Diagnosis of Parkinson's disease on the basis of clinical and genetic classification: a population-based modelling study

doi:10.1016/S1474-4422(15)00178-7

The Lancet Neurology

Volume 14, Issue 10, October 2015, Pages 1002-1009

, , , , , , , , , , , , , , , , , , , …

https://doi.org/10.1016/S1474-4422(15)00178-7 Get rights and content

Summary

Background

Accurate diagnosis and early detection of complex diseases, such as Parkinson's disease, has the potential to be of great benefit for researchers and clinical practice. We aimed to create a non-invasive, accurate classification model for the diagnosis of Parkinson's disease, which could serve as a basis for future disease prediction studies in longitudinal cohorts.

Methods

We developed a model for disease classification using data from the Parkinson's Progression Marker Initiative (PPMI) study for 367 patients with Parkinson's disease and phenotypically typical imaging data and 165 controls without neurological disease. Olfactory function, genetic risk, family history of Parkinson's disease, age, and gender were algorithmically selected by stepwise logistic regression as significant contributors to our classifying model. We then tested the model with data from 825 patients with Parkinson's disease and 261 controls from five independent cohorts with varying recruitment strategies and designs: the Parkinson's Disease Biomarkers Program (PDBP), the Parkinson's Associated Risk Study (PARS), 23andMe, the Longitudinal and Biomarker Study in PD (LABS-PD), and the Morris K Udall Parkinson's Disease Research Center of Excellence cohort (Penn-Udall). Additionally, we used our model to investigate patients who had imaging scans without evidence of dopaminergic deficit (SWEDD).

Findings

In the population from PPMI, our initial model correctly distinguished patients with Parkinson's disease from controls at an area under the curve (AUC) of 0·923 (95% CI 0·900–0·946) with high sensitivity (0·834, 95% CI 0·711–0·883) and specificity (0·903, 95% CI 0·824–0·946) at its optimum AUC threshold (0·655). All Hosmer-Lemeshow simulations suggested that when parsed into random subgroups, the subgroup data matched that of the overall cohort. External validation showed good classification of Parkinson's disease, with AUCs of 0·894 (95% CI 0·867–0·921) in the PDBP cohort, 0·998 (0·992–1·000) in PARS, 0·955 (no 95% CI available) in 23andMe, 0·929 (0·896–0·962) in LABS-PD, and 0·939 (0·891–0·986) in the Penn-Udall cohort. Four of 17 SWEDD participants who our model classified as having Parkinson's disease converted to Parkinson's disease within 1 year, whereas only one of 38 SWEDD participants who were not classified as having Parkinson's disease underwent conversion (test of proportions, p=0·003).

Interpretation

Our model provides a potential new approach to distinguish participants with Parkinson's disease from controls. If the model can also identify individuals with prodromal or preclinical Parkinson's disease in prospective cohorts, it could facilitate identification of biomarkers and interventions.

Funding

National Institute on Aging, National Institute of Neurological Disorders and Stroke, and the Michael J Fox Foundation.

Introduction

Accurate diagnosis or prediction of risk by use of simple, non-invasive measures is a rarely realised goal for many complex diseases. For complex progressive diseases such as Parkinson's disease, preclinical diagnosis and low error rates in diagnosis are crucial in clinical trials and the study of disease-altering therapeutic approaches.

Imaging is often deemed the gold standard for identification of typical Parkinson's disease pre-mortem, however, high cost and restricted portability limit the use of this approach. We aimed to develop a portable method to identify patients with Parkinson's disease who show aetiologically typical disease presentation (confirmed by dopamine transporter [DAT] imaging data). We used a combination of factors that vary over the life of an individual, factors that are constant and do not change with time, general indicators of neurodegeneration, and Parkinson's disease-specific measures to create our classification algorithm.

