GRAPHICAL ABSTRACT

INTRODUCTION

Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family and is essential for neuronal survival [1,2]. Its role has been implicated in regulating the endurance and dopamine release of dopaminergic neurons in the substantia nigra, which is tied to the pathogenesis of Parkinson’s disease (PD) [3-6].

The BDNF rs6265 polymorphism (Val66Met polymorphism) is associated with PD [1,2]. Its allele frequency varies by ethnicity and geographic region [1]. Approximately 30%–50% of Caucasian individuals are Met carriers, whereas approximately 60%–70% of Asian individuals are Met carriers [1]. These ethnic disparities may contribute to the heterogeneity of PD phenotypes.

The influence of the BDNF rs6265 polymorphism has been investigated primarily in Caucasian PD cohorts but rarely in Asian populations, and the existing results are inconsistent between Western and Asian studies [6-11]. In a Japanese population, Val homozygotes were more prevalent among PD patients than among controls [6], whereas the opposite finding was reported in a Swedish population, and there was no difference in genotype frequency between PD patients and controls among Whites [7]. In addition to its association with PD risk, its influence on cognition is also controversial. In an Italian PD cohort, Met homozygotes were associated with worse cognition, whereas other studies reported a higher incidence of cognitive impairment or better verbal fluency among Met carriers; in a Greek population, no association between the polymorphism and cognition was observed [8-11]. These studies differed not only in genotype distributions but were also limited by a narrow focus on a single domain—primarily cognition—and by their predominantly cross-sectional design.

The present study aimed to explore how the BDNF rs6265 genotype could shape motor and nonmotor characteristics and longitudinal progression patterns in PD patients in a Korean population.

MATERIALS & METHODS

Patients

The study was approved by the Institutional Review Board of Seoul St. Mary’s Hospital, The Catholic University of Korea (Approval number: KC21OIDI0362). All the subjects provided written informed consent to participate. Research was conducted in accordance with the relevant guidelines and regulations.

This study utilized data from the Korean nationwide hospital-based PD (K-PD) cohort, an observational, prospective, longitudinal cohort study of the Korea National Institute of Health [12]. A total of 247 de novo PD patients diagnosed between July 2015 and May 2023 at a single participating hospital in the K-PD cohort were included. Diagnosis was established on the basis of the Movement Disorder Society (MDS)-PD diagnostic criteria, which was supported by positron emission tomography (PET) imaging studies using ¹⁸F-N-(3-fluoropropyl)-2β-carbomethoxy-3β-(4-iodophenyl)nortropane (¹⁸F-FP-CIT) [13]. Patients exhibited decreased ¹⁸F-FP-CIT uptake in the striatum, primarily in the posterior putamen.

Baseline characteristics, including age at diagnosis, sex, body mass index, disease duration at diagnosis, follow-up duration, and history of hypertension, diabetes mellitus, dyslipidemia, and smoking status, were investigated.

To avoid PD mimics and to minimize the inclusion of patients with cardiovascular dysautonomia or related complications, individuals meeting any of the following criteria were excluded: 1) any symptoms or signs of atypical and/or secondary parkinsonism during follow-up visits; 2) a history of diabetic neuropathy at initial evaluation; 3) a history of symptomatic stroke that could affect general cognition and performance; and 4) a history of heart failure.

Patients were monitored every 3–6 months for a mean follow-up duration of 50.9±23.9 months. PD diagnosis was independently confirmed by two neurologists (S.-W.Y. and J.-S.K.).

DNA analysis: BDNF and APOE genes

DNA was extracted using a NanoDrop^® ND-2000 UV–Vis Spectrophotometer. Samples were genotyped for BDNF rs6265 and apolipoprotein E (APOE) using TaqMan SNP Genotyping Assays obtained from Applied Biosystems. Genomic DNA was diluted to a concentration of 5 ng/μL on PCR plates. PCR was performed in 5 μL of a mixture containing 2 μL of a DNA sample, 0.125 μL of each TaqMan^TM SNP Genotyping Assay (Thermo Fisher Scientific), 2.5 μL of TaqMan^TM Genotyping Master Mix (Thermo Fisher Scientific), and 0.375 μL of distilled water. After the PCR amplification, allelic discrimination was performed on the same machines (QuantStudio 12K Flex Real-Time PCR System). The allelic discrimination was an endpoint plate read.

Imaging acquisition and processing of ¹⁸F-FP-CIT PET

Brain computed tomography (CT) and ¹⁸F-FP-CIT PET images were obtained using a Discovery STE PET/CT scanner (Discovery PET/CT 710, General Electric Healthcare). After 3 hours of intravenous injection of 3.7 MBq/kg ¹⁸F-FP-CIT, brain CT scans were obtained for attenuation correction, followed by a 10-min PET scan.

