Carriers of LRRK2 pathogenic variants show a milder, anatomically distinct brain signature of Parkinson's disease

Abstract

LRRK2 gene variants are a major genetic risk factor for both familial and sporadic Parkinson's disease (PD), opening an unattended window on the disease's mechanisms and potential therapies. Investigating the influence of pathogenic variants in LRRK2 gene on brain structure is a crucial step toward enabling early diagnosis and personalized treatment. Yet, despite its significance, the ways in which LRRK2 genotype affects brain structure remain largely unexplored. Work in this domain is plagued by small sample sizes and differences in cohort composition, which can obscure genuine distinctions among clinical subgroups. In this study, we overcome such important limitations by combining explicit modeling of population background variation and pattern matching. Specifically, we leveraged a large cohort of 641 participants (including 364 with a PD diagnosis) to examine MRI-detectable cortical atrophy patterns associated with the LRRK2 pathogenic variants in people with PD and non-manifesting individuals. LRRK2 PD patients exhibited milder cortical thinning compared to sporadic PD, with notable preservation in temporal and occipital regions, suggesting a distinct pattern of neurodegeneration. Non-manifesting LRRK2 carriers showed no significant cortical atrophy, indicating no structural signs of subclinical PD. We further analyzed the relationship between aggregated alpha-synuclein in cerebrospinal fluid and atrophy. We found that those with evidence of aggregated alpha-synuclein experienced pronounced neurodegeneration and increased cortical thinning, possibly defining another aggressive PD subtype. Our findings highlight avenues for distinguishing PD subtypes, which could lead to more targeted treatment approaches and a more complete understanding of Parkinson's disease progression.

Figure 1:. Complex interrelations between subgroups in ~600 participant population cohort

a. Ratio breakdown of participant subgroups. Our participant sample consists of 364 participants with the diagnosis of Parkinson’s disease and 227 participants without PD diagnosis (inner circular bar plot). Among these, 94 non-PD participants and 76 PD patients carry LRRK2 pathogenic variants indicated by the black edge (middle circular bar plot). Finally, the outer circular bar plot depicts the ratio of asyn SAA positive (red) and asyn SAA negative (blue) participants in each subgroup. b. PD patients carrying LRRK2 pathogenic variants represent the oldest subgroup. Raincloud plots display the age of each participant in one of the four subgroups. c. Proportion of females among subgroups. Non-manifesting carriers with LRRK2 pathogenic variants display the highest proportion of females. The red dotted line marks an equal ratio of male and female participants. d. Different disease duration in PD patients. Disease duration is plotted for PD patients separated by the presence of LRRK2 pathogenic variant. The diversity in key demographic metrics poses a challenge to comparing PD and control subgroups. Abbreviations: HC=healthy controls, NMC=non-manifesting carrier, PD=Parkinson’s disease, sPD=sporadic PD.

Figure 2:. Region-specific models accurately estimate cortical thinning in PD

a. Subgroups used to derive models of general PD alterations. The infographic highlights the HC and sPD subgroups used for the ensuing analysis. b. Selection of carefully matched groups of participants. Only 104 out of 133 HC, resp. 104 out of 288 sPD, participants are carried forward to derive estimates of brain alterations associated with PD. c. Propensity scores enable careful participant matching. Participants from both subgroups are matched based on age, sex, and site using a flexible propensity score matching strategy, resulting in an excellent match. The histograms display frequencies of propensity scores separately for sPD and HC subgroups. d. Estimates of brain structure alterations. Regional thickness for cortical regions and regional volume for subcortical regions are predicted using a series of linear models based on disease diagnosis, age, sex, and site. Colored brains display the diagnosis coefficient estimates highlighting decreased (blue color) or increased (red color) thickness, resp. volume. e. Effect size of diagnosis across all brain regions. The heatmap shows diagnosis beta coefficients among 68 regions based on the Desikan-Killiany cortical parcellation and 16 regions defined by Harvard-Oxford subcortical atlas. Detected cortical thinning and volume loss aligns with established PD changes, underscoring our analytical protocol.

