Male sex accelerates cognitive decline in GBA1 Parkinson’s Disease

doi:10.21203/rs.3.rs-4885538/v1

Download PDF

Article

Male sex accelerates cognitive decline in GBA1 Parkinson’s Disease

https://doi.org/10.21203/rs.3.rs-4885538/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

We evaluated 128 GBA and 432 nonGBA Parkinson’s disease (PD) subjects. Baseline clinical features and dopaminergic activity were assessed, together with 7-year clinical follow-up. Survival analyses assessed the independent and interactive effects of male sex and GBA1 mutations on cognitive impairment.

At baseline, GBA-PD males showed greater motor impairment, sleep disorders and memory deficits, GBA-PD females showed greater dopaminergic denervation. In longitudinal assessment, GBA-PD males showed greater MoCA rate of change per year and greater risk of cognitive impairment than GBA-PD females and nonGBA-PD, also when excluding subjects with LRRK2 mutations. In GBA-PD males, both late age at onset and “severe/mild” GBA variants were associated with increased risk of cognitive impairment.

Male sex and GBA1 carrier status have an additive value in increasing the risk of cognitive decline in PD. The effect of sex on GBA1-related pathology warrants further examination to address future trials design and patients’ selection.

Gender

Dementia

Synucleinopathy

Risk factors

Dopaminergic denervation

Parkinson's disease (PD) is a neurodegenerative condition characterized by a heterogeneous range of motor and non-motor symptoms.¹ Among these, the occurrence of cognitive impairment is one of the main causes of a poorer quality of life.² The different clinical trajectories observed in PD pose considerable challenges in identifying appropriate participants for clinical trials. As a result, there is a pressing need for a more accurate identification of factors that may increase the risk of PD subjects to develop cognitive impairment.

Genetic factors are known to contribute to the susceptibility to cognitive decline and dementia in PD.³ Heterozygous variants in the glucocerebrosidase gene (GBA1), occurring in about 8–12% of PD subjects worldwide, seem to accelerate the neurodegenerative process already in the earlier stages of the disease,⁴ leading to a significant dopaminergic damage and a more severe clinical phenotype.⁵ Indeed, GBA-PD subjects generally have an earlier disease onset, a higher prevalence of non-motor symptoms and a greater risk of progression to dementia compared to non-mutated subjects.⁶

Sex is another established factor that affects incidence, natural history, and clinical manifestations of PD. A recent meta-analysis confirmed the clear male preponderance over females (59% vs 41%),⁷ with men manifesting more severe motor symptoms and faster disease progression than women.⁸ From a pathogenetic perspective, genetic, hormonal, neuroendocrinal and molecular players all contribute towards these sex-related differences.⁹ In particular, steroid sex hormones, and especially oestrogens, seem to play a crucial neuroprotective role and anti-inflammatory function against PD pathology.⁴

GBA1 pathogenic variants are equally detected in PD subjects of both sexes,⁴ surpassing the potential impact of environmental exposures and hormonal influences that likely result in the higher male prevalence characterising idiopathic PD. Thus, it becomes crucial to investigate the potential interactions between GBA1 genotype and sex, and their combined influence on development of cognitive impairment.

In this study, we embarked on a comprehensive exploration of the mutual role of sex and GBA1 mutations in modulating dopaminergic vulnerability as well as the risk to develop dementia in PD subjects. By shedding light on these multifaceted dimensions, we aim to deepen our understanding of PD cognitive manifestations and pave the way for more targeted and effective interventions.

2.1 Subjects

This is an observational cohort study of subjects with a diagnosis of PD. Data used for this study are openly available from Parkinson's Progression Markers Initiative (PPMI) database (www.ppmi-info.org/data), a multicentre, prospective, longitudinal study that aims to identify genetic, blood, cerebral spinal fluid, and imaging biomarkers of PD progression. Data used in this article were last downloaded in February 2024. Included data were acquired between Jul 1, 2010, and Jun 1, 2019, from the PPMI public database (www.ppmi-info.org/access-data-specimens/download-data), RRID:SCR_006431. For up-to-date information on the study, visit www.ppmi-info.org.^10,11

We included 560 PD subjects (mean age in years ± SD: 60.97 ± 10.1; sex [F/M]: 225/335). The selected cohort had an average 7-year follow-up (mean ± SD: 6.87 ± 3.2 years; [min/max: 1/13 years]). Among them, 128(22.8%) carried a GBA1 variant (GBA-PD), while the remaining 432(77.1%) were GBA-negative (nonGBA-PD). PD subjects were further categorised based on their age at symptoms onset (AAO) into early-onset PD (EO-PD), AAO ≤ 50 years (n = 128), and late-onset PD (LO-PD), AAO > 50 years (n = 421).^5,12

Among the included PD, 292 subjects (GBA-PD/non-GBA-PD: 90/202) were acquired at baseline with ¹²³I-FP-CIT-SPECT to image dopamine transporter (DAT) binding. A group of 125 healthy control subjects (HC) (mean age in years ± SD: 59.02 ± 11.5; sex [M/F]: 81/44) was selected from the PPMI as normative ¹²³I-FP-CIT-SPECT dataset. The HC dataset showed no first-degree family history of PD, no significant neurological disorders and cognitive impairment/dementia, a stable Movement Disorders Society-Unified Parkinson's Disease Rating Scale-Part III (MDS-UPDRS-III) scores, no PD diagnosis for 6 years follow-up and were all negative for GBA1 mutations.

2.2 Clinical assessment at baseline

We included demographic, clinical motor and non-motor and neuropsychological assessments collected at baseline for each patient (Supplementary materials). Demographic and clinical features were compared by means of ANOVA with post-hoc Bonferroni correction for continuous variables and Chi-squared tests for categorical variables. Differences in sex distribution in subjects stratified by GBA1 variant type were determined using one-sample tests of equality of proportions.

2.3 Brain Dopaminergic Activity at baseline

We retrieved reconstructed imaging data related to ¹²³I-FP-CIT-SPECT from the PPMI website. Preprocessing of SPECT brain images was conducted using Statistical Parametric Mapping software (SPM12, Wellcome Trust Centre for Neuroimaging, London, UK, available at: https://www.fil.ion.ucl.ac.uk/spm/). See Supplementary Materials for details.