Image processing was conducted using Statistical Parametric Mapping 8 software (SPM8; Wellcome Trust Centre for Neuroimaging) and an in-house automated pipeline program implemented in MATLAB 2015a (MathWorks). A detailed description of the imaging process is provided in Supplementary Material 1. Every patient was evaluated with ¹⁸F-FP-CIT PET images (Figure 1).

UPDRS and MDS-UPDRS

Disease severity was evaluated using the original Unified Parkinson’s Disease Rating Scale (UPDRS) and the MDS-UPDRS. Baseline motor severity was assessed in the OFF medication state, whereas subsequent examinations were conducted in the ON state, when patients exhibited a full response to dopaminergic therapy. A total of 137 (56.8%) patients were initially evaluated using the UPDRS, and 104 (43.2%) patients were assessed using the MDS-UPDRS. Patients were subsequently reevaluated using the same tool, as appropriate. Motor scores from the original UPDRS (Parts II and III) were converted to MDS-UPDRS scores using a previously published conversion method [14]. Total motor scores were calculated by summing the scaled Part II and III scores. Patients were reassessed 2–3 times during follow-up after the baseline investigation (Figure 1).

Head-up tilt test

Every patient was tested in the full resting state. Continuous electrocardiographic and noninvasive blood pressure (BP) monitoring equipment (YM6000, Mediana Tech) was applied to the patients. A supine position was maintained for 20 minutes during the recording of BP and heart rate every 5 minutes before tilting to 60 degrees. At the tilted position, measurements were taken at 0, 3, 5, 10, 15, and 20 minutes. The definitions of supine hypertension, orthostatic hypotension (OH), neurogenic OH (nOH), and supine/orthostatic mean arterial pressure (MAP) are specified in Supplementary Material 2.

Two hundred and forty-one patients were assessed at the time of diagnosis (Figure 1).

¹²³I-MIBG myocardial scintigraphy

¹²³I-meta-iodobenzylguanidine (¹²³I-MIBG) scintigraphy was conducted using a dual-head camera equipped with a low-energy, high-resolution collimator, and data were collected at 30 min (early) and 120 min (late) time points after the injection of 111 MBq of ¹²³I-MIBG. A static image was obtained with a 128×128 matrix. Regions of interest were manually drawn around the heart and mediastinum. Tracer uptake was measured within each region of interest to calculate the heart-to-mediastinum (H/M) ratio for the early and late phases. ¹²³IMIBG scintigraphy was reassessed 2–3 times (Figure 1).

Neuropsychological evaluation

The results of the neuropsychological evaluations were evaluated by experienced psychologists who were blinded to the clinical data. Five cognitive domains were assessed by a comprehensive neuropsychological battery, the Seoul Neuropsychological Screening Battery 2nd edition (SNSB-II), and subtests of each domain were selected [15-18]. The constructs of each domain (attention/working memory domain, frontal/executive domain, memory domain, language domain, and visuospatial domain) are detailed in Supplementary Material 3.

The frontal cognitive profile was defined as the average z scores of attention/working memory and frontal/executive domains, and the average of delayed recall, language, and visuospatial domains comprised the nonfrontal summary [17]. Global efficiency was calculated as the mean frontal and nonfrontal scores. Cognition was reassessed 2–3 times (Figure 1).

Clinical instruments

The patients completed the following evaluations: 1) Non-Motor Symptoms Scale (NMSS), 2) Montgomery-Asberg Depression Rating Scale (MADRS), 3) Epworth Sleepiness Scale (ESS), 4) Parkinson’s Disease Sleep Scale-2 (PDSS-2), 5) REM Sleep Behavior Disorder Screening Questionnaire (RBDSQ), and 6) 39-Item Parkinson’s Disease Questionnaire (PDQ39). Brief descriptions of each instrument are provided in Supplementary Material 4.

NMSS and PDQ39 summary index (PDQ39SI) were separately investigated in the longitudinal analyses because each signified different aspects of the disease. The patients were reinterviewed 2–3 times with the same questionnaires (Figure 1). All the questionnaires were evaluated in a blinded manner to the clinical information of the patients.

Composite scores

The NMSS, MADRS, ESS, PDSS-2, and RBDSQ were selected to represent the composite of the nonmotor quality of PD. Each tool was standardized by median absolute deviation to adjust for the different scales. The average standardized score was estimated to indicate the overall nonmotor burden. With the available resources in this study, the composite score was preplanned to incorporate the affection and sleep domains with the NMSS because its construct of those domains lacked specificity. The composite scores were calculated for each time point.

Statistical analyses

Statistical analyses were conducted with jamovi software (version 2.6.44; The jamovi project, https://www.jamovi.org/) and R (version 4.4.3; R Foundation for Statistical Computing) for Mac. Independent t-tests or Mann–Whitney U tests were performed for continuous variables when appropriate, and Fisher’s exact tests were used for categorical variables. Analysis of covariance, partialized by age, sex, and disease duration at onset, was performed for between-group marginal mean comparisons.

