UNIVERSITY OF THE WITWATERSRAND JOHANNESBURG FACULTY OF HEALTH SCIENCES Trend of Pre-antiretroviral Therapy HIV-1 Drug Resistance in Kilombero and Ulanga Antiretroviral Cohort, South-Western Tanzania, for over 15 years (2005-2020). Alex NTAMATUNGIRO A thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy JANUARY 2024 ii DECLARATION I, Alex Ntamatungiro declare that this thesis is my original work and has not been submitted before for any degree or award at this or any other university. Where information has been derived from other sources, I confirm that this has been duly acknowledged. Parts of this work have been published as open access articles, I (and my co-authors) retained the copyright for that work. Alex Ntamatungiro 17th January 2024 iii I dedicate this work to my wife Maria, and our children Ethan, and Alvin iv PUBLICATIONS ARISING FROM THIS RESEARCH PROJECT 1. Ntamatungiro AJ, Kagura J, Weisser M, Francis JM. Pre-treatment HIV-1 drug resistance in antiretroviral therapy-naive adults in Eastern Africa: a systematic review and meta- analysis. J Antimicrobial Chemotherapy. 2022 Nov 28;77(12): 3231-3241.doi: 10.1093/jac/dkac338. PMID: 36225089. https://pubmed.ncbi.nlm.nih.gov/36225089/ My contribution to the paper Conceptualization of the study, generation of the hypothesis, conducting literature searches and articles reviews for inclusion eligibility, statistical analyses, synthesizing and presenting findings, writing all drafts of the paper, addressing co-authors’ and peer reviewers’ comments and checking proofs for the final accepted and published manuscript. 2. Ntamatungiro AJ, Anna Eichenberger, James Okuma, Fiona Vanobberghen, Robert Ndege, Namvua Kimera, Joel M. Francis, Juliana Kagura, Maja Weisser, on behalf of the KIULARCO Study Group: Transitioning to dolutegravir in a programmatic setting: Virological outcomes and associated factors in the Kilombero and Ulanga antiretroviral cohort in rural Tanzania: Open Forum Infectious Diseases, Volume 10, Issue 7, July 2023, ofad321,https://doi.org/10.1093/ofid/ofad321. https://pubmed.ncbi.nlm.nih.gov/37520425 My contribution to the paper Conceptualization of the study, generation of the hypothesis, planning, and running all lab experiments, and statistical analyses, synthesizing and presenting findings, writing all drafts of the paper, addressing co-authors’ and peer reviewers’ comments and checking proofs for the final accepted and published manuscript. 3. Ntamatungiro AJ, Kagura J, Mnzava D, Ndege R, Okuma J, Vanobberghen F, Francis JM, Weisser M, on behalf of the KIULARCO Study Group: Recent HIV-1 infection and pre-treatment drug-resistance in recently infected, treatment-naïve persons in rural Tanzania: Submitted to the Journal of Public Health - http://jpubhealth.oupjournals.org (under review). My contribution to the paper Conceptualization of the study, generation of the hypothesis, planning, and running all lab experiments, and statistical analyses, synthesizing and presenting findings, writing all drafts of the paper, and addressing co-authors’ comments and checking proofs for the final submitted manuscript. https://doi.org/10.1093/ofid/ofad321 v 4. Ntamatungiro AJ, Francis Joel, Ndege Robert, Okuma James, Tschumi Nadine, Vanobberghen Fiona, Metzner Karin, Paris Daniel, Weisser Maja and Kagura Juliana on behalf of the KIULARCO Study Group: Trends in prevalence of the nucleoside/nucleotide reverse-transcriptase inhibitor (NRTI) associated mutations in newly diagnosed ART naive patients in the Kilombero and Ulanga Antiretroviral Cohort during the 15 years after ART rollout in Tanzania. To be submitted to the Journal of AIDS. My contribution to the paper Conceptualization of the study, generation of the hypothesis, planning, and running all lab experiments, and statistical analyses, synthesizing and presenting findings, writing all drafts of the paper, addressing co-authors’ comments and preparing the final manuscript for submission. PRESENTATIONS ARISING FROM THIS RESEARCH PROJECT 1. Ntamatungiro A, Eichenberger A, Okuma J, Vanobberghen F, Ndege R, Kimera K, Francis J, Kagura J, Weisser. Transitioning to dolutegravir: virological outcomes in an HIV cohort in rural Tanzania (AS-EACS-20221-00296): 18th European AIDS Conference (EACS) October 27-30, 2021, London, United Kingdom (ePoster presentation). 2. Ntamatungiro A, Eichenberger, Okuma J, Vanobberghen F, Mapesi H, Ndege R, Mnzava D, Kimera N, Weisser M. Transitioning to Dolutegravir in a programmatic setting: Virological outcomes in the Kilombero and Ulanga Antiretroviral Cohort in rural Tanzania: 9th Annual Joint Advanced Seminars Conference 29th July 2022, Kampala, Uganda (Oral presentation). 3. Ntamatungiro A, Neumann k, METZNER K, PARIS D, and WEISSER M. HIV-1 Drug Resistance Surveillance in Tanzania: Swiss TPH Virtual Symposium 25th October 2022 (Oral presentation). 4. Ntamatungiro A, Francis JM, Mnzava D, Ndege R, Njau PS, Okuma J, Vanobberghen F, Kagura J, Weisser M, on behalf of the KIULARCO Study Group. Recent HIV-1 Infection and Pre-Treatment HIV Drug-Resistance in Recently Infected Adults Initiating Antiretroviral Therapy in Rural Tanzania: Wits School of Public Health Research Day and CARTA Conference 14-15 September 2023, Johannesburg, South Africa (Poster presentation). vi ABSTRACT Introduction Pre-treatment HIV drug-resistance (PDR) may result in increased risk of virological failure and subsequently acquisition of new HIV drug resistant mutations. With recent increase in antiretroviral therapy (ART) coverage and periodic modifications of the guidelines for HIV treatment, monitoring changes in levels of PDR is critical, particularly in under-sampled areas, such as rural Tanzania. This PhD project aimed to determine the trend and patterns of PDR in the Kilombero and Ulanga antiretroviral cohort (KIULARCO), analyse the impact of recent HIV-1 infection, and dolutegravir rollout in rural Tanzania. Methods The study comprised a systematic review and meta-analysis of primary studies about prevalence of PDR among ART-naive people living with HIV (PLHIV) (³15 years old), published between 2017 and 2022. The data had to be in one or several of the countries of Eastern Africa, namely, Ethiopia, Kenya, Malawi, Rwanda, Mozambique, Tanzania, and Uganda. Thereafter, cross- sectional analyses of data on newly HIV-1-diagnosed ART-naïve adults (aged ≥ 15 years), enrolled in the on-going prospective clinic-based observational rural antiretroviral cohort- KIULARCO focusing on various aspects of PDR. Multivariate logistic regressions were used to determine the factors associated with recent HIV-1 infection, and viral suppression at 12-months in patients initiating dolutegravir-based ART in the KIULARCO. Results Overall, the pooled prevalence estimate of any PDR was 10.0% (95% CI: 7.9%–12.0%, I2 =88.9%) among 22 studies in the general adults’ population, which was higher than the previously reported prevalence of 8.7% using data available until 2016 in the Eastern Africa region. PDR was mainly driven by non-nucleoside reverse transcriptase inhibitors (NNRTI); whereas the pooled prevalence of PDR to nucleoside reverse transcriptase inhibitors (NRTI) was 2.6% (95% CI: 1.8%–3.4%, I2=69.2%). Remarkably, PDR to NRTIs in a sub-population of recently HIV-1 infected PLHIV in the KIULARCO was high at 12.5%. Also, there was a notable tendency to an increasing prevalence of PDR to NRTI, with the overall prevalence of 2.1% in the first five-year period (2005-2009) of the ART program in Tanzania, and 3.4 % in the most recent period (2019-2022). Moreover, there was no PDR to the dolutegravir co-administered NRTI in those with viremia ≥50 copies/mL, at one year, in patients initiating dolutegravir-based ART in the KIULARCO 2 years after dolutegravir roll. Notably, dolutegravir-based ART was associated with >2 times the odds of viral suppression compared to NNRTI-based ART with an adjusted odds ratio (aOR) of 2.10 (95% CI 1.12-3.94). vii Conclusions There is notable level of PDR to NRTI among general adults’ population in Eastern Africa region, that was high among recently HIV-1 infected PLHIV in a representative rural Sub-Saharan Africa setting. Hence, routine surveillance of pre-existing resistance to the DTG co-administered NRTI remains particularly important, in resource-limited settings, to prevent risk of failure of newer antiretroviral agents such as dolutegravir, which would be detrimental to Tanzania and other low- and middle-income countries for the aim to “end AIDS by 2030”. Our results underline the benefit of programmatic uptake of dolutegravir -based ART in low- and middle-income countries. viii ACKNOWLEDGEMENTS First and foremost, I am extremely grateful to my supervisors, Dr. Juliana Kagura, Dr. Joel Francis, and Prof. Maja Weisser for their invaluable advice, continuous support, and patience during my PhD study. Their immense knowledge and plentiful experience have encouraged me in all the time of my academic research and daily life. I would also like to thank Dr. Fiona Vanobberghen, Dr. James Okuma, and Dr. Anna Eichenberger for their contributions to the KIULARCO platform and advice on my study. My gratitude extends to the Consortium for Advanced Research Training in Africa (CARTA) for the funding opportunity to undertake my studies at the the Faculty of Health Sciences, University of the Witwatersrand, I would like to thank all the members of the KIULARCO Study Group. Finally, I would like to express my sincere gratitude to my wife and my children. Without their tremendous understanding and encouragement in the past few years, it would be impossible for me to complete my study. ix TABLE OF CONTENTS DECLARATION .................................................................................................................... ii PUBLICATIONS ARISING FROM THIS RESEARCH PROJECT ...................................... iv PRESENTATIONS ARISING FROM THIS RESEARCH PROJECT .................................... v ABSTRACT ........................................................................................................................... vi ACKNOWLEDGEMENTS .................................................................................................. viii LIST OF FIGURES ............................................................................................................ xiii LIST OF TABLES ............................................................................................................... xiii LIST OF ABBREVIATIONS ................................................................................................ xv PART I ................................................................................................................................... 1 GENERAL INTRODUCTION ................................................................................................ 1 CHAPTER 1 ........................................................................................................................... 1 1.1. BACKGROUND OF STUDY ....................................................................................... 1 1.1.1 HIV burden and the evolution of antiretroviral therapy initiation guidelines ................... 1 1.1.2 HIV drug resistance ........................................................................................................... 2 1.1.2.1 Factors associated with development of HIV drug resistance .......................................................... 3 1.1.2.2 HIV drug resistance testing .............................................................................................................. 3 1.1.3 The burden of pre-treatment HIV drug-resistance ............................................................ 5 1.2 LITERATURE REVIEW ............................................................................................. 6 1.2.1 Introductory Information .................................................................................................. 