UBC Faculty Research and Publications

Comparison of atazanavir/ritonavir and darunavir/ritonavir based antiretroviral therapy for antiretroviral… Antoniou, Tony; Szadkowski, Leah; Walmsley, Sharon; Cooper, Curtis; Burchell, Ann N; Bayoumi, Ahmed M; Montaner, Julio S G; Loutfy, Mona; Klein, Marina B; Machouf, Nima; Tsoukas, Christos; Wong, Alexander; Hogg, Robert S; Raboud, Janet Apr 11, 2017

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

Item Metadata

Download

Media
52383-12879_2017_Article_2379.pdf [ 583.5kB ]
Metadata
JSON: 52383-1.0362101.json
JSON-LD: 52383-1.0362101-ld.json
RDF/XML (Pretty): 52383-1.0362101-rdf.xml
RDF/JSON: 52383-1.0362101-rdf.json
Turtle: 52383-1.0362101-turtle.txt
N-Triples: 52383-1.0362101-rdf-ntriples.txt
Original Record: 52383-1.0362101-source.json
Full Text
52383-1.0362101-fulltext.txt
Citation
52383-1.0362101.ris

Full Text

RESEARCH ARTICLE Open AccessComparison of atazanavir/ritonavir anddarunavir/ritonavir based antiretroviraltherapy for antiretroviral naïve patientsTony Antoniou1,2,3*, Leah Szadkowski4, Sharon Walmsley3,4, Curtis Cooper5, Ann N. Burchell1,2,3,Ahmed M. Bayoumi1,3, Julio S. G. Montaner6,7, Mona Loutfy3,8,9, Marina B. Klein10, Nima Machouf11,Christos Tsoukas10, Alexander Wong12, Robert S. Hogg6,13, Janet Raboud3,4 and The Canadian Observational Cohort(CANOC) collaborationAbstractBackground: Atazanavir/ritonavir and darunavir/ritonavir are common protease inhibitor-based regimens fortreating patients with HIV. Studies comparing these drugs in clinical practice are lacking.Methods: We conducted a retrospective cohort study of antiretroviral naïve participants in the CanadianObservational Cohort (CANOC) collaboration initiating atazanavir/ritonavir- or darunavir/ritonavir-based treatment.We used separate Fine and Gray competing risk regression models to compare times to regimen failure (compositeof virologic failure or discontinuation for any reason). Additional endpoints included virologic failure,discontinuation due to virologic failure, discontinuation for other reasons, and virologic suppression.Results: We studied 222 patients treated with darunavir/ritonavir and 1791 patients treated with atazanavir/ritonavir.Following multivariable adjustment, there was no difference between darunavir/ritonavir and atazanavir-ritonavir in therisk of regimen failure (adjusted hazard ratio 0.76, 95% CI 0.56 to 1.03) Darunavir/ritonavir-treated patients were at lowerrisk of virologic failure relative to atazanavir/ritonavir treated patients (aHR 0.50, 95% CI 0.28 to 0.91), findings drivenlargely by high rates of virologic failure among atazanavir/ritonavir-treated patients in the province of British Columbia.Of 108 discontinuations due to virologic failure, all occurred in patients starting atazanavir/ritonavir. There was nodifference between regimens in time to discontinuation for reasons other than virologic failure (aHR 0.93; 95% CI 0.65to 1.33) or virologic suppression (aHR 0.99, 95% CI 0.82 to 1.21).Conclusions: The risk of regimen failure was similar between patients treated with darunavir/ritonavir and atazanavir/ritonavir. Although darunavir/ritonavir was associated with a lower risk of virologic failure relative to atazanavir/ritonavir,this difference varied substantially by Canadian province and likely reflects regional variation in prescribing practicesand patient characteristics.Keywords: HIV, Antiretroviral therapy, Protease inhibitor, Atazanavir, Darunavir* Correspondence: tantoniou@smh.ca1Keenan Research Centre, Li Ka Shing Knowledge Institute, St. Michael’sHospital, Toronto, ON, Canada2Department of Family and Community Medicine, St. Michael’s Hospital,Toronto, ON, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Antoniou et al. BMC Infectious Diseases  (2017) 17:266 DOI 10.1186/s12879-017-2379-8BackgroundProtease inhibitors remain important options for the treat-ment of HIV infection [1, 2]. When administered with lowdoses of ritonavir, protease inhibitors impose a high gen-etic barrier against the selection of drug-resistant variantsof HIV and are therefore especially reliable options forpatients for whom poor antiretroviral adherence is antici-pated [3, 4]. Because of once-daily dosing and low rates ofgastrointestinal adverse effects relative to other membersof their class [5, 6], ritonavir-boosted atazanavir anddarunavir were, until recently, designated as ‘preferred’protease inhibitor-based options for the treatment of anti-retroviral naïve patients in the United States Depatmentof Health and Human Services (DHHS) HIV treatmentguidelines [2]. However, atazanavir/ritonavir was reclassi-fied as an ‘alternative’ to darunavir/ritonavir in the mostrecent iteration of the DHHS guidelines [2] based on highrates of discontinuation due to toxicity among patientstreated with atazanavir/ritonavir in ACTG 5257, a ran-domized trial comparing the efficacy of atazanavir/ritona-vir-, darunavir/ritonavir- and raltegravir-based therapy [7].Despite these findings, questions remain about thecomparative effectiveness of atazanavir/ritonavir anddarunavir/ritonavir in clinical practice. Although ran-domized trials are essential for generating evidenceabout efficacy required to inform clinical guidelines, in-dividuals enrolled in these studies are often not repre-sentative of patients treated in routine care [8, 9].Observational studies can address this limitation by pro-viding evidence of the comparative effectiveness and tol-erability of different treatment regimens in patients whoare treated in clinical practice. Although observationalstudies comparing the tolerability of atazanavir/ritonavirand darunavir/ritonavir have been conducted, no suchstudies have specifically compared the effectiveness ofthese drugs [10–12]. Accordingly, we conducted a retro-spective cohort study comparing the effectiveness anddurability of atazanavir/ritonavir- and daruanavir/ritona-vir-based regimens among antiretroviral naïve patientsenrolled in a longitudinal