Cutaneous α-synuclein is correlated with autonomic impairment in isolated rapid eye movement sleep behavior disorder

Abstract

Study Objectives

To define the clinical implications of cutaneous phosphorylated α-synuclein (p-syn) and its association with subjective and objective measures of autonomic impairment and clinical features including antidepressant use in isolated rapid eye movement (REM) sleep behavior disorder (iRBD).

Methods

Twenty-five iRBD patients had quantified neurological and cognitive examinations, olfactory testing, questionnaires, autonomic function testing, and 3 punch skin biopsies (distal thigh, proximal thigh, neck). Skin biopsies were stained for the pan-axonal marker PGP 9.5 and co-stained with p-syn, and results were compared to 28 patients with Parkinson’s disease (PD) and 18 healthy controls. Equal numbers of iRBD patients on and off antidepressants were recruited. The composite autonomic severity scale (CASS) was calculated for all patients.

Results

P-syn was detected in 16/25 (64%) of iRBD patients, compared to 27/28 (96%) of PD and 0/18 controls. The presence of p-syn at any biopsy site was correlated with both sympathetic (CASS adrenergic r = 0.6, p < 0.05) and total autonomic impairment (CASS total r = 0.6, p < 0.05) on autonomic reflex testing in iRBD patients. These results were independent of the density of p-syn at each site. There was no correlation between p-syn and antidepressant use.

Conclusions

In patients with iRBD, the presence of cutaneous p-syn was detected in most patients and was associated with greater autonomic dysfunction on testing. Longitudinal follow-up will aid in defining the predictive role of both skin biopsy and autonomic testing in determining phenoconversion rates and future disease status.

Statement of Significance.

This is the first study to report on cutaneous p-syn findings in a US cohort of iRBD patients, the first to correlate p-syn positivity with objective markers of autonomic function in this population, and the first to report on the association of antidepressant use and cutaneous p-syn positivity. Our results confirmed a high positivity rate of cutaneous p-syn and demonstrated even higher rates of autonomic dysfunction in iRBD patients. Those patients with markers of sympathetic adrenergic impairment on autonomic testing were more likely to be p-syn positive, raising the possibility of a shared pathophysiological mechanism. Furthermore, there was no correlation between cutaneous p-syn positivity and antidepressant use in iRBD patients. These data will be important in collaboration with other centers to better inform the use of skin biopsy and autonomic testing as potential biomarkers in disease-modifying trials.

Introduction

The association of isolated rapid eye movement (REM) sleep behavior disorder (iRBD) with neurodegenerative disorders of α-synuclein (α-syn) deposition—including Parkinson’s disease (PD), multiple system atrophy (MSA), and dementia with Lewy bodies (DLB)—has been well established in longitudinal studies, with >80% of patients phenoconverting to one of these diseases within their lifetime [1–3]. The pathological hallmark of the α-synucleinopathies is the presence of phosphorylated α-syn (p-syn) within the central and peripheral nervous system. Recent evidence suggests that p-syn deposition is common in iRBD and occurs within cutaneous autonomic nerves [4–7], though its association with phenoconversion is unclear.

Autonomic dysfunction is also a common prodromal feature of the α-synucleinopathies, including those with iRBD [8]. Prior studies in PD have demonstrated an association between cutaneous p-syn and autonomic dysfunction [9], however, the association between autonomic dysfunction and cutaneous p-syn deposition in iRBD is unknown. The aims of this study were thus to determine (1) the positivity rates of cutaneous p-syn in a group of carefully phenotyped iRBD patients, (2) the relationship between p-syn deposition and standardized measures of autonomic function, and (3) the relationship between p-syn deposition and clinical characteristics including disease duration, autonomic symptom burden, and antidepressant use, given the incompletely understood phenomenon of drug-induced RBD. We hypothesized that cutaneous p-syn would correlate with autonomic dysfunction as seen on standardized autonomic reflex testing in iRBD patients and would be unrelated to antidepressant use. Ultimately, these data may prove useful to include in longitudinal studies to better assess phenoconversion risk.

