Parkinsonism and Related Disorders 20S1 (2014) S76–S79
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Quantification of a-synuclein in cerebrospinal fluid: How ideal is this biomarker for Parkinson’s disease? Brit Mollenhauer a,b, * a Paracelsus-Elena-Klinik,
Center of Parkinsonism and Movement Disorders, Kassel, and b University Medical Center, Department of Neurosurgery and Neuropathology, Germany
article info
summary
Keywords: Parkinson’s disease a-Synuclein Cerebrospinal fluid Biomarker
The quantification of a-synuclein (aSyn) in cerebrospinal fluid (CSF) has been proposed as a diagnostic biomarker for Parkinson’s disease and other aSyn-related diseases, such as multiple system atrophy and dementia with Lewy bodies. Most studies show decreased levels of aSyn in diseased CSF samples compared to control samples, but discrepant findings and overlapping values have been a major limitation for the use of CSF aSyn as a biomarker. This review addresses the current knowledge and investigates whether CSF aSyn is an ideal biomarker that can detect fundamental neuropathology features. It will also discuss whether CSF aSyn has been validated in neuropathologically confirmed cases, whether it shows a diagnostic sensitivity and whether it has a specificity above 80%. The review of current literature will also determine if sampling CSF aSyn is reliable, reproducible, noninvasive, simple to perform, inexpensive, and whether it has been investigated by at least two independent studies. CSF aSyn appears to meet most of these criteria, which have been proposed for ideal biomarkers, but further validation of this and other markers is needed to best introduce a panel of biomarkers in the early and differential diagnosis of Parkinson’s disease. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Neurodegenerative diseases, like dementia or movement disorders, are an increasing problem for our ageing population. This is especially true given the lack of available neuroprotective agents. However, several factors limit the development of new therapies: first, there are tremendous costs associated with introducing new drugs in clinical settings; second, efficacy testing, regulatory approvals, and the recruitment and conduction of clinical trials need several years and large numbers of subjects before reaching patients; third, due to the lack of objective biomarkers, the selection of study participants relies on clinical diagnosis, which includes a 10–20% misdiagnosis rate in Parkinson’s disease (PD); finally, clinical trials rely on clinical measures alone, which lack of sensitivity and objectivity. Recent clinical futility trials with promising neuroprotective agents have failed to reach the clinical end-point, despite promising preclinical results [1]. Therefore, a biological indicator reflecting the underlying pathogenesis and progressing pathology of the disease is urgently needed. Such a marker could be brain imaging criteria, surrogate markers, or based on the measurements of specific compounds in biological fluids. In Alzheimer’s disease (AD) * Correspondence: Paracelsus-Elena-Klinik, Klinikstrasse 16, 34128 Kassel, Germany. Tel.: +49 561 6009 200; fax: +49 561 6009 126. E-mail address: [email protected] (B. Mollenhauer). 1353-8020/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
the presence of medial temporal lobe atrophy, a specific pattern on functional neuroimaging with PET, and abnormal proteins in cerebrospinal fluid (CSF) have been introduced to complement the clinical criteria [2]. The quantification of CSF proteins [i.e. tau protein, phosphorylated tau protein and amyloid-b (Ab)] reflects pathological alterations of the brain (i.e., neurofibrillary tangles and amyloid plaques). In parallel, the measurement of aSyn as a biomarker for PD was interrogated. 2. Introduction of a-synuclein in cerebrospinal fluid as biomarker The 140 amino acid-long and approximately 16 kDa aSyn protein is the major constituent of intracellular Lewy bodies (LBs) in the brains of patients with PD and dementia with Lewy bodies (DLB). It is also the major constituent of glial cytoplasmic inclusions in brains from patients with multiple system atrophy (MSA) [3]. Missense mutations in the aSyn-encoding SNCA gene as well as multiplication events (duplication or triplication) of the SNCA gene have been found to cause genetic forms of parkinsonism, thereby following a gene dosage effect [4]. Therefore, several groups have hypothesized that the quantification of aSyn in extracellular fluids may reflect the disease state and disease rate (severity), and may also serve as a prognostic marker for aSyn-related diseases. After some methodological challenges were overcome that had been hampering the detection methods, full-length aSyn has
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
been detected in biological fluids, including plasma, conditioned cell media, and CSF [5]. An increasing number of research groups have quantified total aSyn in the CSF, with discrepant findings. A consensus is emerging showing that aSyn-related disorders show lower levels of total aSyn in the CSF. However, the value of CSF aSyn as a biomarker has yet to be determined. A consensus report for biomarkers in AD suggested that the ideal biomarker should detect a fundamental feature of neuropathology, that it should be validated in neuropathologically confirmed cases, that it should have a diagnostic sensitivity >80% (for detecting AD), and that it should have a specificity of >80% for distinguishing other dementias. Biomarker collection should also be reliable, reproducible, noninvasive, simple to perform, inexpensive, and it should be investigated by at least two independent studies [6]. The following sections will critically review how CSF aSyn could be beneficial as a biomarker and what limitations are still present.
3. CSF aSyn detects a fundamental feature of neuropathology In AD, Ab level in the CSF represents Ab plaque formation and correlates with plaque load, which is the neuropathological hallmark of AD and can be shown with in vivo amyloid imaging [7]. Classical PD is associated with LB-positive and LB-negative loss of dopamine neurons in the substantia nigra. By definition, the semi-quantitative count of aSyn-positive LBs in the cortex is essential for a “definite” diagnosis of DLB according to consensus criteria [8]. LBs do not represent a specific marker for an underlying process of PD or DLB, which has been shown in more than 900 autopsied patients by Parkkinen et al. [9]. In this study, 106 subjects showed aSyn-positive inclusions at autopsy, suggesting a synucleinopathy-related disorder. Only 30% of these 106 donors with “incidental LB disease” in the brain had any clinical evidence of a neurodegenerative disease process by clinical review. It is therefore believed that LB formation is rather a secondary event. Just recently it was shown that neuronal dysfunction and cell loss precedes LB pathology. The detection of insoluble, smaller presynaptic oligomers of human aSyn, which precede LB formation and are highly abundant in affected areas of the brain, has recently been proposed as a more sensitive and specific biomarker of aSynassociated PD and DLB at autopsy [10].
