The Lancet MAY 02, 2019

Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER2): final results of a multicentre, prospective, observational study

Alison J Rodger, Valentina Cambiano, Tina Bruun, et al.

Bottom Line

The PARTNER2 study demonstrated that among gay male serodifferent couples, there were zero linked HIV transmissions during periods of condomless sex when the HIV-positive partner maintained a viral load suppressed by antiretroviral therapy.

Key Findings

1. Among 783 gay male couples followed for a median of 1.6 years, there were zero linked HIV transmissions identified over 76,991 reported acts of condomless anal sex.
2. The primary analysis yielded an estimated HIV transmission rate of zero per 100 couple-years of follow-up.
3. The upper limit of the 95% confidence interval for the transmission rate was 0.23 per 100 couple-years, indicating that the risk is clinically negligible when viral load is suppressed.
4. Despite 15 HIV-negative partners acquiring HIV infection during the study period, phylogenetic analysis confirmed none were genetically linked to the HIV-positive study partner, indicating acquisition occurred from external sexual encounters.
5. The data provide evidence of equivalent safety for gay men as previously established for heterosexual couples, supporting the 'Undetectable equals Untransmittable' (U=U) consensus.

Study Design

Design
Prospective Observational Study
N/A
Sample
783
Patients
Duration
1.6 yr
Median
Setting
Multicenter, Europe
Population Gay male serodifferent couples where one partner was HIV-positive on ART and the other was HIV-negative.
Intervention Condomless anal sex while the HIV-positive partner maintained a viral load of less than 200 copies/mL.
Comparator None (Prospective observational design assessing incidence).
Outcome Number of genetically linked HIV transmissions from the HIV-positive partner to the HIV-negative partner.

Study Limitations

The study was observational rather than a randomized controlled trial, relying on participant self-reporting for sexual behavior and adherence.
While the total number of sexual acts was large, the statistical power remains limited for rare subgroups or scenarios involving high-risk factors such as specific types of sexual acts combined with undiagnosed sexually transmitted infections.
Participants were recruited from clinical settings and may not be fully representative of all MSM populations, potentially limiting generalizability to those outside of established HIV care networks.
The upper bounds of the confidence intervals, while low, mean that the transmission risk cannot be mathematically excluded as strictly zero in all conceivable scenarios.

Clinical Significance

These findings provide high-level evidence that effective antiretroviral therapy resulting in viral suppression renders an individual sexually noninfectious. This supports the global U=U campaign, reduces HIV-related stigma, and guides clinical counseling by clarifying that condomless sex does not pose a risk of HIV transmission for serodifferent couples when the HIV-positive partner is virally suppressed.

Historical Context

The PARTNER2 study served as the essential follow-up to the original PARTNER1 study, which provided strong evidence for HIV prevention via treatment in heterosexual couples but had narrower statistical power to definitively rule out transmission risks in gay male couples practicing anal sex. Together, these trials revolutionized the clinical and public health understanding of HIV as a chronic, manageable, and non-transmissible condition under effective treatment.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Based on the biological mechanisms of HIV replication and the results of the PARTNER2 study, how does effective antiretroviral therapy (ART) achieve a state where the risk of sexual transmission is effectively zero?

Key Response

ART works by inhibiting key enzymes—reverse transcriptase, integrase, and protease—which prevents the production of new virions. When ART is consistently taken, the viral load in both the blood and genital secretions falls below the limit of detection (typically <50 copies/mL, though PARTNER2 used a threshold of <200 copies/mL). At these levels, the concentration of virus is insufficient to establish an infection in the mucosal tissues of a partner, forming the basis for the 'Undetectable = Untransmittable' (U=U) principle.

Resident
Resident

A serodifferent gay couple presents to the clinic; the HIV-positive partner has been virally suppressed for 12 months. They ask if they still need to use condoms to prevent HIV transmission. According to PARTNER2, what are the three essential criteria you must confirm before advising them that the risk is zero?

Key Response

Clinical application of PARTNER2 requires confirming: 1) The HIV-positive partner is on a stable ART regimen with excellent adherence; 2) The viral load is suppressed (defined as <200 copies/mL) and has been for at least six months; and 3) Viral load monitoring is performed regularly (every 4-6 months). Notably, PARTNER2 found that the presence of other STIs did not lead to any linked HIV transmissions, though condom use is still recommended for STI prevention.

Fellow
Fellow

PARTNER2 specifically focused on MSM (men who have sex with men). How did the statistical power and the upper limit of the 95% confidence interval in this study provide more definitive evidence for this population compared to the earlier PARTNER1 study?

