Pharmacokinetic approaches to standardize antiviral exposure in cerebrospinal fluid.

Objectives: HIV has been shown to persist in the central nervous system (CNS) in persons on antiretroviral therapy (ART). Our objective was to use pharmacokinetic (PK) modeling to estimate cerebrospinal fluid (CSF) exposure from time-variant concentrations of various antiretrovirals of ART regimens and to standardize CSF metrics, including maximum concentration [C_MAX], area under the curve [AUC], and trough [C_Trough].

Methods: Advancing Clinical Therapeutics Globally (ACTG) A5321 is a prospective cohort study of HIV-1 reservoirs in persons with HIV. Plasma and CSF antiretroviral (ARV) concentrations were measured in 74 participants who were receiving ART. PK modeling (Pmetrics) was performed for nine ARVs. Relative CSF penetration for each ARV was estimated by comparing CSF C_MAX and AUC to plasma C_MAX and AUC (i.e., C_MAXmethod and AUC_method). The CSF C_Trough for each ARV was compared with in vitro literature values of HIV inhibitory concentration values (IC_{50, 90, or 95}).

Results: Emtricitabine exhibited the highest median relative CSF penetration (C_MAXmethod, 46.3%; AUC_method, 72%) and dolutegravir had the lowest CSF penetration (C_MAXmethod, 0.57%; AUC_method, 0.57%). Tenofovir, lamivudine, atazanavir, and raltegravir had median estimated CSF C_Trough concentrations less than IC_{50, 90, or 95}. Interparticipant variability of relative CSF penetration based on exposures ranged from 160% for lamivudine to approximately 9% for dolutegravir.

Conclusions: PK modeling successfully standardized ARV CSF concentrations to a given time point (i.e., C_MAX or C_Trough) to allow estimation of CSF penetration. This approach provides uniformity for the assessment of exposure, for the estimation of whether desired therapeutic drug goals are obtained in the CSF, and for further studies to investigate whether CSF exposure metrics calculated using this method are associated with measures of HIV persistence.