CSF Neurofilament Levels Predicts Neurological Deterioration

Brian Hoyle

March 04, 2011

March 4, 2011 (Boston, Massachusetts) — In a study designed to explore the potential of several molecules as biomarkers for HIV-mediated cognitive impairment, researchers at John Hopkins University, in Baltimore, Maryland, have documented "surprisingly high levels" of neuronal heavy chain neurofilament (neurofilament H) in the cerebrospinal fluid (CSF) of HIV-infected patients who were cognitively normal.

Because neurofilament H levels in CSF are indicative of neuron damage, the use of this stable protein might provide a biomarker for impending cognitive decline and point the way to early initiation of neuroprotective treatments, researchers announced here at the 18th Conference on Retroviruses and Opportunistic Infections.

"Currently, there are no surrogate markers for HIV-associated neurocognitive disorders. However, there is a critical need for such markers, not only for the purpose of diagnosis, but also for monitoring the effects of treatment and for use in clinical trials. If we had reliable surrogate markers, we could conduct clinical trials in smaller numbers of patients over shorter periods of time. This could result in substantial cost savings and will give us the opportunity to screen many more therapeutic agents. CSF neurofilament may be one such surrogate marker," Avindra Nath, MD, professor of neurology at Johns Hopkins University, told Medscape Medical News.

The study was prompted by recent reports indicating that more than half of HIV-infected people whose virological status is being controlled by antiretroviral therapy display HIV-associated neurocognitive disorders. To minimize cognitive decline, sensitive surrogate markers of neuronal injury are needed, Dr. Nath noted.

In seeking a marker, the investigators obtained CSF from 76 HIV-infected individuals with varying degrees of neurocognitive impairment, from 9 patients with multiple sclerosis, and from 9 patients with other neurological disorders, including pseudotumor cerebri and normal pressure hydrocephalus.

The samples were analyzed for neurofilament H, protein carbonyl, and nitrotyrosine using Western blot–utilizing antineurofilament H antisera and a commercial slot-blot-based detection kit with the appropriate antibody. Each protein was quantified in each sample using infrared imaging.

The quantity of neurofilament H was significantly higher in CSF from all of the HIV-positive patients than from individuals with other neurological deficits and HIV-infected individuals who tested cognitively normal, Dr. Nath reported.

The elevation in neurofilament H levels was quantitatively similar in HIV-infected individuals whose cognition was assessed as normal, cognitively affected but asymptomatic, and mildly impaired. In those with HIV-associated dementia, the levels were marginally lower, but still elevated, than in patients with multiple sclerosis and other neurological disorders.

To determine the significance, if any, of the marginal reduction will require an examination of greater numbers of subjects, Dr. Nath cautioned.

No significant correlation was evident with age (30 to ~60 years) or levels of neurofilament H or nitrotyrosine; however, protein carbonyl levels increased with age.

The determination of neurofilament H levels 6 months apart, accompanied by cognitive testing, revealed that individuals with declining cognitive function had the highest CSF levels of neurofilament H, relative to those whose cognitive function was unchanged or improved.

The presence of neurofilament in CSF indicates that the neurons are being injured. This protein is only produced by neurons; hence, its presence in CSF is indicative of neuronal injury, Dr. Nath explained. "The surprising finding here is that we found evidence of elevated neurofilament in the HIV-infected patients who had normal or symptomatic neurocognitive impairment."

"This has very important implications. It means that neuronal injury occurs very early in the course of the illness. Hence, we should consider therapeutic interventions very early as well. This is also supported by the observation that patients who had further increases in neurofilament levels 6 months later appeared to have a decrease in neurocognitive function," Dr. Nath told Medscape Medical News.

These findings might spur a rethink of the design of clinical trials in this population.

"In the past, we have targeted patients with much more severe cognitive impairment for such studies. However, we should try to conduct such trials much earlier in patients with much milder cognitive impairment. It might also be possible to stratify these patients based on their CSF neurofilament levels for such studies," Dr. Nath said.

"Studies such as this hearken back to the debate surrounding active vs inactive disease — in this case with respect to cognition. Cognitively asymptomatic individuals are a conundrum," Ronald J. Ellis, MD, from the University of California at San Diego, told Medscape Medical News.

The study analyses were carried out to avoid this pitfall, according to Dr. Nath.

Dr. Nath and Dr. Ellis have disclosed no relevant financial relationships.

18th Conference on Retroviruses and Opportunistic Infections (CROI): Abstract 407. Presented March 2, 2011.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.