Methods

Twenty-five participants with iRBD were prospectively recruited from the Stanford neurology and sleep medicine clinics. The diagnosis of iRBD was based on ICSD-3 criteria [10] with confirmation of REM sleep without atonia (RSWA) on attended video polysomnography (vPSG) with a combination of submentalis, upper extremity, and lower extremity electromyography (EMG). RSWA was evaluated according to SINBAR criteria [11], using a 27% cut-off for loss of REM sleep atonia using a combination of chin and upper extremity EMG. All patients with an apnea-hypopnea index (AHI) of >15/h were excluded unless obstructive sleep apnea was adequately treated with positive airway pressure therapy. AHI scoring was based on 4% desaturation criteria. Exclusion criteria for a diagnosis of iRBD included signs of parkinsonism, ataxia, autonomic failure, dementia, and other medical conditions or medications associated with RBD, except antidepressant use. Patients were excluded if they had a pre-existing diagnosis of autonomic failure or known causes of autonomic dysfunction, such as uncontrolled diabetes, vitamin B12 deficiency, or alcohol abuse. We also excluded those with a history of post-traumatic stress disorder (PTSD), given the association of RSWA with trauma-associated sleep disorder in those on antidepressants. To assess the relationship of p-syn positivity to antidepressant use, a similar number of patients on and off antidepressant medication were recruited.

A comparison group of 28 patients with mild PD and 18 healthy controls were selected from an existing cohort at the Beth Israel Deaconess Medical Center (BIDMC). All patients with PD were diagnosed by a movement disorders specialist based on current Movement Disorder Society clinical diagnostic criteria [12].

Neurological evaluation, cognitive testing, and olfaction

All participants at Stanford and BIDMC had a complete neurological examination including the Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) part III motor exam [13] and the modified Hoehn & Yahr scale (H&Y) [14]. Only patients classified as H&Y I–II were included in the PD group. Both H&Y and MDS-UPDRS examination scores were measured during the off stage after holding PD medications overnight. iRBD patients were evaluated with both the MDS-UPDRS III and the Unified Multiple System Atrophy Rating Scale (UMSARS) [15]. Particular attention was paid to the presence of subtle motor abnormalities including mild bradykinesia, decreased blinking frequency or reduced facial expressions, reduced unilateral arm-swing, or mild postural instability. iRBD patients underwent assessment of olfaction with the University of Pennsylvania Smell Identification Test-40 (UPSIT-40) [16], and assessment of cognitive function with the Montreal Cognitive Assessment (MoCA) [17].

Autonomic, sleep, and quality of life questionnaires

iRBD patients completed a battery of questionnaires including the Composite Autonomic Symptom Scale-31 (COMPASS-31), a 31-item questionnaire addressing orthostatic, vasomotor, secretomotor, gastrointestinal, urinary, and pupillomotor domains [18]. The severity of orthostatic intolerance was assessed with the 10-item Orthostatic Hypotension Questionnaire (OHQ) [19]. Sleepiness was assessed with the Epworth Sleepiness Scale (ESS) [20]. Quality of life was assessed with the Schrag Quality of Life (QoL) scale [21]. All participants (iRBD, PD, and controls) completed the RBD single question screen (RBD1Q) to assess dream-enacting behaviors [22].

Autonomic function testing

All participants underwent a standard battery of autonomic reflex tests in a quiet, temperature-controlled room. Participants were instructed to withhold caffeine, alcohol, and nicotine consumption from the previous evening. Continuous electrocardiographic RR intervals were recorded from three precordial electrodes. Beat-to-beat BP was measured with finger plethysmography. BP readings were confirmed with an automated cuff sphygmomanometer over the brachial artery. As a measure of sympathetic cholinergic function, participants underwent quantitative sudomotor axon reflex testing (QSART) using a Q-Sweat machine (WR Medical Electronics, Stillwater, MN). Standard measurement sites of the forearm, proximal leg, distal leg, and foot were selected. As a measure of parasympathetic function, participants were verbally coached to breathe at six cycles per minute, and respiratory sinus arrhythmia during deep-paced breathing was calculated from the average of the three longest RR intervals during expiration divided by the average of the three shortest RR intervals during inspiration (expiratory:inspiratory [E:I] ratio). Participants performed a standardized Valsalva maneuver, maintaining an expiratory pressure of >30 mmHg for at least 10 s. Participants were then tilted upright to an angle of 70-degrees for at least 10 min. Data were collected and analyzed with Test works software (WR Medical Electronics). The composite autonomic severity scale (CASS), a well-defined quantification of autonomic function testing results comprised of cardiovagal, adrenergic, and sudomotor domains [23], was calculated for each patient.