Section snippets

Study design and participants

Figure 1 shows a summary of our workflow. Table 1 describes the cohorts we used and further details are available in the appendix: the Parkinson's Progression Marker Initiative (PPMI), the Parkinson's Disease Biomarkers Program (PDBP), the Parkinson's Associated Risk Study (PARS), 23andMe, the Longitudinal and Biomarker Study in PD (LABS-PD), and the Morris K Udall Parkinson's Disease Research Center of Excellence cohort (Penn-Udall). PPMI and PDBP are case-control studies that use a shared set

Results

To accompany this report, and to help with replication and extension of our work, the code and training data for this predictive model and some validation data have been made publicly available online.

Each of five factors that we included made significant contributions to the information content of the integrative predictive model. In comparisons of the standardised beta coefficients within the regression model, the UPSIT score was responsible for 63·1% of the explained variance, followed by

Discussion

We have designed an accurate, non-invasive method to discriminate patients with Parkinson's disease from controls. The studies that we assessed vary in their design, recruitment, and implementation; however, our results and validation suggest that the model might be useful in future. The model we developed includes hyposmia, which is often considered an indicator of neurodegeneration, in addition to genetic, clinical, and demographic data.¹ This approach makes use of the growing wealth of data

References (20)

DS Goldstein et al.
Olfactory dysfunction in pure autonomic failure: implications for the pathogenesis of Lewy body diseases
Parkinsonism Relat Disord
(2009)
RL Doty et al.
Presence of both odor identification and detection deficits in Alzheimer's disease
Brain Res Bull
(1987)
C Paisán-Ruíz et al.
Cloning of the gene containing mutations that cause PARK8-linked Parkinson's disease
Neuron
(2004)
A Zimprich et al.
Mutations in LRRK2 cause autosomal-dominant parkinsonism with pleomorphic pathology
Neuron
(2004)
N Sobel et al.
An impairment in sniffing contributes to the olfactory impairment in Parkinson's disease
Proc Natl Acad Sci USA
(2001)
CH Adler et al.
Olfactory function in restless legs syndrome
Mov Disord
(1998)
A Siderowf et al.
Risk factors for Parkinson's disease and impaired olfaction in relatives of patients with Parkinson's disease
Mov Disord
(2007)
S Li et al.
Variation in Variables that Predict Progression from MCI to AD Dementia over Duration of Follow-up
Am J Alzheimer's Dis
(2013)
AW Michell et al.
Biomarkers and Parkinson's disease
Brain
(2004)
F Lazarini et al.
Adult neurogenesis restores dopaminergic neuronal loss in the olfactory bulb
J Neurosci
(2014)

There are more references available in the full text version of this article.

Cited by (163)

