The role of cohort studies in drug development: clinical evidence of antiviral activity of serotonin reuptake inhibitors and HMG-CoA reductase inhibitors in the central nervous system.

BACKGROUND: Effective antiretroviral therapy (ART) has reduced the incidence of HIV-associated neurocognitive impairment (HNCI) but its prevalence remains high. Clinical trials have yet to identify a consistently effective treatment for HNCI, other than ART, but in vitro data support that some drugs approved by the Food and Drug Administration (FDA) for other indications might benefit individuals with HNCI. Some of these drugs, such as serotonin reuptake inhibitors (SRIs) and HMG-CoA reductase inhibitors (statins), may do so by reducing HIV replication in the CNS and are already widely used by HIV-infected individuals.
METHODS: Six-hundred fifty-eight HIV-infected participants of the CHARTER cohort had a baseline assessment, which included comprehensive neuropsychological (NP) testing and HIV RNA measurements in plasma and cerebrospinal fluid (CSF). Four-hundred sixty-seven (71%) subjects used ART, 195 (30%) used SRIs, and 63 (10%) used statins.
RESULTS: SRI users were less likely to have HIV RNA levels in CSF above 50 copies (c)/mL (29 vs. 37% in non-SRI users, OR 0.69, p = 0.05). This association was most evident for three of the seven SRIs (citalopram, sertraline, and trazodone, or "antiviral" SRIs, combined 25 vs. 38% in non-SRI users, OR 0.56, p = 0.01) and was strongest in those not taking concomitant ART (61 vs. 83%, OR 0.31, p = 0.01). "Antiviral" SRI users also performed better on NP tests (median global deficit score 0.37 vs. 0.47, p = 0.04). Statin users were also less likely to have HIV RNA levels in CSF above 50 c/mL (16 vs. 37%, p < 0.001) but, in contrast to SRIs, the association was strongest in those taking ART (2 vs. 18%, p < 0.001). Statin use was not associated with better NP performance. Multivariate analyses indicated that the use of "antiviral" SRIs-but not statins-was associated with undetectable HIV RNA levels in CSF and better NP performance.
CONCLUSIONS: SRIs may reduce HIV replication in CSF and improve NP performance. This was particularly true for three SRIs-supporting differences in antiviral efficacy between drugs-in individuals who were not taking ART. In contrast, statins were not associated with lower HIV replication in CSF in multivariate analyses and were not associated with better NP performance. These analyses support the value of large observational cohort studies in identifying FDA-approved drugs that may be worth further investigation.