This review is a part of my essay that I wrote for the course of Neural Stem Cells & Neurodegenerative Diseases that I took during my Ph.D. program. I think it might be helpful for people working in this area in the sense that It gives information about collection of recent studies about biomarkers in Parkinson’s Disease.
Many studies have been performed to investigate robust biomarkers to be used in diagnosing Parkinson’s Disease (PD) and PD Dementia (PDD) in early stages. Different types of biomarkers based on neuroimaging, genetic, clinical and blood test results have been investigated. Exploring these biomarkers from different perspectives may give valuable information to see complex picture of the PDD.
Neuropsychological Test Results
According to a clinical study (Anita, P. et al), PDD patients had lower scores than cognitively normal PD patients did in tests evaluating word memory, attention, psychomotor speed, visuospatial skills and executive functions. In addition, cognitively normal PD patients and healthy controls performed equally well in tests evaluating working memory, attention and executive functions. They suggest that impairments in both PD and PDD exist however different patterns of cognitive impairment may serve as biomarkers in predicting onset of dementia in PD.
Blood Test Results
In another study (Lin, Y.S. et al), blood-based biomarker about cognitive decline in PD and Alzheimer’s Disease (AD) was investigated. Neuroflament light chain (NFL) has been presented as a blood-based biomarker. NFL is a subunit of neurofilaments and they are cytoskeletal proteins of neurons. NFL also appears abundantly in myelinated axons. In case of axonal damage, NFLs are released to cerebrospinal fluid (CSF) and then they are delivered to blood. Higher NFL level in the blood means higher axonal degeneration occurs. They observed that plasma NFL was significantly higher in PDD and AD groups than in healthy control, MCI and non-demented PD groups. They also find a correlation between plasma NFL and poor cognition however they didn’t observe a correlation between motor symptoms and plasma NFL.
CSF Test Results
In another study (Halbgebauer, S.), it is proposed that modified serpinA1 in CSF can be a biomarker to differentiate cognitively impaired PD from cognitively intact PD. SerpinA1 is a type of enzyme inhibitor and has a duty of protecting tissues from enzymes of inflammatory cells. Unfortunately, role of serpinA1 in dementia development is not known. However, it is known that this protein can polymerize and form aggregates (Yamasaki, M.). SerpinA1 charge isoforms in CSF were analyzed with capillary isoelectric focusing immunoassay (CIEF-immunoassay). In this study, PDD and cognitively intact groups can be differentiated whereas MCI and cognitively normal groups are not differentiated as much as the first case.
Neuropsychiatric Inventory (NPI) scores and α-synuclein protein levels
In the next study ( Bougea, A.), groups of AD, PDD and Dementia with Lewy Bodies (DLB) tried to be differentiated by using Neuropsychiatric Inventory (NPI) scores and α-synuclein protein levels in CSF. NPI includes test evaluating the status of delusions, hallucinations, agitation, depression, anxiety, euphoria, apathy, disinhibition, irritation, aberrant motor behavior, night behavior and eating-appetite behavior. Higher scores have a meaning that neuropsychiatric problems are more frequent. PDD and DLB groups showed higher scores in NPI hallucinations compared to AD group. DLB group showed higher scores in NPI agitation compared to PDD group. DLB group showed higher scores in NPI sleep compared to AD group. Lastly DLB group showed higher total NPI scores compared to AD and PDD groups. They also concluded that there is no correlation between α-synuclein protein levels in CSF and NPI scores meaning that both biomarkers predict disease status independently.
Magnetic Resonance Imaging Biomarkers
In another study (Darrell T.H. Li), it is suggested that iron deposition in limbic structures can be a biomarker for cognitive impairment in PDD. By the help of Quantitative Susceptibility Mapping (QSM) MRI, iron levels in different brain regions of Healthy Controls (HC), PD cognitively normal and PDD groups were imaged. Region of Interest (ROI) based analysis revealed that PDD group have higher magnetic susceptibility in left hippocampus compared to cognitively normal PD group. In addition, PDD group have higher magnetic susceptibility in left and right hippocampus compared to HC. Lastly, non-demented PD group have higher magnetic susceptibility in right hippocampus and right thalamus. Despite this study reveals that there exists a correlation between iron deposition in limbic structures and cognitive decline, it remains unexplained that how excessive iron deposition cause cognitive dysfunction.
In addition to QSM MRI technique, MR Spectroscopic Imaging (MRSI) technique has been also used in many studies. By the help of MRSI technique, metabolic changes occurring in substantia nigra and putamen had been examined. Unfortunately, getting clean information from the substantia nigra with MRSI is a compelling process because of small size, iron content and location of the region (Pyatigorskaya, N.). Even though some groups claimed that metabolic alterations exist in substantia nigra of PD patients by using 3T MR system (Groger, A., Hattingen, E.), these findings could not be verified by using neither 4T (Oz, G.) nor 7T (Emir, U. E.) MR system. In addition, some other studies investigating metabolic changes in substantia nigra of PD and HC, reported conflicting results about NAA/Cr (Nie, K., Levin, B. E., Groger, A.). For this reason, examining the metabolites in other regions such as putamen may be an option. Substantia nigra sends dense dopaminergic afferent projections to this structure thus investigating putamen may give clues about metabolic changes of PD (Rudkin, T. M.). However, contradictions still exist among results of different studies investigating putamen. Decreased NAA/Cr ratio is observed in PD group relative to HC in one study (Abe, K.), while other studies suggested that the ratio remains unchanged (Watanabe, H., Davie, C. A.).
Concluding Remarks and Future Perspectives
Many researchers tried to find different types of biomarkers using different tools. Some of these biomarkers are informative in classifying HC, cognitively normal PD and PDD whereas, some of them give valuable information about the discrimination of AD and DLB from PDD. Because AD, DLB and PDD have many common clinical symptoms, consulting reliable biomarkers enhances correct diagnosis hence suitable treatment.
Despite the biomarkers that were found are able to discriminate PDD from the other groups, these are not capable of clustering individuals having MCI condition. Because MCI condition is a major risk factor for evolving to dementia, identifying individuals having this condition is very crucial in slowing down the progression of the disease by providing suitable treatment.
If the biomarkers discussed in this review can be used together with multivariate statistical models or machine learning models (see Outstanding Questions), diagnosing MCI status correctly in high sensitivity may be possible.
PDD is a complex heterogenous disease because of different pathologies involved such as α-synuclein, amyloid-β plaques and tau containing neurofibrillary tangles. In addition, as a second component, mutations in different genes such as SNCA, GBA, APOE ε4, MAPT, COMT and serpinA1 make the pathology more complex. Moreover, as a third component, individuals having different cognitive reserve, education and age are influential on course of the disease.
Exploring quantitative biomarkers by utilizing different types of measurements enables us to understand the complex picture drawn by these interactions of the above components.
Outstanding Questions
Can a mathematical model predict PDD on early stage by using all the mentioned biomarkers together?
Can mentioned biomarkers be also informative about MCI status?
Can these biomarkers provide intuition about treatment of PDD?