Skin biopsy

One 3 mm, punch skin biopsy sample was taken from the right proximal thigh, right distal thigh, and right C7 paraspinal area after local anesthesia with 2% lidocaine. Skin biopsies were placed in Zamboni fixative solution for 24 h and then placed in cryoprotectant solution (20% glycerol and 20% 0.4 M Sorensen buffer).

Immunohistochemistry

Biopsy tissues were cut by freezing microtome into 50-μm sections. A total of six tissue sections were analyzed per biopsy (for a total of 18 sections per patient across the 3 biopsy sites). Fluorescent immunohistochemical staining was performed using previously published methods for co-visualizing all p-syn positive nerve fibers with the pan-axonal marker protein gene product 9.5 (PGP 9.5) [24]. Briefly, 50 μm thick tissue sections were placed in primary antibody solution (anti-phosphorylated α-synuclein—Covance, NJ, USA) for 24 h in free-floating 96 well plates, followed by incubation with biotin-conjugated anti-mouse (Jackson ImmunoResearch Lab) then visualized with streptavidin-conjugated fluorescence dye Cy3 (Jackson ImmunoResearch). The sections were then washed and incubated with rabbit anti-protein gene product 9.5 (UltraClone, Isle of Wight, UK) for 12 h at room and visualized by rabbit conjugated fluorescence dye Cy2 (Jackson ImmunoResearch Lab). All test samples were run with both positive and negative controls.

Confocal imaging

All tissue sections were imaged by confocal microscopy (Zeiss LSM5 Pascal Exciter; Carl Zeiss, Thornwood, NY). A series of images of optical sections were acquired at 3-μm intervals throughout the depth of the 50-μm section as a z-stack (Lens Plan–Apochromat ×20/0.8; Carl Zeiss).

Determination of p-syn positivity and nerve fiber density

As described previously [25], a nerve fiber was only considered positive for p-syn if (1) the p-syn completely co-localized with the PGP 9.5 positive nerve fiber, (2) was visible and tracked in and out of the plane of focus with the PGP 9.5 positive nerve fiber using single-frame images, and (3) all regions of tissue damage (from crush or staining artifact) were excluded. Raters were blinded to all clinical data and diagnoses. Intra-epidermal nerve fiber density (IENFD) was calculated using standard methodology using 50 μm florescent immunostained nerve fibers [26]. A total of four tissue sections from each biopsy site were counted. Results are reported as nerve fibers/millimeters.

Standard protocol approvals, registrations, and patient consents

The study was approved by the institutional review boards of Stanford University and BIDMC and all participants signed informed consent.

Statistical analysis

Categorical variables are presented as percentages, and continuous variables as median and interquartile range for non-Gaussian variables, as confirmed by the Shapiro-Wilk test for normality. For non-Gaussian variables, group-wise comparisons of continuous variables were performed using the Kruskal-Wallis test (multiple group comparisons) and post hoc pairwise comparisons were performed with the Wilcoxon rank sum test, only in instances where the Kruskal-Willis p value was below the significance threshold. Comparatively, for normally distributed variables, group-wise comparisons of continuous variables were performed using ANOVA (multiple group comparisons) and post hoc pairwise comparisons were performed with the t-test, only in instances where the ANOVA p-value was below the significance threshold. χ² or Fisher’s exact test (when counts fell below five in any category) was used to compare categorical variables between groups. Spearman correlation coefficients were calculated between the p-syn positivity at any biopsy site (expressed as a binary variable) and various clinical features of interest. A statistical threshold of α = 0.05 was set and Bonferroni correction for multiple comparisons was performed for each major analysis. All methods were implemented in the R programming language v4.0.3 (Vienna, Austria).

Data availability statement

Raw data that supports the findings of this study were generated at the neurology departments of Stanford and BIDMC. Derived data supporting the findings of this study are available from the corresponding author on request.

