Cerebrospinal fluid inhibitory quotients of antiretroviral drugs in HIV-infected patients are associated with compartmental viral control.

Abstract

Despite the efficacy of highly active antiretroviral treatment (HAART), a large proportion of human immunodeficiency virus (HIV)-infected patients may develop moderate neurocognitive impairment. Antiretroviral drug passage into the central nervous system may be relevant for preventing and treating HIV-associated neurocognitive disorder; nevertheless, clear cerebrospinal fluid (CSF) pharmacodynamic targets are not known. HAART-treated adults with wild-type HIV were prospectively enrolled. CSF concentrations (measured by mass spectrophotometric methods) and inhibitory quotients (CSF concentrations divided by in vitro 50% and 95% inhibitory concentrations) were compared among different drugs and related to CSF HIV RNA levels. CSF escape was defined as CSF HIV RNA >50 copies/mL despite contemporary plasma HIV RNA below that threshold. One hundred twenty-seven patients (91 male [71.7%], 93 white [73.2%], with a median age of 46 years [interquartile range, 40.5-54.5 years]) provided 174 paired CSF and plasma samples. Twice-daily darunavir, once-daily darunavir, and efavirenz had the highest CSF 95% inhibitory quotients (18.5, 8.2, and 6.4, respectively). Higher nadir CD4 cell count (P = .01) and plasma HIV RNA <50 copies/mL (P < .001) were independent predictors of controlled CSF HIV RNA. Optimal drug exposure (CSF detectable drugs and 95% inhibitory quotient >1) was protective for CSF escape (P = .01). Cerebrospinal fluid 95% inhibitory quotients may be used to compare antiretroviral drug compartmental exposure; they deserve longitudinal studies to assess the adequacy of CSF drug concentrations in treated HIV-infected patients.