Symmetric and Profound Monoaminergic Degeneration in Parkinson’s Disease with Premotor REM Sleep Behavior Disorder

INTRODUCTION

Idiopathic rapid eye movement (REM) sleep behavior disorder (iRBD), characterized by dream enactment behavior and loss of REM sleep atonia, serves as a robust prodromal indicator for Parkinson’s disease (PD), with approximately 70% developing synucleinopathy within 10 years [1–3]. However, from the PD standpoint, premotor RBD is only observed in a subset of newly diagnosed patients, with an estimated prevalence ranging from 25% to 40%, and many PD patients experience RBD after the onset of motor symptoms [4–7].

The heterogeneous manifestation of RBD in PD has prompted efforts to subtype the disease based on its presence. In PD, the co-presence of RBD is associated with a ‘diffuse and/or malignant’ subtype of disease, along with a higher risk of dementia, psychiatric comorbidity, and autonomic dysfunction [8–12]. The recent α-Synuclein Origin and Connectome (SOC) Model of PD suggests that premotor RBD signifies a “body-first” PD subtype characterized by a bottom-up spread of peripheral synucleinopathy through the brainstem [13]. This subtype is expected to exhibit a more symmetrical nigrostriatal degeneration pattern compared to the “brain-first” subtype, where pathology likely begins at a unilateral site inside the brain.

Meanwhile, a small yet distinct subgroup of longstanding iRBD patients, either undergoing phenoconversion after more than 10 years or remaining nonconverted for an extended period, is consistently observed across multiple prospective studies [14, 15]. However, considering the anatomical proximity of RBD’s key brainstem neuronal network to nigral dopaminergic neurons [16], the remarkably prolonged lag period between RBD and motor parkinsonism raises questions about whether an alternative neurodegenerative pathomechanism should be contemplated [17] rather than the straightforward progression of premotor RBD into “body-first” PD, at least in some instances among these long-lagged PD cases.

Given these considerations, in a cohort of newly diagnosed PD patients undergoing FP-CIT positron emission tomography (PET), we aimed to analyze potential differences in brain monoaminergic denervation patterns based on the presence and temporal profile of premotor RBD. Specifically, we hypothesized that, in the premotor RBD group, the proposed pathomechanism of “body-first” PD would lead to (i) more symmetrical nigrostriatal dopaminergic denervation and (ii) more profound pontine monoaminergic degeneration, both in terms of absolute values and relative ratios to nigrostriatal denervation, as the pathological propagation is suggested to be ‘bottom-up’ [13]. Furthermore, we hypothesized that the symmetric nigrostriatal and severe pontine denervation patterns expected for “body-first” PD would be more pronounced in the subgroup with a temporal gap of less than 10 years between the onset of RBD symptoms and motor parkinsonism compared to the group with an extensive lag period.

MATERIALS AND METHODS

RESULTS

A total of 150 subjects with de novo PD were included in this study. The demographics and clinical characteristics are presented in Table 1. The mean age at PD onset was 66.88±5.667 years, and at PET scan 67.98±5.536 years. Ninety-eight subjects had probable RBD at the time of PD onset (PD^RBD +), 43 of whom reported that the onset of RBD preceded that of PD by 10 years or more (PD^RBD +≥10Y).

Striatal dopaminergic asymmetry and striatum-to-pons ratios of monoaminergic denervation by probable premotor RBD in PD

Although no statistically significant differences were observed between the PD^RBD + and PD^RBD– groups concerning sex, age at PD onset, and age at PET imaging, distinctions emerged in striatal and pontine FP-CIT uptakes, demonstrating significantly lower values in both the striatum VOIs and the pons in the PD^RBD + group (p_FDR < 0.05). Analyzing striatal asymmetry between groups revealed a more symmetric pattern of FP-CIT binding in the putamen (p_FDR < 0.001) and the caudate nucleus (p_FDR < 0.05) in the PD^RBD + group compared to the PD^RBD– group, while ventral striatal AIs did not exhibit significant differences (Fig. 1a). Furthermore, the PD^RBD + group exhibited a higher striatum-to-pons ratio of FP-CIT binding in all striatal VOIs, i.e., putamen/pons, compared to the PD^RBD– group (p_FDR < 0.05; Fig. 1b), indicating a relatively pronounced decrease in pontine SURs over the striatum in the PD^RBD + group.