Canadian cohort study.MethodsStudy populationThe Canadian Observational Cohort (CANOC) collabor-ation is a multisite cohort study of antiretroviral-naïveHIV positive patients initiating combination antiretro-viral therapy (cART) after January 1, 2000 [13]. The col-laboration currently includes 8 participating cohortsfrom Ontario, Quebec and British Columbia. Criteria forinclusion into CANOC were documented HIV infection,residence in Canada, age 18 years and older, initiation ofa first antiretroviral regimen comprised of at least threeindividual agents, and at least one HIV-1 RNA viral loadand CD4 count measurement within one year prior totreatment initiation. Antiretroviral data collectionmethods vary by site, and include abstraction frompatient charts and linkage with provincial prescriptiondatabases. Nonnominal data were submitted from eachparticipating site to the coordinating center in Vancouver,British Columbia, Canada.Participants were eligible for inclusion in this analysisif they initiated atazanavir/ritonavir- or darunavir/ritona-vir-based antiretroviral therapy, did not have a viral loadless than or equal to 200 copies/mL at or before cARTinitiation and had at least one follow-up viral loadmeasurement available after treatment initiation. Weexcluded patients who initiated atazanavir withoutconcomitant ritonavir.Outcome measuresThe primary outcome of the study was time to regimenfailure, defined as a composite of virologic failure anddiscontinuation for any reason [Table 1, A (includes B1)or B2]. Similar to ACTG 5257, we defined virologic fail-ure as a viral load >1000 copies/mL at or after week 16but before week 24, or a viral load >200 copies/mL at orafter week 24 (A) [7]. Patients who never suppressed orwho suppressed and subsequently rebounded were in-cluded in this definition. We defined discontinuation asstopping atazanavir or darunavir for more than 60 days.We did not consider changes to the nucleoside back-bone, ritonavir, or the addition of other antiretroviralmedications as discontinuations. All discontinuationswere hierarchically classified as due to virologic failure(B1)or other reasons (B2). Patients who died were con-sidered to have met a competing risk; otherwise patientswere censored at the first occurrence of a gap in viralload measurements exceeding two years, the lastrecorded antiretroviral stop date if followed by a viralload ≤50 copies/mL, or the last available viral loadmeasurement.Table 1 Outcome and Competing Risk DefinitionsOutcome Definition Competing RisksPrimary AnalysisRegimen Failure First A (includes B1)or B2deathSecondary AnalysesVirologic Failure First A (includes B1) B2, deathDiscontinuation Due toVirologic FailureFirst B1 B2, deathDiscontinuation Due toOther ReasonsFirst B2 A (includes B1), deathVirologic Suppression First C B, deathA. Virologic FailureB. Discontinuation (hierarchically classified as B1: Due to virologic failure, orB2: Due to other reasons)C. Virologic SuppressionAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 2 of 10We conducted several secondary analyses (Table 1).First, we examined time to virologic failure (A) and timeto discontinuations due to virologic failure (B1) separ-ately, considering death and discontinuations for otherreasons (B2) as competing risks. Next, in the absence ofspecific adverse event data, we examined time to discon-tinuation of either darunavir or atazanavir for reasonsother than virologic failure (B2). Such discontinuationsmay have occurred because of toxicities attributable tothe protease inhibitor. Individuals who experienced viro-logic failure (A) and those who died were considered tohave met a competing risk. Finally, we examined time tovirologic suppression (C), defined as time to the first ofat least two consecutive viral load measurements below50 copies/mL at least 30 days apart. Patients who diedor discontinued atazanavir or darunavir for any reason(B) were considered to have met a competing risk. Weused the same censoring rules as in the primary analysisin all secondary analyses.Statistical analysisAll analyses were conducted with SAS 9.4 (SAS Institute,Cary, North Carolina, USA) and R 3.3.1 (R DevelopmentCore Team, Vienna, Austria). We compared baselinecharacteristics between the two regimen groups usingchi-square tests for categorical variables and Wilcoxonrank sum tests for continuous variables. For eachoutcome, cumulative incidence functions taking intoaccount competing risks were determined by regimenand compared using Gray’s Test for Equality [14].We used multivariable Fine and Gray competing risk re-gression [15] to estimate the association between treatmentwith darunavir/ritonavir relative to atazanavir/ritonavir andeach outcome. We adjusted our models for age, sex andmen who have sex with men (MSM) status, race, baselineviral load and CD4 count, calendar year of treatmentinitiation, nucleoside analogue backbone (emtricitabine/tenofovir versus other backbones), Canadian province ofresidence, and history of injection drug use (IDU). Becauseof collinearity beween IDU and co-infection with hepatitisC, we adjusted our models for the former variable only. Forcovariates with large amounts of missing data, separatecategories for missing were created when these variableswere included in the regression models.We conducted several sensitivity analyses to test therobustness of our findings. First, we conducted analysesby provincial subgroup (British Columbia vs Ontario/Quebec) because of regional differences in prescribingpatterns and characteristics of CANOC participants[13]. Specifically, antiretroviral treatment guidelines inBritish Columbia recommended efavirenz or atazanavir/ritonavir as first-line regimens during our study period,favouring the latter for patients who use drugs or withmental health illness and preserving darunavir/ritonavirfor treatment failures. In contrast, no such recommenda-tions were in place in Ontario and Quebec during thestudy period. Further, participants in