WEB ANNEX B. SYSTEMATIC LITERATURE REVIEW AND NETWORK META-ANALYSIS ASSESSING FIRST-LINE ART TREATMENTS

Transcription

1 WEB ANNEX B. SYSTEMATIC LITERATURE REVIEW AND NETWORK META-ANALYSIS ASSESSING FIRST-LINE ART TREATMENTS Steve Kanters, Jeroen Jansen, Michael Zoratti, Jamie Forrest, Brittany Humphries, Jonathon Campbell In: Updated recommendations on first-line and second-line antiretroviral regimens and post-exposure prophylaxis and recommendations on early infant diagnosis of HIV: interim guidelines. Supplement to the 2016 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection

2 WHO/CDS/HIV/18.25 World Health Organization 2018 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. Kanters S, Jansen J, Zoratti M, Forrest J, Humphries B, Campbell J. Web Annex B. Systematic literature review and network meta-analysis assessing first-line antiretroviral treatments In: Updated recommendations on first-line and second-line antiretroviral regimens and post-exposure prophylaxis and recommendations on early infant diagnosis of HIV: interim guidelines. Supplement to the 2016 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. Geneva: World Health Organization; 2018 (WHO/CDS/HIV/18.25). Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at Sales, rights and licensing. To purchase WHO publications, see To submit requests for commercial use and queries on rights and licensing, see Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. 2

3 The mention of specific companies or of certain manufacturers products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication. This publication forms part of the WHO guideline entitled Updated recommendations on first-line and second-line antiretroviral regimens and post-exposure prophylaxis and recommendations on early infant diagnosis of HIV: interim guidelines. Supplement to the 2016 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. It is being made publicly available as supplied by those responsible for its development for transparency purposes and information, as required by WHO (see the WHO handbook for guideline development, 2nd edition (2014)). 3

4 CONTENTS Abbreviations Executive Summary Introduction Objectives Methodology Systematic literature review Sources Search strategy Study selection Study quality Data extraction Analyses Network meta-analyses Evaluation of consistency between direct and indirect comparisons Node definitions and backbone adjustments Models Adjusted analysis Presentation of results Software Adults and adolescents Systematic literature review study selection Analysis set study selection Results Viral suppression Increase in CD4 cell counts Mortality AIDS defining illnesses

5 Discontinuations Discontinuations due to adverse events Treatment-related and emergent adverse events Treatment-related and treatment-emergent serious adverse events Regimen substitutions GRADE tables TB Co-infected Individuals The INSPIRING trial Systematic literature review study selection Results Efficacy Tolerability Safety GRADE tables Pregnant and breastfeeding women Systematic literature review study selection Summary of the evidence base GRADE tables Children and adolescents Summary of the evidence base GRADE tables Discussion Conclusions Appendix A: Search strategy Appendix B: Search strategies for sub populations Appendix C: Trends in observed treatment effects across follow-up times Appendix D: List of included studies Appendix E: Characteristics of included studies Appendix F: Patient characteristics in included studies Appendix G: Quality assessments of included studies

6 Appendix H: Network diagrams TB Subpopulation Appendix I: Included trials by analysis Appendix J: Cross tables Appendix K: GRADE summary tables for children and adolescents References

7 Tables Table 1: Scope of the literature review in PICOS form Table 2: Summary of the GRADE quality of evidence assessments for all outcomes for the DTG vs EFV comparison Table 3: Summary of the GRADE quality of evidence assessments for all outcomes for the EFV400 vs EFV comparison Table 4: Summary of the GRADE quality of evidence assessments for all outcomes for the DTG vs EFV400 comparison Table 5: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 24 weeks from the fixed-effects network meta-analyses in HIV-TB co-infected patients Table 6: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 24 weeks from the fixed-effects network metaanalyses in HIV-TB co-infected patients Table 7: Data for treatment comparisons of interest for discontinuations due to adverse events outcome in HIV-TB co-infected patients Table 8: Data for treatment comparisons of interest for discontinuations due to adverse events outcome in HIV-TB co-infected patients Table 10: Data for treatment comparisons of interest for the treatment-emergent serious adverse events Table 11: Summary of the GRADE quality of evidence assessments for outcomes in the first-line treatment of HIV-TB co-infected patients Table 12: Summary of the Tsepamo study of DTG/TDF/FTC vs EFV/TDF/FTC in pregnant and breastfeeding women initiated on first-line ART Table 13: Summary of evidence among pregnant and breastfeeding women on first-line ART. 72 Table 14: Summary of the GRADE quality of evidence assessments for outcomes in the first-line treatment of pregnancy and breastfeeding women Table 15: Studies of RAL-based regimens for neonates and infants

8 Table 16: Summary of the GRADE quality of evidence assessments for efficacy, safety, and tolerability outcomes for the use of DTG vs EFV, LPV/r, and RAL in the first-line treatment of children and adolescents Table 17: Summary of the GRADE quality of evidence assessments for efficacy, safety, and tolerability outcomes for the use of RAL vs EFV, LPV/r, and NVP in the first-line treatment of children and adolescents Table 18: Systematic literature search strategy Table 19: Systematic literature search to identify non-rct study designs Table 20: Systematic literature review terms to identify the specific sub-populations of interest Table 21: List of included studies with corresponding publications Table 22: List of included studies with corresponding publications for the TB sub-population 118 Table 23: List of included studies with corresponding publications for the children and adolescent sub-population Table 24: List of included studies with corresponding publications for the pregnant and breastfeeding women Table 25: Trial characteristics for principal systematic literature review Table 26: Trial characteristics for studies selected in the systematic literature review among TB co-infected patients Table 27: Patient characteristics across the 76 randomized controlled trials included in the principal analysis Table 28: Patient characteristics in the principal analysis in the systematic literature review among TB co-infected patients Table 29: Cochrane risk of bias quality assessment for randomized controlled trials, arranged by review sub-population Table 30: Critical appraisal of non-randomized studies using the Tool to Assess the Risk of Bias in Cohort Studies, developed by the Clinical Advances through Research and Information Translation (CLARITY) group

9 Table 31: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 48 weeks from the fixed-effects network meta-analyses Table 32: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 96 weeks from the fixed-effects network meta-analyses Table 33: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 144 weeks from the fixed-effects network meta-analyses Table 34: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 48 weeks from network meta-analyses Table 35: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 96 weeks from network meta-analyses Table 36: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 144 weeks from network meta-analyses Table 37: Cross table of odds ratios with 95% credible intervals from network meta-analyses comparing ARVs in terms of discontinuation Table 38: Cross table of odds ratios with 95% credible intervals from network meta-analyses comparing ARVs in terms of discontinuation due to adverse events Table 39: Cross table of odds ratios with 95% credible intervals from the network meta-analyses comparing ARVs in terms of treatment-related adverse events Table 40: Cross table of odds ratios with 95% credible intervals from the network meta-analyses comparing ARVs in terms of treatment-emergent adverse events Table 41: Cross table of odds ratios with 95% credible intervals from network meta-analyses comparing ARVs in terms of treatment emergent serious adverse events Table 42: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 48 weeks from the fixed-effects network meta-analyses in HIV- TB co-infected patients Table 43: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 48 weeks from network meta-analyses in HIV-TB co-infected patients

10 Table 44: Cross table of odds ratios with 95% credible intervals from network meta-analyses comparing ARVs in terms of mortality in HIV-TB co-infected patients in HIV-TB co-infected patients Table 45: GRADE summary of evidence for the DTG vs EFV comparison in children and adolescents Table 46: GRADE summary of evidence for the DTG vs LPV/r comparison in children and adolescents Table 47: GRADE summary of evidence for the DTG vs RAL comparison in children and adolescents Table 48: GRADE summary of evidence for the RAL vs EFV comparison in children and adolescents Table 49: GRADE summary of evidence for the RAL vs LPV/r comparison in children and adolescents Table 50: GRADE summary of evidence for the RAL vs NVP comparison in children and adolescents

11 Figures Figure 1: Flow diagram for principal systematic literature review on adults and adolescents Figure 2: Network of all studies included in the principal analysis Figure 3: Forest plot of select ARVs comparisons with respect to viral suppression at A. 48 weeks and B. 96 weeks according to fixed-effects network meta-analysis Figure 4: Forest plot of select ARVs comparisons with respect to mean change in CD4 cell counts at A. 48 weeks and B. 96 weeks according to fixed-effects network meta-analysis Figure 5: Forest plot of select ARVs comparisons with respect to mortality according to fixedeffects network meta-analysis Figure 6: Forest plot of select ARVs comparisons with respect to the proportion of patients developing AIDS defining illnesses according to fixed-effects network meta-analysis Figure 7: Forest plot comparing pair-wise and NMA estimated relative effects of different ARVs with respect discontinuations (all cause) Figure 8: Forest plot comparing pair-wise and NMA estimated relative effects of select ARVs with respect discontinuations due to adverse events Figure 9: Forest plot of select ARVs comparisons with respect to A. treatment related adverse events and B. treatment emergent adverse events according to fixed-effects and randomeffects network meta-analysis Figure 10: Forest plot of select ARVs comparisons with respect to A. treatment related serious adverse events and B. treatment emergent serious adverse events according to fixed-effects network meta-analysis Figure 11: Forest plot comparing pair-wise and NMA estimated relative effects of different ARVs with respect regimen substitution (48 weeks) Figure 12: Modified FDA snapshot analysis of the percentage of participants (95% CI) with HIV-1 RNA <50 copies/ml Figure 13: Flow diagram for principal literature review on TB co-infected individuals and firstline ART regimens Figure 14: Complete network of evidence for patients with HIV-TB co-infection

12 Figure 15: Flow diagram for principal systematic literature review on pregnant and breastfeeding women and first line ART regimens Figure 16: Flow diagram for principal systematic literature review on adolescents and first line ART regimens Figure 17: Proportion of patients experiencing a treatment emergent serious adverse event or adverse event Figure 18: Proportion of patients experiencing a treatment related serious adverse event or adverse event Figure 19: Proportion of patients discontinuing treatment and discontinuing treatment due to an adverse event Figure 20: Proportion of patients that died and developed an AIDS defining illness Figure 21: Proportion of patients switching treatments during study Figure 22: Network diagram of trials informing viral suppression at 48 weeks (A); 96 weeks (B); and 144 weeks (C) Figure 23: Network diagram of the 66 trials informing mean change from baseline in CD4 cell counts at 48 weeks (A); 96 weeks (B); and 144 weeks (C) Figure 24: Network diagram of the trials informing mortality Figure 25: Network diagram of the comparative trials reporting AIDS defining illnesses outcome Figure 26: Network diagram of the trials informing discontinuation due adverse events among first-line HIV patients Figure 27: Network diagram of the trials informing retention among first-line HIV patients Figure 28: Network diagram of the trials informing the treatment-related adverse events analysis Figure 29: Network diagram of the trials informing the treatment-emergent adverse events analysis Figure 30: Network diagram of the trials informing the treatment-related serious adverse events analysis

13 Figure 31: Network diagram of the trials informing drug emergent serious adverse events among first-line HIV patients Figure 32: Network diagram of the trials informing regimen substitutions Figure 33: Network diagram of trials informing viral suppression at the 24-week (A) and 48- week (B) timepoints in HIV-TB co-infected patients Figure 34: Network diagram of the trials informing mean change from baseline in CD4 cell counts at 24-week (A) and 48-week (B) timepoints in HIV-TB co-infected patients Figure 35: Network diagram of the trials informing Mortality in HIV-TB co-infected patients. 176 Figure 36: Network diagram of the comparative trials reporting AIDS defining illnesses outcome in HIV-TB co-infected patients Figure 37: Network diagram of the trials informing discontinuation due adverse events in HIV- TB co-infected patients Figure 38: Network diagram of the trials informing retention among first-line HIV patients Figure 39: Network diagram of the trials informing the treatment-related serious adverse events analysis in HIV-TB co-infected patients Figure 40: Network diagram of the trials informing treatment emergent serious adverse events in HIV-TB co-infected patients

14 Abbreviations 3TC Lamivudine ABC Abacavir ADI AIDS-defining illness AIDS Acquired immunodeficiency syndrome ATV/r Ritonavir-boosted atazanavir ART Antiretroviral therapy ARV Antiretroviral agent/drug AZT Zidovudine BIC Bictegravir bid Twice daily CD4 Cluster of differentiation 4 CENTRAL Cochrane Central Register of Controlled Trials CI Confidence interval CLARITY Clinical Advances through Research and Information Translation CrI Credible interval CROI Conference on Retroviruses and Opportunistic Infections d4t Stavudine ddi Didanosine DIC Deviance information criterion DOR Doravirine DRV/r Ritonavir-boosted darunavir DTG Dolutegravir EFV Efavirenz Standard 600mg once daily dose EFV400 Efavirenz 400mg once daily dose EMBASE Excerpta Medica database ETR Etravirine EVG Elvitegravir FTC Emtricitabine 14

15 GRADE HBV HIV IAS IDU INSTI IQR ITT LPV/r MCMC MEDLINE NFV NMA NNRTI NRTI MD OR PI/r PICOS PLHIV qd RAL RCT RPV RR SD SE SLR TAF Grading of Recommendations Assessment, Development and Evaluation Hepatitis B virus Human immunodeficiency virus International AIDS Society Injection drug user Integrase strand transfer inhibitors Interquartile range Intention to treat Ritonavir-boosted lopinavir Markov Chain Monte Carlo Medical Literature Analysis and Retrieval System Online Nelfinavir Network meta-analysis Nonnucleoside reverse transcriptase inhibitor Nucleoside reverse transcriptase inhibitors Mean difference Odds ratio Ritonavir-boosted protease inhibitor Population, interventions, comparisons, outcomes, study design People living with HIV Once daily Raltegravir Randomized-controlled trial Rilpivirine Relative risk Standard deviation Standard error Systematic literature review Tenofovir alafenamide 15

16 TB TDF XTC Tuberculosis Tenofovir disoproxil fumarate Lamivudine or Emtricitabine 16

17 Executive Summary Background: In 2015, the WHO conducted evidence synthesis to update the 2013 Consolidated guidelines on the use of antiretrovirals for treating and preventing HIV. At the time, the combination of efavirenz, tenofovir disoproxil and lamivudine (or emtricitabine) [EFV + TDF + XTC] was the preferred first-line therapy. Results of the 2015 systematic literature review (SLR) and network meta-analysis (NMA) revealed improved tolerability and efficacy with dolutegravir (DTG) and low-dose efavirenz (EFV400). Despite this evidence, DTG and EFV400 were recommended as alternative first-line regimens rather than the preferred treatment. This was due to the high price of DTG and uncertainty around subpopulations; rendering it difficult to recommend for low and middle income countries under the public health approach. With numerous changes, including the availability of generic fixed dose combinations of DTG +TDF + XTC and a growing evidence base among sub-populations, we sought to update the SLR and NMA in order to determine the efficacy and safety of DTG and EFV400 relative to EFV. Objective: The objective of this project is to compare the efficacy and safety of first-line ART regimens. Given the knowledge accumulated through previous guidelines and knowledge of current literature results, this project is centred on the following research question: 1. Should DTG be recommended as the preferred first-line antiretroviral agent in combination with age-appropriate backbone (TDF + XTC for adults and adolescents) for the treatment of HIV? 2. Should EFV400 be preferred over EFV (standard-dose) for the first-line antiretroviral agent in combination with age-appropriate backbone for the treatment of HIV? Methods: Systematic database searches were conducted on 12 February 2018 to identify publications reporting on relevant randomised controlled trials (RCTs) in the following databases: MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials through Ovid. The current systematic review is an update on a review completed in May 2015 (Global Evaluative Sciences). Further manual searches of the 2016, 2017 and 2018 Conference on Retroviruses and Opportunistic Infections (CROI), the 2016 AIDS conference, and the 2017 International AIDS Society (IAS) conference were conducted. Additional studies were identified through a review of clinical trial registries and the reference lists of identified publications. All steps of the SLR were conducted independently and in duplicate. The same searches as in 2015 were used in Study eligibility was expanded to include newer treatments. Namely bictegravir, doravirine and tenofovir alafenamide. Data were extracted for trial characteristics, subject characteristics and outcomes. The outcomes included: viral suppression at all available thresholds, mean 17

18 change in CD4 cell counts, mortality, AIDS defining illnesses, retention, discontinuation due to adverse events, treatment related adverse events and regimen switching. This process was repeated for the TBHIV co-infection, pregnant and breastfeeding women, and children sub-populations. Data were analysed using NMA similar to in the 2015 evidence synthesis. NMA build upon the more traditional pairwise meta-analyses by considering all treatment simultaneously for the analysis of a single outcome. The analyses were performed in a Bayesian framework. For each outcome of interest, fixed- or random-effects models were applied. Given that the research questions focus on the third agents of ART with a specific backbone (XTC+ TDF), we chose to define the nodes in terms of specific antivirals rather than specific ART regimens. Defining nodes according to a single ARV rather than the full regimen significantly simplified the interpretation of modelling and results. The analyses used an arm-specific meta-regression adjustment to account for differences in backbones, which was critical given the importance of the SINGLE trial which compares DTG to EFV, but with different backbones. This approach allowed us to use SINGLE rather than ignore it. Analyses restricting to comparisons with the no differences in backbones were conducted as sensitivity analyses. Additionally, we conducted analyses that made adjustments for differences in baseline CD4, HIV RNA and proportion of males. To assess the overall quality of evidence, we used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system for rating overall quality of evidence with adjustments for the NMA methods employed. Adjustments to the direct or indirect estimate quality of evidence using network estimates could be rated up for gained precision and rated down for evidence of lack of transitivity. Results: A total of 2815 citations were identified through database searches for the SLR update. Based on the original review (May 2015) and the current update, 163 publications describing 90 trials were identified and included in the systematic literature review for adults and adolescents. Of the 17 new trials added to the evidence base, 4 included DTG (ARIA, GS-US ; GS-US , and SSAT066), 3 included DOR (DRIVE AHEAD, DRIVE FORWARD, and Study), 2 included BIC, and 3 were endonodal on EVG/c comparing TAF to TDF. Despite the large number of trials in the evidence base as a whole, there are three key trials that very much inform the comparisons of interest: SINGLE (DTG + ABC + XTC vs EFV + TDF + XTC), SPRING-1 (DTG + 2 NRTIs vs EFV + 2 NRTIs) and ENCORE1 (DTG + TDF + XTC vs EFV + TDF + XTC). These are the trials involved in the head to head comparisons of interest. No evidence was available to describe the use of EFV400 in any of the sub-populations and as such only the comparison of DTG vs EFV was considered in the sub-populations. 18

19 An analysis set was formed using a subset of the SLR. On the basis of our 2015 analysis, we removed studies of indinavir, fosamprenavir, unboosted atazanavir, saquinavir, nelfinavir, and triple NRTIs from the analysis set. Five trials were removed for having a raltegravir backbone, which could not be handled in the model. From the review update, we excluded: GS-US used a non-fda approved dose of BIC; and GS-US was an endonodal trial that did not connect to the overall network (cobicistat boosted darunavir). The network was well-connected, with EFV serving as the most well-connected node. Overall, the principal analysis set of studies included 65 trials in which 33,148 patients were randomized to 151 treatment arms (12 treatments). A combination of direct and indirect evidence was available for all treatment comparisons except those involving EFV400, bictegravir and rilpivirine. Results were quite similar to those from the 2015 review. All analyses appeared to meet the consistency assumption for NMA. There were a few differences. We note that while no analysis required metaregression adjustments previously, there was one analysis that did require an adjustment for imbalances in the proportion of males; namely, the analysis for discontinuations. Another difference is that the fixed-effects model was more often favoured in these analyses, while the random-effects were more commonly used in the 2015 analyses. This suggests a reduction in heterogeneity and it may be due to the removal of the older treatment nodes. There was high quality evidence of improved viral suppression (odds ratio [OR]: 1.93; 95% credible interval [CrI]: 1.52, 2.47 at 96 weeks), discontinuations (OR: 0.49; 95% CrI: 0.44, 0.62) and discontinuations due to AEs (OR: 0.30; 95% CrI: 0.19, 0.47) for DTG relative the EFV. This was supported by moderate quality evidence of improvements in CD4 cell counts (mean difference: 22.87; 95% CrI: 8.29, 37.40), and both treatment-related (OR: 0.33; 95% CrI: 0.25, 0.44) and treatment-emergent AEs (OR: 0.63; 95% CrI: 0.38, 1.11). Due to low numbers of events, imprecise estimates and some risk of bias, there was on low to very low quality evidence for efficacy at 144 weeks, mortality and ADIs, SAEs and regimen substitutions. For EFV400 relative to standard dose EFV, high quality evidence was obtained for discontinuations due to AEs (OR: 0.42; 95% CrI: 0.22, 0.77). Otherwise, efficacy and safety tended to have moderate quality evidence due to imprecision. Similar to above, mortality and ADIs, treatment related AEs and SAEs, and regimen substitutions had low to very low quality evidence. Finally with respect to adults and adolescents, due to the indirectness, there was no high quality evidence comparing DTG to EFV400. There was moderate evidence of DTG leading to less discontinuations and having better long-term viral suppression. 19

20 The results of the systematic literature review failed to identify a wealth of evidence for the treatment of patients with HIV-TB co-infection. However, an interim analysis from the ongoing INSPIRING trial was identified from the CROI 2018 conference. INSPIRING (NCT ) is a Phase III, open-label randomized controlled trial enrolling HIV-TB co-infected adult patients for treatment with twice-daily DTG 50 mg or once-daily EFV 600 mg. It included 113 patients and only the 24 week interim analysis results were available. The evidence base consisted of 1378 patients enrolled in 13 treatment arms across 6 RCTs. The evidence was limited to 5 treatments: NVP, DTG, EFV, and RAL (400 mg; 800 mg). No evidence was identified for patients treated with EFV400. There was no statistically significant difference between DTG and EFV (OR: 0.54; 95% CrI: 0.19, 1.57) or between RAL400 and RAL800; however, the estimate suggests lower odds of suppression (in accordance with the FDA Snapshot algorithm). This difference appears to be driven by the larger number of discontinuations among the DTG arm of the INSPIRING trial. DTG appeared to lead to larger increases in CD4 (mean difference: cells/mm 3 ; 95% CrI: 14.93, 89.61). Outside of the moderate quality evidence supporting DTG relative to EFV with respect to change in CD4 cell counts, the evidence was of low to very low quality based on very small event counts and short follow-up. Two studies were identified with respect to pregnant and breastfeeding women: the DolPHIN 1 trial and the Tsepamo study. The Tsepamo study was a large cohort study of 1,729 pregnant women initiating DTG/TDF/XTC and 4,593 women initiating EFV/TDF/XTC in Botswana. The proportion of pregnancies with any adverse birth outcome was similar across treatment arms with 33.2% of DTG-managed pregnancies and 35.0% of EFV-managed pregnancies resulting in an adverse outcome. Similarly, severe birth outcomes were reported in 10.7% of DTG-managed and 11.3% of EFV-managed pregnancies. For a variety of safety outcomes, there was moderate quality evidence due to the risk of bias associated with an observational study, though with respect to safety, a large cohort study such as this one is among the most desirable studies to have. Finally for children, there was a real lack of evidence with respect to DTG, which was only reported in a single study among treatment experienced patients (i.e., not eligible). No comparative studies included RAL. However, two RAL studies were identified in neonates and infants, which both concluded that RAL was tolerable among infants. Given the lack of comparative evidence in the children and adolescent subgroup, we reference the results of the adult and adolescent review and downgrade the quality of evidence by one level for indirectness. 20

21 Discussion: The purpose of this study was to support the 2018 update to the consolidated guidelines on the use of antiretrovirals for treating and preventing HIV with respect to the choice of first-line ART. The two questions of interest were whether DTG and/or EFV400, each with an XTC + TDF backbone, should be the preferred first-line ART regimen rather than their current designation of alternative first-line ART regimen. This extensive systematic literature review and network meta-analysis to evaluate the comparative efficacy and safety of these and other ART regimens drew strong conclusions about the improved efficacy and tolerability of DTG relative to EFV. Moreover, the evidence synthesis supports the use of DTG among sub-populations, which was not the case in Specifically, results of this study suggest comparable safety among pregnant women initiating treatment and results were not demonstrably worse among TB-HIV co-infected individuals. Unfortunately, evidence was lacking with respect to first-line children. Overall, the evidence more strongly supports the choice of DTG as the preferred first-line regimen. Our study has numerous strengths and limitations. First, the use of NMA allowed for analytic adjustments to account for differences in backbones and provide an unbiased estimate of the comparison between DTG and EFV despite the critical trial having different backbones. Second, by combining direct and indirect evidence, some of findings can be seen as having stronger evidence than previously perceived when strong findings are supported by both sources of evidence. With respect to limitations, first, the evidence for the comparisons of interest continued to be somewhat limited in subpopulations. For EFV400, it was completely missing. Most notably for DTG, there was an absence of evidence within children. This was also the case in people pre-exposed to ARVs, though that was somewhat expected. Even in pregnancy and TB, much of the evidence is still to come. Second, some significant outcomes were limited by a very low number of events, including mortality, regimen substitutions, serious adverse events, and ADIs. This influenced the precision of our estimates with respect to these outcomes and, in some cases, precluded the conduct of evidence synthesis through NMA. Third, treatment-related adverse events were both inconsistently defined and inconsistently reported. This limitation was mitigated by considering both treatment-related and treatment-emergent adverse events. Additionally, studies of shorter duration are, by their nature, less likely to identify adverse events than longer-term trials. Despite this, the evidence was collected through a rigorous systematic review process in accordance with the practices and recommendations set forth by the Cochrane Collaboration, including both broad and targeted searches of the literature, critical appraisal of the identified studies, and consultation with subject matter experts. 21

