Abstract
Neurofilament light chain (NFL) level in biofluids is a sensitive measure of axonal damage and a promising biomarker in neurodegenerative diseases. In Parkinson’s disease (PD), NFL can distinguish PD from other parkinsonian disorders, and NFL concentration is associated with disease severity, risk of progression, and survival. To determine whether serum NFL at baseline in de novo PD predicts motor decline, differentially impacts specific motor features, predicts cognitive decline, and predicts loss of dopamine terminals, here we evaluated 376 de novo PD patients from the PPMI database and analyzed the effect of baseline serum NFL levels on progression over eight years of motor impairment measured with the UPDRS, cognitive function measured with the MoCA, and putamen dopamine transporter (DAT) binding ratio measured with DaTscan. In longitudinal mixed effects models that controlled for age, gender, disease duration, and levodopa equivalent drug dose, higher levels of serum NFL at baseline were associated with greater increases of UPDRS-III and total UPDRS scores, with greater worsening of postural instability and gait disorder (PIGD) scores but not tremor scores over time. In contrast, baseline serum NFL was not associated with significant progression of MoCA scores in this de novo PD cohort. Higher baseline serum NFL was associated with greater reduction of putamen DAT binding ratio over time. Together, these findings show that baseline serum NFL levels predict the rate of motor decline, the accumulation of PIGD clinical features, and the progression of dopamine transporter loss in the early stage of PD.
Keywords: Neurofilament light chain, Parkinson’s disease, Dopamine transporter uptake, Motor decline, PIGD, Longitudinal study
1. Introduction
Fluid analytes that reflect severity and progression of idiopathic Parkinson’s disease (PD) have long been sought to guide clinical decision making and clinical trials. To this end, neurofilament light chain (NFL) has been proposed as a sensitive biomarker of neuroaxonal injury in various neurological disorders, including Alzheimer’s disease, frontotemporal dementia, amyotrophic lateral sclerosis, cerebral small vessel disease and multiple sclerosis [1]. Although serum NFL levels are elevated in atypical parkinsonian disorders compared to PD [2], higher NFL concentrations in PD in blood or cerebrospinal fluid have been associated with greater disease severity, shorter survival, and higher risk of progression of motor and cognitive impairments [3,4].
Recent work examining the impact of NFL on progression of PD has found that NFL predicts progression specifically in the postural instability gait disorder (PIGD) subtype of PD [5], a clinical variant associated with poorer prognosis and cognitive impairment [6]. As patients often transition between PIGD and tremor subtypes over time [7–9], however, it remains possible that NFL has prognostic value in predicting the rate of progression of PIGD clinical features in all PD patients, regardless of their dominant clinical features.
Dopamine transporter (DAT) imaging [10] provides an antemortem measure of the integrity of nigrostriatal dopaminergic terminals that are lost in PD. Although reduced DAT uptake in the putamen correlates with the severity of motor impairment, including all cardinal motor symptoms except tremor [11], DAT uptake and motor function can dissociate, for example in the context of optimized dopamine replacement therapy, as well as due to variance in duration of disease at the group level, and in the clinical assessment of motor function. Recent studies have investigated the potential association between DAT uptake and NFL levels cross-sectionally, but the results have been inconsistent [3,12]. Despite the value of this biomarker of dopamine system integrity, the relationship between baseline NFL and longitudinal decline of DAT uptake has not yet been evaluated.
To address these knowledge gaps, here we sought to determine in patients with de novo PD followed for up to 8 years in the PPMI database whether serum NFL levels at baseline predict (1) progression of motor disability overall, (2) progression of PIGD versus tremor clinical features, (3) progression of global cognitive function, and (4) change in dopamine transporter (DAT) uptake over time.
2. Methods
2.1. Participants
The participants of this study were recruited into the Parkinson Progression Markers Initiative (PPMI) database (available at http://www.ppmi-info.org). All recruited participants were designed to have follow-up examinations each year after their baseline visit. The PPMI study was approved by the Institutional Review Board of all participating sites; all participants provided written informed consent before inclusion.