British Columbiadiffer from those in Ontario and Quebec in importantways, including mode of HIV acquisition and hepatitis Ccoinfection [13]. We therefore reasoned that residualconfounding due to associated unmeasured variablessuch as mental health illness and socioeconomic statuscould occur. Second, we examined whether outcomesvaried by baseline viral load (< 100,000 copies/mL or>100,000 copies/mL). Third, because approximately one-quarter of atazanavir-treated patients started treatmentbetween 2004 and 2006, we replicated our analyses,restricting to patients who started treatment on or afterJanuary 1, 2010. Fourth, patients who had a viral load of≤50 copies/mL and who were switched to a single tableregimen were re-defined as having discontinued treat-ment for simplification and included as competing risks.Finally, we replicated our analyses following multiple im-putation for missing values of the race, IDU and MSMvariables. To our knowledge, no methods currently existto impute data specifically for Fine & Gray competingrisk regression models, and so a substantive model com-patible version of fully conditional specification was usedto impute data intended for cause-specific competingrisk regression models [16]. For each outcome, all covar-iates used in the original multivariable models were usedto impute missing values five different times. Fine &Gray models were then run on each imputed datasetand the resulting parameter estimates and varianceswere combined using Rubin’s rules [17, 18].ResultsDuring the study period, we identified 1791 eligible pa-tients whose first cART regimen included atazanavir/ri-tonavir and 222 eligible patients whose first regimenincluded darunavir/ritonavir. Patients treated with ataza-navir/ritonavir were more likely to be female (19% vs.13%; p = 0.02), co-infected with hepatitis C (30% vs.13%; p < 0.0001) and report injection drug use as a riskfactor for HIV infection (29% versus 9%; p < 0.0001)(Table 2). These differences were mitigated whenrestricting comparisons to participants from Ontarioand Quebec only. The median (interquartile range, IQR)duration of follow-up was 3.5 (1.6, 5.4) years and 1.5(0.7, 2.3) years in the atazanavir/ritonavir and darunavir/ritonavir treated participants respectively (Table 2).The cumulative incidence of regimen failure (A or B2)one year following the initiation of treatment was 0.29 (95%confidence interval (CI) 0.27 to 0.31) for patients initiatingatazanavir/ritonavir and 0.21 (95% CI 0.15 to 0.27) for thosestarting darunavir/ritonavir (p < 0.01, Fig. 1a and b). A totalof 43 patients experienced a competing risk of death. Fol-lowing multivariable adjustment, there was no differenceAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 3 of 10Table 2 Demographic and clinical characteristics at initiation of combination antiretroviral therapy by regimenaOverall Sample Ontario & Quebec Sub-analysisTotalN = 2013AtazanavirN = 1791DarunavirN = 222p AtazanavirN = 544DarunavirN = 183pDemographicsAge 40 (33–47) 40 (33–47) 40 (32–46) 0.25 38 (32–45) 39 (32–46) 0.49< 35 601 (30%) 521 (29%) 80 (36%) 0.03 189 (35%) 69 (38%) 0.0235–44 746 (37%) 680 (38%) 66 (30%) 217 (40%) 53 (29%)≥ 45 666 (33%) 590 (33%) 76 (34%) 138 (25%) 61 (33%)Male 1632 (81%) 1439 (81%) 193 (87%) 0.02 434 (80%) 162 (89%) <.01Gender & MSMFemale 373 (19%) 344 (19%) 29 (13%) <.0001 110 (20%) 21 (11%) <.01MSM 764 (38%) 646 (36%) 118 (53%) 293 (54%) 113 (62%)Male, not MSM 479 (24%) 451 (25%) 28 (13%) 93 (17%) 22 (12%)Missing 397 (20%) 350 (20%) 47 (21%) 48 (9%) 27 (15%)RaceWhite 531 (26%) 492 (27%) 39 (18%) <.001 146 (27%) 34 (19%) <.0001Black 128 (6%) 117 (7%) 11 (5%) 84 (15%) 11 (6%)Aboriginal Peoples 109 (5%) 101 (6%) 8 (4%) 11 (2%) 7 (4%)Other 145 (7%) 133 (7%) 12 (5%) 41 (8%) 9 (5%)Missing 1100 (55%) 948 (53%) 152 (68%) 262 (48%) 122 (67%)Injection drug useNo 1207 (60%) 1053 (59%) 154 (69%) <.0001 426 (78%) 142 (78%) 0.03Yes 542 (27%) 521 (29%) 21 (9%) 59 (11%) 11 (6%)Missing 264 (13%) 217 (12%) 47 (21%) 59 (11%) 30 (16%)Endemic countrybNo 537 (27%) 406 (23%) 131 (59%) <.0001 406 (75%) 131 (72%) 0.02Yes 99 (5%) 80 (4%) 19 (9%) 80 (15%) 19 (10%)Missing 1377 (68%) 1305 (73%) 72 (32%) 58 (11%) 33 (18%)ProvinceBritish Columbia 1286 (64%) 1247 (70%) 39 (18%) <.0001 – –Ontario 355 (18%) 264 (15%) 91 (41%) 264 (49%) 91 (50%) 0.78Quebec 372 (18%) 280 (16%) 92 (41%) 280 (51%) 92 (50%)ClinicalHepatitis CNo 1349 (67%) 1168 (65%) 181 (82%) <.0001 449 (83%) 155 (85%) 0.75Yes 559 (28%) 530 (30%) 29 (13%) 71 (13%) 20 (11%)Missing 105 (5%) 93 (5%) 12 (5%) 24 (4%) 8 (4%)CD4 count (cells/mm3) 220 (120–330) 220 (120–320) 280 (130–370) <.01 230 (149–310) 286 (130–370) <.01<200 869 (43%) 795 (44%) 74 (33%) <.001 214 (39%) 60 (33%) <.0001200–349 693 (34%) 620 (35%) 73 (33%) 234 (43%) 60 (33%)350–499 287 (14%) 237 (13%) 50 (23%) 69 (13%) 45 (25%)≥500 164 (8%) 139 (8%) 25 (11%) 27 (5%) 18 (10%)Log 10 (VL copies/mL) 4.91 (4.41–5.11) 4.90 (4.42–5.05) 4.94 (4.36–5.44) 0.03 4.82 (4.35–5.26) 4.90 (4.26–5.39) 0.43VL ≥ 100,000 copies/mL 881 (44%) 780 (44%) 101 (45%) 0.58 209 (38%) 77 (42%) 0.38Antoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 4 of 10between darunavir/ritonavir and atazanavir-ritonavir in therisk of regimen failure (adjusted hazard ratio 0.76, 95% CI0.56 to 1.03) (Table 3). This finding was similar in sensitiv-ity analyses (Table 4).The one-year cumulative incidence of virologic failure(A) was 0.09 (95% CI 0.08 to 0.10) for patients startingatazanavir/ritonavir and 0.04 (95% CI 0.02 to 0.08) forthose starting darunavir/ritonavir (p = 0.02, Fig. 1a and b).Forty-three patients died prior to virologic failure. Follow-ing multivariable adjustment, the risk of virologic failurewas lower in darunavir/ritonavir-treated patients (adjustedhazard ratio 0.50, 95% CI 0.28 to 0.91) (Table 3). However,this finding was driven primarily by high failure ratesamong atazanavir/ritonavir-treated patients in the prov-ince of British Columbia. In a sensitivity analysis byCanadian province, there was