CADTH Health Technology Review

Risk of Sexual Transmission of HIV With Antiretroviral Therapy and Variable or Suppressed Viral Load

Technology Review

Authors: Sara Khangura, Anusree Subramonian, Melissa Severn

External Reviewers

This document was externally reviewed by content experts who have granted permission to be cited.

Dr. Marissa L Becker, MSc (Public Health), MD, FRCPC

Associate Professor

University of Manitoba

Dr. Sharon Walmsley, MSc, MD, FRCPC

Professor of Medicine, University of Toronto

University Health Network

Key Messages

• The absolute risk of HIV transmission through sexual activity, reported as incidence rates of observed seroconversions across person-years of follow-up, ranged widely across the included studies of sex partners living with HIV with variable (or unreported) viral loads.

• Two studies included sex partners living with HIV with suppressed viral loads of less than 200 copies/mL, reporting no phylogenetically linked seroconversions in either study.

• Most studies in this report included data from heterosexual serodiscordant sex partners and did not report follow-up viral loads of sex partners living with HIV.

• The included studies incorporated variable methods, investigated a broad variety of populations and exposures, and demonstrated a wide range of methodological strengths and limitations.

Context and Policy Issues

HIV has been a major public health issue since the recognition of AIDS, and the discovery of the virus in the early 1980s.¹ Globally, HIV continues to be 1 of the most common causes of death in low-income countries, though the rates are decreasing significantly.² WHO reports that there were 1.5 million new HIV infections worldwide in 2021.³ In Canada, it is estimated that there were 62,700 people living with HIV (PLWH) at the end of 2020 and 1,520 new infections occurred in 2020.⁴ The majority of those new infections are estimated to have occurred through sexual transmission.⁴

Current clinical guidelines recommend that antiretroviral therapy (ART) should be started in all PLWH, ideally within 7 days of diagnosis, if the patient is ready and there are no concomitant opportunistic infections.⁵ ART drugs act by inhibiting various levels of the viral life cycle, thereby suppressing the viral replication in human plasma.⁶ Several studies among serodiscordant couples (where 1 partner is living with HIV and the other partner is HIV negative) have shown that early initiation of ART can prevent sexual transmission of HIV.^7-9 This led to the development of the global campaign “Undetectable = Untransmittable,” which seeks to build scientific consensus behind growing evidence that PLWH who are adherent to treatment and maintain an undetectable viral load are unable to transmit the virus through sexual activity.¹⁰

Disclosure of HIV status by a PLWH to a sexual partner is legally required in Canada where there is a “realistic possibility” of HIV transmission. Nondisclosure of HIV status has rendered consent to sexual activity “not obtained in law” when the sex partner living with HIV (LWH) is found to have not been clear and transparent about their status.¹¹

A systematic review (SR) published in 2018 by the Public Health Agency of Canada (PHAC)¹² found that when the sex partner LWH takes ART and maintains a viral load of < 200 copies/mL (i.e., a suppressed viral load) on consecutive viral load testing, the risk of sexually transmitting HIV is “negligible.”¹² The authors also concluded that the risk of sexual transmission is “low” when the sex partner LWH takes ART and has variable levels of viral load. The findings of the review were based on studies among serodiscordant couples conducted from 2007 to 2017.¹² With the emergence of new evidence since then, especially in the men-who-have-sex-with-men (MSM) population,^13,14 an updated review of the evidence is warranted.

The purpose of this report is to identify evidence that has been published since the 2018 PHAC SR¹² regarding the risk of HIV transmission associated with sex when the sex partner LWH takes ART with variable levels of viral load as well as with a suppressed viral load.

Research Questions

What are the absolute risks of HIV transmission associated with sex when the following preventive measures are taken to reduce the risk of transmission between a PLWH and a sex partner who is HIV negative:

1. the sex partner living with HIV takes ART with variable levels of viral load?

2. the sex partner living with HIV takes ART and has a suppressed viral load of less than 200 copies/mL on consecutive measurements every 4 to 6 months?

Methods

Literature Search Methods

The literature search strategy used in this report was developed for a previous SR.¹² For the current report, search terms were added to the search strategy to expand the identification of articles relevant to the LGBTQ2S+ community.

For this update, published literature was identified by searching the following bibliographic databases: MEDLINE All via Ovid, Embase via Ovid, the Cochrane Central Register of Controlled Trials (CENTRAL) via Ovid, and Scopus. Duplicates were removed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were antiretroviral therapy, sex partners, and HIV. The following clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov and Health Canada’s Clinical Trials Database.

CADTH-developed search filters were applied to limit retrieval to health technology assessments, SRs, meta-analyses (MAs), indirect treatment comparisons, and any types of clinical trials or observational studies. Conference abstracts and preprints were excluded from the search results. Retrieval was not limited by language.

The update search was run on November 22, 2022, to capture any articles published or made available since the initial search dates.

Grey literature (literature that is not commercially published) was identified by searching sources listed in relevant sections of the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature reference.¹⁵ Google was used to search for additional internet-based materials.

Selection Criteria

Titles and abstracts for all citations were screened by 2 independent reviewers, with full-text articles retrieved for any citation identified as potentially relevant by either independent reviewer. The full text of each potentially relevant study was read by 1 reviewer and assessed for eligibility. The final selection of full-text articles was based on the inclusion criteria presented in Table 1.

Table 1: Selection Criteria

Exclusion Criteria

Articles were excluded if they did not meet the selection criteria outlined in Table 1, were duplicate publications, or were published before 2017.¹² SRs in which all relevant studies were captured in either the PHAC review,¹² or other more recent or comprehensive SRs, were excluded. Primary studies retrieved by the search were excluded if they were captured in 1 or more of the included SRs.

Synthesis Approach

Relevant data from eligible studies were abstracted into tables and organized by 2 reviewers, with the information used to inform the narrative summary of findings. The narrative summary of findings was structured and presented by the research questions posed in this report, with a focus on absolute risk of HIV transmission in serodiscordant sex partners for whom the partner LWH is treated with ART. The limitations of the included studies were also summarized, as observed by the reviewers and as reported by the study authors.

Critical Appraisal of Individual Studies

Each included study was critically appraised by 1 reviewer using either the AMSTAR 2 (A MeaSurement Tool to Assess Systematic Reviews 2)¹⁶ tool for SRs, or the Downs and Black checklist¹⁷ for randomized and nonrandomized studies. Summary scores were not calculated for the included studies; rather, the strengths and limitations of each included publication were described narratively.

Summary of Evidence

Quantity of Research Available

A total of 5,571citations were identified in the electronic database search. Following screening of titles and abstracts, 5,485 citations were excluded and 86 potentially relevant reports from the electronic search were retrieved for full-text review. One potentially relevant publication was retrieved from the grey literature search for full-text review. Of these potentially relevant articles, 72 publications were excluded for various reasons, while 15 studies met the inclusion criteria and were included in this report.^13,14,18-30 These comprised 1 overview of SRs and/or MAs,¹⁸ and 14 nonrandomized studies.^13,14,19-30 Appendix 1 presents the PRISMA³¹ flow chart of the study selection.

Additional references of potential interest are provided in Appendix 5.

Summary of Study Characteristics

The overview of reviews¹⁸ included 10 SRs and/or MAs — 2 of which were not described in the SR by PHAC.¹² Overview of reviews or “umbrella review” refers to evidence synthesis of existing systematic reviews.³² Results of the 2 relevant SRs included in the overview are summarized in the current report. Of the 14 primary studies,^13,14,19-30 only data describing characteristics and findings relevant to the current report were abstracted and summarized herein.

