On HIV-1 infection in the central nervous system

Abstract

HIV-1 (hereafter referred to as HIV) is a neurotropic virus that invades the central nervous system (CNS). The CNS is then susceptible to an acute infection and later to CNS opportunistic diseases and AIDS dementia complex (ADC). HIV does not infect the neurons, and neuropathogenesis is largely mediated by the immune system. A cerebrospinal fluid (CSF) assay cannot predict which patient will develop dementia, but when established, immune markers and CSF viral load are generally elevated. It would be of great value to identify patients at risk before the development of dementia, either with a CSF marker or another test. Such markers would also provide insight into pathogenesis and provide a means of studying antiretroviral treatment effects on CNS HIV infection. We compared 41 HIV-infected patients to 41 controls in three reaction time tests in order to determine whether HIV-infected patients without AIDS or dementia have cerebral dysfunction, as reflected by impaired reaction times. HIV-infected patients were slower to react than controls but the results were within the normal range. However, 15% of all HIV patients had clearly pathological values on two or three tests, indicating that a subgroup of patients without AIDS has CNS involvement early in the course of infection, while the majority performs normally. For the development of drug resistance to be avoided it is crucial to repress HIV replication. We examined the effect of highly active antiretroviral treatment (HAART) in 74 antiretroviral-naïve patients where lumbar punctures (LPs) had been performed before and after treatment initiation. After three months on HAART, 85% of the patients had HIV RNA below 400 copies/mL in CSF, and 80% of the patients had reached below this level in plasma. This study shows that HAART is generally effective in reducing HIV RNA, not only in plasma but also in CSF. To explore the possibility of neuron-specific light chain neurofilament protein (NFL) as a marker of HIV-induced neurodegeneration and ADC, we analyzed CSF NFL in 210 HIV-infected patients. Patients with ADC had markedly elevated NFL concentrations, compared to neuroasymptomatic individuals, and levels correlated with the severity of ADC. NFL was also elevated in patients with opportunistic CNS diseases, and was moderately raised in patients with very low CD4 T-cell counts, indicating subclinical brain injury. We propose the use of NFL as a neuronal marker of HIV-related neurodegeneration that will also correlate with the severity of ADC. In order to evaluate the effect of HAART on NFL, we analyzed this marker in 53 patients before and after treatment initiation. At baseline, NFL was elevated in 21 patients, decreasing significantly at 3 months and one year on HAART. Thirty-two patients had normal values and all but one, who showed a temporary increase, remained normal on treatment. In 8 patients studied for 2 to 10 years on HAART, 7/8 showed sustained response with normal NFL during follow-up. These findings confirm the beneficial effect of HAART in CNS HIV infection, adding important information on pathogenesis and suggesting that NFL can be used to monitor CNS treatment effects.