OPEN IN VIEWER

Fig. 1

Striatal asymmetry and striatum-to-pons ratios of monoaminergic degeneration in drug-naïve Parkinson’s disease stratified by premotor RBD. The PD^RBD + group displays (a) a more symmetric nigrostriatal degeneration and (b) higher striatum-to-pons ratios of FP-CIT binding, indicating a marked pontine monoaminergic degeneration compared to the PD^RBD– group. (c) Putaminal asymmetry is primarily driven by the severity of least-affected hemisphere (LAH) degeneration. (d) The symmetric degeneration pattern of the striatum is more pronounced in the subgroup of PD^RBD + with less than 10 years of premotor RBD history.

DISCUSSION

In this cohort of de novo PD patients, our investigation reveals that individuals with probable premotor RBD (PD^RBD +) exhibit significantly more symmetrical striatal dopaminergic denervation than those without premotor RBD (PD^RBD–). The relatively higher severity of LAH denervation in PD^RBD +, compared to PD^RBD–, contributes to the symmetry, while MAH severity is comparable between the two groups. Within the PD^RBD + subgroups, individuals with a premotor RBD duration of less than 10 years exhibit a particularly more symmetrical pattern than the PD^RBD– group. Additionally, the PD^RBD + group shows lower pontine FP-CIT uptake than the PD^RBD– group, even after adjusting for striatal uptake (i.e., a higher striatum/pons ratio), suggesting a link between premotor RBD and early degeneration of the pontine monoaminergic nuclei.

We observe profound degeneration of the striatal LAH in the de novo PD^RBD + group, despite comparable severity in the MAHs when compared to the PD^RBD– group. This aligns with the expectations for “body-first” PD: a symmetric and higher overall burden of nigrostriatal degeneration at the time of diagnosis [13, 30]. The SOC model postulates that the initial gastrointestinal pathology in the “body-first” subtype ascends via bilateral dorsal motor nuclei of the vagus, resulting in an extended prodromal phase and a more extensive accumulation of bilateral pathology until motor onset [13]. Given the relatively stereotypical trajectory of dopamine degeneration once initiated [31–34], the comparable severity in the MAHs indicates that the initial dopamine degeneration likely started at similar timepoints in the past across the PD^RBD + and PD^RBD– groups. However, in the PD^RBD + group, it may have progressed more symmetrically, leading to a greater involvement of the LAH at the diagnostic stage. Previous research reported symmetric degeneration in the PD^RBD + group compared to the PD^RBD– group, but as the groups significantly differed in ages [35], the ability to compare the absolute severity of neurodegeneration in the LAH or the MAH was constrained. Our cohort, comprising age- and disease-duration-matched groups, each with over 40 subjects, enabled a quantitative comparison of dopaminergic neurodegeneration and symmetry in the postulated “body-first” and “brain-first” groups of de novo patients. Our findings also align with the pathological observation that the extent and density of synuclein pathology are higher in PD patients with probable RBD [36].

The extrastriatal investigation of FP-CIT binding reveals reduced pontine SURs in the PD^RBD + group, further supporting the assumptions in the SOC model that “body-first” PD has more damage in the pontine modulatory neurotransmitter systems, such as the raphe nucleus (RN) or the locus coeruleus (LC), in the early stages [13]. FP-CIT has affinity for serotonin (SERT) and noradrenaline transporters (NAT) in addition to dopamine transporters (DAT) [37]. Therefore, its binding at the pons level, where DATs are scarce, mainly reflects the somatodendritic availability of SERTs in the RN or the NATs in the LC, while at the striatal level, it reflects the predominant DATs at the nigral nerve terminals [37–39]. The higher striatum-to-pons ratio of SURs in the PD^RBD + group further supports the notion that degeneration of pontine monoaminergic nuclei is more severe, even after adjusting for the degree of nigrostriatal dopaminergic degeneration. Apart from the SOC model, this aligns with classical evidence that brainstem serotonergic and noradrenergic alterations contribute to RBD pathogenesis, along with cholinergic systems [16].