Additional details regarding the characteristics of the included publications are provided in Appendix 2.

Study Design

The overview of reviews was published in 2021, and sought SRs and MAs published between 2007 and 2019; there was no description of overlap between the studies included in the overview.¹⁸ The 14 primary studies were published between 2018 and 2022, all of which employed nonrandomized study designs.^13,14,19-30 Seven of the nonrandomized studies used a prospective cohort design^{13,14,19,21-23,25} and 7 used a retrospective method.^20,24,26-30

Country of Origin

The lead author of the included overview of the reviews was from Iran.¹⁸ Countries represented by the eligible primary studies included Brazil,^13,27 China,^{19,24,26,28-30} Ethiopia,²⁰ India,^22,23 Kenya,²⁵ Rwanda,²¹ and 2 international studies, 1 of which was conducted in 14 European countries,¹⁴ and the other which including participants from Australia, Brazil, and Thailand.¹³

Patient Populations

The overview sought SRs and MAs that described adult couples and/or sex partners who were HIV serodiscordant, including both couples and/or sex partners who were heterosexual as well as MSM.¹⁸ No other characteristics of the study participants described in the included SRs and MAs were reported.

The 14 primary studies also included couple were HIV serodiscordant and/or sex partners who were either heterosexual and/or MSM.^{13,14,19-21,23-30} None of the studies identified specified investigation of women who have sex with women. Whereas 11 of the studies included data for couples with only heterosexual serodiscordant (HSSD) sex partners,^19-21,23-30 2 reported data only for MSM sex partners.^13,14 One study did not specify whether participant couples were heterosexual or MSM.²² Of the heterosexual study cohorts, the sex composition of couples and/or sex partners who are serodiscordant by HIV status varied across 8 of the studies, with either the male or female sex partner LWH.^19,21,25-30 Two of the heterosexual study cohorts included only males LWH,^23,24 and 1 included only females LWH.²⁰ The ages of the study participants were not reported consistently across the included studies, with some studies reporting age only for sex partners LWH,^{13,19,26,28-30} others reporting age for all study participants,^{14,20-22,24,25,27} and others not reporting on age at all.^18,23

Study settings included local, hospital, government and/or university-based clinics;^{13,19,21,24,26} clinics specializing in the treatment of HIV, AIDS, and/or ART;^22,23,27 hospitals;²⁶ clinical research sites and facilities;^14,25 and public health institutions.²⁰ Three of the retrospective cohort studies reported the use of data from Guangxi, China, but did not provide detail describing the setting(s) from which participants were recruited.^28-30 Finally, the presence or extent of potential overlap between patient populations across these studies (given their proximity in setting and jurisdiction) was unclear.

Exposures and Comparators

All included studies reported data on serodiscordant sex partners in whom the sex partner LWH was treated with ART.^13,14,18-30 Types of ART regimens were not specified in most of the studies,^{13,14,18,20-23,25-30} although 2 studies did provide this information, describing lopinavir and ritonavir-based, raltegravir-based, efavirenz-based, and other (unspecified) regimens.^19,24 For sex partners LWH and already using ART at baseline, the duration of use was described in 4 studies.^14,21,24,25 Data describing ART use and adherence across follow-up were reported in 5 studies.^{13,14,20,23,25}

Six of the primary studies reported data describing viral load at baseline in sex partners LWH,^{13,14,19,23,25,27} with 5 of these further specifying subgroups of participants with lowered viral loads.^{13,14,19,25,27} Studies describing subgroups of participants with a low viral load characterized participants at thresholds of either 200 copies/mL^13,14 or 400 copies/mL,^19,25 with 1 study reporting genital viral load in female sex partners LWH at a threshold of 40 copies/mL.²⁷ Notably, 1 of the studies reporting viral load at a threshold of 400 copies/mL also included a definition of viral load at a threshold of 1,000 copies/mL, with a lack of clarity as to how these apparently variable thresholds were applied.²⁵ Of the studies reporting any data describing viral load, 2 provided data at both baseline and follow-up,^13,25 whereas 4 provided baseline data only.^14,19,23,27 Per the research questions and study eligibility criteria for this report, data from studies describing findings in study participants LWH who demonstrated viral loads of less than 200 copies/mL across study follow-up were used to answer the second research question; those who did not specify study participants LWH who demonstrated a viral load of less than 200 copies/mL, or did not specify viral load clearly or at all, were used to answer the first research question.

Pre-exposure prophylaxis (PrEP) is a preventive therapy taken by sex partners who are HIV negative that can reduce the risk of HIV transmission in serodiscordant couples,³³ and was featured in 4 of the studies included in this report.^13,14,24,25. One of the studies included sex partners who were HIV negative, all of whom initiated PrEP across study follow-up.²⁵ Two studies separated outcomes data for serodiscordant couples in whom the sex partner who was HIV negative received PrEP from those who did not.^13,24 Another study measured the use of PrEP among study participants and excluded outcomes data for serodiscordant couples in whom the participants who were HIV negative were receiving PrEP.¹⁴

Eight studies included a comparison group^{18,19,21,24,25,28-30} and 7 studies did not.^{13,14,20,22,23,26,27} Of the 8 studies describing a comparator, 5 bore relevance to the current report (i.e., use of ART versus no use of ART)^18,21,28-30 and 3 did not present comparative data of relevance (i.e., comparisons with exposures of no direct relevance to the research questions).^19,24,25 Relevant comparison groups included those in which the sex partner LWH was not receiving ART,^18,21 or was ART treatment-naive at baseline.^28-30 Irrelevant comparison groups included those in which the sex partners LWH were both receiving ART, but the partners who were HIV negative either did or did not have hepatitis C virus,¹⁹ or were or were not receiving PrEP.²⁴ Another study used a hypothetical comparison cohort.²⁵

Outcomes

All 15 studies included in this report described seroconversions observed in serodiscordant sex partners.^13,14,18-30 Seven studies summarized follow-up time using total person- or couple-years accrued,^{13,14,19,25,26,28,30} and 3 summarized follow-up time using the mean or median years observed per participant couple.^20,21,27 Five studies did not report a summary value describing observed follow-up time.^18,22-24,29 Exposure data of relevance to the research questions posed in this report included describing viral load in sex partners LWH across study follow-up, using consecutive measurement of viral load across a 3 to 12 month time frame, which was reported by 3 of the included studies.^13,14,25

Thirteen of the included studies described the absolute risk of HIV transmission as seroconversions observed across follow-up in person- or couple-years, reported using incidence rates expressed as number of seroconversions observed per 100 person- or couple-years.^{13,14,18-23,25-29} One study did not report incidence rates, describing the absolute number of seroconversions observed over a 24-week course of follow-up per participant couple.²⁴ Another study reported incidence rates across person-years of follow-up for serodiscordant couples among whom the sex partner LWH either was or was not receiving ART (i.e., combined results),³⁰ which were not relevant to the current report. However, this study did report numbers of seroconversions observed at both 6- and 12-month intervals of follow-up in sex partners LWH who were receiving ART (comparing these with seroconversions observed among those who were not receiving ART),³⁰ and these findings are included in this report.

Single-arm studies described the absolute risk of HIV transmission across serodiscordant couples in whom those with HIV all received ART,^{13,14,19,20,22,23,26,27} whereas comparative studies reported and compared data on the transmission risk across serodiscordant couples in whom those with HIV did or did not receive ART.^18,21,28-30

In addition, viral or phylogenetic linkage among observed seroconversions was assessed by 3 studies,^13,14,21 providing confirmatory information supporting the occurrence of HIV transmission between the sex partners under study (as opposed to another source of transmission).