Cerebrospinal fluid biomarkers profile in scans without evidence of dopaminergic deficits (SWEDD)
2023, IBRO Neuroscience Reports
A small proportion of patients with clinical parkinsonism have normal transporter-single photon emission computed tomography (DaTSPECT) which is commonly defined as scans without evidence of dopaminergic deficits (SWEDD). A better understanding of SWEDD can improve the current therapeutic options and appropriate disease monitoring.
We aimed to assess CSF biomarkers levels including α-synuclein (α-syn), amyloid βeta (Aβ1–42), total tau (t-tau), and phosphorylated tau (p-tau) in SWEDD and investigate the longitudinal alteration in the CSF profile.
In total, 406 early-stage PD, 58 SWEDD, and 187 healthy controls (HCs) were entered into our study from PPMI. We compared the level of CSF biomarkers at baseline, six months, one year, and two years. Furthermore, the longitudinal alteration of CSF biomarkers was explored in each group using linear mixed models.
There was no significant difference in the level of CSF α-syn Aβ1–42, t-tau, and p-tau between HCs and SWEDD at different time points. Investigating the level of CSF α-syn in PD and SWEDD showed a significant difference at one (p = 0.016) and two years (p = 0.006). Also, we observed a significant difference in the level of CSF Aβ1–42 between SWEDD and PD at one year (p = 0.012). Moreover, there was a significant difference in the level of CSF t-tau between SWEDD and PD subjects at one (p = 0.013) and two years (p = 0.017). Furthermore, there was a significant difference in the level of CSF p-tau between SWEDD and PD groups at two years visits (p = 0.030). Longitudinal analysis showed a significant decrease after one (p = 0.029) and two years (p = 0.002) from baseline in the level of CSF α-syn only in the PD group. Also, we observed that the level of CSF Aβ1–42 significantly increased after one year in SWEDD (p = 0.031) and decreased after two years from baseline in PD subjects (p = 0.005). Moreover, there was a significant increase in the level of CSF t-tau after two years (p = 0.036) and CSF p-tau after six months from baseline in SWEDD subjects (p = 0.011).
This finding suggests a faster neurodegeneration process in PD patients compared to SWEDD at least based on these biomarkers. Future studies with longer follow-up duration and more sample sizes are necessary to validate our results.
What have we learned from genome-wide association studies (GWAS) in Parkinson's disease?
2022, Ageing Research Reviews
After fifteen years of genome-wide association studies (GWAS) in Parkinson's disease (PD), what have we learned? Addressing this question will help catalogue the progress made towards elucidating disease mechanisms, improving the clinical utility of the identified loci, and envisioning how we can harness the strides to develop translational GWAS strategies. Here we review the advances of PD GWAS made to date while critically addressing the challenges and opportunities for next-generation GWAS. Thus, deciphering the missing heritability in underrepresented populations is currently at the reach of hand for a truly comprehensive understanding of the genetics of PD across the different ethnicities. Moreover, state-of-the-art GWAS designs hold a true potential for enhancing the clinical applicability of genetic findings, for instance, by improving disease prediction (PD risk and progression). Lastly, advanced PD GWAS findings, alone or in combination with clinical and environmental parameters, are expected to have the capacity for defining patient enriched cohorts stratified by genetic risk profiles and readily available for neuroprotective clinical trials. Overall, envisioning future strategies for advanced GWAS is currently timely and can be instrumental in providing novel genetic readouts essential for a true clinical translatability of PD genetic findings.
Immune-related biomarkers for Parkinson's disease
2022, Neurobiology of Disease
Despite the increasing number of studies on Parkinson's disease and it being the second most common neurodegenerative disorder in the world, no established diagnostic markers or disease modifying therapies are available. Understanding the mechanisms involved in its pathogenesis and identifying markers capable of diagnosing or tracking progression of PD is greatly needed. Among the several factors identified to be involved in Parkinson's disease, the immune system has had increasingly growing evidence that presents a fresh avenue to investigate the pathology of the disease. The involvement of the immune system in the pathology of Parkinson's disease has been linked to an interaction between the peripheral and central nervous system immune response. Whether this involvement is due to an immune response being a cause or consequence of Parkinson's disease pathology is still a matter of debate. Players investigated include cytokines, chemokines, and immune-cells found in both the central and peripheral immune system. Herein, we discuss advances in the current literature on these immune-related markers and their potential use as markers for Parkinson's disease diagnosis and progression.
Microglial inhibition alleviates alpha-synuclein propagation and neurodegeneration in Parkinson’s disease mouse model
2024, npj Parkinson's Disease
Moving towards the detection of frailty with biomarkers: A population health study
2024, Aging Cell
Genetic Associations and Functional Pathways in Parkinson's Disease: A Meta-Analytic Insight into LRRK2 Mutation Impact
2024, Research Square

View all citing articles on Scopus

View full text

ArticlesDiagnosis of Parkinson's disease on the basis of clinical and genetic classification: a population-based modelling study

Summary

Background

Methods

Findings

Interpretation

Funding

Introduction

Section snippets

Study design and participants

Results

Discussion

Parkinsonism Relat Disord

Brain Res Bull

Neuron

Neuron

An impairment in sniffing contributes to the olfactory impairment in Parkinson's disease

Proc Natl Acad Sci USA

Olfactory function in restless legs syndrome

Mov Disord

Risk factors for Parkinson's disease and impaired olfaction in relatives of patients with Parkinson's disease

Mov Disord

Variation in Variables that Predict Progression from MCI to AD Dementia over Duration of Follow-up

Am J Alzheimer's Dis

Biomarkers and Parkinson's disease

Brain

Adult neurogenesis restores dopaminergic neuronal loss in the olfactory bulb

J Neurosci

Articles
Diagnosis of Parkinson's disease on the basis of clinical and genetic classification: a population-based modelling study