Journal of Neural Transmission (2018) 125:1417–1432

https://doi.org/10.1007/s00702-018-1910-4

NEUROLOGY AND PRECLINICAL NEUROLOGICAL STUDIES - REVIEW ARTICLE

Diagnostic biomarkers for Parkinson’s disease at a glance: where are

we?
Ilaria Cova1 · Alberto Priori2 

Received: 16 February 2018 / Accepted: 24 July 2018 / Published online: 25 August 2018
© The Author(s) 2018

Abstract
Parkinson’s disease (PD) is a neurodegenerative disorder whose aetiology remains unclear: degeneration involves several
neurotransmission systems, resulting in a heterogeneous disease characterized by motor and non-motor symptoms. PD causes
progressive disability that responds only to symptomatic therapies. Future advances include neuroprotective strategies for
use in at-risk populations before the clinical onset of disease, hence the continuing need to identify reliable biomarkers
that can facilitate the clinical diagnosis of PD. In this evaluative review, we summarize information on potential diagnostic
biomarkers for use in the clinical and preclinical stages of PD.

Keywords  Parkinson’s disease · Diagnosis · Biomarkers

Introduction and surgery aim merely to control symptoms for several

years. Current treatment strategies focus on dopamine
Parkinson’s disease (PD) is the second most common neu- replacement to correct or at least partially correct motor
rodegenerative disorder after Alzheimer’s disease: the inci- signs caused by dopamine deficiency. The key neuropathol-
dence rate is 8–18 per 100,000 person-years in prospective ogy in PD is Lewy body deposition (abnormal aggregates
population-based studies and the prevalence increases with of a misfolded protein called α-synuclein) and consequently
age, the main known risk factor (Lee and Gilbert 2016). neuronal dysfunction, involving many other brain areas and
Typical PD symptoms include motor features (bradykinesia, neurotransmitter systems. In their early research, Brack et al.
postural disturbances, rigidity or tremor or both) and non- proposed a staging scheme based on rostro-caudal patho-
motor features (hyposmia, sleep disorders, autonomic, neu- logical progression (Braak et al. 2003). In this study, they
ropsychiatric and sensory symptoms). The diagnosis of PD suggested that in the earliest stages, PD damage is confined
depends mostly on clinical motor findings (Postuma et al. to non-dopaminergic structures in the lower brainstem, the
2015), which appear when half of the substantia nigra (SN) olfactory bulb or perhaps the peripheral autonomic nervous
dopamine neurons are lost (Cheng et al. 2010). PD is, there- system, accounting for the early appearance of non-motor
fore, often diagnosed clinically when disease progression symptoms (Chaudhuri et al. 2011). This non-motor theory
is already advanced. Hence the diagnosis is notably inac- has disadvantages because it applies mainly to the subset of
curate (from nearly 74% accuracy reached by nonexperts, to patients with young onset and long disease duration. Equally
80–84% by movement disorder experts) (Rizzo et al. 2016). controversial is the connection between Lewy pathology and
Difficulties arise mainly in distinguishing PD from other par- clinical parkinsonian features (Rietdijk et al. 2017).
kinsonisms. To date, no treatment exists for PD; medications At present, we lack standardized international criteria
supporting PD diagnosis at a preclinical stage: research
efforts are, therefore, seeking ways to detect biomarkers
* Alberto Priori
[email protected] for the early diagnosis of PD. Research in recent years,
prompted by epidemiological data on risk factors and pro-
1
Neurology Unit, L. Sacco University Hospital, Milan, Italy dromal biomarkers, has proposed diagnostic criteria based
2
Department of Health Sciences, “Aldo Ravelli” Research upon the likelihood of prodromal disease (with 80% cer-
Center for Neurotechnology and Experimental Brain tainty) but applicable only for research purposes (Berg
Therapeutics, University of Milan and ASST Santi Paolo e et al. 2015). Their usefulness consists in defining target
Carlo, Milan, Italy

13
Vol.:(0123456789)

1418 I. Cova, A. Priori

populations for eventual future disease-prevention trials with clinical progression and have an increased risk to develop
disease-modifying drugs (Mahlknecht et al. 2016). disability and dementia (Rajput et al. 2009). Motor impair-
ment is routinely assessed with a clinical rating scale, such
as the Unified Parkinson’s disease rating scale (UPDRS) or
Definition of biomarkers revised version of the Movement Disorder Society (MDS-
UPDRS). The disease-modifying efficacy of a therapy
Although the term ‘biomarker’ is often used indiscrimi- that also provides a symptomatic effect is troublesome to
nately to describe any change in gene or protein expression detect. An acute levodopa challenge supports the clinical
the NIH Biomarkers Definitions Working Group defines it diagnosis of PD with a specificity that seems to increase in
more appropriately as “a characteristic that is objectively patients with mild symptoms (Merello et al. 2002).
measured and evaluated as an indicator of normal bio- Subthreshold parkinsonism detected by experts in PD
logical processes, pathogenic processes or pharmacologic assessment (i.e., UPDRS > 3 excluding tremor action) is
responses to a therapeutic intervention” (Biomarkers Defi- considered as a marker of prodromal disease that is unlikely
nitions Working Group 2001). to progress to manifest PD (Berg et al. 2015).
Several excellent reports have examined how to use and Because PD measures are subjective and rater-depend-
qualify clinical biomarkers (Freeman et al. 2010). Bio- ent, two decades ago researchers began to consider using
markers for PD could for example be used to diagnose objective monitoring systems, which lead to the develop-
PD (diagnostic markers), predict the risk of PD or dis- ment of wearable technology (such as watches, bracelets)
ease progression (prognostic markers), describe disease and connected devices (such as smartphone apps). These
severity (staging markers) and support treatment choice tools gather valuable early diagnostic data on kinematic vari-
(theragnostic markers). Diagnostic markers can be useful ables in a non-clinical setting, potentially enhancing clinical
to recognize PD before motor features become evident or diagnosis and care (Del Din et al. 2016; Zhan et al. 2018).
when motor or non-motor signs or both are still insufficient Besides motor features, some non-motor symptoms
to define disease (prodromal phase) or even to detect an (NMS) appear specific for PD as well as for other synu-
asymptomatic population at risk of PD in whom neurode- cleinopathies, neurodegenerative diseases characterized
generation is expected to begin (preclinical phase). Diag- by the abnormal accumulation of α-synuclein aggregates
nostic markers could also help to differentiate PD from in the nervous system (Lewy body dementia and multiple
other parkinsonian syndromes, insofar as misdiagnosis system atrophy). NMS may appear in a prodromal disease
often takes place early in the disease and diagnostic confir- stage (Chaudhuri and Schapira 2009). Early NMS reflect
mation needs autopsy reports. The difficulty in identifying degeneration in extra-nigral areas before the loss of nigral
early diagnostic criteria for PD depends on the fact that no dopamine neurons and include olfactory dysfunction, REM
real biomarker can yet predict illness onset. sleep behaviour disorder, autonomic dysfunction (such as
In this evaluative review, we will, therefore, summarize constipation), and depression (Berg et al. 2013) (Fig. 1).
the most referred and promising diagnostic biomarkers
now under investigation for PD (Tables 1, 2). Olfactory dysfunction