Results

Demographics and clinical features

Patients with iRBD, PD, and controls were of similar age, gender, and body mass index (Table 1). Most iRBD and PD patients were male. The average disease duration in the iRBD group was 7.54 ± 6.51 years, with similar numbers of patients on and off antidepressant medication. Of those taking antidepressants, five were taking a selective serotonin reuptake inhibitor (SSRI: citalopram, escitalopram, trazadone), five were taking a serotonin and norepinephrine reuptake inhibitor (SNRI: duloxetine, venlafaxine, desvenlafaxine), and three were taking a serotonin and dopamine reuptake inhibitor (SDRI: bupropion). The mean duration of antidepressant use in these patients was 12.7 years. Mild cognitive impairment, as defined by a MoCA score of <26, was present in 5/25 (20%) of iRBD patients, however, mean and median scores were normal and there was no statistical difference when compared to PD patients and controls. Olfactory function was abnormal in 20/25 (80%) of iRBD patients. These results are summarized in Table 1 and the Supplementary table.

Table 1.

Demographics and clinical features

	iRBD (n = 25)	PD (n = 28)	Controls (n = 18)	P	Pairwise comparison
Age (years)	66.2 ± 7.62	62.46 ± 5.9	63.5 ± 5.85	0.12	n/a
Male (%)	17/25 (68%)	20/28 (71%)	10/18 (63%)	0.83	n/a
BMI (kg/m2)	25.7 [24.5,30]	24.1 [23.3,26.6]	26.5 [25.5,28.6]	0.058	n/a
Antidepressant use (%)	12/25 (48%)	9/28 (32%)	2/18 (11%)	0.035	n/a
RBD1Q positive (%)	25/25 (100%)	23/28 (82%)	0/18 (0%)	1.34 × 10 −13	Ctrl <PD < iRBD
MDS-UPDRS III	0 [0,1]	29 [22.3,41.0]	0 [0,0]	1.03 × 10 −12	Ctrl < iRBD < PD
Hoehn-Yahr	0 [0,0]	2 [2,2]	0 [0,0]	1.86 × 10 −9	iRBD, Ctrl < PD
MoCA	28 [27,28]	26 [25,27]	28 [27,28]	0.096	n/a
Hyposmia (%)	20/25 (80%)	n/a	n/a	n/a	n/a

Comparisons of continuous measures with Kruskal-Wallis test and categorical data with χ²/Fisher’s exact test. Group-wise differences that remained significant after Bonferroni correction for multiple comparisons (α = 0.00625) are bolded and italicized.

iRBD, isolated REM sleep behavior disorder; PD, Parkinson’s disease; BMI, body mass index; RBD1Q, RBD single item question screen; MDS-UPDRS, Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale; MoCA, Montreal Cognitive Assessment.

Autonomic questionnaire results

iRBD patients reported a mean COMPASS-31 score of 20.97/100 ± 17.69, suggestive of mild autonomic failure, with greatest scores in the orthostatic (10.08/40 ± 10.27), gastrointestinal (5.86/25 ± 4.82), secretomotor (2.74/15 ± 3.99), and pupillomotor (0.88/5 ± 1.18) domains. Vasomotor (0.56/15 ± 1.17) and bladder domains (0.84/10 ± 1.37) were the least affected. The mean OHQ total score was 7.08 ± 6.89.

Skin biopsy results

Cutaneous p-syn was found in 16/25 (64%) of iRBD patients, 27/28 (96%) PD patients, and 0/18 controls (Figure 1). In most cases, only 1–2 of the 3 biopsy sites were positive. At any single biopsy site, we found a positivity rate of 40% across individuals with iRBD, with equal deposition across proximal and distal sites. We found no correlation between IENFD and cutaneous p-syn deposition in either PD or iRBD patients.

Figure 1.

Figures 1A–1D are images of cutaneous nerve fibers that contain phosphorylated α-synuclein. The first row of figures A1–D1 are stained with the pan-axonal marker protein gene product 9.5, with nerve fibers in green. The second row of figures A2–D2 reveal structures that stain for phosphorylated α-synuclein. The third row of figures A3–D3 are the merged images revealing phosphorylated α-synuclein within the nerve fibers. Images A1–A3 are from a patient with PD. Structures B1–B3 and C1–C3 are from patients with iRBD. Images D1–D3 are from a healthy control subject. Phosphorylated α-synuclein in seen within nerve fibers of the patient with PD and from the patients with IRBD, but not from the healthy control subject. Scale bar (yellow line in row A3–D3) is 50 μm. PD, Parkinson’s disease, iRBD, isolated REM sleep behavioral disorder.

