Article Text

Download PDFPDF

Research paper
Subtypes of mild cognitive impairment in patients with Parkinson's disease: evidence from the LANDSCAPE study
  1. Elke Kalbe1,
  2. Sarah Petra Rehberg1,
  3. Ines Heber2,
  4. Martin Kronenbuerger2,
  5. Jörg B Schulz2,3,
  6. Alexander Storch4,5,
  7. Katharina Linse4,
  8. Christine Schneider4,
  9. Susanne Gräber6,
  10. Inga Liepelt-Scarfone6,
  11. Daniela Berg6,7,
  12. Judith Dams8,
  13. Monika Balzer-Geldsetzer8,
  14. Rüdiger Hilker9,
  15. Carola Oberschmidt9,
  16. Karsten Witt7,
  17. Nele Schmidt7,
  18. Brit Mollenhauer10,
  19. Claudia Trenkwalder10,
  20. Annika Spottke11,
  21. Sandra Roeske11,
  22. Hans-Ulrich Wittchen12,
  23. Oliver Riedel13,
  24. Richard Dodel8
  1. 1Medical Psychology, Neuropsychology and Gender Studies, Center for Neuropsychological Diagnostics and Intervention (CeNDI), University Hospital Cologne, Cologne, Germany
  2. 2Department of Neurology, University Hospital, RWTH University Aachen, Aachen, Germany
  3. 3JARA Brain Institute 2, RWTH University and Forschungszentrum Jülich, Germany
  4. 4Division of Neurodegenerative Diseases, Department of Neurology, Technische Universität Dresden, Dresden, Germany
  5. 5Department of Neurology, University of Rostock, Rostock, Germany
  6. 6German Center of Neurodegenerative Diseases and Hertie Institute for Clinical Brain Research, Tübingen, Germany
  7. 7Department of Neurology, Christian Albrecht University, Kiel, Germany
  8. 8Department of Neurology, Philipps University Marburg, Marburg, Germany
  9. 9Department of Neurology, J.W. Goethe University, Frankfurt/Main, Germany
  10. 10Paracelsus-Elena Clinic, Centre of Parkinsonism and Movement Disorders, Kassel, Germany
  11. 11Department of Neurology, University Hospital Bonn, and German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany
  12. 12Institute of Clinical Psychology and Psychotherapy, Technische Universität Dresden, Dresden, Germany
  13. 13Department of Clinical Epidemiology, Leibniz Institute for Prevention Research and Epidemiology, Bremen, Germany
  1. Correspondence to Professor Elke Kalbe, Medical Psychology | Neuropsychology and Gender Studies, Center for Neuropsychological Diagnostics and Intervention (CeNDI), University Hospital Cologne, Kerpener Str. 68, Cologne D-50937, Germany; elke.kalbe{at}


Objective Inconsistent results exist regarding the cognitive profile in patients with Parkinson's disease with mild cognitive impairment (PD-MCI). We aimed at providing data on this topic from a large cohort of patients with PD-MCI.

Methods Sociodemographic, clinical and neuropsychological baseline data from patients with PD-MCI recruited in the multicentre, prospective, observational DEMPARK/LANDSCAPE study were analysed.

Results 269 patients with PD-MCI (age 67.8±7.4, Unified Parkinson's Disease Rating Scale (UPDRS-III) scores 23.2±11.6) were included. PD-MCI subtypes were 39.4% non-amnestic single domain, 30.5% amnestic multiple domain, 23.4% non-amnestic multiple domain and 6.7% amnestic single domain. Executive functions were most frequently impaired. The most sensitive tests to detect cognitive dysfunctions were the Modified Card Sorting Test, digit span backwards and word list learning direct recall. Multiple stepwise regression analyses showed that global cognition, gender and age, but not education or disease-related parameters predicted PD-MCI subtypes.

Conclusions This study with the so far largest number of prospectively recruited patients with PD-MCI indicates that non-amnestic PD-MCI is more frequent than amnestic PD-MCI; executive dysfunctions are the most typical cognitive symptom in PD-MCI; and age, gender and global cognition predict the PD-MCI subtype. Longitudinal data are needed to test the hypothesis that patients with PD-MCI with specific cognitive profiles have different risks to develop dementia.