4. CSF aSyn has been validated in some neuropathologically confirmed cases CSF aSyn has been shown to be decreased in a small series of 13 neuropathologically verified cases with DLB [11] versus 21 cases of AD or other diseases that were verified by autopsy (corticobasal degeneration, cerebrovascular disease or frontotemporal dementia). This study did not include PD. However, the levels of CSF aSyn did not correlate with LBs [11]. The analysis of neuropathologically confirmed cases is important because the UK Brain Bank Criteria for the clinical diagnosis of PD can only help us diagnose PD disease at a confidence level of “probable”; “definite” PD is only diagnosed by autopsy. Therefore, the lack of neuropathologic data is a major drawback of many biomarker studies. Correlation with neuropathological findings is extremely important to help diagnose different forms of typical PD and other neurodegenerative disorders more accurately. It is also essential for exploring the mechanisms and biological markers for those diseases that clinically also present with parkinsonism (see also below) [11].
S77
5. The diagnostic sensitivity of CSF aSyn is up to 93%, but the specificity is much lower for distinguishing PD from other movement disorders Depending on the needs (higher sensitivity or higher specificity) of the study investigating aSyn in CSF, the sensitivity ranged from 70% to 93%, while the specificity ranged from 39% to 83% for the diagnosis of PD in studies with advanced PD subjects and neurological controls [11,12]. Most challenging in the diagnosis of PD is its early diagnosis, when the accuracy of the clinical criteria is only 80–90%, even when diagnosed by movement disorders experts [13]. In a single-center study with early denovo PD subjects and healthy controls, the sensitivity reached 91% with a poor specificity of 25%. The corresponding area-underthe-curve value was 0.65 (confidence interval, 0.554–0.750). This observation was also recently shown in a multicenter setting [14]. The latter study also showed for the first time higher levels of CSF aSyn in PD subjects with an akinetic rigid phenotype compared to a tremor-dominant phenotype, which could be responsible for the tremendous overlap of single values of recently analyzed cohorts [5]. These and other potential modifiers of aSyn-levels, such as circadian fluctuations, dopaminergic therapy, and other pharmacotherapy, have to be identified. Another diagnostic challenge is the differential diagnosis between diseases presenting with parkinsonism, e.g. progressive supranuclear palsy (PSP), MSA, corticobasal degeneration (CBD). Because low levels of CSF aSyn have also been reported for other synuclein-related diseases (DLB and MSA), this differentiation remains difficult [11,15]. Most likely, combinations with other aSyn species and other CSF markers need to be considered to improve the diagnostic performance. Combining aSyn with AD biomarkers, such as amyloid-b (Ab1–42) as well as total and phosphorylated tau protein (T-tau; P-tau181P) helps to differentiate PD from other neurological disorders, and it might help to improve PDD/DLB versus AD in the differential diagnosis [11,16,17]. In the same way that phosphorylated tau protein is more specific for AD, post-translational modifications of aSyn may increase the diagnostic accuracy for PD. aSyn in LBs has been shown to be phosphorylated (at S87, S129 or Y125), ubiquitinated (K12, K21, K23), truncated (at its C-terminus), and oxidized (by tyrosine nitration). Besides the monomeric aSyn species mentioned above, oligomeric and post-translationally modified aSyn can be detected in the CSF. However, it is largely unknown to what extent monomeric and oligomeric aSyn levels and post-translational modifications reflect the aSyn’s condition in the CNS or if it correlates with disease progression or severity [18]. Other potential biomarker candidates have been proposed in smaller cross-sectional cohorts at single centers and need further validation. 6. Is the quantification of CSF aSyn reliable and reproducible? Since the quantification of CSF aSyn has been carried out with different assays, the test quality criteria reliability and reproducibility of CSF aSyn varies and needs to be proven with one assay across different laboratories. A single study using one protocol for an assay with the same samples that involved 18 different laboratories is currently under way. It is part of the EUBIOMARKAPD project. In one published assay [19] the test/re-test reliability was shown to be >90%. Evidence that assay results are easily repeatable at a single center by different operators on different days and between centers is necessary for wider assay usage and for clear interpretation of data collected in several different studies. Inter- and intra-plate variability with a co-efficient of signal variation that exceeds 10% often leads to discrepancies when comparing results between
S78
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
operators, between centers, and more importantly, between samples from the same diagnostic group (see above). These quality control measures are specifically important when clinical cohorts need to be analyzed on more than one plate, which is often the case in larger validation cohorts. 7. Is CSF aSyn as biomarker noninvasive? CSF originates from the choroid plexus within the ventricles of the central nervous system (CNS). The affected areas of PD pathology are part of this relevant area, which contributes to the composition of CSF. While 80% of CSF proteins are derived from the peripheral blood (e.g., plasma) up to 20% of the CSF proteome are synthesized and released by cells of the CNS. Therefore, cellular and biochemical alterations of brain metabolism/pathology have the potential to leave a corresponding “marker” in the CSF. The optimal biomarkers reflect aspects proximal to the disease process, which is true for CSF aSyn, which has been shown to derive from neurons of the CNS [20]. Therefore, the CSF seems like it would be the optimal place to find biomarkers for neurological diseases, but the relative invasiveness of lumbar punctures is still the major drawback in CSF biomarkers. CSF can be easily and safely