The Lancet June 15, 2019

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

Alison J Rodger, Valentina Cambiano, Tina Bruun, Pietro Vernazza, Simon Collins, et al.

Bottom Line

The PARTNER2 study provided definitive clinical evidence that the risk of HIV transmission through condomless anal sex is effectively zero when the HIV-positive partner is maintained on virally suppressive antiretroviral therapy.

Key Findings

1. During 1,593 eligible couple-years of follow-up (median 2.0 years), serodifferent gay couples reported engaging in condomless anal sex a total of 76,088 times.
2. There were 15 incident HIV infections observed among the initially HIV-negative partners.
3. Phylogenetic analysis demonstrated that 0 of these 15 infections were linked to the primary HIV-positive partner (0 linked transmissions).
4. The incidence rate of linked within-couple HIV transmission was zero, with an upper 95% confidence limit of 0.23 per 100 couple-years of follow-up.

Study Design

Design
Prospective Observational Cohort
N/A
Sample
782 couples
Patients
Duration
2.0 yr
Median
Setting
14 European countries
Population Serodifferent gay male couples where the HIV-positive partner was taking suppressive ART and both partners reported recent condomless anal sex.
Intervention Condomless anal sex during periods when the HIV-positive partner maintained a suppressed plasma HIV-1 RNA load (<200 copies/mL).
Comparator N/A (Observational single-arm assessing incidence rate)
Outcome Incidence rate of phylogenetically linked HIV-1 transmission from the HIV-positive to the HIV-negative partner.

Study Limitations

The observational nature of the study relied on self-reported questionnaires for the frequency and type of sexual acts.
The occurrence of 15 unlinked HIV infections highlights that a significant proportion of participants engaged in condomless sex outside the primary relationship, indicating that broader prevention tools (like PrEP and condoms) remain critical for comprehensive STI and non-partner HIV protection.
Median follow-up time was relatively short (2.0 years), although the sheer volume of condomless sex acts (over 76,000) generated robust statistical confidence.

Clinical Significance

By demonstrating zero linked transmissions across more than 76,000 condomless anal sex acts, PARTNER2 definitively validated the 'Undetectable = Untransmittable' (U=U) principle for men who have sex with men (MSM). This provides foundational, unassailable evidence for modern HIV public health messaging, helping to dismantle HIV-related stigma, drastically improve the psychological well-being of people living with HIV, incentivize early ART initiation, and scientifically challenge outdated HIV criminalization laws.

Historical Context

Prior to PARTNER2, the landmark HPTN 052 trial and the first phase of the PARTNER study (PARTNER1) established that ART prevents sexual transmission of HIV in heterosexual couples. However, because receptive anal sex carries a significantly higher biological risk of HIV transmission than vaginal sex, rigorous statistical evidence was needed to confirm if viral suppression could completely eliminate transmission risk among MSM. PARTNER2 was specifically designed as an extension study to definitively close this evidence gap, providing the statistical precision necessary to confirm U=U for gay men.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the biological mechanism by which antiretroviral therapy (ART) prevents sexual transmission of HIV, and what does the concept U=U mean in the context of the PARTNER2 study?

Key Response

ART suppresses viral replication by inhibiting key viral enzymes like reverse transcriptase, integrase, or protease, dropping the viral load in blood and genital secretions to undetectable levels. U=U stands for Undetectable equals Untransmittable, a concept definitively supported by PARTNER2 showing zero linked transmissions when the HIV-positive partner's viral load is under 200 copies per mL.

Resident
Resident

How should the results of the PARTNER2 study change your clinical counseling for a newly diagnosed HIV-positive MSM patient and their HIV-negative partner regarding condom use, PrEP, and STI prevention?

Key Response

Residents should counsel that while suppressive ART effectively eliminates HIV transmission risk (making PrEP for the negative partner unnecessary if the positive partner is consistently undetectable and they are strictly monogamous), it does not protect against other STIs like syphilis or gonorrhea. Condom use is still recommended depending on the couple's outside sexual networks.

