Bayesian estimation of HIV acquisition dates for prevention trials.

Area of Science:

• Virology
• Immunology
• Biostatistics

Background:

• Accurate estimation of HIV acquisition timing is essential for HIV prevention trials to determine antibody levels at the time of infection.
• The Antibody-Mediated Prevention (AMP) Studies demonstrated the efficacy of broadly neutralizing antibodies in preventing HIV acquisition.

Purpose of the Study:

• To develop and evaluate a pipeline for estimating the date of detectable HIV acquisition (DDA) using diagnostic and viral sequence data.
• To improve the accuracy of HIV acquisition timing estimates in participants of HIV prevention trials.

Main Methods:

• Developed a Bayesian strategy combining REN sequence, GP sequence, and diagnostic data streams.
• Evaluated the pipeline's performance using PacBio viral sequence data from 41 participants in the FRESH and RV217 acute HIV acquisition cohort studies.
• Defined "true DDA" as the midpoint between last-negative and first-positive RNA diagnostic tests.

Main Results:

• Diagnostic data alone resulted in a bias of 2.4 days and RMSE of 7.9 days.
• Combining sequence and diagnostic data improved estimates (bias 1.5 days, RMSE 6.9 days).
• Restricting sequence-based estimation to samples within 5 weeks post-DDA further reduced bias to 0.2 days (RMSE 7.8 days).

Conclusions:

• The developed pipeline enhances the accuracy of HIV acquisition timing estimates.
• Sequence-based estimation, particularly with samples collected within 5 weeks post-DDA, offers improved precision.
• Monthly sampling frequency is suggested for future clinical trials using sequence-based acquisition timing estimation.