collected at the lower end of the central nervous system. A routine lumbar puncture is minimally invasive and very well tolerated, especially in patients with neurodegenerative disease. Further development and usage of new instruments, such as an ‘atraumatic’ (“Sprotte”) needle for CSF collection, has significantly decreased the incidence of postlumbar puncture headache. Nevertheless, to rely on CSF for biomarker development is far from optimal. However, this may not be necessary for PD in the future. It is well known that the aSyn pathology in PD is present in the peripheral nervous system and organs, which points towards a systemic disease [21]. Therefore, the chances are good that a peripheral marker will be detected in easily accessible biological fluids, which are still proximal to the disease process. The possibility of quantifying a biomarker in other biological fluids or tissue has already been shown in peripheral blood [22,23] where it is known that the red blood cells are the major source of aSyn [24]. Studies on aSyn in peripheral blood have shown discrepant findings, and it needs to be investigated whether aSyn in blood is still a proximal marker for PD. aSyn has also been quantified in other biological fluids [23]. For example, the presence of aSyn in the salivary glands has been shown [25]. Due to the early involvement of the gastrointestinal tract in PD, different tissue biopsies (from the colon and stomach) have also been analyzed for aSyn. Besides technical problems with the depth of biopsy, the invasiveness of such a biopsy needs to be opposed to a lumbar puncture. If Braak’s hypothesis is true, then the gastrointestinal tract may be a source for a very early marker in PD, even before the central nervous system is affected. Independently from aSyn, oxidative stress parameters can be quantified, even in urine. These parameters have shown differences in PD subjects compared to controls [26]. Future studies will address whether better accessible peripheral fluids can serve as biomarkers, or if they can substitute or stand above the current CSF tests. 8. CSF aSyn is inexpensive The quantification of a laboratory marker is not expensive when a standardized assay is available (costs vary from $20 to $30), especially when compared to imaging markers. However, the establishment of an assay is time consuming and costly. Several different assays have been described by using different antibodies, sample preparation, and calibrators, which accounted for discrepant results [5]. Even if the same commercially available standardized assay is used, the results of CSF biomarkers in AD across laboratories
can vary and there is need for a global quality control program [27]. This is important in determining whether a marker will be incorporated in multicenter trials. The advantage of markers in biological fluids for multicenter approaches lies in the convenience of shipment to a central laboratory, but the pre-analytical handling needs to be standardized and monitored. 9. Has CSF aSyn been investigated by at least two independent studies? As shown above, an increasing number of investigators have quantified aSyn in the CSF in the past few years. The decrease in synuclein-related disorders has been shown for PD, DLB and MSA. 10. Summary CSF aSyn fits some criteria for an ideal biomarker according to the consensus report, but there are still aspects to address and major drawbacks for the introduction in the clinical routine. There is tremendous overlap of single values and also of diagnostic performance when applied as single marker. Combination with other markers, e.g. tau protein or oligomeric aSyn, is promising, but more accurate biomarkers need to be detected and assays (singleand multiplex) need to be developed. The exploration of new biological marker candidates is not restricted to the CSF, so other biological fluids should be included, such as blood and saliva. Also, in vivo functional imaging, smell test, polysomnography and other diagnostic approaches can contribute to an algorithm for the best and early differential diagnosis of PD and other neurodegenerative diseases, in addition to marker candidates in biological fluids. CSF aSyn needs to be validated in more subjects with available neuropathology data. Due to the poor accuracy of clinical criteria for “probable” PD, there is an urgent need for biomarker studies involving “definite” cases and correlation with available neuropathologic findings. Also, the mechanism underlying the reduction of CSF aSyn in aSyn-related disorders needs to be explored further to understand the underlying process of PD [11]. It is still unclear whether CSF aSyn can serve as a progression marker of the disease. A progression marker is vital for clinical trials with potentially neuroprotective agents as an objective read-out. One of the most intriguing questions in the PD biomarker field is the status of markers in the premotor disease state. At the time of the clinical diagnosis, the neurodegenerative process has already advanced, and the patient is showing more than 50% dopaminergic cell loss. This diagnosis is based on motor symptoms identified in the UK Brain Bank Criteria. Non-motor symptoms, such as hyposmia or REM-sleep behavior disorder (RBD) may precede motor PD and could thus be studied in terms of potential biomarkers. Up to 82% of older men diagnosed with idiopathic RBD develop parkinsonism and dementia [28]. Therefore, biomarkers are currently being studied in subjects with RBD and hyposmia. The predictive potential of CSF biomarkers in PD was also investigated within asymptomatic leucine-rich repeat kinase 2 (LRRK2) mutation carriers at risk for developing PD. In these cases, CSF aSyn failed to correlate with striatal dopaminergic function by positron emission tomography (PET) [29]. Larger longitudinal multicenter studies for subjects at risk for PD (i.e., subjects with RBD, hyposmia and asymptomatic gene mutation carriers) are currently under way [14,30]. These will validate CSF aSyn and other biomarkers for improving the clinical routine of PD patients and therapeutic clinical trials with neuroprotective agents. Conflict of interests The author has no conflict of interest to declare.
