Cerebrospinal Fluid Iron‐Ferritin Ratio as a Potential Progression Marker for Parkinson's Disease

Longitudinal PD CSF samples were subjected to ICP-MS and the total amount of iron and other bioelements was quantified. Additionally, ferritin and protein biomarkers of neurodegeneration were measured. Over time, mean iron levels significantly increased while levels of ferritin decreased. Biomarkers as surrogates for disease progression and drug response are urgently needed for Parkinson's disease (PD). In PD patients' brains, iron was shown to preferentially accumulate in the substantia nigra, and this accumulation increased over the disease course.1, 2 Iron accumulation has also been demonstrated in-vivo applying QSM-MRI.3 On the contrary, levels of the iron-binding protein ferritin were reduced in almost all evaluated brain regions in PD patients.4 In neurons, iron is mainly neuromelanin-bound, whereas ferritin-bound iron is more abundant in glia.5 We hypothesized that longitudinal assessment of total cerebrospinal fluid (CSF) iron and ferritin levels could be used to reflect progressive brain iron dyshomeostasis and disease progression in individual PD patients. For the first time, we report on a longitudinal CSF analysis of iron, ferritin, and additional bioactive elements (arsenic [As], copper [Cu], magnesium [Mg], nickel [Ni], selenium [Se], strontium [Sr], zinc [Zn]) with an implication in PD pathogenesis.6-9 In addition, several protein markers of neurodegeneration were quantified. Baseline and 1-year follow-up CSF samples of 20 PD patients were subjected to bioelement determination by mass-spectrometry [inductively coupled plasma optical emission spectrometry (ICP-OES), inductively coupled plasma-sector field mass spectrometry (ICP-sf-MS)].10 Nephelometric analysis was used for ferritin quantification. Amyloid beta 1–40/1–42, total-tau, and phospho-tau were quantified using ELISA and alpha-synuclein using SIMOA techniques (see Supplementary Material). Wilcoxon matched pairs signed-rank test was used to evaluate longitudinal differences, and correlation analysis was performed using Spearman's rho. A permission of the local ethics committee was obtained, and written consent was provided by all patients or care givers. In line with our main hypothesis, we report an increase in total iron over time (P = 0.002) while levels of ferritin decreased (P = 0.04). Levels of nickel also showed a significant decrease (P = 0.03). Application of an iron/ferritin ratio or an iron/ferritin/nickel score showed even more pronounced changes (P = 0.006; P = <0.0001, respectively; Fig. 1). All other bioelements and neurodegeneration markers remained stable over 1 year. To exclude influences by the antiparkinsonian medication, a correlation analysis was performed, which revealed no association between changes in the levodopa equivalent dose (LED) and the changes in iron, ferritin, and nickel levels over time (P > 0.05). Interestingly, there was a highly significant and strong correlation between baseline iron and baseline alpha-synuclein levels (r = 0.78, P < 0.0001, Supplementary Figure S1). Although this might reflect the metal binding affinity of alpha-synuclein, both might also be independently regulated in the course of disease. There was no correlation between baseline ferritin and alpha-synuclein levels after adjusting for multiple testing (see Supplementary Material). Due to a compensation by a significant increase in LED (P = 0.0003), no significant worsening in motor, nonmotor, and cognitive scores could be detected at the 1-year follow-up (P > 0.05) in our cohort, which mainly included early-stage PD patients. Therefore, correlation analyses between changes in CSF parameters and changes in clinical parameters were not applicable. As a limitation of this study, we therefore cannot establish a direct association of iron and ferritin levels to motor and nonmotor scores. In conclusion, this study yields first evidence from longitudinal data of individual patients for the potential of iron and ferritin as progression marker in PD. A validation of our findings in a larger cohort, more advanced PD patients and a longer follow-up period is warranted. We gratefully appreciate the participation of our patients in this study. We thank our Parkinson's and study nurses Gudrun Leyerer and Jennifer Heinemann for their excellent assistance. We also thank Peter Lange for providing helpful advice on assay-related issues. Open Access funding enabled and organized by Projekt DEAL. Design, conceptualization, and execution of the study: F.M., B.M., D.W., S.C., M.S., M.B., I.Z., P.L./Design, execution, and interpretation of the biostatistical analysis: F.M., P.L./Drafting the manuscript: F.M., P.L./Revising the manuscript: F.M., B.M., D.W., S.C., M.S., M.B., I.Z., P.L. All authors read and approved the final manuscript. FM received speaker honoraria from Bial. Paul Lingor is supported by the NUM research network, the Bavarian Staatsministerium für Wissenschaft und Kunst (StMWK), the Deutsche Forschungsgemeinschaft, the Bundesministerium für Bildung und Forschung (BMBF), the Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), the Deutsche Gesellschaft für Muskelkranke, the Else Kröner Fresenius Stiftung and received consulting or speakers honoraria from AbbVie, Alexion, Bial, Desitin, ITF Pharma, Stadapharm. The datasets analyzed during the current study are available from the corresponding author on reasonable request. APPENDIX S1: Supporting information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.