no difference in the risk ofvirologic failure among participants in Ontario andQuebec (adjusted hazard ratio 0.90, 95% CI 0.44 to 1.86)(Table 4). In the analysis of participants from BritishColumbia, the point estimate of the adjusted hazard ratioof virologic failure associated with darunavir was similar(adjusted hazard ratio = 0.41, 95% CI (0.09, 1.78)) to thatfrom the analysis of the entire cohort but the finding wasnot statistically significant due to the infrequent use ofdarunavir in this province (Table 4).The cumulative incidence of discontinuation due tovirologic failure (B1) in atazanavir/ritonavir and daruna-vir/ritonavir-treated patients was 0.01 (95% CI 0.01 to0.02) and 0.00, respectively (p = 0.01, Fig. 1a and b). Thelack of events in darunavir/ritonavir treated patients re-sulted in adjusted hazard ratios of 0.00 for darunavir/ritonavir vs. atazanavir/ritonavir when modelling thisoutcome.The cumulative incidence of discontinuation for rea-sons other than virologic failure (B2) was 0.2 (95% CI0.18 to 0.22) for patients initiating atazanavir/ritonavirand 0.17 (95% CI 0.12 to 0.22) for patients initiatingdarunavir/ritonavir (p = 0.16, Fig. 1a and b). Forty-threepatients experienced a competing risk of death. Therewas no difference between darunavir/ritonavir and ataza-navir/ritonavir in the risk of discontinuation for reasonsother than virologic failure after adjusting for confound-ing variables (adjusted hazard ratio 0.93, 95% CI 0.65 to1.33) (Table 3). These results were similar in sensitivityanalyses (Table 4).The cumulative incidence of virologic suppression (C)at 1 year after treatment initiation was 0.73 (95% CI 0.70to 0.75) for participants whose first regimen includedatazanavir/ritonavir and 0.73 (95% CI 0.66 to 0.79) forthose whose first regimen included darunavir/ritonavir(p = 0.62, Fig. 1c and d). Twenty-four patients died priorto achieving virologic suppression. Following multivari-able adjustment, there was no difference in the time tovirologic suppression (adjusted hazard ratio 0.99, 95% CI0.82 to 1.21) according to treatment regimen (Table 3).Results were similar in sensitivity analyses (Table 4).In sensitivity analyses of patients who initiated treat-ment on or after January 1, 2010, the adjusted hazardsratios for treatment group were very similar to those ofthe main analysis considering patients starting on orafter January 1, 2003 (Table 4). However, the estimateswere less precise because of the smaller sample size.Table 2 Demographic and clinical characteristics at initiation of combination antiretroviral therapy by regimena (Continued)Year of cART initiation 2009 (2007–2010) 2008 (2006–2010) 2011 (2010–2011) <.0001 2008 (2007–2009) 2010 (2010–2011) <.00012003–2006 474 (24%) 473 (26%) 1 (0%) <.0001 121 (22%) 1 (1%) <.00012007–2009 799 (40%) 767 (43%) 32 (14%) 307 (56%) 32 (17%)2010–2012 740 (37%) 551 (31%) 189 (85%) 116 (21%) 150 (82%)First NRTIs3TC/ABA 499 (25%) 441 (25%) 58 (26%) <.0001 259 (48%) 53 (29%) <.00013TC/TDF 285 (14%) 284 (16%) 1 (0%) 36 (7%) 1 (1%)FTC/TDF 1172 (58%) 1009 (56%) 163 (73%) 230 (42%) 129 (70%)Other 57 (3%) 57 (3%) 0 (0%) 19 (3%) 0 (0%)Follow upYears of Follow up 3.10 (1.45–5.16) 3.50 (1.63–5.38) 1.48 (0.69–2.25) <.0001 3.73 (2.20–5.32) 1.68 (0.86–2.46) <.0001VLs per year 4.67 (3.61–6.21) 4.68 (3.63–6.16) 4.50 (3.55–6.56) 0.71 3.88 (3.12–4.76) 4.32 (3.44–5.76) <.0010–3 274 (14%) 246 (14%) 28 (13%) 0.43 121 (22%) 26 (14%) <.0013–6 1187 (59%) 1062 (59%) 125 (56%) 358 (66%) 115 (63%)> 6 552 (27%) 483 (27%) 69 (31%) 65 (12%) 42 (23%)MSM men who have sex with men, VL viral load, cART combination antiretroviral therapy, NRTI nucleoside reverse transcriptase, 3TC lamivudine, ABC abacavir,TDF tenofovir, FTC emtricitabinea Medians (IQRs) are presented for continuous variables and frequencies (percent) for categorical variablesb Indicates patients who have immigrated to Canada from a country with a high prevalence of HIVAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 5 of 10DiscussionIn our analysis of more than two thousand antiretroviralnaïve patients from three Canadian provinces, patientsinitiating darunavir/ritonavir were at a lower risk ofvirologic failure (A) and subsequent discontinuationsdue to virologic failure (B1) than patients treated withatazanavir/ritonavir. In contrast, we observed no differ-ences in time to regimen failure (A or B2), virologicsuppression (C) or to discontinuation for reasons otherthan virologic failure (B2).Our findings differ from those of ACTG 5257, a ran-domized trial comparing darunavir/ritonavir, atazanavir/ritonavir- and raltegravir-based antiretroviral therapy [7].Specifically, this trial found a higher incidence of tolerabil-ity discontinuation among patients randomized to ataza-navir/ritonavir, mediated primarily by participant-drivenregimen change for jaundice or hyperbilirubinemia andnon-hepatobiliary gastrointestinal side effects. Althoughwe did not have data regarding the exact reasons for dis-continuation, we found no difference in the risk of discon-tinuations for reasons other than virologic failure (B2), anoutcome which includes toxicity-driven discontinuation.In a sensitivity analysis, there was no difference in the riskof discontinuations for reasons other than (i) virologic fail-ure or (ii) simplification, an outcome that may moreclosely approximate toxicity-driven discontinuation. Thediscrepancy between the results of our analysis and thoseof the ACTG 5257 trial may be due in part to differencesin the calendar year periods of atazanavir and darunavirinitiation in our study. Because atazanavir-treated patientsstarted treatment approximately three years earlier thandarunavir-treated patients, with approximately one-quarter starting between 2004 and 2006, we speculatedthat discontinuations due to jaundice or asymptomatichyperbilirubinemia were tempered by a lack of potent, tol-erable treatment alternatives during this period. Most not-ably, 70% of patients starting atazanavir/ritonavir