Additional outcome measures bearing some relevance to this report included those describing relative risk of HIV transmission, often comparing groups with or without exposure to ART;^18,21,28-30 and sexually transmitted infections observed during follow-up.¹³ One study described severe adverse events associated with ART use,²⁵ but these data were not directly related to the research questions posed by this report.

Summary of Critical Appraisal

Overview of Systematic Reviews

The overview of SRs and MAs demonstrated both strengths and limitations,¹⁸ as assessed using the AMSTAR checklist,¹⁶ as well as additional items of particular relevance to overviews of reviews. Strengths included clearly reported eligibility criteria to judge the inclusion or exclusion of SRs, information on reasons for exclusion, sources of funding and conflict of interest statements for the overview study, and a published protocol, registered a priori with the PROSPERO database. Clarity in reporting is essential to ensure that readers can assess and judge the validity and reliability of the study’s findings. Methodological strengths included a comprehensive electronic literature search strategy, duplicate study selection, risk of bias assessments, and appropriate assessments of heterogeneity between the SRs, including a discussion of the impact of heterogeneity on the interpretation of the findings. Methodological rigour in SRs (including overviews of SRs) is critical to ensuring that confidence can be placed in the validity and reliability of the findings.

The overview of SRs and MAs study also demonstrated important limitations.¹⁸ As it concerned clarity in reporting, SRs excluded from the overview were not listed or described, and there were insufficient details describing the included SRs, making interpretation of the overview challenging. Furthermore, there was no mention of the risk of transitivity or publication bias, and the sources of funding for the included SRs were not reported. Methodological and other limitations included no mention of a search of grey literature, no description of an assessment of overlap between the included SRs, no description of duplicate data abstraction, and considerable heterogeneity identified among the included SRs. These missing details and methodological features make it difficult for the reader to assess the extent to which the findings of the overview are reliable and/or valid.

Primary Studies

Critical appraisal of the nonrandomized cohort studies summarized in this report focused on assessing the clarity of reporting, external and internal validity, and study power.¹⁷ All 14 studies demonstrated some strengths and limitations in either reporting or methodological rigour,^13,14,19-30 and these are summarized in the following.

Clarity of Reporting

Clarity and completeness in reporting of empirical data collection and analyses are critical for the reader to assess the extent to which a study was conducted rigorously, and whether the findings are valid and reliable. A description of the study objectives and main outcomes were included in all 14 primary study reports.^13,14,19-30 Details describing study participants and exposures were reported with sufficient detail in 6 studies,^{13,14,19,20,24,25} whereas 8 studies had relevant information missing on either 1 or more sex partner(s) of the serodiscordant couples, or with regard to the exposure of interest (i.e., the specific details of ART use, including regimens and/or adherence across study follow-up).^21-23,26-30 Study participant loss to follow-up (including characteristics of participants) was not clearly described by most studies;^14,20-28,30 though, 2 studies did report these data appropriately, including numbers and characteristics of those lost to follow-up and/or appropriate adjustments to account for follow-up losses.^13,29 The lack of a clear description of participant loss to follow-up limits the extent to which the internal validity of the study can be assessed, as it remains unclear whether participants who did not complete the study follow-up were or were not similar to those for whom data collection were complete. Similarly, all but 1 study²⁵ did not report information describing any adverse effects caused by ART,^{13,14,19-24,26-30} which, while not a particular focus of the current report, poses a limitation on the reader’s ability to assess the possible impact of this variable on the exposure of interest for these included studies.

Other descriptions of the exposure of interest (i.e., ART use across time, regimens, and/or adherence) were provided with variable levels of detail across the included studies.^13,14,19-30 A sufficient description of the use of ART is essential to interpreting the findings of the studies, as the numbers of participants, their use over time, and their adherence to ART are critical details needed to answer the research questions concerning transmission of HIV. While 7 studies provided some or clear information describing ART use, regimens, and/or adherence,^{13,14,19-21,24,25} 7 studies provided no or few details describing this exposure.^22,23,26-30

Other study features (including participant characteristics and study outcomes) were generally included in and/or reported clearly across most of the studies;^13,14,19-29 however, 1 study report was missing — or provided unclear descriptions of — critical information necessary to interpret the study findings (e.g., missing denominators and unclear HIV status of the partners who were not LWH), which limits the interpretation and utility of the study findings.³⁰

External Validity

Establishing external validity is important to ensuring that the findings from a study are generalizable to the broader population (i.e., outside of the study context). The representativeness of the study populations and the care they received was variable across the included studies.^13,14,20-30 Whereas 6 studies demonstrated some or clear evidence of external validity (e.g., recruitment of consecutive patients from clinics or health facilities used by most or all of the population of interest [i.e., serodiscordant sex partners]),^14,19-22,28 the representativeness of the study populations and/or settings was unclear for 8 of the included studies,^{13,23-27,29,30} limiting the generalizability of their findings. Examples of these limitations included a study population that was older and cohabitating and may not bear generalizability to a broader and more representative population;¹⁹ study settings that were limited to research activities and may not be generalizable to a real-world health care setting;²⁵ patient recruitment that was limited to urban settings with patients who were generally well-connected to adequate health care resources, and may not therefore bear sufficient generalizability to nonurban or disadvantaged populations;¹³ and, a lack of sufficient detail needed to assess the extent to which the health care settings were representative of the broader population of interest.³⁰

Notably, information describing numbers and/or characteristics of patients who chose not to participate in the research during recruitment was either not reported or not clearly reported for all of the 14 primary studies.^13,14,20-30 Data describing nonparticipants is important to an understanding of external validity (i.e., these ensure that the study population bears characteristics that are sufficiently similar to the source population to render findings that are representative and generalizable).

Internal Validity

Internal validity is necessary to establish confidence in the findings, demonstrated by methods that increase confidence in an association between the effect(s) observed with the exposures and/or interventions under study. None of the 14 primary studies summarized in this report used either randomization or blinding in their methods;^13,14,19-30 though, these methodological limitations are necessary features of cohort studies investigating the exposure of interest (i.e., ART in serodiscordant couples). Nonetheless, the absence of randomization and blinding introduce a limitation to the internal validity of the studies, as the effects observed on the outcomes of interest cannot be isolated from any possible effects imposed by the researchers’, clinicians’, and study participants’ knowledge of the exposures and interventions experienced by the participants.

The effect of viral load in sex partners LWH is critical to understanding risk of transmission in couples in serodiscordant relationships. Viral load measured at either baseline and/or across follow-up was reported by 6 of the primary studies,^{13,14,19,23,25,27} but was missing or not clearly described in the report of findings for 8 of the included studies.^{20-22,24,26,28-30} Furthermore, of the 6 studies reporting any data on viral load, 3 did not report, or clearly report, data across follow-up, describing viral load at baseline only^19,23,27 (with 1 of these including data on viral load for the entire study cohort, and no data specific to the subgroup of study participants relevant to this report²³). This lack of data describing viral load across follow-up introduces an important limitation to understanding its effect on the risk of HIV transmission.