Statistics from


About one-fourth of the patients with Parkinson’s disease (PD) suffer from mild cognitive impairment (MCI),1 and even higher frequency rates (39.4%) have been reported in patients with more advanced disease severity.2 PD-MCI is of utmost clinical relevance, as it limits the patients' quality of life3 and has been identified as a predictor for the conversion to PD dementia (PDD).4 Furthermore, its understanding may foster the knowledge on underlying pathomechanisms and development of interventions.

Data suggest that patients with PD-MCI have a specific cognitive profile that differs from that of patients with MCI due to Alzheimer's disease (AD). While in the latter group memory deficits are most typical,5 reviews indicate that executive dysfunctions—ascribed to dopaminergic dysfunction in the ‘cognitive’ frontostriatal circuit—may be the most frequent symptom in PD-MCI,6 although conflicting data exist.2 ,7–11 More generally, studies indicate that a higher proportion of patients with PD-MCI suffer from non-amnestic than from amnestic cognitive deficits,2 ,7 ,12 ,13 albeit some studies find amnestic dysfunction to be more prominent.8 Finally, it is still unclear whether patients with PD-MCI more often suffer from single-domain2 ,7 ,9 ,13 or multiple-domain impairment.12 ,14 Remarkably, most prospective studies so far included rather small study samples, making further studies with large prospective study cohorts necessary.

One source of heterogeneity regarding frequencies and specific characteristics of PD-MCI subtypes is the application of different criteria for the definition of PD-MCI as well as their operationalisation15 including the selection of neuropsychological tests. Notably, the recently published criteria of the Movement Disorder Society (MDS) for PD-MCI16 suggest a large number of tests which vary greatly among cognitive domains and which may additionally vary substantially in their sensitivity to detect cognitive impairment. Information on which tests are most sensitive in the diagnosis of PD-MCI would be of great clinical use, but it has rarely been addressed.

Data indicate that different cognitive patterns in patients with PD may also reflect different neuropathological subtypes of PD that are in turn associated with specific motor symptoms and phenotypes.8 ,12 ,13 Importantly, these patterns may be related differentially to the risk to develop PDD.4 ,17 Therefore, more detailed knowledge from large-scale studies about clinical and sociodemographic correlates of PD subtypes which are easily available in clinical practice is needed and may help identify patients at risk for unfavourable disease prognosis and PDD.

Thus, the aim of this paper is to report on the cognitive profile of patients with PD-MCI using data from a large prospective study on cognition in patients with PD: the DEMPARK/LANDSCAPE study.18 More specifically, we aim at providing the relative prevalence and neuropsychological characteristics of PD-MCI subtypes as well as defining neuropsychological tests which are most sensitive to detect cognitive impairment in PD-MCI. Additionally, we aim at identifying variables predicting PD-MCI subtypes.



The DEMPARK/LANDSCAPE project (for details, see ref. 18) is a multicentre, prospective, observational cohort study, including patients with PD without cognitive impairment, with PD-MCI, and PDD, as well as patients with Lewy body dementia in Germany. For this substudy, baseline data of all patients with PD-MCI were included. Cognitive assessment was performed during patients’ ‘on’ medication state.

Clinical and neuropsychological assessment

All scales with abbreviations are displayed in table 1; for details and references, please see ref. 18. Demographics, medical history and levodopa-equivalent daily dose (LEDD) were documented. Disease severity was rated using the Hoehn and Yahr (H&Y) scale. Furthermore, parts I, III and IV of the Unified Parkinson’s Disease Rating Scale (UPDRS) were used. Depression, apathy, quality of life and health status were evaluated using the Geriatric Depression Scale (GDS-15), Apathy Evaluation Scale (AES), Parkinson's Disease Questionnaire (PDQ-39) and EuroQol (EQ-5D), respectively. All patients were cognitively screened with the Mini Mental State Examination (MMSE) and the Parkinson Neuropsychometric Dementia Assessment (PANDA) and received an elaborate neuropsychological test battery covering tests to examine memory, executive functions, attention, visuospatial functions and language. Verbal short-term memory and long-term memory were assessed with the subtests ‘wordlist learning’ and ‘wordlist recall’ of the German version of the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD-Plus) test battery, respectively. Executive functions were evaluated using the Modified Card Sorting Test (MCST; subtests ‘categories’, ‘non-perseverative errors’ and ‘perseverative errors’) and the subtests ‘semantic word fluency’ and ‘formal-lexical word fluency’ of the CERAD-Plus test battery. Attention was tested with the Brief Test of Attention (BTA) and the Stroop Color and Word Test (subtests ‘color naming’, ‘word reading’ and ‘interference’). Visuospatial functions were examined using subtests 7 (mental rotation) and 9 (spatial imagination) of the German test battery ‘Leistungsprüfungssystem’ (LPS). Language was investigated with the Boston Naming Test of the CERAD-Plus test battery.