22 Dolutegravir in combination with lamivudine/emtricitabine and tenofovir disoproxil fumarate is an effective, safe and tolerable ART regimen. Across a variety of outcomes, evidence strongly suggests that it is superior to the current efavirenz-based preferred first-line ART regimen. With a new affordable generic fixed dose combination and comparable outcomes among sub-populations, the evidence supports the choice of a dolutegravir based preferred first-line regimen. Conclusions regarding low-dose efavirenz are unchanged since Low-dose efavirenz appears to be more tolerable, but with lack of evidence in sub-populations it is likely best to be considered an alternative first-line regimen. 22

23 1. Introduction The efficacy and safety of initial HIV antiretroviral therapy (ART) has important improved over the years and more than 17 million people living with HIV (PLHIV) are currently receiving life-saving ART. 1 Improvements in potency, tolerability, simplicity and availability of first-line ART have resulted in increased life expectancy and quality of life for PLHIV, when treatments are accessed in a timely and consistent manner. Hence, the selection of first-line ART has important public health and programmatic implications. With the effectiveness and safety of regimens as key considerations, many ART programmes, particularly in low- and middle-income settings, are influenced by the World Health Organization (WHO) ART guidelines. 2 Clinical guidelines are developed through multi-step processes that ensure that they are feasible within the current clinical environment and that they are evidence based. A key step involves evidence synthesis whereby all of the evidence is collected and analysed so as to provide an overview of the therapeutic landscape. In 2015, the WHO conducted evidence synthesis to update the 2013 Consolidated guidelines on the use of antiretrovirals for treating and preventing HIV. At the time, the hypothesis was that integrase inhibitor (INSTI) based regimens or low-dose efavirenz (EFV400) based regimens would challenge preferred recommended first-line regimen. The 2013 guidelines recommended, for adults and adolescents, a first-line ART consisting of two nucleos(t)ide reverse transcriptase inhibitors (NRTIs) and a non-nucleoside reverse transcriptase inhibitor (NNRTI). 3 In particular, the combination of efavirenz, tenofovir disoproxil and lamivudine (or emtricitabine) [EFV + TDF + XTC] was the preferred option as first-line therapy. 3 Results of the 2015 systematic literature review (SLR) and network meta-analysis (NMA) revealed improved tolerability and efficacy with INSTIs and EFV400, with dolutegravir (DTG) having the highest estimated tolerability and safety. 4 Despite this evidence, DTG and EFV400 were recommended as alternative first-line regimens rather than the preferred treatment. 5 This was due to uncertainty around sub-populations and an expensive price rendering it difficult to recommend for low and middle income countries. With numerous changes, including the availability of generic fixed dose combinations of DTG +TDF + XTC, the 2018 SLR and NMA aimed to determine the efficacy and safety of DTG and EFV400 relative to EFV600. This report presents the findings of an updated SLR and NMA that was commissioned to shed light on this very topic. 23

24 2. Objectives The objective of this project is to compare the efficacy and safety of first-line ART regimens. Given the knowledge accumulated through previous guidelines and knowledge of current literature results, this project is centred on the following research question: 3. Should DTG be recommended as the preferred first-line antiretroviral agent in combination with age-appropriate backbone (TDF + XTC for adults and adolescents) for the treatment of HIV? 4. Should EFV400 be preferred over EFV (standard-dose) for the first-line antiretroviral agent in combination with age-appropriate backbone for the treatment of HIV? 24

25 3. Methodology 3.1. Systematic literature review Table 1 describes the PICOS (population, interventions, comparator, outcomes, study design) criteria used to guide the selection of studies that were included in this systematic literature review. Note that both research questions are captured by this single PICOS. Table 1: Scope of the literature review in PICOS form Criteria Definition Population Inclusion criteria: Treatment-naïve adults and adolescents (12 years and above) living with HIV Subgroups of interest: Children Adolescents Pregnant and breastfeeding women TB co-infected patients People with prior exposure to ARVs Interventions DTG + 2NRTI EFV NRTI Raltegravir (RAL) + 2NRTI Elvitegravir boosted with cobicistat EVG/c + 2NRTI Bictegravir (BIC) + 2 NRTI Doravirine (DOR) + 2NRTI Rilpivirine (RPV) + 2 NRTI Nevirapine (NVP) + 2 NRTI Darunavir boosted with ritonavir (DRV/r) + 2 NRTI Atazanavir boosted with ritonavir (ATV/r) + 2 NRTI Lopinavir boosted with ritonavir (LPV/r) + 2 NRTI Comparator EFV NRTI Outcomes Viral suppression at 48 and 96 weeks Change from baseline CD4 at 48 and 96 weeks 25

26 Mortality Retention Discontinuations due to adverse events Treatment emergent adverse events Severe adverse events Development of drug resistance Study design Inclusion criteria: Randomized controlled trials (RCTs) Additionally, for subgroups: Single-arm non-randomized controlled trials Prospective and retrospective cohort studies Case-control studies Controlled and uncontrolled longitudinal studies (cohorts or case series) Language Only studies published in English will be included *Note: Except for DTG, EFV400 and EFV600 treatments are required to provide indirect evidence The population listed above, treatment-naïve adults and adolescents, represents the principal analysis for this evidence synthesis project. This systematic literature review includes reviews and analyses for a variety of sub-populations: Adolescents and children; TB co-infected patients; Pregnant and breastfeeding women; and people with prior exposure to ARVs. Although the principal inclusion criteria described above was broad enough to capture these sub-populations, less restrictive criteria on study design were required to obtain meaningful evidence on them. Thus, additional searches were conducted for each sub-population, with the exception of adults and people with prior exposure to ARVs. Note that no RCTs pertaining to or describing patients with prior exposure to ARVs were identified and hence no evidence base exists within the identified parameters for this sub-population. No analyses were conducted for this sub-population either. Treatments were differentiated according to the specific drugs, doses, and frequencies of administration. The only drugs that were considered interchangeable were lamivudine (3TC) and emtricitabine (FTC) due to their molecular likeness, referred to here as XTC. Non-standard doses were not considered reason for exclusion at the study selection process; however, non-standard doses that did not serve as connectors (i.e. were not compared to two or more treatments of interest) were excluded in the final 26

27 selection stage (following full text selection). ART regimens with a single antiviral agent and those with two agents that included one or more NRTI were not considered eligible. Similarly, with the exception of boosted regimens, ART regimens with four or more agents were not eligible (e.g. NNRTI+PI+2NRTI). Trials that had mixed backbones were included if the backbones were equally distributed across arms. Trials where backbones were selected prior to randomization were considered eligible. Trials failing to report on backbone distribution or reporting imbalanced backbone distributions were excluded. Further details on how regimens were defined for analytical purposes are provided below (Section 3.2.3). The eligibility criteria remained generally unchanged relative to the 2015 SLR; however, there were a few additions with respect to interventions. Two new treatments, BIC and DOR, were added to the network. The motivation for adding them is that they might provide additional indirect evidence to the comparisons of interest and that there may be some secondary utility to understanding their efficacy and safety relative to the treatments of interest. Additionally, the use of tenofovir alafenamide (TAF) as a backbone was now permitted given that the evidence base for this treatment has grown substantially since Sources A comprehensive systematic search of the literature was conducted on using the following databases: Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica database (EMBASE), and Cochrane Central Register of Controlled Trials (CENTRAL). The current systematic review is an update on a review completed in May 2015 (Global Evaluative Sciences). Therefore, searches were restricted from 01 January 2015 to the search date, 12 February Further manual searches of the 2016, 2017 and 2018 Conference on Retroviruses and Opportunistic Infections (CROI), the 2016 AIDS conference, and the 2017 International AIDS Society (IAS) conference were conducted. Conference abstracts identified through the EMBASE search were eligible for inclusion. Additional studies were identified through a review of clinical trial registries and the reference lists of identified publications Search strategy The general search strategy involved identifying papers according to the population of interest, and the inclusion of interventions and comparators of interest, and the restriction to randomized controlled trials. Population was identified as having HIV or AIDS and not being treatment experienced or failing 27

28 treatment. Our search further restricted on publication types that were not of interest (i.e., newsletters and reviews). The specific search strategy is presented in Appendix A. The additional search strategies for each subpopulation are presented in Appendix B Study selection Two investigators, working independently, scanned all abstracts and proceedings identified in the literature search. The same two investigators independently reviewed abstracts and proceedings potentially relevant in full-text. If any discrepancies occurred between the studies selected by the two investigators, a third investigator provided arbitration. Full-text screening was conducted for each specific question Study quality The validity of individual randomized controlled trials was assessed using the Risk of Bias instrument, endorsed by the Cochrane Collaboration. 6 This instrument is used to evaluate 7 key domains: sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessors; incomplete outcome data; selective outcome reporting; and other sources of bias. The validity of non-randomized studies, including single-arm trials, cohort studies, and observational study studies, were evaluated using the Tool to Assess the Risk of Bias in Cohort Studies, developed by the Clinical Advances through Research and Information Translation (CLARITY) group at McMaster University. This 8-item instrument is used to evaluate various aspects of the research design and study execution, including selection of patients, differences in patient characteristics, and the assessment of outcomes. We employed the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system for rating overall quality of evidence Most recently, GRADE has issued guidance on network meta-analysis. 13 We applied GRADE in the following manner. First, the GRADE system was applied to the direct evidence as is done with pairwise meta-analyses. If only indirect evidence existed, we used the NMA estimate and evaluated the shortest indirect pathway with the largest number of trials. As neither of these combined direct and indirect evidence, we referred to this stage as the assessment of uncombined evidence. Specifically, for each outcome, the rating began as high-quality evidence and were potentially rated down by one or more of five categories of limitations: (1) risk of bias, (2) consistency, (3) directness, (4) imprecision, and (5) reporting bias. The second stage involved rating the combined evidence, which is the NMA evidence for comparisons assessed according to pairwise meta- 28

29 analyses in the first stage. In this stage, we began with the score from the first stage. We rated down if the comparison was within a loop in which there was evidence on inconsistency (i.e. lack of transitivity) or if the estimate became imprecise. Alternatively, we rated up if a direct estimate that was graded down for imprecision in the first stage and became precise in the NMA. The quality of evidence for each main outcome can be determined after considering each of these elements, and categorized as either high (We are very confident that the true effect lies close to that of the estimate of the effect), moderate (We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different), low (Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect ), or very low (We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect) Data extraction Two investigators, working independently, extracted data on study characteristics, interventions, patient characteristics at baseline, and outcomes for the study populations of interest for the final list of selected eligible studies. Any discrepancies observed between the data extracted by the two data extractors were resolved by involving a third reviewer and coming to a consensus. Data is provided in a Microsoft Excel Workbook with sheets corresponding to the different information categories. For each continuous outcome, the change from baseline at the end of the randomized phase was extracted, along with the corresponding sample size, standard deviation (SD) for mean change from baseline and measures of uncertainty (i.e. standard error (SE), 95% confidence intervals (CI), and p-value) for all relevant intervention groups. If the change from baseline was not provided, we extracted the score at the follow-up time point of interest and the baseline score, and calculated the change in such cases, the standard error of change was estimated by combining the standard errors at both time points and using an outcome specific correlation coefficient (ρ) in the following equation: SE(θ) = Var(Y 1 ) + Var(Y 2 ) 2SD 1 SD 2 ρ / n The outcome specific correlation can be obtained by first deriving the correlation from studies that reported both change and measurements at the both time points. Given that this was not available for 29

30 the outcomes of interest, we used the conservative measure of 0.5. In cases where interquartile ranges (IQR) were provided, the length of the IQR was divided by 1.35 to estimate standard deviation. If the SE was not reported, it was calculated according to the following hierarchy: based on the reported 95% CI by intervention group; SD by intervention group along with sample size; 95% CI of the difference between intervention groups; p-values by intervention groups; p-values for the difference between intervention groups. In cases where standard deviations were not provided, the average standard deviation among reported studies was used. Measures of dispersion were imputed for trials in which dispersion measures were not reported. Mean standard deviation was used for imputation and standard errors were derived from these Analyses In this section, we describe the methods used to conduct the statistical analyses Network meta-analyses When sufficient data were available for quantitative evidence synthesis, a conventional pairwise metaanalysis was employed as a first step. When multiple treatments were available within the evidence base, we employed network meta-analyses (NMA). All NMAs were conducted within the Baian framework using Baian hierarchical models. Under the assumption of consistency, the NMA model relates the data from the individual studies to basic parameters reflecting the (pooled) relative treatment effect and safety profiles between interventions. Based on these parameters, the relative treatment effects between each of the contrasts in the network were obtained. For each outcome and subgroup of interest, fixed or random-effects models were applied. Model selection was conducted using the deviance information criterion (DIC) according to NICE conventions. 15 The DIC provides a measure of model fit that penalizes for model complexity. Through the use of the DIC, the fixed effects models were often favoured. Model fit was also assessed using leverage plots and any outliers identified in this fashion were investigated further. The model with the best fit was chosen as the primary analysis model. In situations with very limited and sparse data, a narrative review was used as an alternative to quantitative analysis. The latter were restricted to the sub-population analyses Evaluation of consistency between direct and indirect comparisons Prior to the NMA, the consistency between direct and indirect comparisons was evaluated for networks that consisted of closed loops. For each of the comparisons (i.e. contrasts) that were part of a closed 30

31 loop made up of more than 1 RCT, we split the available trials into direct and indirect information. For each contrast in question, two (pooled) relative treatment effect estimates were obtained, one with independent-means (or independent-effects) models using only the trials providing direct comparisons, and one based on an NMA of the remaining trials providing only indirect evidence. This iterative technique is called edge-splitting. 16 The difference in estimates generated by the two sets of evidence was evaluated with the Bucher test for inconsistency Node definitions and backbone adjustments Given that the research questions for this project concern third agent antivirals (i.e., non-backbone antivirals); we chose to define the nodes in terms of specific antivirals rather than specific ART regimens. All treatments with multiple standard doses or frequency of administration were not differentiated on this basis. For example, nevirapine 200 mg twice daily (bid) was considered equivalent to nevirapine 400 mg once daily (qd). The only treatment with multiple doses that were distinguished in the analysis was efavirenz (600 mg qd) and low dose efavirenz (400 mg qd). Defining nodes according to a single ARV rather than the full regimen importantly simplified the interpretation of modelling and results. Nonetheless it is important to account for differences in backbone therapies. RCTs that use the same backbone in all trial arms do not require any adjustment in terms of backbones; however, RCTs employing different backbones require adjustments in order to properly measure the effect attributable to the antiviral agent comparison being estimated. Two approaches were used to address differences in backbone regimens. First, backbone regimens were categorized as TDF + XTC (the reference category), TAF + XTC, abacavir (ABC) + XTC, zidovudine (AZT) + XTC, and as other. The other category included treatments such as stavudine (d4t) and didanosine (ddi) as well as the agents contained in the previous categories. We used arm-specific meta-regression to adjust estimates according to differences in backbones according to these categories. The alternative approach was to simply reduce the evidence base to trials that did not differ with respect to backbones. The most notable trial to differ in backbones was the SINGLE trial comparing EFV to DTG, 18 which is central to the research questions. Otherwise, trials that differed in backbones tended to be older or to be endonodal. Endonodal trials are those that compare a node to itself. Indeed, some trials differing in backbone only were included to improve the backbone meta-regression adjustments. Such trials, comparing EFV to EFV, were only of interest in the analysis using meta-regression adjustments for 31

32 differences in backbones. The adjusted model served as the primary analysis; however, in outcomes where differences in backbones were restricted to endonodal trials or a few older trials with dated regimens, the restricted model was used instead Models All outcomes were either binary or continuous. Viral suppression and CD4 outcomes were frequently reported at multiple time points and were analysed separately for each of the three time points of interest: 24 weeks, 48 weeks, and 96 weeks. The remaining outcomes tended to be reported at a single time point, which varied and typically coincided with trial duration. During the feasibility assessment stage, the relationship between follow-up time and outcomes was explored. The figures in Appendix C consider trends in both proportions and odds ratios across time. The odds ratios are the more important consideration given they represent the effect being modelled. In these figures odds ratios at multiple time points within a single trial were connected to further help determine whether follow-up time is an effect modifier to relative treatment effects. As can be seen in these figures, the odds ratios tend to be stable over time or include an equal amount of downward and upward trends. On this basis, we modelled the relative treatment effects on all remaining variables using the outcomes combined across multiple time points. For studies reporting one of these outcomes at multiple time points, the values at longest follow-up were used. For binary outcomes (mortality, AIDS defining illnesses, viral suppression, loss to follow-up, serious adverse events, and regimen substitutions) we used a logistic regression model with the logit link function and a binomial likelihood. We chose to present results as odds ratios (OR) for these models so as to avoid the ceiling effect that limits relative risks (RR) for outcomes with proportions around 0.8 to To test for the presence of heterogeneity both the fixed-effects and random-effects models were employed. For the random-effects model, the conventional non-informative prior, a uniform distribution between 0 and 2, was applied to the between-trial standard deviation. 15 For continuous outcomes (increase in CD4 count) we used linear regression models with an identity link and normal likelihood. The data was arm based, and we modelled the differences in change from baseline between all informed treatment comparisons. Estimates of comparative efficacy were represented as mean differences. 32

33 Adjusted analysis Adjusted analyses came in two flavours. First, we conducted meta-regression adjustments to evaluate whether differences in baseline CD4, baseline log transformed viral load, and proportion of males impacted relative efficacy and safety estimated. Second, we conducted sensitivity analyses. For viral suppression, we deemed the intention-to-treat (ITT) outcomes for our primary analysis and considered the per-protocol outcomes as a sensitivity analysis. Additionally, multiple cut-off values were reported for the definition of viral suppression. Newer trials tend to use a cut-off of <50 copies/ml, but some trials used higher cut-off values, <200 and <400 copies/ml, due to limited sensitivity of older assays. While the cut-off does affect the absolute count, we found no evidence to suggest that these alter relative treatment effects. Thus, for the primary analysis, all trials were included regardless of cut-off used, and as a sensitivity analysis only trials using the <50 copies/ml were included. In trials where multiple cut-off values were reported, <50 copies/ml was favoured to <200 copies/ml, which was favoured to <400 copies/ml Presentation of results The primary output of the Baian NMA are posterior distributions of the relative treatment effects between all interventions in the networks, e.g. odds ratios for discontinuation and mean change from baseline in CD4 cell counts. The results for all outcomes are presented with NMA cross-tables as OR or mean differences (MD). The posterior distributions of relative treatment effects and modelled outcomes were summarized by the median and 95% credible intervals (CrIs), which were constructed from the 2.5 th and 97.5 th percentiles. As this project pertains to questions particular to three treatments, forest plots are used to describe these in the main text and cross tables are provided in the Appendices for a more in-depth look Software The parameters of the different models were estimated using a Markov Chain Monte Carlo (MCMC) method implemented in the JAGS software package. A first series of iterations from the JAGS sampler were discarded as burn-in and the inferences were based on additional iterations using two chains. Convergence of the chains was confirmed by the Gelman-Rubin statistic. All analyses were performed using R version ( and JAGS version 4.3 (OpenBUGS Project Management Group). 33

34 4. Adults and adolescents 4.1. Systematic literature review study selection Following the PICOS outlined in Table 1 and employing the search strategies specified in Appendix A, a total of 2815 citations were identified through database searches for the SLR update; of these 204 were selected for full-text review. The flow diagram for study inclusion is presented Figure 1. Based on the original review (May 2015) and the current update, 163 publications describing 90 trials were identified and included in the systematic literature review We note that these trials are restricted to those of the primary analysis. Trials pertaining to subpopulations, such as TB, have been excluded from this study selection. Further details on these sub-populations are provided in Sections 5, 6 and 7. Of the 17 new trials added to the evidence base, 4 included DTG (ARIA, GS-US ; GS-US , and SSAT066), 34,55,124,148 3 included DOR (DRIVE AHEAD, DRIVE FORWARD, and Study), 58,59,99,117 2 included BIC, 55,148 and 3 were endonodal on EVG/c comparing TAF to TDF. 24,147,149 With respect to the DTG trials, ARIA compared DTG to ATV/r among a sample of women only; SSAT066 compared DTG to RAL and EVG/c; while the remaining two trials were comparisons to BIC. Standard dose efavirenz continued to be central as it was included in 5 trials; however, there were no new trials comparing EFV to DTG or to EFV400. Just as in 2015, ENCORE1 was the only trial that included EFV400, pitting it against standard dose EFV. The NAMSAL trial (NCT ) is ongoing aiming to compare EFV400 to DTG and results are expected later this year. As such, despite the large number of trials in the evidence base as a whole, there are three key trials that very much inform the comparisons of interest: SINGLE (DTG + ABC + XTC vs EFV + TDF + XTC), SPRING-1 (DTG + 2 NRTIs vs EFV + 2 NRTIs) and ENCORE1 (DTG + TDF + XTC vs EFV + TDF + XTC). 18,21,22,64,159,166,170,171 These are the trials involved in the head to head comparisons of interest. They were all multinational, double-blind, placebo-controlled randomized trials. SRING-1 was a Phase II trial, while the others were Phase III trials. Spring-1 randomized 205 patients, SINGLE randomized 833 patients and ENCORE1 randomized 636 patients. Writing in The Lancet Infectious Diseases, the ENCORE1 study investigators concluded that EFV400 was non-inferior to EFV over 96 weeks with fewer treatment-related adverse events and supported the use of EFV400 in routine care. While positive conclusions were also drawn for the SINGLE trial, there was confounding due to the difference in backbones. This is why the adjustments for differences in backbones were critical to this evidence synthesis. 34

35 We note that no evidence was available to describe the use of EFV400 in any of the sub-populations and as such only the comparison of DTG vs EFV was considered in the sub-populations. The full list of included trials, including the studies included in the various sub-population reviews, is provided in Appendix D. 35

36 Figure 1: Flow diagram for principal systematic literature review on adults and adolescents

37 4.2. Analysis set study selection Feasibility, applicability, and relevance considerations led to the removal of several studies or study arms from the analysis set. Following our 2015 analysis, it became clear that older treatments that are no longer used in most settings had very little connectivity to the network and therefore provided negligible additional information regarding the research questions. In fact, at times they displayed evidence of loop inconsistency (disagreement between direct and indirect evidence). As such, we chose to remove studies of indinavir, fosamprenavir, unboosted atazanavir, saquinavir, nelfinavir, and triple NRTIs from the analysis set. To be clear, these were included in the SLR, but were removed from the analysis set. Five trials were excluded for having a RAL backbone, which could not be handled in the model. From the review update, GS-US and GS-US were excluded: GS-US used a non-fda approved dose of BIC; and GS-US was an endonodal trial that did not connect to the overall network (Cobicistat boosted DRV). Figure 2 presents the analysis set represented as a network diagram.