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
References [1] Elm JJ, Goetz CG, Ravina B, Shannon K, Wooten GF, Tanner CM, et al. A responsive outcome for Parkinson’s disease neuroprotection futility studies. Ann Neurol 2005;57:197–203. [2] Dubois B, Feldman HH, Jacova C, Dekosky ST, Barberger-Gateau P, Cummings J, et al. Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 2007;6:734–46. [3] Spillantini MG, Goedert M. The alpha-synucleinopathies: Parkinson’s disease, dementia with Lewy bodies, and multiple system atrophy. Ann N Y Acad Sci 2000;920:16–27. [4] Farrer MJ. Genetics of Parkinson disease: paradigm shifts and future prospects. Nat Rev Genet 2006;7:306–18. [5] Mollenhauer B, El-Agnaf OM, Marcus K, Trenkwalder C, Schlossmacher MG. Quantification of alpha-synuclein in cerebrospinal fluid as a biomarker candidate: review of the literature and considerations for future studies. Biomark Med 2010;4:683–99. [6] Trojanowski JQ, Growdon JH. A new consensus report on biomarkers for the early antemortem diagnosis of Alzheimer disease: current status, relevance to drug discovery, and recommendations for future research. J Neuropathol Exp Neurol 1998;57:643–4. [7] Fagan AM, Mintun MA, Mach RH, Lee SY, Dence CS, Shah AR, et al. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Abeta42 in humans. Ann Neurol 2006;59:512–9. [8] McKeith IG, Dickson DW, Lowe J, Emre M, O’Brien JT, Feldman H, et al. Diagnosis and management of dementia with Lewy bodies: Third report of the DLB consortium. Neurology 2005;65:1863–72. [9] Parkkinen L, Kauppinen T, Pirttila T, Autere JM, Alafuzoff I. Alpha-synuclein pathology does not predict extrapyramidal symptoms or dementia. Ann Neurol 2005;57:82–91. [10] Schulz-Schaeffer WJ. Neurodegeneration in Parkinson disease: moving Lewy bodies out of focus. Neurology 2012;79:2298–9. [11] Mollenhauer B, Locascio JJ, Schulz-Schaeffer W, Sixel-Doring F, Trenkwalder C, Schlossmacher MG. alpha-Synuclein and tau concentrations in cerebrospinal fluid of patients presenting with parkinsonism: a cohort study. Lancet Neurol 2011;10:230–40. [12] Hong Z, Shi M, Chung KA, Quinn JF, Peskind ER, Galasko D, et al. DJ-1 and alphasynuclein in human cerebrospinal fluid as biomarkers of Parkinson’s disease. Brain 2010;133:713–26. [13] Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–4. [14] Kang JH, Irwin DJ, Chen-Plotkin AS, Siderowf A, Caspell C, Coffey CS, et al. Association of cerebrospinal fluid beta-amyloid 1–42, T-tau, P-tau181, and alpha-synuclein levels with clinical features of drug-naive patients with early Parkinson disease. JAMA Neurol 2013 Aug 26 [Epub ahead of print]. ¨ [15] Hall S, Ohrfelt A, Constantinescu R, Andreasson U, Surova Y, Bostrom F, et al. Accuracy of a panel of 5 cerebrospinal fluid biomarkers in the differential diagnosis of patients with dementia and/or parkinsonian disorders. Arch Neurol 2012;69:1445–52.
S79
[16] Aerts MB, Esselink RA, Abdo WF, Bloem BR, Verbeek MM. CSF alpha-synuclein does not differentiate between parkinsonian disorders. Neurobiol Aging 2012; 33:430 e1–3. [17] Parnetti L, Chiasserini D, Bellomo G, Giannandrea D, De Carlo C, Qureshi MM, et al. Cerebrospinal fluid Tau/alpha-synuclein ratio in Parkinson’s disease and degenerative dementias. Mov Disord 2011;26:1428–35. [18] Schmid AW, Fauvet B, Moniatte M, Lashuel HA. Alpha-synuclein posttranslational modifications as potential biomarkers for Parkinson’s disease and other synucleinopathies. Mol Cell Proteomics 2013 Aug 23 [Epub ahead of print]. [19] Mollenhauer B, Cullen V, Kahn I, Krastins B, Outeiro TF, Pepivani I, et al. Direct quantification of CSF alpha-synuclein by ELISA and first cross-sectional study in patients with neurodegeneration. Exp Neurol 2008;213:315–25. [20] Mollenhauer B, Trautmann E, Otte B, Ng J, Spreer A, Lange P, et al. a-Synuclein in human cerebrospinal fluid is principally derived from neurons of the central nervous system. J Neural Transm 2012;119:739–46. [21] Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging 2003;24:197–211. [22] Shi M, Sui YT, Peskind ER, Li GG, Hwang H, Devic I, et al. Salivary tau species are potential biomarkers of Alzheimer’s disease. J Alzheimers Dis 2011;27(2):299– 305. [23] Devic I, Hwang H, Edgar JS, Izutsu K, Presland R, Pan C, et al. Salivary alphasynuclein and DJ-1: potential biomarkers for Parkinson’s disease. Brain 2011; 134:e178. [24] Scherzer CR, Grass JA, Liao Z, Pepivani I, Zheng B, Eklund AC, et al. GATA transcription factors directly regulate the Parkinson’s disease-linked gene alpha-synuclein. Proc Natl Acad Sci U S A 2008;105:10907–12. [25] Del Tredici K, Hawkes CH, Ghebremedhin E, Braak H. Lewy pathology in the submandibular gland of individuals with incidental Lewy body disease and sporadic Parkinson’s disease. Acta Neuropathol 2010;119:703–13. [26] Bolner A, Pilleri M, De Riva V, Nordera GP. Plasma and urinary HPLC-ED determination of the ratio of 8-OHdG/2-dG in Parkinson’s disease. Clin Lab 2011;57:859–66. [27] Mattsson N, Blennow K, Zetterberg H. Inter-laboratory variation in cerebrospinal fluid biomarkers for Alzheimer’s disease: united we stand, divided we fall. Clin Chem Lab Med 2010;48:603–7. [28] Iranzo A, Serradell M, Vilaseca I, Valldeoriola F, Salamero M, Molina C, et al. Longitudinal assessment of olfactory function in idiopathic REM sleep behavior disorder. Parkinsonism Relat Disord 2013;19:600–4. [29] Shi M, Furay AR, Sossi V, Aasly JO, Armaly J, Wang Y, et al. DJ-1 and aSYN in LRRK2 CSF do not correlate with striatal dopaminergic function. Neurobiol Aging 2012;33:836.e5–7. [30] Mollenhauer B, Trautmann E, Sixel-Doring ¨ F, Wicke T, Ebentheuer J, Schaumburg M, et al. Nonmotor and diagnostic findings in subjects with de novo Parkinson disease of the DeNoPa cohort. Neurology 2013;81:1226–34
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Quantification of a-synuclein in cerebrospinal fluid: How ideal is this biomarker for Parkinson’s disease? Brit Mollenhauer a,b, * a Paracelsus-Elena-Klinik,
Center of Parkinsonism and Movement Disorders, Kassel, and b University Medical Center, Department of Neurosurgery and Neuropathology, Germany
article info
summary
Keywords: Parkinson’s disease a-Synuclein Cerebrospinal fluid Biomarker