Table 1

Demographic, clinical and neuropsychological characteristics of the PD-MCI total sample and PD-MCI subtypes

Definition of PD-MCI

Diagnosis of PD-MCI and their subtypes was made according to the established MCI criteria that were available at the time of study set-up19 and was operationalised by (1) cognitive dysfunctions reported by the patient, (2) no significant impairment in activities of daily living according to medical history and (3) cognitive dysfunctions, defined as scores ≤1.5 SD below normative mean values in at least one of the tests used for diagnosis (table 1). With regard to this cut-off score, few exceptions (n=18) were made according to expert's ratings if clinicians found that clear cognitive impairment was evident despite performance above this cut-off score (e.g. in highly educated individuals) or if a performance below this cut-off score was still evaluated as ‘within normal range’ of this specific patient. Patients were categorised into one of the four PD-MCI subtypes (amnestic or non-amnestic (aMCI, naMCI) with single or multiple domains (sd, md) affected) depending on their cognitive profile with one test affected being sufficient for a domain to be regarded as dysfunctional.

Statistical analyses

Statistical analyses were carried out by study-independent statisticians (using SAS 9.4 software). As missing data were <10% in all scales, no values were imputed. Variables were tested for normal distribution and homogeneity with Kolmogorov-Smirnov and Levene's test, respectively. Mean or median values, SDs or ranges, and frequencies for sociodemographic, clinical and neuropsychological variables were calculated. Gender distribution in the total sample and between subtypes was analysed using χ2 tests. Sociodemographic and clinical data were compared between PD-MCI subtypes using Kruskal-Wallis tests; neuropsychological test scores between PD-MCI subtypes were analysed using one-way analysis of (co)variance (AN(C)OVA). Significance level was set at p≤0.05. In case of significance, Tukey's post hoc tests with additional Bonferroni correction were performed. To identify predictors for PD-MCI subtypes, a stepwise backward multinomial logistic regression analysis was performed with PD-MCI subtype as dependent variable (aMCI-md as reference category), and age, gender, education, UPDRS-III, disease duration, LEDD, PANDA and GDS-15 scores as predictors. For PD-MCI subtypes, frequencies of impairment in single domains and combinations of domains were calculated, including only tests used for diagnosis. Frequencies of impairment in all neuropsychological tests were calculated for the total sample.


Overall sample

Out of the 665 patients with PD included in the DEMPARK/LANDSCAPE baseline cohort, data of 654 patients allowed for classification into PD with intact cognition (n=267, 40.2%), PD-MCI (n=292, 43.9%) or PDD (n=97, 14.6%); 9 patients (1.3%) could not be classified into one of these categories. For the analyses of the present study, only data sets of patients which allowed for a classification of a specific PD-MCI subtype were included, resulting in a study sample of 269 (40.5%) patients with PD-MCI. Sociodemographic and clinical characteristics of the patients with PD-MCI are shown in table 1. Mean age was 67.8±7.4. Significantly more men (68.4%) than women were included. Median H&Y stage indicated mild bilateral motor symptoms. Mean MMSE, PANDA, GDS-15 and AES scores were within normal ranges. About two-thirds of the patients with PD-MCI had executive impairment (65.3%), followed by visuospatial (36.3%), memory (33.5%), attention (25.8%) and language impairment (6.5%).

PD-MCI subtypes

Relative prevalence rates, sociodemographic, clinical and neuropsychological characteristics of the PD-MCI subtypes are displayed in table 1. A majority of patients (39.4%) were classified as naMCI-sd subtype, followed by aMCI-md (30.5%), naMCI-md (23.4%) and aMCI-sd (6.7%). All subtypes included more male patients, with a significantly different gender distribution in patients. The proportion of women was significantly higher in the aMCI-md subtype compared to naMCI-md. Global cognition was lowest in patients with aMCI-md and significantly reduced compared with naMCI-sd (MMSE, PANDA) and naMCI-md (PANDA). Concordantly, UPDRS I scores for cognition were significantly worse (higher) in patients with aMCI-md compared with patients with naMCI-sd. Neuropsychological differences between PD-MCI subtypes are indicated in table 1.