The olfactory system is one of the earliest structures affected

in PD and hyposmia is one of the NMS in PD (Xiao et al.
Clinical signs and symptoms 2014). The “olfactory vector hypothesis” supports the idea
that the olfactory pathway could be an entry-point for cer-
The most important clinical diagnostic and prognostic tain pathogens with a putative role in starting the disease
markers in PD are still the motor symptoms, crucial also (Doty 2008). Given that this system’s particular anatomy
for monitoring the response to symptomatic therapy. The enables it to bypass the blood–brain barrier, some have also
physical examination finding that best correlates with suggested that PD might be a primary olfactory disorder
nigrostriatal dopaminergic loss and the core sign of par- (Hawkes et al. 1999; Doty 2008). Some findings neverthe-
kinsonism is bradykinesia, associated with rest tremor or less excluded neurodegeneration involving dopamine neu-
rigidity, or both (Postuma et al. 2015). Since Hoen and rons in the olfactory bulb in patients with PD; some evidence
Yahr first described it in 1967, several studies suggested even indicates a paradoxical increase in dopamine neurons in
that tremor is a progression marker for benign PD (Marras some PD patients compared with controls (Duda 2010). Not
et al. 2002). Others underline differences in the clinical all patients with PD experience olfactory problems, although
course and prognosis between tremor-dominant (TD) and sensitivity rates range from 45% up to 96% (Haehner et al.
postural instability gait disorder (PIGD) or akinetic-rigid 2009). These differences may reflect the type, or the com-
subtypes (Stebbins et al. 2013), who show a more rapid bination of olfactory tests used or both, and their normative

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1419

Table 1  Major diagnostic biomarkers under investigation in Parkinson’s disease

Markers Diagnostic test sta- References Testing ­costa Invasiveness Disadvantages com-
tistics ments

Clinical biomarkers
 Mild motor symp- SE 89.3% Rizzo et al. (2016) Low Low Late indicators
toms SP 65.8%
 Levodopa challenge SE 70.9% Merello et al. (2002) Low Low Late indicator
SP 81.4%
 Olfactory dysfunc- SE 60–100% Berg et al. (2015) Low Low Lack of standardization
tion SP 72–94% in olfactory tests
Unclear lead time
 REM behaviour sleep SE 50% Gagnon et al. (2002) Low for screens Low Differences between
disorder SP 40–65% within and Postuma et al. Moderate for PSG clinically or PSG-
10 years (2009) proven
 Excessive daytime SE 21–23% Berg et al. (2015) Low Low Very low predic-
sleepiness SP 87–92 tive positive value
(0.5–3.7%)
 Constipation SE 10–50% Berg et al. (2015) Low Low Very common in the
SP 75–96% general population
 Depression SE 0.2–45% Berg et al. (2015) Low Low Very common in the
SP 75–99.9 general population
 Visual dysfunctions N/A Armstrong (2015) Low Low Few studies
Fluid biomarkers N/A Moderate or high Minimally or moder- Few studies or scarce
ately invasive reproducibility
Tissue biomarkers Depends on harvesting Berg et al. (2015) Moderate or high Moderately invasive Few studies or scarce
method reproducibility
Imaging biomarkers
 MRI 3 T (7-Tesla) SE 94.6% (97.7%) Mahlknecht et al. High Low Available only in a
SP 94.4% (94.6%) (2017) research context
 DAT imaging In hyposmic subjects: Iranzo et al. (2010) and High Minimally invasive High-risk population
SE 77.8% Jennings et al. (2014) bias in main studies
SP 94.2% In RBD Not useful to differenti-
cohort SE 53.6% SP ate between neurode-
98.4% generative parkinso-
nian disorders
 Substantia nigra SE 44.4–82.4% Berg et al. (2013) and Moderate or high Low 10–20% people have
hyperechogenicity SP 82.5–87.2% Iranzo et al. (2014) an inadequate bone
window
 Genetic biomarkers Depends on gene Berg et al. (2015) Moderate Minimally invasive Few cases of PD owing
mutation to pure monogenetic
mutation; different
penetrance and age-
dependent expres-
sivity
 Meta-iodo-benzyl- SE 90.2% Chung and Kim (2015) Moderate Minimally invasive May be negative in
guanidine cardiac SP 81.9% genetic Parkinson
scintigraphy
Neurophysiological N/A Priori et al. (2004), Moderate Low Few data on healthy
biomarkers Swann et al. (2015) controls
and Wang et al.
(2016)
Metabolomic biomark- N/A Chen et al. (2016) and Moderate or high Minimally invasive Insufficient data
ers Kori et al. (2016)
Inflammatory bio- N/A Chen et al. (2016) Moderate or high Minimally or moder- Insufficient data
markers ately invasive

MRI magnetic resonance imaging, N/A not available

a
 Low = visit or questionnaire; moderate = requires a low-cost examination (< 200 US $); high = requires an expensive evaluation (> 200 US $)
Postuma and Berg (2016)

13

1420 I. Cova, A. Priori

Table 2  Biomarkers for the prodromal stage, diagnosis and progression of Parkinson’s disease (PD)
Prodromal Diagnostic Progression

Clinical Clinical Clinical

 Slight MS (parkinsonism)  MS (parkinsonism)  MS (tremor, postural instability)
 NMS (hyposmia, RBD, EDS, constipation,  NMS (hyposmia, RBD, EDS, constipation,  NMS (hyposmia, cognitive decline, dysauto-
depression, visual dysfunctions) depression, visual dysfunctions) nomia)
Biological fluids Biological fluids Biological fluids
 Blood (e.g., uric acid)  CSF (α-synuclein species, DJ-1, LRKK2,  CSF (α-synuclein, A ß oligomers, p-tau)
metabolites)  Blood (e.g., uric acid, EGF, IGF-1, BDNF)
 Blood (e.g., uric acid, apolipoprotein A1,
LRKK2, inflammation factors)
 Saliva (α-synuclein species)
 Urine (biopyrin, LRKK2)
Pathology Pathology –
 Biopsies (GI tract, salivary glands, olfactory  Microbiota (gut, oral)
mucosa, skin)
Imaging Imaging Imaging
 Nuclear imaging (nigrostriatal degeneration  Structural imaging (nigral degeneration on  Functional imaging (hypometabolism patterns
at DaTscan) MRI at ≥ 3 T) on FDG-PET)
 Transcranial ultrasound (substantia nigra  Functional imaging (remapping of cerebral
hyperechogenicity) connectivity at f-MRI)
 Nuclear imaging (nigrostriatal degenera-
tion at DaTscan, microglial activation at
PET imaging, post-ganglionic sympathetic
nerve damage on cardiac and salivary gland
scintigraphy)
– Neurophysiology –
 Altered ERG, VEP, LFPs
– Omics –
 Proteomics, genomics, lipidomics, glycomics,
metabolomics, interactomics