Autonomic function testing results

Autonomic function testing was abnormal in 21/25 iRBD patients (84%). Sudomotor testing was abnormal in 8/25 (32%), with 5/8 (63%) exhibiting a non-length-dependent pattern. The E:I ratio, a measure of cardiovagal function, was abnormal in 11/25 (44%). The phase IV overshoot of the Valsalva maneuver, a measure of sympathetic adrenergic function, was reduced in 11/25 (44%). Sympathetic adrenergic failure with orthostatic hypotension (OH) was seen in 7/25 (28%) on head-up tilt table testing, with five patients exhibiting classic OH (blood pressure declines within 3 min of upright tilt) and two delayed OH (blood pressure declines after 3 min of upright tilt). Adrenergic and total CASS scores were significantly greater in iRBD patients when compared to controls, and similar to those with mild PD (Table 2).

Table 2.

Autonomic testing results

	iRBD (n = 25)	PD (n = 28)	Controls (n = 18)	P	Pairwise comparisons
Sympathetic cholinergic (sudomotor) function
QSART volume forearm	1.19 [0.93,1.81]	0.76 [0.36,1.06]	0.71 [0.42,1.01]	0.011	n/a
QSART volume proximal leg	0.64 [0.41,1]	0.91 [0.38,1.47]	0.97 [0.86,1.47]	0.041	n/a
QSART volume distal leg	0.55 [0.36,0.8]	0.57 [0.38,1.28]	1.13 [0.61,1.51]	0.079	n/a
QSART volume foot	0.41 [0.15,0.62]	0.49 [0.19,1.12]	1.11 [0.61,1.3]	0.015	n/a
Cardiovagal function
E:I ratio	1.15 [1.09,1.2]	1.13 [1.08,1.23]	1.28 [1.19,1.36]	3.72 × 10−3	n/a
Valsalva ratio	1.45 ± 0.26	1.37 ± 0.14	1.49 ± 0.16	0.20	n/a
Sympathetic adrenergic function
Valsalva baseline to P2 minimum (mmHg)	−31 [−49, −17.75]	−28 [−39.5, −21.75]	−20 [−24.5, −16]	0.013	n/a
Valsalva P2 recovery (mmHg)	8.42 ± 12.3	−21.32 ± 16.04	−2.75 ± 6.48	6.32 ×10−11	PD < Ctrl < iRBD
Valsalva P4 overshoot (mmHg)	11.21 ± 10.28	−2.36 ± 18.97	14.94 ± 8.02	1.10 ×10−3	PD < Ctrl
Tilt baseline SBP (mmHg)	143.32 ± 11.16	135.04 ± 16.4	127.5 ± 10.42	1.77 ×10−3	Ctrl < iRBD
Tilt baseline HR (bpm)	63 [60,68]	67.5 [61,75]	63 [61.7,70]	0.27	n/a
Tilt ΔSBP (mmHg)	−18 [−39, −5]	−20.5 [−31.25, −13]	−12.5 [−14.75, −8]	0.049	n/a
CASS scores
CASS sudomotor	0 [0–1.25]	0 [0–1.25]	0 [0,0]	0.014	n/a
CASS cardiovagal	1 [0–2]	1 [0–1.25]	0 [0–1]	0.064	n/a
CASS sympathetic	1.5 [0–3]	1 [1–3.25]	0 [0–1]	4.78 × 10−3	n/a
CASS total	3 [1–5]	2 [1–5]	1 [0–1.25]	8.68 ×10−4	Ctrl < PD, iRBD

Comparisons of continuous measures with Kruskal-Wallis test and categorical data with χ²/Fisher’s exact test. Group-wise differences that remained significant after Bonferroni correction for multiple comparisons (α = 0.0031) are bolded and italicized.

RBD, isolated REM sleep behavior disorder; PD, Parkinson’s disease; QSART, quantitative sudomotor axon reflex test; E:I, expiratory:inspiratory ratio; P2, phase 2 Valsalva; P4, phase 4 Valsalva; CASS, composite autonomic severity score.