In this study, 376 de novo PD patients with available serum NFL data at baseline and with two or more visits were included. Four percent of subjects had two timepoints of non-missing data; 3.5% had three time points; 6.4% had four time points; 11.4% had five time points; 15.2% had six time points; 33.2% had seven time points; 20.7% had eight time points; and 5.6% had nine time points.
2.2. Clinical assessments and dopamine transporter imaging
PD-related signs and symptoms were assessed with the Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (UPDRS) every year from enrollment up to 8 years. Motor function was measured with the UPDRS part 3 (UPDRS-III) in the off-state. Cognitive function was measured with the Montreal Cognitive Assessment (MoCA).
To exclude the possibility that certain clinical features at baseline might be associated with different durations of follow-up in the study, PD participants were classified into short and long follow-up subgroups by median split of their maximum follow-up years (median, 6 years): If the maximum follow-up time was no more than 6 years, the subject was classified into the short follow-up subgroup. Otherwise, the subject was classified into the long follow-up subgroup. Baseline age, gender, disease duration, total UPDRS and UPDRS-III scores were identical between the two subgroups (p > 0.05, Supplementary Table 1). Subsequent analyses were performed on the entire dataset.
PIGD scores and tremor scores were computed according to previously defined criteria [13]. Briefly, the PIGD measure includes five items, which are freezing, walking and balance in UPDRS part 2 (UPDRS-II), and gait, freezing of gait, and postural stability in UPDRS-III; the tremor measure includes 11 items, which are tremor in UPDRS-II, and postural tremor, kinetic tremor, rest tremor and rest constancy in UPDRS-III. The ratio of tremor score to PIGD score was used to define tremor-dominant (TD) patients (ratio ≥ 1.15; N = 264), indeterminate patients (0.9 < ratio <1.15, N = 43) and PIGD-dominant patients (ratio ≤ 0.90, N = 69) [13]. The percentages of patients in each motor subtype at baseline and during the follow-up period are shown in Supplementary Table 2. Akinetic-rigid scores were computed by adding eight items including rigidity, finger tapping, hand movements, pronation-supination movements of hands, toe tapping, leg agility, arising from chair, and body bradykinesia in UPDRS-III [14].
DaTscan SPECT images were acquired at enrollment and in follow-up years one, two, and four. 355 participants had at least two DaTscan SPECT images. DAT binding ratios in the putamen were calculated as the count density in the putamen divided by that of the occipital cortex, which served as reference region. All participants were free of anti-parkinsonism drugs at enrollment, and levodopa equivalent drug dose (LEDD) was calculated at each follow-up visit [15]. Disease duration was defined as the duration at enrollment since the onset of symptoms. Serum NFL levels at baseline were measured with the Simoa Human NF-light Advantage kit by using the Single Molecule Array in a fully automated SIMOA® HD-1 analyzer (Quanterix, Lexington, MA, USA). Additional details about processing of the samples can be found in the PPMI biologic manual (http://www.ppmi-info.org).
2.3. Statistical analyses
All clinical, serum NFL, CSF NFL, and neuroimaging data included in this study were simultaneously downloaded from the PPMI database on June 30, 2020. Correlations between serum NFL levels and clinical measurements at baseline were assessed via Spearman correlation analysis. Differences of serum NFL levels by sex were compared with the Wilcoxon rank sum test. To analyze the effect of serum NFL on motor progression, longitudinal mixed effects analysis was performed for the dependent variable of motor severity, reflected by different measurements including UPDRS-III scores, total UPDRS scores, PIGD scores, tremor scores, or akinetic-rigid scores, each in a separate analysis. Model selection was based on limited backward stepwise elimination (p > 0.05 for removal from the model), in which we ran pretest checks on fixed covariates and higher order interactions, and in which we removed them if they showed only chance relations and therefore were not confounders. The primary fixed effect predictors were levels of serum NFL at baseline, years in the study, and their interactions. The fixed covariates included sex, age at baseline, disease duration at baseline, their interactions with years in the study, and time varying LEDD. An intercept term and linear rate of change across time per subject were the random terms for the mixed effects model. For the purpose of analyzing the effect of serum NFL on changes of DAT uptake deficits, we ran the same longitudinal mixed effects model for the dependent variable of DAT uptake. As before, only those covariates reaching statistical significance were considered as confounders and were included in the final models.