The naMCI-sd subtype with executive dysfunctions was the most frequent subtype (23.4%), followed by naMCI-sd with visuospatial dysfunction and naMCI-md with combined executive and visuospatial dysfunctions (9.7% and 10%, respectively; figure 1).

Figure 1

Percentage of single and combinations of domain impairment in PD-MCI subtypes. A, attention; aMCI-md, amnestic multiple-domains subtype; aMCI-sd, amnestic single-domain subtype; E, executive functions; L, language; M, memory; naMCI-md, non-amnestic multiple-domains subtype; naMCI-sd, non-amnestic single-domain subtype; other, domains or domain combinations <4%, PD-MCI, Parkinson's disease with mild cognitive impairment; V, visuospatial functions.

PANDA score (χ²=19.39, df=3, p=0.0002), gender (χ²=9.87, df=3, p=0.0197) and age (χ²=7.91, df=3, p=0.048) significantly predicted PD-MCI subtype. Regarding PANDA scores, the aMCI-md subtype differed significantly from naMCI-sd (β=0.1616, SE=0.04, Wald=16.00, p<0.001) and naMCI-md (β=0.1706, SE=0.05, Wald=13.57, p=0.0002). Regarding gender, the aMCI-md subtype differed significantly from naMCI-sd (β=−0.8511, SE=0.41, Wald=4.36, p=0.0368) and naMCI-md (β=−1.564, SE=0.52, Wald=8.99, p=0.0027), indicating female gender as a predictor of aMCI-md. For age, the aMCI-md subtype differed significantly from aMCI-sd (β=−0.098, SE=0.05, Wald=5.58, p=0.0182).

Most sensitive neuropsychological tests

For the identification of the most sensitive tests to assess cognitive dysfunction in PD-MCI, all tests used in the DEMPARK/LANDSCAPE cohort were analysed (figure 2). Overall, the tests most frequently impaired were the MCST (number of categories: 43.5%), the digit span backwards (36.1%) and the word list direct recall (31.2%). Concordantly, within the executive domain, the number of categories of the MCST (43.5%) was the most sensitive indicator for executive dysfunctions, followed by the digit span backwards (36.1%) and non-perseverative errors in the MCST (27.5%). Semantic fluency was impaired more frequently than letter fluency (20.1% vs 13%). To assess memory dysfunctions, word list direct recall was the most efficient (31.2% impairment), followed by the recall of the figures (27.5%) and word list delayed recall (19.7%). The BTA was much more sensitive (20.1% impairment) to assess attention than the Stroop test (6.3%, subtest color word reading). Regarding visuospatial deficits, the frequency of impairment was almost equal for figure copying (21.2%) and mental rotation (20.4%), while spatial imagination was impaired in only 5.2% of the sample.

Figure 2

Percentages of patients with PD-MCI impaired in neuropsychological tests, clustered for the five cognitive domains: memory, executive functions, attention, visuospatial functions and language. A, attention; BTA, Brief Test of Attention; E, executive functions; L, language; LPS, German test battery ‘Leistungsprüfsystem’ (subtests 7 and 9); M, memory; MCST, Modified Card Sorting Test; MMSE-language, Mini Mental Status Examination (items 22, 23, 24, 28, 29); PD-MCI, Parkinson's disease with mild cognitive impairment; TMT-B/A, Trail Making Test—Index: scores of version B divided by scores of version A; V, visuospatial functions. For references of the neuropsychological tests, see ref. 18.


The main findings of this study with the so far largest number of prospectively recruited patients with PD-MCI (n=269) are that (1) executive dysfunctions are the most typical cognitive symptom in PD-MCI, (2) non-amnestic deficits generally occur more frequently than amnestic impairment, (3) multiple-domain PD-MCI is more frequent than single-domain PD-MCI, (4) global cognition, gender and age predict subtypes of PD-MCI and (5) commonly used cognitive tests vary greatly in their sensitivity to detect dysfunctions.