The quantification of a-synuclein (aSyn) in cerebrospinal fluid (CSF) has been proposed as a diagnostic biomarker for Parkinson’s disease and other aSyn-related diseases, such as multiple system atrophy and dementia with Lewy bodies. Most studies show decreased levels of aSyn in diseased CSF samples compared to control samples, but discrepant findings and overlapping values have been a major limitation for the use of CSF aSyn as a biomarker. This review addresses the current knowledge and investigates whether CSF aSyn is an ideal biomarker that can detect fundamental neuropathology features. It will also discuss whether CSF aSyn has been validated in neuropathologically confirmed cases, whether it shows a diagnostic sensitivity and whether it has a specificity above 80%. The review of current literature will also determine if sampling CSF aSyn is reliable, reproducible, noninvasive, simple to perform, inexpensive, and whether it has been investigated by at least two independent studies. CSF aSyn appears to meet most of these criteria, which have been proposed for ideal biomarkers, but further validation of this and other markers is needed to best introduce a panel of biomarkers in the early and differential diagnosis of Parkinson’s disease. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Neurodegenerative diseases, like dementia or movement disorders, are an increasing problem for our ageing population. This is especially true given the lack of available neuroprotective agents. However, several factors limit the development of new therapies: first, there are tremendous costs associated with introducing new drugs in clinical settings; second, efficacy testing, regulatory approvals, and the recruitment and conduction of clinical trials need several years and large numbers of subjects before reaching patients; third, due to the lack of objective biomarkers, the selection of study participants relies on clinical diagnosis, which includes a 10–20% misdiagnosis rate in Parkinson’s disease (PD); finally, clinical trials rely on clinical measures alone, which lack of sensitivity and objectivity. Recent clinical futility trials with promising neuroprotective agents have failed to reach the clinical end-point, despite promising preclinical results [1]. Therefore, a biological indicator reflecting the underlying pathogenesis and progressing pathology of the disease is urgently needed. Such a marker could be brain imaging criteria, surrogate markers, or based on the measurements of specific compounds in biological fluids. In Alzheimer’s disease (AD) * Correspondence: Paracelsus-Elena-Klinik, Klinikstrasse 16, 34128 Kassel, Germany. Tel.: +49 561 6009 200; fax: +49 561 6009 126. E-mail address: [email protected] (B. Mollenhauer). 1353-8020/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
the presence of medial temporal lobe atrophy, a specific pattern on functional neuroimaging with PET, and abnormal proteins in cerebrospinal fluid (CSF) have been introduced to complement the clinical criteria [2]. The quantification of CSF proteins [i.e. tau protein, phosphorylated tau protein and amyloid-b (Ab)] reflects pathological alterations of the brain (i.e., neurofibrillary tangles and amyloid plaques). In parallel, the measurement of aSyn as a biomarker for PD was interrogated. 2. Introduction of a-synuclein in cerebrospinal fluid as biomarker The 140 amino acid-long and approximately 16 kDa aSyn protein is the major constituent of intracellular Lewy bodies (LBs) in the brains of patients with PD and dementia with Lewy bodies (DLB). It is also the major constituent of glial cytoplasmic inclusions in brains from patients with multiple system atrophy (MSA) [3]. Missense mutations in the aSyn-encoding SNCA gene as well as multiplication events (duplication or triplication) of the SNCA gene have been found to cause genetic forms of parkinsonism, thereby following a gene dosage effect [4]. Therefore, several groups have hypothesized that the quantification of aSyn in extracellular fluids may reflect the disease state and disease rate (severity), and may also serve as a prognostic marker for aSyn-related diseases. After some methodological challenges were overcome that had been hampering the detection methods, full-length aSyn has
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
been detected in biological fluids, including plasma, conditioned cell media, and CSF [5]. An increasing number of research groups have quantified total aSyn in the CSF, with discrepant findings. A consensus is emerging showing that aSyn-related disorders show lower levels of total aSyn in the CSF. However, the value of CSF aSyn as a biomarker has yet to be determined. A consensus report for biomarkers in AD suggested that the ideal biomarker should detect a fundamental feature of neuropathology, that it should be validated in neuropathologically confirmed cases, that it should have a diagnostic sensitivity >80% (for detecting AD), and that it should have a specificity of >80% for distinguishing other dementias. Biomarker collection should also be reliable, reproducible, noninvasive, simple to perform, inexpensive, and it should be investigated by at least two independent studies [6]. The following sections will critically review how CSF aSyn could be beneficial as a biomarker and what limitations are still present.
3. CSF aSyn detects a fundamental feature of neuropathology In AD, Ab level in the CSF represents Ab plaque formation and correlates with plaque load, which is the neuropathological hallmark of AD and can be shown with in vivo amyloid imaging [7]. Classical PD is associated with LB-positive and LB-negative loss of dopamine neurons in the substantia nigra. By definition, the semi-quantitative count of aSyn-positive LBs in the cortex is essential for a “definite” diagnosis of DLB according to consensus criteria [8]. LBs do not represent a specific marker for an underlying process of PD or DLB, which has been shown in more than 900 autopsied patients by Parkkinen et al. [9]. In this study, 106 subjects showed aSyn-positive inclusions at autopsy, suggesting a synucleinopathy-related disorder. Only 30% of these 106 donors with “incidental LB disease” in the brain had any clinical evidence of a neurodegenerative disease process by clinical review. It is therefore believed that LB formation is rather a secondary event. Just recently it was shown that neuronal dysfunction and cell loss precedes LB pathology. The detection of insoluble, smaller presynaptic oligomers of human aSyn, which precede LB formation and are highly abundant in affected areas of the brain, has recently been proposed as a more sensitive and specific biomarker of aSynassociated PD and DLB at autopsy [10].
