Article In Brief
Elevated levels of neurofilament protein in cerebrospinal fluid appeared to be a biomarker for subsequent development of mild cognitive impairment (MCI) among people who had normal cognition when enrolled and tested. The risk of developing MCI was more than three times higher for those with high levels of NfL compared with those with low levels.
Neurofilament light (NfL) protein has been the focus of many studies in the past few years as a potential biomarker for early neurodegeneration in a wide range of neurologic conditions. Now a new analysis from the community-based Mayo Clinic Study of Aging adds one more condition to the list — mild cognitive impairment (MCI).
The study found that elevated NfL in the cerebrospinal fluid (CSF) was a good predictor of MCI among people who had normal cognition when enrolled and tested. The risk of developing MCI was more than three times higher for those with high levels of NfL compared with those with low levels.
The study, which was reported November 12 in JAMA Neurology, also found that while NfL was a good predictor of MCI, other biomarkers for neurodegeneration, including total tau (T-tau), phosphorylated tau (P-tau), and neurogranin (Ng), were not.
The ability of NfL to predict risk for MCI appeared to occur independently of a person's CSF amyloid-beta 42 (AB42) status, a finding that suggests that “NfL is a risk factor for cognitive impairment both for those on the pathway to Alzheimer's disease and those not on the AD pathway,” the researchers reported. But NfL's apparent nonspecificity for AD does not mean it doesn't have prognostic value; rather, NfL could provide different or additional information that could be useful in the assessment of cognitive decline, the researchers surmised.
“I think the main message is that CSF NfL is a promising prognostic biomarker for neurodegeneration,” said the study's senior author Michelle M. Mielke, PhD, professor of epidemiology and neurology at the Mayo Clinic in Rochester, MN. “This is not yet ready for the clinic, but we think that NfL might be a better measure than CSF tau as a marker of neurodegeneration.”
Having a blood test to measure for NfL might be preferable to CSF testing in real-life practice, but that too is only in the research stage. Dr. Mielke said her team will soon present a related study, with a smaller sample of study participants, that compares both CSF and plasma NfL for predicting cognitive decline and brain atrophy.
NfL is a presumed biomarker of axonal degeneration, making it of interest to researchers investigating a number of neurodegenerative conditions, including MCI, frontotemporal dementia, AD, motor neuron disease, concussion, and parkinsonian disorders. Previous research has indicated that elevated CSF NfL is associated with greater risk of progression from MCI to AD dementia or further cognitive decline, but its role as a prognostic marker has not been fully assessed in people with normal cognition, the authors of the current study said.
The Mayo Study on Aging
The Mayo Study on Aging is a prospective, population-based study of MCI involving residents of Olmsted County, MN. The new analysis includes a subset of 648 participants without cognitive impairment who were enrolled in the study between 2004 and 2015. They had a median age of 72.3 and all had CSF testing results and at least one follow-up visit. The median follow-up was 3.8 years.
The analysis aimed to determine whether CSF NfL and Ng were associated with risk of MCI and how CSF NfL or Ng compared with CSF T-tau, P-tau, or AB42. During the course of the study, 14.8 percent of participants developed MCI.
“There were no associations between CSF Ng, T-tau, or P-tau and risk of MCI,” the researchers reported. “The combination of elevated CSF NfL and T-tau levels for risk of MCI was no better than CSF NfL alone. As expected, low CSF AB42 levels were associated with an increased risk of MCI. However, the associations between CSF NfL and risk for MCI were independent of CSF AB42.”
The researchers concluded: “Taken together, our findings suggest that CSF is a valuable biomarker of early neurodegeneration but not specific for AD pathology.” NfL may be the preferred “N” marker in the new A/T/N (amyloid/tau/neurodegeneration) biomarker classification research framework proposed for Alzheimer's disease, they said.
The study's strengths included its community-based population and the large number of cognitively normal participants who were followed, the researchers noted.