To analyze the relationship between change in DAT uptake and change in motor function across time, we performed a longitudinal mixed effects analysis for the time-varying dependent variable of motor features, reflected by different measurements including total UPDRS, UPDRS-III, PIGD, tremor, and akinetic-rigid scores, each in a separate analysis. Time-varying DAT binding ratio in putamen was included as the primary fixed effect predictor. Years in the study, sex, baseline age and disease duration, their interactions with years in the study, and time varying LEDD were included as the fixed covariates. By controlling for years in the study in the model in addition to DAT uptake, we excluded the possible emergence of a spurious relation of time-varying DAT uptake to the dependent variable that is solely the result of each of these variables independently changing across time but being otherwise statistically unrelated. The random terms for the mixed effects model included an intercept term and linear rate of change across time per subject.
To analyze the effect of serum NFL levels on cognitive function over time, we performed a longitudinal mixed effects analysis for the dependent variable of MoCA scores. Levels of serum NFL at baseline, years in the study, and their interaction were included as the primary fixed effect predictors. Age and disease duration at baseline, sex, years of education, their interactions with years in the study, and time varying treatment for cognitive dysfunction with acetylcholine esterase inhibitors were included as the fixed covariates. Once again, an intercept term and linear rate of change across time per subject were the random terms for the mixed effects model. We also ran a statistical analysis for CSF NFL levels similar to those involving the serum NFL levels. Given the lower sample size of de novo PD subjects with CSF samples, this study focused on the relation of serum NFL levels to progression of motor features.
Residuals from model fixed effect predictions and combined fixed and random predictions were checked for conformance to normality assumptions. A value of p < 0.05 was considered statistically significant. Standard error is reported unless stated otherwise. Statistical analysis was performed in R software (version 3.6.3, available at http://www.r-project.org).
3. Results
3.1. Demographic and clinical features
The demographic and clinical features of all 376 de novo PD participants are shown in Table 1. The average age at baseline was 62.2 ± 9.8 years (SD), and the percentage of males was 63.8%. PD patients were newly diagnosed and drug naïve, with mean duration since symptom onset of 2.0 ± 2.0 years (SD). The average follow-up time was 5.7 ± 1.6 years (SD).
Table 1.
Study population demographic and clinical data at baseline.
| Clinical characteristics | De novo PD |
|---|---|
| No. of individuals (N) | 376 |
| Age (mean years, SD) | 62.2 (9.8) |
| No. of males (N, %) | 240 (63.8) |
| Disease duration (mean years, SD) | 2.0 (2.0) |
| Total UPDRS (mean, SD) | 32.2 (13.1) |
| (median, range) | 31 (7, 70) |
| UPDRS-III (mean, SD) | 20.9 (8.9)”” |
| (median, range) | 20 (4, 51) |
| PIGD score (mean, SD) | 1.1 (1.1) |
| (median, range) | 1 (0, 6) |
| Tremor score (mean, SD) | 5.4 (3.5) |
| (median, range) | 5 (0, 20) |
| Akinetic-rigid score (mean, SD) | 13.6 (7.3) |
| (median, range) | 12 (2, 34) |
| MoCA (mean, SD) | 27.1 (2.4) |
| (median, range) | 27 (17, 30) |
| Serum NFL levels (mean pg/ml, SD) | 13.0 (7.1) |
| (median, range) | 11.5 (1.8, 76.6) |
| DAT binding ratio (mean, SD) | 0.8 (0.3) |
| (median, range)a | 0.8 (0.2, 2.2) |
355 participants underwent baseline DaTscan SPECT.
At baseline, serum NFL concentrations positively correlated with age (rho = 0.60, p < 0.001) but did not differ by sex (p = 0.6). No association was detected between serum NFL levels and disease duration (rho = 0.04, p = 0.5). Higher levels of serum NFL were associated with more severe parkinsonism, as measured by total UPDRS scores and UPDRS-III scores (total UPDRS: rho = 0.17, p < 0.001; UPDRS-III: rho = 0.16, p = 0.001). Higher levels of serum NFL were also associated with greater cognitive impairment, as measured by MoCA scores (rho = −0.21, p < 0.001). In this cohort, serum NFL levels at baseline were not significantly associated with the putamen DAT binding ratio, and only a weak negative trend was observed (rho = − 0.09, p = 0.092).