Key Response

Anal intercourse carries a significantly higher per-act transmission risk than vaginal intercourse. PARTNER1 lacked sufficient power to make definitive claims for MSM. PARTNER2 collected data on over 76,000 condomless sex acts specifically in MSM couples. The study reported zero linked transmissions, and the upper bound of the 95% confidence interval for the transmission rate was only 0.23 per 100 couple-years, providing the robust statistical confidence needed to recommend U=U for anal sex as confidently as for vaginal sex.

Attending
Attending

How should the 'zero linked transmissions' result of PARTNER2 shift the clinical dialogue from 'risk reduction' to 'risk elimination,' and what are the broader public health implications for HIV stigma and PrEP utilization?

Key Response

The findings transition HIV management from a harm-reduction model to a definitive prevention model. Clinicians can now state with scientific certainty that suppressed viral load eliminates the risk of sexual transmission. This helps de-stigmatize HIV-positive individuals by removing the 'vector' label. It also allows for nuanced discussions regarding PrEP; if the HIV-positive partner is consistently suppressed, the HIV-negative partner may choose to discontinue PrEP, assuming the relationship is monogamous and suppression is verified.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

PARTNER2 utilized phylogenetic analysis to verify the source of any new HIV infections. Critically evaluate why this methodology was essential to the study's validity and what 'linked' versus 'unlinked' infections reveal about the study cohort's behavior.

Key Response

In a cohort of serodifferent couples, new infections in the negative partner can occur from outside the primary relationship. Phylogenetic analysis compares the pol and env sequences of the virus from both partners. If the sequences are significantly different, the infection is 'unlinked.' In PARTNER2, 15 new infections occurred, but all were unlinked to the suppressed partner. Without phylogenetics, the study would have falsely reported 15 failures of ART to prevent transmission, leading to incorrect conclusions about the efficacy of U=U.

Journal Editor
Journal Editor

As a peer reviewer, how would you evaluate the threat of 'attrition bias' in PARTNER2, considering the long-term follow-up required to observe transmission events in a prospective cohort?

Key Response

A tough reviewer would flag that couples who remained in the study (783 couples provided data) might be more adherent to ART or more health-conscious than those lost to follow-up. If the couples who dropped out were those with poor adherence or fluctuating viral loads, the 'zero transmission' result might be over-optimistic. However, the study's large number of sex acts and the use of 'couple-years of follow-up' as the denominator helps mitigate this, but the editor would require a transparent sensitivity analysis of those lost to follow-up.

Guideline Committee
Guideline Committee

Based on the PARTNER2 findings, should international guidelines (e.g., WHO or CDC) move from recommending ART 'to reduce transmission' to 'to prevent transmission,' and how does this impact the 'Level of Evidence' for MSM populations?

Key Response

PARTNER2 provides 'Level 1a' evidence (large prospective multicentre study) that ART prevents transmission. Current CDC and WHO guidelines have already been updated to reflect that 'effectively no risk' exists when U=U is achieved. The PARTNER2 results specifically allow guidelines to align the recommendations for MSM with those for heterosexual couples, removing the previous caveat that data for anal sex was less certain. It justifies a Grade 1A recommendation for U=U as a primary prevention strategy.

Clinical Landscape

Noteworthy Related Trials

2011

HPTN 052

n = 1,763 · NEJM

Tested

Early initiation of antiretroviral therapy (ART)

Population

HIV-serodifferent heterosexual couples

Comparator

Delayed initiation of ART

Endpoint

Linked HIV transmission to the HIV-uninfected partner

Key result: Early ART initiation reduced HIV transmission risk by 96% in serodifferent heterosexual couples.
2016

PARTNER1

n = 1,166 · JAMA

Tested

Suppressive antiretroviral therapy (ART)

Population

HIV-serodifferent couples (men who have sex with men and heterosexuals)

Comparator

Observational cohort (condomless sex)

Endpoint

Linked HIV transmission to the HIV-uninfected partner

Key result: No linked HIV transmissions were observed among couples who had condomless sex when the HIV-positive partner was virally suppressed.
2018

Opposites Attract

n = 343 · Lancet HIV

Tested

Suppressive antiretroviral therapy (ART)

Population

HIV-serodifferent gay male couples

Comparator

Observational cohort (condomless sex)

Endpoint

Linked HIV transmission to the HIV-uninfected partner

Key result: Zero linked HIV transmissions occurred over the study period among gay couples with virally suppressed HIV-positive partners.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis