The Association of Total and Differential Leukocyte Counts with Incident Parkinson’s Disease in UK Biobank

Abstract ImportanceBiomarkers for the early detection of Parkinson’s disease (PD) are needed; emerging evidence implicates immune dysregulation. Altered leukocyte differentials and C-reactive protein (CRP), common markers of immune function, occur in patients with PD. Whether these changes drive pathogenesis remains unclear. ObjectivesWe sought to identify whether peripheral immune dysregulation can be seen as a pre-diagnostic feature of PD, and whether it appears to play a causal role. DesignWe examined the relationship between differential leukocyte count and other markers of acute inflammation at enrolment, and incident cases of PD, in UK Biobank, a longitudinal cohort study. We used Mendelian randomization to establish whether differences in leukocyte subsets have a causal influence on risk of PD. SettingThe UK Biobank; a population-based cohort with over 500,000 participants aged 40-69 recruited in the UK between 2006 and 2010. ParticipantsPD cases were defined as individuals with an ICD-10 coded diagnosis of PD. Cases were defined as ‘incident’ if their age at diagnosis was greater than their age at recruitment to UKB. ‘Controls’ were defined as all other individuals in the dataset after applying the above exclusions. After applying exclusion criteria for pre-existing health conditions that can influence blood counts, 507 incident PD cases and 328,280 controls were included in the analysis. ExposureBlood cell markers (absolute and relative counts) and other markers of inflammation (CRP and albumin) were obtained from blood tests of participants taken at the initial assessment visit. ResultsLower lymphocyte count was associated with increased risk of incident PD. There was weaker evidence of association between lower eosinophil and monocyte counts, lower CRP, and higher neutrophil counts on risk of incident PD. The association between lymphopenia and incident PD remained robust to sensitivity analyses. Mendelian randomization did not reveal a clear causal effect of low lymphocyte count on PD risk, but a weak trend in that direction was seen. Conclusions and relevanceIn this large, prospective setting, lower lymphocyte count was associated with higher risk of subsequent PD diagnosis. The absence of a clear causal effect indicates that lymphopenia might be a consequence of prodromal PD. Key pointsQuestionIs the leukcoyte differential count a feature of pre-diagnostic Parkinson’s disease?FindingsIn the UK Biobank, a longitudinal cohort study with over 500,000 participants, lower lymphocyte count was associated with a 23% increased odds of incident PD, a significant difference.MeaningPre-diagnostic Parkinson’s disease is associated with lower lymphocyte counts; this could enhance positive predictivity in algorithms designed to identify groups at higher risk of future Parkinson’s disease.