MS motor symptoms, NMS non-motor symptoms, RBD REM behaviour sleep disorder, EDS excessive daytime sleepiness, CSF cerebrospinal
fluid, LRKK2 leucine-rich repeat kinase 2, EGF epidermal growth factor, IGF-1 insulin growth factor, BDNF brain-derived neurotrophic factor,
GI gastrointestinal, DaTscan dopamine transporter imaging technique, MRI magnetic resonance imaging, f-MRI functional-magnetic resonance
imaging, FDG-PET fluoro-2-deoxy-d-glucose positron-emission tomography, ERG electroretinogram, VEP visual evoked potential, LFPs long
field potentials

data. Equally important, the age distribution of PD and in essential tremor (Busenbark et al. 1992). Other studies
controls varies between different investigations. Nor can found that PD patients with predominant PIGD lose their
we ignore that hyposmia is a common phenomenon during sense of smell more often than those with TD (Stern et al.
aging (Welge-Lüssen 2009): smell dysfunction affects more 1994). In conclusion, hyposmia can be useful as a preclini-
than 50% of people aged 65 years and over and about 75% cal marker of disease in combination with other biomarkers
of subjects older than 80 years (Doty et al. 1984). Hypos- to increase the predictive value but not alone. For example,
mia can also be the early expression of cognitive disorders: combined assessment of hyposmia, motor asymmetry and a
some evidence also describes severe olfactory deficits as typical finding at ultrasound (midbrain hyper-echogenicity)
a predictive feature of PD with dementia (PDD) (Takeda could improve diagnostic accuracy in early PD (Poewe and
et al. 2014). Subjective olfactory impairment (through infor- Mahlknecht 2012).
mation from history-taking or self-reported olfactory ques-
tionnaires) may not overlap with an objective injury and a REM behaviour sleep disorder
loss of awareness of hyposmia may be associated with early
cognitive impairment in PD (Kawasaki et al. 2016). Smell is Another early PD feature indicating brainstem involvement,
less impaired, however, in other common causes of parkin- supported by the finding of Lewy body pathology, is rapid
sonism (Wenning et al. 1995), in synucleinopathies such as eye movement (REM) sleep behaviour disorder (Boeve et al.
multiple system atrophy (MSA), and in tauopathies such as 2007). REM sleep behaviour disorder (RBD) is a parasom-
corticobasal degeneration (CBD) and progressive supranu- nia which involves abnormal behaviour caused by a loss of
clear palsy (PSP); an olfactory dysfunction is infrequent also the physiological motor inhibition (usually present during

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1421

Fig. 1  Parkinson’s disease progression. Non-motor symptoms (NMS) manifest later. From Kalia and Lang (https​://doi.org/10.1016/S0140​
may emerge in a pre-motor or prodromal stage reflecting degenera- -6736(14)61393​-3) 2015;386:896–912, copyright (2015) with per-
tion in extra-nigral areas before the loss of nigral dopamine neurons. mission from Elsevier
Conversely, the motor symptoms currently required for diagnosis

REM sleep); this disorder leads to varying degrees of com- unintentionally or at inappropriate times. EDS is among the
plex motor activity which ranges from sleep talking to vio- most common sleep-related patient symptoms, affecting an
lent dream enacting behaviours potentially harmful for the estimated 20% of the population (Pagel 2009). It can mani-
patient or bed partner. The reference standard to confirm fest as primary hypersomnia of central origin or more com-
RBD is a polysomnogram (PSG). One way to avoid this monly secondary to sleep disorders (for example, sleep dep-
resource-demanding procedure is to interview the patient’s rivation and obstructive sleep apnoea), to medication effects
partner with specific questionnaires (Al-Qassabi et  al. or to other medical and psychiatric conditions.
2017). RBD affects 0.38–2.01% of people, but its preva- EDS is a common feature in PD and can affect up to 50%
lence is higher in synucleinopathies (Jiang et al. 2016) and of patients taking dopamine agonist medication (Hobson
is an early NMS that appears decades before motor involve- et al. 2002). A prospective follow-up study of incident clini-
ment not only in PD, but also in dementia with Lewy bod- cal PD in men has shown that the presence of EDS increased
ies (DLB), or MSA (Claassen et al. 2010). The presence the risk of disease up to threefold (Abbott et al. 2005). EDS
of RBD increases the risk of these neurodegenerative dis- could also appear during the course of the disease; besides
orders developing in up to 40–65% of patients at 10 years the dopamine agonist dose, disparate factors such as male
(Postuma et al. 2009) and is the prodromal marker with the gender, poorer night-time sleep, cognitive impairment, hal-
highest positive likelihood ratio (LR): patients with posi- lucinations and autonomic dysfunction, have been associated
tive PSG are 130 times more likely than those without to with higher EDS scores over time (Zhu et al. 2016). In two
have PD. Positive LRs dramatically decrease when RBD is population-based studies, EDS showed a positive LR 2.2 to
assessed by means of screen questionnaire even with > 80% predict PD, so despite its low VPP it can be considered a
specificity, from 130 to 2.2 (Postuma et al. 2015). Despite prodromal marker (Berg et al. 2015).
its high specificity, RBD is present in less than half of all
cases of PD (Gagnon et al. 2002), so it is an early sign with
scarce sensitivity. In summary, RBD alone is unreliable as a Insomnia
diagnostic biomarker. Combining the presence of RBD as a
prodromic NMS with other markers could further increase Insomnia is defined as the difficulty to initiate or maintain
its predictive value. sleep or the presence of early awakenings or both problems.
It is the most common sleep disturbance in PD patients,
Excessive daytime sleepiness manifesting in up to 80% of cases, and is multifactorial
(due to neurodegeneration, depression, anxiety, motor
Excessive daytime sleepiness (EDS) consists in the inability fluctuations, PD drugs and their withdrawal) (Al-Qassabi
to maintain wakefulness during the day, with sleep occurring et al. 2017). Insomnia may arise at any PD stage, but its