6 1.2.2 Research Gap ...................................................................................................................... 7 1.2.3 Key concepts ....................................................................................................................... 8 1.2.3.1 Transmitted and pre-treatment HIV-1 drug resistance ..................................................................... 8 1.2.3.2 HIV-1 drug-resistant minority variants ............................................................................................ 9 1.2.3.3 Recent HIV infections testing assays and their limitations. ........................................................... 10 1.2.3.4 Recent HIV infection and drug resistance ...................................................................................... 10 1.2.3.5 HIV-1 subtypes and circulating strains in Tanzania ........................................................................ 11 1.2.3.6 Pre-treatment HIV drug resistance to the dolutegravir co-administered Nucleoside reverse transcriptase inhibitors ...................................................................................................................................... 11 1.2.3.7 Trends of HIV-1 pretreatment HIV drug-resistant mutations ........................................................ 12 1.2.4 Conceptual framework and concept maps ....................................................................... 13 1.2.5 Methods for studies of HIV drug resistance prevalence .................................................. 15 1.3 AIM OF THE PhD WORK ........................................................................................ 16 1.4 OUTLINES OF THIS PhD THESIS ......................................................................... 16 CHAPTER 2 ......................................................................................................................... 17 x 2.0 Methodology .............................................................................................................. 17 2.1 Research settings .............................................................................................................. 17 2.2 Study population .............................................................................................................. 18 2.5 Laboratory analysis .......................................................................................................... 21 2.5.1 RNA Extraction ................................................................................................................................. 21 2.5.2 Reverse Transcription and PCR ...................................................................................................... 21 2.5.3 HIV Genotypic resistance testing .................................................................................................... 21 2.6 Bioinformatics analysis .................................................................................................... 22 2.6.1 Sequence reads assembly, analysis, and annotation ............................................................................ 22 2.6.2 Drug Resistance analysis .................................................................................................................... 22 2.7 Ethical approval ................................................................................................................. 22 PART 2 ................................................................................................................................. 24 EMPIRICAL CHAPTERS .................................................................................................... 24 CHAPTER 3 ......................................................................................................................... 25 A systematic review and meta-analysis to assess the levels of pretreatment HIV-1 drug resistance in antiretroviral therapy- naïve in Eastern Africa (2017-2022). ................................................... 25 Introduction .................................................................................................................................. 25 Background .................................................................................................................................. 27 Methods ........................................................................................................................................ 28 Study design ..................................................................................................................................................... 28 Search Strategy and Selection Criteria ......................................................................................................... 28 Study Selection process and Extraction ........................................................................................................ 28 Data Analysis ................................................................................................................................................... 29 Characteristics of Included studies ................................................................................................................ 29 The Overall Prevalence of Pretreatment Drug Resistance-Associated Mutations .................................... 35 PDR of RT inhibitors ...................................................................................................................................... 37 PDR of PIs ........................................................................................................................................................ 37 PDR of INSTIs ................................................................................................................................................. 37 Discussion ..................................................................................................................................... 40 Conclusion .................................................................................................................................... 42 CHAPTER 4 ......................................................................................................................... 48 Recent HIV-1 infection and pre-treatment drug-resistance in recently infected, treatment-naïve persons in rural Tanzania. ........................................................................................................... 48 Introduction .................................................................................................................................. 48 Abstract ......................................................................................................................................... 50 Background .................................................................................................................................. 51 Methods ........................................................................................................................................ 52 Study design and population .......................................................................................................................... 52 Study area ........................................................................................................................................................ 52 Inclusion criteria ............................................................................................................................................. 52 Study procedures ............................................................................................................................................. 52 Laboratory procedures ................................................................................................................................... 52 Definitions and covariates .............................................................................................................................. 53 Statistical analysis ........................................................................................................................................... 53 Ethical Considerations .................................................................................................................................... 54 xi Results .......................................................................................................................................... 54 Baseline characteristics ................................................................................................................................... 54 Recent HIV infection and associated factors ................................................................................................ 56 Pretreatment HIV-1 drug resistance mutations ........................................................................................... 61 Discussion ..................................................................................................................................... 61 Conclusion .................................................................................................................................... 63 CHAPTER 5 ......................................................................................................................... 65 Trends in prevalence of the nucleoside reverse-transcriptase inhibitor associated mutations in newly diagnosed antiretroviral therapy-naive patients in the Kilombero and Ulanga Antiretroviral cohort in Tanzania. ............................................................................................... 65 Introduction .................................................................................................................................. 65 Abstract ............................................................................................... Error! Bookmark not defined. Background .................................................................................................................................. 68 Methods ........................................................................................................................................ 68 Setting and population .................................................................................................................................... 68 Data collection ................................................................................................................................................. 69 Laboratory analysis ........................................................................................................................................ 69 Ethics approval ................................................................................................................................................ 69 Results and Discussion ................................................................................................................. 70 Conclusion .................................................................................................................................... 71 CHAPTER 6 ......................................................................................................................... 75 Virological-outcomes at one-year, and its predictors after initiating dolutegravir (DGT)-based antiretroviral therapy (ART), and prevalence of pretreatment HIV-1 drug resistance among patients failing on DTG-based ART in a rural cohort, in Tanzania. ............................................ 75 Introduction .................................................................................................................................. 