Efficacy and safety of long-acting cabotegravir compared with daily oral tenofovir disoproxil fumarate plus emtricitabine to prevent HIV infection in cisgender men and transgender women who have sex with men 1 year after study unblinding: a secondary analysis of the phase 2b and 3 HPTN 083 randomised controlled trial

Abstract

Background: Injectable cabotegravir was superior to daily oral tenofovir disoproxil fumarate plus emtricitabine for HIV prevention in two clinical trials. Both trials had the primary aim of establishing the HIV prevention efficacy of long-acting injectable cabotegravir pre-exposure prophylaxis (PrEP) compared with tenofovir disoproxil fumarate plus emtricitabine daily oral PrEP. Long-acting PrEP was associated with diagnostic delays and integrase strand-transfer inhibitor (INSTI) resistance. This report presents findings from the first unblinded year of the HIV Prevention Trials Network (HPTN) 083 study.

Methods: The HPTN 083 randomised controlled trial enrolled HIV-uninfected cisgender men and transgender women at elevated HIV risk who have sex with men, from 43 clinical research sites in Africa, Asia, Latin America, and the USA. Inclusion criteria included: a negative HIV serological test at the screening and study entry, undetectable HIV RNA levels within 14 days of study entry, age 18 years or older, overall good health as determined by clinical and laboratory evaluations, and a creatinine clearance of 60 mL/min or higher. Participants were randomly allocated to receive long-acting injectable cabotegravir or daily oral tenofovir disoproxil fumarate plus emtricitabine PrEP. After study unblinding, participants remained on their original regimen awaiting an extension study. HIV infections were characterised retrospectively at a central laboratory. Here we report the secondary analysis of efficacy and safety for the first unblinded year. The primary outcome was incident HIV infection. Efficacy analyses were done on the modified intention-to-treat population using a Cox regression model. Adverse events were compared across treatment groups and time periods (blinded vs unblinded). This trial is registered with ClinicalTrials.gov, NCT02720094.

Findings: Of the 4488 participants who contributed person-time to the blinded analysis, 3290 contributed person-time to the first unblinded year analysis between May 15, 2020, and May 14, 2021. Updated HIV incidence in the blinded phase was 0·41 per 100 person-years for long-acting injectable cabotegravir PrEP and 1·29 per 100 person-years for daily oral tenofovir disoproxil fumarate plus emtricitabine PrEP (hazard ratio [HR] 0·31 [95% CI 0·17-0·58], p=0·0003). HIV incidence in the first unblinded year was 0·82 per 100 person-years for long-acting PrEP and 2·27 per 100 person-years for daily oral PrEP (HR 0·35 [0·18-0·69], p=0·002). Adherence to both study products decreased after study unblinding. Additional infections in the long-acting PrEP group included two with on-time injections; three with one or more delayed injections; two detected with long-acting PrEP reinitiation; and 11 more than 6 months after their last injection. Infection within 6 months of cabotegravir exposure was associated with diagnostic delays and INSTI resistance. Adverse events were generally consistent with previous reports; incident hypertension in the long-acting PrEP group requires further investigation.

Interpretation: Long-acting injectable cabotegravir PrEP retained high efficacy for HIV prevention in men and transgender women who have sex with men during the first year of open-label follow-up, with a near-identical HR for HIV risk reduction between long-acting injectable cabotegravir and daily oral tenofovir disoproxil fumarate plus emtricitabine PrEP during the first year after unblinding compared with the blinded period. Extended follow-up further defined the risk period for diagnostic delays and emergence of INSTI resistance.

Funding: Division of AIDS at the National Institute of Allergy and Infectious Diseases, ViiV Healthcare, and Gilead Sciences.

Figure 1:. Pharmacological and virological data for HIV infections in the long-acting cabotegravir group identified in the extended analysis

HIV genotyping results are shown to the right of each horizontal bar. Major resistance mutations are shown for nucleoside or nucleotide reverse-transcriptase inhibitors, non-nucleoside reverse-transcriptase inhibitors, protease inhibitors, and integrase strand-transfer inhibitors. The first visit with a genotyping result is shown with a single red square; later visits with genotyping results are shown with two red squares. The pink shading denotes participants for whom infections occurred more than 3 years from enrolment, and who were therefore excluded from prespecified efficacy analyses. BLQ=below the limit of quantification. CAB=cabotegravir. LLOQ=lower limit of quantification. PA-IC₉₀=protein-adjusted 90% cabotegravir inhibitory concentration. WT=wild type. *Group B includes infections that occurred with no recent cabotegravir exposure (>6 months after the last long-acting cabotegravir injection). ^†Group BR includes infections that occurred with no recent cabotegravir exposure (>6 months after the last cabotegravir injection) in participants who restarted long-acting cabotegravir before the study site detected the infection. ^‡Group D includes infections that occurred despite on-time (<10 weeks between all) injections. §Group DX includes infections that occurred in participants who received at least one delayed cabotegravir injection (≥10 weeks between injections, but not meeting group B criteria).

Figure 2:. Study product adherence during the blinded phase, first unblinded year, and combined period

Pharmacokinetic adherence testing was performed on samples collected at weeks 4, 9, 17, 33, 57, 81, 105, 129, 153, and 177, and on a quarterly schedule for 48 weeks and then annually if participants transitioned early to open-label tenofovir disoproxil fumarate plus emtricitabine. The median number of plasma and dried blood spots samples per participant was 5 (IQR 4–7) during the updated primary blinded study, 2 (1–2) during the first unblinded year, and 7 (5–8) during the combined period. The tenofovir diphosphate in dried blood bar shows the proportion of a randomly selected subset of the tenofovir disoproxil fumarate plus emtricitabine group participants with intraerythrocytic tenofovir diphosphate concentrations measured in dried blood spots in each adherence category; these data represent average dosing over the previous 1–2 months. The green boxed numbers above each histogram bar represent the aggregate of four or more doses per week; this dosing frequency is anticipated to provide high levels of rectal protection against HIV acquisition. Each participant selected for adherence testing could have up to eight samples included in this summary. Values for week 4 were adjusted for days on therapy, as steady-state drug concentrations were not yet achieved. The plasma tenofovir bar shows the proportion of a randomly selected subset of the tenofovir disoproxil fumarate plus emtricitabine group participants with plasma tenofovir in each adherence category; these data represent average dosing over the past 72 h. The blue boxed numbers above each histogram bar represent the aggregate of quantifiable tenofovir in plasma. Cabotegravir injection coverage was defined as an initial injection providing 6 weeks of anticipated protective coverage and subsequent injections providing 10 weeks of protective coverage. BLQ=below the limit of quantification. CAB=cabotegravir. DBS=dried blood spots. LLOQ=lower limit of quantification. TFV=tenofovir. TFV-DP=tenofovir diphosphate.

Figure 3:. Pharmacological and virological data for HIV infections in the oral tenofovir disoproxil fumarate plus emtricitabine group that were identified in the extended analysis

HIV genotyping results are shown for the first visit with a viral load greater than 500 copies per mL. HIV genotyping results are shown to the right of each bar. These include major mutations that confer resistance to nucleoside or nucleotide reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors, and protease inhibitors. BLQ=below the limit of quantification. LLOQ=lower limit of quantification. TFV-DP=tenofovir diphosphate.