A voxel-wise multiple linear regression model was employed in SPM12 to compare GBA-PD and nonGBA-PD, and to direct compare groups within the same sex. The model included age at acquisition, asymmetry index (AI), MDS-UPDRS-III and Levodopa equivalent daily dose (LEDD) as covariates. We set our voxel-wise significance threshold at p < 0.05 (uncorrected) and a minimum cluster extent of 100 voxels. We compared the distributions of parametric binding potentials across groups through ANCOVA with post-hoc Bonferroni correction, adjusted for age at acquisition, AI, MDS-UPDRS-III and LEDD.

2.4 Cognitive trajectories at follow-up

To assess the rate of cognitive decline over time, we calculated MoCA deflection by subtracting the baseline MoCA score from the score at the last available follow-up and dividing the result by the number of years between assessments. Between-group differences in MoCA deflection were assessed to determine if any of the four groups exhibited a faster rate of cognitive decline.

We used a linear mixed-effects model (lme) to estimate the rate of change in the MoCA scores available at each time point across: 1) males and females, 2) patients with PD who were carriers of the GBA1 mutation and those who did not carry the mutation, and 3) EO-PD and LO-PD. We also considered the interactions among all the above conditions, and we run the same set of analyses by excluding PD subjects with LRRK2 mutations. The difference between the first MoCA scores and each subsequent assessment was used as the time scale. The lme models were adjusted for age (excluding for AAO model), years of education, disease duration and MDS-UPDRS-III at baseline. The participant-specific random effect was used to account for correlations between repeated measurements from the same participants. Lme has been chosen as a valuable tool to effectively handle correlated data points within the same participant, even where there are varying numbers of observations per participant, missing data, or uneven intervals between measurements.¹³

Moreover, we assessed the probability of remaining free from cognitive impairment using the Kaplan–Meier method and Cox regression model.¹⁴ To ensure the validity of the Cox model, we tested the proportional hazards assumption, which was found to be satisfied in our analysis. This indicates that the risk ratios estimated by the Cox model are stable over time, supporting the appropriateness of this method for our study. In detail, we estimated Hazard Ratio (HR) associated with sex, GBA1 genotype, GBA variants, and AAO. We also conducted a Kaplan-Meier survival analysis to investigate the combined effect of sex and GBA1 genotype on the AAO.

Survival analyses included the assessment of risk of cognitive impairment, which was defined as change in MoCA score from a baseline of > 26 to the development of a score ≤ 26.¹⁵ Covariates included age (excluding for AAO model), years of education, disease duration and MDS-UPDRS-III at baseline.

During the 13-year follow-up period, not all the 560 subjects included in the study completed all control assessments regularly. To address this, we performed sensitivity analyses to replicate Cox regression models considering a subpopulation of 440 PD subjects who had homogeneous clinical assessments at baseline and 3-year follow-up and excluding patients carrying LRRK2 mutations. Further details are available in the Supplementary materials.

All statistical analyses were performed with Statistical Package for Social Sciences (SPSS) version 28 and Python packages. The significance threshold p < 0.05 was established for all tests.

3.1 Baseline features

The demographic and clinical characteristics of the whole GBA-PD (M/F: 66/62) and nonGBA-PD (M/F: 269/163) groups are shown in Table 1. Upon GBA1 variant classification, 25% subjects carried “risk” variants, 56.2% “mild”, 10.9% “severe”, and 6.2% “unknown”. The N409S “mild” variant emerged as predominant, being detected in 50% of mutated individuals (Supplementary Table 1).

Table 1

Demographic and clinical features of GBA-PD and nonGBA-PD groups.
	GBA-PD	nonGBA-PD	p-value^a	M	F	p-value^b
N (%)	128 (22.8%)	432 (77.2%)	-	335 (59.8%)	225 (40.2%)	-
Age at baseline	59.86 ± 10	61.30 ± 10.1	0.159	61.01 ± 10.3	60.92 ± 9.9	0.916
AAO	56.3 ± 10.3	58.5 ± 10.7	0.040	58.4 ± 10.8	57.3 ± 10.4	0.254
Education (years)	16.05 ± 3.3	15.51 ± 3.6	0.127	16.13 ± 3.2	14.89 ± 3.8	0.000
Disease Duration (years)	1.73 ± 2.1	0.88 ± 1.6	0.000	0.84 ± 1.5	1.43 ± 2	0.000
LEDD	320.13 ± 315	301.55 ± 332.9	0.576	334.4 ± 373.3	263.39 ± 243	0.013
Clinical assessment
MDS-UPDRS-III	22.95 ± 10.8	20.03 ± 9.5	0.003	21.54 ± 10.4	19.44 ± 9	0.014
SCOPA-AUT	16.29 ± 11.2	14.84 ± 9.6	0.164	12.96 ± 8.7	18.56 ± 11.1	0.000
RBDSQ	5.32 ± 3.4	4.19 ± 2.7	0.000	4.64 ± 3	4.16 ± 2.7	0.055
RBD disorders (RBDSQ ≥ 6)	53 (41.4%)	120 (28%)	0.108	112 (33.4%)	61 (27%)	0.410
GDS	5.58 ± 1.6	5.39 ± 1.6	0.238	5.36 ± 1.5	5.55 ± 1.7	0.164
Cognitive assessment - at baseline
MoCA	26.62 ± 2.4	26.86 ± 2.7	0.363	26.68 ± 2.5	26.99 ± 2.8	0.187
BJLO	11.23 ± 3.1	11.95 ± 2.9	0.016	12.35 ± 2.9	10.95 ± 2.9	0.000
LNS	11.61 ± 2.8	11.36 ± 2.7	0.371	11.56 ± 2.6	11.20 ± 2.9	0.133
Semantic fluency	50.28 ± 11.4	50.99 ± 10.5	0.511	50.32 ± 10.7	51.59 ± 10.6	0.167
HVLT, immediate recall	46.22 ± 10.5	45.99 ± 10.7	0.833	44.75 ± 10.6	47.97 ± 10.5	0.000
HVLT, delayed recall	46.45 ± 10.3	45.14 ± 11.4	0.246	43.99 ± 11	47.60 ± 11	0.000
HVLT, retention	48.59 ± 10.5	46.90 ± 11.7	0.144	46.47 ± 11.7	48.51 ± 11.1	0.040
HVLT, recognition	46.74 ± 11.4	45.96 ± 10.9	0.362	44.46 ± 10.9	48.18 ± 10.9	0.000
SDMT	44 ± 11.1	44.75 ± 9.2	0.459	43.98 ± 9.6	45.48 ± 9.7	0.073
Cognitive assessment - at last follow-up
MoCA	25 ± 4.3	25.55 ± 4.3	0.242	25.05 ± 4.4	26.01 ± 4.1	0.010
LNS	10.52 ± 3.6	10.81 ± 3.2	0.383	10.59 ± 3.2	10.98 ± 3.2	0.179
Semantic fluency	48.76 ± 12.2	50.72 ± 12.2	0.111	49.23 ± 12.8	51.81 ± 11.1	0.014
HVLT, immediate recall	42.46 ± 12.2	44.79 ± 11.7	0.052	41.95 ± 11.2	47.69 ± 12	0.000
HVLT, delayed recall	43.60 ± 12.6	43.62 ± 13.1	0.991	41.90 ± 12.9	46.16 ± 12.7	0.000
HVLT, retention	46.31 ± 12.5	44.71 ± 13	0.216	44.21 ± 13.5	46.35 ± 11.9	0.054
HVLT, recognition	43.60 ± 12.8	45.34 ± 11.8	0.152	43.77 ± 11.9	46.68 ± 12.1	0.005
MoCA deflection	-0.68 ± 1.5	-0.27 ± 1	0.000	-0.42 ± 1.1	-0.28 ± 1.2	0.126
Cognitive Impairment	62 (48.4%)	166 (38.4%)	0.283	146 (43.5%)	82 (36.4%)	0.427
Abbreviations: AAO, age at symptoms onset; BJLO, Benton Judgment of Line Orientation; GBA, Glucocerebrosidase; GDS, Geriatric Depression Scale; HVLT, Hopkins Verbal Learning Test-Revised; LEDD, levodopa-equivalent daily dose; LNS Letter-Number Sequencing; MoCA, Montreal Cognitive Assessment; MDS-UPDRS-III, Movement Disorders Society-Unified Parkinson’s Disease Rating Scale pt. III; N, Number; PD, Parkinson's disease; RBDSQ, REM Sleep Behavior Disorder Screening Questionnaire; SCOPA-AUT, Scales for Outcomes in PD-Autonomic; SDMT, Symbol Digit Modalities Test.
^a= comparison between GBA-PD and nonGBA-PD patients, regardless of sex.
^b= comparison between all PD males and all PD females, regardless of genotype.
Significant p-values (p < 0.05) are reported in bold

At baseline, the two groups were comparable for age and educational level, but GBA-PD showed an earlier AAO than nonGBA-PD. GBA-PD showed longer disease duration, a significantly more severe motor profile and higher Rapid Eye Movement Sleep Behaviour Disorder Questionnaire (RBDSQ) score than nonGBA-PD. On the cognitive side, MoCA scores were comparable between GBA-PD and nonGBA-PD, although GBA-PD showed lower performance on visuospatial function (Benton Judgment of Line Orientation - BJLO).

When assessing sex-related differences, regardless of GBA1 carrier status, PD males showed more severe RBD and motor symptoms, while females showed greater autonomic dysfunction. PD males also showed lower memory performance than females, while the opposite was observed for visuospatial scores, with no differences in MoCA scores between sexes (Table 2).

Table 2

Demographic and clinical features of GBA-PD and nonGBA-PD groups divided by sex.
	GBA-PD F	GBA-PD M	nonGBA-PD F	nonGBA-PD M	Comparison across the four groups
	GBA-PD F	GBA-PD M	nonGBA-PD F	nonGBA-PD M	p-value	Post Hoc
N (%)	62 (11%)	66 (11.8%)	163 (29.2%)	269 (48%)	0.000	-
Age at baseline	60.10 ± 10.2	59.64 ± 9.9	61.23 ± 9.7	61.34 ± 10.4	0.560	-
AAO	56.15 ± 10.7	56.34 ± 10.1	57.78 ± 10.3	58.9 ± 10.9	0.149	-
Education (years)	15.48 ± 3.9	16.59 ± 2.6	14.67 ± 3.8	16.02 ± 3.4	0.000	nonGBA-PD F < GBA-PD M nonGBA-PD F < nonGBA-PD M
Disease Duration (years)	2 ± 2.3	1.48 ± 2	1.21 ± 1.9	0.68 ± 1.3	0.000	GBA-PD F > nonGBA-PD F GBA-PD F > nonGBA-PD M
LEDD	284.2 ± 288.1	354.41 ± 337.4	255.46 ± 224	329.59 ± 382	0.078	-
Clinical assessment
MDS-UPDRS-III	21.21 ± 9.7	24.58 ± 11.6	18.77 ± 8.6	20.79 ± 9.9	0.001	GBA-PD M > nonGBA-PD F GBA-PD M > nonGBA-PD M
SCOPA-AUT	18.85 ± 11.4	13.88 ± 10.6	18.42 ± 11	12.69 ± 8	0.000	GBA-PD F > GBA-PD M GBA-PD F > nonGBA-PD M nonGBA-PD F > GBA-PD M nonGBA-PD F > nonGBA-PD M
RBDSQ	4.65 ± 3	5.95 ± 3.5	3.98 ± 2.6	4.32 ± 2.8	0.000	GBA-PD M > nonGBA-PD F GBA-PD M > nonGBA-PD M
RBD disorders (RBDSQ ≥ 6)	20 (32.2%)	33 (50%)	41 (25%)	79 (29%)	0.013	-
GDS	5.73 ± 1.8	5.44 ± 1.4	5.48 ± 1.6	5.33 ± 1.5	0.355	-
Cognitive assessment - at baseline
MoCA	27.03 ± 2.4	26.23 ± 2.4	26.97 ± 3	26.80 ± 2.6	0.243	-
BJLO	10.54 ± 2.9	11.89 ± 3.1	11.11 ± 2.9	12.47 ± 2.8	0.000	GBA-PD F < GBA-PD M GBA-PD F < nonGBA-PD M nonGBA-PD F < nonGBA-PD M
LNS	11.55 ± 2.7	11.67 ± 2.8	11.07 ± 3	11.54 ± 2.6	0.295	-
Semantic fluency	49.55 ± 11.2	50.97 ± 11.6	52.36 ± 10.2	50.16 ± 10.5	0.148	-
HVLT, immediate recall	47.84 ± 11.1	44.70 ± 9.7	48.02 ± 10.3	44.77 ± 10.8	0.006	nonGBA-PD M < nonGBA-PD F
HVLT, delayed recall	49.79 ± 9.3	43.30 ± 10.3	46.77 ± 11.5	44.15 ± 11.2	0.000	GBA-PD M < GBA-PD F nonGBA-PD M < GBA-PD F
HVLT, retention	51.50 ± 9.2	45.86 ± 11	47.37 ± 11.5	46.62 ± 11.9	0.016	GBA-PD M < GBA-PD F nonGBA-PD M < GBA-PD F
HVLT, recognition	50.11 ± 9.9	43.58 ± 11.9	47.44 ± 11.2	44.68 ± 10.6	0.000	GBA-PD M < GBA-PD F nonGBA-PD M < GBA-PD F
SDM	45.80 ± 10.1	42.36 ± 11.9	45.4 ± 9.6	44.38 ± 9	0.133	-
Cognitive assessment - at last follow-up
MoCA	26.02 ± 3.7	24 ± 4.7	26.01 ± 4.2	25.27 ± 4.3	0.013	GBA-PD M < nonGBA-PD F
LNS	11.10 ± 3	9.98 ± 4.1	10.93 ± 3.2	10.74 ± 3.2	0.196	-
Semantic fluency	50.42 ± 11.1	47.20 ± 13	52.34 ± 11.1	49.73 ± 12.7	0.024	GBA-PD M < nonGBA-PD F
HVLT, immediate recall	47.37 ± 11.8	37.85 ± 10.7	47.81 ± 12.1	42.96 ± 11.2	0.000	GBA-PD M < GBA-PD F GBA-PD M < nonGBA-PD F GBA-PD M < nonGBA-PD M nonGBA-PD M < GBA-PD F nonGBA-PD M < nonGBA-PD F
HVLT, delayed recall	46.90 ± 12.2	40.50 ± 12.4	45.88 ± 13	42.24 ± 13	0.001	GBA-PD M < GBA-PD F GBA-PD M < nonGBA-PD F nonGBA-PD M < nonGBA-PD F
HVLT, retention	47.60 ± 10.3	45.11 ± 14.3	45.88 ± 12.4	44 ± 13.3	0.180	-
HVLT, recognition	46.48 ± 12.6	40.89 ± 12.5	46.76 ± 11.9	44.48 ± 11.7	0.005	GBA-PD M < GBA-PD F GBA-PD M < nonGBA-PD F
MoCA deflection	-0.28 ± 1.4	-1.05 ± 1.4	-0.28 ± 1.1	-0.27 ± 0.9	0.000	GBA-PD M < GBA-PD F GBA-PD M < nonGBA-PD F GBA-PD M < nonGBA-PD M
Cognitive Impairment	24 (38.7%)	38 (57.6%)	58 (35.5%)	108 (40%)	0.075	-
Abbreviations: AAO, age at symptoms onset; BJLO, Benton Judgment of Line Orientation; GBA, Glucocerebrosidase; GDS, Geriatric Depression Scale; HVLT, Hopkins Verbal Learning Test-Revised; LEDD, levodopa-equivalent daily dose; LNS Letter-Number Sequencing; MoCA, Montreal Cognitive Assessment; MDS-UPDRS-III, Movement Disorders Society-Unified Parkinson’s Disease Rating Scale pt. III; N, Number; PD, Parkinson's disease; RBDSQ, REM Sleep Behavior Disorder Screening Questionnaire; SCOPA-AUT, Scales for Outcomes in PD-Autonomic; SDMT, Symbol Digit Modalities Test.
Post-hoc comparison survived Bonferroni correction.
Significant p-values (p < 0.05) are reported in bold.

GBA-PD males had significantly more severe REM sleep disturbances than all other groups, and the worst motor scores, albeit significant only vs. non-GBA-PD, regardless of sex. Other inter-group comparisons at baseline confirmed the former observations of worse autonomic dysfunction, worse visuospatial performance and better memory performance in females compared to males.

There was no difference in sex-specific distribution of GBA1 carriers (59 men, 50%) and across certain variant subgroups ("Risk" = 18 males, 56.2%, and "Mild" = 35 males, 48.6%) (Supplementary Table 1). However, we observed a higher prevalence of males for the "severe" variants (9 males, 64.3%), and selected variants prevailed in either sex (e.g. T369M in males and E326K in females). The different subgroups defined by the genetic variants were similar in terms of MoCA scores at baseline.

3.2 Dopamine Transporter Imaging at baseline

GBA-PD showed reduced dopamine uptake in putamen [MNI XYZ Coordinates: -28; 0; -2/30; 8; -4] as compared to nonGBA-PD (Fig. 1a).

NonGBA-PD females showed a higher striatal AI than all the other subgroups. Of note, we found significant reduction between GBA-PD females and nonGBA-PD females at voxel level [MNI XYZ Coordinates: 22; 0; -4/-24; 0; -4]. No differences were observed between males in the GBA-PD and nonGBA-PD groups.

Off-line statistical comparison showed that despite a milder motor clinical profile (Supplementary Table 2), GBA-PD females showed the greatest dopaminergic impairment, which was significantly worse than nonGBA-PD males (Fig. 1b).

3.3 Cognitive trajectories at follow-up

3.3.1 Cognitive Deflection Scores

During the follow-up period (mean ± SD: 6.87 ± 3.2 years), 228 out of 560 PD patients (40.7%) experienced cognitive impairment, with MoCA scores ≤ 26 at follow-up assessment; this subgroup included 146 males (out of 335 = 43.6%), 66 GBA1 carriers (out of 128 = 51.6%) and 206 LO-PD (out of 560 = 36.8%) patients. “Severe” and “mild” GBA1 variants were associated with cognitive impairment in 57.1% and 54.3% of cases, respectively. Among the GBA1 carriers of “mild” variants, 64 (88.8%) had the N409S variant, and of them 33 (51.6%) developed cognitive impairment. Conversely, individuals carrying "risk" variants (40.6%) and those with "unknown" variants (25%) exhibited a lower prevalence of cognitive impairment (Supplementary Table 1).