Viral, or phylogenetic, linkage of HIV transmission for serodiscordant couples in whom seroconversion is observed is critical to understanding whether transmission occurred between the sex partners under study or not.³⁴ This variable was investigated and reported in 3 of the 14 studies,^13,14,21 but was absent from the others.^19,20,22-30 While self-report of exclusivity or monogamous sex partnerships may mitigate the limitations imposed by the absence of this variable, confirmation of transmission between the couple and/or sex partners under study cannot otherwise be validated, representing an important limitation to the internal validity, and thus interpretation, of the study findings.

The use of PrEP in 3 of the studies included in this report^13,24,25 introduces a potential threat to their internal validity with regard to reported estimates of the effect of ART on the risk of HIV transmission (i.e., serodiscordant couples in whom the partner who is HIV negative is using PrEP may demonstrate different rates of seroconversion than couples in whom no PrEP is being used). Nonetheless, 2 of the studies reported outcomes data separately for participants who were HIV negative who were or were not administered PrEP, which allows for a clearer assessment of how this exposure may or may not have impacted the findings of the studies.^13,24 One study included participants who were HIV negative and were were administered PrEP, which introduces an important risk of confounding with regard to the findings on HIV transmission in these serodiscordant couples.²⁵ Of the remaining studies summarized in this report, 1 excluded participant data with any exposure to PrEP (increasing confidence in the internal validity of the findings),¹⁴ whereas the others made no mention of exposure (or no exposure) to PrEP, leaving uncertainty as to whether the findings may or may not have been impacted by this potentially confounding factor.^18-23,26-30

For the 3 studies that reported actual losses to follow-up, proportions of participants lost or not completing follow-up ranged from between 11% and 35%.^13,21,25 Significant loss to, or failure to complete, study follow-up can introduce a risk of confounding, potentially introducing limitations on the internal validity of the studies.

Study Power

Sample size and/or study power is critical to understanding a study's capacity to detect clinically important effect(s).³⁵ Study power was not addressed (or not clearly described) by most of the included studies;^{19-21,23,25-27} though, 3 of the studies did provide information addressing and/or describing study power.^13,14,22,24 Of these, 1 study acknowledged that it was underpowered due to an insufficiently small sample size²⁴ and 2 studies reported an insufficient number of accrued couple-years to demonstrate a clinically meaningful effect.^13,14

Additional details regarding the strengths and limitations of the included publications are provided in Appendix 3.

Summary of Findings

Thirteen of the 15 included studies in this report presented data describing the absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner LWH takes ART with variable (or unreported) levels of viral load.^18-21,23-30 One of the 15 studies presented data describing the absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner LWH takes ART and has a suppressed viral load of less than 200 copies/mL on consecutive measurements every 4 to 6 months.¹⁴ One of the 15 studies presented data of relevance to both research questions.¹³

Appendix 4 presents detailed and tabulated study findings.

Absolute Risk of HIV Transmission Associated With Sex Among Serodiscordant Couples When the Sex Partner LWH Takes ART With Variable Levels of Viral Load

Overview of SRs and MAs

Of the 2 SRs summarized in the overview study that were not previously included in the 2018 SR by PHAC,¹² 1 described eligible data on absolute risk, reporting a pooled incidence rate of 2.13 (95% confidence interval [CI], 0.00 to 4.63) seroconversions across an unspecified number of person-years in an unreported number of serodiscordant heterosexual couples for whom the sex partner LWH was receiving ART.¹⁸ The findings from the second SR did not include data describing absolute risk of seroconversion.¹⁸

Primary Studies

Incidence of HIV transmission reported as seroconversions observed across 100 person-years of follow-up in serodiscordant couples for whom the sex partner LWH was receiving ART and demonstrating variable (or unspecified) levels of viral load ranged widely across 12 of the 13 included primary studies,^13,19-30 with the rate of seroconversions ranging between 0 (upper limit of 95% CI = 0.84)¹³ and 6.4 (95% CI, 4.64 to 8.76)²⁰ seroconversions per 100 person-years of follow-up reported for all patients included in the studies (i.e., subgroup findings not included).

Three cohort studies in which the sex partner LWH was receiving ART with variable (or unreported) levels of viral load described observing no seroconversions (or, no phylogenetically linked seroconversions) across study follow-up.^13,24,25 Two of these studies were prospective^13,25 and 1 was retrospective,²⁴ with 2 investigating HSSD pursuing pregnancy through natural conception^24,25 and 1 investigating MSM sex partners.¹³ Sample sizes ranged from between 74 couples with 146.4 years of person follow-up²⁵ to 343 couples with 588.4 couple-years of follow-up.¹³ In all 3 of these studies, some or all of the sex partners who were HIV negative were receiving PrEP during the study follow-up^13,24,25 (with 2 studies reporting findings separately by PrEP status of the partner who was HIV negative and reporting zero seroconversions in both subgroups).^13,24 Viral load was also monitored in sex partners LWH across study follow-up in 2 of the studies reporting no observed seroconversions,^13,25 with 1 study reporting 75% of participants consistently demonstrating a viral load of less than 200 copies/mL.¹³ The other study reporting on viral load across follow-up (and reporting no observed seroconversions) indicated that suppressed viral load (described variably across the study report as either < 400 copies/mL and < 1,000 copies/mL) was demonstrated during 74.3% of follow-up clinic visits in serodiscordant couples who achieved pregnancy, and during 64.6% of follow-up clinic visits in couples who did not achieve pregnancy.²⁵ Phylogenetic linkage analysis of observed seroconversions was conducted in 1 of the studies (with nonlinked seroconversions not included in the findings).¹³ HIV prevention methods included encouraging participants to limit condomless sex to the periovulatory period only,^24,25 monitoring of adherence to ART in the sex partner LWH,^13,25 and the sex partner who was HIV negative taking PrEP.^13,24,25 Another retrospective cohort study of 186 heterosexual couples across a total of 507.7 person-years also reported an observation of no seroconversions among the subgroup of 119 participants who received ART within 1 month of HIV diagnosis.²⁶

Of the 6 studies reporting at least 1 or more observed seroconversion(s) (but less than 2 seroconversions per 100 person-years of follow-up) across all study participant couples for whom the sex partner LWH was receiving ART with variable (or unreported) levels of viral load,^{19,21-23,26,28} 3 studies found incidence rates of less than 1 per 100 person-years of follow-up^19,21,26 and 3 studies reported incidence rates greater than 1 but less than 2 per 100 person-years of follow-up.^22,23,28 Four of these 6 studies were prospective in their design,^19,21-23 and 2 used retrospective study methods,^26,28 with eligible data reported from a range of 90²² to 7,713²⁸ study participants, and a range of study follow-up between a median of 1.4 years (standard deviation = 1.2)²¹ and 18,985.29 person-years.²⁸ None of the 6 studies reporting observed seroconversions of greater than 0 but less than 2 per 100 person-years of follow-up reported information on the use of PrEP in sex partners who were HIV negative,^{19,21-23,26,28} and 1 reported information on viral linkage of seroconversions,²¹ whereas the remaining 5 did not.^{19,22,23,26,28} Of the 1 study that demonstrated viral linkage for a proportion of all seroconversions observed across the study follow-up, 19 seroconversions were observed among 2,032 serodiscordant couples for whom the sex partner LWH was receiving ART, with 14 of these 19 found to demonstrate viral linkage. An incidence rate of 0.63 (95% CI, 0.38 to 1.00) was reported for these 19 seroconversions (which comprised a combination of seroconversions with and without demonstrated viral linkage [i.e., no incidence rate was reported for the subgroup of 14 virally linked seroconversions]).²¹ One additional study reported a finding of 1.6 (95% CI, 1.0 to 2.2) observed seroconversions per 100 person-years in the subgroup of serodiscordant couples in whom the sex partner LWH received early ART.²⁹