residedin British Columbia, a province where the epidemic isdriven primarily by injection drug use, thereby potentiallydeterring clinicians from switching patients to regimenswith lower barriers to resistance or higher rates ofgastrointestinal side effects. In contrast, ACTG 5257Fig. 1 Cumulative incidence functions of time to events and competing risksAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 6 of 10participants initiated treatment between 2009 and2011, a period marked by the availability of potenttreatment alternatives conducive to supportingparticipant-driven requests for regimen change. Al-though sensitivity analyses restricted to participantsinitiating cART between 2010 and 2012 yielded simi-lar results, we speculate that this reflects a desire topreserve darunavir/ritonavir and other new options assecond-line regimens for patients treated with ataza-navir/ritonavir in British Columbia.Our finding of an increased risk of virologic failureamong patients receiving atazanavir/ritonavir alsocontrasts with the results of ACTG 5257, in which nodifference was observed in this outcome. A possible ex-planation relates to inter-study differences in the preva-lence and distribution of baseline characteristics knownto adversely affect adherence and sustained virologicsuppression. Specifically, relative to ACTG 5257 ourstudy population included a greater proportion of pa-tients who had a history of injection drug use (27%Table 3 Multivariable Fine and Gray models of time to eventsVirologic Failure orOther DiscontinuationVirologic Failure Discontinuation due toVirologic FailureOtherDiscontinuationVirologic SuppressionaHR (95% CI) p aHR (95% CI) p aHR (95% CI) p aHR (95% CI) p aHR (95% CI) pDRV/RIT 0.76 (0.56,1.03) 0.08 0.50 (0.28,0.91) 0.02 0.00 (0.00,0.00) <.0001 0.93 (0.65,1.33) 0.68 0.99 (0.82,1.21) 0.96FTC/TEN 1.02 (0.86,1.21) 0.84 1.30 (0.99,1.70) 0.06 1.41 (0.87,2.28) 0.16 0.87 (0.70,1.08) 0.20 0.99 (0.87,1.12) 0.82Age (per 10y) 0.89 (0.83,0.95) <.001 1.02 (0.91,1.15) 0.70 0.89 (0.73,1.09) 0.27 0.86 (0.79,0.94) <.001 1.04 (0.98,1.10) 0.17Gender & MSMFemale Ref. Ref. Ref. Ref. Ref.MSM 0.68 (0.55,0.84) <.001 0.73 (0.51,1.03) 0.08 0.46 (0.24,0.89) 0.02 0.79 (0.61,1.03) 0.08 1.20 (1.00,1.45) 0.05Male, not MSM 0.72 (0.59,0.88) <.01 0.90 (0.64,1.25) 0.51 0.70 (0.42,1.17) 0.17 0.76 (0.59,0.98) 0.03 1.06 (0.87,1.28) 0.57Missing 0.74 (0.56,0.98) 0.03 0.80 (0.47,1.35) 0.40 0.64 (0.27,1.55) 0.33 0.83 (0.60,1.15) 0.26 1.13 (0.90,1.42) 0.31RaceWhite Ref. Ref. Ref. Ref. Ref.Black 0.89 (0.64,1.23) 0.47 1.41 (0.88,2.26) 0.15 1.18 (0.47,2.95) 0.72 0.64 (0.40,1.04) 0.07 1.16 (0.88,1.54) 0.29Aboriginal Peoples 1.39 (1.05,1.85) 0.02 1.14 (0.72,1.82) 0.57 1.46 (0.71,2.99) 0.30 1.32 (0.91,1.89) 0.14 0.68 (0.51,0.92) 0.01Other 0.89 (0.67,1.18) 0.42 1.04 (0.68,1.59) 0.86 0.79 (0.36,1.75) 0.56 0.82 (0.56,1.19) 0.29 1.15 (0.94,1.40) 0.17Missing 1.17 (0.98,1.39) 0.08 0.75 (0.55,1.01) 0.06 0.70 (0.43,1.14) 0.15 1.38 (1.12,1.72) <.01 0.93 (0.80,1.08) 0.35IDUNo Ref. Ref. Ref. Ref. Ref.Yes 1.68 (1.39,2.02) <.0001 1.46 (1.05,2.04) 0.03 1.64 (0.92,2.94) 0.09 1.50 (1.20,1.88) <.001 0.69 (0.58,0.82) <.0001Missing 1.11 (0.83,1.49) 0.48 1.16 (0.68,1.96) 0.58 1.10 (0.41,2.93) 0.85 1.04 (0.74,1.45) 0.83 0.91 (0.72,1.14) 0.40ProvinceBC Ref. Ref. Ref. Ref. Ref.ON 0.95 (0.77,1.16) 0.60 1.11 (0.81,1.54) 0.52 0.67 (0.34,1.31) 0.24 0.90 (0.69,1.17) 0.44 0.97 (0.82,1.15) 0.73QC 0.82 (0.66,1.02) 0.08 1.26 (0.88,1.80) 0.21 1.03 (0.51,2.08) 0.93 0.68 (0.51,0.90) <.01 1.02 (0.86,1.21) 0.83Baseline CD4 (per 100)< 200 Ref. Ref. Ref. Ref. Ref.200–349 0.83 (0.71,0.96) 0.01 0.93 (0.73,1.20) 0.58 1.16 (0.76,1.77) 0.48 0.84 (0.69,1.02) 0.08 1.18 (1.04,1.33) <.01350–499 0.80 (0.63,1.01) 0.06 0.52 (0.33,0.84) <.01 0.60 (0.23,1.56) 0.29 1.00 (0.76,1.31) 0.99 1.35 (1.14,1.61) <.001≥ 500 1.07 (0.80,1.45) 0.64 0.79 (0.44,1.42) 0.44 0.95 (0.33,2.69) 0.92 1.22 (0.87,1.71) 0.25 1.36 (1.08,1.71) <.01Baseline VL ≥ 100,000 1.20 (1.05,1.38) <.01 1.74 (1.38,2.20) <.0001 2.27 (1.49,3.48) <.001 0.95 (0.80,1.13) 0.57 0.61 (0.55,0.68) <.0001Calendar Year2003–2006 Ref. Ref. Ref. Ref. Ref.2007–2009 0.96 (0.79,1.17) 0.69 0.68 (0.50,0.92) 0.01 0.41 (0.24,0.71) <.01 1.11 (0.87,1.42) 0.39 0.96 (0.82,1.12) 0.582010–2012 1.12 (0.88,1.42) 0.35 0.88 (0.61,1.28) 0.50 0.64 (0.33,1.24) 0.19 1.03 (0.77,1.38) 0.84 0.67 (0.56,0.81) <.0001AHR adjusted hazard ratio, DRV/r darunavir/ritonavir, ATZ/r atazanavir/ritonavir, FTC emtricitabine, TDF tenofovir, NRTI nucleoside reverse transcriptase inhibitor,MSM men who have sex with men, VL viral loadAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 7 of 10versus 7%), were coinfected with hepatitis C (28% versus7.8%) and with baseline CD4 counts below 350 cells/mm3 (77% versus 8.8%). Notably, these characteristicswere disproportionately represented among patientstreated with atazanavir/ritonavir in our study, likely pre-disposing these patients to virologic failure.Inter-provincial differences in patient characteristicsand prescribing practices may also account for disparatefindings between our study and ACTG 5257. Specific-ally, a higher prevalence of injection drug use and hepa-titis C was observed among participants in BritishColumbia relative to Ontario and Quebec. Although weadjusted for injection drug use in our analysis, we lackeddata regarding co-existing mental health illness, othersubstance use and socioeconomic status; residual con-founding is therefore possible. Also, as noted earlier, ata-zanavir/ritonavir was the recommended proteaseinhibitor for antiretroviral naïve patients in BritishColumbia during the study period, particularly for pa-tients with co-morbid illness or social circumstancesthat could predispose them to treatment failure. In con-trast, no such recommendations were in place in On-tario and Quebec during the study period. The impact ofthese selection biases was evident in sensitivity analysesrestricted to participants from Ontario and Quebec, inthat no differences between regimens were observed