Our finding that executive dysfunctions are the most typical cognitive symptom in PD-MCI is in line with recent reviews and original studies1 ,6 ,9 ,13 and reflects the fact that the ‘cognitive’ frontostriatal loop projecting from the dorsal striatum to the dorsolateral prefrontal cortex, which is related to executive functions, is affected early in patients with PD.17 ,20 Likewise, executive dysfunctions are most typical in newly diagnosed, drug-naïve patients with PD.21 However, contradictory findings exist indicating that memory is the domain most frequently impaired in PD-MCI,2 ,8 ,10 also in drug-naïve patients with PD.7 ,22 Although methodical aspects might be responsible for part of this heterogeneity, the variability of cognitive patterns might additionally point to different subtypes of PD which could be represented in different proportions of PD subtypes in the study cohorts. Notably, it has been suggested that different cognitive profiles in PD-MCI reflect different neuropathological pathways. More specifically, according to the ‘dual syndrome hypothesis’,17 dopaminergic deficits in frontostriatal circuits are related to predominantly executive dysfunction, while widespread Lewy body pathological changes and pronounced cholinergic depletion typically cause memory and more ‘posterior cortical’ (prominently visuospatial) deficits. Importantly, these distinct patterns relate differentially to the risk to develop PDD deficits: while patients with mainly executive dysfunctions typically remain stable, those with posterior dysfunction progress more rapidly to dementia.23 ,24 Interestingly, our patients with the amnestic subtype with additional domains being affected (aMCI-md) showed the lowest performance in global cognitive scores, indicating that they might be ‘closest’ to developing dementia.

Although a recent review concluded that single-domain PD-MCI is more frequent than multiple-domain PD-MCI,1 our result that the multiple-domain subtype occurs more often is in line with some other studies.10 ,12 ,14 These differences may be related to the cohorts’ characteristic (e.g. whether only patients with early PD were included) and the definition of PD-MCI (e.g. whether 1 or 1.5 SDs below normative mean values are used as cut-off scores for impairment25). The fact that our study cohort is very heterogeneous regarding disease duration and motor severity, together with the cut-off score of 1.5 SDs for impairment, may explain the higher frequency of multiple-domain impairment in our study.

Prediction of PD-MCI subtypes has only rarely been studied yet. However, our data are in line with one other study showing that disease duration, LEDD, the UPDRS motor score and mood do not predict PD-MCI subtype.13 Instead, we identified higher age (which was no predictor in the other study13), female gender and lower global cognition (both of which were not tested in that study13) as predictors for being classified as aMCI-md subtype. Predicting aMCI-md with variables that are easily available in clinical routine could be useful, as this subtype scored lowest on global cognition and, as discussed above, may thus be ‘closest’ to dementia.17 Remarkably, the fact that female gender predicts an amnestic PD-MCI subtype fits to recent data from the DEMPARK/LANDSCAPE study including patients with PD, PD-MCI and PDD that memory declines more in women than in men.26

Our data demonstrate that cognitive tests differ substantially in their potential to detect impairment in PD-MCI. The most sensitive tests in our study are recommended by the MDS16 for PD-MCI diagnosis, but notably, those recommendations also include tests that seem less efficient, such as the Stroop test and the letter fluency tests, demonstrating the need for a more specific selection of tests for clinical and scientific purposes. The fact that word list learning was substantially more deteriorated than the delayed recall condition is intriguing and might be based on dysfunctional executive components during learning (while retrieval is less impaired), such as the use of semantic clustering strategies, as has been demonstrated in patients with PD.27 A relevant executive component that is frequently impaired in patients with PD28 which has already been shown to contribute to reduced semantic clustering during learning in (non-PD) patients with MCI is working memory.29 Thus, working memory which was dysfunctional in many of our patients, tested with the digit span backwards task, might have contributed to the learning deficits of our patients. Given the fact that working memory is a very basic function that is relevant for learning and other higher cognitive functions30 and that it is frequently impaired in patients with PD, the use of the digit span task which is very time economic can clearly be recommended.