38 Figure 2: Network of all studies included in the principal analysis Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. The network was well-connected, with EFV serving as the most well-connected node. Overall, the principal analysis set of studies included 65 trials in which 33,148 patients were randomized to 151 treatment arms (12 treatments). A single study compared EFV to EFV400, with no indirect evidence identified. A combination of direct and indirect evidence was available for all other treatment comparisons except BIC and RPV. We kept both so as to have comparisons of these with the remainder of the therapeutic landscape, but their inclusion provided no additional information to the comparisons of interest. Summaries of trial characteristics, patient characteristics, and critical appraisal quality assessments are presented in Appendices E, F, and G, respectively. With respect to trial characteristics, trials ranged from 38

39 Phase II to IV, but most were Phase III, double-blinded and multinational. With respect to baseline characteristics, there were a number of notable differences across trials. Most notably, sex varied from all females to all males. Mean CD4 varied from as low as 102 cells/mm 3 (PHIDISA II) to cells/mm 3 (GS-US ). Similarly, baseline HIV RNA varied from 4.28 (Epzicom-Truvada) to 5.48 (ADVANZ; a new trial). There were also notable differences with respect risk groups and other markers of disease severity, but age, sex, CD4 and viral load were the best reported and the ones that were explored further through meta-regression. Overall, study quality was generally high (i.e., low risk of bias). Exceptions were restricted to open-label trials having a high risk of bias due to blinding and some of the more recent trials that were only reported upon in posters having insufficient information to determine with certainty that the risk of bias was either low or high Results Results were quite similar to those from the 2015 review. We note that while no analysis required metaregression adjustments previously, there was one analysis that did require an adjustment for imbalances in the proportion of males; namely, the analysis for discontinuations. For all other analyses, the unadjusted model was favoured. Another difference is that the fixed-effects model was more often favoured in these analyses, while the random-effects were more commonly used in the 2015 analyses. This suggests a reduction in heterogeneity and it may be due to the removal of the older treatment nodes. In the following subsections we present results for each outcome. The GRADE tables summarizing the overall evidence are then presented in Section 4.4. Prior to providing details regarding the analysis of each outcome individually, consider a few general remarks. The network diagrams for each analysis are provided in Appendix H. To further facilitate the reader in identifying which trials were used for the analyses of each outcome, Appendix I lists the trials and indicates which outcome analysis each trial was included in. All analyses appeared to meet the consistency assumption for NMA. There were very few exceptions and these were restricted to specific loops in the treatment related SAEs. In our judgement, these did not justify a need to use an alternative analysis to the NMA. Nonetheless, we adjusted the GRADE score by reducing the score for network transitivity where these concerns arose. 39

40 Viral suppression Viral suppression was among the best reported outcomes. The definition of viral suppression was a composite of the various thresholds reported in the included studies (i.e. <20-50 copies/ml; <200 copies/ml; <400 copies/ml). The <50 ml threshold was favoured and was by far the most commonly reported threshold. Sensitivity analyses restricting the evidence base to only trials using the <50 ml threshold yielded similar results, thus supporting the composite approach (results not shown). Additionally, many trials used the Food and Drug Administration Snapshot algorithm, whereby discontinuations are considered failures. This approach was used throughout the evidence base, even in trials that did not explicitly use this approach. A consequence of this approach is that differences in viral load suppression can either be driven by differences in efficacy (i.e., improved ability of the drug to suppress the virus) or differences in tolerability (i.e., an increased propensity to stay on the drug) or both. As such, while we consider viral suppression an efficacy outcome, in actuality it is difficult to disaggregate efficacy from tolerability and this should be kept in mind while interpreting the results. The reported analyses are restricted to intention to treat (ITT) results, but per-protocol results were also analysed (results not shown). The per-protocol results did not resolve the issue around determining whether differences are due to efficacy or tolerability. We note that Walmsley et al note that a large difference between DTG and EFV in the SINGLE trial was due to tolerability. The networks used for analysis at the 48, 96, and 144-week timepoints are presented in Error! Reference source not found. in Appendix H. Notably, evidence on EFV400 was only available at the 46 and 96- week time points. For the analysis of viral suppression at 48 weeks, evidence was derived from 53 trials of 115 treatments arms including 26,410 patients. The results of the fixed-effects NMA for the comparisons of interest are presented in Figure 3, below, and all comparisons in Error! Reference source not found. of Appendix J. Based on the available evidence, DTG was statistically significantly more effective than standard dose EFV in achieving viral suppression at 48 weeks (OR: 1.86; 95% CrI: 1.44, 2.40). In fact, it was statistically superior to all other treatments except EFV400 (OR 1.61, 95% CrI 0.97, 2.70) against which it was marginally significantly better. The comparison between DTG and EFV400 was based only on indirect evidence. EFV400 was not statistically distinguishable from standard dose EFV. For the analysis of viral suppression at 96 weeks, evidence was derived from 28 trials of 63 treatments arms including 16,495 patients. The results of the fixed-effects NMA for the comparisons of interest are also presented in Figure 3, below, and all comparisons in Error! Reference source not found. of Appendix J. Treatment was DTG was associated with a higher proportion of patients achieving viral 40

41 suppression compared to all other treatments, though the comparison to RPV was not statistically significant (OR 1.44, 95% CrI 0.98, 2.10). At this time point, treatment with DTG was associated with a statistically significantly higher proportion of patient achieving viral suppression compared to EFV400 (OR 2.00, 95% CrI 1.20, 3.37). Similarly to at 48 weeks, EFV400 was not statistically distinguishable from standard dose EFV. The analysis of viral suppression at 144 weeks was based on evidence from 6 trials of 13 treatment arms enrolling 5,274 patients. Given that EFV400 was not available at this time point, results of the fixedeffects NMA for viral suppression at 144 weeks are only presented in Error! Reference source not found. of Appendix J. DTG continued to demonstrate superior viral suppression, although only half the comparisons were statistically significant. This was likely due to the smaller number of trials and patients. Nonetheless, among the significant comparisons was that against EFV (OR: 1.44; 95% CrI: 1.08, 1.94). Figure 3: Forest plot of select ARVs comparisons with respect to viral suppression at A. 48 weeks and B. 96 weeks according to fixed-effects network meta-analysis Increase in CD4 cell counts The network of evidence for change in CD4 cell count at 48 weeks is based on 44 trials comprising 94 treatment arms including 23,789 patients. The results of the fixed-effects NMA for the comparisons of 41

42 interest are presented in Figure 4, below, and all comparisons in Error! Reference source not found. of Appendix J. Based on the available evidence, DTG was statistically significantly more effective than standard dose EFV in increasing CD4 at 48 weeks (MD: 22.87; 95% CrI: 8.29, 37.40), as was EFV 400 (MD: 25.43; 95% CrI: 6.93, 43.97). As a result, both treatments were comparable to one another. Both treatments had higher estimated increases in CD4 than almost all other treatments, albeit not statistically significant. The network of evidence for change in CD4 cell count at 96 weeks is based on 22 trials comprising 47 treatment arms including 15,134 patients. The results of the fixed-effects NMA for the comparisons of interest are also presented in Figure 4, below, and all comparisons in Error! Reference source not found. of Appendix J. Results at this time point were very similar to those at 48 weeks, suggesting an improvement over the first year that is sustained in the second. The network of evidence for change in CD4 cell count at 144 weeks is based on 7 trials comprising 15 treatment arms including 7,019 patients. Given that EFV400 was not available at this time point, results of the fixed-effects NMA for viral suppression at 144 weeks are only presented in Error! Reference source not found. of Appendix J. The improvement in CD4 for DTG relative to EFV increased to cells/mm 3 (95% CrI: 19.51, 79.39); however, we caution that network was much more sparse than at earlier time points. 42

43 Figure 4: Forest plot of select ARVs comparisons with respect to mean change in CD4 cell counts at A. 48 weeks and B. 96 weeks according to fixed-effects network meta-analysis Mortality The evidence base for mortality in first-line treatment consisted of 21,604 patients enrolled in 29 trials consisting of 62 treatment arms. The network of evidence is presented in Error! Reference source not found. of Appendix H. Although mortality was considered a very important outcome by the guideline development group, the trials were underpowered for this outcome. Mortality across trials was low, with the exception of PHIDISA II which compared EFV (106/872, 12.2%) to LPV/r (102/873, 11.7%). In all, there were 391 deaths in the evidence base, but many comparisons with 0 events can render estimates unreliable. Consider that there were only 5 deaths in each of DTG and EFV400 in the evidence base As a result Given the small number of events, there are limitations to synthesizing the evidence through an NMA which may produce wide credible intervals. However, select comparisons are presented in Error! Reference source not found.. There was no statistically significant difference between mortality outcomes in patients treated with DTG, EFV, and EFV

44 Figure 5: Forest plot of select ARVs comparisons with respect to mortality according to fixed-effects network meta-analysis AIDS defining illnesses The evidence base for AIDS defining illnesses in first-line treatment consisted of 9,722 patients enrolled in 18 trials consisting of 40 treatment arms. The network of evidence is presented in Error! Reference source not found. of Appendix H. Similar issues to those seen in mortality were present in the ADI analysis. For example, 12.3% (14/144) of patients enrolled in the EFV arm of the Altair trial reported ADIs while no patients (0%) reported ADIs in the EFV arm of GS-US (0/352), SPRING-1 (0/50), and Protocol 004 (0/38) trials. Few ADIs were reported for patients treated with DTG (SPRING-1, 2/51, 3.9%) and for patients treated with EFV400 (ENCORE1, 14/321, 4.4%). Given the small number of events, there are limitations to synthesizing the evidence through an NMA which may produce wide and nonmeaningful CrIs. However, select comparisons are presented in Error! Reference source not found.. There was no statistically significant difference between mortality outcomes in patients treated with DTG, EFV, and EFV400. Figure 6: Forest plot of select ARVs comparisons with respect to the proportion of patients developing AIDS defining illnesses according to fixed-effects network meta-analysis 44

45 Discontinuations The evidence base for all-cause discontinuations (retention) was based on 26,399 patients enrolled across 120 treatment arms in 54 trials. The network of evidence is presented in Error! Reference source not found. of Appendix H. The results of the fixed-effects NMA for the comparisons of interest are presented in Figure 8, below, and all comparisons in Error! Reference source not found. of Appendix J. Recall that this was the outcome that required an adjustment for the proportion of males in the trials. Based on the available evidence, DTG was statistically significantly more effective than standard dose EFV in preventing discontinuations (OR: 0.49; 95% CrI: 0.44, 0.62). In fact, it was statistically superior to all other treatments except EVG/c, BIC, and DOR against which it still had a lower estimate of discontinuation. The comparison between DTG and EFV400 was based only on indirect evidence. EFV400 was not statistically distinguishable from standard dose EFV (OR: 0.91; 95% CrI: 0.50, 2.08); however, we note that DTG did appear to be superior to EFV400. Figure 7: Forest plot comparing pair-wise and NMA estimated relative effects of different ARVs with respect discontinuations (all cause) 45

46 Discontinuations due to adverse events The evidence base for discontinuations due to adverse events was based on 54 trials of 26,165 patients enrolled in 118 treatment arms. The network of evidence is presented in Error! Reference source not found. of Appendix H. Figure 8: Forest plot comparing pair-wise and NMA estimated relative effects of select ARVs with respect discontinuations due to adverse events The results of the fixed-effects NMA for the comparisons of interest are presented in Figure 9, below, and all comparisons in Error! Reference source not found. of Appendix J. It is for this outcome that both DTG and EFV400 shine, with both having the lowest odds of discontinuation due to adverse events and both being superior to standard dose EFV. This time EFV400 and DTG were not statistically differentiable Treatment-related and emergent adverse events The evidence for treatment-related adverse events included 15,599 patients enrolled in 61 treatment arms across 27 trials and that for emergent adverse events included 18,915 patients in 70 treatment arms across 32 trials. The results of the fixed-effects NMAs are presented in Error! Reference source not found. and Error! Reference source not found.. Key comparisons are presented in Figure 9. While none of the treatments were distinguishable with respect to treatment emergent AEs, both DTG and EFV400 had lower odds of leading to a treatment-related AE. Moreover, DTG had lower odds than EFV400. Overall, BIC had the lowest odds of treatment-related AEs, followed by DOR, DRV/r and then DTG. Treatments were generally less distinguishable with respect to emergent AEs. 46

47 Figure 9: Forest plot of select ARVs comparisons with respect to A. treatment related adverse events and B. treatment emergent adverse events according to fixed-effects and random-effects network meta-analysis 47

48 Treatment-related and treatment-emergent serious adverse events The evidence for treatment-related SAEs was based on 8,041 patients enrolled in 31 trial arms across 15 trials and for treatment-emergent SAEs was based on 26,706 patients enrolled in 98 trial arms across 45 trials. With only 81 treatment related SAEs reported across the evidence base, there were too few events to obtain reliable estimates. Results of the analysis are nonetheless presented in Figure 10. Treatment emergent SAEs did not have the same limitation. Full results are presented in Error! Reference source not found.. Again, neither DTG nor EFV400 distinguished themselves from EFV, though the estimates were lower. Figure 10: Forest plot of select ARVs comparisons with respect to A. treatment related serious adverse events and B. treatment emergent serious adverse events according to fixed-effects network metaanalysis Regimen substitutions The evidence base for regimen substitutions by 48 weeks was based on 9,263 patients enrolled in 18 trials across 41 treatment arms. The network of evidence is illustrated in Error! Reference source not found. of Appendix H. Figure 11 displays some of the key comparisons. Results of this analysis do not match what is seen in practice. That is to say that in practice, regimen substitutions are less common than with EFV. It is important to note that there was no direct evidence supporting the DTG to EFV comparison and that there were very few observed regimen substitutions in the trials that did.

49 Figure 11: Forest plot comparing pair-wise and NMA estimated relative effects of different ARVs with respect regimen substitution (48 weeks) 4.4. GRADE tables The summary of evidence and overall quality of evidence for each outcome are presented for the DTG to EFV comparison in Table 2, for the EFV400 to EFV comparison in Table 3, and for the DTG to EFV400 comparison in Table 4. There was high quality evidence of improved viral suppression, discontinuations and discontinuations due to AEs for DTG relative the EFV. This was supported by moderate quality evidence of improvements in CD4 cell counts, and both treatment-related and treatment-emergent AEs. Due to low numbers of events, imprecise estimates and some risk of bias, there was on low to very low quality evidence for efficacy at 144 weeks, mortality and ADIs, SAEs and regimen substitutions. Due to the high propensity of zero counts, results of the mortality, ADIs, treatment-related SAEs and regimen substitutions should be interpreted with great care (if not ignored). For EFV400 relative to standard dose EFV, high quality evidence was obtained for discontinuations due to AEs. Otherwise, efficacy and safety tended to have moderate quality evidence due to imprecision. Similar to above, mortality and ADIs, treatment related AEs and SAEs, and regimen substitutions had low to very low quality evidence. Finally, due to the indirectness, there was no high quality evidence comparing DTG to EFV400. There was moderate evidence of DTG leading to less discontinuations and having better long-term viral suppression. 49

50 Table 2: Summary of the GRADE quality of evidence assessments for all outcomes for the DTG vs EFV comparison Uncombined Estimates Combined Estimates Outcome Direct Effect Risk of Bias Incons istenc y Indirec tness Impre cision Public ation Bias Quality of direct evidenc e Odds ratio (95% CrI) Absolute effects Indirect evidenc e precisio n Networ k Transiti vity Overall quality of evidenc e Viral supp. at 48 weeks 1.79 (1.25, 2.58) High 1.86 (1.44, 2.40) 74 per 1,000 (47 to 98) 0 0 High Viral supp. at 96 weeks 1.65 (1.21, 2.24) High 1.93 (1.52, 2.47) 94 per 1,000 (63 to 121) 0 0 High Viral supp. at 144 weeks 1.44 (1.08, 1.92) Low 1.44 (1.08, 1.92) 39 per 1,000 (-11 to 83) 0 0 Low Change in CD4 at 48 weeks (36.80, 80.71) High cells/ml (5.89, 43.18) 0-1 Modera te Change in CD4 at 96 (13.83, High cells/ml Modera

51 Uncombined Estimates Combined Estimates Indirect Overall Quality Networ Incons Public evidenc quality Outcome Direct Effect Risk of Indirec Impre of direct Odds ratio Absolute k istenc ation e of Bias tness cision evidenc (95% CrI) effects Transiti y Bias precisio evidenc e vity n e weeks 69.69) (2.95, 48.74) te Change in (16.70, Low cells/ml Low CD4 at ) (20.12, weeks 79.58) per 0 0 Mortality (0.01, 4.16) Low (0.09, 4.87) 1,000 (-12 to 9) Low per 0 0 AIDS defining (0.24, Low (0.46, 1,000 Low illness ) 66.45) (-13 to 550) per 0 0 Discontinuati (0.44, 0.79) High (0.39, 0.62) 1,000 High ons (-54 to - 101) Discontinuati per

52 Uncombined Estimates Combined Estimates Indirect Overall Quality Networ Incons Public evidenc quality Outcome Direct Effect Risk of Indirec Impre of direct Odds ratio Absolute k istenc ation e of Bias tness cision evidenc (95% CrI) effects Transiti y Bias precisio evidenc e vity n e ons due to AE (0.15, 0.45) High (0.19, 0.47) 1,000 (-30 to -54) High per 0 0 Emergent (0.77, 1.58) Low (0.49, 1.23) 1,000 Low SAEs (-45 to 6) per 0 0 Emergent AEs (0.27, 1.54) Modera te (0.38, 1.11) 1000 (-160 to 70) Modera te per 0-1 Treatmentrelated SAEs (0.05, 1.03) Very (0.02, 1,000 Very low ) (-12 to 822) low per 0 0 Treatmentrelated AEs (0.29, 0.50) Modera te (0.25, 0.44) 1000 (-256 to - Modera te 170) Regimen substitutions Very 6.20 (0.87, 436 per 1, Very 52

53 Uncombined Estimates Combined Estimates Indirect Overall Quality Networ Incons Public evidenc quality Outcome Direct Effect Risk of Indirec Impre of direct Odds ratio Absolute k istenc ation e of Bias tness cision evidenc (95% CrI) effects Transiti y Bias precisio evidenc e vity n e low 74.94) (-53 to 925) low Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 53

54 Table 3: Summary of the GRADE quality of evidence assessments for all outcomes for the EFV400 vs EFV comparison Outcome Direct Effect Uncombined Estimates Combined Estimates Risk of Incons Indirec Imprec Public Quality of Odds Absolute Indirect Network Overall Bias istency tness ision ation direct ratio effects evidenc Transitivit quality Bias evidence (95% CrI) e y of precisio eviden n ce Viral supp. at per weeks (0.75, 1.79) Moderate (0.74, 1,000 Modera 1.79) (-48 to 71) te Viral supp. at per weeks (0.61, 1.52) Moderate (0.61, 1,000 Modera 1.52) (-48 to 71) te Change in CD4 at 48 (6.57, 43.43) Moderate cells/ml Modera weeks (-1.27, te 51.49) Change in CD4 at 96 (2.95, 49.05) Moderate cells/ml Modera weeks (4.30, te 49.77) Mortality per 1,

55 AIDS defining illness Discontinuati ons Discontinuati ons due to AE Emergent SAEs Emergent AEs Treatmentrelated SAEs Treatmentrelated AEs (0.28, 3.36) Low (0.11, 7.69) (0.78, 4.94) Low (0.82, 5.57) (0.50, 1.68) Moderate (0.50, 1.68) (0.23, 0.79) High (0.22, 0.77) (0.40, 1.22) Moderate (0.33, 1.47) (0.47, 1.43) Moderate (0.31, 2.13) (0.09, 2.63) Very low (0.01, 15.13) (0.51, 0.96) Low (0.51, (-23 to 8) Low 20 per 0 0 1,000 Low (-4 to 81) -6 per 0 0 1,000 Modera (-64 to 96) te -35 per 0 0 1,000 High (-50 to -13) -30 per 0 0 1,000 Modera (-64 to 21) te -30 per 0 0 1,000 Modera (-133 to 44) te -6 per 0-1 1,000 Very (-15 to 15) low -80 per 0 0 1,000 Low 55

56 0.96) (-144 to -9) Regimen per 0 0 substitutions (0.33, 1.33) Moderate (0.32, 1,000 Modera 1.32) (-42 to 17) te Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 56

57 Table 4: Summary of the GRADE quality of evidence assessments for all outcomes for the DTG vs EFV400 comparison Uncombined Estimates Combined Estimates Outcome Direct Effect Risk of Bias Inconsiste ncy Indirectne ss Imprecisio n Publicatio n Bias Quality of direct evidence Odds ratio (95% CrI) Absolute effects Indirect evidence precision Network Transitivity Overall quality of evidence Viral supp. at 48 weeks Viral supp. at 96 weeks Low Moderate 1.61 (0.97, 2.70) 2.00 (1.20, 3.37) 53 per 1,000 (-4 to 126) 100 per 1,000 (23 to 194) Low Moderate Change in CD4 at 48 weeks Change in CD4 at 96 weeks Mortality AIDS defining illness Discontinuations Discontinuations due to AE Emergent SAEs Emergent AEs Treatment-related SAEs Treatment-related AEs Regimen substitutions Low Low Very low Very low Moderate Low Low Low Very low Low Very low (0.04, 13.30) 3.88 (0.19, 41.03) 0.45 (0.23, 0.88) 0.72 (0.33, 1.58) 1.12 (0.46, 2.68) 0.78 (0.26, 2.41) (0.02, 4256) 0.48 (0.31, 0.73) 9.43 (1.16, ) cells/ml (-33.16, 31.78) cells/ml (-33.63, 31.3) -4 per 1,000 (-14 to 27) 103 per 1,000 (56 to 530) -43 per 1,000 (-54 to 30) -7 per 1,000 (-31 to 9) 8 per 1,000 (-47 to 52) -82 per 1,000 (-173 to 30) 131 per 1,000 (-9 to 827) -134 per 1,000 (-215 to -58) 240 per 1,000 (8 to 779) Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. Low Low Very low Very low Moderate Low Low Low Very low Low Very low 57

58 The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 58

59 5. TB Co-infected Individuals The results of the systematic literature review failed to identify a wealth of evidence for the treatment of patients with HIV-TB co-infection. However, an interim analysis from the ongoing INSPIRING trial was identified from the CROI 2018 conference. Given the direct relevance and impact the findings of this trial have in relation to the research question, we first provide an overall of the interim findings and supplement this with an NMA of all available evidence to provide an overview of the evidence landscape The INSPIRING trial In this review, the evidence for HIV-TB co-infected patients treated with DTG is based on a 24-week interim analysis from the INSPIRING trial which was presented at the CROI 2018 conference. INSPIRING (NCT ) is a Phase III, open-label randomized controlled trial enrolling HIV-TB coinfected adult patients for treatment with twice-daily DTG 50 mg or once-daily EFV 600 mg. Patients were receiving rifampin-based TB therapy. The trial is anticipated to be completed by the end of December A summary of the evidence on the DTG vs EFV comparison is presented in Treatment with DTG led to relative increases in CD4 cell counts but was not distinguishable from EFV with respect to viral suppression (Figure 12). By 24 weeks, DTG was well tolerated though the evidence on safety and tolerability was based on very few events. As discussed in the previous section, viral suppression was defined using the FDA Snapshot algorithm and differences here were driven by differences in discontinuations. The difference in discontinuations were not due to AEs

60 Figure 12: Modified FDA snapshot analysis of the percentage of participants (95% CI) with HIV-1 RNA <50 copies/ml Adapted from Dooley et al 2018 (CROI 2018) 60

61 Systematic literature review study selection A subgroup systematic literature review was conducted to describe and synthesize the evidence for the first-line treatment of HIV-TB co-infected patients. While the primary search strategy did not exclude patients with TB co-infection, we supplemented this search with a more sensitive, targeted strategy (Appendix B). The flow of information diagram is presented in Error! Reference source not found.. Overall, 10 studies were identified.

62 Figure 13: Flow diagram for principal literature review on TB co-infected individuals and first-line ART regimens 62

63 As with the principal analysis, some studies were excluded from the analysis set. In this case, three trials were non-comparative and could not be used in the analysis. These are ANRS 129 BKVIR, HIV-TB Pharmagene and TB-HAART. This was determined at the feasibility stage and could not be determined at the SLR stage. Also, the Sinha et al, 2013 was a prelude to Sinha et al, 2017 and was therefore not used in the analyses. The complete network of evidence for the analysis set of the HIV-TB co-infected sub-population is presented in Error! Reference source not found.. The evidence base consisted of 1378 patients enrolled in 13 treatment arms across 6 RCTs. The evidence was limited to 5 treatments: NVP, DTG, EFV, and RAL (400 mg; 800 mg). No evidence was identified for patients treated with EFV400. Figure 14: Complete network of evidence for patients with HIV-TB co-infection Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors.