75 Background .................................................................................................................................. 77 Methods ........................................................................................................................................ 78 Study design and setting ................................................................................................................................. 78 Study population ............................................................................................................................................. 78 Data collection ................................................................................................................................................. 79 Laboratory measurements ............................................................................................................................. 79 Definitions ........................................................................................................................................................ 79 Study Outcomes ............................................................................................................................................... 80 Statistical methods .......................................................................................................................................... 80 Ethical considerations ..................................................................................................................................... 81 Results .......................................................................................................................................... 81 Study Population and Baseline Characteristics ............................................................................................ 81 Virological Outcome ....................................................................................................................................... 84 Factors associated with virological suppression ........................................................................................... 86 HIV-1 drug resistance associated mutations ................................................................................................. 88 Side effects and pregnancy outcomes ............................................................................................................ 89 Discussion ..................................................................................................................................... 89 Conclusion .................................................................................................................................... 92 PART 3 ................................................................................................................................. 98 GENERAL DISCUSSION .................................................................................................... 98 xii CHAPTER 7 ......................................................................................................................... 99 Summary, Overall discussion, and Conclusions ........................................................................... 99 Highlights of the key findings ........................................................................................................................... 99 The major conclusions from this PhD research. .............................................................................................. 101 RECOMMEDANTION .................................................................................................................................. 102 References: ......................................................................................................................... 103 Appendices ......................................................................................................................... 123 xiii LIST OF FIGURES CHAPTER 1 Figure 1.1 Evolution of first line ART in Tanzania…………………………………………….…1 Figure 1.2 WHO recommended programmatic shift to integrase-based therapy…………………2 Figure 1.3 Concept map of pre-Antiretroviral Therapy HIV-1 drug resistance and associated factors…………………………………………………………………………………………….12 CHAPTER 2 Figure 2.1. Study area for paper II-IV; Kilombero and Ulanga districts in Morogoro region, South-Western Tanzania................................................................................................................16 Figure 2.2. Patients’ enrolment into the Kilombero and Ulanga Antiretroviral cohort by year………………………………………………………………………………………………………..…17 Figure 2.3 Schematic presentation of HIV-1 viral genome……………………………………...18 CHAPTER 3 Figure 1 Flow diagram of search results and screening……………………………………………………………………………..…………..26 Figure 2 Overall prevalence of PDR. Note: the red dotted line indicates the overall prevalence of PDR. ……………………………………………………………………………………………..32 Figure 3. Prevalence of PDR to Non-nucleoside reverse transcriptase inhibitor………………………………………………………………………………………......34 Figure 4. Prevalence of PDR to nucleoside reverse transcriptase inhibitor……………..………35 Figure 5. Prevalence of PDR to protease inhibitor…...………………………..………..……….36 CHAPTER 4 Figure 1 Study participants flowchart……………….……………………………………......….51 CHAPTER 6 Figure 1. Patient flowchart for those initiating DTG or NNRTI-based regimens …………..…..78 Figure 2. Virologic outcome at 12 months after initiation of reverse transcriptase inhibitor (NNRTI)‐ and dolutegravir (DTG)‐based antiretroviral therapy (ART) in treatment naïve…….81 LIST OF TABLES CHAPTER 2 Table 1. Sample size calculations……………………………………………………………….20 Table 2. Sequencing primers information………………………………………………………..22 xiv CHAPTER 3 Table 1. The characteristics of included datasets with reverse transcriptase and/or protease sequences from ART-naïve adults………………………………………….……………………27 Table S1. Characteristics of datasets with Reverse Transcriptase (RT)/AND OR Protease (PR) sequences of ART-naïve adults in eight Eastern Africa Countries (2005-2022) ……...………..40 CHAPTER 4 Table 1. Characteristics of study participants……………………………………...…………….52 Table 2. Risk factors associated with recent HIV infection ………………………..……….…..54 Table 3. Distribution of Pretreatment Drug Resistance-Associated Mutations- and HIV-1 subtype(s) among recently HIV-1 infected people living with HIV………………….…………56 CHAPTER 5 Table 1. Prevalence of pre-antiretroviral treatment HIV drug resistance mutations detected in antiretroviral-naïve adult in “2005-2009” and “2019-2022………………………...……...…….68 Table 2. Prevalence of HIV-1 subtypes detected in antiretroviral-naïve adult in “2005-2009” and “2019-2022” …………………………………………………………………...………...………69 CHAPTER 6 Table 1. Patients’ characteristics at initiation of NNRTI- or DTG-based ART regimens………79 Table 2. Factors associated with viral suppression (HIV-1 RNA <50 copies/ml) at 12 months (5-15 month) among treatment-naïve patients initiating NNRT-based or DTG-based ART……82 Table 3. Patterns of mutations detected in the reverse transcriptase and protease region of the HIV-1 pol -sequences in participants on DTG -based ART with VL>=50 at 12 months….…....85 Table S1. Patients’ characteristics at initiation for patients without and with 12-month viral measurement………………………………………………………………………...…….……..90 Table S2. Factors associated with viral suppression (HIV-1 RNA <50 copies/ml) at 12 months (+/- 3 months) among patients initiating NNRT-based or DTG-based ART…………….…...…92 xv LIST OF ABBREVIATIONS ABC Abacavir ART Antiretroviral Therapy CDCI Chronic Diseases Clinic CTC Care and Treatment Centre d4T Stavudine ddI didanosine DTG Dolutegravir EFV Efavirenz FTC Emtricitabine HIV-1 Human Immunodeficiency Virus type 1 HIVdb HIV Drug Resistance Database HIVDR HIV-1 Drug Resistance HVL HIV viral load INSTI Integrase Inhibitors KIULARCO Kilombero & Ulanga Antiretroviral Cohort 3TC Lamivudine LMIC Low- and Middle-Income Countries NACP National AIDS Control Care Programme NGS Next-Generation Sequencing NNRTI Non-Nucleoside Reverse Transcriptase Inhibitor NRTI Nucleoside Reverse Transcriptase Inhibitor PCR Polymerase Chain Reaction PDR Pre-treatment HIV-1 drug resistance PI Protease Inhibitors PLHIV People living with HIV. PMTCT Prevent mother-to-child transmission PR Protease RHI Recent HIV infection RLS Resources Limited Settings RNA Ribonucleic Acid RT Reverse transcriptase RT-PCR Reverse transcriptase polymerase chain reaction SSA Sub-Saharan Africa xvi TDF Tenofovir Disoproxil Fumarate TDR Transmitted HIV-1 drug resistance TLD Tenofovir/Lamivudine/Dolutegravir WHO World Health Organization 1 PART I GENERAL INTRODUCTION 1 CHAPTER 1 1.1. BACKGROUND OF STUDY 1.1.1 HIV burden and the evolution of antiretroviral therapy initiation guidelines Currently about 38 million people are living with HIV globally and two thirds of these reside in Sub-Saharan Africa (SSA)(1). The HIV prevalence in Tanzania has stabilized at 4.7 % (about 1.7 million people living with HIV (PLHIV) in 2021) and new infections have been reduced to a still high number of 54 000 new HIV infections in 2021. The global epidemiology of HIV infection has changed markedly because of the rapid expanding access to antiretroviral therapy (ART). For example, by the end of 2021, 28.7 million people were accessing ART(2), up from 7.8 million [6.9–7.9 million] in 2010. Also, the ART coverage increased from 24% in 2010 to 75% in 2021(2). By the end of 2021 about 86% of PLHIV in Tanzania were receiving antiretroviral treatment(1). There has been major improvement in antiretroviral medication with replacement of nucleoside analogues (NRTIs) with long-term toxicity like stavudine (d4T) and didanosine (ddI) to more favorable NRTIs like tenofovir disoproxil fumarate (TDF)(3). Until recently most low- and middle-income countries (LMIC) were using 2 NRTIs and 1 non-nucleoside reverse transcriptase inhibitor (NNRTI) as first line treatment recommended by World Health Organization (WHO)(4,5). The preferred regimen was tenofovir disoproxil fumarate (TDF) + lamivudine (3TC) or emtricitabine (FTC) + efavirenz (EFV) as a fixed dose combination. Figure 1.1 Evolution of first line ART in Tanzania 2 In many high-income countries, integrase-inhibitors as a new drug class have already replaced NNRTIs in first line treatment(6). Available since 2014, integrase inhibitors (INSTI) in several randomized clinical trials have shown faster viral suppression, higher genetic barrier to developing resistance and fewer side effects compared to NNRTIs in treatment-naïve and experienced patients(7,8). Dolutegravir (DTG) an integrase inhibitor for treatment of HIV infection became available to LMIC by the end of 2017, as the generic fixed dose combination tenofovir/lamivudine/dolutegravir (TLD), and since then has been rolled-out in many countries in SSA(9,10). Tanzania has been transitioning to dolutegravir-based first-line regimens since January 2019. Figure 1.2 WHO recommended programmatic shift to integrase-based therapy. 1.1.2 HIV drug resistance HIV drug resistance refers to the ability of viruses to continue multiplying even in presence of antiretroviral drugs that usually inhibits its replication. The phenomenon occurs mainly when the virus develops changes in its genetic information, referred to as “mutations”, which alter HIV proteins, that the virus uses to replicate, including the reverse transcriptase, protease, and integrase enzymes. Antiretroviral Therapy in Tanzania, Treatment naive patients 2NRTIs + INSTI • TDF/3TC /DTG= TLD March 2019 2NRTIs + NNRTI • TDF/3TC /EFV= TLE 3 HIV drug resistance mutations may naturally emerge due to random errors occuring because of the high viral replication rates, or during episodes of incomplete viral suppression, occurring mainly due to suboptimal compliance and irregular drug taking behaviour, which allows the virus to replicate and develop new HIV drug resistant mutations over time. This is known as acquired drug resistance. In some cases, people acquire a strain of HIV that already has the mutations conferring resistance to antiretroviral (ARV), that is known as transmitted drug resistance (TDR). While TDR are resistances detected in patients with no prior history of HIV drug exposure, a synonym term “pre-treatment HIV drug resistance” (PDR) OR “pre-antiretroviral therapy” refers to resistance detected in ARV drug-naive people initiating ART or people with prior ARV drug exposure initiating or reinitiating first-line ART(11). 