A limitation of this study is that PD-MCI diagnosis was not based on MDS criteria16 as these were not available at study set-up. Thus, the proportion of patients with PD-MCI-sd for which performance in one test in our but two tests in MDS criteria needs to be deteriorated might be overestimated. One could try to apply the new criteria to the data of our study. However, our aim was to present the data as it was assessed in each participating centre of the study. Furthermore, only one language test was used in our study, so that the ‘two tests per domain’ criterion of the MDS criteria cannot be fulfilled. Additionally, while patients have to report subjective cognitive dysfunctions according to Petersen criteria which we used, this aspect is less strict in MDS criteria in which a relative’s or clinician’s judgement on suspected cognitive decline is sufficient. Finally, we would like to emphasise that, despite the shortcoming of not using the ‘new’ PD-MCI criteria, we are convinced that our study is able to provide valuable data to the field, as data can be compared to other important large studies on PD-MCI which, for example, used mean z scores for each cognitive domain which partly relied on only one test.2 ,7 As a second limitation, statistically, the small group size of the aMCI-sd subtype may have confounded the statistical comparisons between this and the other subtypes, so the findings related to the aMCI-sd subtype should be interpreted with caution.

Strengths of our study are the large sample size, the analysis of the most sensitive tests to assess cognitive dysfunctions in patients with PD-MCI which will be of clinical use and data on the prediction of PD-MCI subtypes which is rare so far.

Longitudinal data from this and other prospective large cohort studies will have to test the hypothesis that different PD-MCI subtypes correspond to distinct neuropathological syndromes which differentially relate to PDD, and whether patients with the aMCI-md subtype are at highest risk to develop PDD.17


The authors wish to thank all patients participating in the study and all staff members in the recruiting centres who contributed to the study. We especially thank the study nurses Daniela Probst (Aachen), Cecile Bosredon (Dresden), Simone Schmidt (Dresden), Gerda Engel (Marburg), as well as the doctoral students Josephine Christ and Hannah Glonnegger (Tübingen) and Annette Petrelli (Vechta/Cologne).



  • EK and SPR contributed equally to this study.

  • Contributors AlS, AnS, BM, CO, CS, CT, DB, IL-S, KL, KW, MK, NS, SG and RH were responsible for the recruitment of patients included in the DEMPARK/LANDSCAPE project. AlS, CS and MK were responsible for the care and assistance of patients included in the DEMPARK/LANDSCAPE project. AlS, AnS, CO, CS, H-UW, IH, IL-S, KL, MK, SG, SR and RHR contributed to the acquisition of data in the DEMPARK/LANDSCAPE project. AlS, BM, CT, EK, H-UW, JBS, KW, MB-G, OR, RD and RH were responsible for the conceptualisation and design of the DEMPARK/LANDSCAPE project. BM, CT, DB, KW, MB-G, RD and RH were responsible for the funding, supervision and coordination of the DEMPARK/LANDSCAPE project. EK, H-UW and SPR were responsible for the concept and design of this specific subproject. JD was responsible for the statistical analyses described in the article. EK, JD, H-UW and SPR contributed to the interpretation of the data presented in the article. EK and SPR wrote the final version of the article. All authors contributed to the revision of the article.

  • Funding The DEMPARK study was funded by an unrestricted grant from Novartis and a grant from the International Parkinson Fonds (Deutschland) GmbH (IPD). The continuation of the study (LANDSCAPE) is part of the Competence Network Degenerative Dementias and is funded by the German Ministry for Education and Research (BMBF; project number 01GI1008C).

