine and clinical outcome in children. Lancet 1992;339:15-19. 307. Koza l N J , Shafer R S , Winter M A , et al. A mutation in HIV-1 reverse transcriptase and decline in C D 4 + lymphocyte numbers in long-term zidovudine recipients. The Journal o f Infectious Diseases 1993;167:526-532. 308. Erice A , Mayers D L , Strike D G , et al. B r i e f report: primary infection caused by zidovudine-resistant human immunodeficiency virus type 1 (HIV-1) . N e w England Journal o f Medicine 1993;328:1163-1165. 309. Fang G , Burger H , Grimson R, et al. Maternal plasma HIV-1 R N A : a determinant and threshold for mother-to-child transmission. Proc. Natl . Acad . Sci . U S A 1995;92:12100-12104. 310. Anonymous. HIV-1 seroconversion after occupational exposure despite early prophylactic zidovudine therapy. Lancet 1993;341:1077-1078. 311 .Dournon E, Matheron S, Rozenbaum W , et al. Effects o f zidovudine in 365 consecutive patients with A I D S or AIDS-related complex. Lancet 1988;ii: 1297-1302. 312. Graham N M H , Zeger S L , Park L P , et al. Effect o f zidovudine and Pneumocystis carinii pneumonia prophylaxis on progression o f HIV-1 infection to A I D S . Lancet 1991;338:265-269. 313. Boucher C A B , Tersmette M , Lange J M A , et al. Zidovudine sensitivity o f human immunodeficiency viruses from high-risk symptom-free individuals during therapy. Lancet 1990;336:585-590. 314. Richman D D , Grimes J M , Lagakos S W . Effect o f stage of disease and dug dose on zidovudine susceptibilities o f isolates o f human immunodeficiency virus. Journal o f Acquired Immune Deficiency Syndromes 1990;3:743-746. 315. Asjo B , Morfeldt-Manson L , Albert J, et al. Replicative capacity o f human immunodeficiency virus from patients with varying severity o f HIV-1 infection. Lancet 1986;ii:660-662. 3 16.Tersmette M , De Goede R, A l B J M , et al. Differential syncytium-inducing capacity o f Human immunodeficiency virus isolates: frequent detection of syncytium-inducing isolates inpatients with acquired immunodeficiency syndrome ( A I D S ) and AIDS-related complex. Journal o f V i ro logy 1988;62:2026-2032. 317. Koot M , Vos A H V , Keet R P M , et al. HIV-1 biological phenotype in long-term infected individuals evaluated with an M T - 2 cocultivation assay. A I D S 1992;6:49-54. 318. Tersmette M , Lange J M A , De Goede R, et al. Association between biological properties o f human immunodeficiency virus variants and risk for A I D S and A I D S mortality. Lancet 1989;1:983-985. 319. Fitzgibbon JE , Gaur S, Frenkel L D , et al. Transmission from one child to another o f human immunodeficiency virus type 1 with a zidovudine-resistance mutation. N e w England Journal o f Medicine 1993;329(25): 1835-1841. 320.Ogino M T , Dankner W M , and Spector S A . Development and significance o f zidovudine resistance in children infected with human immunodeficiency virus. The Journal o f Pediatrics 1993; 123(1): 1-8. 105 321. Koot M , Keet I P M , Vos A H V , et al. Prognostic value of HIV-1 syncytium-inducing phenotype for rate of C D 4 + cell depletion and progression to A I D S . Annals o f Internal Medicine 1993; 118:681-688. 322. Conway B , A l l e n U D , Cassol S, et al. Evaluation of novel zidovudine ( Z D V ) resistance testing methodologies [abstract], X I International Conference on A I D S . Vancouver, July 7-12,1996. Th.B.4342. 323.Sei S, Kleiner D E , K o p p J B , et al. Quantitative analysis o f viral burden in tissues from adults and children with symptomatic human immunodeficiency virus type 1 infection assessed by polymerase chain reaction. The Journal o f Infectious Diseases 1994;170:325-333. 324. Piatak M , Saag M S , Yang L C , et al. High levels o f HIV-1 in plasma during all stages o f infection determined by competitive P C R . Science 1993;259:1749-1754. 325. Katzenstein T L , Pedersen C , Nielsen C , et al. Longitudinal serum HIV-1 R N A quantification: correlation to viral phenotype at seroconversion and cl inical outcome. A I D S 1996; 10(2): 167-173. 326. Nielsen C , Bruun L , Mathiesen L R , et al. Development of resistance to zidovudine ( Z D V ) and didanosine (ddl) in HIV-1 from patients in Z D V , ddl and alternating Z D V / d d l therapy. A I D S 1996;10(6):625-633. 327. M c G a v i n C , Land S, Sebire K , Hooker D , et al . Syncytium-inducing phenotype and zidovudine susceptibility o f HIV-1 isolated from post-mortem tissue. A I D S 1996; 10(l):47-53. 328. Boom R, Sol C , Salimans M , et al. A rapid and simple method for purification o f nucleic acids. Journ. o f C l i n . Mic rob io l . 1990:28:495-503. 329. Mofenson L M , Wol insky S M . Current Insights Regarding Vertical Transmission, from: Pizzo and Wilfert eds: Pediatric A I D S . The challenge of HIV-1 infection in infants, children and adolescents 2 n d edition, Wil l iams and Wi lk in s , Baltimore, M D 1994. 330. Wor ld Health Organization. A I D S Epidemic Update: December 1998. 331. Health Canada. H I V and A I D S in Canada: Surveillance report to June 30, 1998. Divis ion of H I V / A I D S , Bureau of H I V / A I D S , S T D and T B , L C D C , Health Canada, November 1998. 332. H I V / A I D S Epi Update. Perinatal Transmission o f H I V . Bureau o f H I V / A I D S , S T D and T B Update Series. L C D C , M a y 1999. 333. Connor, R I, Ho, D D . Bio logy and Molecular Bio logy o f H I V , from: Pizzo and Wilfert eds: Pediatric A I D S . The challenge o f HIV-1 infection in infants, children and adolescents 2 n d edition, Wil l iams and Wi lk ins , Baltimore, M D 1994. 334. Fauci, A S, Moderator. Immunopathogenic mechanisms in Human Immunodeficiency Virus ( H I V ) Infection. A n n Intern M e d 1991: 114:678-693. 335. Viscarello, R R, Landers D V . Advances in prenatal diagnosis o f HIV-1 infection, from: Pizzo and Wilfert eds: Pediatric A I D S . The challenge o f HIV-1 infection in infants, children and adolescents 2 n d edition, Wi l l i ams and Wi lk ins , Baltimore, M D 1994. 336. Rene van Erp. Organon Teknika, N A S B A System Training Manual March 1994. 106 337. Kievits , T, van Gemen, B , van Strijp, D , Schukkink, R, Dirckx , M , Adriaanse, H , Malek , L , Sooknanan, R, and Lens, P. N A S B A isothermal enzymatic in vitro nucleic acid amplification optimized for the diagnosis o f HIV-1 infection. J. V i r o l . Methods 1991: 35:273-286. 338. Palumbo, P E. H I V / A I D S in infants, children and adolescents: Antiretroviral therapy o f H I V infection in children. Pediatric C l in ic s o f North America 2000:47(1): 155-169. 339. Cocchi , F, D e V i c o , A L , Garzino-Demo A , A r y a S K , Gal lo R C , Lusso P. Identification o f R A N T E S , M l P - l a and M I P - 1 P as the major HIV-supressive factors produced by C D 8 + T cells. Science 1995:270:1811-1815. 107 

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.831.1-0090070/manifest

Comment

Related Items