Nancy Padian

This session presented the results of several large pragmatic, effectiveness studies related to treatment and focused on implementation issues such as decentralized care, integrated care, use of intensive methods to minimize loss to follow-up, and community engagement.

The Eswatini MaxART study (Okello), the Ya Tsie Botswana Project (Makhema) and SEARCH (Havlir) in rural Kenya and Uganda are universal-treatment cluster RCTs, all with deep community engagement. In addition to facility-based care, services were offered in homes, mobile clinics and community health fairs. All demonstrated the importance of LTFU as a major conduit to treatment as substantial numbers of new treatment initiates knew they were infected, but were not engaged in care. Ya Tsie and MaxART demonstrated reduced viral load with early treatment. Although not so for SEARCH, its integrated approach to health was apparent in significant improvements in TB and hypertension control. Another integrated approach to HIV care was reported in Vancouver (Barker) where infected substance users were more likely to adhere to HIV treatment and have reduced viral load when retained in methadone maintenance therapy. This suggests the importance of integrating HIV care with addiction services.

Although some studies were able to demonstrate significant findings, they also highlighted the challenges of conducting effectiveness trials with active control groups in the context of changing WHO guidelines for standard of care.

Graeme Meintjes

This session featured abstracts specifically selected by the conference co-chairs given their importance and implications for global HIV treatment and prevention programmes. The primary endpoint results of three large community-based randomised trials investigating the impact of multi-faceted, enhanced HIV treatment and prevention strategies were presented.

The community-randomized trial, BCPP/Ya Tsie, was conducted in 30 communities in Botswana. It evaluated the impact of a package of interventions on population-level HIV incidence. The interventions included community-based HIV testing, linkage-to-care, and expanded ART criteria (with universal ART from June 2016). The intervention resulted in a 31% reduction in community HIV incidence. The SEARCH trial in East Africa evaluated an HIV “test and treat” strategy with universal ART, randomizing 32 communities. The intervention did not result in reduced HIV incidence, but significant increases in ART initiation and virologic suppression and reductions in HIV-mortality (21% reduction) and HIV-TB incidence (59% reduction) were observed. MaxART, a stepped-wedge randomized-controlled health systems trial conducted in Swaziland, had an intervention offering ART regardless of CD4 count. Clinical mentoring and community mobilization support was also provided. 86% retention was observed under the intervention compared with 80% under standard of care at 12 months.

BCPP/Ya Tsie showed that it is possible to increase ART coverage and reduce HIV incidence in a high-prevalence generalized epidemic, even when high baseline ART coverage exists. In contrast, in SEARCH no reduction in HIV incidence was observed with the intervention, possibly due to certain interventions also being offered to control communities and evolution of ART guidelines affecting standard of care. However, mortality and morbidity benefits of the intervention in SEARCH were substantial. The findings of MaxART are reassuring given the concern that UTT may result in worse retention in care; the opposite was found.

Michael Marco

AIDS 2018 Co-chairs' Choice: Report on PARTNERS2 Study. Treatment as prevention (TasP) is a paradigm shift in HIV prevention modalities. For sero-different couples, TasP research has been undertaken to determine the risk of HIV transmission during condomless sex when the index partner is virally suppressed on ART.

The original PARTNERS1 observational study enrolled sero-different heterosexual and gay couples. For gay male sero-different couples, zero cases of HIV transmission were documented. However, the upper 95% confidence limit of the rate for gay men was 0.84 /100 CYFU. PARTNERS2 was a follow-up study aimed at providing a more precise estimates of transmission risk in gay partnerships. This TasP study, was, in essence, to determine if being undetectable is equated with being untransmissable. PARTNERS2 had 1) very strict inclusion and exclusion criteria; 2) the ability to phylogenetically link transmissions during eligible CYFU; 3) a robust sample size (783 gay couples); 4) multiple sites (conducted at 75 European clinical sites); 5) median 1.6 years of follow-up; and 6) a high degree of risk of HIV transmission. Among serodifferent gay couples who had sex ~77,000 times without condoms with undetectable viral load (<200 copies), there were zero phylogenetically-linked transmissions during ~1600 CYFU.

PARTNERS2 study is a watershed moment in HIV prevention research history. It is simply one of the most important HIV observational prevention studies to date. Many of the strengths of its design are listed above. The one limitation is its generalizability: 89% of the participants were white European males. Nevertheless, it’s hard to argue with zero transmissions. With such robust data confirming the results of other studies, the jury is now in: Undetectable = Untransmissable (U=U)!