Abstract

Bevirimat [3-O-(3',3'-dimethylsuccinyl)betulinic acid] is the first in a new class of anti-human immunodeficiency virus (HIV) drugs that inhibit viral maturation by specifically blocking cleavage of the Gag capsid (CA) precursor, CA-SP1, to mature CA protein, resulting in defective core condensation and release of immature noninfectious virions. Four cohorts of six HIV-infected adults, with CD4 counts of >200 and plasma viral loads of 5,000 to 250,000 transcripts/ml and not currently receiving antiretroviral therapy, were randomized to receive a single oral dose of placebo, 75, 150, or 250 mg of bevirimat. Thirty blood samples for drug concentrations and 20 HIV RNA measures were collected from each subject over a 20-day period. Candidate pharmacokinetic/pharmacodynamic models were fit to individual subjects by maximum likelihood followed by Bayesian estimation; model discrimination was by corrected Akaike's Information Criterion. The bevirimat pharmacokinetics was well described by an oral two-compartment linear model (r(2), 0.98), with a mean (percent coefficient of variation) half-life of 60.3 (13.6) h and apparent oral clearance of bevirimat from the plasma compartment of 0.17 (18) liters/h. HIV RNA was modeled as being produced in infected CD4 cells, with bevirimat inhibiting infection of new CD4 cells thru a Hill-type function (r(2), 0.87). Single oral doses of bevirimat were well tolerated and demonstrated a dose-dependent reduction in viral load. The average maximum reduction from baseline following the 150- and 250-mg doses was greater than 0.45 log(10), with individual patients having reductions of greater than 0.7 log(10). No bevirimat resistance mutations were detected during the course of the study.