1.1.2.1 Factors associated with development of HIV drug resistance The main factor that can lead to uncontrolled viral replication, is poor adherence to ART(12). If ART are taken at sub-optimal level—and it is not diagnosed quickly, the virus can replicate and develop HIV drug resistant mutations. Other factors may include poor absorption: HIV drugs need to be taken consistently on schedule. Additionally, drugs must be absorbed properly into the bloodstream to an optimal drug level, to adequately inhibit the virus replication. Another factor is the pharmacokinetics profile of the antiretrovirals: which refers “to how the drugs are absorbed, distributed, broken down and eliminated from the body. Interactions between drugs can interfere with absorption. Some HIV medications require a booster to maintain high enough levels in the body (for example, protease inhibitors boosted with ritonavir)”(13,14). 1.1.2.2 HIV drug resistance testing HIV drug resistance can be tested by examining the virus genetic sequences to see whether it contains mutations on the drug targets regions of the viral genome, mainly in the reverse transcriptase, protease, and integrase enzymes. For example, the K65R mutation that reduces TDF and ABC susceptibility, is resulted from the lysine (K)-to-arginine (R) substitution at residue 65 (K65R) in human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT), results from a single nucleotide G-to-A transition (AAA to AGA)(15). Genotypic HIV drug resistance testing is achieved by using genetic analyzers equipment. The main sequencing methods are viral population-Sanger bases Sequencing (SBS), and high throughput deep sequencing method or Next Generations sequencing (NGS). Additionally, to archive genotypic resistance testing there need to be established laboratory protocols and bioinformatics tools for mutations annotations, such as the Stanford University’s HIV Drug Resistance Database (HIVdb)(16,17). NGS compared to SBS allows for the detection of low-frequency mutation not identified by SBS. Also, NGS https://hivdb.stanford.edu/ 4 generates massive genomic data with large genome coverage compared to SBS, which allows details analysis of variants(18). Phenotypic test is another method to test for the HIV drug resistance, that measures the behaviour of HIV in the presence of specific medications. A sample of HIV from a blood test is exposed to various concentrations of different drugs in a laboratory to see whether the virus can still replicate. Its ability to do so is compared against wild-type virus, which is known to be susceptible to all antiretrovirals(19). 5 1.1.3 The burden of pre-treatment HIV drug-resistance Recent data from ART-naïve cohorts indicate varying trends in prevalence of PDR across regions globally. For instance, the analysis report of over 33,000 pooled sequences worldwide between 1996 and 2016, indicated an increasing trend of HIV drug resistance within non-B subtype virus populations(20). Importantly, evidence from European HIV-cohorts where HIV is predominantly subtype B, PDR reports indicate a stable at <7%, to a declining trend overtime(21–23). The declining prevalence of PDR in recent studies among western population, is likely to be explained by the routine virologic monitoring including HIV drug resistance testing, allowing early detection of HIV drug resistant mutations and switch to an active drug regimen, as well as an increasing usage of ART regimens with a higher genetic barrier to developing resistance, for example, the use of INSTI(24). Conversely, available data from studies in SSA as in many other LMICs show a rise of PDR. A systematic review and meta-analysis involving ART-naïve adult populations in Eastern and Southern Africa reported a significantly increased PDR over time(11,25,26). However, the study declares several important limitations: For example, in Eastern Africa less than 30% of included studies come from rural settings, so the analysis reflects mostly the prevalence of PDR primarily in urban and peri-urban areas. Elsewhere, differences in HIV drug resistance burden between urban and rural settings have been documented(27,28). Also, the large amount of the included studies is not nationally representative, which makes extrapolation challenging. Similarly, sampling in the East African studies encompassed only studies published in 2005-2009, during which period ART represented in the current WHO guidelines, e.g., tenofovir-based ART were still relatively new in many LMICs. In Tanzania like many other settings in SSA, data on PDR is scarce and geographically scattered(29–36). This situation may threaten the HIV control program, including the implementation of the test and treat policy, where newly diagnosed HIV patients are put on a lifelong ART. Elsewhere, evidence show that expansion of ART program appear to coincide with increase in PDR rates(37). Hence, a comprehensive analysis of the burden and trends of PDR for the period of ART rollout is of paramount importance. As for the current ART regimen recommendation in most LMICs, there is high risk of some patients remaining on dolutegravir monotherapy. Despite of high genetic barrier to developing resistance of dolutegravir, resistance to dolutegravir monotherapy has been reported in some studies in western countries(38–40), where this drug has been in use for some-time. Similarly, it is essential to monitor the rate of emergence and spread of PDR to protease inhibitors (PI). This is 6 because, under the current recommendation, HIV infected infants are started by default on PI based ART regimen. Thus, it’s equally essential to monitor the trend of PI associated HIV drug resistant mutations, for effective implementation of the Prevention of Mother to Child Transmissions (PMTCT) programme program(41–43). 1.2 LITERATURE REVIEW 1.2.1 Introductory Information The purpose of this literature review was to find out all published information related to the PDR and gaining insight on the rates of emergence and spread of HIV-1 drug resistant strains- among ART-naïve population, with the focus on SSA. The review was to establish the gap in knowledge on this topic, and assist focusing the research direction, to be able to generate useful scientific information that may substantially contribute to the existing body of knowledge on the topic among the scientific community. Furthermore, the literature review was aimed at familiarizing with various methodological approaches that has been employed to study the emergence of PDR among ART-naïve HIV infected populations, with the aim of selecting and employing the most appropriate method(s) for the current research study. To archive the literature review, a systematic organized search notes was developed in a literature review matrix, as an invaluable writing resource for literature review chapter. The literature review matrix includes information about the author, year of publication, sampling year, country, features of study population, sample size, proportions (time-point prevalence) of HIV-1 drug resistance mutations by ART class and HIV-1 subtypes. Eligible studies had to report data about prevalence of pre-antiretroviral(s) therapy in adults (> 15years) with no prior history of ART exposure; quantitative studies involving genotyping of the reverse transcriptase with or without protease region of the pol gene and the intergrase gene. 7 Furthermore, in this chapter, the prevalence of PDR among newly diagnosed HIV patients in low- and middle-income countries (LMIC) is described, with a particular emphasis on Tanzania. We do so by describing the research gap, which is pertaining to the sparsity of PDR data in Tanzania as in other SSA. Such information is essential not only for any individual appropriate HIV management, but also for the national control program to be able to evaluate the control intervention(s), as HIV-1 drug-resistance may affect the control efforts, for both reducing HIV transmission, morbidity and mortality attributed to HIV(44,45) The following section, include description of the research gaps, also the keys concepts and issues are explained, these includes TDR and PDR; Minority HIV-drug resistance mutations; Recent HIV infections with pre-ART drug resistant mutant virus, and dynamics of HIV-1 subtypes and circulating strains in Tanzania. The concept map is described in the next section, it depicts the conceptual relationship between different exposures and PDR status among HIV infected individuals. Thereafter, in the following sections, the various methods used to study the PDR are discussed. The overall focus of the study is on ART naïve PLHIV. 1.2.2 Research Gap The ART programme in Tanzania has completed 17 years(35), a period in which the national guidelines for ART initiation have undergone periodic modifications based on the WHO recommendations. Likewise, there has been nearly 200 folds increase in the number of PLHIV on ART to date, from about 5000 PLHIV that were on ART at the beginning on the free ART program in 2005 in Tanzania(1). The rapid expansion of ART coverage may coincide with changes in the prevalence of HIV drug resistance among HIV infected ART-naïve population(37). Also, ART regimens may impact differently on the prevalence rates of HIV drug resistance among PLHIV(46). This is mainly because different ART regimens differ in their genetic barrier to developing resistance, pharmacokinetic-profile, and tolerability(47). Even so, PDR data in Tanzania is scarce, mainly based in urban areas and come from studies with only small numbers of participants being evaluated in each study(29,30,34–36,48–52) Therefore, it is difficult to generalize the trends in prevalence of PDR in individual demographic and population groups. This situation could limit the effectiveness of HIV prevention approaches, such as those limiting onwards HIV transmission. This thesis aimed to improve availability of local evidence and support to adapt preventive measures and effective response to the threats of HIV drug resistance. 8 Additionally, as dolutegravir rollout continues in LMIC, information from this thesis contributes to evidence demonstrating the efficacy of dolutegravir alongside a compromised NRTI backbone under programmatic conditions in resources limited setting. Information generated in this study provides additional knowledge on the change in prevalence of PDR for over a full decade of ART rollout, in a representative African rural and semi urban setting, from the era of roll-out to the scaling up of ART in many places of Africa. Data generated are of great importance for the Tanzanian National AIDS Control Program, as well as for the region. The study contributes to the essential data needs for mitigating transmitted HIV-1 drug resistance in resources limited setting. 