ABSTRACT The long-term objective of the proposed work is to provide evidenced-based recommendations for minimizing metabolic adverse events, particularly weight gain, in a diverse population of people living with HIV (PLWH) by identifying and quantifying variability in drug exposure that may increase risk in patient subgroups. We will consider a broad array of modifying factors of drug exposure, including demographic characteristics, prior laboratory values, body anthropometrics, frailty phenotype, and pharmacogenomics. Our focus is on the integrase strand transfer inhibitors (INSTIs) dolutegravir (DTG) and bictegravir (BIC), as well as tenofovir alafenamide (TAF). The MWCCS includes diverse PLWH underrepresented in Phase III clinical trials and thus, in the sponsor-developed population pharmacokinetic models of the drug. In AIM 1, we will enroll diverse PLWH from four MWCCS sites to collect pharmacokinetic data and further refine the knowledge of factors that influence DTG, BIC, and TAF pharmacokinetics. Drug concentrations will be measured in blood plasma and peripheral blood mononuclear cells (TAF only) in the UNC Center for AIDS Research Clinical Pharmacology and Analytical Chemistry Laboratory (CFAR CPAC). Nonlinear mixed effects models will be used to develop a comprehensive PK model for each drug of interest. Using these models, then, in AIM 2, we will retrospectively measure drug concentrations in repository specimens (using the same methods and laboratory as AIM 1, and predict drug clearance in men and women from initiation of DTG, BIC, and/or TAF. These drug clearances, predicted for multiple visits per participant, will then be analyzed as the predictors of body weight gain, increased waist to hip ratio, and increased insulin resistance (as measured by HOMA-IR) over the course of treatment on the drug of interest, with multiple measures of drug exposure. The hypothesis is that those PLWH with higher drug exposure (via slower drug clearance) will be more likely to experience the metabolic adverse events of these drugs. Upon completion, we expect to have the underpinnings of a model-based risk estimator developed for further prospective validation.