Our results illustrate that the striatal DAT symmetry of PD^RBD + is particularly pronounced in the PD^RBD +<10Y subgroup. Large-scale multicenter studies on iRBD report overall phenoconversion rates of 70–75% after 10–12 years [1, 3], indicating that at least 10% may remain nonconverted after a decade. We specifically focused on this minority of RBD patients with a prolonged lag period between the RBD and motor parkinsonism [17]. We hypothesized that this group may comprise a mix of patients, including 1) those with a markedly gradual progression of prodromal body-first parkinsonian pathology, and 2) those with non-progressive RBD and an independent development of brain-first PD. The results, indicating the most prominent “body-first-like” striatal symmetry trend in the PD^RBD +<10Y group, may support this hypothesis. Our findings suggest a possibility that a subset of these patients may experience a predominant “brain-first” pathomechanism independently occurring from RBD, because otherwise the PD^RBD +≥10Y group should show reduced putaminal FP-CIT binding and symmetry as in the PD^RBD +<10Y group. As part of ongoing efforts to subtype PD based on premotor RBD, considering the duration of preceding RBD may prove beneficial in identifying patient subgroups at a higher risk of faster progression and/or dementia.

Some limitations of our study should be addressed. Firstly, the absence of video polysomnography (PSG) confirmation of RBD poses a limitation. The RBD Single-Question Screen, the RBD1Q, is a versatile tool to screen for RBD in the clinics and poses an estimated specificity of 87% and sensitivity of 94% for PSG-confirmed RBD [19]. Still, the proportion of RBD-positive cases at diagnosis in our study was relatively higher than previously reported figures [4]. This could have been influenced by the questionnaire-based nature of the data and our requirement for both the patient and the bed partner to be present to answer the questionnaire. In other words, the inclusion of false-positive cases with secondary RBD and exclusion of cases without a bed partner, with or without RBD, might have impacted the rate in either way. We speculate that future investigations in a PSG-confirmed cohort may settle these limitations, providing valuable insights and validating our findings. Additionally, the inclusion criteria for PD duration and age at scan, implemented to control group differences, might have unintentionally affected the positivity rate. Second, the accuracy of the reported time of RBD onset may have been affected by recall bias. However, it is essential to note that 1) bed partners confirmed the onset of RBD in all cases, enhancing the reliability of reported onset times, and 2) the majority of cases (87.3%) specifying an RBD–PD interval of less than 10 years clarified that it fell within the 1–5-year range. This suggests a lower likelihood of confusion between durations less than 10 years and more than 10 years by patients and their bed partners. Thirdly, the absence of quantitative evaluation data for motor and nonmotor symptoms at the time of PET scan hinders the exploration of clinical correlations. Lastly, as the VOIs were delineated on the PET images using the template, and individual MRIs were not available, unrecognized atrophy could have influenced the FP-CIT SURs, particularly in the brainstem. A structural evaluation combined with PET analysis in the future is warranted.

In conclusion, our study highlights the association between premotor RBD and pronounced, symmetric degeneration of nigrostriatal dopaminergic nerve terminals in de novo PD, particularly evident in individuals developing PD within 10 years of RBD onset. Furthermore, premotor RBD is correlated with significant degeneration of the pontine monoaminergic nuclei. These findings indicate that premotor RBD in PD is linked to a subtype of the disease characterized by a more profound neurodegeneration, potentially suggesting a “widespread, bottom-up” pathophysiological mechanism.