Matrix metalloproteinase‐10 as a novel biomarker of neurodegeneration in Alzheimer’s disease

Abstract

AbstractBackgroundAlzheimer's disease (AD) is driven by cycles of protein misfolding, inflammation, and neurovascular and metabolic dysfunction. Although cerebrospinal fluid (CSF) tau and Amyloid‐β (Aβ) levels are strong indicators of AD pathology, they are inadequate for measuring disease progression or drug response. We investigated a collection of biomarkers that may represent active pathways in AD to identify novel targets for the diagnosis, staging, and pathophysiological profiling of AD.MethodImmunoassays (MesoScale Discovery, Quanterix) were used to measure a targeted selection of over 20 biomarkers in a well‐characterized cohort of 240 paired CSF and plasma samples obtained from the Penn ADCC and Center for Neurodegenerative Disease Research. The cohort was comprised of individuals with diverse diagnoses, including mild cognitive impairment (MCI), AD, amyotrophic lateral sclerosis (ALS), frontotemporal dementia (FTD), and cognitively unimpaired controls (CU‐N). Statistical modeling of AD severity, associations between biofluids, and differential diagnosis were performed. Regressions were adjusted for covariates, including age and sex.ResultCSF MMP‐10, FABP‐3, NfL, and YKL‐40 were significantly elevated in AD compared to CU‐N (p<0.005). MMP‐10, a regulator of blood‐brain barrier integrity, correlated significantly with CSF tau (p<0.005). The established neurodegeneration biomarkers FABP‐3 and NfL correlated significantly (p< 0.05) with CSF AD biomarkers. The neuroinflammatory marker YKL‐40 correlated significantly with CSF tau and ptau‐181 in both CSF and plasma (p<0.05). Linear regression identified a significant correlation between CSF and plasma MMP‐10 in CU‐N (p<0.005) and AD (p<0.005), but not in MCI. CSF MMP‐10 could distinguish CU‐N (AUC: 0.90), MCI (AUC: 0.71), and ALS (AUC: 0.78) from AD. Addition of FABP‐3, NfL, YKL‐40, and MMP‐10 to a logistic model increased classification accuracy between AD and CU‐N in both CSF (AUC: 0.99) and plasma (AUC: 0.788).ConclusionDevelopment of a broad biomarker panel may help clarify the interrelated pathophysiologies active in AD onset and progression. Here we highlight MMP‐10 as a novel biomarker that could provide insight on underlying vascular pathology contributing to AD pathogenesis. MMPs have also been implicated in Aβ‐induced proinflammatory response and tau aggregation. Further studies are warranted to assess the physiological significance of MMP‐10 in AD and its mechanistic role in neurodegeneration.