64 5.2. Results As compared with the principal analysis, there were less analytical adjustments used here. There were no adjustments for differences in backbones and the network was too sparse to allow for metaregression adjustments for baseline characteristics. All network diagrams for specific analyses are provided in Appendix H. In the following subsections, we present results for analyses that involved DTG. For all other outcomes, cross-tables are provided in Appendix J as a reference Efficacy For the analysis of viral suppression in HIV-TB co-infected patients at 24 weeks, 3 trials including 382 patients across 7 trial arms informed the network of evidence. Results of the fixed-effects NMA are presented in Table 5. There was no statistically significant difference between DTG and EFV or between RAL400 and RAL800; however, the estimate suggests lower odds of suppression (in accordance with the FDA Snapshot algorithm). As previously mentioned, this difference appears to be driven by the larger number of discontinuations among the DTG arm of the INSPIRING trial. All treatments were associated with a higher proportion of patients achieving viral suppression compared to NVP, though this difference was not statistically significant compared to DTG (OR 1.64). Table 5: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for viral suppression at 24 weeks from the fixed-effects network meta-analyses in HIV-TB co-infected patients EFV 1.86 (0.64, 6.33) 0.51 (0.21, 1.19) 0.46 (0.19, 1.09) 3.09 (1.28, 7.98) 0.54 (0.16, 1.57) DTG 0.27 (0.06, 1.10) 0.24 (0.05, 0.98) 1.64 (0.36, 6.88) 1.95 (0.84, 4.74) 3.67 (0.91, 16.51) RAL (0.35, 2.27) 6.07 (1.77, 22.19) 2.19 (0.92, 5.39) 4.14 (1.02, 18.72) 1.12 (0.44, 2.89) RAL (2.00, 25.13) 0.32 (0.13, 0.78) 0.61 (0.15, 2.76) 0.16 (0.05, 0.57) 0.15 (0.04, 0.50) NVP 64

65 Values represent the effect of the treatment lower on the diagonal to the one higher on it. Bold values indicate comparisons that are statistically significant. Odds ratios above 1 indicate higher efficacy in viral suppression. DTG: dolutegravir; EFV: efavirenz; NVP: nevirapine; RAL400: raltegravir 400; RAL800: raltegravir 800 The network of evidence for change in CD4 count at 24 weeks in HIV-TB co-infected patients was based on 3 trials of 8 treatment arms consisting of 371 patients. The network of evidence at the 24-, 48-, and 96-week time points are presented in Error! Reference source not found. of Appendix H. Table 6: Cross table of odds ratios with 95% credible intervals comparing the relative efficacy of ARVs for mean change in CD4 cell counts at 24 weeks from the fixed-effects network meta-analyses in HIV- TB co-infected patients EFV (-89.61, ) (-23.84, 35.48) DTG (14.93, 89.61) (10.72, ) NVP (-35.48, 23.84) ( , ) Values represent the effect of the row treatment relative to the column treatment. Bold values indicate comparisons that are statistically differentiable. DTG: dolutegravir; EFV: efavirenz; NVP: nevirapine; Results of the fixed effects NMA are presented in Error! Reference source not found.. Treatment with DTG was associated with statistically significant increases in CD4 cell count compared to all other treatments in the network. No other comparisons were statistically significant Tolerability The evidence base for discontinuations due to adverse events was based on 4 trials of 524 patients enrolled in 9 treatment arms. The network of evidence is presented in Error! Reference source not found. of Appendix H. A summary of the evidence, arranged by treatment and trial, is presented in Error! Reference source not found.. The proportion of patients with a discontinuation due to adverse events varied across treatments. There are limitations to synthesizing evidence by NMA when there are a small number of events and analyses may generate non-meaningfully wide CrIs. With no such events observed in the DTG arm, an NMA was conducted for the GRADE table (below) but results should be disregarded. 65

66 Table 7: Data for treatment comparisons of interest for discontinuations due to adverse events outcome in HIV-TB co-infected patients Trials EFV DTG RAL400 RAL800 NVP INSPIRING 1/44 (2.3%) 0/69 (0%) Swaminathan et al, /59 (1.7%) 2/57 (3.5%) ANRS Reflate TB trial 3/51 (5.9%) 0/51 (0%) 3/51 (5.9%) N2R 3/71 (4.2%) 4/71 (5.6%) EFV: efavirenz; NVP: nevirapine; RAL400: raltegravir 400; RAL800: raltegravir 800; DTG: dolutegravir The evidence base for all-cause discontinuations (retention) was based on 2,839 patients enrolled across 13 treatment arms in 6 trials. The network of evidence is presented in Error! Reference source not found. of Appendix H. A summary of the evidence, arranged by treatment and trial, is presented in Error! Reference source not found.. The number of patients who discontinued treatment varied across treatments. There are limitations to synthesizing evidence by NMA when there are a small number of events and analyses may generate non-meaningfully wide CrIs. Table 8: Data for treatment comparisons of interest for discontinuations due to adverse events outcome in HIV-TB co-infected patients Trials EFV DTG RAL400 RAL800 NVP LPV/r INSPIRING 2/44 (4.5%) 5/69 (7.2%) Swaminathan et al, /59 (6.8%) 10/57 (17.5%) ANRS Reflate TB trial 6/51 (11.8%) 5/51 (9.8%) 9/51 (17.6%) CARINEMO 52/285 (18.2%) 43/285 (15.1%) N2R 9/71 (12.7%) 16/71 (22.5%) 66

67 EFV: efavirenz; NVP: nevirapine; RAL400: raltegravir 400; RAL800: raltegravir 800; DTG: dolutegravir; LPV/r: ritonavir-boosted lopinavir Safety The network of evidence for all treatment emergent SAEs (i.e. all SAEs) is presented in Error! Reference source not found. of Appendix H. This analysis was based on 2,726 patients enrolled in 5 trials consisting of 11 treatment arms. Data collected from trials is presented in Error! Reference source not found.. Again, small sample sizes were a major limiting factor. The odds of experiencing an SAE while on DTG were 0.48 times those of experiencing an SAE while on EFV (95% CrI: 0.12, 1.90). There were more events when it came to overall AEs and here, DTG was found to be safer than EFV (OR: 0.26; 95% CrI: 0.08, 0.84). Table 9: Data for treatment comparisons of interest for the treatment-emergent serious adverse events Trials EFV DTG RAL400 RAL800 NVP LPV/r INSPIRING 5/44 (11.4%) 4/69 (5.8%) Swaminathan et al, /59 (6.8%) 5/57 (8.8%) ANRS Reflate TB trial 19/51 (37.3%) 17/51 (33.3%) 17/51 (33.3%) CARINEMO 70/288 74/285 (24.3%) (26%) EFV: efavirenz; NVP: nevirapine; RAL400: raltegravir 400; RAL800: raltegravir 800; DTG: dolutegravir; LPV/r: ritonavir-boosted lopinavir 67

68 5.3. GRADE tables The summary of findings (GRADE) table, comparing DTG to EFV, is presented in Error! Reference source not found.. Most comparisons were informed by low or very low-quality evidence, though the change in CD4 cell count at 24 weeks was based on moderate quality evidence. Risk of bias was due to single trial used for the comparison of interest being small and of short duration. Table 10: Summary of the GRADE quality of evidence assessments for outcomes in the first-line treatment of HIV-TB co-infected patients Outcome Direct Effect Uncombined Estimates Combined Estimates Risk of Bias Inconsisten Indirectnes Imprecisio Publication Quality of Odds Ratio Absolute effects Indirect Network Overall cy s n Bias direct (95% CrI) evidence Transitivity quality of evidence precision evidence Viral suppression per 1, at 24 weeks (0.18, 1.67) Low (0.16, 1.57) (-225 to 75) Low Change in CD4 at weeks (14.96, 91.04) Moderate (14.93, 89.61) Moderate Discontinuations per 1, (0.30, 8.85) Low (0.35, 15.02) (-78 to 132) Low Discontinuations per 1, due to AE (0.01, 5.24) Very Low (0.00, 2.17) (-53 to 5) Very Low Emergent SAEs per 1, (0.12, 1.90) Very Low (0.11, 1.90) (-183 to 72) Very Low Emergent AEs per 1, (0.08, 0.84) Low (0.06, 0.73) (-338 to 30) Low Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate.

69 The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 69

70 6. Pregnant and breastfeeding women 6.1. Systematic literature review study selection A subgroup systematic literature review was conducted to describe and synthesize the evidence for the first-line treatment of pregnant and breastfeeding women. While the primary search strategy did not exclude pregnant and breastfeeding women, we supplemented this search with a more sensitive, targeted strategy (Appendix B). The flow of information diagram is presented in Figure 15. Overall, 15 studies were identified Summary of the evidence base In the previous review (2015), 9 studies in treatment-naïve pregnant and breastfeeding women were identified. However, no closed networks of evidence could be established (i.e. the evidence could not be synthesized through an NMA). Several studies were identified through the current update, but most were excluded from the analysis as they could only be synthesized through a descriptive summary. With respect to the research question, which is focused on the efficacy and safety of DTG relative to EFV, we identified two studies of relevance: the DolPHIN 1 trial and the Tsepamo study. The DolPHIN 1 trial is an open-label, phase II/III randomized controlled pilot study comparing DTG/TDF/XTC to EFV/TDF/XTC (standard of care). This is primarily a pharmacokinetics study with limited clinical outcomes and a very small sample size with 8 patients in each treatment arm in the current interim analysis. A larger, phase III trial is underway (DolPHIN-2, NCT ). This open-label randomized controlled trial was planned to begin in January 2018 and is anticipated to be completed in March 2021 with an estimated enrolment of 250 adult patients in South Africa and Uganda. The Tsepamo study was a large cohort study of pregnant women initiating DTG/TDF/XTC or EFV/TDF/XTC across 8 government hospitals in Botswana. A large sample of patients was enrolled, with 1,729 patients treated with DTG and 4,593 treated with EFV. It is noteworthy that Botswana was the first country to recommend DTG/TDF/XTC for initiation in pregnancy. The proportion of pregnancies with any adverse birth outcome was similar across treatment arms with 33.2% of DTG-managed pregnancies and 35.0% of EFV-managed pregnancies resulting in an adverse outcome. Similarly, severe birth outcomes were reported in 10.7% of DTG-managed and 11.3% of EFV-managed pregnancies. A summary of the outcomes from the Tsepamo study is presented in Table 12.

71 71

72 Figure 15: Flow diagram for principal systematic literature review on pregnant and breastfeeding women and first line ART regimens

73 Table 11: Summary of the Tsepamo study of DTG/TDF/FTC vs EFV/TDF/FTC in pregnant and breastfeeding women initiated on first-line ART Outcome Dolutegravir/ Efavirenz/ Unadjusted Adjusted RR TDF/FTC TDF/FTC RR (95% CI) (N=1729) (N=4593) (95% CI) Any Adverse Birth Outcome 576 (33 3%) 1611 (35 0%) Any Severe Adverse Birth Outcome 186 (10 8%) 520 (11 3%) Preterm birth (<37 weeks) 309 (18 0%) 844 (18 5%) Very preterm birth (<32 weeks) 66 (3 8%) 160 (3 5%) 0 95 (0 88,1 03) 0.95 (0 81,1 11) 0 97 (0 87,1 10) 1 10 (0 83,1 45) 0.94 (0.87, 1.02) 0.93 (0.79,1.11) 0.98 (0.87,1.11) 1.09 (0.82,1.45) Small for Gestational Age (<10th %tile weight-for-gestational age) 297 (17 4%) 838 (18 5%) 0 94 (0 83,1 06) 0.94 (0.83,1.06) Very small for Gestational Age (<3rd %tile weight-for-gestational age) 104 (6 1%) 302 (6 7%) 0 91 (0 74,1 13) 0.91 (0.74,1.13) Stillbirth 39 (2 3%) 105 (2 3%) Neonatal death (<28 days) 21 (1 2%) 60 (1 3%) RR: Relative risk; 95% CI: 95% confidence interval 0 99 (0 69,1 42) 0 93 (0 57,1 53) 0.99 (0.69,1.42) 0.96 (0.58,1.57) Two additional studies of relevance were identified (IMPAACT 1026s and EPPICC/PANNA) though they were not restricted to the treatment of first-line patients. In both studies, pregnant women were initiated on DTG-based regimens, with 29 pregnancies in IMPAACT 1026s and 84 pregnancies in EPPICC/PANNA. These studies were included in the evidence base to inform safety outcomes of interest. A summary of the evidence for select outcomes of interest in patients treated with DTG and EFV is presented in Error! Reference source not found.. It shows similar estimates to those seen in the Tsepamo study and suggests that the results of the Tsepamo study may be generalisable to high and middle-income settings.

74 Table 12: Summary of evidence among pregnant and breastfeeding women on first-line ART Outcome DTG/TDF/FTC EFV/TDF/FTC Unadjusted OR Study (95% CI) Viral suppression < 50 copies/ml 5 (62.5%) 4 (50%) 1.25 (0.52, 3.00) DolPHIN 1 (2 weeks post-partum) Still births 1 (12.5%) 0 (0%) -- DolPHIN 1 1 (1.2%) EPPICC/PANNA Congenital abnormalities 4 (4.9%) EPPICC/PANNA Pre-term birth 4 (13.8%) IMPAACT 1026s 11 (13.8%) EPPICC/PANNA Low birth weight (<2.5 kg) 4 (13.8%) IMPAACT 1026s 13 (16.9%) EPPICC/PANNA Very low birth weight (<1.5 kg) 1 (3.4%) IMPAACT 1026s 0 (0%) EPPICC/PANNA HIV transmissions 0 (0%) IMPAACT 1026s OR: Odds ratio; 95% CI: 95% confidence interval 74

75 6.3. GRADE tables The summary of findings (GRADE) table, comparing DTG to EFV, is presented in Error! Reference source not found.. Most outcomes were evaluated to be of moderate quality evidence due to the risk of bias associated with an observational study, though with respect to safety, a large cohort study such as this one is among the most desirable studies to have. The evidence for viral suppression 2-weeks post-partum was considered to be of very low quality largely due to the very small sample sizes. Table 13: Summary of the GRADE quality of evidence assessments for outcomes in the first-line treatment of pregnancy and breastfeeding women Outcome Risk of Inconsist Indirect- Imprecisi Publicati Relative Absolute DTG EFV Quality bias ency ness on on Bias risks effect sample sample of size size evidence Viral supp. ( per 5 (62.5%) 4 (50%) weeks post- (0.52, 3.00) 1,000 Very low partum) (260 to 1000) Any Adverse per Birth (0.87, 1.02) 1,000 (33 3%) (35 0%) Moderat Outcome (-50 to 5) e Any Severe per Adverse (0.79,1.11) 1,000 (10 8%) (11 3%) Moderat Birth Outcome (-26 to 9) e Preterm birth per

76 (0.87,1.11) 1,000 (-25 to 18) (18 0%) (18 5%) Moderat e Very preterm per 1, (3 8%) 160 birth (0.82,1.45) (-7 to 14) (3 5%) Moderat e Small for per Gestational (0.83,1.06) 1,000 (17 4%) (18 5%) Moderat Age (-31 to 12) e Very small for per 1, Gestational (0.74,1.13) (-16 to 11) (6 1%) (6 7%) Moderat Age e Stillbirth per 1, (2 3%) 105 (0.69,1.42) (-9 to 8) (2 3%) Moderat * e Neonatal death (0.58,1.57) -1 per 1,000 (-7 to 6) 21 (1 2%) 60 (1 3%) Moderat e Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. 76

77 The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 77

78 7. Children and adolescents 7.1. Summary of the evidence base A subgroup systematic literature review was conducted to describe and synthesize the evidence for the first-line treatment of children. While the primary search strategy did not exclude patients with TB coinfection, we supplemented this search with a more sensitive, targeted strategy (Appendix B). The flow of information diagram is presented in Error! Reference source not found.. Figure 16: Flow diagram for principal systematic literature review on adolescents and first line ART regimens

79 The evidence base for these patients was limited. Although 13 studies were identified, few of them provided relevant information. Children are a more complex population with multiple categories. These are defined as neonates, infants and children (age 0-3 years), children (age 3-10 years) and adolescents (age years). There were six comparisons of interest in this category: DTG vs EFV for adolescents and children DTG vs LPVr for infants and children DTG vs RAL for infants and children RAL vs EFV for children RAL vs LPV/r for infants and children RAL vs NVP for neonates There were no studies identified among first-line children, adolescents or infants that included DTG. Indeed, the only DTG-focused study among children and adolescents was IMPAACT P1093, which enrolled treatment-experienced patients. However, this study was cited by the US FDA as supporting evidence for approving DTG use among children. The results of this trial are presented in the accompanying, second-line report. No comparative studies included RAL. However, two RAL studies were identified in neonates and infants (Error! Reference source not found.). In the 24-week IMPAACT P1110 study, evidence from 42 patients suggested that daily RAL was safe and well-tolerated during the first 6 weeks of life. The multinational, non-randomized phase I/II IMPAACT P1066 study enrolled 5 cohorts of patients receiving various doses and delivery methods (i.e. coated tablet, chewable tablet, granules for suspension) and reported on safety outcomes and change in CD4 cell count. However, while the overall number of enrolled patients was high (N=153), most cohorts consisted of relatively few patients. Therefore, conclusions based on this study alone must be made with caution. Table 14: Studies of RAL-based regimens for neonates and infants Trial Sample Age Study Follow- Outcomes Conclusion size design up (weeks) IMPAACT 42 Eligible: Single-arm 24 -Safety and Daily RAL was P1110 Birth to 48 trial tolerability safe and well 79

80 NCT hours tolerated during the first 6 weeks of life. IMPAACT 153 Eligible: 4 Comparative 48 -Immunologic Only available P1066 weeks to cohort response from NCT years -Safety and ClinicalTrials.gov tolerability record -Retention 7.2. GRADE tables Given the lack of comparative evidence in the children and adolescent subgroup, we reference the results of the adult and adolescent review and downgrade the quality of evidence by one level for indirectness. Table 13 presents the summary of findings (GRADE) table for efficacy, safety, and tolerability outcomes for comparisons to DTG. Treatment with DTG was associated with a higher proportion of patients achieving viral suppression at both 48 and 96 weeks compared to either EFV, LPV/r, or RAL, though this was based on moderate (vs EFV) and low-quality evidence (LPV/r; RAL). The evidence on changes in CD4 cell count was mixed, however, with statistically significant increase at all time points compared to EFV (low quality evidence) and not statistically significant differences compared to either LPV/r or RAL based on very low-quality evidence. Given the lack of evidence identified, conclusions regarding comparisons on mortality and ADIs should be avoided. However, treatment was DTG was generally associated with fewer discontinuations, though only the comparison with EFV was considered to be moderate quality. Evidence on serious adverse events, both treatmentrelated and emergent, was based on low or very low-quality evidence. These findings, particularly the relatively few studies identified relevant to the research question in children and adolescents, highlights the need for ongoing research on the use of DTG in the first-line treatment of HIV-infected children. The GRADE table for the efficacy, safety, and tolerability outcomes for comparisons to RAL is presented in Error! Reference source not found.. Treatment with RAL was associated with a higher proportion of patients achieving viral suppression at both 48 and 96 weeks compared to EFV, LPV/r, and NVP, though 80

81 this was largely based on low quality evidence. No statistically significant differences were observed with respect to change in CD4 cell count based on low or very low-quality evidence. However, treatment with RAL was associated with fewer discontinuations, discontinuations due to adverse events, and fewer emergent serious adverse events compared to EFV (moderate quality evidence), LPV/r (low quality evidence), and NVP (low quality evidence). The evidence for emergent adverse events, treatmentrelated SAEs, and treatment-related adverse events was sparse and was largely considered to be of very low quality. As with the evidence for DTG, the evidence on the use of RAL in the first-line treatment of children is sparse and was generally considered to be low quality according to the GRADE assessments. Again, we highlight the need for ongoing research to further inform this evidence base. A detailed summary of the GRADE assessments for the evidence in children and adolescents is presented in Appendix K. 81

82 Table 15: Summary of the GRADE quality of evidence assessments for efficacy, safety, and tolerability outcomes for the use of DTG vs EFV, LPV/r, and RAL in the first-line treatment of children and adolescents DTG VS. EFV DTG VS. LPV/r DTG VS. RAL Odds Absolute Overall Odds ratio Absolute Overall Odds ratio Absolute Overall ratio effects quality of (95% CrI) effects quality (95% CrI) effects quality of (95% CrI) evidence of evidence evidence Viral suppression at 48 weeks 1.86 (1.44, 2.40) 74 per 1000 (47 to 98) Moderate 2.70 (1.92, 3.70) 136 per 1000 (92 to 184) Low 1.37 (1.03, 1.82) 33 per 1000 (3 to 64) Low Viral suppression at 96 weeks 1.93 (1.52, 2.47) 94 per 1000 (63 to 121) Moderate 2.70 (1.75, 4.00) 162 per 1000 (107 to 223) Low 1.32 (0.98, 1.79) 34 per 1000 (4 to 66) Low Viral suppression at 144 weeks 1.44 (1.08, 1.92) 39 per 1000 (-11 to 83) Low (0.01, 100) 31 per 1000 (-16 to 84) Very low Change in CD at 48 weeks - cells/ml (5.89, Low - cells/ml (-28.99, Very low - cells/ml (-15.04, Very low 43.18) 25.86) 23.85) Change in CD4 at 96 weeks cells/ml Low cells/ml Very low cells/ml Very low

83 DTG VS. EFV DTG VS. LPV/r DTG VS. RAL Odds Absolute Overall Odds ratio Absolute Overall Odds ratio Absolute Overall ratio effects quality of (95% CrI) effects quality (95% CrI) effects quality of (95% CrI) evidence of evidence evidence (2.95, (-54.43, (-27.41, 48.74) 18.83) 40.74) Change in CD at 144 weeks - cells/ml (20.12, Low - cells/ml (-89.80, Very low - cells/ml (-79.64, Very low 79.58) ) 95.93) Mortality 0.64 (0.09, 4.87) -4 per 1000 (--9 to 12) Very low 0.47 (0.05, 4.17) -4 per 1000 (-10 to 12) Very low 0.87 (0.13, 7.14) -1 per 1000 (-8 to 15) Low AIDS defining illnesses 8.08 (0.46, 66.45) 127 per 1000 (-13 to 550) Very low 7.14 (0.32, ) 127 per 1000 (-12 to 551) Very low 8.33 (0.30, ) 126 per 1000 (-16 to 552) Very low Discontinuations 0.49 (0.39, 0.62) -43 per 1000 (-54 to -30) Moderate 0.44 (0.32, 0.60) -65 per 1000 (-104 to -38) Low 0.76 (0.58, 0.99) -8 per 1000 (-25 to 4) Low Discontinuations due to AEs 0.30 (0.19, 0.47) -78 per 1000 (-101 to - Moderate 0.23 (0.11, 0.46) -107 per 1000 (-146 to -69) Low 0.72 (0.37, 1.41) -9 per 1000 (--57 to --3) Low 83

84 DTG VS. EFV DTG VS. LPV/r DTG VS. RAL Odds ratio (95% CrI) Absolute effects Overall quality of evidence Odds ratio (95% CrI) Absolute effects Overall quality of Odds ratio (95% CrI) Absolute effects Overall quality of evidence evidence 54) Emergent SAEs per -10 per per (0.38, Very low (0.37, 0.74) (-49 to 12) Low (0.94, 1.47) Low 1.11) (-45 to 6) (-36 to 17) Emergent AEs per -110 per 33 per (0.49, 1000 ( Low (0.56, 1.14) Very low (0.64, 1.22) Low 1.23) to -70) (-168 to -52) (-12 to 78) Treatmentrelated SAEs (0.02, 1000 (0.08, ( per per 126 per 12.5 Very low Very low (0.02, 1000) Very low 507.2) (-12 to 822) 1000) (-8 to 817) to 821) Treatmentrelated AEs per per -265 per (0.25, 1000 ( (-356 Very low (0.21, 0.52) (-295 to - Very low (0.16, 0.50) Very low 0.44) to -170) to -174) 151) Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. 84

85 The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 85

86 Table 16: Summary of the GRADE quality of evidence assessments for efficacy, safety, and tolerability outcomes for the use of RAL vs EFV, LPV/r, and NVP in the first-line treatment of children and adolescents RAL vs EFV RAL vs LPV/r RAL vs NVP Odds ratio Absolute Overall Odds ratio Absolute Overall Odds ratio Absolute Overall (95% CrI) effects quality of (95% CrI) effects quality (95% CrI) effects quality evidence of of evidence evidence Viral suppression at 48 weeks 1.36 (1.04, 1.78) 41 per 1000 (9 to 70) Low 1.96 (1.43, 2.78) 103 per 1000 (56 to 153) Low 1.56 (1.14, 2.17) 61 per 1000 (20 to 104) Low Viral suppression at 96 weeks 1.47 (1.11, 1.97) 59 per 1000 (23 to 92) Moderate 2.08 (1.35, 3.03) 128 per 1000 ( 71 to 190) Low 2.50 (1.47, 4.35) 166 per 1000 (75 to 266) Low Viral suppression at 144 weeks 1.50 (0.08, 29.57) 61 per 1000 (13 to 104) Low (0.03, ) 143 per 1000 (29 to 275) Very low Change in CD at 48 weeks (3.34, cells/ml Very low (-32.17, cells/ml Very (-6.87, cells/ml Very 37.54) (3.34 to 20.40) ( to low 37.89) (-6.87 to low 37.54) 20.40) 37.89) Change in CD at 96 weeks (-0.24, cells/ml Low (-55.27, cells/ml Very (-4.38, cells/ml Very 86

87 44.66) (-0.24 to 6.57) ( to low 74.13) (-4.38 to low 44.66) 6.57) 74.13) Change in CD at 144 weeks (-27.18, cells/ml Low (-88.18, cells/ml Very (-60.93, cells/ml Very 86.33) ( to 83.19) ( to low ) (-60.9 to low 86.33) 83.19) 113.5) Mortality 0.72 (0.17, 3.13) -3 per 1000 (-8 to 4) Low 0.53 (0.09, 2.63) -3 per 1000 (-9 to 4) Very low 0.51 (0.08, 3.45) -7 per 1000 (-23 to 3) Very low AIDS defining illnesses 0.98 (0.29, 4.09) -2 per 1000 (-15 to 26) Low 0.89 (0.19, 4.00) -1 per 1000 (-16 to 27) Very low 3.7 (0.22, ) 12 per 1000 (-32 less to 40) Very low Discontinuations per 0.50 (0.54, -49 per Moderate (0.43, 0.79) 1000 Low (0.35, 0.69) Low 0.89) 1000 (-119 to per 1000 (-75 to -20) 38) (-156 to- 56) Discontinuations per per 0.24 due to AEs (0.24, 1000 Moderate (0.16, 0.66) 1000 Low (0.10, 0.53) -85 per 1000 Low 0.78) (-49 to -17) (-95 to -27) (-153 to -40) Emergent SAEs per per 0.36 (0.35, 1000 Moderate (0.32, 0.62) 1000 Low (0.22, 0.58) -99 per 1000 Low 0.55) (-32 to -10) (-34 to -6) (-166 to -44) Emergent AEs per per per

88 (0.69, 1.11) 1000 (-193 to - 102) Treatmentrelated 0.77 SAEs (0.02, -1 per ) (-11 to 20) Treatmentrelated 1.23 AEs (0.81, 50 per ) (-29 to 133) Low (0.68, 1.18) 1000 (-199 to 87) Very low (0.29, 0.68) (-235 to -99) Low Very low (0.01, 6 per 1000 Very (0, ) 0 per 1000 Very ) (-89 to 28) low (-889 to 25) low Very low (0.69, 2.00) 41 per 1000 Very (0.85, 2.50) 78 per 1000 Very (-62 to 146) low (-26 to 183) low Legend: Uncombined estimates represent either direct estimates, if available, or indirect NMA estimates otherwise. Combined estimates are NMA estimates for comparisons where direct estimates were available. For uncombined estimates start with high quality evidence. -1 symbolizes a choice to rate down (e.g. high quality to moderate quality evidence); 0 symbolizes choice to not rate down; -- = not applicable because the NMA estimate is the only estimate. The final quality of evidence updates that of the uncombined evidence. The quality can be moved up if the uncombined score was penalized for precision, which was overcome in network estimates. It can be moved down if the estimates are no longer precise or if there is evidence of inconsistency in loops containing the comparison (i.e. violation of transitivity). Precision We rated down for precision if the confidence interval crossed 1.1 or 0.9. Consistency We assessed the consistency for direct treatment comparisons using I 2 estimates and visual inspection of point estimates. An I 2 of 75% or higher indicates considerable heterogeneity. This was conducted along the shortest indirect pathway with the largest number of trials for indirect estimates. Risk of Bias For direct estimates we rated down for risk of bias if the majority of studies within a comparison were considered to be at high risk of bias and similarly along the principal indirect pathway for indirect estimates. Indirectness Estimates obtained solely from indirect evidence were rated down for indirectness. 88

