Time course of cerebrospinal fluid responses to antiretroviral therapy: evidence for variable compartmentalization of infection

OBJECTIVES

To compare the kinetics and magnitude of HIV-1 RNA responses to antiretroviral therapy (ART) in the cerebrospinal fluid (CSF) and plasma.

DESIGN

Repeated lumbar punctures (LPs) were performed after the initiation or change in ART in 15 HIV-1-infected subjects, with the focus on two phases of response: an acute phase within the first 11 days, for which crude estimates of viral RNA half-lives and decay rates were derived and CSF:plasma relative decay ratios quantitatively analysed; and a longer-term phase beyond 4 weeks that was descriptively assessed.

RESULTS

In 13 subjects studied during the acute phase, the crude HIV-1 RNA half-life was longer (median 2.0 compared with 1.9 days), the decay rate slower (median 0.13 compared with 0.16 log10 copies/day) and, most notably, the variability greater (intraquartile range of half-life 1.8-4.3 compared with 1.7-2.1 days) in the CSF than in the plasma. A slower decay in the CSF correlated with lower initial blood CD4 T lymphocyte counts (P = 0.001). Seven of 11 subjects studied at 4 weeks or later, including some with slower acute-phase CSF responses, showed greater or more durable viral suppression in the CSF.

CONCLUSION

Divergent acute-phase viral kinetics in the CSF and plasma, and proportionally greater long-term decrements in CSF HIV-1 RNA in slow early-responders or poor overall plasma responders indicate variable compartmentalization of CSF infection, consistent with a model of two prototypes of CSF infection: short-lived, transitory infection that predominates in early HIV-1 infection and longer-lived, more autonomous CSF infection predominating in late HIV-1 infection. Additional studies will be needed to define more precisely the acute and longer-term CSF kinetics in different clinical settings and to assess this model.

Effects of chronic zidovudine administration on CNS function and virus burden after perinatal SIV infection in rhesus monkeys

Continuous intravenous administration of zidovudine (AZT) has been reported to improve cognitive function in HIV-infected pediatric patients (Pizzo et al., 1988). The effects of long-term zidovudine treatment in the perinatally infected pediatric population, including antiviral efficacy and effects on cognitive and motor function has not been systematically examined. These questions were addressed in rhesus macaque infants infected at birth with SIVSMM/B670, a primate model for infantile HIV infection and disease (Eiden et al., 1993a). Continuous or intermittent administration of AZT during the first 6 months following infection resulted in about a doubling of lifespan, a delay in the occurrence of motor impairment, and lower virus burden and quinolinic acid levels in cerebrospinal fluid (CSF) following administration of the antiviral drug.