Data from: Plasma extracellular vesicle synaptic proteins as biomarkers of clinical progression in patients with Parkinson’s disease

Hong, Chien-Tai 1 ; Chung, Chen-Chih1 ; Yu, Ruan-Ching2; Chan, Lung1

Published Sep 28, 2026 on Dryad. https://doi.org/10.5061/dryad.4qrfj6qh9

Data files

Sep 28, 2026 version files 70.59 KB

For_linear_mixed_model_20220630_for_Grace.sav

55.50 KB
README.md

15.08 KB

Abstract

Synaptic dysfunction plays a key role in Parkinson’s disease (PD), and plasma extracellular vesicle (EV) synaptic proteins are emerging as biomarkers for neurodegenerative diseases. Assessment of plasma EV synaptic proteins for their efficacy as biomarkers in Parkinson's Disease (PD) and their relationship with disease progression was conducted. In total, 144 participants were enrolled, including 101 people with PD (PwP) and 43 healthy controls (HCs). The changes in plasma EV synaptic protein levels between baseline and 1-year follow-up did not differ significantly in both PwP and HCs. In PwP, the changes in plasma EV synaptic protein levels were significantly associated with the changes in unified PD rating scale (UPDRS) part II and III scores. Moreover, PwP with elevated levels (first quartile) of any one plasma EV synaptic proteins (synaptosome-associated protein 25, growth-associated protein 43 or synaptotagmin-1) had significantly greater disease progression in UPDRS part II score and the postural instability and gait disturbance subscore in UPDRS part III than did the other PwP after adjustment for age, sex, and disease duration. The promising potential of plasma EV synaptic proteins as clinical biomarkers of disease progression in PD was suggested. However, a longer follow-up period is warranted to confirm their role as prognostic biomarkers.

Updated: 2026-09-24

Study and associated publication

This dataset contains longitudinal demographic, clinical, and plasma extracellular vesicle protein measurements from participants with Parkinson's disease and controls.

Associated publication: Hong CT, Chung CC, Yu RC, Chan L (2024). Plasma extracellular vesicle synaptic proteins as biomarkers of clinical progression in patients with Parkinson's disease. eLife.

The data file contains 171 distinct participant numbers. The published study analysed the 144 participants with complete data: 101 participants with Parkinson's disease and 43 controls. The file therefore contains a broader set of records than the published analysis cohort. It does not include a separate indicator identifying membership in the published analysis cohort.

Data file

Filename: For_linear_mixed_model_20220630_for_Grace.sav

Property	Value
Format	IBM SPSS Statistics system file (`.sav`)
Size	55,505 bytes
Records	395
Variables	17, all stored numerically
Distinct participant numbers	171
Public file writer	pyreadstat 1.3.6
Public revision date	2026-09-24
Character encoding	UTF-8

This public revision replaces source participant identifiers with number and exact age with the categorical variable Age_group. Age-category labels and missing-value definitions are embedded in the SPSS file. The variables are defined below. The original source file was created using IBM SPSS Statistics 26.

Record structure and visits

Each row contains a participant's record for a visit. number links records belonging to the same participant. It is a replacement number from 1 to 171, assigned once per participant and retained across visits. Original source identifiers and the identifier-to-number lookup are not included in the public file. Follow identifies the first, second, or third visit, scheduled at one-year intervals:

Follow = 1: first visit (baseline; elapsed time 0 years).
Follow = 2: second visit (one year after the first visit).
Follow = 3: third visit (two years after the first visit).

These codes identify scheduled visits; the file does not contain assessment dates or exact elapsed intervals. The published article reports baseline and one-year follow-up, while this file also includes third-visit records.

Group	First visit (`1`)	Second visit (`2`)	Third visit (`3`)	Total records	Participants
Control (`PD = 0`)	67	67	11	145	69
Parkinson's disease (`PD = 1`)	102	103	45	250	102
Total	169	170	56	395	171

Visit counts are counts of records. There are 3 participants with one record, 112 with two records, and 56 with three records. One number/Follow combination appears twice in non-identical rows, so these columns are not a unique key in this file. No completely identical duplicate rows occur.

