Clinical Epidemiology of HIV

Neurocognitive complications of HIV remain common in spite of widespread use of HAART. Rippeth and colleagues (2004) noted a 38 percent prevalence of neurocognitive impairment in HIV+ HAART era patients who were not primary drug abusers. The presence of such impairment was a highly specific predictor of HIV encephalitis in those followed to autopsy (Cherner et al., 2002) underscoring the validity of neurocognitive diagnosis as reflecting HIV-associated processes in the brain. In addition, Langford et al. (2002) reported that some patients failing HAART developed a severe form of HIV-associated leukoencephalopathy that was observed only rarely in the pre-HAART era.

HIV Neuroadaptation and Neurovirulence

Previously, HNRC investigators observed that HIV populations in the CNS were genetically distinct from virus present in blood or lymphoid tissues (Wong et al., 1997). Recently, Strain and colleagues (2005) observed discordant resistance patterns between the CSF and plasma, and identified positions in C2-V3 as being associated with the presence of such discordant resistance, reflecting autonomous replication of HIV and the independent evolution of drug resistance in the CNS. Discordance was associated with severity of neurocognitive deficits, and having the amino acid serine at position five of the V3 loop was specifically associated with neurocognitive impairment (Pillai et al., 2006). These findings emphasize the importance of further studies on the genetics and molecular mechanisms of viral neuroadaptation on the one hand, and treatment implications on the other.

Host Genetics

HIV neuropathogenesis appears to involve combinations of toxicity of viral products such as tat and nef and other viral factors that disturb expression of host genes that control critical cellular functions including metabolism, trophic and apoptotic mechanisms, and chemokine and cytokine signaling, to name a few. HNRC investigators observed elevations of MCP-1 associated with neurocognitive impairment in some individuals with HIV, particularly in persons with a genetic variant (-2578G) of the MCP-1 promoter (Letendre et al., 2004). Thus, host genetic factors determining local chemokine expression in CNS tissues may explain some of the differences between subjects in susceptibility to HIV cognitive impairment.

Role of Co-factors

HIV infected persons in the US, other Western countries and the developing world frequently have comorbidities, such as abuse of methamphetamine and viral co-pathogens (eg, HCV). During the current funding period a series of investigations focused on the joint and separate effects of methamphetamine dependence on HIV-associated neurocognitive disturbance. Rippeth et al. (2004) noted that cognitive impairment was more prevalent in both methamphetamine and HIV positive individuals, and there was evidence for an additive effect in those who were dually affected. In a translational study, Langford et al. (2003) found that those dying with HIV and a history of methamphetamine addiction had greater loss of calbindin immunostaining interneurons than in those with either condition alone. Additionally, this study also demonstrated more severe microglial reactions in those with histories of methamphetamine dependence, suggesting a methamphetamine-associated immunologic stimulation. This notion received support from a finding that MCP-1 in the CSF is selectively elevated in methamphetamine abusers. Very recent work indicates that brain samples of persons dying with methamphetamine dependence exhibit upregulation of interferon inducible genes, providing a potential mechanism for methamphetamine-induced inflammation in the CNS (Everall et al., 2005). In the same work, Everall and colleagues found that several genes related to the cytoskeleton (synapsin II, MAP1B, PAK1, vimentin) are downregulated in brain samples from those dying with HIV dementia. These observations may provide a mechanism for our previous observations that dendritic simplification is a strong correlate of severity of HIV-associated neurocognitive impairment (Everall et al., 1999; Masliah et al., 1997).

Effects of Treatment on Neurocognitive Function

Despite HAART, some cognitively impaired individuals do not respond to treatment with improvement in cognitive function. To assess factors that might influence such recovery, Letendre and colleagues (2004) evaluated whether CSF drug penetration and CSF virologic suppression influenced the extent of neuropsychological improvement during ART. Subjects on regimens containing greater numbers of CSF penetrating drugs showed significantly greater reduction in CSF viral load. Additionally, those on drugs achieving greater suppression of CSF HIV levels were more likely to improve neuropsychologically. These data, combined with the observations on discordant resistance between CSF and plasma compartments point to the importance of further research on “CNS targeted” therapies.

Implications of Neurocognitive Impairment on Everyday Functioning

Although neurocognitive complications are generally not as severe in the HAART era as previously, these milder impairments substantially affect everyday functioning. For example, Marcotte and colleagues showed that both laboratory-assessed and on-road driving performance were worse in HIV positive people with neurocognitive impairment (Marcotte et al., 2004). Heaton et al. (2004) noted that such neurocognitive impairments were associated with reduced level of employment and increased difficulties in everyday functioning. In these studies, novel, objective measures of everyday functioning improved upon the prediction of “real world” functioning provided by neuropsychological testing alone. In addition, cognitive neuropsychological methods may also improve the ability to detect functional impairment (Woods et al., 2005). Such data are of substantial public health importance, given the transition of HIV into a chronic, perhaps lifelong disease in which treatment that substantially modifies medical morbidity does not necessarily translate into similar CNS protections.