Random slope and intercept mixed modeling, controlled by disease duration at onset, was applied to the repeated measures of the longitudinal data with a constrained residual covariance structure (nlme package; version 3.1). Statistical significance was defined as a two-tailed p-value <0.05. The exact method of model estimation is further described in Supplementary Material 5.

Data sharing

Anonymized data generated during the current study are available from the corresponding author upon reasonable request from individuals affiliated with research or health care institutions.

RESULTS

Baseline characteristics across BDNF genotypes

Patients with early PD (n=247) were enrolled and followed (Table 1 and Figure 1). The mean age at diagnosis was 67.0±9.3 years, and 131 patients (53.0%) were male. The median disease duration at diagnosis was 1.0 year (interquartile range [IQR], 0.5–1.5 years), and the patients were followed for an average of 50.9± 23.9 months. Forty-seven patients (19.0%) were APOE ε4 allele carriers. These baseline characteristics did not differ between the BDNF genotypes. The prevalence of the homozygous valine (Val/Val) and methionine (Met/Met) genotypes were 31.2% and 19.8%, respectively. The percentage of heterozygotes (Val/Met) was 49.0%. Overall, approximately 69% of the study population were Met carriers. The proportions of BDNF genotypes did not differ significantly across the evaluation tools, and no dose-dependent effects of alleles were observed across clinical characteristics (Supplementary Tables 1 and 2).

Motor scores indicated mild parkinsonism at enrollment (n=241/247, 97.6%). Converted MDS-UPDSR Parts II and III, and their totals were 5.7 (IQR, 3.0–10.0), 19.7±11.3, and 26.5 ±15.3, respectively. No between-group differences were observed (Table 1).

Two hundred forty-one (241/247, 97.6%) and two hundred twenty-nine (229/247, 92.7%) patients underwent head-up tilt tests and ¹²³I-MIBG myocardial scintigraphy, respectively (Table 1). Sixty-six patients (66/241, 27.4%) had an orthostatic MAP below 75 mm Hg (MAP75). Supine hypertension was present in 12.0% (29/241) of patients, and 23.7% (57/241) exhibited hypotension during upright tilt. Thirty-five patients (35/241, 14.5%) had nOH with a standing MAP75. These frequencies did not differ between the BDNF genotypes. The Met carrier group demonstrated significantly lower H/M ratios (both early and late).

Two hundred twenty-four (224/247, 90.7%) and one hundred ninety-seven (197/247, 79.8%) patients completed the initial comprehensive neuropsychological examinations and questionnaires, respectively (Table 1). Their global and subdomains of cognition were relatively preserved. Genotypic comparisons revealed no significant differences, except in the nonfrontal cognitive profile, in which the Val/Val group demonstrated poorer performance. The Digit Span Forward test was more preserved in Val homozygotes; however, this did not translate to a significant difference in the overall frontal profile.

The nonmotor burden was moderate (NMSS, 26.9±23.2). The nonmotor symptoms, composite scores, and quality of life assessments did not differ across the BNDF subtypes (Table 1). Comparisons of subregional presynaptic dopamine transporter densities across the BDNF genotypes also revealed no significant differences (Table 2).

Longitudinal analyses

The initial evaluations were longitudinally reassessed (Figure 2 and Supplementary Table 3). Mixed-effects models were applied to discern genotypic differences in trajectory patterns. With respect to the average disease duration at diagnosis (2.8 years), Val homozygotes and Met carriers did not differ in motor severity; however, the Met carrier group had a significant linear progression of motor severity (effect of disease duration: Model 1A vs. Model 1B vs. Model 1C; 95% confidence interval [CI] 0.61–1.18 vs. 0.59–1.78 vs. 0.59–1.78) (Supplementary Table 3). The Val/Val genotype significantly contributed a positive quadratic effect to average motor progression, and as the disease advanced, motor worsening exceeded the trajectory of the Met carrier after approximately 3 years of follow-up (interaction effect on disease duration [2]: Model 1A vs. Model 1B vs. Model 1C; 95% CI 0.05–0.40 vs. 0.11–0.83 vs. 0.11–1.83) (Supplementary Table 3). In summary, BDNF interacted with motor severity, resulting in faster motor progression in Val homozygotes than in Met carriers after 3 years of follow-up.

On average, the Met carriers showed a linear association with worsening NMSS, PDQ39SI, and nonmotor composite scores across disease spans (effect of disease duration: Model 2A vs. Model 2B vs. Model 2C; 95% CI 1.52–5.07 vs. 0.88–2.64 vs. 0.08–0.21) (Supplementary Table 3). The Met carrier genotype exerted a greater effect on the progression of nonmotor burden.