1.2.3 Key concepts 1.2.3.1 Transmitted and pre-treatment HIV-1 drug resistance HIV transmitted drug-resistance (TDR), or primary HIV drug-resistance, refers to drug resistance in HIV infected persons with no prior history of ART exposure, mainly acquired during initial infection with HIV drug-resistant variant(s), or “during subsequent re-infection with a resistant virus, referred to as superinfection”(11). Whereas pretreatment HIV drug-resistance (PDR) or pre- antiretroviral therapy is drug-resistance in HIV infected person prior ART initiation, which can either be a result of transmitted drug-resistant variant(s) or acquired during previous ART exposure, such as in women given ART to prevent mother-to-child transmission of HIV. PDR can arise either through transmission of drug-resistant HIV-1 variants from a person with acquired drug resistance, or by transmission of primary drug-resistant HIV from another ART naïve person(11). PDR may compromise the effectiveness of the initial ART, as evidenced in previous systematic reviews, that the odds of virological failure is higher in patients with PDR mutant variants compared to their counterpart infected with wild-type virus only [45]. Likewise, maternal PDR may diminish efficacy of the regimen to prevent mother-to-child transmission (PMTCT) and future treatment options. Thus, information of the levels and trends of PDR is essential for monitoring of HIV control programs, such as PMCTC. Pre-ART HIV drug-resistant variant(s) may revert to wild-type virus, or be overgrown by wild type variants, as such may exist at low levels(53–55). Hence PDR studies not considering patient populations with early HIV infection may underestimate the levels of mutations. Similarly, some HIV drug resistance associated mutations are known to confer viral fitness equally or more than the wild HIV variant(s). For example, the K103N mutations have been shown to persist for years even without drug pressure, which is not surprising that K103N is commonly detected in recently infected individuals(23). This may partly explain the majority of NNRTI transmitted mutations, observed among ART-naïve patients infected with HIV drug resistant variants. 9 Until recent, PDR was mainly driven by NNRTI associated HIV drug resistant mutations. In many setting in LMICs prevalence of PDR to NNRTIs exceeds 10%(56), which underscore the importance of WHO recommendation to replace NNRTI with a much more potent-the intergrase inhibitors, i.e., Dolutegravir-based ART regimen. Notably, NRTI associated PDR is reportedly rising, but in many setting to a level of about 5%. Similarly, PI associated PDR are reassuring low in LMICs. PI are characterized by relatively high genetic barrier to developing resistance, also until recent the usage of PI was relatively still very low, thus with recent increasing patients switched or started on PI-based regimen, there is need to systematically evaluate emergence and spread of mutations conferring resistance to PI. 1.2.3.2 HIV-1 drug-resistant minority variants HIV-1 Minority variants are virus that exist in very low proportional within the viral population, that may go undetected when using standard viral population sequencing methods(57). For example, in a detection methods comparison study, HIV-1 drug-resistance prevalence rates of 8% and 31.3% were detected among recently infected HIV-1 patients using standard population-based Sanger sequencing and Next-Generation sequencing (NGS) respectively(18). Similarly, a pooled analysis of 10 studies involving ART-naïve patients with no detectable HIV-1 drug-resistance mutations using population based-Sanger Sequencing, showed that 14% harbored NRTI/NNRTI mutations by ultra-sensitive assay(18). Clinical significance of minority HIV-1 variants is still an active area for investigation. However, some studies have shown that minority HIV drug resistant mutations prior to ART initiation are associated with more than two-folds increase in the risk of virologic failure [(55,58–62). For example, a retrospective samples analysis of the anonymized population in Uganda, reported 64% of failing patients without detectable HIV-1 drug resistance associated mutations as determined by population-based Sanger sequencing, harbored minority HIV drug resistant variants. Which may explain the continued virology failure observed in that population(63). Likewise, Rupérez et al., while studying a population of pregnant women in Mozambique, observed that all emerged drug resistance associated mutations during pregnancy were minority mutants, underscoring the importance of minority HIV-1 variants(64). Nevertheless, such virologic/clinical effects of minority HIV-1 variants have not been apparent in some studies(65). Minority HIV-1 drug resistant variants may go undetected with standard population-based genetic analysis methods. However, may emerge especially under drug pressure. Pimenta et al, while studying a population of HIV-1 infected ART-naive pregnant women in Sao Paulo, Brazil, did not find any HIV drug resistant mutations (66). A similar, recent study in Uganda did not detect HIV drug resistant mutations among ART-naive Ugandan women initiating Option B+ (67). The levels 10 of pretreatment HIV drug resistant mutations in above studies, may have been underestimated. Firstly, the analytical technique employed detect the majority viral population, at about 20% of the virus population(18). Whereas HIV drug resistant mutations may exist at low levels, particularly in the absence of drug pressure. A similar study in Mozambique, found that 10% of pretreatment HIV drug resistant mutations among pregnant women using NGS, which is the genetic analysis method able to detects minority HIV drug resistant mutations to as lower threshold as 1% of the viral population. In this study, all the emerged HIV drug resistant mutations during pregnancy were minority variants, which underscore the importance of minority HIV drug resistant mutations (64). 1.2.3.3 Recent HIV infections testing assays and their limitations. Several assays and testing strategies based on HIV antigens, or HIV antibodies and HIV viral RNA have been developed over the years to distinguish recent from chronic HIV infections. The advantage of these assays is that they can be carried out retrospectively on stored blood samples from cross-sectional studies, and they are cheaper and simpler to perform than following cohort studies. However, there are several limitations of the assays that currently limit their widespread applicability, the most important of which relate to the misclassification of the tests- resulting in false recent HIV infection, estimation of the window period, which varies substantially by HIV subtype and host population and low reproducibility. To overcome these misclassification of tests, new HIV-1 recent infection testing algorithm for HIV-1 Limiting Antigen (LAg) Avidity test have included HIV viral load test, and antiretroviral test results(68). 1.2.3.4 Recent HIV infection and drug resistance Recent HIV infection (RHI) is considered the phase of ≤6 months after infection, during which anti-HIV antibodies become detectable(69). People with RHI may be more likely to transmit HIV to others due to relatively high HIV viral load (HVL) in blood and genital secretions, absence, or few HIV specific blocking antibodies, fewer viral quasispecies diversity and possibly high transmissible, and which may possibly contain HIV-1 drug resistant variants(70). Data on PDR among recently HIV infected ART-naive patients from routine surveillance is crucial to inform interventions to control the HIV epidemic. However, prevalence estimates of HIV-1 drug resistance among recently HIV infected ART-naive patients in Tanzania has not yet been described. Previous studies show that HIV resistant variants are significantly higher in recently HIV infected population than in those with chronic HIV infections(70). For example, PDR prevalence of 20.9% vs 8.6 % was documented in recently HIV infected population and chronic HIV infection respectively(71). Also, a study in North-West India reported a moderate to high prevalence of PDR at 16% in patients with RHI, in setting where earlier studies have shown lower level of PDR prevalence(72). Likewise, a study in Rio de Janeiro, Brazil documented TDR 11 prevalence rate of 16.3% among acutely/recently HIV-infected individuals, which was higher than those found in chronically HIV-infected individuals in Brazil(73). Similarly, relatively higher prevalence of TDR was reported in a population-based household survey, that involve adult patients with RHI in 2012 in Kenya(74). Hypothetically, higher levels of PDR may be observed when targeting the recent HIV infected population, which is key consideration while estimating the burden of PDR. 1.2.3.5 HIV-1 subtypes and circulating strains in Tanzania The distribution and dynamics of HIV-1 viral subtype, matched to HIV drug resistant mutations or changing ART regimens over time has not been described in Tanzania, and in many other SSA countries. Previous studies conducted in Tanzania at the beginning of rollout of free ART in Tanzania have reported subtype A1 being the most prevalent, followed by subtype C(32,34,35) and D, CRF-CD and URF have been reported in the most recent studies(32). Nonetheless, these studies were only conducted in urban settings and had only small numbers of participants been evaluated. Therefore it is difficult to generalise the trends of HIV variants-linked to HIV drug resistance mutations or changing ART regimens over time, in individual demographic and population groups. Likewise, previous studies have evidently demonstrated that certain HIV subtype are more prone to developing mutations. For instance, HIV subtype C is known to be more prone to develop K65R mutation, which confer intermediate to high level resistance to Tenofovir. Thus, it is essential to monitor the emergence and spread of the mutations such as K65R. Thus, molecular epidemiological studies of HIV subtypes linked to long-term clinical and epidemiological data from under-sampled areas such as rural Tanzania, are key component in assessing and addressing the threat of HIV drug resistance. 1.2.3.6 Pre-treatment HIV drug resistance to the dolutegravir co-administered Nucleoside reverse transcriptase inhibitors The current new first line treatment recommended by the WHO, the integrase inhibitor dolutegravir in a fixed dose combination with tenofovir/lamivudine (TLD). Dolutegravir is known for its high and rapid viral suppression and a high genetic barrier to developing resistance. As such, TLD is predicted to curb the number of new HIV infections and the spread of resistance – leading the way to control of the HIV pandemic. Most evidence on efficacy of dolutergravir-based regimens comes from randomized controlled trials conducted in high-income countries, with a low background level of HIV-drug resistance. In 12 resource-limited settings, with limited or absent viral load and HIV drug resistance testing, long delays in the management of uncontrolled viraemia, could facilitate emergence of resistance to the co-administered NRTIs and possibly dolutegravir. Also, pre-existing HIV resistances to drugs used in combination might lead to dolutergravir functional monotherapy and potential for emergence of dolutergravir resistance(75). European cohorts have shown that, virological efficacy of dolutergravir monotherapy associates with virologic failures and risk of dolutegravir resistance(76). Analysis from the NADIA and DAWNING trial shows that no evident positive effect of inactive NRTI backbone used in combination with dolutegravir; Although observed emerging intermediate -or -high level dolutegravir resistance within the 48 months follow-up raises concern(77,78). Certainly, surveillance of pre-existing resistance to the co-administered NRTIs remains particularly important. 