89 8. Discussion The purpose of this study was to support the 2018 update to the consolidated guidelines on the use of antiretrovirals for treating and preventing HIV with respect to the choice of first-line ART. The two questions of interest were whether DTG and/or EFV400, each with an XTC + TDF backbone, should be the preferred first-line ART regimen rather than their current designation of alternative first-line ART regimen. This extensive systematic literature review and network meta-analysis to evaluate the comparative efficacy and safety of these and other ART regimens drew strong conclusions about the improved efficacy and tolerability of DTG relative to EFV. Moreover, the evidence synthesis supports the use of DTG among sub-populations, which was not the case in Specifically, results of this study suggest comparable safety among pregnant women initiating treatment and results were not demonstrably worse among TB-HIV co-infected individuals. Unfortunately, evidence was lacking with respect to first-line children. Overall, the evidence more strongly supports the choice of DTG as the preferred first-line regimen. Despite strong evidence of improved efficacy and tolerability of DTG relative to EFV in the 2015, 4 other factors prevented its recommendation as the preferred first-line regimen. 5 These factors were: the unavailability of a fixed-dose combination with XTC and TDF, the high price, and the uncertainty around its use in sub-populations. All of these factors have now been overcome. A generic fixed-dose combination with TDF and lamivudine, referred to as TLD, 181 is now available and studies among subpopulations have begun to report results. It is in this context that this study was undertaken. For adults and adolescents, the principal population, there were only a few new trials with the treatments of interest and no new trials for the comparisons of interest. It is therefore not at all surprising that the conclusions were similar to the previous set of analyses. There continues to be high quality evidence of improved viral suppression, discontinuations and discontinuations due to AEs for DTG relative the EFV. Across all outcomes, the results of DTG were favourable, with the only exceptions being in outcomes that are plagued with very low number of events. While we demonstrated improved safety with DTG relative to EFV in terms of overall AEs, it is understood that the specific adverse events experienced with each differ, where DTG is more likely to lead to headaches and EFV has a higher propensity for neuropsychiatric side effects. 182 The evidence on EFV400 compared to standard dose EFV (600mg qd) comes entirely from the ENCORE1 trial. 21 Evidence from this trial suggests that low dose EFV is non-inferior to the standard dose, with

90 respect to efficacy and safety. It also suggests improved retention compared to standard dose EFV. Our study also displayed improved discontinuations due to adverse events among EFV400. Despite this difference, comparisons of low dose EFV to DTG using indirect evidence showed that DTG tended to have better retention and lower discontinuations due to adverse events. Moreover, these showed that DTG was more effective with respect to viral suppression. Critical to the public health approach that is favoured by the WHO, is the ability to prescribe treatment regardless of TB co-infection, pregnancy and ideally to children as well. By having the simplest treatment algorithm, we can ensure that task shifting and non-centralized care can continue in low-income settings, which has been so critical to the success of the fight against HIV/AIDS. No eligible studies were identified with respect to EFV400 among sub-populations and thus the second research question could not be tackled outside of the principal population. Only a handful of eligible studies providing insights on the first research question (DTG vs EFV) were identified. None were identified for children, so we relied on information on adults and adolescents to inform children and adolescents. Although we downgraded the strength of evidence for indirectness, it may not be too much of a stretch to believe the efficacy observed in adults would translate into children aged 3 years or more. For infants, the comparisons of interest were with RAL and a similar was used. Although a trial was available comparing DTG to EFV among TB-HIV co-infected patients, it must be recognized that the trial was both small and that only 24 week results were available. As such the evidence was of low quality. Evidence appears to suggest negligible difference between treatments. Pregnancy was the sub-population with the richest evidence base. Pregnancy was also a focal point of discussion during the Guideline Development Group meetings, but not exactly in line with this research. While the evidence base was convincing with respect to the use of DTG as a first-line regimen among pregnant women, there is a signal that DTG in preconception may be problematic. 183 That issue falls outside the scope of the current study. Our study has numerous strengths and limitations. First, the use of NMA allowed for analytic adjustments to account for differences in backbones and provide an unbiased estimate of the comparison between DTG and EFV despite the critical trial having different backbones. Second, by combining direct and indirect evidence, some of findings can be seen as having stronger evidence than previously perceived when strong findings are supported by both sources of evidence. With respect to limitations, first, the evidence for the comparisons of interest continued to be somewhat limited in subpopulations. For EFV400, it was completely missing. Most notably for DTG, there was an absence of 90

91 evidence within children. This was also the case in people pre-exposed to ARVs, though that was somewhat expected. Even in pregnancy and TB, much of the evidence is still to come. Second, some significant outcomes were limited by a very low number of events, including mortality, regimen substitutions, serious adverse events, and ADIs. This influenced the precision of our estimates with respect to these outcomes and, in some cases, precluded the conduct of evidence synthesis through NMA. Third, treatment-related adverse events were both inconsistently defined and inconsistently reported. This limitation was mitigated by considering both treatment-related and treatment-emergent adverse events. Additionally, studies of shorter duration are, by their nature, less likely to identify adverse events than longer-term trials. Despite this, the evidence was collected through a rigorous systematic review process in accordance with the practices and recommendations set forth by the Cochrane Collaboration, including both broad and targeted searches of the literature, critical appraisal of the identified studies, and consultation with subject matter experts. 9. Conclusions Dolutegravir in combination with lamivudine/emtricitabine and tenofovir disoproxil fumarate is an effective, safe and tolerable ART regimen. Across a variety of outcomes, evidence strongly suggests that it is superior to the current efavirenz-based preferred first-line ART regimen. With a new affordable generic fixed dose combination and comparable outcomes among sub-populations, the evidence supports the choice of a dolutegravir based preferred first-line regimen. Conclusions regarding low-dose efavirenz are unchanged since Low-dose efavirenz appears to be more tolerable, but with lack of evidence in sub-populations it is likely best to be considered an alternative first-line regimen. 91

92 Appendix A: Search strategy The search strategy presented in Table A1 was used for MEDLINE and EMBASE via OVID. The same search strategy was adapted to the other search engines. Searches were restricted from 01 January 2015 to the search date, 12 February Table 17: Systematic literature search strategy No. Term Comments 1. exp HIV/ or exp HIV Infection/ HIV/AIDS terms 2. (HIV Infections OR HIV?1* OR HIV?2* OR HIV infect* OR human immuno?deficiency virus OR human immune?deficiency virus).ti,ab. 3. ((human immun*) AND (deficiency virus)).ti,ab. 4. (acquired immuno?deficiency syndrome OR AIDS OR acquired immunedeficiency syndrome OR acquired immune deficiency).ti,ab. 5. ((acquired immun*) AND (deficiency syndrome)).ti,ab. 6. (Salvage therapy).ti,ab. Treatment failure 7. Exp Treatment Failure/ and experienced 8. (Treatment-experienced OR Antiretroviral experienced OR ARTexperienced OR Experienced patients).ti,ab. 9. (treatment switch*).ti,ab. 10. (or/1-5) not (or/6-9) Population Final 11. exp Antiretroviral Therapy, Highly Active/ Intervention and 12. exp Integrase Inhibitors/ comparators 13. exp HIV Reverse Transcriptase/ 14. exp Reverse Transcriptase Inhibitors/ 15. Exp Anti-HIV Agents/ 16. exp HIV Protease Inhibitors/ 17. (atazanavir OR Reyataz OR a OR BMS or atv*).ti,ab. 18. (cobicistat OR GS-9350 OR Tybost).ti,ab. 19. (dolutegravir OR Tivicay OR a OR S?GSK OR GSK ).ti,ab. 20. (darunavir OR Prezista OR TMC114 OR a or drv*).ti,ab. 92

93 21. (Elvitegravir OR GS-9137 OR Vitekta).ti,ab. 22. (emtricitabine OR Emtriva OR Coviracil OR a604004).ti,ab. 23. (lopinavir OR ABT-378 OR a or lpv*).ti,ab. 24. (nevirapine OR Viramune OR a600035).ti,ab. 25. (ritonavir OR Norvir OR a696029).ti,ab. 26. (raltegravir OR Isentress OR MK-0518 OR a608004).ti,ab. 27. (efavirenz OR Efavir OR Sustiva OR Stocrin OR Efcure OR Efferven OR Estiva OR Evirenz OR Viranz OR a699004).ti,ab. 28. (Trizivir OR Aluvia OR Kaletra OR Stribild OR triumeq).ti,ab. 29. or/11-28 Intervention and comparators final 30. (Randomized Controlled Trial or Controlled Clinical Trial).pt. Randomized 31. (Clinical Trial or Clinical Trial, Phase II or Clinical Trial, Phase III or Clinical Trial, Phase IV).pt. controlled trial terms 32. Multicenter Study.pt. 33. Randomized Controlled Trial/ or Randomized Controlled Trials as Topic/ or "Randomized Controlled Trial (topic)"/ 34. Controlled Clinical Trial/ or Controlled Clinical Trials as Topic/ or "Controlled Clinical Trial (topic)"/ 35. Clinical Trial/ or Phase 2 Clinical Trial/ or Phase 3 Clinical Trial/ or Phase 4 Clinical Trial/ 36. Clinical Trials as Topic/ or Clinical Trials, Phase II as Topic/ or Clinical Trials, Phase III as Topic/ or Clinical Trials, Phase IV as Topic/ 37. "Clinical Trial (topic)"/ or "Phase 2 Clinical Trial (topic)"/ or "Phase 3 Clinical Trial (topic)"/ or "Phase 4 Clinical Trial (topic)"/ 38. or/30-37 Study design final and 29 and 38 Complete Search 40. (healthy adj3 volunteer*).ti,ab. Features of 41. (healthy adj3 subject*).ti,ab. undesired 42. (cohort or observational study or case-control*).ti,ab. publications not (40 or 41 or 42) 93

94 not (cost minimi* or cost-utilit* or health utility* or economic evaluation* or economic review* or cost outcome or cost analys?s or economic analys?s or budget* impact analys?s).ti,ab not (review or letter or meta-analysis or case report or case series or posters or News or Newspaper article or meeting abstracts or lectures or interview or historical article or handbooks or guidelines or guidebooks or essays or editorial or comment or clinical conference or catalogs or case reports).pt. mp denotes multi-purpose and implies a search through all fields;.sh. denotes a Medical Subject Heading (MeSH) term; ti,ab. denotes a search for terms in title and abstract; exp denotes explode and implies that a term and a collection of variations on that term are searched for; * is used for truncation;? is a single space wildcard term. 94

95 Appendix B: Search strategies for sub populations As discussed in Section 3, there were sub-populations for which we supplemented the principal search with observational studies (principally cohort studies). The sub-populations included: TB co-infected patients, HBV co-infected patients, and pregnant and breastfeeding women. These supplemental searches replaced the study design section of the principal search with the study design terms shown in Table A1. Table 18: Systematic literature search to identify non-rct study designs No. Term Comments 1. (Nonrandom* or non random* or non-random* or quasi-random* or quasirandom*).ti,ab,hw. 2. cohort studies/ or cohort analysis/ 3. longitudinal studies/ or longitudinal study/ 4. prospective studies/ or prospective study/ 5. follow-up studies/ or follow up/ or followup studies/ 6. retrospective studies/ or retrospective study/ 7. observational study/ 8. quasi experimental methods/ or quasi experimental study/ 9. (quasi adj (experiment or experiments or experimental)).ti,ab. 10. ((non experiment or nonexperiment or non experimental or nonexperimental) adj3 (study or studies or design or analysis or analyses)).ti,ab. Observational study design terms Additionally, the searches terms in Table A2 were added to the study population terms. Each section in the table was conducted independently, not all at once. Table 19: Systematic literature review terms to identify the specific sub-populations of interest No. Term Comments 1. exp Tuberculosis/ TB terms 2. Mycobacterium tuberculosis/ 3. (tuberculosis or tb).ti,ab. 95

96 4. exp Hepatitis B/ HBV terms 5. exp Hepatitis B virus/ 6. (HBaSq).ti,ab,kw. 7. exp Pregnant Women/ Pregnant and 8. exp Mothers/ breastfeeding 9. exp Breast Feeding/ women terms 10. exp Infectious Disease Transmission, Vertical/ 11. Adolescent/ Adolescents 12. (adolescent* or teen*).tw. 96

97 Appendix C: Trends in observed treatment effects across follow-up times In this section we present exploratory plots investigating the effect of follow-up time on relative treatment effects. Figure 17: Proportion of patients experiencing a treatment emergent serious adverse event or adverse event

98 Legend: The left panels present proportions reported at each time point. The right panels present the odds ratios for each outcome over time. Connected points indicate a pair of odds ratios from a single RCT. Figure 18: Proportion of patients experiencing a treatment related serious adverse event or adverse event Legend: The left panels present proportions reported at each time point. The right panels present the odds ratios for each outcome over time. Connected points indicate a pair of odds ratios from a single RCT. 98

99 Figure 19: Proportion of patients discontinuing treatment and discontinuing treatment due to an adverse event Legend: The left panels present proportions reported at each time point. The right panels present the odds ratios for each outcome over time. Connected points indicate a pair of odds ratios from a single RCT. 99

100 Figure 20: Proportion of patients that died and developed an AIDS defining illness Legend: The left panels present proportions reported at each time point. The right panels present the odds ratios for each outcome over time. Connected points indicate a pair of odds ratios from a single RCT. 100

101 Figure 21: Proportion of patients switching treatments during study Legend: The left panels present proportions reported at each time point. The right panels present the odds ratios for each outcome over time. Connected points indicate a pair of odds ratios from a single RCT. 101

102 Appendix D: List of included studies In this appendix, we present the complete list of studies and corresponding publications, arranged by review. Table 20: List of included studies with corresponding publications Trial ID Author Title Year 089 Study 32,85,86 Bertz, R.J. et al 32 Malan, D.R. et al 86 Pharmacokinetics and pharmacodynamics of atazanavircontaining antiretroviral regimens, with or without ritonavir, in patients who are HIV-positive and treatment-naive Efficacy and safety of atazanavir, with or without ritonavir, as part of once-daily highly active antiretroviral therapy regimens in antiretroviral-naive patients Malan, D.R. et al week efficacy and safety of atazanavir, with and without ritonavir, in a HAART regimen in treatment-naive patients Study 58,59 Efficacy and safety of doravirine 100 mg qd vs. Efavirenz 600 Gatell et al 58 mg qd with TDF/FTC in art-naive HIV-infected patients: Gatell et al 59 Week 24 results Doravirine 100mg qd vs efavirenz +TDF/FTC in art-naive HIV+ patients: Week 48 results Kappelhoff, B.S. et al 74 Are adverse events of nevirapine and efavirenz related to plasma concentrations? NN 74,163,164 van Leth, F. et al 164 Nevirapine and efavirenz elicit different changes in lipid profiles in antiretroviral-therapy-naive patients infected with HIV Comparison of first-line antiretroviral therapy with regimens van Leth, F. et al 163 including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: A randomised open-label trial, the 2nn 2004 study ACTG A ,66,136 Gulick, R.M. et Three- vs four-drug antiretroviral regimens for the initial al 65 treatment of HIV-1 infection: A randomized controlled trial 2006 Gulick, R.M. et Triple-nucleoside regimens versus efavirenz-containing

103 al 66 regimens for the initial treatment of HIV-1 infection Ribaudo, H.J. et al 136 Efavirenz-based regimens in treatment-naive patients with a range of pretreatment HIV-1 RNA levels and cd4 cell counts 2008 ACTG A ,137 Haubrich, R.H. et al 69 Riddler, S.A. et al 137 Metabolic outcomes in a randomized trial of nucleoside, nonnucleoside and protease inhibitor-sparing regimens for initial HIV treatment Class-sparing regimens for initial treatment of HIV-1 infection Daar, E.S. et al 44 Atazanavir plus ritonavir or efavirenz as part of a 3-drug regimen for initial treatment of HIV type-1: A randomized trial 2011 ACTG A ,62,146 Gotti, D. et al 62 Increase in standard cholesterol and large HDL particle subclasses in antiretroviral-naive patients prescribed efavirenz compared to atazanavir/ritonavir 2012 Sax, P.E. et al 146 Abacavir/lamivudine versus tenofovir DF/emtricitabine as part of combination regimens for initial treatment of HIV: Final results 2011 Efficacy and tolerability of 3 nonnucleoside reverse ACTG A Lennox, J.L. et al 82 transcriptase inhibitor-sparing antiretroviral regimens for treatment-naive volunteers infected with HIV-1: A 2014 randomized, controlled equivalence trial A randomized, controlled trial of initial anti-retroviral ACTION 77 Kumar, P.N. et al 77 therapy with abacavir/lamivudine/zidovudine twice-daily compared to atazanavir once-daily with lamivudine/zidovudine twice-daily in HIV-infected patients 2009 over 48 weeks (ess100327, the action study) Immune reconstitution in severely immunosuppressed Advanz-3 98 Miro et al 98 antiretroviral-naive HIV-1-infected patients starting efavirenz, lopinavir-ritonavir, or atazanavir-ritonavir plus tenofovir/emtricitabine: Final 48-week results (the advanz-3 trial)

104 AI Murphy, R.L. et al 106 Dose-ranging, randomized, clinical trial of atazanavir with lamivudine and stavudine in antiretroviral-naive subjects: 48-week results 2003 Albini et al, Albini, L. et al 20 A randomized, pilot trial to evaluate glomerular filtration rate by creatinine or cystatin c in naive HIV-infected patients after tenofovir/emtricitabine in combination with atazanavir/ritonavir or efavirenz 2012 ALERT 152 Smith, K.Y. et al 152 Fosamprenavir or atazanavir once daily boosted with ritonavir 100 mg, plus tenofovir/emtricitabine, for the initial treatment of HIV infection: 48-week results of alert 2008 ALTAIR 128 Puls, R.L. et al 128 Efavirenz versus boosted atazanavir or zidovudine and abacavir in antiretroviral treatment-naive, HIV-infected subjects: Week 48 data from the Altair study 2010 Fixed-dose combination dolutegravir, abacavir, and lamivudine versus ritonavir-boosted atazanavir plus ARIA 124 Porteiro et al 124 tenofovir disoproxil fumarate and emtricitabine in previously untreated women with HIV-1 infection (aria): Week results from a randomised, open-label, non-inferiority, phase 3b study Arathoon, E. et al 23 Effects of once-daily darunavir/ritonavir versus lopinavir/ritonavir on metabolic parameters in treatmentnaive HIV-1-infected patients at week 96: Artemis 2013 ARTEMIS 23,52,79,96,115,119,16 0 Fatkenheuer, G. et al 52 Lazzarin, A. et al 79 Artemis 96-week comparison of liver tolerability of oncedaily darunavir/ritonavir (DRV/r) versus lopinavir/ ritonavir (LPV/r) in treatment-naive patients Artemis: 192-week efficacy and safety of once daily darunavir/ritonavir (DRV/r) versus lopinavir/r (LPV/r) in treatment-naive HIV-1-infected adults Mills, A.M. et al 96 Once-daily darunavir/ritonavir vs. Lopinavir/ritonavir in treatment-naive, HIV-1-infected patients: 96-week analysis 2009 Orkin, C. et Final 192-week efficacy and safety of once-daily

105 al 115 darunavir/ritonavir compared with lopinavir/ritonavir in HIV- 1-infected treatment-naive patients in the ARTEMIS trial Efficacy and safety of once-daily darunavir/ritonavir versus Ortiz, R. et al 119 lopinavir/ritonavir in treatment-naive HIV-1-infected 2008 patients at week 48 Termini, R. et al 160 Use of once-daily darunavir/r (800/100mg) in treatmentnaive patients co-infected with hepatitis b and/or c. Data from the ARTEMIS study 2009 Comparison of lipid profile with nevirapine versus Podzamczer, D. et al 123 atazanavir/ritonavir, both combined with tenofovir DF and emtricitabine (TDF/FTC), in treatment-naive HIV-1-infected 2009 patients: ARTEN study week-48 results Lipid profiles for nevirapine vs. Atazanavir/ritonavir, both ARTEN 121,123,153 Podzamczer, D. et al 121 combined with tenofovir disoproxil fumarate and emtricitabine over 48 weeks, in treatment-naive HIV infected patients (the ARTEN study) Soriano, V. et al 153 Nevirapine versus atazanavir/ritonavir, each combined with tenofovir disoproxil fumarate/emtricitabine, in antiretroviral-naive HIV-1 patients: The ARTEN trial 2011 Moyle, G.J. et al week results of abacavir/lamivudine versus tenofovir/emtricitabine, plus efavirenz, in antiretroviralnaive, HIV-1-infected adults: Assert study 2013 Randomized comparison of renal effects, efficacy, and safety ASSERT 104,125,158 Post, F.A. et al 125 with once-daily abacavir/lamivudine versus tenofovir/emtricitabine, administered with efavirenz, in antiretroviral-naive, HIV-1-infected adults: 48-week results 2010 from the assert study Stellbrink, H.J. et al 158 Comparison of changes in bone density and turnover with abacavir-lamivudine versus tenofovir-emtricitabine in HIVinfected adults: 48-week results from the assert study 2010 ATADAR 90,91 Martinez et Differential body composition effects of protease inhibitors

106 al 91 recommended for initial treatment of HIV infection: A randomized clinical trial Martinez, E. et al 90 Early lipid changes with atazanavir/ritonavir or darunavir/ritonavir 2014 ATLANTIC 162 van Leeuwen, R. et al 162 A randomized trial to study first-line combination therapy with or without a protease inhibitor in HIV-1-infected patients 2003 Avihingsanon et al, Avihingsanon, A. et al 27 Efficacy of tenofovir disoproxil fumarate/emtricitabine compared with emtricitabine alone in antiretroviral-naïve HIV-HBV coinfection in Thailand 2010 BASIC 168 Vrouenraets, S.M. et al 168 Randomized comparison of metabolic and renal effects of saquinavir/r or atazanavir/r plus tenofovir/emtricitabine in treatment-naive HIV-1-infected patients 2011 Johnson, M. and Moyle, G. 73 Castle study: 96-week efficacy & safety of ATV/r versus LPV/r in antiretroviral-naive HIV-1-infected patients 2009 Malan, N. et al 87 Gastrointestinal tolerability and quality of life in antiretroviral-naive HIV-1-infected patients: Data from the castle study 2010 Once-daily atazanavir/ritonavir compared with twice-daily CASTLE 73,87,101,1 02 Molina, J.M. et al 102 lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naive HIV-1- infected patients: 96-week efficacy and safety results of the 2010 castle study Once-daily atazanavir/ritonavir versus twice-daily Molina, J.M. et al 101 lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naive HIV-1- infected patients: 48 week efficacy and safety results of the 2008 castle study CCTG Bowman, V. et al week results of a pilot randomized study of an nucleoside reverse

107 transcriptase inhibitor (NRTI)-sparing regimen of raltegravir (RAL) + lopinavir/ritonavir (LPV/r) versus efavirenz/tenofovir disoproxil fumarate /emtricitabine (EFV/TDF/FTC) in antiretroviral CNA DeJesus, E. et al 46 Abacavir versus zidovudine combined with lamivudine and efavirenz, for the treatment of antiretroviral-naive HIVinfected adults 2004 CNA Vibhagool, A. et al 167 CNAAB Staszewski, S. et al 157 CNAF Matheron, S. et al 93 COMBINE 53 Fisac, C. et al 53 COMBINE 122 Podzamczer, D. et al 122 Triple nucleoside treatment with abacavir plus the lamivudine/zidovudine combination tablet (com) compared to indinavir/com in antiretroviral therapy-naive adults: Results of a 48-week open-label, equivalence trial (cna3014) Abacavir-lamivudine-zidovudine vs indinavir-lamivudinezidovudine in antiretroviral-naive HIV-infected adults: A randomized equivalence trial Triple nucleoside combination zidovudine/lamivudine/abacavir versus zidovudine/lamivudine/nelfinavir as first-line therapy in HIV- 1-infected adults: A randomized trial A comparison of the effects of nevirapine and nelfinavir on metabolism and body habitus in antiretroviral-naive human immunodeficiency virus-infected patients: A randomized controlled study A randomized clinical trial comparing nelfinavir or nevirapine associated to zidovudine/lamivudine in HIV-infected naive patients (the combine study) CTN177 67,68 Harris, M. et al 67 A randomized, open-label study of a nucleoside analogue reverse transcriptase inhibitor-sparing regimen in antiretroviral-naive HIV-infected patients 2009 Harris, M. et al 68 NRTI Sparing Trial (CTN 177): Antiviral and Metabolic Effects of