We found that the levels of MoCA deflection were significantly higher in GBA-PD compared to nonGBA-PD. No significant difference in MoCA deflection was found between males (mean ± SD: -0.42 ± 1.1 points/year) and females (Table 1).

GBA-PD males showed a significantly steeper MoCA deflection not only when compared to nonGBA-PD, both males and females, but also when compared to GBA-PD females (Table 2).

3.3.2 Mixed Linear Model

The linear mixed-effects models showed that sex and AAO were associated with significant changes in MoCA scores across time. Notably, male PD patients had significantly faster declines in global cognition compared to female PD patients; and LO-PD compared to EO-PD patients. A significant interaction was observed between AAO and sex, with LO-PD males showing significantly higher declines in global cognition than EO-PD females, EO-PD males and almost significant than LO-PD females. No significant difference was observed among GBA-PD and nonGBA-PD. However, a significant interaction between sex and GBA1 positivity was found in the change of MoCA scores over time. GBA-PD males had significantly greater decline in global cognition than GBA-PD females and nonGBA-PD females. We observed a significant interaction in the decline of MoCA scores, due to a higher rate of decline in LO-GBA-PD group compared to EO-GBA-PD and EO-nonGBA-PD (Table 3). All the above results were replicated in the subcohort of PD patients without LRRK2 mutations.

Table 3

Summary of the main effects in models comparing the rates of change in MoCA scores.
	Estimate (95% CI)	p-value	Estimate (95% CI)*
Sex	0.55 (0.09 to 1.02)	0.019	0.87 (0.3 to 1.4)	0.000
GBA	0.26 (-0.30 to 0.80)	0.355	0.25 (-0.30 to 0.8)	0.385
AAO	1.78 (1.26 to 2.31)	0.000	2.07 (1.50 to 2.49)	0.000
Interaction
Sex*GBA (reference: GBA-PD Males)		0.041		0.005
GBA-PD Female	1.09 (0.13 to 2.05)	0.026	1.20 (0.23 to 2.17)	0.015
nonGBA-PD Female	1.05 (0.26 to 1.84)	0.009	1.35 (0.51 to 2.18)	0.002
nonGBA-PD Male	0.62 (-0.12 to 1.37)	0.099	0.55 (-0.19 to 1.31)	0.147
Sex*AAO (reference: LO-PD Males)		0.000		0.000
EO-PD Female	2.07 (1.31 to 2.84)	0.000	2.51 (1.66 to 3.35)	0.000
LO-PD Female	0.47 (-0.05 to 0.99)	0.077	0.86 (0.28 to 1.45)	0.004
EO-PD Male	1.89 (1.20 to 2.57)	0.000	2.27 (1.53 to 3.02)	0.000
GBA*AAO (reference: LO-GBA-PD)		0.000		0.000
EO-GBA-PD	2.14 (1.06 to 3.23)	0.000	2.16 (1.05 to 3.28)	0.000
EO-nonGBA-PD	2.14 (1.38 to 2.89)	0.000	2.41 (1.60 to 3.23)	0.000
LO-nonGBA-PD	0.44 (-0.17 to 1.06)	0.158	0.36 (-0.28 to 1.01)	0.269
Linear mixed-effects models adjusted for age, disease duration, education and MDS-UPDRS-III at baseline.
Abbreviations: MoCA, Montreal Cognitive Assessment; HR, Hazard Ratio; PD, Parkinson disease; GBA, patients with PD who carry the GBA1 mutation; AAO, Age at symptoms onset; EO-PD; PD patients with early age at onset; LO-PD PD patients with late age at onset.
*Analysis performed in a subgroup of patients without LRRK2 mutations.
Significant p-values (p < 0.05) are reported in bold.

3.3.3 Survival analysis – Main effects

GBA1 positivity produced a significant risk of cognitive impairment over time (Fig. 2b), which did not change when the model was adjusted for sex. Both “severe” and “mild” GBA1 variants were significantly associated with an increased risk of cognitive impairment. Of note, “risk” variants showed an HR below 1 indicative of a protective effect than the other variants (Table 4 and Supplementary Fig. 1).

Table 4

Summary of the Cox regression results on risk of cognitive impairment (MoCA score < 26).
	HR (95% CI)	p-value
Sex	1.19 (0.89 to 1.60)	0.235
GBA	1.82 (1.33 to 1.48)	0.000
Severe&Mild	3.40 (1.68 to 6.88)	0.001
Risk	0.46 (0.23 to 0.93)	0.030
Unknown	0.44 (0.15 to 1.27)	0.131
AAO	2.36 (1.54 to 3.61)	0.000
Interaction
Sex*GBA	2.03 (1.41 to 2.91)	0.000
Male Sex*Severe/Mild	3.24 (1.43 to 7.35)	0.005
Female Sex*Severe/Mild	3.21 (0.83 to 12.48)	0.092
Sex*AAO	1.64 (1.24 to 2.16)	0.000
GBA*AAO	1.83 (1.33 to 2.53)	0.000
Cox regression models adjusted for age, disease duration, education MDS-UPDRS-III at baseline.
Abbreviations: MoCA, Montreal Cognitive Assessment; HR, Hazard Ratio; PD, Parkinson disease; GBA, patients with PD who carry the GBA1 mutation; AAO, Age at symptoms onset; EO-PD; PD patients with early age at onset; LO-PD PD patients with late age at onset.
Significant p-values (p < 0.05) are reported in bold.

When we considered AAO, participants with LO-PD showed a significantly increased risk of cognitive impairment across time (Fig. 2c).

3.3.4 Survival analysis - Interaction effects

We found an interaction between male sex and GBA1 positivity on the risk of cognitive impairment (Fig. 2d). GBA-PD male carriers of "severe" and "mild" variants showed a significantly higher risk of cognitive impairment than “unknown” and “risk” variants (Supplementary Fig. 1).