The mixed model failed to explain any significant effect of BDNF genotype on the progression of cognition (Model 3) (Supplementary Table 3). The overall trajectories of each genotype paralleled global cognition, with an accelerated decline after 3 years of follow-up. When cognition was stratified by frontal vs. nonfrontal profiles, the Val/Val genotype had a predominant effect on the augmented decline in frontal cognition.

Compared with the Met carrier, the Val/Val genotype resulted in a significantly higher late H/M ratio (BDNF estimate: Model 4; 95% CI 0.02–0.20) (Supplementary Table 3). The H/M ratios significantly decreased in an inverse linear fashion over time, with comparable slopes between the groups (Model 4) (Supplementary Table 3). Overall, BDNF genotypes did not affect the progression of myocardial denervation in a genotype-dependent manner, although Val homozygotes demonstrated more preserved myocardium at diagnosis.

DISCUSSION

Baseline characteristics were similar across BDNF genotypes, except that the Met carrier group exhibited more denervated myocardium but relatively more preserved nonfrontal cognition at diagnosis. Longitudinal analyses demonstrated genotype-specific increases in motor and cognitive severity. Compared with the Met carriers, the homozygous Val group showed accelerated motor progression and faster cognitive decline in the frontal domain after 3 years of follow-up. These findings suggest that the BDNF rs6265 polymorphism may influence distinctive clinical progression patterns in the Korean population.

The prevalence of the Met variant is higher in Asian populations than in Caucasian populations [1,2]. An ethnically homogeneous Korean population has been reported to harbor approximately twice the Met allele frequency of its Caucasian counterpart [19]. Our data support this ethnic difference, underscoring the need for population-specific research to determine how BDNF genotypes influence PD in Korea.

BDNF is involved in the survival and maintenance of several types of neurons (including dopaminergic neurons), the formation of synaptic connections and the regulation of synaptic plasticity. Its synthesis and maturation occur both in the intra- and extracellular environment; the immature pro-BDNF isoform forms into mature BDNF (m-BDNF). In adulthood, m-BDNF plays a dominant role by binding to the tropomyosin-related kinase B (TrkB) receptor, whereas pro-BDNF promotes apoptosis upon binding to the receptors of neurotrophin p75 (p75NTR) and sortilin [1,2,20]. Neurons exposed to elevated levels of pro-BDNF or reduced levels of m-BDNF may undergo long-term depression, synaptic retraction, and cell death, which may be exacerbated by aging through the depletion of extracellular proteases responsible for BDNF maturation [1,2,20]. The Met allele impairs m-BDNF production, which contributes to the loss of nigral dopaminergic neurons in PD [1,2].

If experimental evidence demonstrated a dose-dependent detrimental effect of the Met allele on the survival of dopaminergic neurons, a stronger rationale for investigating genotype-specific effects on clinical outcomes would have been established. In the absence of such literature, the genotypes were dichotomized into the homozygous Val group and the Met allele carrier group for subsequent analyses.

Early-stage PD patients who exhibited mild parkinsonism at the time of cohort enrollment were included. This facilitated cross-sectional comparisons of baseline characteristics, as differing disease stages could distort interpretation. Initial workups did not reveal any significant genotypic differences, except for cardiac sympathetic denervation and the cognitive profile. Motor severity, orthostatic stress, nonmotor aspects, and quality-of-life measures were comparable across genotypes.

Cardiac denervation and nOH independently reflect disease burden and progression [21-24]. They result primarily from peripheral dysautonomia but also mirror the central pathology [25-27]. OH implies diffuse neuropathology [23,27]. While orthostatic dysregulation did not differ between the genotypes in this study, the Met allele carriers had a higher incidence of nOH with MAP75 and more damaged myocardial denervation. This suggests that the BNDF polymorphism could correspond to a noradrenergic subtype, which is relevant to the body-first model [26,28]. This requires validation in future studies because our longitudinal analyses failed to confirm that Met allele carriers are fast progressors [26,29].

In the longitudinal analyses, the Met carrier genotype was used as a reference because it contrasted with the Caucasian frequency and was associated with milder motor severity [1,30]. While the motor severity of the Met carrier group steadily increased over time, the Val/Val genotype exceeded that of the Met allele genotype after accelerating beyond 5–6 years of disease duration. Previous studies suggested that these findings were due to abrupt changes in vulnerability resulting from enhanced degeneration and the loss of compensatory mechanisms [31,32]. This finding was not anticipated considering its physiologic function [2] but could be partially explained by its age-dependent role [20,33]. In aged controls older than 65 years, the Val/Val genotype may promote pro-BDNF secretion, which has a detrimental effect on neuronal vulnerability [20].

The effect of the BDNF polymorphism on quality of life and nonmotor burden over the follow-up period was not significant in this study. Although the Met allele genotype significantly worsened in a linear fashion, it did not differ from the Val genotype.