108 Nevirapine (NVP) + Lopinavir/ritonavir (LPV/r) vs. Zidovudine/lamivudine (AZT/3TC) + NVP vs. AZT/3TC + LPV/r DAUFIN 133 Rey, D. et al 133 tenofovir/lamivudine/nevirapine combination in naive HIV- High rate of early virological failure with the once-daily 1-infected patients 2009 DAYANA 78 Landman, R. et al 78 Evaluation of four tenofovir-containing regimens as first-line treatments in Cameroon and Senegal: The ANRS Dayana trial 2014 DRIVE Orkin et al 117 AHEAD 117 DRIVE Molina et al 99 FORWARD 99 Similar Efficacy and Safety By Subgroup in DRIVE-AHEAD: DOR/3TC/TDF vs EFV/FTC/TDF Doravirine is non-inferior to darunavir/r in phase 3 treatment naïve trial at week ECHO 103 Molina, J.M. et al 103 Rilpivirine versus efavirenz with tenofovir and emtricitabine in treatment-naive adults infected with HIV-1 (echo): A phase 3 randomised double-blind active-controlled trial 2011 Efficacy of 400 mg efavirenz versus standard 600 mg dose in Amin, J. et al 21 HIV-infected, antiretroviral-naive adults (encore1): A randomised, double-blind, placebo-controlled, non inferiority trial ENCORE1 21,22,64 Amin, J. et al 22 Efficacy and safety of efavirenz 400 mg daily versus 600 mg daily: 96-week data from the randomised, double-blind, 2015 placebo-controlled, non-inferiority ENCORE1 study Encore1 Study Group 64 Efficacy and safety of efavirenz 400 mg daily versus 600 mg daily: 96-week data from the randomised, double-blind, placebo-controlled, non-inferiority encore1 study 2015 Epzicom- Truvada 111,112 Nishijima, T. et al 112 Nishijima, T. et al 111 Abacavir/lamivudine versus tenofovir/emtricitabine with atazanavir/ritonavir for treatment-naive Japanese patients with HIV-1 infection: A randomized multicenter trial Once-daily darunavir/ritonavir and abacavir/lamivudine versus tenofovir/emtricitabine for treatment-naive patients with a baseline viral load of more than copies/ml

109 Once-daily dolutegravir versus darunavir plus ritonavir in FLAMINGO 37,10 0 Clotet, B. et al 37 Molina, J. et al 100 antiretroviral-naive adults with HIV-1 infection (FLAMINGO): 48-week results from the randomised open-label phase 3b study Once-daily dolutegravir versus darunavir plus ritonavir for treatment-naive adults with HIV-1 infection (FLAMINGO): 96 week results from a randomised, open-label, phase 3b study GEMINI 169 Walmsley, S. et al 169 Gemini: A noninferiority study of saquinavir/ritonavir versus lopinavir/ritonavir as initial HIV-1 therapy in adults 2009 GESIDA Berenguer, J. et al 31 Didanosine, lamivudine, and efavirenz versus zidovudine, lamivudine, and efavirenz for the initial treatment of HIV type 1 infection: Final analysis (48 weeks) of a prospective, randomized, noninferiority clinical trial, gesida Bictegravir versus dolutegravir, each with emtricitabine and GS-US-141- Sax et al 144 tenofovir alafenamide, for initial treatment of HIV infection: A randomised, double-blind, phase 2 trial 143,144 Sax et al 143 Randomized trial of bictegravir or dolutegravir with FTC/TAF for initial HIV therapy 2017 Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir versus co-formulated efavirenz, emtricitabine, and Sax, P.E. et al 145 tenofovir for initial treatment of HIV-1 infection: A randomised, double-blind, phase 3 trial, analysis of results 2012 after 48 weeks.[erratum appears in lancet aug GS-US ;380(9843):730] 0102 A randomized, double-blind comparison of single-tablet 145,178,180 Wohl, D.A. et al 178 regimen elvitegravir/cobicistat/emtricitabine/tenofovir df versus single-tablet regimen efavirenz/emtricitabine/tenofovir df for initial treatment of 2014 HIV-1 infection: Analysis of week 144 results Zolopa, A. et al 180 A randomized double-blind comparison of co-formulated elvitegravir/cobicistat/emtricitabine/tenofovir disoproxil

110 fumarate versus efavirenz/emtricitabine/tenofovir disoproxil fumarate for initial treatment of HIV-1 infection: Analysis of week 96 results Co-formulated elvitegravir, cobicistat, emtricitabine, and DeJesus, E. et al 48 tenofovir disoproxil fumarate versus ritonavir-boosted atazanavir plus co-formulated emtricitabine and tenofovir disoproxil fumarate for initial treatment of HIV-1 infection: A 2012 randomised, double-blind, phase 3, non-inferiority trial Nathan, C. et al 108 Long-term efficacy and safety of elvitegravir/cobicistat/emtricitabine/tenofovir df versus atazanavir plus ritonavir plus emtricitabine/tenofovir 2014 Orkin, C. et al 118 A randomized study comparing a three- and four-drug HAART regimen in first-line therapy (quad study) 2005 Week 144 efficacy and safety data: Elvitegravir/cobicistat/emtricitabine/tenofovir df (stribild) GS-US Orkin, C. et al 114 demonstrates durable efficacy and differentiated safety compared to atazanavir boosted by ritonavir plus ,108,114,116,118,13 emtricitabine/tenofovir df at week 144 in treatment-naive 8,140 HIV-1-infected patients Week 96 efficacy and safety data: Elvitegravir/cobicistat/ Orkin, C. et al 116 emtricitabine/tenofovir df (quad) compared to atazanavir boosted by ritonavir plus emtricitabine/tenofovir df in 2013 treatment-naive HIV-1-infected patients Elvitegravir/cobicistat/emtricitabine/tenofovir df (quad) has Rockstroh, J. et al 138 durable efficacy and differentiated safety compared to atazanavir boosted by ritonavir plus emtricitabine/tenofovir 2012 df at week 96 in treatment-naive HIV-1-infected patients A randomized, double-blind comparison of coformulated Rockstroh, J.K. et al 140 elvitegravir/ cobicistat/emtricitabine/tenofovir df vs ritonavir-boosted atazanavir plus coformulated 2013 emtricitabine and tenofovir df for initial treatment of HIV-1 110

111 infection: Analysis of week 96 results GS-US Cohen, C. et al 38 Randomized, phase 2 evaluation of two single-tablet regimens elvitegravir/cobicistat/emtricitabine/tenofovir disoproxil fumarate versus efavirenz/emtricitabine/tenofovir disoproxil fumarate for the initial treatment of HIV infection 2011 GS-US Gilead Science 61 Renal Effect of Stribild or Other Tenofovir DF-containing Regimens Compared to Ritonavir-boosted Atazanavir Plus Abacavir/Lamivudine in Antiretroviral Treatment-naive HIV-1 Infected Adults 2016 GS-US-292- Tenofovir alafenamide vs. Tenofovir disoproxil fumarate in 0102 Sax et al 147 single tablet regimens for initial HIV-1 therapy: A randomized phase 2 study Arribas et al 24 Significant efficacy and long-term safety difference with TAFbased STR in naive adults 2017 Brief report: A randomized, double-blind comparison of Wolh et al 176 tenofovir alafenamide versus tenofovir disoproxil fumarate, each coformulated with elvitegravir, cobicistat, and 2016 emtricitabine for initial HIV-1 treatment: Week 96 results GS-US-292- Tenofovir alafenamide versus tenofovir disoproxil fumarate, 0104; GS-US-292- Sax et al 149 coformulated with elvitegravir, cobicistat, and emtricitabine, for initial treatment of HIV-1 infection: Two randomised, ,26,149,176,1 double-blind, phase 3, non-inferiority trials 77 Brief report: Randomized, double-blind comparison of tenofovir alafenamide (taf) vs tenofovir disoproxil fumarate Wohl et al 26 (tdf), each coformulated with elvitegravir, cobicistat, and 2017 emtricitabine (e/c/f) for initial HIV-1 treatment: Week 144 results Wohl et al 177 Tenofovir alafenamide (TAF) in a single-tablet regimen in initial HIV-1 therapy 2015 GS-US Mills et al 95 Tenofovir alafenamide versus tenofovir disoproxil fumarate in the first protease inhibitor-based single-tablet regimen for

112 95 initial HIV-1 therapy: A randomized phase 2 study Bictegravir, emtricitabine, and tenofovir alafenamide versus GS-US-380- dolutegravir, abacavir, and lamivudine for initial treatment 1489 Gallant et al 55 of HIV-1 infection (gs-us ): A double-blind, multicentre, phase 3, randomised controlled non-inferiority trial Coformulated bictegravir, emtricitabine, and tenofovir GS-US-380- alafenamide versus dolutegravir with emtricitabine and 1490 Sax et al 148 tenofovir alafenamide, for initial treatment of HIV infection (gs-us ): A randomised, double-blind, multicentre, phase 3, non-inferiority trial HEAT 151 Smith, K.Y. et al 151 Randomized, double-blind, placebo-matched, multicenter trial of abacavir/lamivudine or tenofovir/emtricitabine with lopinavir/ritonavir for initial HIV treatment 2009 Initio Trial Virological and immunological outcomes at 3 years after INITIO 42 International Co-ordinating starting antiretroviral therapy with regimens containing nonnucleoside reverse transcriptase inhibitor, protease 2006 Committee 42 inhibitor, or both in INITIO: Open-label randomised trial Japanese Anti- HIV-1 QD Therapy 70 Honda, M. et al 70 Open-label randomized multicenter selection study of once daily antiretroviral treatment regimen comparing ritonavirboosted atazanavir to efavirenz with fixed-dose abacavir and lamivudine 2011 The klean study of fosamprenavir-ritonavir versus lopinavir- KLEAN 51 Eron Jr, J. et al 51 ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: A 2006 randomised non-inferiority trial LAKE Study 50 Echeverria, P. et al 50 Similar antiviral efficacy and tolerability between efavirenz and lopinavir/ritonavir, administered with abacavir/lamivudine (kivexa), in antiretroviral-naive patients: A 48-week, multi-centre, randomized study (LAKE study)

113 Li et al, Li, T. et al 83 Three generic nevirapine-based antiretroviral treatments in chinese HIV/aids patients: Multicentric observation cohort 2008 M Walmsley, S. et al 173 Maggiolo et al, Maggiolo, F. et al 84 Lopinavir-ritonavir versus nelfinavir for the initial treatment of HIV infection Once-a-day therapy for HIV infection: A controlled, randomized study in antiretroviral-naive HIV-1-infected patients MASTER 161 Torti, C. et al 161 efavirenz combination in HIV-positive patients upon starting Early virological failure after tenofovir + didanosine + antiretroviral therapy 2005 METABOLIK 19 Aberg, J.A. et al 19 Metabolic effects of darunavir/ritonavir versus atazanavir/ritonavir in treatment-naive, HIV type 1-infected subjects over 48 weeks 2012 NEAT001/ANR S Raffi, F. et al 129 Ritonavir-boosted darunavir combined with raltegravir or tenofovir-emtricitabine in antiretroviral-naive adults infected with HIV-1: 96-week results from the NEAT001/ANRS143 randomised non-inferiority trial 2014 NEWART 47 Dejesus, E. et al 47 A randomised comparison of safety and efficacy of nevirapine vs. Atazanavir/ritonavir combined with tenofovir/emtricitabine in treatment-naive patients 2011 Twenty-four-week safety and tolerability of nevirapine vs. Dart Trial Team 45 Abacavir in combination with zidovudine/lamivudine as firstline antiretroviral therapy: A randomized double-blind trial 2008 (NORA) Nevirapine/zidovudine/lamivudine has superior NORA 45,105,109 Munderi, P. et al 105 immunological and virological responses not reflected in clinical outcomes in a 48-week randomized comparison with abacavir/zidovudine/lamivudine in HIV-infected Ugandan 2010 adults with low cd4 cell counts Ndembi, N. et al 109 Viral rebound and emergence of drug resistance in the absence of viral load testing: A randomized comparison

114 OzCombo1 36 Carr, A. et al 36 between zidovudine-lamivudine plus nevirapine and zidovudine-lamivudine plus abacavir A randomised, open-label comparison of three highly active antiretroviral therapy regimens including two nucleoside analogues and indinavir for previously untreated HIV-1 infection: The OzCombo1 study 2000 Randomized, open-label, comparative trial to evaluate the OzCombo2 54 French, M. et al 54 efficacy and safety of three antiretroviral drug combinations including two nucleoside analogues and nevirapine for 2002 previously untreated HIV-1 infection: The ozcombo 2 study PEARLS 35 Campbell, T.B. et al 35 Efficacy and safety of three antiretroviral regimens for initial treatment of HIV-1: A randomized clinical trial in diverse multinational settings 2012 Impact of lamivudine on HIV and hepatitis b virus-related Matthews, G.V. et al 94 outcomes in HIV/hepatitis b virus individuals in a randomized clinical trial of antiretroviral therapy in southern 2011 PHIDISA II 94,132 Africa Ratsela, A. et al 132 A randomized factorial trial comparing 4 treatment regimens in treatment-naive HIV-infected persons with aids and/or a cd4 cell count <200 cells/mul in South Africa 2010 Examination of noninferiority, safety, and tolerability of PROGRESS 134,13 Reynes, J. et al 134 lopinavir/ritonavir and raltegravir compared with lopinavir/ritonavir and tenofovir/ emtricitabine in antiretroviral-naive subjects: The progress study, 48-week results Reynes, J. et al 135 Lopinavir/ritonavir combined with raltegravir or tenofovir/emtricitabine in antiretroviral-naive subjects: 96- week results of the progress study 2013 Protocol Gotuzzo, E. et ,88,89,107 al 63 Sustained efficacy and safety of raltegravir after 5 years of combination antiretroviral therapy as initial treatment of HIV-1 infection: Final results of a randomized, controlled,

115 phase II study (protocol 004) Rapid and durable antiretroviral effect of the HIV-1 integrase Markowitz, M. et al 89 inhibitor raltegravir as part of combination therapy in treatment-naive patients with HIV-1 infection: Results of a week controlled study Markowitz, M. et al 88 Sustained antiretroviral effect of raltegravir after 96 weeks of combination therapy in treatment-naive patients with HIV-1 infection 2009 Murray, J.M. et al 107 Antiretroviral therapy with the integrase inhibitor raltegravir alters decay kinetics of HIV, significantly reducing the second phase 2007 Radar study: Raltegravir combined with boosted darunavir Bedimo, R. 28 has similar safety and antiviral efficacy as tenofovir/emtricitabine combined with boosted darunavir in 2011 antiretroviral-naive patients RADAR Bedimo, R. et al 29 Standard Triple Therapy Controls HIV Better Than Raltegravir/Darunavir at 48 Weeks RADAR Study 2013 The RADAR study: Week 48 safety and efficacy of raltegravir Bedimo, R.J. et al 30 combined with boosted darunavir compared to tenofovir/emtricitabine combined with boosted darunavir in 2014 antiretroviral-naive patients. Impact on bone health SEARCH Phanuphak, N. et al 120 A 72-week randomized study of the safety and efficacy of a stavudine to zidovudine switch at 24 weeks compared to zidovudine or tenofovir disoproxil fumarate when given with lamivudine and nevirapine 2012 SENC 113 Nunez, M. et al 113 SENC (Spanish efavirenz vs. Nevirapine comparison) trial: A randomized, open-label study in HIV-infected naive individuals 2002 Sierra-Madero et al, Sierra-Madero, J. et al 150 Prospective, randomized, open label trial of efavirenz vs lopinavir/ritonavir in HIV+ treatment-naive subjects with cd4+<200 cell/mm 3 in Mexico

116 Brief report: Dolutegravir plus abacavir/lamivudine for the Walmsley et al 170 treatment of HIV-1 infection in antiretroviral therapy-naive patients: Week 96 and week 144 results from the single randomized clinical trial.[erratum appears in j acquir 2015 immune defic syndr jan 1;71(1):E33] SINGLE 18, Walmsley, S. et al 172 Dolutegravir regimen statistically superior to tenofovir/emtricitabine/efavirenz: 96-wk data 2014 Walmsley, S. et al 171 Dolutegravir Regimen Statistically Superior to Efavirenz/Tenofovir/Emtricitabine: 96-Week Results From the SINGLE Study (ING114467) 2014 Walmsley, S.L. et al 18 Dolutegravir plus abacavir-lamivudine for the treatment of HIV-1 infection 2013 SOLO 60 Gathe, J.C., Jr. et al 60 SOLO: 48-week efficacy and safety comparison of once-daily fosamprenavir /ritonavir versus twice-daily nelfinavir in naive HIV-1-infected patients 2004 SPARTAN 75 Kozal, M.J. et al 75 A nucleoside- and ritonavir-sparing regimen containing atazanavir plus raltegravir in antiretroviral treatment-naive HIV-infected patients: SPARTAN study results 2012 Stellbrink, H.J. Dolutegravir in antiretroviral-naive adults with HIV-1: 96- et al 159 week results from a randomized dose-ranging study 2013 SPRING-1 159,166 van Lunzen, J. et al 166 Once daily dolutegravir (S/GSK ) in combination therapy in antiretroviral-naive adults with HIV: Planned interim 48-week results from SPRING-1, a dose-ranging, 2012 randomised, phase 2b trial Once-daily dolutegravir versus raltegravir in antiretroviral- Raffi, F. et al 131 naive adults with HIV-1 infection: 48-week results from the 2013 randomised, double-blind, non-inferiority SPRING-2 study SPRING-2 130,131 Once-daily dolutegravir versus twice-daily raltegravir in Raffi, F. et al 130 antiretroviral-naive adults with HIV-1 infection (SPRING-2 study): 96-week results from a randomised, double-blind, 2013 non-inferiority trial 116

117 Once-Daily Dolutegravir (DTG; GSK ) Has a Renal SPRING-2/ SINGLE 43 Curtis, LD. et al 43 Safety Profile Comparable to Raltegravir (RAL) and Efavirenz in Antiretroviral (ART)-Naive Adults: 48 Week Results From 2013 SPRING-2 (ING113086) and SINGLE (ING114467) SPRING-2/ SINGLE/ FLAMINGO/ SAILING 97 Min, S. et al 97 Efficacy and Safety of Dolutegravir (DTG) in Hepatitis (HBV or HCV) Co-infected Patients: Results from the Phase 3 Program 2014 Body fat and other metabolic effects of atazanavir and Squires et al, ,155 Jemsek, J.G. et al 72 Squires, K. et al 155 efavirenz, each administered in combination with zidovudine plus lamivudine, in antiretroviral-naive HIVinfected patients Comparison of once-daily atazanavir with efavirenz, each in combination with fixed-dose zidovudine and lamivudine, as initial therapy for patients infected with HIV SSAT Bracchi et al 34 A randomized comparison of integrase inhibitors with TDF/FTC on renal markers 2017 Star study: Single-tablet regimen rilpivirine/ Cohen, C. et al 39 emtricitabine/tenofovir DF maintains noninferiority to efavirenz/emtricitabine/tenofovir DF and has minimal 2013 impact on fasting lipids in art-naive adults through week 96 Week 48 results from a randomized clinical trial of STaR 39,40,165 Cohen, C. et al 40 rilpivirine/ emtricitabine/tenofovir disoproxil fumarate vs. Efavirenz/emtricitabine/ tenofovir disoproxil fumarate in 2014 treatment-naive HIV-1-infected adults Van Lunzen et al 165 Rilpivirine vs. Efavirenz-based single-tablet regimens in treatment-naive adults: Week 96 efficacy and safety from a randomized phase 3b study 2016 START i 156 Squires, K.E. et al 156 A comparison of stavudine plus lamivudine versus zidovudine plus lamivudine in combination with indinavir in antiretroviral naive individuals with HIV infection: Selection

118 of thymidine analog regimen therapy (start i) Efficacy of raltegravir versus efavirenz when combined with DeJesus, E. et al 49 tenofovir/emtricitabine in treatment-naive HIV-1-infected patients: Week-192 overall and subgroup analyses from 2012 STARTMRK Lennox, J.L. et al 80 Raltegravir versus efavirenz regimens in treatment-naive HIV-1-infected patients: 96-week efficacy, durability, subgroup, safety, and metabolic analyses 2010 Safety and efficacy of raltegravir-based versus efavirenz- STARTMRK 49,80, 81,141,142 Lennox, J.L. et al 81 based combination therapy in treatment-naive patients with HIV-1 infection: A multi-centre, double-blind randomised controlled trial 2009 Durable efficacy and safety of raltegravir versus efavirenz Rockstroh, J.K. et al 141 when combined with tenofovir/emtricitabine in treatmentnaive HIV-1-infected patients: Final 5-year results from 2013 STARTMRK Long-term treatment with raltegravir or efavirenz combined Rockstroh, J.K. et al 142 with tenofovir/emtricitabine for treatment-naive human immunodeficiency virus-1-infected patients: 156-week 2011 results from STARTMRK STARTMRK/ BENCHMRK 139 Rockstroh, J. et al 139 Safety and efficacy of raltegravir in patients co-infected with HIV and hepatitis B and/or C virus: Complete data from phase iii double-blind studies 2012 Gallant, J.E. et al 57 Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: A 3- year randomized trial 2004 Study ,71 Long-term renal safety of tenofovir disoproxil fumarate in Izzedine, H. et al 71 antiretroviral-naive HIV-1-infected patients. Data from a double-blind randomized active-controlled multi-centre 2005 study Study Arribas, J.R. et Tenofovir disoproxil fumarate, emtricitabine, and efavirenz

119 934 25,56,126 al 25 compared with zidovudine/lamivudine and efavirenz in treatment-naive patients: 144-week analysis Gallant, J.E. et al 56 Tenofovir DF, emtricitabine, and efavirenz vs. Zidovudine, lamivudine, and efavirenz for HIV 2006 Tenofovir disoproxil fumarate, emtricitabine, and efavirenz Pozniak, A.L. et al 126 versus fixed-dose zidovudine/lamivudine and efavirenz in antiretroviral-naive patients: Virologic, immunologic, and 2006 morphologic changes - a 96-week analysis Evaluation of cardiovascular biomarkers in a randomized SUPPORT 76 Kumar, P. et al 76 trial of fosamprenavir/ritonavir vs. Efavirenz with abacavir/lamivudine in underrepresented, antiretroviral naive, HIV-infected patients (support): 96-week results SWATCH 92,110 Martinez- Picado, J. et al 92 Negredo, E. et al 110 Alternation of antiretroviral drug regimens for HIV infection. A randomized, controlled trial Alternation of antiretroviral drug regimens for HIV infection. Efficacy, safety and tolerability at week 96 of the swatch study Rilpivirine versus efavirenz with two background nucleoside THRIVE 41 Cohen, C.J. et al 41 or nucleotide reverse transcriptase inhibitors in treatmentnaive adults infected with HIV-1 (thrive): A phase 3, 2011 randomised, non-inferiority trial TMC278- C ,175 Pozniak, A.L. et al 127 Wilkin, A. et al 175 Efficacy and safety of tmc278 in antiretroviral-naive HIV-1 patients: Week 96 results of a phase IIb randomized trial Long-term efficacy, safety, and tolerability of rilpivirine (RPV, tmc278) in HIV type 1-infected antiretroviral-naive patients: Week 192 results from a phase IIb randomized trial Tshepo 174 Wester, C.W. et al 174 Non-nucleoside reverse transcriptase inhibitor outcomes among combination antiretroviral therapy-treated adults in Botswana 2010 WAVES 154 Squires et al 154 Integrase inhibitor versus protease inhibitor based regimen

120 Zhang et al, Zhang et al 179 for HIV-1 infected women (WAVES): A randomised, controlled, double-blind, phase 3 study Randomized clinical trial of antiretroviral therapy for prevention of HAND 2017 Table 21: List of included studies with corresponding publications for the TB sub-population Trial ID Author Title Year ANRS Reflate TB trial 184 Grinsztejn, B. et al 184 Raltegravir for the treatment of patients co-infected with HIV and tuberculosis (ANRS Reflate TB): A multicentre, phase 2, non-comparative, open-label, randomised trial 2014 ANRS 129 Lortholary, O. BKVIR 185 et al Tenofovir DF/emtricitabine and efavirenz combination therapy for HIV infection in patients treated for tuberculosis: The ANRS 129 BKVIR trial 2016 CARINEMO 186 Bonnet, M. et al 186 Nevirapine versus efavirenz for patients co-infected with HIV and tuberculosis: A randomised non-inferiority trial 2013 HIV-TB Habtewold, A. Pharmagene 187 et al Is there a need to increase the dose of efavirenz during concomitant rifampicin-based antituberculosis therapy in sub-saharan Africa? The HIV-TB Pharmagene study 2015 INSPIRING 188 Dooley et al INSPIRING: Safety and efficacy of dolutegravir-based ART in TB/HIV coinfected adults at week A randomized trial comparing plasma drug concentrations Manosuthi, W. et al 189 and efficacies between 2 nonnucleoside reversetranscriptase inhibitor-based regimens in HIV-infected 2009 N2R 189,190 patients receiving rifampicin: The n2r study Hepatotoxicity in patients co-infected with tuberculosis Mankhatitham, W. et al 190 and HIV-1 while receiving non-nucleoside reverse transcriptase inhibitor-based antiretroviral therapy and 2011 rifampicin-containing anti-tuberculosis regimen Sinha et al, Sinha, S. et Nevirapine versus efavirenz-based antiretroviral therapy

121 al 191 regimens in antiretroviral-naive patients with HIV and tuberculosis infections in India: A pilot study Sinha et al, Sinha, S. et al 192 Nevirapine- versus efavirenz-based antiretroviral therapy regimens in antiretroviral-naive patients with HIV and tuberculosis infections in India: A multi-centre study 2017 Swaminathan et al, Swaminathan, S. et al 193 Efficacy and safety of once-daily nevirapine- or efavirenzbased antiretroviral therapy in HIV-associated tuberculosis: A randomized clinical trial 2011 Efficacy and safety of antiretroviral therapy initiated one TB-HAART 194 Amogne, W. et al week after tuberculosis therapy in patients with cd4 counts < 200 cells/mul: TB-HAART study, a randomized clinical 2015 trial 121

