Evaluation of ATNPD framework and biofluid markers to predict cognitive decline in early Parkinson's disease

Background and ObjectivesIn Parkinsons disease (PD), Alzheimers disease (AD) co-pathology is common and clinically relevant. However, the longitudinal progression of AD cerebrospinal fluid (CSF) biomarkers - {beta}-amyloid 1-42 (A{beta}42), phosphorylated tau 181 (p-tau181) and total tau (t-tau) - in PD is poorly understood, and may be distinct from clinical AD. Moreover, it is unclear if CSF p-tau181 and serum neurofilament light (NfL) have added prognostic utility in PD, when combined with CSF A{beta}42. First, we describe longitudinal trajectories of biofluid markers in PD. Second, we modified the AD {beta}-amyloid/tau/neurodegeneration (ATN) framework for application in PD (ATNPD) using CSF A{beta}42 (A), p-tau181 (T), and serum NfL (N), and tested ATNPD prediction of longitudinal cognitive decline in PD. MethodsParticipants were selected from the Parkinsons Progression Markers Initiative (PPMI) cohort, clinically-diagnosed with sporadic PD or as normal Controls, and followed annually for 5 years. Linear mixed effects models (LMEM) tested the interaction of diagnosis with longitudinal trajectories of analytes (log-transformed, FDR-corrected). In PD, LMEMs tested how baseline ATNPD status (AD [A+T+N{+/-}] vs. not) predicted clinical outcomes, including Montreal Cognitive Assessment (MoCA; rank-transformed, FDR-corrected). ResultsParticipants were 364 PD and 168 Controls, with comparable baseline mean ({+/-}SD) age (PD=62{+/-}10; Control=61{+/-}11]; Mann-Whitney-Wilcoxon: p=0.40) and gender distribution (PD=231 males [63%]; Control=107 males [64%]; chi-square: p=1.0). PD had overall lower CSF p-tau181 ({beta}=-0.16, 95%CI=-0.23 - -0.092, p=2.2e-05) and t-tau than Controls ({beta}=-0.13, 95%CI=-0.19 - -0.065, p=4.0e-04), but not A{beta}42 (p=0.061) or NfL (p=0.32). Over time, PD had greater increases in serum NfL than Controls ({beta}=0.035, 95%CI=0.022 - 0.048, p=9.8e-07); PD slopes did not differ from controls for CSF A{beta}42 (p=0.18), p-tau181 (p=1.0) or t-tau (p=0.96). Using ATNPD, PD classified as A+T+N{+/-} (n=32; 9%) had consistently worse cognitive decline, including on global MoCA ({beta}=-73, 95%CI=-110 - -37, p=0.00077), than all other ATNPD statuses including A+ alone (A+T-N-; n=75; 21%). DiscussionIn early PD, CSF p-tau181 and t-tau were low compared to Controls and did not increase over 5 year follow-up. Even so, classification using modified ATNPD (incorporating CSF p-tau181 with CSF A{beta}42 and serum NfL) may identify biologically-relevant subgroups of PD to improve prediction of cognitive decline in early PD.