This cohort did not demonstrate an effect of BDNF on global cognition, which is consistent with the findings of a previous study [34]. However, distinct genotypic patterns of cognitive deterioration were observed. A dynamic relationship was observed with the Val homozygote group. In the early stage, the attention/working memory domain was preserved by its physiologic maintenance of dopamine levels, but as the disease progressed, the Val/Val genotype showed accelerated worsening of frontal cognition because of its age-dependent detrimental effect [1,2,20,33]. This interpretation of an inverse relationship across disease spans was also supported by a previous longitudinal study with the COMT polymorphism [34]. The initial nonfrontal cognitive domain, which included delayed recall, was more preserved in the Met carriers [35]. The effect of BDNF was not affected by APOE genotype.

Cardiac denervation progressed linearly with disease duration, without the conditional influence of BDNF genotype.

This study encountered several limitations, including a substantial number of dropouts during the longitudinal follow-up, incomplete assessments conducted by certain individuals, and discrepancies in the timing of individual evaluations, which led to varying intervals. These shortcomings hindered the authors from disentangling the effect of dopaminergic supplementation across specific genotypes. It was paramount to adjust for the dopaminergic influence on motor severity, but owing to large dropouts, a mixed model with an additional time-varying predictor (levodopa-equivalent dose) at level 1 (within-person) was anticipated to fail in estimating the model. The study was thus predesigned to construct a parsimonious model with only one level-1 predictor, disease duration. As this study aimed to observe genotype-specific natural disease progression in real-world clinical practice, our model sufficiently aligned with this objective. Motor severity was not evaluated during the OFF state, except at baseline. This might have led to a biased estimate of the association between genotype and motor progression because of unstandardized medication effects on each individual patient. However, there is no standardized method to ensure a total washout OFF state because of the long-duration response to levodopa in treated PD patients [36]. The optimal medical response is also subjective for each patient; thus, the study investigated the optimal motor ON state for standardized measurements. It was also not feasible to withdraw medications long enough to obtain a full OFF state in clinical practice for chronic PD. This study did not rate Hoehn and Yahr staging to evaluate overall motor progression because it did not fully represent the clinical status, and its clinimetric properties were not well suited for the purpose of this study [37]. Finally, the effect of cognition-enhancing medications was not adjusted in the models, which could be the source of type II error in Model 3 (Supplementary Table 3).

The strength of this study is that the cohort comprises a relatively large population with extensive examinations and a lengthy follow-up period. A previous study used a longitudinal design to examine the influence of BDNF polymorphisms; however, the follow-up duration was too short, allowing assessment of changes in outcome measures from baseline at only a single time point. Limited domains of PD have also been investigated [33]. In contrast, our study’s prolonged follow-up periods enabled the two neurologists to exclude atypical parkinsonian disorders that could mimic PD in their early diagnosis and maintain the validity of the cohort. The dataset also included both clinical metrics that fully captured the motor and nonmotor domains of the disease and objective surrogates of central and peripheral pathobiology. Extensive examinations conducted across the disease course allowed the prediction of protracted disease progression trajectories that would not have been possible with shorter follow-up periods. The mixed model adequately analyzed these unevenly spaced longitudinal data, which inevitably entailed missing data and autoregression. By providing cross-sectional and within-person longitudinal data, our study uniquely captured both early-stage PD at baseline and disease progression over time stratified by genotype. These findings were unprecedented compared with those of previous cross-sectional and longitudinal studies [10,11,35]. The orthostatic challenge was specified by the MAP at the upright position (MAP75). Neurogenic OH with MAP75 further detailed cardiovascular dysautonomia to fully portray clinically significant OH. Our data also excluded the confounding effect of APOE polymorphism, as it occurred even across the BDNF genotypes [38].

It has been widely reported that the frequencies of BDNF genotypes differ across ethnicities, and their effects on PD patients in a Korean cohort have rarely been investigated. The considerable differences in myocardial denervation at diagnosis and nonmotor burden and motor progression might suggest a differential ethnic role of BDNF polymorphism in disease evolution in PD among the Korean population.

Supplementary Materials

Notes

Conflicts of Interest

The authors have no financial conflicts of interest.

Funding Statement

This research was supported by the “Korea National Institute of Health” research project (2024ER100202 awarded to Joong-Seok Kim). The Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2021R1I1A1A01050492/RS-2021-NR065151 awarded to Sang-Won Yoo) supported this. This was also supported by the NRF grant funded by the Korea government (Ministry of Science and ICT, RS-2024-00452428 awarded to Sang-Won Yoo), and the Ministry of Science, ICT and Future Planning (NRF-2017R1D1A1B06028086/ RS-2017-NR027859 awarded to Joong-Seok Kim).

Acknowledgments

We thank our patients for their understanding and generosity.