  • Competing interests EK reports grants from International Parkinson Fonds (Deutschland) GmbH (IPD), grants from Novartis and grants from German Federal Ministry for Education and Research (BMBF) during the conduct of the study, as well as grants from International Parkinson Fonds (Deutschland) GbmH, personal fees from Abbvie GmbH Deutschland, personal fees from Novartis Pharma GbmH Deutschland and from German Federal Ministry of Education and Research outside the submitted work. SPR reports grants from German Federal Ministry for Education and Research during the conduct of the study. IH reports grants from German Federal Ministry for Education and Research during the conduct of the study. AlS reports grants from the Bundesministerium für Wirtschaft und Technologie (Federal Ministry for Economy and Technology), the Deutsche Forschungsgemeinschaft (German Research Association), the Helmholtz-Association, the NCL Foundation and the Novartis Foundation during the conduct of the study. He has received unrestricted research grants from TEVA Pharma and Global Kinetics Cooperation (GKC, Melbourne, Australia), and honoraria for presentations/advisory boards/consultations from Desitin, Abbvie, GKC, Mundipharma, Zambon, UCB, Britannia Ltd. UK, Pfizer Ltd. UK, Teva, Meda, GSK, Medtronic, Messe Karlsruhe, Neuro-Depesche, Lund University and Volkswagen Foundation, and royalties from Kohlhammer Verlag and Elsevier Press outside the submitted work. He serves as an editorial board member of Stem Cells, Stem Cells International, Open Biotechnology Journal, Frontiers in Ageing Neuroscience and JBC, The Journal of Biological Chemistry. KL reports grants from German Federal Ministry for Education and Research, and grants from Novartis during the conduct of the study, as well as grants from German Ministry for Education and Research (grant no. 16SV5843) outside the submitted work. IL-S reports grants from International Parkinson Fonds (Deutschland) GmbH (IPD), grants from Novartis and grants from German Federal Ministry for Education and Research during the conduct of the study, as well as grants from Johnson & Johnson and grants from European Commission (grant no. H2020-TWINN-2015) outside the submitted work. DB reports grants from Janssen Pharmaceutica, grants and personal fees from TEVA Pharma GmbH, grants and personal fees from UCB Pharma GmbH, grants from Michael J. Fox Foundation, grants from ParkinsonFonds Deutschland gGmbH, grants from German Parkinson's Disease Association (dPV), grants from German Federal Ministry for Education and Research, grants from BMWi, grants from EU, grants and personal fees from Novartis Pharma GmbH, grants and personal fees from Lundbeck and grants from Boehringer Ingelheim Pharma GmbH outside the submitted work. RD reports personal fees and non-financial support from Abbott/Abbvie, personal fees and non-financial support from Academy of Finland, grants from Baxter, non-financial support from Betreuungsverein Biedenkopf, grants from German Federal Ministry for Education and Research, non-financial support from British Society of Immunology, personal fees from Contingo Consulting, grants and non-financial support from Deutsche Gesellschaft für Neurologie, non-financial support from Deutsche Parkinson Gesellschaft, grants from Deutsche Parkinson Vereinigung, personal fees from DZNE Deutsches Zentrum für neurodegenerative Erkrankungen, grants from Faber-Stiftung, grants from Hector-Stiftung, grants from International Parkinson Fonds, non-financial support from KNDD Kompetenznetz Degenerative Demenzen, personal fees from Lilly, non-financial support from Movement Disorder Society, personal fees from Novartis, personal fees from Thieme Verlag, grants from UKGM, non-financial support from Universität Duisburg-Essen, personal fees and non-financial support from Vitos Klinik Bad Emstal, personal fees from Med Update, personal fees from GroupH, personal fees from Piramal, personal fees from Studienstiftung des deutschen Volkes, personal fees from National Board of Health and Healthcare Schweden, grants from EU Horizon 2020, personal fees from Axon Neuroscience, non-financial support from Universitätsklinikum Tübingen, non-financial support from Forschungszentrum Jülich, non-financial support from European Academy of Neurology (EAN), non-financial support from Universität Regensburg, grants from AOK Gesundheitskasse Hessen and grants from AOK Gesundheitskasse Sachsen und Thüringen, outside the submitted work. KW reports grants from German Federal Ministry of Education and Research during the conduct of the study. NS reports grants from German Federal Ministry of Education and Research during the conduct of the study. BM reports grants from TEVA Pharma, grants from Desitin, grants from Boehringer Ingelheim, grants from GE Healthcare, personal fees from Bayer Schering Pharma AG, personal fees from Roche, personal fees from AbbVie, personal fees from TEVA-Pharma, personal fees from Biogen, personal fees from GlaxoSmithKline, personal fees from Orion Pharma, grants from BMBF, EU, Dt. Parkinsonvereinigung, Michael J. Fox Foundation for Parkinson's Research outside the submitted work. CT reports personal fees from Paracelsus-Elena-Klinik, Kassel, and personal fees from Medical University of Goettingen outside the submitted work. AnS11 reports grants from German Federal Ministry for Education and Research (BMBF; grant no. 01GI1008C), grants from Novartis and grants from International Parkinson Fonds (Deutschland) GmbH (IPD) during the conduct of the study.

  • Patient consent Obtained.

  • Ethics approval The DEMPARK/LANDSCAPE study was approved by the Ethics Committee of Philipps University Marburg (DEMPARK: approval no. 178/07 26 March 2009; LANDSCAPE: approval no. 25/11 18 October 2011) as well as the local ethics committees of all participating centres.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.