3.2. Predictive effects of baseline serum NFL levels on the rate of PD-related functional decline
In longitudinal analysis, elevated levels of serum NFL at baseline were associated with a significantly faster increase in UPDRS-III scores over 8 years (p = 0.003). Similar results were found for total UPDRS scores (p < 0.001). For each increase of 10 pg per milliliter (pg/ml) in serum NFL at baseline, UPDRS-III scores and total UPDRS scores climbed each year by 0.52 ± 0.17 points and 0.96 ± 0.28 points, respectively (Fig. 1A and B). Interestingly, compared to females, males had a steeper increase of UPDRS-III and total UPDRS scores over time (UPDRS-III, p = 0.024; total UPDRS scores: p = 0.005), with an approximate growth of 0.54 ± 0.24 excess points for UPDRS-III scores and 1.10 ± 0.38 excess points for total UPDRS scores per year in males compared to females.
Fig. 1.
Baseline serum NFL levels in de novo PD are associated with disease progression. Predicted values from model fixed effects are shown for each of the following clinical features for three levels of baseline serum NFL (adjusted for other predictor and covariate terms): (A). Total UPDRS score; (B). UPDRS-III score; (C). PIGD score; (D). DAT binding ratio in putamen. The predictive values of NFL reflect low (1 standard deviation below mean), mean, and high (1 standard deviation above mean) levels.
3.3. Predictive effects of baseline serum NFL levels on accumulation of PD motor features
To determine whether baseline serum NFL might differentially predict progression of specific clinical features of this de novo PD cohort, we next evaluated the impact of baseline serum NFL levels on progression of PIGD scores, tremor scores, and akinetic-rigid scores. Interestingly, higher levels of serum NFL at baseline were associated with a faster increase of PIGD scores over 8 years (p < 0.001), with an elevation of 2.46 ± 0.43 points per decade in patients with a 10 pg/ml increase of serum NFL at baseline (Fig. 1C). In contrast, serum NFL levels at baseline were not associated with progression of tremor scores. Higher levels of serum NFL at baseline were associated with a faster increase of akinetic-rigid scores over time (p = 0.006, Supplementary Fig. 1), but this association was not significant in the initial model when controlling for all covariates.
We next set out to evaluate whether the NFL-associated accumulation of PIGD impairments was restricted to PIGD-dominant patients defined at baseline (see Methods) or was a more general finding in the de novo PD cohort. Within the PIGD-dominant subgroup but not the tremor-dominant subgroup, higher baseline NFL levels predicted faster increases of total UPDRS and UPDRS-III scores (total UPDRS: p = 0.006; UPDRS-III: p = 0.006). However, in both PIGD-dominant and tremor-dominant subgroups, higher baseline NFL levels were associated with faster worsening of PIGD scores (PIGD subgroup: p < 0.001; TD subgroup: p = 0.039). In contrast, and as expected from the analyses above, baseline NFL levels were not associated with worsening tremor scores in either subgroup. Thus, the impact of baseline NFL on PIGD impairments was not subtype specific and appeared to affect the de novo PD cohort more broadly.
3.4. Predictive effects of baseline serum NFL levels on the rate of cognitive decline
We built a mixed effects longitudinal model to evaluate whether baseline serum NFL levels could predict the progression of cognitive impairment in this cohort of de novo PD patients. We found that levels of serum NFL at baseline were not significantly associated with the rate of cognitive decline as measured by MoCA scores over time (p = 0.15).