forany outcome, including virologic failure.Strengths of our analysis include the size of the studypopulation, the diversity of the participants in CANOCand the ability to compare clinical outcomes between in-dividuals initiating atazanavir/ritonavir and darunavir/ri-tonavir in a clinical practice setting. However, severallimitations of our study merit emphasis. Most notably,as with all observational studies, our findings may bebiased by residual inter-group differences in baselinevariables and unmeasured confounders. In addition, asnoted earlier, differences in the timing of darunavir andatazanavir availability may have precluded our ability toobserve differences in the risk of treatment discontinu-ation. The nature of antiretroviral data in a retrospectivecohort also imposes some limitations when consideringdiscontinuation. Some sites have prescription based datawith stop dates calculated from the days supplied by theprescription. These stop dates are inaccurate when a pa-tient is not perfectly adherent. Other sites provide startand stop dates abstracted from patient charts which aresubject to recall and documentation errors. Since 70% ofparticipants who had atazanavir as part of their firstregimen resided in British Columbia, where antiretro-viral records are prescription-based compared to 18% ofparticipants who had darunavir as part of their startingregimen, this may have resulted in an overestimate oftime to discontinuation among participants on atazana-vir. Finally, we lacked data regarding exact reasons fortreatment discontinuation.ConclusionsIn conclusion, darunavir/ritonavir and atazanavir/ritonavirwere of similar effectiveness in the treatment of antiretro-viral naïve patients. Differences in the risk of virologic fail-ure between darunavir/ritonavir and atazanavir/ritonavirvaried substantially by province due to regional differencesin prescribing patterns and patient characteristics. Ourdata provide both a comparison of the effectiveness ofthese two protease inhibitors in the clinical setting and anillustration of the potential magnitude and impact ofselection bias in a cohort study setting.AbbreviationsCANOC: Canadian observational cohort; CART: Combination antiretroviraltherapy; CI: Confidence interval; DHHS: Department of health and humanservices; IDU: Injection drug use; IQR: Interquartile range; MSM: Men whohave sex with menTable 4 Adjusted hazard ratios for darunavir/ritonavir vs. atazanavir/ritonavir from sensitivity analysesAnalysis Virologic Failure orOther DiscontinuationVirologic Failure Discontinuations dueto Virologic FailureOtherDiscontinuationVirologicSuppressionaHR (95% CI) p aHR (95% CI) p aHR (95% CI) p aHR (95% CI) p aHR (95% CI) p0 Original Model 0.76 (0.56,1.03) 0.08 0.50 (0.28,0.91) 0.02 0.00 (0.00,0.00) <.0001 0.93 (0.65,1.33) 0.68 0.99 (0.82,1.21) 0.961 Subgroup by ProvinceON/QC Only 0.73 (0.51,1.06) 0.10 0.90 (0.44,1.86) 0.78 0.00 (0.00,0.00) <.0001 0.72 (0.47,1.11) 0.14 0.99 (0.78,1.27) 0.96BC Only 0.69 (0.34,1.42) 0.32 0.41 (0.09,1.78) 0.23 0.00 (0.00,0.00) <.0001 0.93 (0.43,2.04) 0.86 0.78 (0.46,1.32) 0.352 Subgroup by Baseline VL< 100,000 copies/mL 0.66 (0.42,1.03) 0.07 0.47 (0.18,1.19) 0.11 0.00 (0.00,0.00) <.0001 0.82 (0.49,1.36) 0.44 0.97 (0.76,1.23) 0.78≥ 100,000 copies/mL 0.86 (0.56,1.33) 0.49 0.54 (0.25,1.16) 0.11 0.00 (0.00,0.00) <.0001 1.16 (0.68,2.00) 0.58 1.11 (0.81,1.53) 0.513 2010–2012 Only 0.81 (0.56,1.18) 0.27 0.55 (0.26,1.14) 0.11 0.00 (0.00,0.00) <.0001 0.96 (0.63,1.48) 0.87 0.98 (0.76,1.26) 0.874 Simplification discontinuationsincluded as competing risks0.75 (0.54,1.03) 0.07 0.50 (0.28,0.91) 0.02 0.00 (0.00,0.00) <.0001 0.92 (0.63,1.34) 0.65 0.99 (0.82,1.21) 0.965 SMC-FCS Imputation 0.76 (0.56, 1.04) 0.08 0.5 (0.28, 0.9) 0.02 0.00 (0.00,0.00) <.0001 0.91 (0.62, 1.34) 0.64 0.98 (0.8, 1.2) 0.84aHR adjusted hazard ratio, VL viral loadAntoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 8 of 10AcknowledgementsWe would like to thank all of the participants for allowing their informationto be a part of the CANOC collaboration.The CANOC Collaborative Research Centre includes: CANOC NominatedPrincipal Investigator: Robert Hogg (British Columbia Centre for Excellence inHIV/AIDS, Simon Fraser University).Principal Investigators: Ann N. Burchell (St. Michael’s Hospital, University ofToronto), Curtis Cooper (University of Ottawa, OCS), Deborah Kelly (MemorialUniversity of Newfoundland), Marina Klein (Montreal Chest InstituteImmunodeficiency Service Cohort, McGill University), Mona Loutfy (Universityof Toronto, Maple Leaf Medical Clinic, OCS), Nima Machouf (CliniqueMedicale l’Actuel, Université de Montréal), Julio Montaner (British ColumbiaCentre for Excellence in HIV/AIDS, University of British Columbia), JanetRaboud (University of Toronto, University Health Network, OCS), ChrisTsoukas (McGill University), Stephen Sanche (University of Saskatchewan),Alexander Wong (University of Saskatchewan), Tony Antoniou (St. Michael’sHospital, University of Toronto, Institute for Clinical Evaluative Sciences),Ahmed Bayoumi (St. Michael’s Hospital, University of Toronto), Mark Hull(British Columbia Centre for Excellence in HIV/AIDS), Bohdan Nosyk (BritishColumbia Centre for Excellence in HIV/AIDS, Simon Fraser University).Co-Investigators: Angela Cescon (Northern Ontario School of Medicine),Michelle Cotterchio (Cancer Care Ontario, University of Toronto), CharlieGoldsmith (Simon Fraser University), Silvia Guillemi (British Columbia Centrefor Excellence in HIV/AIDS, University of British Columbia), P. Richard Harrigan(British Columbia Centre for Excellence in HIV/AIDS, University of BritishColumbia), Marianne Harris (St. Paul’s Hospital), Sean Hosein (CATIE), SharonJohnston (Bruyère Research Institute, University of Ottawa), Claire Kendall(Bruyère Research Institute, University of Ottawa), Clare Liddy (BruyèreResearch Institute, University of Ottawa), Viviane Lima (British ColumbiaCentre for Excellence in HIV/AIDS, University of British Columbia), DavidMarsh (Northern Ontario School of Medicine), David Moore (British ColumbiaCentre for Excellence in HIV/AIDS, University of British Columbia), AlexisPalmer (British Columbia Centre for Excellence in HIV/AIDS, Simon FraserUniversity), Sophie Patterson (British Columbia Centre for Excellence in HIV/AIDS, Simon Fraser University), Peter Phillips (British Columbia Centre forExcellence in HIV/AIDS, University of British Columbia), Anita Rachlis(University of Toronto, OCS), Sean B. Rourke (University of Toronto, OCS),Hasina Samji (British Columbia Centre for Excellence in HIV/AIDS), MarekSmieja (McMaster University), Benoit Trottier (Clinique Medicale l’Actuel,Université de Montréal), Mark Wainberg (McGill University, Lady DavisInstitute for Medical Research), Sharon Walmsley (University Health Network,University of Toronto).Collaborators: Chris Archibald (Public Health Agency of Canada Centre forCommunicable Diseases and Infection Control), Ken Clement (CanadianAboriginal AIDS Network), Monique Doolittle-Romas (Canadian AIDS Society),Laurie Edmiston (Canadian Treatment Action Council), Sandra Gardner(OHTN, University of Toronto, OCS), Brian Huskins (Canadian TreatmentAction Council), Jerry Lawless (University of Waterloo), Douglas Lee(University Health Network, University of Toronto, ICES), Renee Masching(Canadian Aboriginal AIDS Network), Stephen Tattle (Canadian WorkingGroup on HIV & Rehabilitation), Alireza Zahirieh (Sunnybrook Health SciencesCentre).Analysts and Staff: Claire Allen (Regina General Hospital), Stryker Calvez(SHARE), Guillaume Colley (British Columbia Centre for Excellence in HIV/AIDS), Jason Chia (British Columbia Centre for Excellence in HIV/AIDS), DanielCorsi (The Ottawa Hospital Immunodeficiency Clinic, Ottawa HospitalResearch Institute), Louise Gilbert (Immune Deficiency Treatment Centre),Nada Gataric (British Columbia Centre for Excellence in HIV/AIDS), KatelynMerritt (British Columbia Centre for Excellence in HIV/AIDS), Lucia Light(OHTN), David Mackie (The Ottawa Hospital), Costa Pexos (McGill University),Susan Shurgold (British Columbia Centre for Excellence in HIV/AIDS), LeahSzadkowski (University Health Network), Chrissi Galanakis (Clinique MédicaleL’Actuel), Benita Yip (British Columbia Centre for Excellence in HIV/AIDS),Jaime Younger (University Health Network), and Julia Zhu (British ColumbiaCentre for Excellence in HIV/AIDS).FundingCANOC is funded by the Canadian Institutes of Health Research (CIHR)through a Centres Grant (Centres for HIV/AIDS Population Health and HealthServices Research [CIHR 02684]), two Operating Grants (HIV/AIDS PriorityAnnouncement [CIHR 134047]; Population and Public Health [CIHR 136882]),and is supported by the CIHR Canadian HIV Trials Network (CTN 242) and aFoundation Grant (Expansion of Antiretroviral Therapy and its Impact onVulnerable Populations in Canada and Global Settings [CIHR 143342]). JR issupported through an Ontario HIV Treatment Network (OHTN) Chair inBiostatistics. TA is supported by a New Investigator Award from OHTN/CIHR.AMB is supported by the Fondation Baxter and Alma Ricard Chair in InnerCity Health. ANB is supported by a CIHR New Investigator Award. CC issupported through an Applied HIV Research Chair from the OHTN. RSH issupported by a University Professorship at Simon Fraser University. MBK issupported by a Chercheur National Career Award from the Fonds derecherche du Québec - Santé (FRQ-S). JSGM is supported by the BritishColumbia Ministry of Health and by the US National Institutes of Health(R01DA036307). SW is supported through a Applied Health Research Chairfrom the OHTN.Availability of data and materialsThe data analysed during the current study are not publicly available due torestrictions from data sharing agreements among our institutions to protectthe privacy of study participants.Authors’ contributionTA and JR conceived of the study, LS conducted statistical analyses, and TA,JR and LS drafted the manuscript. RSH, JR, ANB, CC, MBK, ML, NM, JSGM andCT were involved in data acquisition. All authors reviewed iterations of theanalyses and drafts of the manuscript and approved the final version of themanuscript.Competing interestsCC has served on advisory boards for Abbvie and Gilead Sciences. MBK reportsgrants from Merck and ViiV Healthcare and personal fees for consultancy fromViiV Healthcare, Bristol-Meyers Squibb, and Merck. ML has served on advisoryboards and spoken at CME events for Viiv Healthcare, Abbvie, Merck CanadaInc. and Gilead Sciences. NM has been a speaker for Bristol-Meyers Squibb,Merck, and ViiV Healthcare. JSGM is supported with grants paid to his institutionby the British Columbia Ministry of Health and by the US National Institutes ofHealth (R01DA036307). He has also received limited unrestricted funding, paidto his institution, from Abbvie, Bristol-Myers Squibb, Gilead Sciences, Janssen,Merck, and ViiV Healthcare. JR is co-investigator on two projects with in-kindcontributions from Merck and Gilead Sciences. SW has served on advisoryboards and spoken at CME events for Viiv, Abbvie, Merck, Gilead, Jannsen, andBristol Meyers Squibb. AW has received honouraria for speaking events andadvisory meetings with Gilead Sciences, ViiV Healthcare, Bristol-Myers Squibb,Janssen, Abbvie, and Merck and is an investigator in clinical trials with GileadSciences, Merck and ViiV Healthcare. RSH has received funding from CanadianInstitutes of Health Research, through a Foundation Award, to support thisresearch and reports personal fees from Gilead Sciences. For the remainingauthors no conflicts of interest were declared.Consent for publicationNot applicable.Ethics approval and consent to participateEthics approval of the CANOC collaboration was granted by the SimonFraser University Research Ethics Board (REB) and the University of BritishColumbia REB. Approval from local institutional review boards (IRBs) wasgranted at each participating cohort site, including: Providence Health CareResearch Institute Office