4. CSF aSyn has been validated in some neuropathologically confirmed cases CSF aSyn has been shown to be decreased in a small series of 13 neuropathologically verified cases with DLB [11] versus 21 cases of AD or other diseases that were verified by autopsy (corticobasal degeneration, cerebrovascular disease or frontotemporal dementia). This study did not include PD. However, the levels of CSF aSyn did not correlate with LBs [11]. The analysis of neuropathologically confirmed cases is important because the UK Brain Bank Criteria for the clinical diagnosis of PD can only help us diagnose PD disease at a confidence level of “probable”; “definite” PD is only diagnosed by autopsy. Therefore, the lack of neuropathologic data is a major drawback of many biomarker studies. Correlation with neuropathological findings is extremely important to help diagnose different forms of typical PD and other neurodegenerative disorders more accurately. It is also essential for exploring the mechanisms and biological markers for those diseases that clinically also present with parkinsonism (see also below) [11].
S77
5. The diagnostic sensitivity of CSF aSyn is up to 93%, but the specificity is much lower for distinguishing PD from other movement disorders Depending on the needs (higher sensitivity or higher specificity) of the study investigating aSyn in CSF, the sensitivity ranged from 70% to 93%, while the specificity ranged from 39% to 83% for the diagnosis of PD in studies with advanced PD subjects and neurological controls [11,12]. Most challenging in the diagnosis of PD is its early diagnosis, when the accuracy of the clinical criteria is only 80–90%, even when diagnosed by movement disorders experts [13]. In a single-center study with early denovo PD subjects and healthy controls, the sensitivity reached 91% with a poor specificity of 25%. The corresponding area-underthe-curve value was 0.65 (confidence interval, 0.554–0.750). This observation was also recently shown in a multicenter setting [14]. The latter study also showed for the first time higher levels of CSF aSyn in PD subjects with an akinetic rigid phenotype compared to a tremor-dominant phenotype, which could be responsible for the tremendous overlap of single values of recently analyzed cohorts [5]. These and other potential modifiers of aSyn-levels, such as circadian fluctuations, dopaminergic therapy, and other pharmacotherapy, have to be identified. Another diagnostic challenge is the differential diagnosis between diseases presenting with parkinsonism, e.g. progressive supranuclear palsy (PSP), MSA, corticobasal degeneration (CBD). Because low levels of CSF aSyn have also been reported for other synuclein-related diseases (DLB and MSA), this differentiation remains difficult [11,15]. Most likely, combinations with other aSyn species and other CSF markers need to be considered to improve the diagnostic performance. Combining aSyn with AD biomarkers, such as amyloid-b (Ab1–42) as well as total and phosphorylated tau protein (T-tau; P-tau181P) helps to differentiate PD from other neurological disorders, and it might help to improve PDD/DLB versus AD in the differential diagnosis [11,16,17]. In the same way that phosphorylated tau protein is more specific for AD, post-translational modifications of aSyn may increase the diagnostic accuracy for PD. aSyn in LBs has been shown to be phosphorylated (at S87, S129 or Y125), ubiquitinated (K12, K21, K23), truncated (at its C-terminus), and oxidized (by tyrosine nitration). Besides the monomeric aSyn species mentioned above, oligomeric and post-translationally modified aSyn can be detected in the CSF. However, it is largely unknown to what extent monomeric and oligomeric aSyn levels and post-translational modifications reflect the aSyn’s condition in the CNS or if it correlates with disease progression or severity [18]. Other potential biomarker candidates have been proposed in smaller cross-sectional cohorts at single centers and need further validation. 6. Is the quantification of CSF aSyn reliable and reproducible? Since the quantification of CSF aSyn has been carried out with different assays, the test quality criteria reliability and reproducibility of CSF aSyn varies and needs to be proven with one assay across different laboratories. A single study using one protocol for an assay with the same samples that involved 18 different laboratories is currently under way. It is part of the EUBIOMARKAPD project. In one published assay [19] the test/re-test reliability was shown to be >90%. Evidence that assay results are easily repeatable at a single center by different operators on different days and between centers is necessary for wider assay usage and for clear interpretation of data collected in several different studies. Inter- and intra-plate variability with a co-efficient of signal variation that exceeds 10% often leads to discrepancies when comparing results between
S78
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
operators, between centers, and more importantly, between samples from the same diagnostic group (see above). These quality control measures are specifically important when clinical cohorts need to be analyzed on more than one plate, which is often the case in larger validation cohorts. 7. Is CSF aSyn as biomarker noninvasive? CSF originates from the choroid plexus within the ventricles of the central nervous system (CNS). The affected areas of PD pathology are part of this relevant area, which contributes to the composition of CSF. While 80% of CSF proteins are derived from the peripheral blood (e.g., plasma) up to 20% of the CSF proteome are synthesized and released by cells of the CNS. Therefore, cellular and biochemical alterations of brain metabolism/pathology have the potential to leave a corresponding “marker” in the CSF. The optimal biomarkers reflect aspects proximal to the disease process, which is true for CSF aSyn, which has been shown to derive from neurons of the CNS [20]. Therefore, the CSF seems like it would be the optimal place to find biomarkers for neurological diseases, but the relative invasiveness of lumbar punctures is still the major drawback in CSF biomarkers. CSF can be easily and safely