Variable dictionary

Variable names below match the SPSS file. Observed ranges exclude declared SPSS missing values and are distinct from the possible ranges of the assessment instruments.

Variable	Definition and coding	Units / observed nonmissing range
`number`	Replacement participant number linking repeated observations of the same participant. Treat as a categorical identifier. Original source identifiers are not included.	Consecutive integers 1–171; one number per participant, consistent across visits.
`PD`	Study group: `0` = control; `1` = Parkinson's disease.	Categorical.
`Age_group`	Age category: `1` = <50 years; `2` = 50–59 years; `3` = 60–69 years; `4` = 70–79 years; `5` = ≥80 years; `999` = missing.	Ordered categories; exact ages are not included.
`Sex`	`0` = male; `1` = female.	Categorical. One stored value of `5` falls outside these categories.
`dx_duration`	Duration of Parkinson's disease.	Years; observed range 0–14.
`Follow`	Visit number: `1` = first visit, `2` = second visit, `3` = third visit; visits are one year apart.	Scheduled elapsed time from first visit: 0, 1, and 2 years, respectively.
`UPDRSII`	Unified Parkinson's Disease Rating Scale (UPDRS) Part II total score, assessing activities of daily living. Higher scores indicate greater impairment.	Points; observed range 0–31.
`UPDRSIII`	UPDRS Part III total score, assessing motor impairment. Higher scores indicate greater impairment.	Points; observed range 2–51.
`Tremor_score`	Mean tremor score from UPDRS Part III items 20 and 21, using their seven component ratings.	Mean item score; observed range 0–1.714285714.
`AR_score`	Mean akinetic-rigidity score from UPDRS Part III items 18, 19, 22–26, and 31, using their 16 component ratings.	Mean item score; observed range 0.0625–2.5.
`PIGD_score`	Mean postural instability and gait disturbance score from UPDRS Part III items 27–30.	Mean item score; observed range 0–3.25.
`MMSE`	Mini-Mental State Examination total score. Higher scores indicate better cognitive performance.	Points; possible range 0–30; observed range 0–30.
`MOCA`	Montreal Cognitive Assessment total score, including the education adjustment. Higher scores indicate better cognitive performance.	Points; maximum 30; observed range 1–30.
`MOCA_1`	MoCA first-section score: visuospatial/executive function, comprising alternating trail making, cube copying, and clock drawing. Higher scores indicate better performance.	Points; observed range 0–5.
`SNAP25`	Synaptosome-associated protein 25 signal divided by the HSP70 signal from the same well.	Dimensionless ratio; observed range 0.32–35329.83.
`GAP43`	Growth-associated protein 43 signal divided by the HSP70 signal from the same well.	Dimensionless ratio; observed range 0.79–25094.32.
`Syn1`	Synaptotagmin-1 signal divided by the HSP70 signal from the same well.	Dimensionless ratio; observed range 0.55–19990.68.

Age categories and public-file revision

Exact age has been replaced by Age_group to reduce identification risk in the public data. Codes are ordered categories, not ages in years. The interval boundaries are <50, 50 to <60, 60 to <70, 70 to <80, and ≥80 years.

Code	Age category	Records
1	<50 years	5
2	50–59 years	52
3	60–69 years	199
4	70–79 years	113
5	≥80 years	26
999	Missing age (no occurrences in this revision)	0

The author supplied the correct ages for one erroneous source-age entry and two source-age entries previously coded 999. These corrections were applied before age categorization. All three affected records are in the 60–69-year category. There are no missing age categories in this revision; 999 remains defined as the missing-age code. Source participant identifiers have also been replaced with number; all 395 records, the other 15 variables, and the row order are retained unchanged from the source file.

Age grouping changes the information available for statistical adjustment. Analyses that used exact age in the original study cannot necessarily be reproduced exactly with this public age variable.