Correlations between skin biopsy results, autonomic function, and olfaction in iRBD patients

The presence of cutaneous p-syn at any biopsy site in iRBD patients was moderately correlated with secretomotor symptoms on the COMPASS-31 (r = 0.46, p = 0.03) and cardiovagal impairment on autonomic testing (CASS cardiovagal r = 0.4, p = 0.02), but these findings were not deemed statistically significant after adjusting for multiple comparisons. The presence of p-syn was more robustly correlated with sympathetic impairment (CASS adrenergic r = 0.6, p = 0.002) and total autonomic impairment (CASS total r = 0.6, p = 0.002) on autonomic testing (Figure 2). We found no correlations with other subjective measures of autonomic function including other COMPASS-31 subscores or other measures of autonomic testing including sudomotor function (Table 3). These results were independent of p-syn density at each site.

Figure 2.

Box-and-whisker plots demonstrating CASS subscore and CASS total score distributions based on the presence (PSYN “pos”) vs. absence (PSYN “neg”) of p-syn at any of the 3 biopsy sites among individuals with iRBD. Box bounds represent the upper and lower quartiles, while the contained line represents the median and the whiskers represent the range of the highest and lowest observations.

Table 3.

Correlation analyses of variables and phosphorylated α-synuclein positivity in iRBD patients

Variables	ρ	P
Clinical features analysis (Bonferroni α = 0.007)
Duration of illness	−0.03	0.89
Antidepressant use	−0.11	0.59
MDS-UPDRS III	0.05	0.80
UMSARS Part 1	0.12	0.58
UMSARS Part 2	−0.34	0.11
MoCA	0.23	0.28
UPSIT	−0.23	0.28
Quality of life and symptom analysis (Bonferroni α = 0.003)
ESS	0.01	0.96
Shrag QoL motor	0.01	0.97
Shrag QoL nonmotor	0.17	−0.04
Shrag QoL mood	0.10	0.66
Shrag QoL total score	0.10	0.66
COMPASS-31 orthostatic	0.18	0.39
COMPASS-31 vasomotor	−0.05	0.81
COMPASS-31 secretomotor	0.46	0.02
COMPASS-31 gastrointestinal	0.02	0.93
COMPASS-31 bladder	−0.09	0.66
COMPASS-31 pupillomotor	0.12	0.58
COMPASS-31 total score	0.17	0.41
OHQ total score	0.11	0.61
Autonomic testing analysis (Bonferroni α = 0.004)
E:I ratio	−0.35	0.08
Valsalva BP baseline	−0.11	0.60
Valsalva P2 BP recovery	−0.01	0.98
Valsalva P4 overshoot	−0.31	0.14
Tilt ΔSBP	−0.10	0.64
QSART volume forearm	−0.17	0.44
QSART volume proximal leg	−0.04	0.87
QSART volume distal leg	−0.05	0.81
QSART volume foot	0.09	0.68
CASS sudomotor	0.21	0.32
CASS cardiovagal	0.40	0.06
CASS adrenergic	0.60	0.002
CASS total	0.60	0.002

Correlations that remain significant after Bonferroni correction are bolded and italicized.

iRBD, isolated REM sleep behavior disorder; MDS-UPDRS, Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale; UMSARS, Unified Multiple System Atrophy Rating Scale; MoCA, Montreal Cognitive Assessment; CASS, Composite Autonomic Severity Scale; UPSIT, University of Pittsburg Smell Identification Test; COMPASS, Composite Autonomic Symptom Scale-31; OHQ, Orthostatic Hypotension Questionnaire; SCOPA-AUT, Scale for Outcomes in PD-Autonomic.

Hyposmia was more common in iRBD patients who were p-syn positive, present in 15/16 (94%) of these patients, compared to 4/9 (44%) of the iRBD patients who were p-syn negative. All 15 of the hyposmia, p-syn positive patients had evidence of autonomic dysfunction on autonomic reflex testing. In comparison, all four of the iRBD patients with normal autonomic testing had normal UPSIT scores. However, the correlation between p-syn positivity and hyposmia was not deemed statistically significant after adjusting for multiple comparisons (rho = −.0.23, p = 0.28; Table 3).

Associations between skin biopsy results and antidepressant use

The presence of cutaneous p-syn was independent of both RBD duration and antidepressant use in iRBD patients (Table 3 and Figure 3). In those iRBD patients taking antidepressants that had a positive biopsy, we found no association with either duration or timing of antidepressant use, for example, whether patients had started the antidepressant before or after the onset of their dream enacting behaviors, or class of antidepressant (SSRI, SNRI, or NDRI).