collected at the lower end of the central nervous system. A routine lumbar puncture is minimally invasive and very well tolerated, especially in patients with neurodegenerative disease. Further development and usage of new instruments, such as an ‘atraumatic’ (“Sprotte”) needle for CSF collection, has significantly decreased the incidence of postlumbar puncture headache. Nevertheless, to rely on CSF for biomarker development is far from optimal. However, this may not be necessary for PD in the future. It is well known that the aSyn pathology in PD is present in the peripheral nervous system and organs, which points towards a systemic disease [21]. Therefore, the chances are good that a peripheral marker will be detected in easily accessible biological fluids, which are still proximal to the disease process. The possibility of quantifying a biomarker in other biological fluids or tissue has already been shown in peripheral blood [22,23] where it is known that the red blood cells are the major source of aSyn [24]. Studies on aSyn in peripheral blood have shown discrepant findings, and it needs to be investigated whether aSyn in blood is still a proximal marker for PD. aSyn has also been quantified in other biological fluids [23]. For example, the presence of aSyn in the salivary glands has been shown [25]. Due to the early involvement of the gastrointestinal tract in PD, different tissue biopsies (from the colon and stomach) have also been analyzed for aSyn. Besides technical problems with the depth of biopsy, the invasiveness of such a biopsy needs to be opposed to a lumbar puncture. If Braak’s hypothesis is true, then the gastrointestinal tract may be a source for a very early marker in PD, even before the central nervous system is affected. Independently from aSyn, oxidative stress parameters can be quantified, even in urine. These parameters have shown differences in PD subjects compared to controls [26]. Future studies will address whether better accessible peripheral fluids can serve as biomarkers, or if they can substitute or stand above the current CSF tests. 8. CSF aSyn is inexpensive The quantification of a laboratory marker is not expensive when a standardized assay is available (costs vary from $20 to $30), especially when compared to imaging markers. However, the establishment of an assay is time consuming and costly. Several different assays have been described by using different antibodies, sample preparation, and calibrators, which accounted for discrepant results [5]. Even if the same commercially available standardized assay is used, the results of CSF biomarkers in AD across laboratories
can vary and there is need for a global quality control program [27]. This is important in determining whether a marker will be incorporated in multicenter trials. The advantage of markers in biological fluids for multicenter approaches lies in the convenience of shipment to a central laboratory, but the pre-analytical handling needs to be standardized and monitored. 9. Has CSF aSyn been investigated by at least two independent studies? As shown above, an increasing number of investigators have quantified aSyn in the CSF in the past few years. The decrease in synuclein-related disorders has been shown for PD, DLB and MSA. 10. Summary CSF aSyn fits some criteria for an ideal biomarker according to the consensus report, but there are still aspects to address and major drawbacks for the introduction in the clinical routine. There is tremendous overlap of single values and also of diagnostic performance when applied as single marker. Combination with other markers, e.g. tau protein or oligomeric aSyn, is promising, but more accurate biomarkers need to be detected and assays (singleand multiplex) need to be developed. The exploration of new biological marker candidates is not restricted to the CSF, so other biological fluids should be included, such as blood and saliva. Also, in vivo functional imaging, smell test, polysomnography and other diagnostic approaches can contribute to an algorithm for the best and early differential diagnosis of PD and other neurodegenerative diseases, in addition to marker candidates in biological fluids. CSF aSyn needs to be validated in more subjects with available neuropathology data. Due to the poor accuracy of clinical criteria for “probable” PD, there is an urgent need for biomarker studies involving “definite” cases and correlation with available neuropathologic findings. Also, the mechanism underlying the reduction of CSF aSyn in aSyn-related disorders needs to be explored further to understand the underlying process of PD [11]. It is still unclear whether CSF aSyn can serve as a progression marker of the disease. A progression marker is vital for clinical trials with potentially neuroprotective agents as an objective read-out. One of the most intriguing questions in the PD biomarker field is the status of markers in the premotor disease state. At the time of the clinical diagnosis, the neurodegenerative process has already advanced, and the patient is showing more than 50% dopaminergic cell loss. This diagnosis is based on motor symptoms identified in the UK Brain Bank Criteria. Non-motor symptoms, such as hyposmia or REM-sleep behavior disorder (RBD) may precede motor PD and could thus be studied in terms of potential biomarkers. Up to 82% of older men diagnosed with idiopathic RBD develop parkinsonism and dementia [28]. Therefore, biomarkers are currently being studied in subjects with RBD and hyposmia. The predictive potential of CSF biomarkers in PD was also investigated within asymptomatic leucine-rich repeat kinase 2 (LRRK2) mutation carriers at risk for developing PD. In these cases, CSF aSyn failed to correlate with striatal dopaminergic function by positron emission tomography (PET) [29]. Larger longitudinal multicenter studies for subjects at risk for PD (i.e., subjects with RBD, hyposmia and asymptomatic gene mutation carriers) are currently under way [14,30]. These will validate CSF aSyn and other biomarkers for improving the clinical routine of PD patients and therapeutic clinical trials with neuroprotective agents. Conflict of interests The author has no conflict of interest to declare.