LO-PD males showed a significantly higher risk of cognitive impairment, and the interaction effect between AAO and GBA1 positivity showed a higher risk of cognitive impairment in GBA-LO-PD. See Table 4 and Fig. 2e-f.

The results of this secondary analysis were consistent with our initial findings. The risks associated with sex, GBA1 positivity, and age at symptom onset (AAO), as well as their interactions, mirrored those observed in the Cox model for the broader population (Supplemental Table 6). This consistency indicates that the associations we reported are unlikely to be significantly affected by data censoring or the presence of LRRK2 mutations. Further details can be found in the Supplementary materials.

The interaction between sex and genetics is complex and poorly understood in the context of PD.⁸ Sex-related frequency differences have been reported in genetic forms of PD, with observed variation depending on the specific gene.¹⁶ As regards GBA1, the sex distribution remains controversial,^17–19 with discrepancies depending on the GBA1 variants under investigation.^4,16 For instance, a previous report indicated a preponderance of women among carriers of "severe" variants, while men were more likely to harbour "mild" and "risk" variants.¹⁶ Here, in a large cohort of PD subjects extracted from the PPMI dataset, we failed to detect relevant differences in the prevalence of GBA1 heterozygous carriers between men and women.

The main aim of this study was to investigate the combined role of sex and GBA1 carrier status in the risk of progression toward a cognitive impairment, to address the key question whether GBA1 mutations and sex have an independent or cumulative effect on cognitive outcomes. Our results highlighted, for the first time, that the combination of GBA1 mutations and male sex is associated with a higher risk of cognitive impairment and a steeper MoCA change along the disease course.

This novel finding is in keeping with previous research reporting male sex as a predictor of higher risk of developing cognitive decline²⁰ or even dementia,²¹ as well as with the known association of GBA1 variants with cognitive impairment.^5,22–24 To the best of our knowledge, only a previous study reported similar evidence in a large cohort of 4,923 subjects with primary degenerative parkinsonism, finding that GBA1 variants and male sex were associated with a higher proportion of subjects with PD-dementia and dementia with Lewy bodies than idiopathic PD.⁴

On the other hand, female biological sex seems to exert a protective effect also on GBA-PD condition. Indeed, we found no association between female sex and risk of cognitive decline in GBA-PD subjects, also supported by a slower cognitive decline in GBA-PD females than males. Taken together, these findings suggest the existence of relevant sex-related discrepancies in the manifestation of cognitive dysfunction in GBA-PD. Interestingly, despite a more benign clinical phenotype, GBA-PD females showed greater dopaminergic deficit as compared to GBA-PD males, suggesting that, in the course of the disease, GBA-PD females can counteract pathological brain changes through mechanisms of neural reserve and neural compensation.²⁵ In the general population, women tend to exhibit higher physiological levels of dopamine in the striatum, reflecting differences in basal dopamine system organization and/or neuroanatomy.²⁶ The dopamine system contains a high density of oestrogen receptors, through which hormones exert their protective role on dopaminergic functions.²⁶ Such protective effects of oestrogens are achieved by reducing oxidative stress and mitochondrial dysfunction, limiting neuroinflammation, and preventing the deposition of α-synuclein and neural injury.²⁷ Another aspect under investigation is related to the detrimental role of GBA1 mutations on sphingolipid homeostasis, the latter found to be modulated by means of oestrogen receptor.²⁸ Thus, both environmental and hormonal factors may counteract PD-related pathology over the lifetime of pre-menopausal women, contributing to build a neural reserve through relevant neurobiological effects, even in GBA1 carriers.²⁹ Overall, it is tempting to speculate that a more advanced stage of neurodegeneration is needed in females to reach the same clinical severity observed in GBA-PD males. Future prospective studies - focusing on the influence of hormones on GBA1-related pathology - could lead to a better understanding of the wide motor and cognitive between-sex variability in PD, as well as reveal new therapeutic avenues or preventive strategies.

Besides cognitive impairment, GBA-PD males showed higher occurrences of RBD disorders compared to all other groups. This finding is of particular interest as it is in line with the male predominance of RBD,³⁰ but also with the strong association between the presence of RBD and GBA1 carrier status.^5,31 RBD in PD subjects is considered a marker of a more malignant phenotype, with more rapid progression of motor and non-motor symptoms,³² as well as being related to more severe spreading of α-synuclein pathology at post-mortem assessment.³³

Of note, we found that older age at onset was, independently and in combination with both GBA1 mutations and male sex, associated with future development of cognitive impairment.

An older age at the onset of the disease has already been associated with increased risk of incident dementia in PD, due to a combined effect of the disease and aging process, possibly acting on nondopaminergic brainstem structures (i.e. cholinergic and noradrenergic systems).³⁴ Here, we found that AAO, together with male sex, modulates the detrimental effect of GBA1 mutation on PD cognitive decline.

Another aspect emerging from our analysis is related to the differential effects of GBA1 classes of variants on cognitive decline. Previous evidence showed conflicting results on the relationship between classes of GBA1 mutations and PD phenotype as well as the risk for dementia.^4,24,35,36 Indeed, Cilia et al. (2016) found that carriers of “severe” variants had greater risk for dementia compared to carrier of “mild” variants, even if the latter showed a 2-fold higher risk of dementia than nonGBA-PD subjects.³⁵ Our former study on a large Italian cohort also showed that GBA-PD patients with “severe” variants exhibited greater risk of non-motor symptoms (e.g. cognitive impairment) compared to GBA-PD carrier of “mild” variants.³⁷ Later on, Lunde et al. (2018) reported that “severe” variants are associated with a faster progression to dementia than carriers of “risk” polymorphisms (e.g. E326K),²⁴ a finding not confirmed by Straniero et al. (2020), who found a higher risk of dementia not only in association with “severe” variants, but also with the E326K “risk” variant.⁴

The present study on the PPMI cohort confirms that GBA1 “severe” variants are generally correlated with a more severe clinical phenotype, but it shows that “mild” variants also increase the risk of cognitive decline in PD. Conversely, a much lower proportion of "risk" variant carriers eventually converted to dementia at follow-up, suggesting that this category is characterized by a more benign outcome, especially on the cognitive side. The higher risk of conversion found in carriers of both “mild” and “severe” mutations may partially explain the considerable clinical variability reported among GBA-PD subjects in terms of cognitive dysfunction and motor disability.³⁵ However, we found that 46% of “mild” and 43% of “severe” variant carriers showed stable cognitive profiles at follow-up, suggesting that other risk factors, in combination with genetic risk factors, should be considered to shed light on PD heterogeneity. Overall, this still controversial evidence suggests that the current classification of GBA1 variants, which is based on their role in Gaucher's disease, may not adequately reflect their pathogenic role in PD, and new classification approaches should be investigated. Given these premises, future studies further addressing the issue of heterogeneity within the spectrum of GBA1 genotypes and its relationship with sex should be implemented, also considering the effect on PD motor and non-motor phenotype and clinical trajectories. Moreover, our data should be confirmed in large population-based studies to limit bias in the ascertainment.