Figure 3.

Stacked bar chart demonstrating proportion (summing to 1) of iRBD patients with (PSYN “pos”) and without (PSYN “neg”) p-syn at any of the 3 biopsy sites separated based on antidepressant usage.

Discussion

Cutaneous p-syn was detected in 64% of iRBD patients, similar to positivity rates reported in other iRBD cohorts, which range from 56% to 87% [4–7]. Our analysis resulted in a sensitivity of 64% in iRBD and 96% in PD, with a specificity of 100%. While others have reported a trend toward a proximal-distal gradient of p-syn deposition in iRBD, with higher positivity rates at cervical or thoracic biopsy sites [4, 5], we found an equal deposition of p-syn across proximal and distal sites in our iRBD cohort, and in some cases only distal samples were positive. While these findings may reflect differences in methodology (e.g. the inclusion of both cervical and thoracic biopsy sites by one group [4], or two biopsies instead of one at the cervical site by another [5]), they agree with cutaneous p-syn distribution in other α-synucleinopathies of Lewy body pathology including PD [26], DLB [27], and pure autonomic failure (PAF) [28]. They also emphasize the need to biopsy multiple sites to reach acceptable levels of sensitivity.

Autonomic impairment was common in our iRBD patients, with deficits across sudomotor, cardiovagal, and adrenergic measures of autonomic function testing. These abnormalities are congruent with the pathological findings seen on skin biopsy, as p-syn tends to aggregate within the autonomic nerves innervating sweat glands, vasomotor, and pilomotor structures [29]. iRBD patients with cutaneous p-syn were more likely to exhibit greater deficits on autonomic testing than those without p-syn, with the most robust correlation seen in measures of sympathetic adrenergic function. This is in line with prior studies in iRBD demonstrating sympathetic impairment in markers of heart rate variability spectral analysis, cardiac scintigraphy, and autonomic reflex testing [8].

It has been suggested that the presence of RBD in those with PD is associated with a progressive and malignant subtype of the disease, with more severe motor and non-motor symptoms, including those of autonomic impairment [30]. In support of this theory, one study correlating cutaneous p-syn with autonomic failure in those with PD demonstrated that PD + OH patients had higher p-syn positivity rates than PD − OH patients (90% vs. 38%) [31]. In addition, PD + OH patients were more likely to exhibit a patchy distribution of p-syn across proximal and distal sites, whereas PD − OH patients were more likely to exhibit a proximal-distal gradient. Our findings suggest that iRBD patients with autonomic impairment are also more likely to exhibit patchy cutaneous p-syn positivity, which may be related to a more diffuse spread of p-syn.

The cutaneous autonomic structures involved in p-syn deposition are not involved in cardiovagal or baroreflex function, therefore the abnormalities seen on autonomic testing are suggestive of a more widespread process that likely affects both central and peripheral autonomic structures. Given the anatomical proximity of pontine coeruleus/subcoeruleus REM control nuclei to central autonomic structures in the pons and medulla, p-syn spread along common anatomical pathways likely plays a role. Indeed, PD patients with RBD exhibit more widespread synuclein-driven pathology than PD patients without RBD [32], and are more likely to have OH as well as develop dementia, 36 further supporting the malignant subtype theory.

Despite high rates of objective autonomic impairment on autonomic testing, we found a limited correlation between cutaneous p-syn positivity and symptoms of autonomic impairment as measured by COMPASS-31, SCOPA-AUT, and OHQ scores. Most patients reported mild symptoms on these rating scales. While this may be due to limited sample size, this finding is not surprising, as some PD patients with autonomic impairment exhibit impaired autonomic symptom recognition [33], and it is conceivable that this impairment may extend to iRBD, as others have noted [34]. It also raises the possibility that objective autonomic testing may prove a more reliable biomarker than autonomic questionnaires in the α-synucleinopathies including iRBD.