B. Mollenhauer / Parkinsonism and Related Disorders 20S1 (2014) S76–S79
References [1] Elm JJ, Goetz CG, Ravina B, Shannon K, Wooten GF, Tanner CM, et al. A responsive outcome for Parkinson’s disease neuroprotection futility studies. Ann Neurol 2005;57:197–203. [2] Dubois B, Feldman HH, Jacova C, Dekosky ST, Barberger-Gateau P, Cummings J, et al. Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 2007;6:734–46. [3] Spillantini MG, Goedert M. The alpha-synucleinopathies: Parkinson’s disease, dementia with Lewy bodies, and multiple system atrophy. Ann N Y Acad Sci 2000;920:16–27. [4] Farrer MJ. Genetics of Parkinson disease: paradigm shifts and future prospects. Nat Rev Genet 2006;7:306–18. [5] Mollenhauer B, El-Agnaf OM, Marcus K, Trenkwalder C, Schlossmacher MG. Quantification of alpha-synuclein in cerebrospinal fluid as a biomarker candidate: review of the literature and considerations for future studies. Biomark Med 2010;4:683–99. [6] Trojanowski JQ, Growdon JH. A new consensus report on biomarkers for the early antemortem diagnosis of Alzheimer disease: current status, relevance to drug discovery, and recommendations for future research. J Neuropathol Exp Neurol 1998;57:643–4. [7] Fagan AM, Mintun MA, Mach RH, Lee SY, Dence CS, Shah AR, et al. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Abeta42 in humans. Ann Neurol 2006;59:512–9. [8] McKeith IG, Dickson DW, Lowe J, Emre M, O’Brien JT, Feldman H, et al. Diagnosis and management of dementia with Lewy bodies: Third report of the DLB consortium. Neurology 2005;65:1863–72. [9] Parkkinen L, Kauppinen T, Pirttila T, Autere JM, Alafuzoff I. Alpha-synuclein pathology does not predict extrapyramidal symptoms or dementia. Ann Neurol 2005;57:82–91. [10] Schulz-Schaeffer WJ. Neurodegeneration in Parkinson disease: moving Lewy bodies out of focus. Neurology 2012;79:2298–9. [11] Mollenhauer B, Locascio JJ, Schulz-Schaeffer W, Sixel-Doring F, Trenkwalder C, Schlossmacher MG. alpha-Synuclein and tau concentrations in cerebrospinal fluid of patients presenting with parkinsonism: a cohort study. Lancet Neurol 2011;10:230–40. [12] Hong Z, Shi M, Chung KA, Quinn JF, Peskind ER, Galasko D, et al. DJ-1 and alphasynuclein in human cerebrospinal fluid as biomarkers of Parkinson’s disease. Brain 2010;133:713–26. [13] Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–4. [14] Kang JH, Irwin DJ, Chen-Plotkin AS, Siderowf A, Caspell C, Coffey CS, et al. Association of cerebrospinal fluid beta-amyloid 1–42, T-tau, P-tau181, and alpha-synuclein levels with clinical features of drug-naive patients with early Parkinson disease. JAMA Neurol 2013 Aug 26 [Epub ahead of print]. ¨ [15] Hall S, Ohrfelt A, Constantinescu R, Andreasson U, Surova Y, Bostrom F, et al. Accuracy of a panel of 5 cerebrospinal fluid biomarkers in the differential diagnosis of patients with dementia and/or parkinsonian disorders. Arch Neurol 2012;69:1445–52.
S79
[16] Aerts MB, Esselink RA, Abdo WF, Bloem BR, Verbeek MM. CSF alpha-synuclein does not differentiate between parkinsonian disorders. Neurobiol Aging 2012; 33:430 e1–3. [17] Parnetti L, Chiasserini D, Bellomo G, Giannandrea D, De Carlo C, Qureshi MM, et al. Cerebrospinal fluid Tau/alpha-synuclein ratio in Parkinson’s disease and degenerative dementias. Mov Disord 2011;26:1428–35. [18] Schmid AW, Fauvet B, Moniatte M, Lashuel HA. Alpha-synuclein posttranslational modifications as potential biomarkers for Parkinson’s disease and other synucleinopathies. Mol Cell Proteomics 2013 Aug 23 [Epub ahead of print]. [19] Mollenhauer B, Cullen V, Kahn I, Krastins B, Outeiro TF, Pepivani I, et al. Direct quantification of CSF alpha-synuclein by ELISA and first cross-sectional study in patients with neurodegeneration. Exp Neurol 2008;213:315–25. [20] Mollenhauer B, Trautmann E, Otte B, Ng J, Spreer A, Lange P, et al. a-Synuclein in human cerebrospinal fluid is principally derived from neurons of the central nervous system. J Neural Transm 2012;119:739–46. [21] Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging 2003;24:197–211. [22] Shi M, Sui YT, Peskind ER, Li GG, Hwang H, Devic I, et al. Salivary tau species are potential biomarkers of Alzheimer’s disease. J Alzheimers Dis 2011;27(2):299– 305. [23] Devic I, Hwang H, Edgar JS, Izutsu K, Presland R, Pan C, et al. Salivary alphasynuclein and DJ-1: potential biomarkers for Parkinson’s disease. Brain 2011; 134:e178. [24] Scherzer CR, Grass JA, Liao Z, Pepivani I, Zheng B, Eklund AC, et al. GATA transcription factors directly regulate the Parkinson’s disease-linked gene alpha-synuclein. Proc Natl Acad Sci U S A 2008;105:10907–12. [25] Del Tredici K, Hawkes CH, Ghebremedhin E, Braak H. Lewy pathology in the submandibular gland of individuals with incidental Lewy body disease and sporadic Parkinson’s disease. Acta Neuropathol 2010;119:703–13. [26] Bolner A, Pilleri M, De Riva V, Nordera GP. Plasma and urinary HPLC-ED determination of the ratio of 8-OHdG/2-dG in Parkinson’s disease. Clin Lab 2011;57:859–66. [27] Mattsson N, Blennow K, Zetterberg H. Inter-laboratory variation in cerebrospinal fluid biomarkers for Alzheimer’s disease: united we stand, divided we fall. Clin Chem Lab Med 2010;48:603–7. [28] Iranzo A, Serradell M, Vilaseca I, Valldeoriola F, Salamero M, Molina C, et al. Longitudinal assessment of olfactory function in idiopathic REM sleep behavior disorder. Parkinsonism Relat Disord 2013;19:600–4. [29] Shi M, Furay AR, Sossi V, Aasly JO, Armaly J, Wang Y, et al. DJ-1 and aSYN in LRRK2 CSF do not correlate with striatal dopaminergic function. Neurobiol Aging 2012;33:836.e5–7. [30] Mollenhauer B, Trautmann E, Sixel-Doring ¨ F, Wicke T, Ebentheuer J, Schaumburg M, et al. Nonmotor and diagnostic findings in subjects with de novo Parkinson disease of the DeNoPa cohort. Neurology 2013;81:1226–34