Of the remining 4 primary studies providing data in answer to the first research question posed in this report, 2 studies reported overall incidence rates exceeding 2 per 100 person-years of follow-up.^20,27 Another study did not report incidence rates for the overall study cohort; rather, only for subgroups of participants LWH by timing of ART initiation, finding rates of seroconversion per 100 person-years ranging from between 1.6 (95% CI, 1.0 to 2.2) and 6.7 (95% CI, 5.0 to 8.4) for participants with early to late ART initiation, respectively.²⁹ One additional retrospective study³⁰ reported that there were 573 seroconversions among sex partners and/or couples in whom the sex partner LWH was receiving ART across 12 months of follow-up per couple; however, the number of sex partners LWH and receiving ART (among the overall 5,544 individuals in the study) was unclear from the publication, making any interpretation of the findings challenging. None of these studies reported whether the partners who were HIV negative were receiving PrEP during the study follow-up. Of these 4 studies that described the highest numbers of observed seroconversions (i.e., more than 2 per 100 person-years of follow-up), all were retrospective and investigated heterosexual sex partner couples.^20,27,29,30 Sample sizes and follow-up ranged from between 200 couples with a median of 4.5 years of follow-up (range = 0.3 years to 21.7 years)²⁷ and 13,132 couples accrued and followed from 2003 to 2014.²⁹ ART adherence in sex partners LWH was measured by 1 of the studies,²⁰ and measurement of viral load in sex partners LWH was reported in 1 of the studies.²⁷ There was no viral linkage of seroconversions described by the 4 studies.^20,27,29,30 One study reported on condom use as an HIV-preventive measure used by some participants,²⁰ and another reported findings stratified by timing of ART initiation,²⁹ but no other preventive measures were reported in the 4 studies.^20,27,29,30 Of note, the extent to which the findings were interpretable in 1 of these studies was limited by inconsistent reporting and missing denominators for at least 1 outcome.³⁰

Absolute Risk of HIV Transmission Associated With Sex Among Serodiscordant Couples When the Sex Partner LWH Takes ART and Has a Suppressed Viral Load of Less Than 200 Copies/mL on Consecutive Measurements Every 4 to 6 Months

Overview of SRs and MAs

No novel data (i.e., not previously included in the 2018 PHAC SR)¹² describing the absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner LWH takes ART and has a suppressed viral load of less than 200 copies/mL on consecutive measurements every 4 to 6 months were identified in the overview of SRs and MAs.¹⁸

Primary Studies

Two studies were identified describing the absolute risk of HIV transmission in serodiscordant couples with a sex partner LWH demonstrating a suppressed viral load of less than 200 copies/mL on consecutive measurements across study follow-up. Both studies included MSM participant couples only, representing a combined total of 1,049 study participants and 1,370.4 couple-years of follow-up, and reporting no observed transmissions that were phylogenetically linked.^13,14 Notably, the findings for both of these studies did not include data on participant couples in whom the partner who was HIV negative was receiving PrEP.^13,14

Relative Risk of HIV Transmission Among Serodiscordant Couples When the Sex Partner LWH Does Versus Does Not Take ART

Though not a focus of this report, 5 of the included studies reported relative risks of HIV transmission in serodiscordant couples for whom the sex partner LWH did or did not receive ART.

Overview of SR and MA

Neither of the 2 SRs and MAs included in the overview and eligible for inclusion in this report described incidence rates comparing groups of serodiscordant sex partners for whom the participants LWH did or did not receive ART.¹⁸

Primary Studies

Four studies provided effect estimates describing relative risk rates of HIV transmission in serodiscordant couples, comparing those for whom the sex partner LWH did or did not receive ART.^21,28-30 Hazard ratios in 2 large, retrospective studies ranged from between 0.30 (95% CI, 0.18 to 0.50)²⁹ and 0.48 (95% CI, 0.40 to 0.59),²⁸ demonstrating a statistically significant benefit of ART, with 1 study specifying a 45% reduction in the risk of transmission observed in participants receiving ART.²⁸ Another large prospective study compared incidence rates of seroconversion across groups receiving or not receiving ART, and reported an adjusted rate ratio of 6.9 (95% CI, 3.8 to 12.4), demonstrating a statistically significant effect of ART on reducing the risk of transmission.²¹ Finally, a large retrospective study compared seroconversions across groups of serodiscordant heterosexual couples receiving and not receiving ART at 12 months of follow-up, reporting an odds ratio of 0.50 (95% CI, 0.42 to 0.60), suggesting a statistically significant benefit of not taking ART (i.e., P < 0.0001).³⁰ The authors of the study conceded that this finding is discordant with others in the literature and postulate that uncertainty of the HIV status of the nonindex partners at baseline may have contributed to the unexpected findings.³⁰

HIV seroconversions and rates were also reported across multiple additional subgroup and stratified analyses that are tabulated and detailed in Appendix 4.

Limitations

This review identified 15 studies describing exposure to ART in sex partners LWH and the risk of HIV transmission in serodiscordant sex partners, but is limited by the quality of available evidence, as well as limited available data addressing the various populations and/or exposures.

Regarding the 14 studies addressing absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner LWH takes ART with variable (or unreported) levels of viral load, all 14 studies had some methodological limitations, and some had considerable limitations.^{13,18-21,23-30} The overview of SRs and MAs included in this report was limited by a lack of clarity in its report of findings, a low quality of included SRs relevant to this review, and the degree of overlap with previous work.¹⁸ Specifically, the overview did not describe study characteristic details or absolute risk data for 1 of the SRs relevant to this review, which described an MSM population, and the quality of the other SR, for which a small amount of relevant data were available, was rated as low by the authors of the overview, demonstrating wide CIs and a limited degree of certainty in its results.¹⁸ Data overlap were also identified between those describing the PARTNER 1 study, as summarized in the 2018 SR,¹² and those reported in 1 study included in this report.¹⁴ Study quality for the 13 primary studies addressing absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner LWH takes ART with variable (or unreported) levels of viral load was also limited by the nonrandomized and unblinded methods employed, which limit the internal validity of study findings and the confidence that can be placed in the interpretation and conclusions of their results.^13,18-30 Critical appraisal of the 13 cohort studies rendered variable findings, with some studies demonstrating more (and different) strengths and/or limitations than others.^13,19-30 In particular, limitations were observed for some methodological features that demonstrate internal and external validity,^13,19-30 introducing limitations on the confidence that can be placed in the representativeness, generalizability, reliability, and validity of the findings summarized herein. Nonetheless, particular limitations identified for each study are important to consider when interpreting their findings, including acknowledgement of how these may impact the confidence that can be placed in the conclusions that have been drawn, as well as their utility. For instance, while a larger sample size is an important contributor to the confidence that can be placed in the findings, other study features, such as a retrospective study design, lack of clarity in reporting, and missing data — as was seen in 1 study summarized in this report³⁰ — reduce the confidence that can be placed in the findings and conclusions.

Evidence describing populations of interest was variable in 14 studies included in this report (i.e., 12 of the studies identified and summarized in this report focused only on HSSD,^19-30 with 1 primary study providing data specific to MSM,¹³ and 1 overview of SRs and MAs that provided no data of relevance regarding MSM).¹⁸ The 2018 SR by PHAC also included a limited number of studies describing MSM populations;¹² while the number of studies describing MSM populations has since increased, and now includes relatively large sample sizes as well as mostly robust study methods, there are still a limited number of available studies examining these groups.

