FDG PET Parkinson's disease-related pattern as a biomarker for clinical trials in early stage disease

Abstract

Background: The development of therapeutic interventions for Parkinson disease (PD) is challenged by disease complexity and subjectivity of symptom evaluation. A Parkinson's Disease Related Pattern (PDRP) of glucose metabolism via fluorodeoxyglucose positron emission tomography (FDG-PET) has been reported to correlate with motor symptom scores and may aid the detection of disease-modifying therapeutic effects.

Objectives: We sought to independently evaluate the potential utility of the PDRP as a biomarker for clinical trials of early-stage PD.

Methods: Two machine learning approaches (Scaled Subprofile Model (SSM) and NPAIRS with Canonical Variates Analysis) were performed on FDG-PET scans from 17 healthy controls (HC) and 23 PD patients. The approaches were compared regarding discrimination of HC from PD and relationship to motor symptoms.

Results: Both classifiers discriminated HC from PD (p < 0.01, p < 0.03), and classifier scores for age- and gender- matched HC and PD correlated with Hoehn & Yahr stage (R2 = 0.24, p < 0.015) and UPDRS (R2 = 0.23, p < 0.018). Metabolic patterns were highly similar, with hypometabolism in parieto-occipital and prefrontal regions and hypermetabolism in cerebellum, pons, thalamus, paracentral gyrus, and lentiform nucleus relative to whole brain, consistent with the PDRP. An additional classifier was developed using only PD subjects, resulting in scores that correlated with UPDRS (R2 = 0.25, p < 0.02) and Hoehn & Yahr stage (R2 = 0.16, p < 0.06).

Conclusions: Two independent analyses performed in a cohort of mild PD patients replicated key features of the PDRP, confirming that FDG-PET and multivariate classification can provide an objective, sensitive biomarker of disease stage with the potential to detect treatment effects on PD progression.

Keywords: Biomarker; Classifier; FDG PET; PDRP; Parkinson.

Figures

Graphical abstract

Fig. 1

(a) SSM derived training pattern, (b) NPAIRS CVA derived consensus training pattern, (c) Leave-One-Out consensus test pattern from NPAIRS CVA, and comparison of Leave-One-Out test CV1 mean (d) and individual (e) scores for HC vs. PD subjects. Higher CV1 scores reflect greater magnitude of expression of the PDRP pattern. Red regions represent areas of increased metabolism, blue regions represent areas of decreased metabolism (relative to whole brain metabolism). Error bars show standard error of the mean. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Age-related pattern identified by NPAIRS CVA in the comparison of HC subjects

Fig. 3

Results of NPAIRS CVA Leave-One-Out test results for comparison of age-matched, gender-matched HC and PD subjects. (a) Pattern of regional hypometabolism (blue) and hypermetabolism (red) relative to whole brain glucose metabolism that discriminated groups; b) CV1 scores; c) Correlation between CV1 score and H&Y stage; and (d) Correlation between CV1 score and UPDRS. higher the CV1 scores reflect greater magnitude of expression of the PDRP relative to whole brain shown in a). Horizontal bars in (b) represent group means; Effect size = −1.11 (Hedge's g). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

a) CV1 pattern for classifier trained on PD subjects from Hoehn & Yahr stages I vs. II & III; b) Correlation between CV1 scores and UPDRS motor scores generated through Leave-One-Out testing; c) Correlation between Leave-One-Out CV1 scores and H&Y stage. Higher CV1 scores reflect greater magnitude of expression of the PDRP.