Diagnostic Performance and Clinical Applications of a Commercial Alpha‐Synuclein Seed Amplification Assay in a Tertiary Care Memory Clinic

Abstract Background SAAs represent a promising biomarker of Lewy Body disease (LBD), with high sensitivity (87.3%, 95%CI: 0.755‐0.947) and specificity (97.2%, 95%CI: 0.922‐0.994) for detection of patients in the Parkinson’s Progression Markers Initiative repository with positive DaT SPECT; and reasonable sensitivity (78.7%, 95%CI: 0.66‐0.88) and specificity (89.5%, 95%CI: 0.78‐0.96 referencing clinical diagnoses in participants included within national research repositories. However, test performance in heterogeneous clinical cohorts is unknown. Method Patients with typical (n=122) and rapid progressive dementia (n=76) consented to banking of CSF at the time of clinically‐indicated diagnostic lumbar puncture. Clinical data were deidentified and uploaded to a secure research database and clinical (etiologic) diagnoses established by two neurologists following independent review of available records. SAA (SYNTap™) were run on stored CSF by Amprion Diagnostics (San Diego, CA). Result Median participant age was 69.1 years (18.4‐86.1); 56.3% were female. CSF was sampled median 1.4 years (0‐10.3) after symptom onset. Pathological alpha‐synuclein was detected in 28/122 patients with typically progressive dementia (including 6/18 patients with presumed LBD; sensitivity 33%, specificity 79%) and 11/76 patients with rapid progressive dementia (including 3/9 patients with presumed LBD; sensitivity 33%, specificity 88%). “Positive” SAA was associated with tremor (OR=3.2, 95%CI: 1.3‐7.8), visual hallucinations (OR=6.14, 95%CI 1.6‐23.6), and REM behavior disorder (OR=3.7, 95%CI: 1.4‐9.5) in patients with typical but not rapid progressive dementia. Brain autopsies were completed in 17 patients: Lewy bodies were detected in 2/3 patients with positive SAA and no patients with negative SAA. Conclusion “Positive” SAA demonstrated reasonable specificity but low sensitivity for the diagnosis of LBD. Additional studies in heterogeneous cohorts with neuropathologic data are needed to optimize clinical applications of SAA.