Performance of a seed amplification assay for misfolded alpha-synuclein in cerebrospinal fluid and brain tissue in relation to Lewy body disease stage and pathology burden

Abstract

The development of in vitro seed amplification assays (SAA) detecting misfolded alpha-synuclein (αSyn) in cerebrospinal fluid (CSF) and other tissues has provided a pathology-specific biomarker for Lewy body disease (LBD). However, αSyn SAA diagnostic performance in early pathological stages or low Lewy body (LB) pathology load has only been assessed in small cohorts. Moreover, the relationship between SAA kinetic parameters, the number of αSyn brain seeds and the LB pathology burden assessed by immunohistochemistry has never been systematically investigated. We tested 269 antemortem CSF samples and 138 serially diluted brain homogenates from patients with and without neuropathological evidence of LBD in different stages by the αSyn Real-Time Quaking-Induced Conversion (RT-QuIC) SAA. Moreover, we looked for LB pathology by αSyn immunohistochemistry in a consecutive series of 604 Creutzfeldt-Jakob disease (CJD)-affected brains. αSyn CSF RT-QuIC showed 100% sensitivity in detecting LBD in limbic and neocortical stages. The assay sensitivity was significantly lower in patients in early stages (37.5% in Braak 1 and 2, 73.3% in Braak 3) or with focal pathology (50% in amygdala-predominant). The average number of CSF RT-QuIC positive replicates significantly correlated with LBD stage. Brain homogenate RT-QuIC showed higher sensitivity than immunohistochemistry for the detection of misfolded αSyn. In the latter, the kinetic parameter lag phase (time to reach the positive threshold) strongly correlated with the αSyn seed concentration in serial dilution experiments. Finally, incidental LBD prevalence was 8% in the CJD cohort. The present results indicate that (a) CSF RT-QuIC has high specificity and sufficient sensitivity to detect all patients with LB pathology at Braak stages > 3 and most of those at stage 3; (b) brain deposition of misfolded αSyn precedes the formation of LB and Lewy neurites; (c) αSyn SAA provides "quantitative" information regarding the LB pathology burden, with the lag phase and the number of positive replicates being the most promising variables to be used in the clinical setting.

Keywords: Amygdala; Diagnosis; Lewy bodies; Parkinson’s disease; RT-QuIC; α-Synuclein.

Fig. 1

CSF αSyn RT-QuIC SAA results in αSyn LB+ patients according to Braak LBD stage and LB score. Red dots represent αSyn RT-QuIC SAA negative patients, while blue dots indicate the αSyn RT-QuIC SAA-positive ones. Almost all negative patients exhibited a low LBD (or amygdala-predominant) stage or low LB scores. AP, amygdala-predominant; CSF, cerebrospinal fluid; LBD, Lewy body disease

Fig. 2

αSyn RT-QuIC SAA results in brain homogenates from αSyn LB+ patients. a Progressive reduction of seeding activity in brain homogenates of cingulate gyrus from case #24 along the dilution series. Each curve represents the average RFU values of positive quadruplicates. Standard deviation (SD) was hidden to improve the image's readability. b–f Progressive PAR reduction of αSyn RT-QuIC SAA reactions in different brain areas along the dilution series. The dilution coefficient is expressed in a logarithmic scale. Each point represents the average PAR value of positive quadruplicates at each dilution. Standard deviation (SD) was hidden to improve the image's readability. Linear regression lines were applied to values from patients in the same McKeith stage only when at least three values were available. AP, amygdala-predominant; PAR, protein aggregation rate; RFU, relative fluorescent units; SAA, seed amplification assay

Fig. 3

αSyn seeds quantification in brain samples a Average SD₅₀/mg values in the main McKeith stages in different brain areas. Standard deviation (SD) was hidden to improve the readability of the image. b Correlation analysis between logSD₅₀ score and LB score (r = 0.7903, 95% CI 0.4719 to 0.9263; p < 0.001). AMG, amygdala; BF, basal forebrain; CI, cingulate gyrus; CE, cerebellum; FC, frontal cortex; LB, Lewy body; ME, medulla oblongata; OCC, occipital cortex; SN, substantia nigra; TC, temporal cortex. SD₅₀/mg values are expressed in logarithmic scale

Fig. 4

Age-specific iLBD prevalence rates in the CJD cohort. The green area shows the number of LB-positive cases at αSyn immunohistochemistry in each decade. The red line expresses these data in percentage. iLBD, incidental Lewy body disease, CJD, Creutzfeldt Jakob disease