Objective:

To develop a novel, continuous, quantitative method for analysis of cutaneous phosphorylated alpha-synuclein (P-SYN). 

Background:

Synucleinopathies include Parkinson’s disease (PD), dementia with Lewy bodies (DLB), multiple system atrophy (MSA) and pure autonomic failure (PAF).  Early and accurate diagnosis of synucleinopathies remain a challenge although recent advances using skin immunofluorescent immunostaining of P-SYN from skin biopsies have offered a major diagnostic breakthrough.  There is an ongoing need for an objective, continuous, quantitative measure of P-SYN deposition to monitor disease progression and the response to disease modifying interventions.

Design/Methods:

We created a novel artificial intelligence (AI) driven protein detection algorithm to quantify misfolded proteins within cutaneous axons.  We studied 20 patients each with PD, MSA, DLB, PAF and healthy controls.  All subjects had 3mm skin biopsies at the distal leg, distal thigh and posterior cervical region with immunostaining for P-SYN and with protein gene product 9.5. All tissue sections underwent complete confocal digitization and AI driven quantitation of nerve fibers and intra-axonal P-SYN.

Results:

:  We created an AI driven heatmap of P-SYN overlaying each tissue section to define the location and quantity of P-SYN within nerve fiber subtypes. 300 biopsies from 100 participants were reviewed.  Diagnostic concordance between pathologic review of glass slides and AI augmented reading was 99.3% with P-SYN detected in 80/80 synucleinopathies and 0/20 healthy controls. AI assisted pathologic reading was >99% reproducible and augmented differentiation between synucleinopathy subtypes. 

Conclusions:

We have initiated a program for whole slide digitization combined with artificial intelligence assisted detection of pathologic proteins.  The application output provides heatmap assisted visualization of pathology with quantitative analysis of abnormal proteins by tissue region that improve detection of P-SYN with continuous, quantitative outputs.  Digital pathology with artificial intelligence assisted pathologic quantitation may make a substantial contribution to the neurodegenerative disease field and bring novel diagnostic advances into mainstream medical care.