Exposures of interest were represented and/or described with various levels of detail and relevance across the studies included and summarized in this report. While all of the included studies described use of ART in at least some of the sex partners LWH,^13,18-30 details concerning use of ART across follow-up, types of ART regimens used, and duration of and adherence to therapy were not described — or not described clearly — in many of the studies, limiting the extent to which the role of ART exposure on the rates of HIV transmission can be considered. Additional exposures described may have introduced a risk of confounding to the observed association between ART and HIV transmission risk (e.g., use of PrEP^13,24,25 or hepatitis C virus–positive status¹⁹ among sex partners who are HIV negative) and other exposures, such as condom use and/or timing of unprotected sex, may limit the internal validity of the findings with regard to the outcome of interest.

Viral load data for sex partners LWH were not reported by most studies included in this report, and for those that included these data, the availability and/or relevance was variable (e.g., viral load reported at baseline only and not across follow-up,^19,23,27 or viral load across follow-up by subgroups of participants who did or did not achieve pregnancy using natural conception, as opposed to HIV transmission and/or ART use).²⁵

Finally, data describing phylogenetic or viral linkage among observed seroconversions were not reported by 12 of the 14 included studies addressing the first research question concerning variable viral load.^18-20,22-30 As demonstrated in 2 of the included studies in this report, observed seroconversions may not always demonstrate a linkage between sex partners in the study,^13,14 which could introduce a risk to the internal validity of studies that do not include this information (e.g., overcounting of seroconversions may be assumed between the serodiscordant couples in the study where this information is missing).

Regarding the 2 studies reporting data in answer to the question concerning absolute risk of HIV transmission associated with sex among serodiscordant couples when the sex partner who is HIV positive takes ART and has a suppressed viral load of less than 200 copies/mL on consecutive measurements every 4 to 6 months (1 of which contributed data to the first question, as well), both contributed data from MSM populations, limiting the generalizability of the findings summarized in this report to other populations.^13,14 While this lack of data specific to HSSD that met the criteria for this report represents a gap in recent literature (i.e., published since 2017), it is notable that earlier studies identified and summarized in the 2018 PHAC SR did describe HSSD with sex partners who were HIV positive and receiving ART with a viral load of less than 200 copies/mL.¹² Both studies were affected by considerable participant loss to follow-up, failing to accrue the necessary number of couple-years specified by their respective study protocols, and limiting the study power required to demonstrate an effect.^13,14 In addition, both studies received funding from pharmaceutical companies, which can introduce potential for conflict of interest and risk of bias.^13,14

Overall, a limitation common to all 14 primary studies included in this report was external validity, which was either limited or unclear in many of the primary studies in this report.^{13,14,19-21,23-30} Generalizability of the 15 included studies to the Canadian context was also limited, with no available data generated in Canada, and a broad variety of populations and health care settings with unclear representativeness to a Canadian context.^13,14,18-30 Finally, none of the available evidence in this report described women who have sex with women.

Conclusions and Implications for Decision- or Policy-Making

This aim of this report was to provide a summary of evidence that has been published since 2017 and not previously captured in the 2018 PHAC SR.¹² This report identified 15 eligible studies addressing the risk of sexual transmission of HIV in serodiscordant couples for whom the sex partner LWH was treated with ART.^13,14,18-30 One study was an overview of SRs and MAs¹⁸ and 14 were nonrandomized cohort studies.^13,14,19-30 Fourteen included studies provided data in answer to the first research question posed by this report,^13,14,18-30 and 2 studies provided evidence in answer to the second research question^13,14 (1 of which also contributed data to the first research question¹³).

Regarding the absolute risk of HIV transmission associated with sex among serodiscordant couples in whom the sex partner LWH takes ART with variable (or unreported) measure of viral load, a wide range of seroconversions was observed across the 14 eligible studies reporting relevant data,^13,18-30 with incidence rates from all patients included in the studies ranging between 0 (upper limit of 95% CI = 0.84)¹³ and 6.4 (95% CI, 4.64 to 8.76)²⁰ seroconversions per 100 person-years of follow-up. Of the included studies, 3 studies reported no seroconversions;^13,24,25 7 studies observed rates between 0 and 2 per 100 person-years,^{19,21-23,26,28,29} and 5 studies reported rates exceeding 2 per 100 person-years^{18,20,27,29,30} (with 1 study reporting rates across subgroups only and reporting both rates between 0 and 2, and exceeding 2, per 100 person-years²⁹). Of the 3 studies with no seroconversions reported, 1 was affected by the potentially confounding effect of PrEP in all included participants.²⁵ The 2018 PHAC SR reported a pooled incidence rate of 0.22 (95% CI, 0.14 to 0.33) seroconversions per 100 person-years.¹² It is notable that the majority of included studies in this report describing serodiscordant couples in whom the sex partners LWH had variable (or unreported) viral loads also reported incidence rates ranging between greater than 0 and less than 1 seroconversion per 100 person-years,^{19,21-23,26,28,29} with smaller proportions of included studies reporting rates above^{18,20,27,29,30} or below^13,24,25 this range.

The range of observed seroconversions and rates across studies included in this report among couples for whom patients who were HIV positive had variable (including unreported) viral loads may, in part, be associated with multiple differences across study methods, populations, and exposures, including performance and reporting of phylogenetic linkage analyses that support the observed seroconversions having occurred between the sex partners under study. Additional methodological differences contributing to this broad range of findings may include retrospective versus prospective studies; study sample sizes and the representativeness of the study participants to other populations; duration of and/or adherence to ART (including measurement of the latter); and use of additional HIV-preventive interventions, such as PrEP, condoms, and/or timing of unprotected sex.^13,18-30

Regarding the absolute risk of HIV transmission associated with sex among serodiscordant couples in whom the sex partner LWH was takes ART and has a suppressed viral load of less than 200 copies/mL on consecutive measurements every 4 to 6 months, 2 studies met the inclusion criteria for this report, both of which investigated MSM populations and observed no seroconversions that were phylogenetically linked.^13,14 This finding was consistent with that generated by the 2018 PHAC SR, which reported pooled results from 2 earlier studies that reported no seroconversions across 1,327 observed person-years of follow-up (upper limit of 95% CI = 0.28); a finding that was later updated with additional data to 0 seroconversions (upper limit of 95% CI = 0.13).¹²

While the nonrandomized and unblinded methods of the studies included in this report limit the interpretation, randomized and/or blinded study designs are likely not feasible in the context of investigating association between ART use in sex partners LWH and the risk of HIV transmission associated with sex. Consequently, consideration of the methodological strengths and limitations for available studies is a critical feature of assessing the utility of evidence in this area. For instance, while only 2 studies investigating HIV transmission between serodiscordant MSM sex partners were identified (both of which provided data in answer to this report’s second research question),^13,14 both studies demonstrated important strengths that were not observed in many of the studies reporting data of relevance to the first question, including representation of populations across variable national contexts, and measurement of important exposure variables, such as ART adherence, viral load, and phylogenetic linkage of observed seroconversions.^13,14

Extending from this, populations represented by the available studies are an important consideration when weighing the generalizability of the evidence. Differences in HIV transmission patterns between various populations³⁶ make the need for robust data describing these populations essential to understanding the role of ART and any associated risk of HIV transmission associated with sex between serodiscordant couples. Furthermore, differences in the use and measurement of ART and other exposures across studies are important considerations; for instance, the timing of ART initiation following diagnosis of HIV is an important variable for understanding and assessing the risk of transmission.^37,38 In addition, demonstration of viral linkage is important information that can supplement self-reported data describing sexual and other risk behaviours, providing confirmatory evidence for observed seroconversions that have occurred between the sex partners under study (as opposed to those having occurred from another source).