13

1422 I. Cova, A. Priori

prevalence seems to increase along with the duration (Loddo pre-motor PD. Depression in PD seems related to multiple
et al. 2017). Too little evidence is available to justify includ- neurotransmitter dysfunction, involving not only dopamine
ing insomnia as a prodromal marker of PD. in the substantia nigra, but also serotonin in the raphe nuclei
and noradrenaline in the locus coeruleus (Borgonovo et al.
Constipation 2017). Like constipation, owing to its high prevalence in
the general population, depression is not specific as a stand-
A considerably more sensitive early NMS in PD is consti- alone biomarker of prodromal PD, but can coexist with other
pation (Berg et al. 2013), which affects up to 80% of PD potential markers such as family history and substantia nigra
patients (Noyce et al. 2012). Despite its sensitivity, this hyperechogenicity (Liepelt-Scarfone et al. 2011). Depres-
symptom has a low specificity because it is relatively fre- sion is also tricky to diagnose in a patient with clinically
quent also in the general population and its prevalence manifest PD because it overlaps with motor symptoms,
increases with age (Higgins and Johanson 2004). A popula- such as slowing and bradyphrenia and with other NMS, for
tion study (Savica et al. 2009) suggested that constipation example sleep and appetite disturbance, loss of concentra-
arising also more than 20 years before motor symptom onset tion and interest and impaired libido. Patients with PD can
is associated with an increased risk of PD. The various pub- maintain intact affective responses but may have difficulties
lished studies fail to define constipation in a uniform manner. in translating such feelings into motor phenomena (Rickards
A review published in recent years (Knudsen et al. 2017) has 2005). A disorder often mistaken for depression in PD is
identified 12 different criteria. Among these, the ROME cri- abulia. Abulia can bias the estimated prevalence rates for a
teria appear the most valid tool for investigating colonic and depressive syndrome in PD. Abulia is a disorder of dimin-
anorectal dysfunctions (Drossman and Dumitrascu 2006). ished motivation due to damaged connections between the
These findings notwithstanding, idiopathic constipation is anterior cingulate area and striatum, so that abulic patients
one of the strongest risk factors for PD: men reporting a have no desire to do things without getting bored (Heimer
bowel movement frequency of < 1 per day showed an odds et al. 2008). Some help in differentiating depression from
ratio (OR) for PD of 2.3 compared with those reporting 1 abulia comes from appropriate diagnostic tools (Torbey et al.
per day (Abbott et al. 2001). 2015).
Colorectal transit time (CTT) is prolonged in approxi-
mately 80% of untreated de novo PD patients, but appears Cognition
uncorrelated with subjective constipation symptoms (Knud-
sen et al. 2017). Hence, CTT studied with radio opaque Cognitive impairment in PD is due to a fronto-striatal syn-
markers may be a potential biomarker for prodromal PD. drome which mainly affects executive function in milder
Α-synuclein accumulation and other neurodegenerative disease stages; a widespread failure in neurotransmitter sys-
changes in the enteric nervous system, in addition to signs of tems besides the dopaminergic one, such as noradrenergic,
local inflammation, oxidative stress and increased mucosal serotoninergic and cholinergic systems, is more evident with
permeability, are associated with prolonged CTT and consti- disease progression. Though cognitive decline is a well-
pation in the earliest stages of PD (Scheperjans et al. 2015). known NMS in the late stages of PD, increasing in paral-
These features, supporting the caudal-rostral progression of lel with age and PD duration, mild cognitive impairment
α-synuclein pathology suggested by Braak et al. (2003), led is recognized also in 15–20% of patients in the early PD
some to hypothesize that such environmental factors could stages; approximately 80% of patients have dementia during
act primarily through the gut, implying that gut microbiota the disease course (O’Callaghan and Lewis 2017). Cognition
might play a mediatory role. Subsequent research discover- cannot be considered as a pre-motor biomarker, but it has a
ing these gut alterations in PD is only the tip of the iceberg. key role as both a staging and prognostic biomarker, as well
A promising approach for detecting α-synuclein pathology as in sub-typing and disease stratification. An increase inci-
could be to take gastrointestinal tract biopsies (Cersosimo dence of cognitive impairment is evident in the non-tremor,
2015) thus providing a new, although invasive, diagnostic akinetic-rigid phenotype. Executive dysfunction has been
biomarker for PD. associated with a more severe gait dysfunction, freezing
of gait and postural instability, whereas specific cognitive
Depression domains were differentially related to different PIGD com-
ponents (i.e., visual impairment to a more severe freezing
About 35% of patients with PD suffer from severe depres- of gait, memory impairment to a worse postural instability)
sion (Reijnders et al. 2008). Given that depressive symptoms (Kelly et al. 2015). The NMS most associated with cogni-
may precede the onset of motor symptoms in 30% of cases tive decline are early visual hallucinations, RBD, hypos-
and their incidence increases during the few years preceding mia and depression. Dementia associated with at least one
the diagnosis, depression has a potential role in diagnosing feature among fluctuating cognition, visual hallucinations,

