NEW YORK (360Dx) – Researchers with the Dominantly Inherited Alzheimer Network (DIAN) have found that serum neurofilament light chain (Nfl) levels could aid in diagnosing and monitoring the progression of Alzheimer's disease.
In a study published this week in Nature Medicine, the scientists determined that measuring the rate of change in patient Nfl levels allowed them to distinguish carriers of a mutation linked to familial Alzheimer's more than 16 years before the onset of symptoms.
While Nfl is not specific to Alzheimer's disease, the study indicates that it could be a useful marker for identifying patients likely to develop the condition when used in combination with conventional Alzheimer's markers like amyloid-β or tau, said Mathias Jucker, a professor at the German Center for Neurodegenerative Diseases (DZNE) and senior author on this study.
Nfl is a subunit of the neurofilament proteins that are part of the neuronal cytoskeleton. Changes in Nfl levels in have been linked to a variety of neurological conditions, including multiple sclerosis and amyotrophic lateral sclerosis. As the authors noted, improved assays for Nfl have enabled measurements of the molecule in blood, with researchers finding good correlation between patient blood and cerebrospinal Nsf levels, suggesting that the molecule could prove useful as a blood-based biomarker for neurodegenerative diseases including Alzheimer's.
Blood-based biomarkers for Alzheimer's are an area of great interest, particularly among researchers and drug companies running clinical trials for Alzheimer's drugs. Pharma has had little success tackling the disease to date, and one line of thinking holds that his is due in part to the fact that many trials are done in patients too late in the disease process. Thus, firms are looking for better ways to select early-stage patients for trials and to track their response to therapies.
Blood biomarkers could offer a less expensive and more easily accessible way of identifying early-stage Alzheimer's patients than cerebrospinal fluid and imaging markers currently used.
In the Nature Medicine study the researchers used samples and data from the DIAN, a collection of patients from families carrying mutations in the amyloid beta precursor protein (APP) or presenilin 1 (PSEN1) or 2 (PSEN2) gene, which are strongly associated with Alzheimer's. The study authors noted that because "the age of symptom onset tends to be consistent for a given mutation, it is possible to calculate for participants an estimated years-to-symptom onset (EYO) from the known onset of individuals with the same mutation."
Using Quanterix's Simoa technology, Jucker and his colleagues measured Nfl levels in the CSF and serum of 243 mutation carriers and 162 controls (family members without either of the two mutations), finding that in both CSF and serum Nfl levels were significantly higher in mutation carriers than controls. This elevation was evident in both fluids at an estimated 6.8 years before the onset of symptoms.
The researchers then did a longitudinal analysis looking at 196 subjects in whom samples were collected at between one and five additional times after the baseline sampling. Using this data they calculated the rate of change in Nfl levels for these subjects and found that this rate of change provided an even earlier indication of disease activity, distinguishing mutation carriers from non-carriers at an estimated 16.2 years before the onset of symptoms, almost a decade earlier than absolute Nfl levels alone.
Looking at Nfl levels and rates of change in subjects who were in the process of converting from asymptomatic to symptomatic, the researchers found that rates of change leveled off as subjects became symptomatic, at which point absolute Nfl values appeared to be more informative regarding disease progression.
Jucker said that measuring Nfl in combination with established Alzheimer's markers like Aβ could help better select patients for clinical trials. Increased Aβ levels are apparent in Alzheimer's patients even before altered Nfl levels, Jucker said, but while increased Aβ is necessary for the development of Alzheimer's, not all subjects with elevated Aβ will develop the disease. By actually providing a measure of the neurodegeneration trigged by the Aβ deposits, Nfl could help researchers more accurately determine which subjects actually have Alzheimer's as well as better assess the stage of their disease.
While the research was conducted in a population of patients with familial disease, Jucker said he believed it would be applicable to sporadic Alzheimer's patients, as well.
He noted that other than the earlier presentation of the familial version of the disease, the familial and sporadic forms are essentially the same. He added that smaller studies have made similar findings in subjects with sporadic Alzheimer's.
"I think the evidence is overwhelming that it will work" in sporadic patients, he said.
He allowed, though, that the older age of the sporadic patients could complicate measurements compared to younger patients with familial disease.
"In sporadic [patients], people who are 80 or 90 years old… maybe they will have multiple sclerosis or they will have had a stroke, and so their Nfl levels will also go up" as a result of those conditions, he said. "However, if you have a stroke, for instance, then four or five months later your levels should be down again. So it may be even more important to take the rate of change compared to the absolute levels in sporadic Alzheimer's disease."
Jucker said he was now interested in better determining the most informative windows for measuring rates of Nfl change. He said he was working with the DIAN to get additional longitudinal samples for that work.
Jucker said he was also interested in combining Nfl with blood-based measurements of other Alzheimer's markers. He noted that one of the Nature Medicine coauthors, Randall Bateman, professor of neurology at Washington University in St. Louis and founder of Alzheimer's diagnostics firm C2N diagnostics, has developed assays for measuring levels and kinetics of different Aβ isoforms in plasma.
In a 2017 study in Alzheimer's and Dementia, Bateman and colleagues used immunoprecipitation combined with mass spec to assess the kinetics of three Aβ isoforms, Aβ38, Aβ40, and Aβ42, in plasma, finding that in patients with brain amyloidosis, which is strongly linked to Alzheimer's disease, plasma Aβ42 showed faster turnover and lower concentrations than Aβ38 and Aβ40.
"It would be a big breakthrough to combine [the Nfl measurements] with Aβ measurements in the blood like what [Bateman] does," Jucker said. "So we would have Aβ in the blood and Nfl in the blood, two extremely important readouts. I think that is where we all want to go."
He said that he believed Nfl was already drawing interest among drug firms doing clinical trials, noting that while the Nature Medicine publication was only out this week, the results have been circulating through the research community for some time now.
"I'm sure that in every [Alzheimer's] clinical trial that is being designed or has been designed in the last six months, Nfl is going to be looked at," he said.
Quanterix declined to say whether it was aware of any specific clinical trials using its Simoa technology to measure Nfl in blood, but CEO Kevin Hrusovsky noted that the company "has placed instruments at 23 of the 25 largest pharmaceutical companies, including all of those with significant drug development programs in neurology."
The company currently offers a 4-plex panel for neurology research that includes Nfl, tau, GFAP, and UCH-L1, he added.