Author Contributions

Conceptualization: Sang-Won Yoo, Yun Joong Kim, Joong-Seok Kim. Data curation: Sang-Won Yoo, Dong-Woo Ryu, Yoonsang Oh, Seunggyun Ha, Joong-Seok Kim. Formal analysis: Sang-Won Yoo, Yun Joong Kim, Joong-Seok Kim. Funding acquisition: Sang-Won Yoo, Joong-Seok Kim. Investigation: all authors. Methodology: Sang-Won Yoo, Yun Joong Kim, Joong-Seok Kim. Project administration: Joong-Seok Kim. Resources: all authors. Software: Sang-Won Yoo, Yun Joong Kim, Seunggyun Ha, Joong-Seok Kim. Supervision: Joong-Seok Kim. Validation: all authors. Visualization: Sang-Won Yoo, Yun Joong Kim, Seunggyun Ha, Joong-Seok Kim. Writing—original draft: Sang-Won Yoo. Writing—review & editing: all authors.

Figure 1.

A flowchart of overall evaluations in this longitudinal study. For each investigation, the baseline evaluation was defined as T₀, and subsequent follow-up time points were numbered sequentially (T₁, T₂, T₃). The average intervals with standard deviations between successive assessments were illustrated in the figure and expressed in months. The questionnaire included the Non-Motor Symptoms Scale (NMSS), Montgomery-Asberg Depression Rating Scale, Epworth Sleepiness Scale, Parkinson’s Disease Sleep Scale-2, REM Sleep Behavior Disorder Screening Questionnaire, and 39-Item Parkinson’s Disease Questionnaire (PDQ39) at baseline. The NMSS and PDQ39 summay index were repeatedly assessed. Val, valine; Met, methionine; ¹⁸F-FP-CIT, ¹⁸F-N-(3-fluoropropyl)-2β-carbomethoxy-3β-(4-iodophenyl)nortropane; PET, positron emission tomography; CT, computed tomography; MAPstanding, orthostatic mean arterial pressure; nOH, neurogenic orthostatic hypotension; MDS-UPDRS, Movement Disorder Society–Unified Parkinson’s Disease Rating Scale; ¹²³I-MIBG, ¹²³I-meta-iodobenzylguanidine; SNSB, Seoul Neuropsychological Screening Battery.

Figure 2.

The trajectory of motor, non-motor, quality of life, and cognition across disease duration, stratified by BDNF genotypes. A: Motor function (Model 1). Part II and III represented the respective MDS-UPDRS subsets of motor scores. The total motor scores represented the sum of Part II and III motor scores. B: Nonmotor function and quality of life (Model 2). The composite score represented the average of standardized scores of NMSS, MADRS, ESS, PDSS-2, and RBDSQ to reflect the nonmotor overall burden. C: Cognitive function (Model 3). The frontal cognition was defined as the average z-scores of attention/working memory and frontal/executive cognitive domains, and the average of delayed recall, language, and visuospatial domains comprised the nonfrontal cognition. Global cognition was defined as the mean of the frontal and nonfrontal cognitive scores. D: Cardiac denervation (Model 4). *p<0.05. Statistical results for each model are presented in Supplementary Table 3. Val, valine; Met, methionine; MDS-UPDRS, Movement Disorder Society–Unified Parkinson’s Disease Rating Scale; NMSS, Non-Motor Symptoms Scale; PDQ39SI, 39-Item Parkinson’s Disease Questionnaire summary index; MADRS, Montgomery-Asberg Depression Rating Scale; ESS, Epworth Sleepiness Scale; PDSS-2, Parkinson’s Disease Sleep Scale-2; RBDSQ, REM Sleep Behavior Disorder Screening Questionnaire; HME, early heart-to-mediastinum ratio; HML, late heart-to-mediastinum ratio.