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1423

RBD and parkinsonism allow a diagnosis of DLB. No major controversial results (Andersen et al. 2016). Phosphorylated
clinical differences exist between DLB and PDD, both of α-synuclein combined with total α-synuclein concentrations
which belong both to the Lewy body dementia. The DLB in CSF may help in distinguishing PD from atypical parkin-
Consortium has developed an arbitrary diagnostic distinction sonisms (MSA, PSP) (Wang et al. 2012), but further studies
(McKeith et al. 2017) based on the temporal appearance of are needed in independent cohorts of patients.
motor and cognitive symptoms (the “1 year-rule” states that Other major potential CSF biomarkers studied include
cognitive decline manifests before or within a year after the DJ-1, whose mutations are a rare cause of PD (Hong et al.
onset of parkinsonism in DLB). 2010), Aβ42, which seems to correlate with cognitive
impairment, and different forms of tau and neurofilament
Visual dysfunction light chains that might help to differentiate PD from other
α-synucleinopathies (such as MSA) and primary tauopa-
Among other NMS, PD patients may have various visual thies (PSP and CBD) (Magdalinou et al. 2014). Despite
disturbances that can be sensitive markers of disease, includ- these encouraging findings, future research needs to seek
ing changes in colour vision and contrast sensitivity possibly more data. Owing to the degeneration in catecholaminergic
due to a loss of dopaminergic amacrine retinal cells, and neurons, some report lower levels of dopamine metabolites,
difficulties with complex visual tasks (for example men- such as dihydroxyphenyl acetate (DOPAC) and homovan-
tal rotation and emotion recognition) caused by a cortical illic acid (HVA) and noradrenergic metabolites including
visuoperceptual dysfunction (Weil et al. 2016). In a study dihydroxyphenylglycol (DHPG) and 3-methoxy-4-hydroxy-
conducted in 2010, Diederich et al. (2010) noted that defi- phenylglycol (MHPG) in CSF from patients with PD than
cits in colour and contrast sensitivity seem to discriminate from controls (Lee et al. 2017). In a case–control study (217
early diagnosis of PD within 3 years better than other NMS vs. 26 subjects) that lacked healthy elderly controls, others
do. Visuo-perceptual deficits rarely arise in other parkin- proposed as a possible state and trait biomarker for PD, the
sonisms and could serve to distinguish idiopathic PD from ratio of the purine metabolite xanthine over HVA (LeWitt
other diseases. Another visual dysfunction reported in PD et al. 2011).
is a decrease in retinal nerve-fibre layer thickness on optical Because lumbar puncture is an invasive procedure, and
coherence tomography (Moschos et al. 2011). Some evi- increasing evidence suggests that α-synuclein passes through
dence also suggests as a retinal biomarker a mathematical the blood–brain barrier, recent efforts concentrate on assay-
model quantifying foveal symmetry and breadth (Slotnick ing it in plasma rather than CSF. This research has yielded
et al. 2015). More data are needed on the prevalence of ambiguous results owing to the various assay techniques, the
visual markers in the preclinical stage of PD to allow their different protein forms assayed (total, oligomeric and phos-
future use as biomarkers. phorylated) and the small populations enrolled (El-Agnaf
et al. 2006; Lee et al. 2006; Li et al. 2007; Duran et al. 2010;
Shi et al. 2010; Foulds et al. 2013); others also suggested
Biological fluids gender-related differences in plasma α-synuclein expression
in PD (Caranci et al. 2013). Blood has been a target also
Altered α-synuclein metabolism in the central nervous sys- for assaying DJ-1, a protein involved in the pathogenesis of
tem (CNS) has a central role in the pathogenesis of PD and PD through oxidative stress and mitochondrial dysfunction.
may manifest also in the periphery. Studies conducted in Assaying total DJ-1 levels in PD patients has produced con-
recent years have focused on determining α-synuclein spe- troversial results (Hong et al. 2010; Shi et al. 2010), whereas
cies in different fluids and tissues. Although α-synuclein is fractions of specific DJ-1 isoforms differed in PD patients
mainly expressed by neuronal cells as a cytoplasmic protein and healthy controls. Although some researchers suggested
in its native form or in aggregated pathological (oligomeric, high oxidized DJ-1 protein levels in red blood cells as a
phosphorylated) forms (Abd-Elhadi et al. 2016), because it potential biomarker for early PD, the study enrolled a small
also has access to the extracellular space it can be detected sample (Ogawa et al. 2014). Fractions of specific isoforms
in cerebrospinal fluid (CSF). Because observations in recent of 4-hydroxy-2-nonenal-modified DJ-1 differed significantly
years have linked CSF α-synuclein levels to PD, some have in patients with PD and controls (Lin et al. 2012) and within
speculated that they could reflect disease severity (Hall stages of PD, suggesting a potential as a staging biologi-
et al. 2015), though others disagree (van Dijk et al. 2014). cal marker of disease. These findings merit confirmation in
CSF α-synuclein levels seem to decrease with age also in further studies.
healthy subjects (Koehler et al. 2015). Total α-synuclein is Another blood biomarker is serum urate (one of the major
decreased in synucleinopathies, but is not specific for PD, anti-oxidants in humans), levels of which have been known
whereas data for aggregated forms of α-synuclein yield for 20 years to be lower in PD than in healthy persons (Davis

13

1424 I. Cova, A. Priori

et al. 1996). Urate has been proposed as a staging marker in autonomic neurons in the gastrointestinal apparatus may
the PRECEPT cohort (Schwarzschild et al. 2008): clinical precede CNS pathology (Greene 2014). Another potential
progression was slower in a cohort of 399 de novo PD men gut biomarker emerges from studies investigating faecal
(but not in women) with the highest serum urate level than microbiome (brain-gut-microbiota axis), changes in whose
in men with the lowest levels. complex equilibrium can be linked to the development of
Other unbiased blood-based biomarkers for PD include several diseases, including PD. Nine studies over recent
epidermal growth factor, whose level correlates with cogni- years have suggested an intestinal dysbiosis in PD due to an
tive performance in PD (Chen-Plotkin et al. 2011) and apoli- imbalance in the bacterial strains in favour of pro-inflam-
poprotein A1, which seems to decrease the risk of develop- matory species, (Scheperjans 2017). Indeed, GI dysfunc-
ing the disease. Apolipoprotein-1 negatively correlates with tion manifests in over 80% of PD subjects. An inflammatory
dopaminergic denervation in subjects with hyposmia and a condition, associated with a reduction in intestinal cell adhe-
family history of PD (Qiang et al. 2013) and its level has a sion (Clairembault et al. 2015), may result in a dysfunction
maximum sensitivity of 71% and specificity of 60% in differ- involving the intestinal barrier (leaky gut syndrome). These
entiating patients with PD from healthy controls (Swanson changes may be the gateway allowing synuclein pathol-
et al. 2015). ogy to spread through the enteric nervous system toward
Another potentially readily accessible body fluid is saliva; the CNS, as the Braak hypothesis suggested (Braak et al.
two studies reported a reduction in total α-synuclein lev- 2003); as a possible mechanism, others have postulated
els and elevation in oligomeric α-synuclein and DJ-1 in the prion-like propagation (Chauhan and Jeans 2015) (Fig. 2).
saliva of PD patients compared with controls (Masters et al. All the studies published until now have shown significant
2015; Vivacqua et al. 2016), confirming previously reported differences between the microbiota in established PD and
results (Stewart et al. 2014). controls favouring proinflammatory species in the patients’
The search for fluid PD biomarkers focusses also on the group (Tremlett et al. 2017). Combined with other measures,
urine, as in the BIOFIND study (Kang et al. 2016). In a these distinctions may help in developing clinically useful
case–control study (92 patients vs. 65 healthy controls) oth- microbial biomarkers for PD. Longitudinal studies starting
ers found elevated excretion of urinary biopyrin, an oxida- in the premotor phase would help to investigate the causality
tive metabolite of bilirubin (Luan et al. 2015), but this prom- of such alterations. Evidence obtained in recent years under-
ising finding remains unreplicated. lines that alterations in gut microbiome precede the motor
symptoms of PD, given that they were found also in RBD
(Heintz-Buschart et al. 2017), the NMS associated with the
Pathology major likelihood to progress to a synucleinopathy.
Oral microbiota seems to differ in bacterial taxa from
Given the known presence of extr-anigral PD pathology, patients with PD compared with controls, whereas nasal
research efforts have focused on demonstrating Lewy type microbiota were similar in two studies (Pereira et al. 2017;
α-synuclein deposition in peripheral tissues, especially in Heintz-Buschart et al. 2017). Last, some evidence supports
olfactory bulb, gastrointestinal tract (GI) (such as subman- assaying α-synuclein deposition in cutaneous autonomic
dibular gland, oesophagus, stomach and colon) and skin nerves (Wang et al. 2013): whether tissue α-synuclein might
(Lee et al. 2017). Yet tissue biopsies are unlikely to help be a marker for disease progression awaits evidence from
distinguish PD from other synucleinopathies (Mehta and longitudinal studies.
Adler 2016). A review published in recent years (Schnei-
der et al. 2016) evaluated the sensitivity and specificity of
total and phosphorylated α-synuclein detected in biopsies Imaging
from various tissues from PD patients, despite the difficul-
ties in analysing the many disparate harvesting methods Progressive dopaminergic cell loss in the pars compacta
(such as vivo vs. post-mortem samples) used in the vari- (SNpc) is the hallmark of PD and has been studied in detail
ous studies. Tissues obtained from the GI tract and sali- histologically (Jellinger 2012). Unlike conventional mag-
vary glands showed a better sensitivity and specificity for netic resonance imaging (MRI) acquired with a 1 or 1.5 T
total α-synuclein than did skin and or olfactory mucosa or scanner, ultra-high field 7 T (7 T) MRI, by providing bet-
bulb, the most promising tissue biomarker for PD, whereas ter spatial resolution and contrast, can now detect typical
phosphorylated α-synuclein resulted inadequate. This find- PD-related changes in SNpc morphology (Lehéricy et al.
ing is not surprising given that the gut could be a starting 2014) (Fig. 3). Neuropathologically, dopamine-containing
point for PD through which the disease may spread along neurons are distributed in calbindin-poor zones (nigrosomes)
the brain–gut axis (Mulak and Bonaz 2015). Some obser- and in the calbindin-positive region (matrix). Nigrosome-1
vations imply that α-synuclein pathology in peripheral indicates the area of maximal dopamine depletion and is

