Purpose

Thus far, a breakthrough in Parkinson’s disease (PD) therapeutics has been slowed, in part, by the lack of a biomarker which can objectively detect brain changes related to PD and sensitively monitor these changes as the disease progresses. Prior work from our group demonstrated that free-water (FW) in the posterior substantia nigra (PSN), a novel metric derived from diffusion MRI (dMRI), is elevated in PD compared to controls across single- and multi-site cohorts, increases over 1 year in PD but not in controls, and predicts changes in parkinsonian symptoms1,2. Here, we used data from the Parkinson’s Progression Marker Initiative (PPMI) database, a multicenter study aimed to identify biomarkers in de novo PD, to test the feasibility of FW in PSN as a measure of disease progression in PD patients with a 4-year follow-up.

Methods

We included baseline and follow-up dMRI scans from patients with PD who had undergone nuclear imaging to confirm striatal dopamine deficit and were drug naïve at enrollment. PD patients were grouped into three cohorts: 1) PD imaged at baseline and 12 months (N = 103), 2) PD imaged at baseline, 12 months, 24 months, and 48 months (N = 46), and 3) a subset of PD who underwent both dMRI and DaTscan SPECT imaging at baseline, 12 months, 24 months, and 48 months (N = 30). Diffusion MRI scans were acquired at various sites using a Siemens 3T scanner and had the following parameters: b = 1000 s/mm2, 64 diffusion gradient directions with one b0 image, image matrix = 116 x 116 x 72, flip angle = 90º, and a voxel resolution of 1.98 x 1.98 x 2 mm. Upon preprocessing, a bi-tensor model was used, which calculates the signal attenuation as the sum of attenuations arising from two compartments: one that models FW and a tissue compartment that models either gray matter or a single bundle of white matter3. Using this model, FW and FW-corrected fractional anisotropy (FAt) maps were computed. Next, the b0 image was registered to a standardized T2-weighted template, and the resulting transformation matrix was then applied to the FW and FAt maps. Bilateral regions of interest (ROIs) were hand-drawn in the anterior substantia nigra (ASN) and posterior substantia nigra (PSN) on the b0 image of each subject in MNI space at each time point, and used to extract values from the corresponding FW and FAt maps1,2. We examined: a) 1-year changes in FW/FAt in 103 PD, b) 4-year changes in FW/FAt in the subset of 46 PD who were imaged 4 times over 4 years, and c) the relation between 4-year changes in FW and striatal binding ratio (SBR) in 30 PD who had undergone both diffusion MRI and DaTscan at each time point. We also report required sample size at 90% power for future clinical trials that would use relative changes in FW in the PSN.

Results

Results demonstrate that 1) FW in PSN increases over 1 year, 2) FW in PSN steadily increases over 4 years, 3) sex and baseline FW predict 4-year changes in FW, 4) 4-year changes in FW correlate with changes in SBR in the putamen. There was no time by site interaction. Also, no changes were detected in FW in ASN, or in FAt measures in ASN/PSN. The power analysis for a 2-arm drug study using relative change in FW in PSN over 2 years as an outcome measure indicated that to detect a significant effect of a neuroprotective agent with 90% power and 50% predicted change, 176 subjects are required across groups.

Discussion

Results extend previous single-site longitudinal findings in PD by showing that FW levels in PSN increase over 1 year in a large multi-site study of de novo PD. Importantly, FW in PSN continues to increase with disease progression over the course of 4 years., and results are consistent across PPMI sites, validating the feasibility and reliability of FW imaging as a progression marker of PD. Current findings also encourage sex as a factor to consider in the conduct of future clinical trials and most importantly suggest that imaging-based sample sizes are smaller than those based on clinical outcomes previously used in clinical trials in PD (e.g.1176 subjects for the ADAGIO study4, 336 subjects in the ongoing STEADY-PD trial NCT02168842).

Conclusion

We validate FW in PSN as a non-invasive imaging marker of disease progression in PD in a multi-site 4-year study, a measure which could be used in clinical trials of disease-modifying therapies.

Acknowledgements

The authors would like to thank the PD patients participating in the Parkinson’s Progression Marker Initiative (PPMI), an international multicenter study funded by the Michael J. Fox Foundation for Parkinson’s Research and funding partners including GE Healthcare, Pfizer, Genentech, Merck, Abbvie, Biogen, Covance, TEVA, UCB (http://www.ppmi-info.org/about-ppmi/who-we-are/study-sponsors/).