This study has some limitations that should be acknowledged. First, in the survival analysis, cognitive impairment did not occur in 50% (effective 40.7%) of the sample, which restrict the interpretation and generalizability of the results. Additionally, the impact of concomitant medications, such as anticholinergics and comorbidities, was not considered for analysis, which may have influenced cognitive decline and confounded the results.

In conclusion, we confirm that GBA1 variants are the major risk factor associated with cognitive impairment, however, this effect is particularly evident in association with the male sex. Indeed, we found that, among GBA1 carriers (mainly of “severe” and “mild” variants), PD males showed the greatest risk of develop cognitive impairment over time. These elements should be considered when interpreting the current literature and planning future studies. Understanding the role of genetic variants on the course of cognitive decline over PD progression will foster a more accurate disease prognosis and may help to a better future clinical trials design and patients’ selection. In particular, the effect of sex on GBA1 mutation should be considered in the emerging therapeutic strategies targeting GBA1-regulated pathways.³⁸

Author contributions

SPC and EMV contributed to the conception and design of the study. SPC, AG, RM, GC, CG, RC contributed to the acquisition and analysis of data. SPC, AG, RM contributed to drafting the text and preparing the ﬁgures. MA, APi, AP, FB, DP, CT, EMV Interpretation of the data; critical revision of the manuscript.

Acknowledgements

SPC is supported by #NEXTGENERATIONEU (NGEU) and funded by the Ministry of University and Research (MUR), National Recovery and Resilience Plan (NRRP), project MNESYS (PE0000006) – A multiscale integrated approach to the study of the nervous system in health and disease (DN. 1553 11.10.2022).

CT has received personal fees for participating in advisory boards for Eli Lilly.

Funding: PPMI – a public-private partnership – is funded by the Michael J. Fox Foundation for Parkinson’s Research and funding partners, including 4D Pharma, Abbvie, AcureX, Allergan, Amathus Therapeutics, Aligning Science Across Parkinson's, AskBio, Avid Radiopharmaceuticals, BIAL, Biogen, Biohaven, BioLegend, BlueRock Therapeutics, Bristol-Myers Squibb, Calico Labs, Celgene, Cerevel Therapeutics, Coave Therapeutics, DaCapo Brainscience, Denali, Edmond J. Safra Foundation, Eli Lilly, Gain Therapeutics, GE HealthCare, Genentech, GSK, Golub Capital, Handl Therapeutics, Insitro, Janssen Neuroscience, Lundbeck, Merck, Meso Scale Discovery, Mission Therapeutics, Neurocrine Biosciences, Pfizer, Piramal, Prevail Therapeutics, Roche, Sanofi, Servier, Sun Pharma Advanced Research Company, Takeda, Teva, UCB, Vanqua Bio, Verily, Voyager Therapeutics, the Weston Family Foundation and Yumanity Therapeutics.

Competing interests

The authors report no competing interests relevant to this study.

Data availability

All the data used in this study are publicly available in the PPMI repository (www.ppmi-info.org/

access-data-specimens/download-data).

The datasets generated and analysed during the current study are not publicly available due to PPMI repository (www.ppmi-info.org/access-data-specimens/download-data).

Ethics approval and consent to participate

Each PPMI participating site received approval from their local ethic committee before study initiation, and written informed consent was obtained from all participants before enrolment, in full compliance with the principles set out by the Declaration of Helsinki. We have obtained permission for publishing our research from the Data and Publication Committee of the PPMI study.