3.5. Associations between baseline serum NFL level and progressive reduction of DAT uptake
To evaluate whether baseline serum NFL level predicted deterioration of striatal dopamine, we next investigated NFL’s effects on longitudinally acquired putamen DAT binding ratio. Compared to the original cohort, and to those without at least two DAT scans (N = 21), participants who underwent at least two DAT scans (N = 355) had comparable age, sex, disease duration and baseline UPDRS scores (p > 0.05 for each contrast, Supplementary Table 3). The average follow-up time for DAT scans was 3.5 ± 1.0 years. Although there was no significant cross-sectional association detected between serum NFL levels and putamen DAT binding ratio at baseline, as noted above, higher levels of serum NFL at baseline were associated with a steeper decline of putamen DAT binding ratio over time (p = 0.013). In this mixed-effects model, for each 10 pg/ml increase in serum NFL at baseline, putamen DAT binding ratio decreased by 0.14 ± 0.06 units per decade (Fig. 1D).
Given the association of baseline serum NFL levels with both worsening motor function and decline in DAT binding ratio over time, we next examined the relationship between changes in DAT binding over time and changes in motor function. Greater loss of DAT binding over time was associated with greater increases of UPDRS-III and total UPDRS scores (both p < 0.001), and specifically with greater increase of akinetic-rigid scores (p < 0.001), but not with changes in PIGD (p = 0.19) or tremor (p = 0.8) scores.
Of note, all of the key results that were significant after backward elimination in this longitudinal study were significant in the initial full model unless otherwise noted. Residuals from fixed and random predicted values reasonably conformed to normality assumptions for all longitudinal models.
3.6. Exploratory findings of CSF NFL levels in de novo PD patients
Among all 376 subjects in the study, only 207 (55%) had CSF NFL levels at baseline (Supplementary Table 4). These were evaluated on an exploratory basis in the Supplement. Serum NFL and CSF NFL were strongly correlated (N = 207, rho = 0.63, p < 0.001). Results with CSF NFL levels were broadly similar to those with serum NFL (Supplementary Fig. 2).
4. Discussion
In this eight-year longitudinal study of de novo PD, baseline serum NFL levels were found to predict decline in overall motor function, worsening of PIGD clinical features but not tremor, and loss of putamen DAT over time. Consistent with previous cross-sectional studies, NFL levels were found to correlate with baseline UPDRS-III scores [3,16,17]. Thus, NFL levels appear to reflect neuronal injury in de novo PD. While some previously reported longitudinal studies of NFL have found higher levels of NFL in blood or CSF to be associated with increased risk of PD motor progression [4], others have found only a trend correlation [18]. By employing a longitudinal design with dynamic analyses of clinical features, evaluating 8 years of longitudinal data compared to up to 3 years in previous studies, and restricting analyses to de novo PD with short (average, 0.6 years) duration of disease (compared to > 1 year in previous studies), we were able to demonstrate that a 10 pg/ml increase in serum NFL level at baseline in PD is associated with an increase of UPDRS-III score of 5 points per decade.
Interestingly, the effect of baseline NFL on progression of global motor impairments indexed with the UPDRS-III was reflected in a similar effect of baseline serum NFL on progression of PIGD clinical impairments, with a trend towards progression of akinetic-rigid impairments as well, but with no effect on progression of tremor. Few studies to date have investigated the associations between NFL and PIGD features, cross-sectionally [3,12] or longitudinally [5]. In a recent longitudinal study, serum NFL at baseline was associated with faster motor decline in a subgroup of patients with the PIGD subtype but not the tremor-dominant subtype defined at baseline [5]. Consistent with evidence of motor subgroup transitions in de novo PD [8] and with longer follow-up times [7], we found that while only 18.4% of patients could be classified into the PIGD subtype at baseline, PIGD clinical features evolved in relation to baseline NFL in the entire cohort and could be detected in both the PIGD-dominant subgroup as well as the tremor-dominant subgroup. As PIGD symptoms are linked to Hoehn and Yahr motor progression and have been closely associated with cognitive impairment in PD [7,19,20], the prognostic value of serum NFL for progression of PIGD symptoms may find utility in clinical trials targeting these patients in the early stage.