We found no correlation between cutaneous p-syn positivity and antidepressant use in iRBD patients. Furthermore, we found no association with duration or timing of antidepressant use, antidepressant class, or onset of dream enacting behaviors. While depression, anxiety, and antidepressant use are common in those with iRBD [35], it is unclear if antidepressant medications are a cause of RSWA and RBD, if they lead to an unmasking of subclinical dream enacting behavior in RBD patients with underlying neurodegenerative disease, or if the prodromal neurodegenerative disease itself leads to depression, anxiety, and subsequent antidepressant use [36]. Several groups have reported an association between antidepressants and RBD, with some reporting a greater association with SSRIs [37], and others reporting an equal association across different classes of antidepressants [38, 39]. While caution should be taken with the relatively small sample size of our cohort, our data failed to find a correlation between antidepressant use and cutaneous p-syn positivity.

While iRBD patients with hyposmia were more likely to be p-syn positive compared to those with normal olfaction, we did not find a statistically significant correlation between hyposmia and p-syn positivity. It should be noted that others have documented a significant association between these two variables (rho = −.0.64, p = 0.003) using Sniffin’ sticks [4], a slightly different olfactory test than the UPSIT utilized in our cohort. The limited sample size of our iRBD cohort may have also accounted for this discrepancy.

Finally, we found no correlation between cutaneous p-syn deposition and IENFD. Similar findings have been reported by others in patients with PD and continue to support the theory that the length-dependent reduction in IENFD and α-synuclein-associated autonomic neuropathy may have different pathogenetic mechanisms.

The limitations of our study include the relatively small sample size, the single center iRBD cohort drawn from a limited geographical area, and the lack of longitudinal follow-up. In addition, due to the use of a previously recruited cohort of PD and controls, questionnaires were not harmonized across institutions and thus symptoms of autonomic impairment could not be compared across groups. Furthermore, while PD and control groups completed the RBD1Q screening questionnaire, vPSG was not performed, therefore it is possible that RBD or isolated RSWA may have gone undetected.

In summary, our study confirms a high positivity rate of cutaneous p-syn and demonstrates even higher rates of autonomic dysfunction in iRBD patients. Those patients with markers of sympathetic adrenergic impairment on autonomic testing were more likely to be p-syn positive, raising the possibility of a shared pathophysiological mechanism. Future efforts will focus on longitudinal follow-up to assess the influence of these results on phenoconversion rates and subtype of the disease. These data will be important in collaboration with other centers to better inform the use of skin biopsy and autonomic testing as potential biomarkers in disease-modifying trials.

Funding

This work was supported by NIH U54-NS065736 (RF) and the Department of Veterans Affairs Office of Academic Affiliations Advanced Fellowship Program in Mental Illness Research and Treatment (LS).

Authors’ Contributions

Mitchell Gordon Miglis: Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data.

Jennifer Zitser: Drafting/revision of the manuscript for content, including medical writing for content.

Logan Douglas Schneider: Drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data.

Emmanuel During: Drafting/revision of the manuscript for content, including medical writing for content.

Safwan Jaradeh: Drafting/revision of the manuscript for content, including medical writing for content.

Roy Freeman: Drafting/revision of the manuscript for content, including medical writing for content; Study concept or design.

Christopher H Gibbons: Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data.

Statistical Analysis performed by: Logan Schneider, MD Stanford/VA Alzheimer's Center, Palo Alto VA Health Care System, Palo Alto, CA.

All authors have seen and approved this manuscript.

Disclosure Statements

Financial Disclosures: M.G. Miglis reports royalties from Elsevier; J. Zitser reports no disclosures; L. Schneider reports grant funding from the Veterans Affairs Palo Alto Health Care System during the conduct of the study, personal fees from Alphabet, Inc., personal fees from Jazz Pharmaceuticals, and personal fees from Harmony Biosciences, outside the submitted work; E. During reports no disclosures relevant to the manuscript; S. Jaradeh reports no disclosures relevant to the manuscript; R. Freeman has received personal compensation and/or stock options for serving on scientific advisory boards of Abide, Akcea, Averitas, Applied Therapeutics, Clexio, Ceracor, Cutaneous NeuroDiagnostics, Lundbeck, Novartis, NeuroBo, Regenacy, Toray, Theravance, and Vertex, and has received personal compensation for his editorial activities (Editor) with Autonomic Neuroscience Basic and Clinical; C. Gibbons has served as a scientific advisor for CND Life Sciences and Lundbeck. Dr. Gibbons has stock options with CND Life Sciences. C. Gibbons has received personal compensation for his editorial activities (Associate Editor) with Autonomic Neuroscience Basic and Clinical.

Nonfinancial Disclosures: All authors report no conflicts of interest.