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1425

considered the most sensitive pathological marker of neuro- PD appear. Even if DaTSCAN is more sensitive than clini-
degeneration in PD (Damier et al. 1999). 7 T MRI displays cal observations, it lacks diagnostic specificity and cannot
a nigral hyperintensity overlapping the neuropathological distinguish PD from other synucleinopathies (MSA, DLB)
characterization of nigrosome 1 and is useful to distinguish or tauopathies (PSP) nor measure disease progression (Per-
patients with PD from controls owing to a loss of hyper- lmutter and Norris 2014). What research needs to do now is
intensity areas on T2*-weighted images, present also in to develop new tracers based on molecular imaging advances
patients with MSA with predominant parkinsonism and thus providing in vivo markers for detecting the underlying
PSP (Kim et al. 2016). This marker could reflect the loss of pathology in PD (Strafella et al. 2017).
melanized neurons and the increase in iron deposition in the Over 90% of PD patients show SN hyperechogenicity
SNc (Jellinger 2012). A metanalysis (including 10 studies) assessed by transcranial ultrasound compared with 10% of
conducted in recent years has established that hyperintensity controls. Given that this echo-feature remains stable during
on iron-sensitive dorsolateral nigral MRI sequences can pro- disease progression, it could act as a vulnerability factor for
vide excellent diagnostic accuracy. Its absence demonstrated PD during patients’ lifetime (Behnke et al. 2010). Because
an overall sensitivity and specificity of 97.7 and 94.6% (3 SN hyperechogenicity is rarely present in degenerative par-
and 7 T) and of 94.6 and 94.4% (3 T only) for PD vs. con- kinsonisms, it differentiates PD from MSA and PSP (Pilotto
trols, thus providing a marker for nigral pathology and for et al. 2015). Like every ultrasound technique, transcranial
a neurodegenerative form of parkinsonian disorders (Mahl- ultrasound is operator-dependent and in up to 10–20% of
knecht et al. 2017). people is also limited by a poor temporal bone window
Because the shape of the degenerated nigrosome-1 and its (Behnke et al. 2010).
surrounding structures in axial imaging on 3 T susceptibil- Meta-iodobenzylguanidine (MIBG) cardiac scintigraphy
ity-weighted MRI in PD has been compared to “a swallow is a technique used to evaluate post-ganglionic presynaptic
tail”, this imaging sign has been proposed as a potential cardiac sympathetic nerves in heart diseases through a false
diagnostic tool for nigral degeneration in PD (Schwarz et al. neurotransmitter analogous to norepinephrine. Given the
2014; Gao et al. 2016). early sympathetic nerve degeneration during PD, reduced
As tools for studying the pathophysiology of PD in vivo, cardiac MIBG uptake is a specific marker for Lewy bod-
functional neuroimaging techniques (fMRI) have promising ies in the autonomic nervous system (Lucio et al. 2013).
potential. Various methods for analysing data for subjects This imaging technique is also useful for differentiating PD
in the resting state (awake without performing any particu- from MSA, a disease that spares postganglionic sympathetic
lar task) have been applied to detect functional connectiv- nerves, but pathologically damages preganglionic neurons.
ity abnormalities in PD. Dopamine depletion causes a cer- The MIBG technique can also discriminate between PSP and
ebral connectivity remapping across the whole brain (better DLB (Lucio et al. 2013). MIBG uptake in the parotid and
identified through “network-based” and “graph analysis” submandibular glands is significantly lower in PD patients
techniques), between specific regions (as shown by “seed- than in controls (Haqparwar et al. 2016).
based” and “effective connectivity” methods) and regional
anomalies (detected by analysing “regional homogeneity”
and “amplitude of low-frequency fluctuations”) (Tahmasian Genetics
et al. 2015). Drawbacks currently limiting resting-state fMRI
as a biomarker for PD include heterogeneous methods and PD has a multifactorial aetiology including lifestyle, envi-
insufficient published works using similar methods. ronment and genetics and, although only about 10% of
Ample information on dopaminergic denervation comes patients report a positive family history, at least 30% of the
from studies using radiotracer imaging in target structures risk of developing this disorder depends directly on genetic
(striatum). These imaging techniques are currently used in factors. Genome-wide association studies have identified at
research and clinical settings to improve the diagnostic accu- least 28 genetic risk loci associated with PD (Taymans et al.
racy for PD in positron emission tomography (PET) and in 2017). Few cases of PD are determined by a pure monogenic
single-photon emission CT (SPECT). The most commonly mutation, and also in these patients, PD development cannot
used radioligands are substances contained in presynaptic be taken for granted given the possible low penetrance and
terminals, such as dopa decarboxylase (18F-fluorodopa age-dependent variable expressivity. Age at onset remains
PET), vesicular monoamine transporter type 2 (11C-dihy- highly variable even in some carriers of high-penetrance
drotetrabenazine) and dopamine transporter (DAT) (123I-β- mutations (Gasser 2009), such as the α-synuclein gene
CIT SPECT and 99mTcTRODAT-1). DAT density has been (SNCA) mutation, recognized as a rare cause of autoso-
used as an imaging biomarker for diagnosing PD since 2011, mal dominantly inherited PD. Leucine-rich repeat kinase
because decreased DAT uptake, expressing nigrostriatal 2 (LRRK2) is another gene of particular interest because
dopamine neuron loss, is evident before motor features of it induces pleomorphic effects, shows high phenotypic