The limitations of the included literature should be considered when interpreting the findings of this report. Furthermore, it is important to consider that this report does not provide a comprehensive summary of all available evidence addressing both research questions as it only summarizes evidence published since 2017. Future research will benefit the existing evidence base and help to inform consensus and policy by considering current gaps (i.e., high-quality, prospective studies with sufficient study power — that is, large sample sizes — representing participant characteristics, including measurement and consideration of viral load across study follow-up), critical exposure and outcomes data (including phylogenetic linkage of observed HIV transmissions), settings, and health care systems that can be generalized to broader populations and real-world contexts.

References

1.Sharp PM, Hahn BH. Origins of HIV and the AIDS pandemic. Cold Spring Harb Perspect Med. 2011;1(1):a006841-a006841. PubMed

2.World Health Organization. The top 10 causes of death. 2020; https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death. Accessed 2023 Jan 23.

3.World Health Organization Global HIV Programme. Key Facts HIV 2022; https://cdn.who.int/media/docs/default-source/hq-hiv-hepatitis-and-stis-library/key-facts-hiv-2021-26july2022.pdf?sfvrsn=8f4e7c93_5. Accessed 2023 Jan 20.

4.Public Health Agency of Canada. Estimates of HIV incidence, prevalence and Canada’s progress on meeting the 90-90-90 HIV targets, 2020. Ottawa, Canada: Public Health Agency of Canada; 2022: https://www.canada.ca/en/public-health/services/publications/diseases-conditions/estimates-hiv-incidence-prevalence-canada-meeting-90-90-90-targets-2020.html. Accessed 2023 Jan 19.

5.Gandhi RT, Bedimo R, Hoy JF, et al. Antiretroviral Drugs for Treatment and Prevention of HIV Infection in Adults: 2022 Recommendations of the International Antiviral Society–USA Panel. JAMA. 2023;329(1):63-84. PubMed

6.Aquaro S, Borrajo A, Pellegrino M, Svicher V. Mechanisms underlying of antiretroviral drugs in different cellular reservoirs with a focus on macrophages. Virulence. 2020;11(1):400-413. PubMed

7.Cohen MS, Chen YQ, McCauley M, et al. Antiretroviral Therapy for the Prevention of HIV-1 Transmission. N Engl J Med. 2016;375(9):830-839. PubMed

8.Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365(6):493-505. PubMed

9.Rodger AJ, Cambiano V, Bruun T, et al. Sexual Activity Without Condoms and Risk of HIV Transmission in Serodifferent Couples When the HIV-Positive Partner Is Using Suppressive Antiretroviral Therapy. JAMA. 2016;316(2):171-181. PubMed

10.Prevention Access Campaign. https://preventionaccess.org/. Accessed 2023 Mar 14.

11.Department of Justice Canada. HIV Non-Disclosure. 2022; https://www.justice.gc.ca/eng/cons/hiv-vih/nd.html#:~:text=The%20Supreme%20Court%20of%20Canada,will%20not%20constitute%20a%20crime. Accessed 2023 Jan 20.

12.LeMessurier J, Traversy G, Varsaneux O, et al. Risk of sexual transmission of human immunodeficiency virus with antiretroviral therapy, suppressed viral load and condom use: a systematic review. CMAJ. 2018;190(46):E1350-E1360. PubMed

13.Bavinton BR, Pinto AN, Phanuphak N, et al. Viral suppression and HIV transmission in serodiscordant male couples: an international, prospective, observational, cohort study. Lancet HIV. 2018;5(8):e438-e447. PubMed

14.Rodger AJ, Cambiano V, Bruun T, et al. Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. Lancet. 2019;393(10189):2428-2438. PubMed

15.Grey matters: a practical tool for searching health-related grey literature. Ottawa (ON): CADTH; 2019: https://www.cadth.ca/grey-matters. Accessed 2022 Nov 22.

16.Shea BJ, Reeves BC, Wells G, et al. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ. 2017;358:j4008. PubMed

17.Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-384. PubMed

18.Davari M, Giwa HB, Nabizade A, Taheri F, Giwa A. Antiretroviral therapy and the risk of sexual transmission of HIV: a systematic review and meta-analysis. HIV Med. 2020;21(6):349-357. PubMed

19.Chen F, Tang H, Han J, et al. Association of HCV Prior Infection and Unprotected Sex on Subsequent HIV Acquisition Risk in the Era of Treatment as Prevention. Front Med (Lausanne). 2022;9:902271. PubMed

20.Bantigen K, Kitaw L, Negeri H, et al. Rate of HIV Seroconversion Among Seronegative Male Partners Living with HIV Positive Women in Addis Ababa, Ethiopia, 2019: A Retrospective Cohort Study. HIV AIDS (Auckl). 2021;13:125-134. PubMed

21.Nyombayire J, Ingabire R, Mukamuyango J, et al. Antiretroviral Therapy Use and HIV Transmission Among Discordant Couples in Nonresearch Settings in Kigali, Rwanda. Sex Transm Dis. 2021;48(6):424-428. PubMed

22.Arjun Bal KP, Singh LR, Singh SK, et al. Seroprevalence and seroconversion rates of human immunodeficiency virus discordant couples attending a tertiary care center: A prospective cohort study. JMS - Journal of Medical Society. 2020;34(2):91-95.

23.Dieckhaus KD, Ha TH, Schensul SL, Sarna A. Modeling HIV Transmission from Sexually Active Alcohol-Consuming Men in ART Programs to Seronegative Wives. J Int Assoc Provid AIDS Care. 2020;19. PubMed

24.Sun L, Liu A, Li J, et al. Is PrEP necessary during natural conception in HIV-1-serodiscordant couples on ART with suppressed viral load? A retrospective cohort analysis. BMC Infect Dis. 2020;20(1):195. PubMed

25.Heffron R, Ngure K, Velloza J, et al. Implementation of a comprehensive safer conception intervention for HIV-serodiscordant couples in Kenya: uptake, use and effectiveness. J Int AIDS Soc. 2019;22(4):e25261. PubMed

26.Ma G, Chen H, Chen J, Jiang Z, Ma T. HIV transmission between spouses living in Lu'an city, Anhui province, China: a longitudinal study. Epidemiol Infect. 2019;147:e205. PubMed

27.Melo MGD, Sprinz E, Gorbach PM, et al. HIV-1 heterosexual transmission and association with sexually transmitted infections in the era of treatment as prevention. Int J Infect Dis. 2019;87:128-134. PubMed

28.Chen H, Yang X, Zhu Q, et al. Treatment for HIV prevention study in southwestern areas of China. Infect Dis Model. 2018;3:249-255. PubMed

29.Liu P, Tang Z, Lan G, et al. Early antiretroviral therapy on reducing HIV transmission in China: strengths, weaknesses and next focus of the program. Sci Rep. 2018;8(1):3431. PubMed

30.Zheng Z, Li Y, Jiang Y, Liang X, Qin S, Nehl EJ. Population HIV transmission risk for serodiscordant couples in Guangxi, Southern China: A cohort study. Medicine (Baltimore). 2018;97(36):e12077. PubMed

31.Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1-e34. PubMed

32.Pollock M, Fernandes RM, Becker LA, Pieper D, Hartling L. Chapter V: Overviews of Reviews. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook. 2022.

