LAUSR

The search for a biomarker or panel of biomarkers for Parkinson’s disease (PD) has become a Holy Grail that is crucial to the future direction of the early detection of the illness and assessing its progression, if disease modification or neuroprotective strategies are to be developed. The search has engaged the sophisticated technologies of genomics, epigenomics, proteomics, and metabolomics just as with other neurodegenerative diseases. Metabolomics has been applied to studies of serum, saliva, urine, and cerebrospinal fluid (CSF) in PD, and a range of metabolic products have been put forward as potential biomarkers based on the changes observed. Comparison of these studies is complicated by the stage of the disease studied, nature of the patient population, extent and type of drug treatment, biological fluid studied, and precise analytical technique used. It is probably fair to say that, so far, the outcomes have not been consistent and that the overlap between values for PD patients and controls has been marked. Moving closer to the affected areas of the brain by studying CSF has been one approach to refining the search. In this edition of Movement Disorders, Trezzi and colleagues (INSERT REF) report on a nontargeted, mass spectrometry–based approach applied to the CSF of early untreated patients with PD compared to matched healthy controls. They show that 16 metabolites in CSF are altered in PD—although seven of these remain to be identified. But, perhaps importantly, they identify changes in fructose, mannose, and threonic acid as having a high impact of the differentiation of patients with PD from healthy controls using logistic regression models. Having examined a machine training set of CSF samples, the technique was applied to an independent study cohort and the findings were largely validated. The researchers suggest that the sensitivity and specificity of this metabolomics panel would allow its use as a diagnostic tool for early PD—even in individual patients. The importance of this study lies in the fact that it provides a glimmer of light at the end of the tunnel. The fact that it was carried out in early untreated patients alleviates the confounding factor that drugs bring to such investigations. The identification of a validated metabolite panel provides a tool that can now be probed in greater detail. The limitations of the study raised by the researchers are also a step forward because it provides a basis for further investigation. Certainly, we need independent validation of the findings and bigger sample-size studies. Can this approach be used to look at disease progression (one part of this study suggests that the predictability may be weaker in patients with later-stage disease than in early PD)? Are the findings specific to PD—do the same changes hold sway in related neurodegenerative diseases? For example, a previous study has shown that some metabolomic changes found in PD are also present in amyotrophic lateral sclerosis—although others are different. But this is all food for thought and furthers progress in the field. It is to be expected that many metabolic products will be changed in a complex illness such as PD. A systematic review by Anderson and colleagues highlights the vast array of changes in all types of biomarkers so far found in CSF in PD. The altered metabolome reported to date probably reflects only a small fraction of the metabolite changes that occur, but sieving this information to extract the fingerprint of PD is essential. That is why a uniformity of approach is required. There seems to be a need for a consortium to be established given that the studies based on CSF carried out so far have studied patients at different stages of the illness, at different times of day and using different analytical approaches and probably in individuals with differing dietary habits and food status. There is even evidence of specific changes in metabolites occurring in subgroups of individuals with PD. The majority of these studies point out that a weakness in their investigation of the metabolome in CSF is the small sample size used—and this in particular emphasises the need for a -----------------------------------------------------------*Correspondence to: Prof. Peter Jenner, Neurodegenerative Diseases Research Group, Institute of Pharmaceutical Sciences, Faculty of Health Sciences and Medicine, NDRG, Hodgkin Building, Guy’s Campus, King’s College, London, SE1 1UL, United Kingdom; Email: [email protected]