To our knowledge, this is the first observation linking NFL levels at baseline to loss of putamen DAT binding ratio over time in PD. This result suggests that neuroaxonal injury reflected in baseline NFL levels and detected at the early phase of disease is a relevant correlate of subsequent rate of dopaminergic neurodegeneration. Of note, the present results linking serum NFL to clinical course in PD are supported by cross-sectional studies where NFL levels in blood and CSF have been found to strongly correlate (r value in the current study of 0.6) [21], and where NFL levels in CSF have been observed to correlate with striatal DAT levels cross-sectionally [3]. It is intriguing that serum NFL has sensitivity for such progression. Together with the observation that changes in DAT binding over time were associated with progression of akinetic-rigid features rather than PIGD or tremor features, supported by prior cross-sectional and longitudinal results linking striatal DAT binding ratio to bradykinesia and rigidity in PD [22–24], these results suggest that loss of nigrostriatal integrity may contribute to the observed association between baseline NFL level and motor progression in PD primarily through its effects on bradykinesia and rigidity.
Limitations of this study include the lack of neuropathological diagnoses of participants with de novo PD. However, reduced striatal DAT uptake on DaTscan increases confidence of the clinical diagnosis of PD in this multi-center clinical study. Although the presence of some missing data in the dataset might have influenced the results, the mixed effects models built in our statistical analyses permitted reasonable estimates of effects in spite of some missing timepoints for some subjects, maximized use of available data, and optimized the power of the analyses [25]. In addition, we found no evidence for differential drop-out of subjects on the basis of baseline clinical features that could contribute to our observations, as baseline demographic and clinical features were comparable in the short follow-up de novo PD subgroup and the long follow-up subgroup. Another limitation is the lack of detailed cognitive testing that reduced sensitivity to detect an effect of NFL on changes in cognitive function over time, although we were able to detect a cross-sectional association between higher NFL levels with lower MOCA scores, as shown previously [4,12,26]. Strengths of this study include the long follow-up period, the relatively large sample size, and the study of a de novo PD cohort, which reduced heterogeneity in baseline disease severity and duration. In addition, multiple features of our mixed effects models touched base with prior observations, including the linear increase in UPDRS scores over time [27].
5. Conclusions
Baseline serum NFL levels in de novo PD predict decline in overall motor function, in PIGD features specifically, and in putamen DAT levels reflecting nigrostriatal dopamine neuropathology over an 8-year period. Given these findings and its ease of acquisition, serum NFL is likely to have value as a biomarker in PD clinical trials, with potential, for example, to discriminate fast from slow progressors. Further studies to understand the pathophysiological mechanisms of NFL as a predictive PD biomarker are needed.
Supplementary Material
Acknowledgements
The authors thank all the participants recruited in this study. Data used in the preparation of this article were obtained from the Parkinson’s Progression Markers Initiative (PPMI) database (https://www.ppmi-info.org/data). For up-to-date information on the study, visit https://www.ppmi-info.org.
Funding
Department of Defense CDMRP/W81XW1810516, NINDS 1R21 NS109833, Lewy Body Dementia Association Research Centers of Excellence Program.
Declaration of competing interest
RY, JL, AG, MQ, BZ have nothing to disclose. SG serves on an Advisory Board for Acadia Pharmaceuticals. He receives funding from NIH grant R01 AG054551, P50 AG005134, R21 NS109833, DOD CDMRP/W81XW1810516, the Farmer Family Parkinson’s Initiative, and the Lewy Body Dementia Association. None of the authors of this manuscript has any potential conflict of interest related to the content of this study.
Footnotes
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.parkreldis.2021.02.008.