UPDRS Part III subscores

The subscores are arithmetic means of individual component ratings, not sums. Item numbers refer to the original UPDRS Part III. Tremor uses item 20 (five body-region ratings) and item 21 (two hand ratings), totalling seven component ratings. PIGD uses items 27 (arising from a chair), 28 (posture), 29 (gait), and 30 (postural stability), totalling four ratings.

AR uses items 18 (speech), 19 (facial expression), 22 (rigidity; five body-region ratings), 23 (finger taps; two sides), 24 (hand movements; two sides), 25 (rapid alternating hand movements; two sides), 26 (leg agility; two sides), and 31 (body bradykinesia/hypokinesia). The sum of these 16 component ratings is divided by 16. The corresponding denominators for tremor and PIGD are 7 and 4, respectively.

For all three subscores, higher scores indicate greater motor impairment. The file contains the derived scores, without the underlying item-level ratings. Some stored scores are rounded to two decimal places; others retain more precision. Tremor and PIGD item selections are described in the authors' 2023 follow-up study.

MoCA scoring

The standard 30-point MoCA comprises seven scored sections:

Section	Maximum points
Visuospatial/executive function	5
Naming	3
Attention	6
Language	3
Abstraction	2
Delayed recall	5
Orientation	6

This seven-section structure is described in Vogel et al. (2015). MOCA stores the education-adjusted total. MOCA_1 stores only the first section, visuospatial/executive function: alternating trail making (1 point), cube copying (1 point), and clock drawing (3 points), for a possible total of 0–5 points. It is a cognitive section score, not a visit code. The other six section scores are not separate variables in this file.

Protein measurements

SNAP25, GAP43, and Syn1 contain ratios of the target protein signal to HSP70 from the same well. These values are relative measurements without concentration units. The associated article describes the plasma extracellular vesicle preparation and protein assay methods.

Missing values

The SPSS file declares 999 as a user-defined missing value for the following 13 variables. These values represent missing observations and must be excluded from numeric summaries and models.

Variable	Records coded `999`	Other records
`Age_group`	0	395
`dx_duration`	154	241
`UPDRSII`	145	250
`UPDRSIII`	145	250
`Tremor_score`	145	250
`AR_score`	145	250
`PIGD_score`	145	250
`MMSE`	23	372
`MOCA`	24	371
`MOCA_1`	23	372
`SNAP25`	72	323
`GAP43`	67	328
`Syn1`	67	328

No system-missing numeric cells are stored. number, PD, Sex, and Follow have no declared user-defined missing values. Zero is not a missing-value code.

Recorded-value exceptions

The following features of the source data remain in this public revision:

Sex contains one value of 5, outside the documented male/female codes. Four participant numbers have differing sex codes across their records.
One participant/visit combination appears in two non-identical rows, as described above.

These features affect sex-based grouping and assumptions of one observation per participant per visit. The file contains no correction flags or adjudicated replacements for these entries.

Opening the file

Open the .sav file in IBM SPSS Statistics or another application supporting SPSS system files and user-defined missing values. For example, using Python and pyreadstat:

import pyreadstat

data, metadata = pyreadstat.read_sav(
    "For_linear_mixed_model_20220630_for_Grace.sav",
    user_missing=False,
)

This converts declared missing values, including Age_group = 999, to the reader's missing-value representation. To inspect stored category codes and missing-value declarations, use user_missing=True and inspect metadata.variable_value_labels and metadata.missing_ranges. SPSS display formats can show fewer decimal places than the values stored in the file.

File version

Public revision dated 2026-09-24: source participant identifiers replaced by number (1–171); three source-age entries corrected using author-supplied ages before exact age was replaced by five categories. All 395 records have an age category. All records, their order, and the other 15 variables are retained unchanged.

SHA-256 checksum of this public .sav file:

86f6e747625eb37ae59f447d0fce766fa7c705cfaa0b982edca3c0a34d645014

Human subjects data

The research protocol was approved by the Joint Institutional Review Board of Taipei Medical University (TMU-JIRB approval no. N201609017 and N201801043). Written informed consent was obtained from all participants for participation in the study.