Safety and tolerability of long-acting cabotegravir injections in HIV-uninfected men (ECLAIR): a multicentre, double-blind, randomised, placebo-controlled, phase 2a trial

Background: Cabotegravir (GSK1265744) is an HIV-1 integrase strand transfer inhibitor with potent antiviral activity and a long half-life when administered by injection that prevented simian-HIV infection upon repeat intrarectal challenge in male macaques. We aimed to assess the safety, tolerability, and pharmacokinetics of long-acting cabotegravir injections in healthy men not at high risk of HIV-1 infection.

Methods: We did this multicentre, double-blind, randomised, placebo-controlled, phase 2a trial at ten sites in the USA. Healthy men (aged 18-65 years) deemed not at high risk of acquiring HIV-1 at screening were randomly assigned (5:1), via computer-generated central randomisation schedules, to receive cabotegravir or placebo. Participants received oral cabotegravir 30 mg tablets or matching placebo once daily during a 4 week oral lead-in phase, followed by a 1 week washout period and, after safety assessment, three intramuscular injections of long-acting cabotegravir 800 mg or saline placebo at 12 week intervals. Study site staff and participants were masked to treatment assignment from enrolment through week 41 (time of the last injection). The primary endpoint was safety and tolerability from the first injection (week 5) to 12 weeks after the last injection. We did analysis in the safety population, defined as all individuals enrolled in the study who received at least one dose of the study drug. This study is registered with ClinicalTrials.gov identifier, NCT02076178.

Findings: Between March 27, 2014, and Feb 23, 2016, we randomly assigned 127 participants to receive cabotegravir (n=106) or placebo (n=21); 126 (99%) participants comprised the safety population. Most participants were men who have sex with men (MSM; n=106 [83%]) and white (n=71 [56%]). 87 (82%) participants in the cabotegravir group and 20 (95%) participants in the placebo group completed the injection phase. Adverse events (n=7 [7%]) and injection intolerability (n=4 [4%]) were the main reasons for withdrawal in the cabotegravir group. The frequency of grade 2 or higher adverse events was higher in participants in the long-acting cabotegravir group (n=75 [80%]) than in those in the placebo group (n=10 [48%]; p=0·0049), mostly due to injection-site pain (n=55 [59%]). No significant differences were noted in concomitant medications, laboratory abnormalities, electrocardiogram, and vital sign assessments. Geometric mean trough plasma concentrations were 0·302 μg/mL (95% CI 0·237-0·385), 0·331 μg/mL (0·253-0·435), and 0·387 μg/mL (0·296-0·505) for injections one, two, and three, respectively, indicating lower than predicted exposure. The geometric mean apparent terminal phase half-life estimated after the third injection was 40 days. Two (2%) MSM acquired HIV-1 infection, one in the placebo group during the injection phase and one in the cabotegravir group 24 weeks after the final injection when cabotegravir exposure was well below the protein-binding-adjusted 90% inhibitory concentration.

Interpretation: Despite high incidence of transient, mild-to-moderate injection-site reactions, long-acting cabotegravir was well tolerated with an acceptable safety profile. Pharmacokinetic data suggest that 800 mg administered every 12 weeks is a suboptimal regimen; alternative dosing strategies are being investigated. Our findings support further investigation of long-acting injectable cabotegravir as an alternative to orally administered pre-exposure prophylaxis regimens.

Funding: ViiV Healthcare.