13

1426 I. Cova, A. Priori

variability among subjects, is also involved in sporadic dis- amplitude at flash ERG, a reduced P50 amplitude and a
ease forms, and is the most common genetic cause of PD in delayed latency at pattern ERG. Visual evoked potentials in
the world (Taymans et al. 2017). LRKK2 could be assayed response to coloured stimuli are also affected in PD; some
in tissues and biofluids relevant to disease and its expres- patient’s visual evoked potentials also show a decreased
sion and phosphorylation levels could have potential as a amplitude along with an increased latency for all chromatic
PD-related biomarker (Taymans et al. 2017). stimuli, above all for blue–yellow horizontal gratings, as
Some monogenic and all sporadic forms of PD share well as an increased P100 latency to a checkboard stimulus.
the same pathological mechanism, namely intraneuronal This last finding outlined a delay in visual processing at a
accumulation of α-synuclein aggregates in Lewy bodies certain visual system level, maybe involving the choliner-
and Lewy neuritis in the SNpc and striatum. Mutated genes gic system (Armstrong 2015). No current evidence suggests
contribute to the pathogenesis of PD through various mecha- that these neurophysiological biomarkers will accelerate the
nisms (some still unknown): some by producing misfolded diagnosis of PD.
proteins (α-synuclein gene duplications, triplications and Subthalamic nucleus local field potentials recorded from
point mutations), others due to dysregulated mitochondrial electrodes implanted intraoperatively in the basal ganglia
homeostasis (Parkin, DJ1, PINK1 and FBXO7 mutations) disclosed oscillations in the β frequency range, the range
or to altered lysosomal degradation (ATP13A2 and GBA that levodopa and deep brain stimulation (DBS) decrease
mutations). Other possible candidate biomarkers for PD while motor symptoms improve (Priori et al. 2004). Some
include the aforementioned misfolded proteins, e.g., SNCA evidence suggests that this β activity, as well as β phase
gene-encoded α-synuclein (Miller and O’Callaghan 2015) coupling to high-frequency oscillation amplitudes is a spe-
(Fig. 4). cific biomarker for the parkinsonian condition, (Wang et al.
2016), but few studies have investigated these patterns in
healthy controls (Swann et al. 2015).
Neurophysiological biomarkers

In neurophysiological studies conducted in the 1980s,

researchers describe defects in visual processing involv-
ing dopamine neurons including altered electroretinogram
(ERG) waves in PD patients: for example, reduced ‘b’ wave

Fig. 2  An abnormal microbiota

in PD (in favour of proinflam-
matory species) can lead to
immune dysregulation and
intestinal nervous system
inflammation and consequently
to a dysfunctional intestinal bar-
rier: α-synuclein pathology can
thus spread through the vagal
nerve from peripheral auto-
nomic neurons in the gastroin-
testinal apparatus to the central
nervous system. From Perez-
Pardo (https​://doi.org/10.1016/j.
ejpha​r.2017.05.042) 2017 with
permission from Elsevier

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1427

Fig. 3  7-Tesla magnetic resonance imaging scan (MRI) scan show- whereas it is not visible in the PD patient, in whom shape analysis
ing mesencephalic anatomy in a control (C) and a patient with Par- shows undulation in the anterolateral perimeter of the subthalamic
kinson’s disease (PD). Nigrosome-1 (white arrowheads) appears as a nucleus (arrow). From Lehéricy et  al. (https​://doi.org/10.1002/
hyperintense pocket (due to a low iron content) in the control subject mds.26043​), 2014 with permission from John Wiley and Sons

Omics information about the status of high-energy phosphates, and

those containing phosphorus (31P MRS), reflecting intracel-
During continued research into PD, the study of disease lular energy (Rango 2015). The extreme heterogeneity in
networks and molecular pathways through a bioinformatics these studies in number of enrolled patients and techniques
approach provides new potential ‘Omics’ biomarkers. used makes MRS, for the time being, only a future candidate
Omics testing can be applied to various biological media in PD biomarker scenery (Fig. 4).
(tissue, CSF, blood, urine) and aims to analyse functional
molecules such as proteins (proteomics), DNA/RNA
(genomics), lipids (lipidomics), carbohydrates (glycom-
ics) and metabolites (metabolomics), interactions between
molecules (interactomics), potentially involved in pathways
that are associated with dopaminergic neurodegeneration
and subsequently PD (Redenšek et al. 2018).
Metabolomics refers to the study of quantitative data on
cellular processes, molecular connections and metabolic
pathways appearing in complex diseases such as neuro-
degenerative disorders. A recent review (Kori et al. 2016)
reported that to date 54 metabolites have been studied as
biomarkers for PD. Some of these metabolites were found
disease-specific and others matched also with other neu-
rodegenerative diseases such as Alzheimer’s disease and
amyotrophic lateral sclerosis. A more specific set of differ-
ent system biomarkers should be studied for PD diagnosis
or prognosis or both.
In vivo brain metabolomics in PD can be studied noninva-
sively by magnetic resonance spectroscopy (MRS) measured
in specified small tissue volumes in the brain; this technique
is useful for increasing insight into the pathophysiological
mechanisms underlying PD and may produce biomarkers for Fig. 4  Genetic risk factors for Parkinson’s disease. Parkinson’s dis-
metabolic dysfunctions reflecting irreversible neuronal dam- ease (PD) has a multifactorial genesis and, although only about 10%
age (Davie 1998). This technique is based on the behaviour of patients have a family history, at least 30% of the risk of PD devel-
oping depends on genetic factors, which differ in strength and allele
of specific nuclei within a magnetic field and on the princi-
frequencies. The size of the bubbles corresponds to population; the
ple that resonant frequencies depend on the chemical envi- size of the rings corresponds to allele frequencies; inheritance is
ronment around the nuclei. Clinical studies mainly focused shown by colours (blue = dominant, yellow = recessive, green = risk
on metabolites containing hydrogen (H MRS), providing loci). From Gasser (https​://doi.org/10.3233/JPD-14050​7) 2015 with
permission of IOS Press