1.2.3.7 Trends of HIV-1 pretreatment HIV drug-resistant mutations The wider coverage of ART may increase the rates of disorderly use of ART, resulting into increased rates of acquired HIV-DRM, which may in turn be transmitted to individuals who have never been exposed to ART(37). Thus, analysis of time-trend in PDR is essential to monitor the ART regimen and evaluate the effectiveness of the program. Available data in SSA shows increasing trend of PDR among general adult population, which is mainly driven by NNRTI associated HIV drug resistant mutations (25,26,79). Also, there has been noticeable increase in NRTIs, particularly the Tenofovir associated HIV drug-resistance mutation(26). e.g., a study conducted in two clinics in Nairobi, Kenya in the year 2006 and 2014 respectively, reported a nearly tripled number of cases of K65R mutations(80). The reverse transcriptase (RT) mutation K65R results in a four-fold decrease in TDF susceptibility and is selected by TDF, didanosine (ddI), d4T and Abacavir (ABC)(81). A pooled analysis of studies on PDR in South Africa shows a relatively stable prevalence of PDR at <5% until 2009, and afterward increased to 11.9% in 2015(26). Similarly, a study involving seven West African and 2 Asian countries, where the prevalence of pre-ART drug-resistance ranged from 9.6% -24.6%(82). On the contrary, data from European HIV cohorts shows a decline or rather stable prevalence of PDR, at <7% (21–23)Examples of such cohorts are United Kingdom (UK) national representative PDR database which show stable or declining prevalence rates of PDR(83,84). Similar trend has been documented by Sallam, et al while studying a population in Iceland between 1996-2012(22), and Machnowska et al, in German with a PDR prevalence declining between 1996-2017(85). The declining prevalence of PDR in recent studies among western population, may be explained by the routinely virologic monitoring, including HIV drug 13 resistance testing and increasingly usage of ART regimen which have higher genetic barriers, e.g., intergrases. Most European HIV cohorts are characterized with increased genetic diversity of HIV variants, where HIV subtype other the traditionally mostly prevalent, subtype B are becoming increasingly(84,86). 1.2.4 Conceptual framework and concept maps Figure 1.3 Concept map of pre-Antiretroviral Therapy HIV-1 drug resistance and associated factors. The concept map depicts the conceptual relationship between different exposures and PDR status among HIV infected patients. The concept map show that different demographic factors may be associated with risk of being infected with a drug-resistance HIV virus, e.g. young age. Likewise, year of HIV infection may be associated with the coverage of ART, whereby increased ART access 1. HIV-1drug resistance associated mutations 2. Demographics: Age sex Education Marital status Occupation Geographical background HIV-1 subtype 5.Virological failure 3. HIV-1 infected individuals: Type of ART regimen, Year of HIV infection, Clinical characterist ics e.g. CD4 count level, Pre-ART resistance. 6. Persistent HIV transmission 4.Treatment failure- Affecting HIV Care & Treatment outcome. 14 may increase use at uncontrolled level, thus increase circulation of resistant virus, which may infect an individual who have never been exposed to ART. The concepts map shows the risk of transmission of resistant virus. 15 1.2.5 Methods for studies of HIV drug resistance prevalence Temporal analysis of HIV drug resistance associated mutation (HIV-DRM) among newly diagnosed HIV patients, enables the evaluation of HIV diversity overtime, including emergence of new HIV variants, e.g., the circulating recombinant forms (CRFs)(87), understanding the rate of change of burden of HIV drug resistant mutations and emergence of new HIV drug resistant mutations (88). Previous observational studies on molecular dynamics of pre-ART HIV-DRM have employed a variety of methods, including prospective observational methodological approach of newly diagnosed HIV infected patients. On the other hand, retrospective study methods, have been used to generate information by analyzing bio-banked specimen(89), and or analysis of data generated during routine treatment and care(90–92). Methods for assessing prevalence and incidence of HIV-DRM are mainly quantitative. Quantitative prospective and retrospective methodological approaches have been used to determine the prevalence and incidence of HIV-DRM over-time. Machnowska et al, and Siemieniu et al, employed prospective and retrospective quantitative methods, respectively(23,93). In a prospective cohort study, Machnowska et al, studied the prevalence of transmitted HIV-DRM considering the changing landscape of HIV-1 subtypes in the German HIV-1 Seroconverter Study Cohort between 1996 and 2017, and assessed the proportion of study patients potentially at risk of ART-failure due to these transmitted HIV drug resistant mutations. Elsewhere, in a retrospective analysis Siemieniu et al., documented an increase in non-subtype B, with notable increase in subtype C and CRFS, in southern Alberta, Canada. However, the major limitation of such retrospective studies, is that there is frequently an absence of data on potential confounding factors as the data was recorded in the past. Similarly, it may be difficult to identify an appropriate exposed cohort and an appropriate comparison group. Data from prospective studies can inform control programs on the rate of emergence of HIV-DRM in programmatic interventions, such as Option B+ which represent a challenge to PMTCT, and emergence of new circulating HIV strains. Prospective methodological approaches allow setting inclusion criteria and may enable to determine if HIV drug resistant mutations are lost during follow-up. Prospective longitudinal studies are expensive and may take long time to accomplish. Conversely, retrospective longitudinal analysis of bio-banked specimen or previously collected data, may provide invaluable information, within short period and at a low cost. 16 1.3 AIM OF THE PhD WORK The overall aim of this thesis was to determine the trend and patterns of pre-Antiretroviral therapy HIV-1 drug resistance in Kilombero and Ulanga Antiretroviral Cohort South-western-Tanzania. 1.4 OUTLINES OF THIS PhD THESIS As an introduction to the thesis, the first two chapters; Chapter 1 include a review of data on PDR, with focus on SSA, that were available before the start of the PhD research (before 2020); and Chapter 2 focus of the study methodology. In chapter 3, A systematic review and meta-analysis was undertaken to assess the trends in PDR associated mutations in Eastern African countries, during a five-year period (2017-2022). In chapter 4, A cross-sectional study was undertaken to determining the proportion of recent HIV- 1 infection, its predictors, and PDR in recently HIV-1 infected persons in the Kilombero and Ulanga antiretroviral cohort, in rural Tanzania. In chapter 5, A cross-sectional study was undertaken to measure the change in prevalence of PDR to NRTIs in the KIULARCO during the 15 years after ART rollout in Tanzania. In chapter 6, A cross-sectional study was undertaken to assess the virological-outcomes at 12 months, and its predictors after initiating DTG-based ART, and prevalence of PDR among patients failing on DTG-based ART. The final chapter 7, is a summary and general discussion of the main research findings of this thesis, followed by some concluding remarks. Appendices are included after the references. Appendix 1 contains a copy of the ethical approval form for conducting my PhD study, Appendix 2 contains the publication (paper I) in the Journal of Antimicrobial Chemotherapy, Appendix 3 contains the publication (paper IV) in the Journal of Open Forum Infectious Diseases. Appendix 4 contains the publication (paper II) under review in the Journal of Public Health. 17 CHAPTER 2 2.0 Methodology 2.1 Research settings The study consists of two parts; Firstly, a systematic review and meta-analysis of primary studies reporting prevalence of pretreatment HIV-1 drug resistance (PDR) in antiretroviral therapy-naïve adults in seven Eastern Africa: Ethiopia, Kenya, Malawi, Rwanda, Mozambique, Tanzania, and Uganda (paper I). Secondly, the studies in paper (II-IV) were embedded in the on-going prospective clinic-based observational Kilombero and Ulanga antiretrovirals cohort (KIULARCO) study of Peole living with HIV (PLHIV) in rural Tanzania. KIULARCO includes participants from two districts, namely, Kilombero and Ulanga in Morogoro region, South-Western Tanzania (Figure 4). The longitudinal follow-up of the cohort is conducted within the Tanzanian National AIDS Control Care Programme (NACP) framework. KIULARCO has been described in detail elsewhere(94– 99). Briefly, this study was carried out within the Chronic Disease Clinic of Ifakara (CDCI), the HIV Care and Treatment Centre (CTC) of the St. Francis Referral Hospital in Ifakara, Southwestern Tanzania. CDCI is a joint project of the hospital, with the Ifakara Health Institute (IHI), the St. Francis Referral Hospital, the Swiss Tropical and Public Health Institute and the University Hospital Basel, Switzerland. CDCI was among the first rural CTCs in Tanzania started in 2005 to provide HIV care and free ART within the NACP framework. CDCI runs a patient cohort, the KIULARCO described above. For more detailed information about patients, ethical permits, and statistical analysis, see respective papers. 18 Figure 2.1. Study area for paper II-IV; Kilombero and Ulanga districts in Morogoro region, South-Western Tanzania. 2.2 Study population In paper I, A Systematic Review and Meta-Analysis of eligible observational studies published between 1 January 2017 and 30 April 2022 about prevalence of PDR among ART-naive PLHIV (>15 years old)(56). In paper II- IV, studies were be embedded in the KIULARCO of the IHI in Tanzania. KIULARCO is one of the largest long-term patients’ cohorts in the East African region, started in 2005(96)(Figure 1). KIULARCO offers a clinical database and a biobank of yearly collected plasma samples collected during routine visits of patients in the past 20 years (have been stored in 19 -80°C freezers at the IHI laboratory in Ifakara, Tanzania), from more than 12,000 PLHIV within a population of about 600,00 inhabitants and an estimated 30,000 patients living with HIV/AIDS in the area (Figure 2). The longitudinal follow-up of the cohort is conducted within the Tanzanian National AIDS Control Care Programme (NACP) framework(96). Newly HIV-1-diagnosed adult individuals (aged ≥ 15 years) prior to ART initiation, who are enrolled in the KIULARCO were included. Study population detailed descriptions is provided in respective papers. Figure 2.2. Patients’ enrolment into the Kilombero and Ulanga Antiretroviral Cohort (KIULARCO) by year. 2.3 Sample size estimation The estimated sample size (In paper III) was calculated based on the expected prevalence of pre- treatment HIV-1 drug resistance of 10%, with prevalence between 3.3% and 5% based on existing published data. Sampling period was, the first five-year period (2005-2009) of the ART program in Tanzania, in the most recent period (2019-2022), in each batch 130 randomly selected samples was included in the study. With this sample, the assumption was to be able to detect the prevalence ≥ 7 %, one sided 5% level of significance, with the power of 80% (Table 1). An addition of 20% for sequencing failure rate to ensure minimal sample size for power is met. 20 Table 1. Sample size calculations Established Prevalence Expected Prevalence Statistical Power Calc. Sample Size* 3.3%** 7% 80% 310 90% 389 7.5% 80% 252 90% 315 10% 80% 122 90% 149 5%*** 7% 80% 1168 90% 1519 7.5% 80% 784 90% 1013 10% 80% 239 90% 301 *the caculated sample sizes are calculated with a one-sided test. **according to the last prevalence study in the KIULARCO cohort in 2009 [14] ***reported transmitted HIVDR prevalence across different sites in Africa [7] In paper IV, sample size estimation was based on the assuming the rate of VF at 10%, a 95% confidence and 80% statistical power, sample size is 230 patients. Adding 20% sequencing failure rate minimal sample size is 276, rounded-up to 280 total sample size is required. 