33.Ayerdi Aguirrebengoa O, Vera Garcia M, Portocarrero Nunez JA, et al. Implementing pre-exposure prophylaxis could prevent most new HIV infections in transsexual women and men who have sex with men. Rev Clin Esp (Barc). 2019;219(7):360-366. PubMed

34.Grabowski MK, Redd AD. Molecular tools for studying HIV transmission in sexual networks. Current opinion in HIV and AIDS. 2014;9(2):126-133. PubMed

35.Serdar CC, Cihan M, Yücel D, Serdar MA. Sample size, power and effect size revisited: simplified and practical approaches in pre-clinical, clinical and laboratory studies. Biochemia medica. 2021;31(1):010502. PubMed

36.UNAIDS. Global HIV & AIDS statistics — Fact sheet. 2022; https://www.unaids.org/en/resources/fact-sheet. Accessed 2023 Jan 23.

37.Nanditha NGA, Dong X, Tafessu HM, et al. A province-wide HIV initiative to accelerate initiation of treatment-as-prevention and virologic suppression in British Columbia, Canada: a population-based cohort study. CMAJ Open. 2022;10(1):E27-E34. PubMed

38.Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in Adults and Adolescents with HIV. 2019. https://clinicalinfo.hiv.gov/en/guidelines/hiv-clinical-guidelines-adult-and-adolescent-arv/initiation-antiretroviral-therapy. Accessed 2023 Jan 20.

39.Rodger A, Bruun T, Weait M, et al. Partners of people on ART - a New Evaluation of the Risks (The PARTNER study): design and methods. BMC Public Health. 2012;12(1):296. PubMed

Appendix 1: Selection of Included Studies

Figure 1: Selection of Included Studies

Appendix 2: Characteristics of Included Publications

Note that this appendix has not been copy-edited.

Table 2: Characteristics of Included Overview of Systematic Reviews

AMSTAR 2 = A MeaSurement Tool to Assess systematic Reviews 2; ART = antiretroviral therapy; HSSD = heterosexual serodiscordant; ; LWH = living with HIV; MA = meta-analysis; MSM = men who have sex with men; NR = not reported; ROB = risk of bias; SR = systematic review.

^aOnly 2 of the included SRs were not overlapped by other sources in this report; thus, data from these 2 SRs are summarized herein.

Appendix 3: Critical Appraisal of Included Publications

Note that this appendix has not been copy-edited.

Table 4: Strengths and Limitations of Overview of Systematic Reviews Using AMSTAR 2¹⁶

AMSTAR 2 = A MeaSurement Tool to Assess systematic Reviews 2; COI = conflict of interest; PRISMA = Preferred Reporting Items in Systematic Reviews and Meta-Analyses; PROSPERO = International prospective register of systematic reviews; SR = systematic review

Table 5: Strengths and Limitations of Clinical Studies Using the Downs and Black Checklist¹⁷

ART = antiretroviral therapy; CD4 = cluster of differentiation 4; CI = confidence interval; HCV = hepatitis C virus; LWH = living with HIV; NR = not reported; PrEP = pre-exposure prophylaxis; STI = sexually transmitted infection

^aThe study cohort included serodiscordant and seroconcordant couples (where both partners were HIV positive at the time of diagnosis and one partner was determined to be infected from the other). Only the characteristics and results of the relevant population i.e., serodiscordant couples where the sex partner LWH takes ART is summarized in the current report.

Appendix 4: Main Study Findings

Note that this appendix has not been copy-edited.

Table 6: Summary of Findings by Outcome — HIV Seroconversion in Single-Arm Studies of Those With Any Viral Load

CI = confidence interval; HSSD = heterosexual serodiscordant; LWH = living with HIV; MSM = men who have sex with men; N = number; NR = not reported; py = person-years

Note: Results of subgroup analysis are presented in Table 9 and Table 10

^aRates of seroconversion was reported in the publication as “per 1000 person years.”

Table 7: Summary of Findings by Outcome — HIV Seroconversion in Single-Arm Studies of Those With a Viral Load of < 200 copies/mL

CI = confidence interval; LWH = living with HIV; MSM = men who have sex with men; N = number; py = person-year

Note: Results of subgroup analysis are presented in Table 9 and Table 10

^aResults for condomless anal sex, phylogenetically linked transmissions only.

Table 10: Summary of Findings by Outcome — HIV Seroconversion in MSM Couples Stratified by Subgroups

ART = antiretroviral therapy; CI = confidence interval; CLAI = condomless anal intercourse; LWH = living with HIV; MSM = men who have sex with men; PrEP = pre-exposure prophylaxis; STI = sexually transmitted infections

Table 11: Summary of Findings by Outcome — HIV Seroconversion ART vs. No ART Stratified by Subgroups With Any Viral Load

AHR = adjusted hazard ratio; ART = antiretroviral therapy; CD4 = cluster of differentiation 4; CI = confidence interval; HR = hazard ratio; LWH = living with HIV; M+/F- = male partner positive, female partner negative; M-/F+ = male partner negative, female partner positive

^aAdjusted for man’s age

^bAdjusted for age, sex, level of education, marital status, occupation, and route of HIV transmission in the sex partner LWH

Table 12: Summary of Findings by Outcome — Incidence of STI

LWH = living with HIV; STI = sexually transmitted infection.

Appendix 5: References of Potential Interest

Scoping Review

Knight J, Kaul R, Mishra S. Risk heterogeneity in compartmental HIV transmission models of ART as prevention in Sub-Saharan Africa: A scoping review. Epidemics. 2022 09;40:100608.

Non-randomized studies

Le Guillou A, Cabie A, Delpierre C, et al. Dramatic decline in new HIV diagnoses in persons born in France in a large nationwide HIV cohort. Public Health. 2021;196():129-134.

Lima VD, Brumme ZL, Brumme C, et al. The Impact of Treatment as Prevention on the HIV Epidemic in British Columbia, Canada. Curr HIV/AIDS Rep. 2020;17(2)():77-87.

Woodson E, Goldberg A, Michelo C, et al. HIV transmission in discordant couples in Africa in the context of antiretroviral therapy availability. AIDS. 2018;32(12)():1613-1623.

Modelling Studies

Zhu J, Rozada I, David J, et al. The potential impact of initiating antiretroviral therapy with integrase inhibitors on HIV transmission risk in British Columbia, Canada. EClinicalMedicine. 2019;13():101-111.

Supervie V, Breban R. Brief Report: Per Sex-Act Risk of HIV Transmission Under Antiretroviral Treatment: A Data-Driven Approach. J Acquir Immune Defic Syndr. 2018 12 01;79(4):440-444.

Additional References

Patterson S, Nicholson V, Milloy MJ, et al. Awareness and Understanding of HIV Non-disclosure Case Law and the Role of Healthcare Providers in Discussions About the Criminalization of HIV Non-disclosure Among Women Living with HIV in Canada. Aids Behav. 2020;24(1):95-113. PubMed

Kesler MA, Kaul R, Loutfy M, et al. Prosecution of non-disclosure of HIV status: Potential impact on HIV testing and transmission among HIV-negative men who have sex with men. PLoS One. 2018;13(2):e0193269. PubMed

Patterson S, Kaida A, Ogilvie G, et al. Awareness and understanding of HIV non-disclosure case law among people living with HIV who use illicit drugs in a Canadian setting. Int J Drug Policy. 2017;43():113-121.