References
- [1].Khalil M, Teunissen CE, Otto M, Piehl F, Sormani MP, Gattringer T, Barro C, Kappos L, Comabella M, Fazekas F, Petzold A, Blennow K, Zetterberg H, Kuhle J, Neurofilaments as biomarkers in neurological disorders, Nat. Rev. Neurol 14 (10) (2018) 577–589. [DOI] [PubMed] [Google Scholar]
- [2].Marques TM, van Rumund A, Oeckl P, Kuiperij HB, Esselink RAJ, Bloem BR, Otto M, Verbeek MM, Serum NFL discriminates Parkinson disease from atypical parkinsonisms, Neurology 92 (13) (2019) e1479–e1486. [DOI] [PubMed] [Google Scholar]
- [3].Backstrom D, Linder J, Jakobson Mo S, Riklund K, Zetterberg H, Blennow K, Forsgren L, Lenfeldt N, NfL as a biomarker for neurodegeneration and survival in Parkinson disease, Neurology 95 (7) (2020) e827–e838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Lin CH, Li CH, Yang KC, Lin FJ, Wu CC, Chieh JJ, Chiu MJ, Blood NfL: a biomarker for disease severity and progression in Parkinson disease, Neurology 93 (11) (2019) e1104–e1111. [DOI] [PubMed] [Google Scholar]
- [5].Ng ASL, Tan YJ, Yong ACW, Saffari SE, Lu Z, Ng EY, Ng SYE, Chia NSY, Choi X, Heng D, Neo S, Xu Z, Keong NCH, Tay KY, Au WL, Tan LCS, Tan EK, Utility of plasma Neurofilament light as a diagnostic and prognostic biomarker of the postural instability gait disorder motor subtype in early Parkinson’s disease, Mol. Neurodegener 15 (1) (2020) 33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].van der Heeden JF, Marinus J, Martinez-Martin P, Rodriguez-Blazquez C, Geraedts VJ, van Hilten JJ, Postural instability and gait are associated with severity and prognosis of Parkinson disease, Neurology 86 (24) (2016) 2243–2250. [DOI] [PubMed] [Google Scholar]
- [7].Alves G, Larsen JP, Emre M, Wentzel-Larsen T, Aarsland D, Changes in motor subtype and risk for incident dementia in Parkinson’s disease, Mov. Disord 21 (8) (2006) 1123–1130. [DOI] [PubMed] [Google Scholar]
- [8].Simuni T, Caspell-Garcia C, Coffey C, Lasch S, Tanner C, Marek K, Investigators P, How stable are Parkinson’s disease subtypes in de novo patients: analysis of the PPMI cohort? Park. Relat. Disord 28 (2016) 62–67. [DOI] [PubMed] [Google Scholar]
- [9].Eisinger RS, Martinez-Ramirez D, Ramirez-Zamora A, Hess CW, Almeida L, Okun MS, Gunduz A, Parkinson’s disease motor subtype changes during 20 years of follow-up, Park. Relat. Disord 76 (2020) 104–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Eshuis SA, Jager PL, Maguire RP, Jonkman S, Dierckx RA, Leenders KL, Direct comparison of FP-CIT SPECT and F-DOPA PET in patients with Parkinson’s disease and healthy controls, Eur. J. Nucl. Med. Mol. Imag 36 (3) (2009) 454–462. [DOI] [PubMed] [Google Scholar]
- [11].Benamer HT, Patterson J, Wyper DJ, Hadley DM, Macphee GJ, Grosset DG, Correlation of Parkinson’s disease severity and duration with 123I-FP-CIT SPECT striatal uptake, Mov. Disord 15 (4) (2000) 692–698. [DOI] [PubMed] [Google Scholar]
- [12].Sampedro F, Perez-Gonzalez R, Martinez-Horta S, Marin-Lahoz J, Pagonabarraga J, Kulisevsky J, Serum neurofilament light chain levels reflect cortical neurodegeneration in de novo Parkinson’s disease, Park. Relat. Disord 74 (2020) 43–49. [DOI] [PubMed] [Google Scholar]
- [13].Stebbins GT, Goetz CG, Burn DJ, Jankovic J, Khoo TK, Tilley BC, How to identify tremor dominant and postural instability/gait difficulty groups with the movement disorder society unified Parkinson’s disease rating scale: comparison with the unified Parkinson’s disease rating scale, Mov. Disord 28 (5) (2013) 668–670. [DOI] [PubMed] [Google Scholar]