Obeso, J. A. et al. Past, present, and future of Parkinson’s disease: A special essay on the 200th Anniversary of the Shaking Palsy. Mov Disord 32, 1264–1310 (2017).
Leroi, I. et al. Behavioural disorders, disability and quality of life in Parkinson’s disease. Age Ageing 40, 614–621 (2011).
Halliday, G. M., Leverenz, J. B., Schneider, J. S. & Adler, C. H. The neurobiological basis of cognitive impairment in Parkinson’s disease. Movement Disorders 29, 634–650 (2014).
Straniero, L. et al. The SPID-GBA study: Sex distribution, Penetrance, Incidence, and Dementia in GBA-PD. Neurol Genet 6, (2020).
Caminiti, S. P., Carli, G., Avenali, M., Blandini, F. & Perani, D. Clinical and Dopamine Transporter Imaging Trajectories in a Cohort of Parkinson’s Disease Patients with GBA Mutations. Mov Disord 37, 106–118 (2022).
Gan-Or, Z., Liong, C. & Alcalay, R. N. GBA-Associated Parkinson’s Disease and Other Synucleinopathies. Curr Neurol Neurosci Rep 18, 44 (2018).
Georgiev, D., Hamberg, K., Hariz, M., Forsgren, L. & Hariz, G.-M. Gender differences in Parkinson’s disease: A clinical perspective. Acta Neurol Scand 136, 570–584 (2017).
Cerri, S., Mus, L. & Blandini, F. Parkinson’s Disease in Women and Men: What’s the Difference? J Parkinsons Dis 9, 501–515 (2019).
Vaidya, B., Dhamija, K., Guru, P. & Sharma, S. S. Parkinson’s disease in women: Mechanisms underlying sex differences. Eur J Pharmacol 895, 173862 (2021).
Parkinson Progression Marker Initiative. The Parkinson Progression Marker Initiative (PPMI). Prog Neurobiol 95, 629–35 (2011).
Marek, K. et al. The Parkinson’s progression markers initiative (PPMI) – establishing a PD biomarker cohort. Ann Clin Transl Neurol 5, 1460–1477 (2018).
Schirinzi, T. et al. Young-onset and late-onset Parkinson’s disease exhibit a different profile of fluid biomarkers and clinical features. Neurobiol Aging 90, 119–124 (2020).
Laird, N. M. & Ware, J. H. Random-effects models for longitudinal data. Biometrics 38, 963–74 (1982).
Cox, D. R. Regression Models and Life-Tables. J R Stat Soc Series B Stat Methodol 34, 187–202 (1972).
Litvan, I. et al. Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Movement Disorders 27, 349–356 (2012).
Ortega, R. A. et al. Differences in Sex‐Specific Frequency of Glucocerebrosidase Variant Carriers and Familial Parkinsonism. Movement Disorders 37, 2217–2225 (2022).
Simuni, T. et al. Correlates of excessive daytime sleepiness in de novo Parkinson’s disease: A case control study. Movement Disorders 30, 1371–1381 (2015).
Setó‐Salvia, N. et al. Glucocerebrosidase mutations confer a greater risk of dementia during Parkinson’s disease course. Movement Disorders 27, 393–399 (2012).
Malek, N. et al. Features of GBA-associated Parkinson’s disease at presentation in the UK Tracking Parkinson’s study. J Neurol Neurosurg Psychiatry 89, 702–709 (2018).
Reekes, T. H. et al. Sex specific cognitive differences in Parkinson disease. NPJ Parkinsons Dis 6, 7 (2020).
Cholerton, B. et al. Sex differences in progression to mild cognitive impairment and dementia in Parkinson’s disease. Parkinsonism Relat Disord 50, 29–36 (2018).
Stoker, T. B. et al. Impact of GBA1 variants on long-term clinical progression and mortality in incident Parkinson’s disease. J Neurol Neurosurg Psychiatry 91, 695–702 (2020).
Szwedo, A. A. et al. GBA and APOE Impact Cognitive Decline in Parkinson’s Disease: A 10-Year Population-Based Study. Movement Disorders 37, 1016–1027 (2022).
Lunde, K. A. et al. Association of glucocerebrosidase polymorphisms and mutations with dementia in incident Parkinson’s disease. Alzheimer’s & Dementia 14, 1293–1301 (2018).
Barulli, D. & Stern, Y. Efficiency, capacity, compensation, maintenance, plasticity: emerging concepts in cognitive reserve. Trends Cogn Sci 17, 502–509 (2013).
Zachry, J. E. et al. Sex differences in dopamine release regulation in the striatum. Neuropsychopharmacology 46, 491–499 (2021).
Bovenzi, R. et al. Sex hormones differentially contribute to Parkinson disease in males: A multimodal biomarker study. Eur J Neurol 30, 1983–1990 (2023).
Marin, R. & Diaz, M. Estrogen Interactions With Lipid Rafts Related to Neuroprotection. Impact of Brain Ageing and Menopause. Front Neurosci 12, 128 (2018).
Subramaniapillai, S., Almey, A., Natasha Rajah, M. & Einstein, G. Sex and gender differences in cognitive and brain reserve: Implications for Alzheimer’s disease in women. Front Neuroendocrinol 60, 100879 (2021).
Dauvilliers, Y. et al. REM sleep behaviour disorder. Nat Rev Dis Primers 4, 19 (2018).
Gan‐Or, Z. et al. GBA mutations are associated with Rapid Eye Movement Sleep Behavior Disorder. Ann Clin Transl Neurol 2, 941–945 (2015).
Fereshtehnejad, S.-M., Zeighami, Y., Dagher, A. & Postuma, R. B. Clinical criteria for subtyping Parkinson’s disease: biomarkers and longitudinal progression. Brain 140, 1959–1976 (2017).
Postuma, R. B. et al. REM sleep behavior disorder and neuropathology in Parkinson’s disease. Movement Disorders 30, 1413–1417 (2015).
Levy, G. et al. Combined effect of age and severity on the risk of dementia in Parkinson’s disease. Ann Neurol 51, 722–729 (2002).
Cilia, R. et al. Survival and dementia in GBA-associated Parkinson’s disease: The mutation matters. Ann Neurol 80, 662–673 (2016).
Liu, G. et al. Genome-wide survival study identifies a novel synaptic locus and polygenic score for cognitive progression in Parkinson’s disease. Nat Genet 53, 787–793 (2021).
Petrucci, S. et al. GBA‐Related Parkinson’s Disease: Dissection of Genotype–Phenotype Correlates in a Large Italian Cohort. Movement Disorders 35, 2106–2111 (2020).
Huh, Y. E., Usnich, T., Scherzer, C. R., Klein, C. & Chung, S. J. GBA1 Variants and Parkinson’s Disease: Paving the Way for Targeted Therapy. J Mov Disord 16, 261–278 (2023).

(Not answered)

Download PDF

Editorial decision: revise
15 Sep, 2024
Review #1 received at journal
06 Sep, 2024
Review #3 received at journal
05 Sep, 2024
Review #2 received at journal
03 Sep, 2024
Reviewer #3 agreed at journal
30 Aug, 2024
Reviewer #2 agreed at journal
29 Aug, 2024
Reviewer #1 agreed at journal
27 Aug, 2024
Reviewers invited by journal
26 Aug, 2024
Editor assigned by journal
14 Aug, 2024
Submission checks completed at journal
13 Aug, 2024
First submitted to journal
13 Aug, 2024
Unknown event
13 Aug, 2024

You are reading this latest preprint version

Male sex accelerates cognitive decline in GBA1 Parkinson’s Disease

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1 Subjects

2.2 Clinical assessment at baseline

2.3 Brain Dopaminergic Activity at baseline

2.4 Cognitive trajectories at follow-up

3. Results

3.1 Baseline features

3.2 Dopamine Transporter Imaging at baseline

3.3 Cognitive trajectories at follow-up

3.3.1 Cognitive Deflection Scores

3.3.2 Mixed Linear Model

3.3.3 Survival analysis – Main effects

3.3.4 Survival analysis - Interaction effects

4. Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1