Pharmacokinetics, pharmacodynamics, efficacy and drug resistance selection of injectable long-acting lenacapavir pre-exposure prophylaxis (PrEP) against HIV

Oral pre-exposure prophylaxis (PrEP) denotes an effective strategy to reduce the risk of HIV infection. However, many individuals encounter difficulties adhering to the once-daily regimen, which highlights the need for a broader portfolio of PrEP options. The novel HIV capsid inhibitor lenacapavir (LEN), when injected every six month, has shown potential in the recently completed clinical trials. However, clinical trials may not enable to accurately estimate prophylactic efficacy and protective concentration benchmarks. Moreover, since LEN may persist up to two years, there may be a risk for de novo resistance emergence after stopping PrEP.

We developed an integrated pharmacokinetic-pharmacodynamic (PK-PD) model of LEN and incorporated observed variability from Phase III clinical data into our analysis. The model was used to quantify prophylactic efficacy against wild type (WT) and resistant virus, as well as to quantify risks of de novo drug resistance emergence when LEN-PrEP is stopped.

We estimated a 95% preventive plasma concentration ( EC ₉₅ ) of 5.8ng/mL. LEN fully prevented infection with WT virus at plasma concentrations above 10ng/mL, which in an ‘average individual’ is achieved within 155hours after the first subcutaneous injection of 927mg and persists for up to 45weeks after the last injection. However, considering PK variability indicates that > 5.8 or > 10ng/mL are not surpassed at all times in all individuals. Full protection against mutant strains carrying the Q67H and N74D mutations was achieved at plasma concentrations of 35ng/mL and 85ng/mL, whereas LEN concentrations of 3095ng/mL, 218ng/mL, 100ng/mL provided full protection against variants carrying Q67H+N74D, Q67H+T107N and Q67H+N74S. The mutant selection window for N74D and all double mutants overlapped with steady-state concentrations for twice-yearly dosing. De novo resistance emergence is possible once LEN concentrations fall below 10ng/mL after the last injection and this temporal window lasts for ≈ 201, 105, 79, 96 and 86days for the Q67H, N74D, Q67H+N74D, Q67H+T107N and Q67H+N74S mutants, respectively, in an ‘average individual’.

Our results highlight a substantial risk for de novo drug resistance emergence when LEN SC injection are stopped, calling for strategies to manage LEN discontinuation.