Fellow
Fellow

In the PARTNER2 study, 15 HIV-negative partners did acquire HIV, but the transmissions were not phylogenetically linked to their main partner. How does phylogenetic analysis work in this context, and why is it crucial for determining the true efficacy of Treatment as Prevention (TasP)?

Key Response

Phylogenetic analysis sequences viral genes (like pol and env) from both partners to construct an evolutionary tree. If the viral strains do not cluster together, the transmission came from outside the relationship. This is critical in TasP studies because without it, outside acquisitions would falsely lower the perceived efficacy of suppressive ART, masking the true zero-transmission rate from the suppressed partner.

Attending
Attending

Beyond the biological prevention of transmission, how does integrating the definitive zero-risk findings of PARTNER2 into your clinical practice alter the psychological care and societal stigma management for patients living with HIV?

Key Response

The U=U message fundamentally transforms patient identity, lifting the profound psychological burden of feeling infectious or dangerous to loved ones. Attendings should use this data not just for epidemiological control, but as a therapeutic tool to improve patient mental health, encourage strict ART adherence, and actively combat internalized and external stigma.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The PARTNER2 study reported a transmission rate of 0.00 with a 95 percent confidence interval upper bound of 0.23 per 100 couple-years. From a statistical and epidemiological standpoint, why is it impossible to prove a risk is mathematically zero, and how do we determine if an upper bound is tight enough to inform public health policy?

Key Response

In frequentist statistics, observing zero events yields a point estimate of zero, but the upper bound of the CI depends heavily on the person-years of follow-up. Proving absolute zero is impossible. The upper bound of 0.23 means the true risk could theoretically be up to 1 in roughly 435 years of condomless sex, demonstrating the necessity of massively powered, multi-year observational cohorts to narrow the CI to a level deemed practically zero by public health experts.

Journal Editor
Journal Editor

As a peer reviewer assessing the PARTNER2 manuscript, what are the primary methodological threats to validity regarding self-reported sexual behavior and viral load monitoring intervals, and how did the authors attempt to mitigate them?

Key Response

A rigorous reviewer would flag recall bias in self-reported condomless anal sex acts and the risk of unmeasured viral load blips between testing intervals. The authors mitigated this through prospective questionnaire designs and frequent viral load testing (censoring data where gaps between tests were too large), though transient viremia remains a theoretical but practically insignificant unmeasured confounder.

Guideline Committee
Guideline Committee

Based on the PARTNER2 results, how should WHO and CDC guidelines regarding HIV Treatment as Prevention (TasP) and PrEP indications for serodiscordant MSM couples be updated, and what Level of Evidence does this study provide?

Key Response

PARTNER2 provides strong prospective cohort evidence that acts as Level I for TasP, given the ethical impossibility of an RCT for this question. Current CDC and WHO guidelines now explicitly endorse U=U. Guidelines should clearly state that PrEP is not indicated for the negative partner in a monogamous relationship if the positive partner has documented, sustained viral suppression for at least 6 months, thereby shifting public health resources to individuals at actual risk.

Clinical Landscape

Noteworthy Related Trials

2011

HPTN 052 Trial

n = 1,763 · NEJM

Tested

Early initiation of ART

Population

HIV-serodiscordant predominantly heterosexual couples

Comparator

Delayed ART initiation

Endpoint

Linked HIV transmission

Key result: Early ART reduced the risk of linked HIV transmission by 96 percent.
2016

PARTNER 1 Study

n = 1,166 · JAMA

Tested

Suppressive ART

Population

HIV-serodiscordant heterosexual and MSM couples

Comparator

None (Observational cohort)

Endpoint

Phylogenetically linked HIV transmission

Key result: Zero linked HIV transmissions occurred after 58,000 condomless sex acts.
2018

Opposites Attract Study

n = 343 · Lancet HIV

Tested

Suppressive ART

Population

HIV-serodiscordant male homosexual couples

Comparator

None (Observational cohort)

Endpoint

Phylogenetically linked HIV infections

Key result: Zero linked HIV transmissions occurred during nearly 17,000 acts of condomless anal sex.

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