2.4 Study procedures In paper I, a systematic review and meta-analysis of pre-treatment HIV-drug resistance (PDR) in antiretroviral therapy-naïve I reviewed all relevant articles, the supervisors independently reviewed 20% of randomly selected articles. Standardized data abstraction form was used to systematically collect data from scientific published articles. Core set of information retrieved from the articles, included bibliographic details, study design, inclusion and exclusion criteria, country and their main demographic, ART exposure history, the year(s) of sample collection, study type, study population, proportion of participants gender, and method for determining pre- treatment HIV-drug resistance (PDR), and reported prevalence of PDR. The systematic review protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) on 18/02/2022 (registration number CRD42022304207). 21 According to the KIULARCO protocol, patients who signed informed consent have blood samples drawn at routine clinic visits, prior to ART initiation and every 12 months thereafter. Plasma is cryopreserved at -800C for research purposes. In brief, data on demographics, clinical presentation, therapy outcome and adherence of ART and treatment outcome are being prospectively captured within an electronic database over the years. In paper II-IV, blood plasma samples of eligible patients were identified from the KIULARCO biobank, unfrozen and tested as detailed in respective papers. 2.5 Laboratory analysis 2.5.1 RNA Extraction Briefly, RNA was extracted from 150 μl of plasma using the PureLink™ Viral RNA/DNA Mini Kit (Invitrogen™, Thermo Fisher Scientific, USA), according to the manufacturer’s protocol(100). 2.5.2 Reverse Transcription and PCR RNA extracts were retrotranscribed and amplified using the HIV-1 Genotyping Kit Amplification Module (Applied Biosystems™, Thermo Fisher Scientific, USA), according to the manufacturer’s protocol(101). The PCR products were visualized on a gel visualization system (To confirm that the 2892 bp length of the pol gene) after being run in a 1% agarose gel at 130 V. 2.5.3 HIV Genotypic resistance testing In Paper II- IV, Sequencing was conducted to determine the presence of HIV drug resistance- associated mutations in the pol gene. Figure 2.3 Schematic presentation of HIV-1 viral genome Sequencing was performed using a validated in-house PCR protocol to determine the HIV-1 drug resistance-associated mutations for Reverse Transcriptase (RT), Protease (PR) and Integrase Inhibitor (INSTI) – mutations(42,95,99,102) A direct sequencing reaction was done using 6 overlapping primers (Table 1). 22 Table 2. Sequencing primers information # Primer Name Sequence §1 F-2150 AGCCAACAGCCCCACCAG 2 F-2232 GGAACTGTATCCTTTAGCTTCCC 3 F-2610 GTTAAACAATGGCCATTGACAGAAGAAA 4 F-3007 GGAAAGGATCACCAGCAATATTCCA 5 R-2605 GGGCCATCCATTCCTGG 6 R-3010 CCATCCCTGTGGAAGCACATTG 7 R-3541 TTCTGTATTTCTGCTATTAAGTCTTTTGATGGGTCA 2.6 Bioinformatics analysis 2.6.1 Sequence reads assembly, analysis, and annotation In Paper II-IV sequence reads assembly and analysis was performed using BioEdit 7.2- program(103), each sequence was manually inspected to ensure sequence quality, and finally annotated according to the Stanford HIV Drug Resistance Database, current version (https://hivdb.stanford.edu/)(104) 2.6.2 Drug Resistance analysis The assessment of HIV drug resistance associated mutations were interpreted according to the Stanford University’s HIV Drug Resistance Database (HIVdb) Program version 9.2 (http://hivdb.stanford.edu). HIV drug resistant mutations conferring low, intermediate, or high- level resistance were considered. The reported protease and reverse transcriptase sequences are deposited in GenBank accession numbers are provided in respective papers. 2.7 Ethical approval This study received ethics approval from the University of the Witwatersrand- Human Research (Medical) Ethics Committee. Ethical approval for the KIULARCO cohort has been obtained from https://hivdb.stanford.edu/ http://hivdb.stanford.edu/ 23 Ifakara Health Institute Review Board IHI/IRB/No: 16 – 2006 and the National Health Research Committee of the National Institute of Medical Research of Tanzania (NIMR/HQ/R.8c/Vol.I/378). Written informed consent for KIULARCO is sought from all participants ≥18years, and for children and adolescents aged <18 years from caregivers. 24 PART 2 EMPIRICAL CHAPTERS 25 CHAPTER 3 A systematic review and meta-analysis to assess the levels of pretreatment HIV-1 drug resistance in antiretroviral therapy- naïve in Eastern Africa (2017-2022). Introduction This chapter describes the prevalence of pre-treatment HIV-1 drug resistance among ART-naïve adult general population in Eastern African countries, within a systematic review and meta- analysis, during a five-year period (2017-2022). Results are presented in research paper I, entitled “Pre-Treatment HIV-1 Drug Resistance in Antiretroviral Therapy-Naïve Adults in Eastern Africa: A Systematic Review and Meta-Analysis”, this has been published as an original article in the Journal of Antimicrobial Chemotherapy. 26 Pre-treatment HIV-1 drug resistance in antiretroviral therapy-naive adults in Eastern Africa: a systematic review and meta-analysis Alex J. Ntamatungiro 1,2*, Juliana Kagura2, Maja Weisser1,3,4 and Joel M. Francis 5 1Ifakara Health Institute, Ifakara, Tanzania; 2Division of Epidemiology and Biostatistics, University of the Witwatersrand, Johannesburg, South Africa; 3Division of Infectious Diseases and Hospital Epidemiology, University Hospital of Basel, Basel, Switzerland; 4Swiss Tropical and Public Health Institute, Basel, Switzerland; 5Department of Family Medicine and Primary Care, University of the Witwatersrand, Johannesburg, South Africa *Corresponding author. E-mail: antamatungiro@cartafrica.org Received 23 May 2022; accepted 12 September 2022 Background: Pre-treatment HIV drug resistance (PDR) may result in increased risk of virological failure and ac- quisition of new resistance mutations. With recently increasing ART coverage and periodic modifications of the guidelines for HIV treatment, there is a need for an updated systematic review to assess the levels of the PDR among adults newly initiating ART in Eastern Africa. Methods:We conducted a systematic search for studies published between 1 January 2017 and 30 April 2022 in the MEDLINE Complete and CINAHL Complete, searched simultaneously using EBSCOhost, and Web of Science. To determine the overall PDR prevalence estimates, we extracted data from eligible articles and analysed preva- lence estimates using Stata 14.2. Results: A total of 22 eligible observation studies were selected. The studies included a total of 5852 ART-naive people living with HIV. The overall pooled prevalence of PDR was 10.0% (95% CI: 7.9%–12.0%, I2=88.9%) and 9.4% (95% CI: 7.0%–11.9%, I2=90.4%) for NNRTIs, 2.6% (95% CI: 1.8%–3.4%, I2=69.2%) for NRTIs and 0.7% (95% CI: 0.3%–1.2%, I2=29.0%) for PIs. Nomajor integrase strand transfer inhibitors (INSTI)-related mutations were identified. Conclusions: We observed a moderate overall PDR prevalence among new ART initiators in this study. PDR to NNRTIs is more prevalent, underscoring the importance of the current WHO recommendation for replacement of NNRTIs by INSTIs. PDR to NRTIs was low but notable, which warrants continuous surveillance of pre-existing resistance to the dolutegravir co-administered NRTI in Eastern Africa. Introduction The ART programme in countries in Eastern Africa has been run- ning for more than 15 years. Over these years, the guidelines for HIV care and treatment including ART initiation have undergone periodic modifications based on the WHO recommendations.1 Also, there has been a remarkable increase in the number of peo- ple living with HIV (PLHIV) on ART to date. For example, globally the ART coverage increased from 24% in 2010 to 65% in 2018. Also, 73% of PLHIV in East and Southern Africa were on ART by the end of 2019.2 Of note, the scale-up of ART rollout may coin- cide with changes in the prevalence of pre-treatment HIV drug resistance (PDR).3 This is mainly due to the existing gaps in HIV prevention interventions, care and treatment cascade, as well as the potency of the antiretroviral regimen and patient factors associated with emergence and spread of HIV drug-resistance mutations, such as inadequate adherence to ART.4,5 Major improvements in antiretroviral medication in terms of safety and pill count have been achieved over the past 15 years of ART rollout worldwide and in low- and middle-income coun- tries (LMICs), e.g. the replacement of NRTIs with long-term tox- icity like stavudine and didanosine to more favourable NRTIs like tenofovir disoproxil fumarate. Until recently, most LMICs were using two NRTIs and one NNRTI as first-line treatment, according to the recommendations by the WHO. The preferred regimen was tenofovir disoproxil fumarate+ lamivudine or emtricitabine+efavirenz as a fixed-dose combination. Dolutegravir, one of the first developed integrase inhibitors for treatment of HIV infection became available to LMICs by the end of 2017, as the generic fixed-dose combination teno- fovir/lamivudine/dolutegravir (TLD). Since then, TLD has been rolled out in many countries in sub-Saharan Africa (SSA). Of note, ART regimens may impact differently on the prevalence rates of PDR. This is mainly because ART regimens differ in their © The Author(s) 2022. Published by Oxford University Press on behalf of British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com 1 of 11 J Antimicrob Chemother https://doi.org/10.1093/jac/dkac338 D o w n lo a d e d fro m h ttp s ://a c a d e m ic .o u p .c o m /ja c /a d v a n c e -a rtic le /d o i/1 0 .1 0 9 3 /ja c /d k a c 3 3 8 /6 7 5 9 8 0 4 b y g u e s t o n 1 8 O c to b e r 2 0 2 2 27 Background The Antiretroviral Therapy programme in countries in Eastern Africa: has been running for more than 15 years. Over these years, the guidelines for HIV care and treatment including ART initiation have undergone periodic modifications based on the WHO recommendations(105). Also, there has been remarkable increase in the number of people living with HIV (PLHIV) on ART to date. For example, globally the ART coverage increased from 24% in 2010 to 65% in 2018. Also 73% of PLHIV East and Southern Africa were on ART by the end of 2019(106). The scale-up of ART rollout may coincide with changes in the prevalence of pre-treatment HIV drug- resistance (PDR)(46). This is mainly due to the existing gaps in HIV preventions interventions, care, and treatment cascade, also potency of antiretroviral regimen and patient factors associated with emergence and spread of HIV drug-resistance mutations, such as inadequate adherence to ART(107,108) Major improvement in antiretroviral medication in terms of safety and pill count have been achieved over the past 15 years of ART rollout worldwide and in low- and middle-income countries (LMIC) – e.g., replacement of NRTI with long-term toxicity like stavudine (d4T) and didanosine (ddI) to more favourable NRTIs like tenofovir disoproxil fumarate (TDF). Until recently, most LMIC were using 2 NRTI and 1 NNRTI as first line treatment according to the recommendations by the WHO. The preferred regimen was tenofovir disoproxil fumarate (TDF) + lamivudine (3TC) or emtricitabine (FTC) + efavirenz (EFV) as a fixed dose combination. Dolutegravir (DTG), one of the first developed integrase inhibitors for treatment of HIV infection became available to LMIC by the end of 2017, as the generic fixed dose combination tenofovir/lamivudine/dolutegravir (TLD). Since then, TLD has been rolled out in many countries in sub-Ssaharan Africa (SSA). It is worth noting that, ART regimens may impact differently on the prevalence rates of PDR. This is mainly because ART regimens differ in their genetic barrier to developing resistance, pharmacokinetic-profile, and tolerability PDR may compromise ART efficacy resulting in increased risk of virological failure, acquisition of new HIV drug resistant mutations and need of ART switch(109,110). Previous review study using available data in eastern Africa, include studies published until 2016(11). However, this was a period during which ART was not widely available, also ART represented in the current WHO guidelines, such as Tenofovir-based ART was still relatively new in many LMIC. Thus, with increasing ART coverage and periodic modifications of the guidelines for HIV care and treatment including ART initiation, there is the need for an updated systematic review to assess the levels of the PDR associated mutations among adults newly initiating ART in eastern Africa countries from 2017 until 2022. 