REFERENCES

• 1. Shen T, You Y, Joseph C, Mirzaei M, Klistorner A, Graham SL, et al. BDNF polymorphism: a review of its diagnostic and clinical relevance in neurodegenerative disorders. Aging Dis 2018;9:523–536.ArticlePubMedPMC
• 2. Urbina-Varela R, Soto-Espinoza MI, Vargas R, Quiñones L, Del Campo A. Influence of BDNF genetic polymorphisms in the pathophysiology of aging-related diseases. Aging Dis 2020;11:1513–1526.ArticlePubMedPMC
• 3. Hyman C, Hofer M, Barde YA, Juhasz M, Yancopoulos GD, Squinto SP, et al. BDNF is a neurotrophic factor for dopaminergic neurons of the substantia nigra. Nature 1991;350:230–232.ArticlePubMedPDF
• 4. Blöchl A, Sirrenberg C. Neurotrophins stimulate the release of dopamine from rat mesencephalic neurons via Trk and p75Lntr receptors. J Biol Chem 1996;271:21100–21107.ArticlePubMed
• 5. Howells DW, Porritt MJ, Wong JY, Batchelor PE, Kalnins R, Hughes AJ, et al. Reduced BDNF mRNA expression in the Parkinson’s disease substantia nigra. Exp Neurol 2000;166:127–135.ArticlePubMed
• 6. Momose Y, Murata M, Kobayashi K, Tachikawa M, Nakabayashi Y, Kanazawa I, et al. Association studies of multiple candidate genes for Parkinson’s disease using single nucleotide polymorphisms. Ann Neurol 2002;51:133–136.ArticlePubMed
• 7. Håkansson A, Melke J, Westberg L, Shahabi HN, Buervenich S, Carmine A, et al. Lack of association between the BDNF Val66Met polymorphism and Parkinson’s disease in a Swedish population. Ann Neurol 2003;53:823.ArticlePubMed
• 8. Guerini FR, Beghi E, Riboldazzi G, Zangaglia R, Pianezzola C, Bono G, et al. BDNF Val66Met polymorphism is associated with cognitive impairment in Italian patients with Parkinson’s disease. Eur J Neurol 2009;16:1240–1245.ArticlePubMed
• 9. Karakasis C, Kalinderi K, Katsarou Z, Fidani L, Bostantjopoulou S. Association of brain-derived neurotrophic factor (BDNF) Val66Met polymorphism with Parkinson’s disease in a Greek population. J Clin Neurosci 2011;18:1744–1745.ArticlePubMed
• 10. Altmann V, Schumacher-Schuh AF, Rieck M, Callegari-Jacques SM, Rieder CR, Hutz MH. Val66Met BDNF polymorphism is associated with Parkinson’s disease cognitive impairment. Neurosci Lett 2016;615:88–91.ArticlePubMed
• 11. Foltynie T, Lewis SG, Goldberg TE, Blackwell AD, Kolachana BS, Weinberger DR, et al. The BDNF Val66Met polymorphism has a gender specific influence on planning ability in Parkinson’s disease. J Neurol 2005;252:833–838.ArticlePubMedPDF
• 12. Oh S, Sohn HY, Seo J, Kang E, Park JK, Moon SY, et al. Profile for Brain Disease Research Infrastructure for Data Gathering and Exploration (BRIDGE) platform. Aging Dis 2026;17:499–514.Article
• 13. Postuma RB, Berg D, Stern M, Poewe W, Olanow CW, Oertel W, et al. MDS clinical diagnostic criteria for Parkinson’s disease. Mov Disord 2015;30:1591–1601.ArticlePubMedPMC
• 14. Goetz CG, Stebbins GT, Tilley BC. Calibration of unified Parkinson’s disease rating scale scores to Movement Disorder Society-unified Parkinson’s disease rating scale scores. Mov Disord 2012;27:1239–1242.ArticlePubMedPDF
• 15. Litvan I, Goldman JG, Tröster AI, Schmand BA, Weintraub D, Petersen RC, et al. Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Mov Disord 2012;27:349–356.ArticlePubMedPMCPDF
• 16. Goldman JG, Holden S, Ouyang B, Bernard B, Goetz CG, Stebbins GT. Diagnosing PD-MCI by MDS Task Force criteria: how many and which neuropsychological tests? Mov Disord 2015;30:402–406.ArticlePubMedPMCPDF
• 17. Yoo SW, Ryu DW, Oh Y, Ha S, Lyoo CH, Kim JS. Unraveling olfactory subtypes in Parkinson’s disease and their effect on the natural history of the disease. J Neurol 2024;271:6102–6113.ArticlePubMedPDF
• 18. Ryu HJ, Yang DW. The Seoul Neuropsychological Screening Battery (SNSB) for comprehensive neuropsychological assessment. Dement Neurocogn Disord 2023;22:1–15.ArticlePubMedPMCPDF
• 19. Pivac N, Kim B, Nedić G, Joo YH, Kozarić-Kovacić D, Hong JP, et al. Ethnic differences in brain-derived neurotrophic factor Val66Met polymorphism in Croatian and Korean healthy participants. Croat Med J 2009;50:43–48.ArticlePubMedPMC
• 20. Erickson KI, Kim JS, Suever BL, Voss MW, Francis BM, Kramer AF. Genetic contributions to age-related decline in executive function: a 10-year longitudinal study of COMT and BDNF polymorphisms. Front Hum Neurosci 2008;2:11.ArticlePubMedPMC
