Pharmacokinetics Simulations for Studying Correlates of Prevention Efficacy of Passive HIV-1 Antibody Prophylaxis in the Antibody Mediated Prevention (AMP) Study

TL;DR

This study uses pharmacokinetics simulations to assess how drug concentration measurements relate to HIV prevention efficacy in clinical trials, focusing on sampling design, adherence, and model accuracy.

Contribution

It introduces a simulation framework to evaluate sampling strategies and adherence effects on PK model estimates in HIV prophylaxis trials.

Findings

Reliable PK estimates are possible with 120 samples per participant.

Sampling schedule coarsening still yields acceptable estimates.

Adherence levels impact the accuracy of PK model parameters.

Abstract

A key objective in two phase 2b AMP clinical trials of VRC01 is to evaluate whether drug concentration over time, as estimated by non-linear mixed effects pharmacokinetics (PK) models, is associated with HIV infection rate. We conducted a simulation study of marker sampling designs, and evaluated the effect of study adherence and sub-cohort sample size on PK model estimates in multiple-dose studies. With m=120, even under low adherence (about half of study visits missing per participant), reasonably unbiased and consistent estimates of most fixed and random effect terms were obtained. Coarsened marker sampling schedules were also studied.