Figure 3:. Parkinson’s patients with LRRK2 pathogenic variants show a milder form of brain atrophy

a. Subgroups used to assess the extent of PD-related atrophy. The infographic indicates the sPD and LRRK2 PD subgroups used for the present analysis. b. Selection of carefully matched groups of PD patients. Only 19 out of 288 sPD, resp. 19 out of 76 LRRK2, participants are selected based on age, sex, disease duration and site using propensity score matching. c. Transferring the model of PD-related atrophy between cohort subgroups. We apply a PD atrophy model, trained on measures from sPD and HC subgroups, to estimate regional thickness and volume in LRRK2 PD and sPD patients (i.e., model transfer). We then average these estimates across participants to derive region-specific measures. d. Model estimates reveal less atrophy in LRRK2 PD patients. We compare region-wise differences of measured and estimated morphometry between the two subgroups of PD patients. Positives suggest preservation of brain structure while negatives point towards pronounced degeneration. The p-value denotes a two-sample t-test of prediction differences. e. LRRK2 pathogenic variants are associated with less cortical thinning in temporal and occipital regions. Region-wise differences between measured and estimated brain metrics are plotted on cortical and subcortical regions for LRRK2 PD participants. Orange color highlights regions with lesser amount of cortical thinning or volume reduction while purple color denotes increased thinning or reduction. f. Relationship of prediction difference with PD atrophy. The diagnosis-related beta coefficients are plotted against prediction differences for every brain region. g. Linking predictions of brain structure to cognitive and genetic measures. The correlations quantify linear association strength across all sPD and LRRK2 PD patients between predictions of brain structure averaged across all regions with MDS-UPDRS III, MoCA, and PRS. h. Relating regional predictions to cognitive and genetic factors. Correlations between MDS-UPDRS III, MoCA, and PRS with predictions of regional morphometry are displayed for all regions. Overall, patients carrying LRRK2 pathogenic variants show less brain atrophy, with preservation especially in the temporal and occipital regions.

Figure 4.. Absence of significant brain structure alterations in LRRK2 non-manifesting carriers

a. The influence of LRRK2 pathogenic variants in participants without PD diagnosis. We analyze the influence of LRRK2 pathogenic variants in HC and LRRK2 NMC participants. b. Selection of carefully matched groups of HC and LRRK2 NMC participants. Only 42 HC and 42 LRRK2 NMC participants are selected based on age, sex, and site using propensity score matching. c. Transferring the model of PD-related atrophy. We apply a PD atrophy model, trained on measures from sPD and HC subgroups, to estimate regional brain thickness and volume in HC and LRRK2 NMC. d. Comparable estimates of brain morphometry in controls. Boxen plots display region-wise differences of measured and estimated morphometry for HC and LRRK2 NMC subgroups. The p-value denotes a two-sample t-test of prediction differences. Unlike in PD patients, the presence of LRRK2 pathogenic variants alone is not associated with significant structural brain alterations in participants without PD diagnosis.

Figure 5.. Aggregated alpha-synuclein proteins associated with increased brain atrophy

a. Measures of asyn SAA across participant subgroups. We analyze the influence of asyn SAA measured in HC, sPD, and LRRK2 PD participants. b. Only 23 out of 149 asyn SAA negative, resp. 23 out of 295 asyn SAA positive, participants are selected based on age, sex, disease duration and site using propensity score matching. c. Model estimates show more atrophy associated with asyn SAA positive status. We compare region-wise differences of measured and estimated morphometry between groups of asyn SAA positive and negative participants. Lower measured compared to predicted thickness and volume suggests pronounced degeneration. d. Differences in vulnerability for degeneration in cortical and subcortical regions associated with positive asyn SAA result. Region-wise differences between measured and estimated brain morphometry for asyn SAA positive are plotted on cortical and subcortical regions. Orange color highlights regions with lesser amount of cortical thinning or volume reduction than expected by the model PD atrophy while purple color denotes increased thinning or reduction. Our findings reveal pronounced cortical thinning specifically linked to asyn SAA positive status, with distinct cortical regions showing increased vulnerability.