28 Methods Study design We conducted a systematic search for studies published during January 1,2017 to April 30, 2022, about prevalence of PDR among ART-naïve PLHIV (>15 years). The data had to be collected in one or several of the countries of eastern Africa (Ethiopia, Kenya, Malawi, Rwanda, Mozambique, Tanzania, and Uganda). The systematic review protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) on 18/02/2022 (registration number CRD42022304207). Search Strategy and Selection Criteria An electronic search was conducted from three databases: MEDLINE Complete and CINAHL Complete searched simultaneously using EBSCOhost and Web of Science. The search advanced search approach was carried out according to PRISMA guidelines (Figure 1)(111), using the following search terms: (HIV OR ‘Human Immunodeficiency Viruses) AND (‘Pre- Antiretroviral Therapy’ OR ‘Pre-Antiretroviral OR ‘Pre-treatment’ OR ‘Pretreatment’ OR antiretroviral OR ART OR drug OR antiviral OR ARV OR ART OR ‘Nucleoside RT Inhibitor’ OR NRTI OR NNRTI OR INSTI OR ‘integrase inhibitors’ OR ‘protease inhibitors’ OR ‘non-nucleoside reverse transcriptase inhibitor’ OR ‘nucleoside reverse transcriptase inhibitor’) AND (‘HIV-1 Drug Resistance’ OR ‘transmitted HIV drug resistance’ OR ‘drug resistance’ OR ‘Resistance Mutations’) AND (Burundi OR Ethiopia OR Kenya OR Malawi OR Rwanda OR Mozambique OR Tanzania OR Uganda OR eastern Africa). We screened studies for eligibility based on title and abstract. Eligible studies had to report original data about the proportion of PDR in a sample of PLHIV (>15 years), with no prior history of ART exposure. Also, studies had be published in English. Publications without primary data such as review articles and letters to editors were excluded. We included quantitative studies involving HIV-1 drug resistance mutations for Reverse Transcriptase (RT), Protease (PR) and, integrase strand transfer inhibitors (INSTI)- mutations. Study Selection process and Extraction Standardized data abstraction form was used to systematically collect data from journal articles. AJN and JMF independently screened articles for selection. This involved an assessment of articles based on titles and abstract. In the case of insufficient information in the title and abstract, the full text of the specific article was retrieved and assessed. JK was involved in case the two reviewers failed to reach a consensus to include the article. We recorded the source, year of sampling, country origin, gender proportion, study period, number of study participants, prevalence of PDR to NRTI, NNRTI, PI, INSTI and HIV subtypes. 29 Data Analysis The statistical analysis was done using STATA 14.2. We assessed heterogeneity using I2 statistics, where I2> 50% heterogeneity was high. Studies quality assessment was done using JBI’s critical appraisal tools. (https://jbi.global/critical-appraisal-tools). A funnel plot and Egger’s test were applied to detect any publication bias. A p-value of <0.05 was considered statistically significant. Results Characteristics of Included studies We identified 889 potential research articles from three databases according to our search strategy. A total of 22 observational studies were selected, representing a total of 5852 PLHIV with no prior history of ART exposure (Figure 1), with available resistance data derived from RT and/or PR and/or Integrase sequences in seven eastern African countries; Ethiopia, Kenya, Malawi, Rwanda, Mozambique, Tanzania, and Uganda (Table 1). All the research articles included in this study involved heterosexual patients’ population, expect one research article that included men who have sex with men (MSM)(112). https://jbi.global/critical-appraisal-tools 30 FIGURE 1 | Flow diagram of search results and screening Identification of studies via databases Articles excluded after assessment of the title and/or abstract (n=418) 152 Non-HIV studies 266 Non-HIV PDR studies Articles excluded after full-text assessment (n =31) 3 PDR in Children 1 Systematic review and Meta regression analysis 1 Methods validation 1 proviral DNA 7 Countries outside eastern Africa 18 Others Studies included in the analysis. (n =22) Studies selected for full-text assessment (n =53) Studies identified through database searching (n=889) MEDLINE Complete (n=410) CINAHL Complete (n=105) Web of Science (n=374) 471 Studies screenend on basis of the title and Abstract 418 Duplicates Excluded 31 Table 1: The characteristics of included datasets with reverse transcriptase and/or protease sequences from ART-naïve adults. AUTHOR, PUBLICATION YEAR, REFERENCE Year of Sampling Country Origin Study Popula tion Females (%) Study Design A Point Prevalence Estimate (%) GRT- Method HIVDRM Identification- Method Any NNRTI NRTI PI HASSAN ET AL., (2019)(113) 2008 Kenya 50 72% Retrospectiv e hospital- based cohort study 24 6.0 8.0 12 NGS HIVdb TELELE ET AL., (2018)(114) 2009-2011 Ethiopia 461 58% Prospective hospital- based cohort study 3.9 1.5 2.0 0.4 PBSS HIVdb HASSAN ET AL., (2018)(112) 2005-2017 Kenya 97 0 % Prospective population- based cohort study 8.2 7.2 2.1 0.0 PBSS HIVdb CROWELL ET AL., (2020)(115) 2013- 2019 Tanzania 154 59.6% Prospective hospital- based cohort study 8.0 6 .0 3.0 1.0 PBSS HIVdb 2013-2019 Kenya 254 59.6% Prospective hospital- based cohort study 7.0 7.0 2.0 0.0 2013-2019 Uganda 291 59.6% Prospective hospital- based cohort study 14 10.0 5.0 1.0 BARABONA ET AL., (2019)(29) 2017 Tanzania 47 75.7% Cross- sectional hospital- based study 29.8 25.5 8.5 0.0 PBSS HIVdb RUDOVICK ET AL., (2018)(30) 2013-2015 Tanzania 235 75.7% Cross- sectional hospital- 4.7 3.0 1.7 0.0 PBSS HIVdb 32 based study VON BRAUN ET AL., (2018)(116) 2015 Uganda 152 62.8% Cross- sectional hospital- based study 5.9 5.3 3.3 0.0 PBSS HIVdb RUTSTEIN ET AL., (2019)(117) 2012–2014 Malawi 45 39% Prospective hospital- based cohort study 24.4 20 6.7 0.0 PBSS HIVdb SILVERMAN ET AL., (2017)(118) 2013–2014 Kenya 815 65.1% Cross- sectional hospital- based study 9.4 9 0.7 0.0 PBSS OLA MILNE ET AL., (2020)(119) 2006-2010 Kenya 97 100% Prospective hospital- based cohort study 6.2 . . . PBSS HIVdb ONYWERA ET AL., (2017)(74) 2012 Kenya 87 72% Cross- sectional population- based study 9.2 6.9 4.6 1.2 PBSS HIVdb and WHO 2009 Mutation list CARNIMEO ET AL., (2021)(120) 2017–2018 Mozambiq ue 196 58.7% Cross- sectional hospital- based survey 16.8 14.8 4.6 0.5 PBSS HIVdb 258 66.7% Cross- sectional hospital- based survey 26.4 25.6 4.3 0.4 PBSS KIROS ET AL., (2020)(121) 2018 Ethiopia 51 54.9% Cross- sectional hospital- based study 9.8 7.8 2.0 0.0 PBSS HIVdb and IAS- USA 2019 Mutation list WATERA ET AL., (2021)(122) 2016 Uganda 335 61.5%* Cross- sectional survey 18.0 13.7 6.0 1.5 PBSS HIVbd 33 BECK ET AL., (2020)(123) 2006 Kenya 303 66.7% Cross- sectional hospital- based cohort analysis 2.6 . . . PBSS OLA 2010 Kenya 97 100% Cross- sectional hospital- based cohort analysis 6.2 . . . 2013–2014 Kenya 679 65.7% Cross- sectional hospital- based cohort analysis 4.4 . . . MCCLUSKEY ET AL., (2019)(124) 2005-2013. Uganda 738 69% Cross- sectional hospital- based cohort analysis 3.5 2.8 1.0 0.0 PBSS HIVbd NEUHANN ET AL, (2020)(125) 2014-2015 Malawi 197 61.9% Cross- sectional hospital- based cohort analysis 14.2 13.7 0.5 0.0 NGS HIVdb ZHOU ET AL., (2018)(126) 2008 Malawi 20 . Cross- sectional hospital- based cohort analysis . 9 2 0 NGS . ARIMIDE ET AL., (2018)(127) 2011–2013 Ethiopia 67 87% Cross- sectional hospital- based survey 6.0 6.0 0.0 0.0 PBSS HIVdb NDASHIMYE ET AL., (2018)(128) 2007-2011 Uganda 87 100% Retrospectiv e hospital- 8.7 8.7 . . PBSS HIVdb 34 HIVdb, HIV Drug Resistance Database; IAS-USA, International AIDS Society-USA; NGS, Next Generation Sequencing; PBSS-Population based Sanger Sequencing; data not analysed or not provided; OLA, Oligonucleotide Ligation Assay; HIVDRM, HIV drug resistance mutation; GRT, Genotypic Resistance Testing. based cohort study UMVILIGIHOZO ET AL., (2021)(129) 2005-2011 Rwanda 21 . Prospective hospital- based cohort study 19 19 . . PacBio RSII HIVdb SANGEDA ET AL., (2019)(52) 2010 Tanzania 18 . Prospective hospital- based cohort study 0.0 0.0 0.0 0.0 PBSS HIVdb 35 The Overall Prevalence of Pretreatment Drug Resistance-Associated Mutations The overall prevalence of PDR was 10.0% (95% CI: 7.9%–12.0%, I2 =88.9%) among 22 studies (Figure 2). Sub-analysis by country shows that the prevalence of PDR in Ethiopia 4.6% (95% CI: 2.4%–6.9%, I2 =12.0%, in 3 studies), in Kenya 7.0% (95% CI: 4.6%–9.3%, I2=79.6%, in 7 studies), in Malawi 17.6% (95% CI: 8.2%–27.1%, I2 =54.9%, in two studies), in Mozambique 21.6% (95% CI: 12.2%–30.9%, I2 =83.8%, in two studies), in Rwanda 19.0% (95% CI: 2.3%–35.8%, in one study), in Tanzania 10.8% (95% CI: 3.0%–18.6%, I2 =86.0%, in three studies) and in Uganda 10.0% (95% CI: 4.1%–16.0%, I2 =93.6%, in five studies). HIV-1 subtypes data is reported in 14(63.6%) of the selected studies; the presence of HIV-1 subtype in Ethiopia, Malawi and Mozambique are predominantly subtype C; in Kenya subtype A and D; in Tanzania subtype A, C and D, whereas in Uganda subtype A and D are prevailing. Publication bias, as examined by the funnel plot and Egger’s test, showed no publication bias in this analysis. 36 FIGURE 2 | Overall prevalence of PDR. Note: the red dotted line indicates the overall prevalence of PDR. 37 PDR of RT inhibitors The prevalence of mutations conferring resistance to NNRTI was 9.4% (95% CI: 7.0%–11.9%, I2=90.4%) in 21 of the included studies (Figure 3). NNRTI resistance reached highest level among different HIV-drugs categories. The prevalence of PDR to NRTI is 2.6% (95% CI: 1.8%–3.4%, I2=69.2%, in 18 studies (Figure 4). Nonetheless, prevalence of PDR to NRTI above 5% have been reported in several countries, including 8.5% in Tanzania(29); 5.2%-6% in Uganda(122) and 6.7% in Malawi(117). Also, population with a high risk of PDR to NRTI are young adults (2.9% in a study from Uganda)(130), and people injecting drugs (15.5% in a study from Kenya)(43). Supplementary Table S1. PDR of PIs Overall, PI resistance was re-assuring low, 0.7% (95% CI: 0.3%–1.2%, I2=29.0%) in 8 studies (Figure 5). High levels of PDR to PI were reported in studies detecting HIV drug resistant mutations using Next Generation Sequencing (NGS). For-example, PDR to PI of 12%and PDR to PI of 15% was documented by studies using NGS in in Kenya and Malawi respectively, whereas analysis of same samples in the later study by consensus sequencing no mutation associated with PI resistance was detected(113,126) PDR of INSTIs We identified four studies that examined HIV-1 INSTI associated resistance in patients prior to ART initiation, which found no participants had mutations associated with high levels of INSTI resistance. 38 FIGURE 3 | Prevalence of PDR to Non-nucleoside reverse transcriptase inhibitor 39 FIGURE 4 | Prevalence of PDR to nucleoside reverse transcriptase inhibitor 40 FIGURE 5 | Prevalence of PDR to protease inhibitor Discussion In this study, we investigated the overall prevalence of PDR in ART naïve adults living with HIV, in seven eastern African countries for over a period of five years. We observed an overall PDR prevalence of 10.0% (95% CI: 7.9%–12.0%); Whereas previous review study using available data in eastern Africa published until 2016, reported overall PDR prevalence estimates of 8.7%(11). Like our finding PDR was highest to the NNRTI drug class, particularly in more recent studies across the region, which underscores the importance