122 Table 22: List of included studies with corresponding publications for the children and adolescent subpopulation Trial ID Author Title Year ARROW Kekitiinwa, A. (NCT ) 195 et al 195 Virologic response to first-line efavirenz-or nevirapinebased antiretroviral therapy in HIV-infected African children 2017 ATN Rudy, B. J. et al 196 Immune reconstitution but persistent activation after 48 weeks of antiretroviral therapy in youth with pretherapy cd4 >350 in ATN CHAPAS Bienczak, A. et al 197 Plasma efavirenz exposure, sex, and age predict virological response in HIV-infected African children 2016 GS-US ; GS-US (NCT ; NCT ) 198,199 Porter, D. P. et al 198 Porter, D. P. et al 199 Lack of emergent resistance in HIV-1-infected adolescents on elvitegravir-based STRS Lack of emergent resistance in HIV-1-infected adolescents on elvitegravir-based single-tablet regimens Gaur, A. et al 200 Safety and efficacy of E/C/F/TAF in HIV-1 infected treatment-naive adolescents 2016 GS-US (NCT ) 200 Gaur, A. H. et al 200 Safety, efficacy, and pharmacokinetics of a single-tablet regimen containing elvitegravir, cobicistat, emtricitabine, and tenofovir alafenamide in treatmentnaive, HIV-infected adolescents: A single-arm, open label trial GS-US (NCT ) Batra, J. et al 201 Kizito, H. et al 202 Week 24 data from a phase 3 clinical trial of E/C/F/TAF in HIV-positive adolescents Week-24 data from a phase 3 clinical trial of E/C/F/TAF in HIV-infected adolescents Shao, Y. et al 203 Week 24 data from a phase 3 clinical trial of E/C/F/TAF in HIV-infected adolescents 2015 IMPAACT P1083 Pinto et al 204 A Phase II/III Trial of Lopinavir/Ritonavir Dosed

123 (NCT ) 204 According to the WHO Pediatric Weight Band Dosing Guidelines IMPAACT P1093. Phase I/II, Multi-Center, Open-Label IMPAACT P IMPAACT Network 205 Pharmacokinetic, Safety, Tolerability and Antiviral Activity of Dolutegravir (GSK ), a Novel Integrase Inhibitor, in Combination Regimens in HIV Infected Infants, Children and Adolescents Virological response and resistances over 12 months MONOD-ANRS Amani-Bosse C et al 206 among HIV-infected children less than two years receiving first-line lopinavir/ritonavir-based antiretroviral therapy in Cote d'ivoire and Burkina Faso: 2017 (NCT ) 206,20 The MONOD-ANRS cohort 7 Pressiat C, et al 207 Suboptimal co-trimoxazole prophylactic concentrations in HIV-infected children according to the WHO guidelines 2017 PAINT Lombaard, J. (NCT ) 208 et al 208 Gopalan, B. P. et Gopalan, B. P. al 209 et al 209 Week 48 safety and efficacy of a rilpivirine (tmc278)- based regimen in HIV-infected treatment-naive adolescents: Paint phase ii trial Sub-therapeutic nevirapine concentration during antiretroviral treatment initiation among children living with HIV: Implications for therapeutic drug monitoring Shiau, S. et al 210 Shiau, S. et al 210 Early age at start of antiretroviral therapy associated with better virologic control after initial suppression in HIV-infected infants

124 Table 23: List of included studies with corresponding publications for the pregnant and breastfeeding women Trial ID Author Title Year BAN 211 Jamieson, D.J. et al 211 Maternal and infant antiretroviral regimens to prevent postnatal HIV-1 transmission: 48-week follow-up of the ban randomised controlled trial 2012 DART 212 Gibb, D. et al 212 Pregnancy and infant outcomes among HIV-infected women taking long-term ART with and without tenofovir in the DART trial 2012 DolPHIN 1 (NCT ) 213 Waitt et al 213 DolPHIN-1: Dolutegravir vs Efavirenz when Initiating Treatment in Late Pregnancy An Interim Analysis 2018 IMPAACT 1026s (NCT ) 214,21 5 Stek, A. et al 214 Mulligan et al 215 Pharmacokinetics of increased dose darunavir during late pregnancy and postpartum Dolutegravir pharmacokinetics in pregnant and postpartum women living with HIV Lamorde et al 216 Lamorde et al 216 Pharmacokinetics, pharmacodynamics and pharmacogenomics of efavirenz 400mg once-daily during pregnancy and postpartum 2017 MMA BANA Study Shapiro R 217 Antiretroviral regimens in pregnancy and breast-feeding in Botswana 2010 (NCT ) 217,21 8 Shapiro, R.L. et al 218 HIV transmission and 24-month survival in a randomized trial of HAART to prevent MTCT during pregnancy and breastfeeding in Botswana 2013 Ngoma Ngoma, M. S. et al 219 Efficacy of who recommendation for continued breastfeeding and maternal cart for prevention of perinatal and postnatal HIV transmission in Zambia 2015 PROMISE 1077BF/1077FF Fowler, M. G. et al 220 Benefits and risks of antiretroviral therapy for perinatal HIV prevention 2016 (NCT ; NCT ) 220,221 Fowler, M.G. et al 221 PROMISE: Efficacy and safety of 2 strategies to prevent perinatal HIV transmission

125 Cohan, D. et al 222 Efficacy and safety of lopinavir/ritonavir versus efavirenz-based antiretroviral therapy in HIV-infected pregnant Ugandan women 2015 Koss, C. A. et al 223 Hair concentrations of antiretrovirals predict viral suppression in HIV-infected pregnant and breastfeeding Ugandan women 2015 PROMOTE (NCT ) Cohan, D. et al 224 Efficacy and safety of LPV/r versus EFV in HIV+ pregnant and breast-feeding Ugandan Women Risk Factors for Preterm Birth among HIV-Infected Koss 225 Pregnant Ugandan Women Randomized to Lopinavir/ritonavir- or Efavirenz-based Antiretroviral 2011 Therapy Lopinavir/Ritonavir-Based Antiretroviral Treatment Natureeba 226 (ART) Versus Efavirenz-Based ART for the Prevention of 2014 Malaria Among HIV-Infected Pregnant Women Samuel Samuel, M. et al 227 Antenatal atazanavir: A retrospective analysis of pregnancies exposed to atazanavir 2014 Giuliano, M. et al 228 High cmv igg antibody levels are associated to a lower cd4+ response to antiretroviral therapy in HIV-infected women 2017 Palombi, L. et al 229 Drug resistance mutations 18 months after discontinuation of nevirapine-based art for prevention of mother-to-child transmission of HIV in Malawi 2015 SMAC Andreotti, M. et al 230 The impact of HBV or HCV infection in a cohort of HIVinfected pregnant women receiving a nevirapine-based antiretroviral regimen in Malawi 2014 Palombi, L. et al 231 Antiretroviral prophylaxis for breastfeeding transmission in Malawi: Drug concentrations, virological efficacy and safety 2012 Giuliano, M. et al 232 Maternal Antiretroviral Therapy for the Prevention of Mother-To-Child Transmission of HIV in Malawi:

126 Maternal and Infant Outcomes Two Years after Delivery Thorne et al 233 Pregnancy and neonatal outcomes following prenatal exposure to dolutegravir [Slides] -- Thorne et al Thorne et al 234 Pregnancy and neonatal outcomes following prenatal exposure to dolutegravir [Poster] -- Bollen et al 235 A comparison of the pharmacokinetics of dolutegravir in pregnancy and postpartum 2014 TSHEPISO 236 Dooley, K. E. et al 236 Pharmacokinetics of efavirenz and treatment of HIV-1 among pregnant women with and without tuberculosis coinfection 2015 Tshepo (NCT ) 237 Bussmann, et al 237 Pregnancy rates and birth outcomes among women on efavirenz-containing highly active antiretroviral therapy in Botswana Zash et al 238 The comparative safety of dolutegravir or efavirenz initiated during pregnancy in Botswana -- Zash et al 238,239 DTG/TDF/FTC Started in Pregnancy is as Safe as Zash et al 239 EFV/TDF/FTC in Nationwide Birth Outcomes 2017 Surveillance in Botswana 126

127 Appendix E: Characteristics of included studies In this section we present the trial and patient characteristics of all RCTs included in the principal analyses. Table 24: Trial characteristics for principal systematic literature review Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Previous review (May 2015) 2NN NVP EFV NVP XTC + d4t XTC + d4t XTC + d4t North and South America, Australia, Europe, South Africa, and Thailand ACTG A5142 (NCT ) EFV LPV/r 2 NRTIs 2 NRTIs US EFV XTC + TDF ACTG A5202 (NCT ) ATV/r EFV XTC + TDF XTC + ABC B US and Puerto Rico ATV/r XTC + ABC ACTG A5257 (NCT ) ATV/r RAL XTC + TDF XTC + TDF US and Puerto Rico

128 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) DRV/r XTC + TDF Albini et al, 2012 EFV ATV/r XTC + TDF XTC + TDF Italy Altair (NCT ) EFV ATV/r XTC + TDF XTC + TDF Australia DAYANA Trial (NCT ) NVP EFV XTC + TDF XTC + TDF Senegal and Cameroon ARTEMIS (NCT ) DRV/r LPV/r XTC + TDF XTC + TDF North and South America, Europe, and Asia NVP XTC + TDF Argentina, Germany, Italy, Mexico, ARTEN NVP ATV/r XTC + TDF XTC + TDF Portugal, Romania, Spain, Switzerland, and the United Kingdom ASSERT (NCT ) EFV EFV XTC + ABC XTC + TDF Europe ATADAR (NCT ) DRV/r ATV/r XTC + TDF XTC + TDF Spain 128

129 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Avihingsanon et al, 2010 EFV EFV XTC + AZT XTC + TDF Thailand CASTLE (NCT ) ATV/r LPV/r XTC + TDF XTC + TDF Africa, Asia, Europe, North America, South America EFV XTC + ABC United States, Europe, South CNA30024 EFV XTC + AZT America, Central America, and Puerto Rico CTN177 (NCT ) LPV/r NVP XTC + AZT XTC + AZT Canada, France, Spain, Argentina DAUFIN (NCT ) NVP NVP XTC + TDF XTC + TDF France ECHO (NCT ) RPV EFV XTC + TDF XTC + TDF USA, Canada, Australia, South Africa, Europe, Asia, and Latin America EFV400 XTC + TDF Argentina, Australia, Chile, Germany, ENCORE1 (NCT ) EFV XTC + TDF China Hong Kong SAR, Israel, Malaysia, Mexico, Nigeria, Singapore, South Africa, Thailand, 129

130 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) and the United Kingdom Epzicom-Truvada (NCT ) ATV/r ATV/r XTC + TDF XTC + ABC Germany, China Hong Kong SAR, Israel, Malaysia, Mexico, Nigeria DTG XTC + ABC FLAMINGO (NCT ) DRV/r DTG XTC + ABC XTC + TDF B Singapore, South Africa, Thailand, and the United Kingdom DRV/r XTC + TDF GESIDA 3903 (NCT ) EFV EFV XTC + ddi XTC + AZT Spain GS-US (NCT ) EVG/c EFV XTC + TDF XTC + TDF North America GS-US (NCT ) EVG/c ATV/r XTC + TDF XTC + TDF Australia, Europe, North America, and Thailand GS-US (NCT ) EVG/c EFV XTC + TDF XTC + TDF the USA HEAT LPV/r XTC + ABC the USA and Puerto Rico 130

131 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) (NCT ) LPV/r XTC + TDF Lake Study (NCT ) EFV LPV/r XTC + ABC XTC + ABC Spain and Italy Li et al, 2008 (NCT ) NVP NVP NVP AZT + ddi XTC + d4t XTC + AZT China Maggiolo et al, 2003 EFV EFV XTC + AZT XTC + ddi Italy LPV/r XTC + AZT MASTER EFV XTC + TDF Italy EFV ddi + TDF METABOLIK DRV/r ATV/r XTC + TDF XTC + TDF US NEWART (NCT ) NVP ATV/r XTC + TDF XTC + TDF US OzCombo 2 NVP XTC + AZT Australia 131

132 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) NVP XTC + d4t NVP ddi + D4T PEARLS (NCT ) EFV EFV XTC + TDF XTC + AZT Brazil, Haiti, India, Malawi, Peru, South Africa, Thailand, the USA and Zimbabwe Protocol 004 (NCT ) RAL EFV XTC + TDF XTC + TDF United States, Canada, Latin America, Thailand, and Australia SEARCH 003 NVP XTC + d4t (NCT ) NVP XTC + AZT Thailand NVP XTC + TDF SENC NVP EFV ddi + d4t ddi + d4t Spain Sierra-Madero et al, EFV XTC + AZT 2010 (NCT ) LPV/r XTC + AZT Mexico SINGLE (NCT ) DTG EFV XTC + ABC XTC + TDF Australia, Belgium, Canada, Denmark, France, Germany, Italy, 132

133 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Netherlands, Romania, Spain, United Kingdom, and United States SPRING-1 (NCT ) DTG EFV 2 NRTIs 2 NRTIs France, Germany, Italy, Russia, Spain, and the USA DTG XTC + ABC SPRING-2 (NCT ) RAL DTG XTC + ABC XTC + TDF Canada, the USA, Australia, and Europe RAL XTC + TDF RPV XTC + TDF United States, Australia, Austria, STaR (NCT ) EFV XTC + TDF B Belgium, Canada, France, Germany, Italy, Netherlands, Portugal, Puerto Rico, Spain, Switzerland, and United Kingdom RAL XTC + TDF Australia, Brazil, Canada, Chile, STARTMRK (NCT ) EFV XTC + TDF Colombia, France, Germany, India, Italy, Mexico, Peru, Spain, Thailand, and the USA Study 903 EFV XTC + TDF South America, Europe, and the USA 133

134 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) (NCT ) EFV XTC + D4T Study 934 (NCT ) EFV EFV XTC + TDF XTC + AZT France, Germany, Italy, Spain, the United Kingdom, and the USA THRIVE (NCT ) RPV EFV 2 NRTIs 2 NRTIs the USA and Puerto Rico, Canada, Australia, Europe, South Africa, Asia, and Latin America TMC278-C204 (NCT ) RPV EFV 2 NRTIs 2 NRTIs B The USA Tshepo NVP EFV 2 NRTIs 2 NRTIs Botswana Japanese Anti-HIV-1 QD Therapy EFV ATV/r XTC + ABC XTC + ABC Japan Review update (2018) DOR XTC + TDF Australia, Belgium, Canada, France, Study (NCT ) EFV XTC + TDF Germany, Netherlands, Russian Federation, Romania, Spain, Puerto Rico, and United States 134

135 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Advanz-3 (NCT ) EFV ATV/r LPV/r XTC + TDF XTC + TDF XTC + TDF Spain DTG ABC + XTC United States, Argentina, Canada, ARIA (NCT ) ATV/r XTC + TDF b France, Italy, Mexico, Portugal, Puerto Ricco, Russian Federation, South Africa, Spain, Thailand, and United Kingdom DRIVE AHEAD (NCT ) DOR EFV XTC + TDF XTC + TDF Africa, Asia, Europe, Latin America, and North America DOR 2 NRTIs Argentina, Australia, Austria, DRIVE FORWARD (NCT ) DRV/r 2 NRTIs Canada, Chile, Denmark, France, Germany, Italy, Romania, Russia, South Africa, Spain, United Kingdom, and United States GS-US (NCT ) EVG/c ATV/r XTC + TDF XTC + TDF Belgium, France, Ireland, Spain, and United Kingdom 135

136 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) EFV XTC + TDF ATv/r XTC + ABC GS-US (NCT ) EVG/c EVG/c XTC + TAF XTC + TDF United States and Puerto Rico EVG/c XTC + TAF United States, Australia, Austria, GS-US (NCT ) EVG/c XTC + TDF Belgium, Canada, Italy, Japan, Puerto Rico, Spain, Switzerland, Thailand, and United Kingdom EVG/c XTC + TAF United States, Canada, Dominican GS-US (NCT ) EVG/c XTC + TDF Republic, France, Italy, Mexico, Netherlands, Portugal, Puerto Rico, Sweden, and United Kingdom BIC XTC + TAF United States, Belgium, Canada, GS-US (NCT ) DTG XTC + ABC Dominican Republic, France, Germany, Italy, Puerto Rico, Spain, and United Kingdom GS-US (NCT ) BIC DTG XTC + TAF XTC + TAF United States, Australia, Belgium, Canada, Dominican Republic, France, 136

137 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Germany, Italy, Puerto Rico, Spain, and United Kingdom SSAT066 (NCT ) RAL DTG EVG/c XTC + TDF XTC + TDF XTC + TDF United Kingdom EVG/c XTC + TDF United States, Belgium, Dominican WAVES (NCT ) ATV/r XTC + TDF Republic, France, Italy, Mexico, Portugal, Puerto Rico, Russian Federation, Thailand, Uganda, and United Kingdom Zhang et al, 2015 (NCT ) NVP EFV XTC + AZT XTC + TDF China ABC: abacavir; ATV/r: ritonavir-boosted atazanavir; AZT: zidovudine; DRV/r: ritonavir-boosted darunavir; DTG: dolutegravir; EFV: efavirenz; EVG/c: Elvitegravir/cobicistat; LPV/r: ritonavir-boosted lopinavir; NRTI: non-nucleoside reverse transcriptase inhibitor; NVP: nevirapine; RAL: raltegravir; RPV: rilpivirine; TDF: Tenofovir disoproxil fumerate; XTC: lamivudine or emtricitabine 137

138 Table 25: Trial characteristics for studies selected in the systematic literature review among TB co-infected patients Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) Previous review (May 2015) ANRS Reflate TB trial EFV RAL400 RAL800 XTC + TDF XTC + TDF XTC + TDF Brazil, France CARINEMO EFV RAL400 XTC + TDF XTC + TDF Mozambique N2R EFV NVP XTC + d4t XTC + d4t Thailand Phidisa II (NCT ) EFV LPV/r EFV ddi + AZT ddi + AZT XTC + d4t South Africa LPV/r XTC + d4t Sinha et al, 2013 EFV NVP 2NRTIs 2NRTIs India Swaminathan et al, EFV XTC + ddi India 138

139 Study ID Treatment Backbone Number Trial Year of Trial Setting Arm randomiz Duration Initiation Phase ed (weeks) 2011 NVP XTC + ddi Review update (2018) ANRS 129 BKVIR EFV XTC + TDF France XTC + HIV-TB Pharmagene EFV d4t/tdf/zd Ethiopia V DTG 2 NRTIs Argentina, Brazil, Mexico, Peru, INSPIRING EFV 2 NRTIs Russian Federation, South Africa, Thailand Sinha et al, 2017 EFV NVP XTC + AZT XTC + AZT India XTC + TB-HAART EFV AZT/d4T/Te Ethiopia nofovir ABC: abacavir; ATV/r: ritonavir-boosted atazanavir; AZT: zidovudine; DRV/r: ritonavir-boosted darunavir; DTG: dolutegravir; EFV: efavirenz; EVG/c: Elvitegravir/cobicistat; LPV/r: ritonavir-boosted lopinavir; NRTI: non-nucleoside reverse transcriptase inhibitor; NVP: nevirapine; RAL: raltegravir; RPV: rilpivirine; TDF: Tenofovir disoproxil fumerate; XTC: lamivudine or emtricitabine 139

140 Appendix F: Patient characteristics in included studies Table 26: Patient characteristics across the 76 randomized controlled trials included in the principal analysis Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) Previous review (May 2015) NVP 34 (0.6) 220 (63%) 44 (20%) 200 (12.24) 4.7 (0.05) 58 (26%) 11 (5%) 2NN EFV 35 (0.37) 400 (64%) 84 (21%) 190 (10) 4.7 (0.04) 117 (29%) 21 (5%) NVP 34 (0.41) 387 (61%) 86 (22%) 170 (9.41) 4.7 (0.04) 102 (26%) 13 (3%) ACTG A5142 EFV 39 (0.57) 250 (81%) (9.42) 4.8 (0.05) LPV/r 37 (0.57) 253 (77%) (9.37) 4.8 (0.05) EFV 39 (0.45) 464 (85%) 70 (15%) 234 (7.94) 4.7 (0.02) ACTG A5202 ATV/r 39 (0.52) 465 (83%) 70 (15%) 224 (8.24) 4.7 (0.03) EFV 37 (0.48) 465 (79%) 88 (19%) 225 (7.59) 4.7 (0.02) ATV/r 38 (0.52) 463 (84%) 83 (18%) 236 (9.43) 4.6 (0.03) ACTG A5257 ATV/r 37 (0.37) 605 (76%) (7.68) 4.6 (0.03) 330 (55%) 15 (3%) RAL 36 (0.37) 603 (75%) (8.1) 4.7 (0.03) 324 (54%) 15 (3%) 140

141 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) DRV/r 37 (0.37) 601 (76%) (7.71) 4.6 (0.03) 323 (54%) 7 (1%) Albini et al, 2012 EFV 41.7 (1.77) 43 (77%) 2 (5%) ATV/r 45.4 (1.63) 48 (81%) 2 (4%) (16.98) (18.21) 4.7 (0.1) -- 4 (9%) 4.6 (0.09) -- 4 (8%) Altair ANRS DAYANA Trial ARTEMIS EFV 37.3 (0.84) 114 (79%) 6 (5%) 227 (8.9) 4.7 (0.06) 60 (53%) -- ATV/r 36.7 (0.83) 105 (71%) 4 (4%) 235 (11.13) 4.8 (0.06) 53 (50%) -- NVP 37 (1.22) 31 (45%) 2 (6%) 191 (10.49) 5.4 (0.1) EFV 40 (1.1) 30 (27%) 6 (20%) 201 (9.24) 5.6 (0.11) DRV/r 36 (0.49) 343 (70%) 27 (8%) 228 (8.06) 4.9 (0.03) LPV/r 35 (0.48) 346 (70%) 35 (10%) 218 (7.66) 4.8 (0.03) NVP 38 (0.71) 188 (81%) 12 (6%) 177 (7.44) 5.1 (0.05) ARTEN NVP 40 (0.77) 188 (87%) 16 (9%) 187 (6.71) 5.1 (0.04) ATV/r 38 (0.68) 193 (84%) 17 (9%) 188 (6.77) 5.1 (0.05) ASSERT EFV 38 (0.74) 192 (83%) 10 (5%) 240 (8.66) 5.0 (0.06)

142 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) EFV 36 (0.69) 193 (80%) 18 (9%) 230 (8.49) 5.1 (0.05) ATADAR Avihingsanon et al, 2010 CASTLE CNA30024 CTN177 DAUFIN DRV/r 37 (0.95) 88 (89%) (18.13) 4.8 (0.08) ATV/r 35 (0.84) 90 (87%) (21.49) 4.8 (0.07) EFV 34 (2.34) 5 (83%) (18.3) 4.91 (0.13) 1 (17%) -- EFV 33 (1.87) 7 (70%) (27.6) 4.90 (0.12) 2 (20%) -- ATV/r 34 (0.51) 440 (69%) 18 (4%) 205 (7.55) 5.0 (0.03) LPV/r 36 (0.49) 443 (69%) 22 (5%) 204 (7.66) 5.0 (0.02) EFV 35 (0.63) 324 (80%) (20.42) 4.8 (0.04) EFV 35 (0.6) 325 (82%) (12.97) 4.8 (0.04) LPV/r 37 (1.81) 24 (88%) (29.8) 4.9 (0.15) NVP 38.1 (1.74) 27 (63%) (28.68) 4.9 (0.14) NVP 41.2 (1.08) 35 (71%) 3 (9%) 195 (15.25) 4.9 (0.09) 9 (26%) 1 (3%) NVP 41.6 (1.63) 36 (78%) 3 (8%) 191 (12.13) 5.0 (0.13) 13 (36%) 0 (0%) ECHO RPV 36 (0.65) 346 (77%) 14 (4%) 240 (9.54) 5.0 (0.05)

143 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) EFV 36 (0.52) 344 (80%) 24 (7%) 257 (8.15) 5.0 (0.04) ENCORE1 EFV (0.56) 321 (69%) 11 (3%) 273 (5.39) 4.8 (0.03) (8%) EFV 35.8 (0.57) 309 (67%) 11 (4%) 272 (5.72) 4.7 (0.04) (7%) ATV/r 35 (1.3) 55 (98%) (12.88) 4.3 (0.07) 49 (89%) -- Epzicom-Truvada ATV/r 39 (1.53) 54 (98%) (10.82) 4.3 (0.08) 47 (87%) -- FLAMINGO GESIDA 3903 GS-US GS-US DTG 35.7 (0.69) 242 (87%) (10) 4.5 (0.05) DRV/r 36.2 (0.68) 242 (83%) (10.95) 4.5 (0.05) EFV 38 (0.65) 186 (77%) 28 (15%) 205 (10.37) 5.0 (0.05) 73 (39%) 24 (13%) EFV 40 (0.77) 183 (76%) 44 (24%) 216 (8.76) 5.0 (0.05) 79 (43%) 31 (17%) EVG/c 38 (0.56) 348 (88%) 28 (8%) 391 (10.11) 4.7 (0.03) 278 (80%) 10 (3%) EFV 38 (0.56) 352 (90%) 24 (7%) 382 (9.07) 4.8 (0.03) 282 (80%) 11 (3%) EVG/c 38 (0.56) 353 (92%) 32 (9%) 351 (7.57) 4.8 (0.03) 275 (78%) 4 (1%) ATV/r 39 (0.52) 355 (89%) 24 (7%) 366 (7.55) 4.8 (0.03) 273 (77%) 7 (2%) GS-US EVG/c 36 (1.28) 48 (92%) 3 (6%) 354 (21.51) 4.6 (0.11)