of Research Services, University Health Network(UHN) REB, Ottawa Hospital REB, University of Toronto HIV REB, Women’sCollege Hospital REB, Véritas IRB and Biomedical C REB of the McGillUniversity Heath Centre (MUHC).Written consent for study participation was obtained from all cohorts except:HAART Observational Medical Evaluation and Research Cohort (aninformation sheet is provided in lieu of a consent form; the IRB approves theretrospective use of anonymous administrative data without requiringconsent); Ottawa Hospital Cohort, UHN and Maple Leaf Medical Clinic (IRB/REBs approve the anonymous use of data retrospectively extracted fromclinical databases without requiring consent); and MUHC and the ElectronicAntiretroviral Therapy Cohort (patients sign a general waiver on starting amedical chart at the hospital, but no specific study-related consent isobtained; the IRB/REBs approve the anonymous use of data retrospectivelyextracted from clinical databases without requiring consent).Antoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 9 of 10Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Keenan Research Centre, Li Ka Shing Knowledge Institute, St. Michael’sHospital, Toronto, ON, Canada. 2Department of Family and CommunityMedicine, St. Michael’s Hospital, Toronto, ON, Canada. 3University of Toronto,410 Sherbourne Street, Toronto, ON ON M4X 1K2, Canada. 4Toronto GeneralResearch Institute, University Health Network, Toronto, ON, Canada. 5TheOttawa Hospital Research Institute, University of Ottawa, Ottawa, ON, Canada.6British Columbia Centre for Excellence in HIV/AIDS, St. Paul’s Hospital,Vancouver, BC, Canada. 7University of British Columbia, Vancouver, BC,Canada. 8Women’s College Hospital Research Institute, Toronto, ON, Canada.9Maple Leaf Medical Clinic, Toronto, ON, Canada. 10McGill University HealthCentre, McGill University, Montreal, Quebec, Canada. 11Clinique Médicalel’Actuel, Montreal, Quebec, Canada. 12Regina Qu’Appelle Health Region,Regina, SK, Canada. 13Simon Fraser University, Burnaby, BC, Canada.Received: 1 November 2016 Accepted: 4 April 2017References1. British HIV Association. British HIV Association guidelines for the treatmentof HIV-1 positive adults with antiretroviral therapy 2015. Available at http://www.bhiva.org/documents/Guidelines/Treatment/2015/2015-treatment-guidelines.pdf. Accessed 24 Jan 2016.2. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines forthe use of antiretroviral agents in HIV-1 infected adults and adolescents.Department of Health and Human Services. Available at http://www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf. Accessed 24 Jan2016.3. Shuter J. Forgiveness of non-adherence to HIV-1 antiretroviral therapy.J Antimicrob Chemother. 2008;61:769–73.4. Zeldin RK, Petruschke RA. Pharmacological and therapeutic properties ofritonavir-boosted protease inhibitor therapy in HIV-infected patients.J Antimicrob Chemother. 2004;53:4–9.5. Molina JM, Andrade-Villanueva J, Echevarria J, Chetchotisakd P, Corral J,David N, et al. Once-daily atazanavir/ritonavir compared with twice-dailylopinavir/ritonavir, each in combination with tenofovir and emtricitabine, formanagement of antiretroviral-naive HIV-1-infected patients: 96-week efficacyand safety results of the CASTLE study. J Acquir Immune Defic Syndr.2010;53:323–32.6. Orkin C, DeJesus E, Khanlou H, Stoehr A, Supparatpinyo K, Lathouwers E, etal. Final 192-week efficacy and safety of once-daily darunavir/ritonavircompared with lopinavir/ritonavir in HIV-1-infected treatment-naïve patientsin the ARTEMIS trial. HIV Med. 2013;14:49–59.7. Lennox JL, Landovitz RJ, Ribaudo HJ, Ofotokun I, Na LH, Godfrey C, et al. Aphase III comparative study of the efficacy and tolerability of three non-nucleoside reverse transcriptase inhibitor-sparing antiretroviral regimens fortreatment-naïve HIV-1-infected volunteers: a randomized controlled trial.Ann Intern Med. 2014;161:461–7.8. Mugavero MJ, May M, Ribaudo HJ, Gulick RM, Riddler SA, Haubrich R, et al.Comparative effectiveness of initial antiretroviral therapy regimens: ACTG5095 and 5142 clinical trials relative to ART-CC cohort study. J AcquirImmune Defic Syndr. 2011;58:253–60.9. Cuzin L, Pugliese P, Allavena C, Katlama C, Cotte L, Cheret A, et al.Comparative effectiveness of first antiretroviral regimens in clinical practiceusing a causal approach. Medicine. 2015;95:e1668.10. Johnston SS, Juday T, Esker S, Espindle D, Chu BC, Hebden T, et al.Comparative incidence and health care costs of medically attended adverseeffects among U.S. Medicaid HIV patients on atazanavir- or darunavir-basedantiretroviral therapy. Value Health. 2013;16:418–25.11. Nishijima T, Hamada Y, Watanabe K, Komatsu H, Kinai E, Tsukada K, et al.Ritonavir-boosted darunavir is rarely associated with nephrolithiasiscompared with ritonavir-boosted atazanavir in HIV-infected patients. PLoS One.2013;8:e77268.12. Aberg JA, Tebas P, Overton ET, Gupta SK, Sax PE, Landay A, et al. Metaboliceffects of darunavir/ritonavir versus atazanavir/ritonavir in treatment-naive,HIV type 1-infected subjects over 48 weeks. AIDS Res Hum Retrovir.2012;28:1184–95.13. Palmer A, Klein M, Raboud J, Cooper C, Hosein S, Loutfy M, et al. Cohortprofile: the Canadian Observational Cohort Collaboration. Int J Epidemiol.2011;40:25–32.14. Gray RJ. A class of k-sample tests for comparing the cumulative incidenceof a competing risk. Ann Stat. 1988;16:1141–54.15. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of acompeting risk. J Am Stat Assoc. 1999;94:496–509.16. Bartlett JW, Taylor JM. Missing covariates in competing risk analysis.Biostatistics. 2016;17:751-63.17. Rubin DB. Inference and missing data. Biometrika. 1976;63:581–92.18. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York: Wiley;1987.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Antoniou et al. BMC Infectious Diseases  (2017) 17:266 Page 10 of 10

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.52383.1-0362101/manifest

Comment

Related Items