• 21. Yoo SW, Kim JS, Oh YS, Ryu DW, Ha S, Yoo JY, et al. Cardiac sympathetic burden reflects Parkinson disease burden, regardless of high or low orthostatic blood pressure changes. NPJ Parkinsons Dis 2021;7:71.ArticlePubMedPMCPDF
• 22. Lee JE, Kim JS, Ryu DW, Oh YS, Yoo IR, Lee KS. Cardiac sympathetic denervation can predict the wearing-off phenomenon in patients with Parkinson disease. J Nucl Med 2018;59:1728–1733.ArticlePubMed
• 23. Fereshtehnejad SM, Romenets SR, Anang JB, Latreille V, Gagnon JF, Postuma RB. New clinical subtypes of Parkinson disease and their longitudinal progression: a prospective cohort comparison with other phenotypes. JAMA Neurol 2015;72:863–873.ArticlePubMed
• 24. De Pablo-Fernandez E, Tur C, Revesz T, Lees AJ, Holton JL, Warner TT. Association of autonomic dysfunction with disease progression and survival in Parkinson disease. JAMA Neurol 2017;74:970–976.ArticlePubMedPMC
• 25. Jain S, Goldstein DS. Cardiovascular dysautonomia in Parkinson disease: from pathophysiology to pathogenesis. Neurobiol Dis 2012;46:572–580.ArticlePubMedPMC
• 26. Horsager J, Borghammer P. Brain-first vs. body-first Parkinson’s disease: an update on recent evidence. Parkinsonism Relat Disord 2024;122:106101.ArticlePubMed
• 27. Udow SJ, Robertson AD, MacIntosh BJ, Espay AJ, Rowe JB, Lang AE, et al. ‘Under pressure’: is there a link between orthostatic hypotension and cognitive impairment in α-synucleinopathies? J Neurol Neurosurg Psychiatry 2016;87:1311–1321.ArticlePubMed
• 28. Ray Chaudhuri K, Leta V, Bannister K, Brooks DJ, Svenningsson P. The noradrenergic subtype of Parkinson disease: from animal models to clinical practice. Nat Rev Neurol 2023;19:333–345.ArticlePubMedPDF
• 29. Borghammer P. The α-synuclein origin and connectome model (SOC Model) of Parkinson’s disease: explaining motor asymmetry, non-motor phenotypes, and cognitive decline. J Parkinsons Dis 2021;11:455–474.ArticlePubMedPMC
• 30. Fischer DL, Auinger P, Goudreau JL, Paumier KL, Cole-Strauss A, Kemp CJ, et al. Bdnf variant is associated with milder motor symptom severity in early-stage Parkinson’s disease. Parkinsonism Relat Disord 2018;53:70–75.ArticlePubMed
• 31. Prange S, Danaila T, Laurencin C, Caire C, Metereau E, Merle H, et al. Age and time course of long-term motor and nonmotor complications in Parkinson disease. Neurology 2019;92:e148–e160.ArticlePubMed
• 32. Yoo SW, Ryu DW, Oh YS, Ha S, Lyoo CH, Kim Y, et al. Estimating motor progression trajectory pursuant to temporal dynamic status of cardiac denervation in Parkinson’s disease. J Neurol 2024;271:2019–2030.ArticlePubMedPDF
• 33. van der Kolk NM, Speelman AD, van Nimwegen M, Kessels RP, IntHout J, Hakobjan M, et al. BDNF polymorphism associates with decline in set shifting in Parkinson’s disease. Neurobiol Aging 2015;36:1605.e1–e6.ArticlePubMed
• 34. Williams-Gray CH, Evans JR, Goris A, Foltynie T, Ban M, Robbins TW, et al. The distinct cognitive syndromes of Parkinson’s disease: 5 year follow-up of the CamPaIGN cohort. Brain 2009;132:2958–2969.ArticlePubMed
• 35. Białecka M, Kurzawski M, Roszmann A, Robowski P, Sitek EJ, Honczarenko K, et al. BDNF G196A (Val66Met) polymorphism associated with cognitive impairment in Parkinson’s disease. Neurosci Lett 2014;561:86–90.ArticlePubMed
• 36. Anderson E, Nutt J. The long-duration response to levodopa: phenomenology, potential mechanisms and clinical implications. Parkinsonism Relat Disord 2011;17:587–592.ArticlePubMed
• 37. Goetz CG, Poewe W, Rascol O, Sampaio C, Stebbins GT, Counsell C, et al. Movement Disorder Society Task Force report on the Hoehn and Yahr staging scale: status and recommendations. Mov Disord 2004;19:1020–1028.ArticlePubMed
• 38. Coughlin DG, Hurtig HI, Irwin DJ. Pathological influences on clinical heterogeneity in Lewy body diseases. Mov Disord 2020;35:5–19.ArticlePubMedPMCPDF

Figure & Data

References

Citations

Citations to this article as recorded by  

• Commentary for “Longitudinal Implications of the BDNF rs6265 Polymorphism for Motor and Nonmotor Features of Parkinson’s Disease in the Korean Population”
  Sun Ju Chung
  Journal of Movement Disorders.2026; 19(2): 242.     CrossRef

Comments on this article