There were also limitations in the study: CSF markers were measured at only one point, meaning the researchers could not determine whether changes in the biomarker levels at different times correlated with cognitive decline. They also did not have a large enough sample to differentiate between cognitive subsets of MCI.
Many questions remain unanswered, Dr. Mielke said, including how NfL measurements would be used to help with clinical decision-making around cognitive decline and Alzheimer's disease.
“Among individuals with elevated brain amyloid, it is important to identify who will develop cognitive impairment and how fast,” Dr. Mielke said. “Because NfL is an indicator of neurodegeneration, individuals with both elevated brain amyloid and NfL may identify patients on the Alzheimer's disease track [who] will decline the fastest. This is important for enrollment into clinical trials and also for family planning purposes.”
“Another use of NfL could be to track disease progression to determine whether a specific therapy is slowing the fate of neurodegeneration for an individual patient,” she said.
Other Biomarker Research
Anne Fagan, PhD, professor of neurology at Washington University School of Medicine in St. Louis, said the Mayo report “is terrific and is very consistent with what others have been reporting on NfL. What makes this paper unique is that it is a population-based cohort. There is no cherry-picking; the participants are from the community.”
Dr. Fagan, whose research focuses primarily on CSF biomarkers of AD, said, “I think it's too soon to consider using NfL in the clinic since there are still too many unknowns in terms of the science as well as the assay performance.”
But the marker is important for basic research into disease pathology — what it is “marking” — and potentially for the design of clinical trials utilizing tau-independent outcome measures of neuronal injury, she said.
“The question becomes, ‘Is NfL better than another marker or redundant with another marker of injury such as tau or VILIP-1 or synaptic markers?’” Dr. Fagan said. Different biomarkers may also be more informative at different stages of the disease. It is noteworthy, she said, that levels of NfL in both CSF and plasma have been reported to be elevated early in the disease, before the appearance of dementia symptoms, a time when drug intervention might be most effective.
“For scientists like myself who are trying to better understand the pathophysiological course of the disease, being able to identify markers that differ even slightly in their temporal patterns is helpful,” she said.
Testing for NfL or other biomarkers might someday prove useful in determining whether a person's slips in thinking or memory are due to neurodegeneration or non-neurodegenerative issues such as depression, medications, or obstructive sleep apnea, said Suzanne Schindler, MD, PhD, assistant professor of neurology at Washington University in St. Louis.
Dr. Schindler said that pinpointing the cause of cognitive decline is not necessarily simple. “Often there is more than one problem contributing,” she said, “and it can be difficult to sort out what is causing cognitive symptoms.”
CSF NfL may be offering information different from other well-established CSF biomarkers, Dr. Schindler said. “CSF total tau, phospho-tau, and Abeta42 reflect Alzheimer disease pathology that is present in the brain for many years before onset of dementia. In contrast, it appears that NfL becomes abnormal closer to the onset of symptoms.”
Dr. Schindler, who collaborates with Dr. Fagan, has been evaluating other so-called “emerging” CSF biomarkers of AD in a cohort of persons with autosomal dominant AD. That analysis has found that Ng, SNAP-25, VILIP-1, and YKL-40 become abnormal about 15 years before the onset of symptoms and may be redundant with t-tau and p-tau in diagnostic testing, Dr. Schindler said.
It will be important to evaluate whether adding new biomarkers such as NfL to established biomarkers will improve diagnosis of disease, she said.
Douglas Galasko, MD, professor in the department of neurosciences at University of California, San Diego (UCSD) and an investigator at the UCSD Shiley-Marcos Alzheimer's Disease Research Center, said that the Mayo Clinic study may “have implications for research studies, because lumbar punctures would not routinely be performed on people with normal cognition unless they were part of a study.”
“It is now possible to measure plasma levels of A-beta42, A-beta40, and NfL with great precision, and it will be interesting to see whether combinations of these markers could pave the way to screening blood tests that might support Alzheimer-prevention studies in future,” he said.