144 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) EFV 35 (2) 23 (91%) 1 (4%) 436 (31.07) 4.6 (0.15) HEAT Lake Study LPV/r 38 (0.53) 343 (84%) 55 (16%) 214 (10.18) 4.9 (0.05) LPV/r 38 (0.51) 345 (80%) 57 (17%) 193 (10.06) 4.8 (0.05) EFV 39 (1.06) 63 (86%) 5 (7%) 193 (15.37) 5.4 (0.74) 30 (47%) -- LPV/r 37 (1.19) 63 (87%) 5 (8%) 191 (16) 5.3 (0.69) 25 (39%) -- NVP 35.8 (1.08) 65 (52%) (9.1) 4.4 (0.08) Li et al, 2008 NVP 40.1 (1.3) 69 (45%) (10.11) 4.5 (0.09) NVP 37.3 (1.19) 64 (48%) (9.19) 4.4 (0.08) Maggiolo et al, 2003 EFV 37 (--) 26 (76%) 21 (64%) 175 (22.91) 5.22 (0.09) 5 (15%) 6 (18%) EFV 40 (--) 28 (82%) 18 (55%) 184 (21.53) 5.21 (0.09) 3 (9%) 10 (28%) LPV/r 41 (2.96) 9 (78%) 2 (22%) 218 (47.41) 4.4 (0.26) 1 (11%) 3 (33%) MASTER EFV 42 (0.94) 10 (100%) 2 (20%) 116 (21.55) 4.6 (0.27) 2 (20%) 4 (40%) EFV 38 (2.68) 11 (73%) 1 (9%) 113 (42.21) 4.8 (0.2) 2 (18%) 2 (18%) METABOLIK DRV/r 36.5 (1.34) 34 (85%) 0 (0%) 267 (17.9) 5 (0.14)

145 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) ATV/r 35 (1.62) 31 (87%) 1 (3%) 316 (27.8) 4.6 (0.13) NEWART NVP 38 (1.11) 75 (87%) -- ATV/r 36 (0.89) 77 (92%) (12.16) (12.68) 4.9 (0.09) (0.08) NVP 40 (1.92) 34 (97%) (50.98) 4.5 (0.16) OzCombo 2 NVP 37 (1.55) 37 (89%) (50.95) 4.6 (0.15) NVP 39 (1.72) 20 (95%) (57.55) 4.7 (0.16) PEARLS EFV 38.3 (2.03) 22 (96%) 57 (11%) 169 (6.54) 5.0 (0.03) EFV 35 (0.4) 519 (54%) 53 (10%) 162 (6.5) 5.0 (0.03) EFV 35 (0.35) 526 (51%) (3.8) 5.1 (0.02) Phidisa II LPV/r 35 (0.39) 526 (54%) (3.81) 5.2 (0.03) EFV 35.3 (0.26) 444 (67%) (4.11) 5.1 (0.02) LPV/r 35.3 (0.26) 440 (68%) (3.62) 5.2 (0.02) Protocol 004 RAL 35.5 (0.26) 444 (69%) 12 (29%) 338 (29.83) 4.6 (0.09)

146 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) EFV 35.6 (0.26) 443 (68%) 14 (37%) 280 (24.98) 4.8 (0.08) NVP 34 (0.97) 41 (76%) 0 (0%) 157 (13.43) 4.9 (0.1) SEARCH 003 NVP 36 (1.04) 38 (68%) 2 (4%) 174 (13.72) 4.9 (0.09) NVP 35 (1.29) 48 (48%) 3 (6%) 154 (12.74) 4.8 (0.09) SENC Sierra-Madero et al, 2010 SINGLE SPRING-1 SPRING-2 NVP 34 (1.13) 49 (43%) 2 (6%) 353 (23.67) 4.4 (0.07) 14 (39%) 14 (39%) EFV 35 (0.98) 51 (45%) 5 (16%) 416 (31.93) 4.4 (0.08) 14 (45%) 10 (32%) EFV 35 (0.9) 36 (78%) 36 (38%) 79 (15.29) LPV/r 35 (1.04) 31 (77%) 43 (46%) (15.37) DTG 36.7 (0.29) 95 (83%) 17 (4%) (7.32) 4.7 (0.04) 277 (67%) 21 (5%) EFV 36 (0.34) 94 (87%) 17 (4%) 339 (7.28) 4.7 (0.04) 297 (71%) 8 (2%) DTG 36 (0.57) 322 (70%) 0 (0%) 327 (20.86) 4.4 (0.1) EFV 36 (0.55) 327 (76%) 1 (2%) 328 (21.07) 4.5 (0.1) DTG 37 (0.92) 51 (88%) 8 (2%) 359 (7.09) 4.5 (0.04) RAL 40 (1.67) 50 (88%) 8 (2%) 362 (7.38) 4.6 (0.03)

147 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) STaR STARTMRK Study 903 Study 934 THRIVE TMC278-C204 Tshepo RPV 33 (0.53) 404 (64%) (8.98) 4.8 (0) EFV 33 (0.55) 401 (66%) (9.35) 4.8 (0.03) RAL 0 (0) 103 (77%) 52 (19%) (7.41) 5 (0.04) EFV 0 (0) 101 (77%) 59 (21%) (7.96) 5 (0.04) EFV 37.6 (0.54) 281 (81%) (11.62) 4.9 (0.04) EFV 36.9 (0.6) 282 (82%) (11.53) 4.9 (0.04) EFV 36 (0.49) 299 (74%) (9.28) 5.0 (0.05) EFV 36 (0.53) 301 (75%) (9.26) 5.0 (0.05) RPV 36 (1.11) 52 (86%) 20 (6%) 263 (8.05) 5.0 (0.04) EFV 39 (0.82) 54 (70%) 17 (5%) 263 (12.36) 5.0 (0.04) RPV 36 (0.47) 340 (74%) 6 (7%) 176 (10.54) 4.8 (0.07) EFV 36 (0.54) 338 (72%) 8 (9%) 207 (20.5) 4.9 (0.06) NVP 36 (1) 93 (70%) (4.31) 5.3 (0.04) EFV 35 (0.89) 89 (67%) (5.3) 5.3 (0.03)

148 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) Japanese Anti-HIV-1 QD Therapy EFV 35 (1.51) 36 (100%) (6.73) 4.6 (0.09) ATV/r 36 (1.53) 35 (100%) (7.47) 4.4 (0.08) Review update (2018) Study (NCT ) Advanz-3 (NCT ) ARIA (NCT ) DRIVE AHEAD (NCT ) DRIVE FORWARD (NCT ) DOR 35 (0.93 a ) 99 (91.7%) (19.6 a ) 4.6 (0.07 a ) EFV 34 (0.71 a ) 101 (93.5%) (19.2 a ) 4.6 (0.07 a ) ATV/r 38.5 (1.2 a ) 27 (90%) 17 (57%) (0.11 c ) 14 (46.7%) 1 (3.3%) EFV 39 (1.6 a ) 21 (72.4%) 12 (41%) (0.11 c ) 12 (41.4%) 1 (3.5%) LPV/r 36.5 (1.57 a ) 25 (83.3%) 13 (44%) (0.11 c ) 14 (46.7%) 0 (0%) DTG 38.1 (0.71 b ) 0 (0%) 11 (4%) 340 (14.08 c ) 4.41 (0.06 c ) 1 (0.004%) 12 (5%) ATV/r 37.8 (0.64 b ) 0 (0%) 9 (4%) 350 (11.55 c ) 4.43 (0.05 c ) 2 (1%) 8 (3%) DOR 32 (0.54 a ) 305 (83.8%) 46 (12.6%) 414 (14.39 a ) 4.4 (0.04 a ) EFV 30 (0.53 a ) 311 (85.4%) 53 (14.6%) 388 (14.98 a ) 4.5 (0.04 a ) DOR 34.8 (0.54 b ) 319 (83%) 36 (9%) 433 (10.6 b ) 4.4 (0.04 b ) DRV/r 35.7 (0.55 b ) 326 (85%) 37 (10%) 412 (11.74 b ) 4.4 (0.04 b )

149 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) ATV/r 34 (1.98 b ) 17 (100%) (51.36 b ) GS-US (NCT ) ATV/r 34 (1.77 b ) 15 (93.8%) (44.78 b ) EFV 34 (2.26 b ) 15 (93.8%) (50.86 b ) EVG/c 36 (1.91 b ) 17 (100%) (41.91 b ) GS-US (NCT ) GS-US (NCT ) GS-US (NCT ) GS-US (NCT ) GS-US (NCT ) EVG/c 35 (1.06 b ) 108 (96.4%) (17.08 b ) 4.63 (0.05 b ) EVG/c 37 (1.39 b ) 57 (98.3%) (27.52 b ) 4.69 (0.08 b ) EVG/c 35 (0.48 b ) 364 (83.7%) (10.73 b ) 4.55 (0.03 b ) EVG/c 36 (0.51 b ) 376 (87%) (10.21 b ) 4.55 (0.03 b ) EVG/c 35 (0.52 b ) 369 (85.6%) (9.96 b ) 4.53 (0.03 b ) EVG/c 36 (0.52 b ) 364 (83.7%) (10.87 b ) 4.5 (0.03 b ) DTG 32 (2.09 c ) 282 (90%) 15 (5%) 443 (12.13 c ) 4.51 (0.03 c ) 250 (79%) 4 (1%) BIC 31 (2.21 c ) 285 (91%) 12 (4%) 450 (11.85 c ) 4.42 (0.04 c ) 251 (80%) 5 (2%) BIC 33 (0.78 c ) 280 (88%) 24 (8%) 440 (12.37 c ) 4.43 (0.04 c ) 237 (74%) 3 (1%) DTG 34 (0.77 c ) 288 (89%) 26 (8%) 441 (12.23 c ) 4.45 (0.03 c ) 250 (77%) 6 (2%) 149

150 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) Sinha et al, 2017 (NCT ) SSAT066 (NCT ) WAVES (NCT ) EFV 36.7 (0.62 b ) 111 (79.3%) 134 (97.8%) 133 (9.46 a ) 5.2 (0.07 a ) NVP 36.7 (0.71 b ) 104 (72.2%) 136 (97.1%) 127 (9.11 a ) 5.5 (0.07 a ) DTG 36 (1.61 a ) 20 (100%) (33.39 d ) EVG/c 31 (1.43 a ) 18 (90%) (33.39 d ) RAL 40 (1.61 a ) 19 (95%) (33.39 d ) ATV/r 35 (0.57 c ) 0 (0%) 13 (5%) 370 (10.73 c ) 4.56 (0.04 c ) EVG/c 34 (0.65 c ) 0 (0%) 12 (4%) 344 (9.59 c ) 4.46 (0.04 c ) Zhang et al, 2015 NVP (13.52 d ) 4.2 (0.07 d ) (NCT ) EFV (13.2 d ) 4.2 (0.07 d ) GS-US ; GS- US EVG/c 33 (0.4 c ) 733 (85%) (6.72 c ) 4.58 (0.02 c ) 652 (75%) 5 (1%) EVG/c 35 (0.4 c ) 740 (85%) (6.31 c ) 4.58 (0.02 c ) 645 (74%) 6 (1%) Median value reported; a SE calculated from the range; b SE calculated from SD; c SE calculated from interquartile range; d SE imputed; ABC: abacavir; ATV/r: ritonavir-boosted atazanavir; AZT: zidovudine; DRV/r: ritonavir-boosted darunavir; DTG: dolutegravir; EFV: efavirenz; EVG/c: 150

151 Elvitegravir/cobicistat; LPV/r: ritonavir-boosted lopinavir; NRTI: non-nucleoside reverse transcriptase inhibitor; NVP: nevirapine; RAL: raltegravir; RPV: rilpivirine; TDF: Tenofovir disoproxil fumerate; XTC: lamivudine or emtricitabine Table 27: Patient characteristics in the principal analysis in the systematic literature review among TB co-infected patients Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) Previous review (May 2015) ANRS Reflate TB trial CARINEMO N2R EFV 35 (1.64 c ) 39 (76) (27.02 c ) 4.9 (0.1 c ) RAL (1.34 c ) 35 (69) (16.91 c ) 4.9 (0.12 c ) RAL (1.03 c ) 38 (75) (29.48 c ) 5.5 (0.04 c ) EFV 33 (0.53 c ) 171 (60) 15 (5) 86 (4.21 c ) 5.5 (0.04 c ) RAL (0.53 c ) 160 (56) 19 (7) 92 (4.56 c ) 5.7 (0.04 c ) EFV 35.7 (0.95 b ) 46 (64.8) (8.09 b ) 5.75 (0.02 c ) NVP 38 (1.07 b ) 49 (69) (6.8 b ) 5.75 (0.03 c ) Sinha et al, 2013 EFV 34.8 (0.47 b ) 319 (83) (7.12 a ) 5.19 (0.06 a )

152 Study ID Treatment Arm Age mean (SE) Males n (%) AIDSdefining illness n (%) Baseline CD4 (cells/mm3) Mean (SE) Baseline viral load (log copies/ml) Mean (SE) Men who have sex with men n (%) Persons who inject drugs n (%) NVP 35.7(0.48 b ) 326 (85) (7.03 a ) 5.52 (0.06 a ) Swaminathan et al, 2011 EFV 34.4 (0.98 b ) 49 (83) (3.66 c ) 5.5 (0.04 c ) NVP 37.8 (1.02 b ) 44 (77) (10.01 c ) 5.7 (0.04 c ) Review update (2018) ANRS 129 BKVIR EFV 34 (1.59 c ) 49 (71) (12.04 c ) 5.4 (0.09 c ) 11 (16) -- HIV-TB Pharmagene EFV 35 (0.67 c ) 109 (52.6) (4.73 c ) 5.23 (0.05 c ) INSPIRING Sinha et al, 2017 DTG 33 (1.07 a ) 39 (57) 69 (100) 208 (25.15 c ) 5.1 (0.07 c ) EFV 32 (1.26 a ) 28 (64) 44 (100) 202 (29.26 c ) 5.24 (0.13 c ) EFV 36.7(0.62 b ) 111 (79.3) 134 (97.8) 133 (9.46 a ) 5.2 (0.07 a ) NVP 36.7(0.71 b ) 104 (72.2) 136 (97.1) 127 (9.11 a ) 5.5 (0.07 a ) TB-HAART EFV 37(0.78 b ) 95 (57) (3.89 c ) 4.9 (0.05 b ) Median value reported; a SE calculated from the range; b SE calculated from SD; c SE calculated from interquartile range; d SE imputed; ABC: abacavir; ATV/r: ritonavir-boosted atazanavir; AZT: zidovudine; DRV/r: ritonavir-boosted darunavir; DTG: dolutegravir; EFV: efavirenz; EVG/c: Elvitegravir/cobicistat; LPV/r: ritonavir-boosted lopinavir; NRTI: non-nucleoside reverse transcriptase inhibitor; NVP: nevirapine; RAL: raltegravir; RPV: rilpivirine; TDF: Tenofovir disoproxil fumerate; XTC: lamivudine or emtricitabine 152

153 Appendix G: Quality assessments of included studies The following table summarizes the critical appraisals for randomized and non-randomized studies using the Cochrane Risk of Bias instrument and the Tool to Assess the Risk of Bias in Cohort Studies, developed by the Clinical Advances through Research and Information Translation (CLARITY) group, respectively. Table 28: Cochrane risk of bias quality assessment for randomized controlled trials, arranged by review sub-population Trial Sequence Allocation Blinding Incomplete Selective Other Population generation concealment outcome data outcome reporting sources of bias Study Unclear Unclear Unclear Low Low Unclear Principal, NN Low Low High Low Low Low Principal, 2015 ACTG A5142 Low Low High Unclear Low Low Principal, 2015 ACTG A5202 Low Low High Low Low Low Principal, 2015 ACTG A5257 Low Low High Low Low Low Principal, 2015 Advanz-3 Low Low High High Low Unclear Principal, 2018 Albini et al, 2012 Low Low Unclear Low Low Low Principal, 2015 Altair Low Low High Low Low Low Principal, 2015 ANRS Reflate TB trial Low Low High Low Low Low HIV/TB ANRS Principal, Low Low High Low Unclear Low DAYANA Trial 2015

154 ARIA Low Low High Low Low Unclear Principal, 2018 ARROW Unclear Unclear High Low Low High Children ARTEMIS Low Low High Low Low Low Principal, 2015 ARTEN Unclear Low High Low Low Low Principal, 2015 ASSERT Unclear Low High High Low Low Principal, 2015 ATADAR Low Low High Low Low Low Principal, 2015 (updated) Avihingsanon et Principal, Low Low High Low Low High al, BAN Low Unclear High High Low High Pregnancy CARINEMO Low Low High Low Low Low HIV/TB CASTLE Low Low High Low Low Low Principal, 2015 CNA30024 Unclear Low Low Low Low Low Principal, 2015 CTN177 Low Low High Low Unclear Low Principal, 2015 DAUFIN Unclear Low High High Low Low Principal, 2015 DolPHIN 1 Unclear Unclear High Low Low High Pregnancy DRIVE AHEAD Unclear Unclear Low Unclear Unclear Unclear Principal, 2018 DRIVE FORWARD Unclear Unclear Unclear Unclear Unclear Unclear Principal, 2018 Principal, ECHO Low Low Low Low Low Low

155 ENCORE1 Low Low Unclear Low Low Low Principal, 2018 ENCORE1 Low Low Low Low Low Low Principal, 2015 Epzicom-Truvada Low Low High Low Low Low Principal, 2015 FLAMINGO Low Low High Low Low Low Principal, 2015 GESIDA 3903 Low Low High Low Low Low Principal, 2015 GS-US Low Low Low Low Low Low Principal, 2018 GS-US Low Low Low Low Low Low Principal, 2015 GS-US Low Low Low Low Low Low Principal, 2015 GS-US Low Low Low Low Low Low Principal, 2015 GS-US Unclear Unclear Unclear Low Low Unclear Principal, 2018 GS-US Low Low Low Low Low Low Principal, 2018 GS-US Low Low Low Unclear Low Unclear Principal, 2018 GS-US Low Low Low Unclear Low Unclear Principal, 2018 GS-US Low Low Low Low Low Low Principal, 2018 GS-US Low Low Low Low Low Low Principal, 2018 GS-US Low Low Low Low Low Low Principal, 155

156 2018 HEAT Unclear Low Low High Low Low Principal, 2015 IMPAACT P Cohort 1 Unclear Unclear High Low Low High Children IMPAACT P Cohort 2 Unclear Unclear High Low Low High Children INSPIRING Unclear Unclear High Unclear Low Low HIV/TB Lake Study Unclear Low High High Low Low Principal, 2015 Li et al, 2008 Unclear Low High High Low Low Principal, 2015 Maggiolo et al, Principal, Low Low Low Low Low Unclear Principal, MASTER Low Low High Low Unclear High 2015 METABOLIK Unclear Low High Low Low Low Principal, 2015 MMA BANA Study Unclear Unclear Unclear Low Low High Pregnancy N2R Unclear Unclear High Low Low Low HIV/TB NEWART Low Low High Low Low Low Principal, 2015 NVP Unclear Low High High High High HIV/TB OzCombo 2 Unclear Low High Low Low Low Principal, 2015 PEARLS Low Low High Low Low Low Principal, 2015 Phidisa II Low Low High Low Low Low Principal, 2015 PHIDISA II Unclear Low High Low Low Low HIV/TB 156

157 PROMISE 1077BF/1077FF Unclear Unclear High Low Low High Pregnancy PROMOTE Unclear Unclear High Low Low High Pregnancy Protocol 004 Unclear Unclear Low Low Low Low Principal, 2015 SEARCH 003 Low Low High High Low Low Principal, 2015 SENC Unclear Low High High Low Low Principal, 2015 Sierra-Madero et Principal, Low Low High High Low Low al, Principal, SINGLE Low Low Unclear Low Low Low 2015 (updated) Sinha et al, 2017 Unclear Low Unclear Unclear Low High Principal, 2018 SPRING-1 Low Low High Low Low Low Principal, 2015 SPRING-2 Low Low Low Low Low Low Principal, 2015 SSAT066 Unclear Unclear Unclear High High Unclear Principal, 2018 STaR Low Low High Low Low Low Principal, 2015 (updated) STARTMRK Low Low Low Low Low Low Principal, 2015 Study 903 Low Low High High Low Low Principal, 2015 Principal, Study 934 Low Low High High Low Low

158 Swaminathan et al, 2011 Low Low High Low Low Low HIV/TB TB-HAART Unclear Unclear High High Low Low HIV/TB THRIVE Low Low Low Low Low Low Principal, 2015 TMC278-C204 Low Low High Low Low Low Principal, 2015 Tshepo Low Low High Unclear Low Low Principal, 2015 Tshepo Unclear Unclear High Unclear High Low Pregnancy Japanese Anti- Principal, Unclear Low High Low Unclear Low HIV-1 QD Therapy 2015 Principal, WAVES Low Low Low Low Low Low 2018 Principal, Zhang et al, 2015 Unclear Unclear Unclear Unclear Unclear Unclear

159 Table 29: Critical appraisal of non-randomized studies using the Tool to Assess the Risk of Bias in Cohort Studies, developed by the Clinical Advances through Research and Information Translation (CLARITY) group Can we Was selection of exposed and Can we be confident Trial non-exposed in the cohorts assessment drawn from the same of exposure? population? HIV-TB Pharmagene Probably Probably ANRS 129 BKVIR N/A Probably DART Definitely Definitely IMPAACT 1026s N/A Probably be Can we be confident confident Can we be that the in the confident Was the outcome assessment in the follow up of Q4* of the assessment of cohorts interest presence or of adequate? was not absence of outcome? present prognostic at start of factors? study? Definitely Definitely Definitely Definitely Definitely Definitely Definitely N/A Definitely Probably Definitely Probably Definitely Definitely N/A Probably no Were co- Interventions similar Population between groups? Probably Definitely HIV/TB Definitely N/A HIV/TB Probably Definitely Pregnancy Probably N/A Pregnancy Ngoma et al, N/A Probably Definitely N/A Probably Definitely Probably N/A Pregnancy

160 2015 no no Samuel et al, 2014 N/A Probably Definitely N/A Definitely Definitely Probably no N/A Pregnancy SMAC Definitely Probably Definitely Probably Probably Probably Definitely Definitely Pregnancy TSHEPISO Definitely Definitely Definitely Definitely Probably no Definitely Definitely Definitely Pregnancy Zash et al, 2018 Definitely Probably Definitely Definitely Probably Probably Probably Definitely Pregnancy Thorne et al, 2017 N/A Probably Definitely N/A Probably Definitely Probably N/A Pregnancy Lamorde et al, 2017 N/A Probably Probably N/A Probably Probably Probably N/A Pregnancy IMPAACT P1030 N/A Definitely Definitely N/A Definitely Definitely Definitely N/A Children IMPAACT P1083 N/A Definitely Definitely N/A Definitely Definitely Definitely N/A Children PACTG1030 N/A Definitely Definitely N/A Probably no Definitely Definitely N/A Children IMPAACT P1110 N/A Definitely Definitely N/A Probably no Definitely Definitely N/A Children IMPAACT P1066 N/A Probably Probably N/A Probably no Probably Definitely N/A Children 160

161 IMPAACT P Cohort I N/A Definitely Definitely N/A Probably Definitely Definitely N/A Children IMPAACT P Cohort IIA N/A Probably Probably N/A Probably no Probably Probably N/A Children Q4: Did the study match exposed and unexposed for all variables that are associated with the outcome of interest or did the statistical analysis adjust for these prognostic variables? 161

162 Appendix H: Network diagrams Figure 22: Network diagram of trials informing viral suppression at 48 weeks (A); 96 weeks (B); and 144 weeks (C). (A)

163 (B) (C) Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the 163

164 number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. Figure 23: Network diagram of the 66 trials informing mean change from baseline in CD4 cell counts at 48 weeks (A); 96 weeks (B); and 144 weeks (C). (A) (B) 164

165 (C) Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 165

166 Figure 24: Network diagram of the trials informing mortality Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 166

167 Figure 25: Network diagram of the comparative trials reporting AIDS defining illnesses outcome Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 167

168 Figure 26: Network diagram of the trials informing discontinuation due adverse events among firstline HIV patients Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 168

169 Figure 27: Network diagram of the trials informing retention among first-line HIV patients Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 169

170 Figure 28: Network diagram of the trials informing the treatment-related adverse events analysis Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 170

171 Figure 29: Network diagram of the trials informing the treatment-emergent adverse events analysis Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 171

172 Figure 30: Network diagram of the trials informing the treatment-related serious adverse events analysis Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 172

173 Figure 31: Network diagram of the trials informing drug emergent serious adverse events among firstline HIV patients Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 173

174 Figure 32: Network diagram of the trials informing regimen substitutions Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 174

175 TB Subpopulation Figure 33: Network diagram of trials informing viral suppression at the 24-week (A) and 48-week (B) timepoints in HIV-TB co-infected patients. (A) (B) Legend: Circles (nodes) in the diagrams represent individual treatments, lines between circles represent availability of head-to-head evidence between two treatments, and the numbers on the lines are the number of RCTs informing each head-to-head comparison. Blue: NNRTIs; Green: Protease inhibitors; Orange: Integrase inhibitors. 175