CSF quinolinic acid levels are determined by local HIV infection: cross-sectional analysis and modelling of dynamics following antiretroviral therapy.

Quinolinic acid (QUIN) is a product of tryptophan metabolism that can act as an endogenous brain excitotoxin when released by activated macrophages. Previous studies have shown correlations between increased CSF QUIN levels and the presence of the AIDS dementia complex (ADC), a neurodegenerative condition complicating late-stage human immunodeficiency virus type 1 (HIV) infection in some patients. CSF QUIN is putatively one of the important molecular mediators of the brain injury in this clinical setting and, more generally, serves as a marker of local macrophage activation. This study was undertaken to examine the relationship of CSF QUIN concentrations to local HIV infection and to define the effects of antiretroviral drug treatment on CSF QUIN using two complementary approaches. The first was an exploratory cross-sectional analysis of a clinically heterogeneous sample of 62 HIV-infected subjects, examining correlations of CSF QUIN levels with CSF and plasma HIV RNA levels and other salient parameters of infection. The second involved longitudinal observations of a subset of 20 of these subjects who initiated new antiretroviral therapy regimens. In addition to descriptive analysis, we used kinetic modelling of QUIN decay in relation to that of HIV RNA to assess further the relationship between CSF QUIN and infection in the dynamic setting of treatment. The cross-sectional studies showed strong correlations of CSF QUIN with both CSF HIV RNA and blood QUIN levels, as well as with elevations in CSF white blood cells, CSF total protein and CSF:blood albumin ratio. In this group of subjects with a low incidence of active, untreated ADC, CSF QUIN did not correlate with ADC stage or measures of quantitative neurological performance. Antiviral treatment reduced the CSF QUIN levels in all the longitudinally followed, treated subjects. Kinetic modelling of CSF QUIN decay indicated that CSF QUIN levels were driven primarily by CSF HIV infection with a lesser contribution from blood QUIN levels. In three subjects with new-onset, untreated ADC, CSF QUIN decay paralleled both CSF HIV decrement and improvement in neurological performance. These studies show that CSF QUIN concentrations relate primarily to active CSF HIV infection and to a lesser extent to plasma QUIN. CSF QUIN serves as a marker of local infection with a wide dynamic range. The time course of therapy-induced changes links CSF QUIN to local infection and supports the action of antiviral therapy in ameliorating immunopathological brain injury and ADC.