- [14].Kang GA, Bronstein JM, Masterman DL, Redelings M, Crum JA, Ritz B, Clinical characteristics in early Parkinson’s disease in a central California population-based study, Mov. Disord 20 (9) (2005) 1133–1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, Clarke CE, Systematic review of levodopa dose equivalency reporting in Parkinson’s disease, Mov. Disord 25 (15) (2010) 2649–2653. [DOI] [PubMed] [Google Scholar]
- [16].Hansson O, Janelidze S, Hall S, Magdalinou N, Lees AJ, Andreasson U, Norgren N, Linder J, Forsgren L, Constantinescu R, Zetterberg H, Blennow K, F. s. Swedish Bio, Blood-based NfL: a biomarker for differential diagnosis of parkinsonian disorder, Neurology 88 (10) (2017) 930–937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Lerche S, Wurster I, Roben B, Zimmermann M, Machetanz G, Wiethoff S, Dehnert M, Rietschel L, Riebenbauer B, Deuschle C, Stransky E, Lieplt-Scarfone I, Gasser T, Brockmann K, CSF NFL in a longitudinally assessed PD cohort: age effects and cognitive trajectories, Mov. Disord 35 (7) (2020) 1138–1144. [DOI] [PubMed] [Google Scholar]
- [18].Mollenhauer B, Dakna M, Kruse N, Galasko D, Foroud T, Zetterberg H, Schade S, Gera RG, Wang W, Gao F, Frasier M, Chahine LM, Coffey CS, Singleton AB, Simuni T, Weintraub D, Seibyl J, Toga AW, Tanner CM, Kieburtz K, Marek K, Siderowf A, Cedarbaum JM, Hutten SJ, Trenkwalder C, Graham D, Validation of Serum Neurofilament Light Chain as a Biomarker of Parkinson’s Disease Progression, Mov Disord, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Burn DJ, Rowan EN, Allan LM, Molloy S, O’Brien JT, McKeith IG, Motor subtype and cognitive decline in Parkinson’s disease, Parkinson’s disease with dementia, and dementia with Lewy bodies, J. Neurol. Neurosurg. Psychiatry 77 (5) (2006) 585–589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Blaszczyk JW, Orawiec R, Duda-Klodowska D, Opala G, Assessment of postural instability in patients with Parkinson’s disease, Exp. Brain Res 183 (1) (2007) 107–114. [DOI] [PubMed] [Google Scholar]
- [21].Bacioglu M, Maia LF, Preische O, Schelle J, Apel A, Kaeser SA, Schweighauser M, Eninger T, Lambert M, Pilotto A, Shimshek DR, Neumann U, Kahle PJ, Staufenbiel M, Neumann M, Maetzler W, Kuhle J, Jucker M, Neurofilament light chain in blood and CSF as marker of disease progression in mouse models and in neurodegenerative diseases, Neuron 91 (1) (2016) 56–66. [DOI] [PubMed] [Google Scholar]
- [22].Pirker W, Correlation of dopamine transporter imaging with parkinsonian motor handicap: how close is it? Mov. Disord 18 (Suppl 7) (2003) S43–S51. [DOI] [PubMed] [Google Scholar]
- [23].Rossi C, Frosini D, Volterrani D, De Feo P, Unti E, Nicoletti V, Kiferle L, Bonuccelli U, Ceravolo R, Differences in nigro-striatal impairment in clinical variants of early Parkinson’s disease: evidence from a FP-CIT SPECT study, Eur. J. Neurol 17 (4) (2010) 626–630. [DOI] [PubMed] [Google Scholar]
- [24].Li W, Lao-Kaim NP, Roussakis AA, Martin-Bastida A, Valle-Guzman N, Paul G, Loane C, Widner H, Politis M, Foltynie T, Barker RA, Piccini P, 11) C-PE2I and (18) F-Dopa PET for assessing progression rate in Parkinson’s: a longitudinal study, Mov. Disord 33 (1) (2018) 117–127. [DOI] [PubMed] [Google Scholar]
- [25].Locascio JJ, Atri A, An overview of longitudinal data analysis methods for neurological research, Dement Geriatr Cogn Dis Extra 1 (1) (2011) 330–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Lin YS, Lee WJ, Wang SJ, Fuh JL, Levels of plasma neurofilament light chain and cognitive function in patients with Alzheimer or Parkinson disease, Sci. Rep 8 (1) (2018) 17368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Holden SK, Finseth T, Sillau SH, Berman BD, Progression of MDS-UPDRS scores over five years in de novo Parkinson disease from the Parkinson’s progression Markers initiative cohort, Mov Disord Clin Pract 5 (1) (2018) 47–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.