13

1428 I. Cova, A. Priori

Inflammation Given that the diagnosis of PD is still based on clinical find-

ings, a single test seems unlikely to satisfy all the functions
Animal and human studies supported the role of inflamma- required by a reliable biomarker for PD. Current research
tion in PD, due to the abnormal protein misfolding, which now aims to strengthen the diagnosis of PD by combining
can activate a proinflammatory state involving both Th1 and various clinical or non-clinical biomarkers or both. A single
Th2 responses (Mosley et al. 2012) and microglial activa- biomarker could play one or multiple roles (preclinical, diag-
tion (Qian and Flood 2008). This inflammatory activation nostic, or prognostic) thus providing variable clinometric
depends also on interactions between aging, environmental properties. A reliable putative biomarker should be repro-
risk factors and genetic factors. Dopaminergic drugs can ducible and replicable and take into account possible con-
modify the biological characteristics of T lymphatic cells founding from sociodemographic or other factors (McGhee
dopamine receptors, by inhibiting the release of cytotoxic et al. 2013). A feature that makes discovering a biomarker
mediators involved in PD development. In their review illus- an even more complex undertaking is the clinical heteroge-
trating immune dysregulation and inflammation response neity in PD from the early preclinical to the more advanced
(Chen et al. 2016) Chen et al. identified several potential PD disease stages. Aggregation of certain symptoms in specific
biomarkers: immunological genes (such as some human leu- clusters (expressed by different phenotypic PD traits) may
kocyte antigen genes), PD-related genes (directly influencing reflect degeneration in distinct neural pathways (Di Bat-
α-synuclein accumulation), PD-related microRNAs (which tista et al. 2018), a pathophysiological change that accounts
modulate PD genes at a post-transcriptional level), specific for different expression and also disease progression. The
antibodies (such as elongation factor 1-alpha-1 and poly (A) variability in findings from biomarker studies might reflect
binding protein cytoplasmic-3), inflammatory factors (such recruiting a non-uniform PD population. Heterogeneity in
as cytokines). Increased interleukin-6 and -8 levels indicate a PD could imply the need to seek specific diagnostic bio-
risk of PD (Polivka et al. 2016); higher levels of soluble TNF markers for variable PD traits, and consequently, different
receptor-1 were linked to early onset PD (Scalzo et al. 2010). theragnostic markers. Research efforts to this aim may be
Among the various available PET radiotracers for imag- very beneficial when research develops neuroprotective ther-
ing microglia, the most currently used is the 18 kDa trans- apies to slow or halt the progression of the disease. Despite
locator protein (TSPO), overexpressed on the outer mito- advances in our insight into the pathogenesis and pathophys-
chondrial membrane of the activated microglia, but usually iology underlying PD, today the neurologist has no 100%
poorly expressed in healthy brains (Gerhard 2016). Previ- reliable clinical marker for use in clinical practice and still
ous studies showed how microglia activation patterns dif- needs to rely on clinical expertise to diagnose PD. This is
fered in distinct parkinsonian syndromes according to how the most pressing problem on which biomarker researchers
neuropathologic changes are distributed in these disorders, need to focus their efforts in the coming years.
suggesting a role of a disease specific marker. Longitudinal
studies are needed to estimate the association between bio- Compliance with ethical standards 
marker evidence for microglial activation with and clinical
disease progression in PD. Whether and if so how microglial Conflict of interest  The authors declare no conflicts of interest.
activation relates to neurodegeneration in PD is an interest-
Open Access  This article is distributed under the terms of the Crea-
ing question that deserves future research. tive Commons Attribution 4.0 International License (http://creat​iveco​
mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribu-
tion, and reproduction in any medium, provided you give appropriate
Conclusion credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made.

Identifying a successful biomarker depends inevitably on

fully understanding the pathophysiology underlying the
disease. In PD, despite remarkable advances in our insight References
into the responsible mechanisms, the aetiology remains
Abbott RD, Petrovitch H, White LR et al (2001) Frequency of bowel
unknown. This drawback makes the search for diagnostic movements and the future risk of Parkinson’s disease. Neurol-
or preclinical biomarkers especially complex to achieve. ogy 57:456–462
Intense research efforts aim now to find multiple biomarkers Abbott RD, Ross GW, White LR et al (2005) Excessive daytime sleepi-
to detect PD as early as possible and to distinguish it from ness and subsequent development of Parkinson disease. Neurol-
ogy 65:1442–1446. https​://doi.org/10.1212/01.wnl.00001​83056​
other parkinsonisms. Although this evaluative review sum- .89590​.0d
marises research that takes the field ahead, given the abun- Abd-Elhadi S, Basora M, Vilas D et  al (2016) Total α-synuclein
dant research it inevitably leaves some studies unmentioned. levels in human blood cells, CSF, and saliva determined by a

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1429

lipid-ELISA. Anal Bioanal Chem 408:7669–7677. https​://doi. Chung EJ, Kim SJ (2015) Metaiodobenzylguanidine myocardial scin-
org/10.1007/s0021​6-016-9863-7 tigraphy in Lewy body-related disorders: a literature review. J
Al-Qassabi A, Fereshtehnejad S-M, Postuma RB (2017) Sleep dis- Mov Disord 8:55–66. https​://doi.org/10.14802​/jmd.15015​
turbances in the prodromal stage of Parkinson disease. Curr Claassen DO, Josephs KA, Ahlskog JE et  al (2010) REM sleep
Treat Options Neurol 19:22. https​: //doi.org/10.1007/s1194​ behavior disorder preceding other aspects of synucleinopathies
0-017-0458-1 by up to half a century. Neurology 75:494–499. https​://doi.
Andersen AD, Binzer M, Stenager E, Gramsbergen JB (2016) Cerebro- org/10.1212/WNL.0b013​e3181​ec7fa​c
spinal fluid biomarkers for Parkinson’s disease—a systematic Clairembault T, Leclair-Visonneau L, Coron E et al (2015) Structural
review. Acta Neurol Scand. https​://doi.org/10.1111/ane.12590​ alterations of the intestinal epithelial barrier in Parkinson’s dis-
Armstrong RA (2015) Oculo-visual dysfunction in Parkinson’s disease. ease. Acta Neuropathol Commun 3:12. https​://doi.org/10.1186/
J Parkinson’s Dis 5:715–726. https:​ //doi.org/10.3233/JPD-15068​ s4047​8-015-0196-0
6 Damier P, Hirsch EC, Agid Y, Graybiel AM (1999) The substantia
Behnke S, Schröder U, Berg D (2010) Transcranial sonography in the nigra of the human brain. II. Patterns of loss of dopamine-
premotor diagnosis of Parkinson’s disease. Int Rev Neurobiol containing neurons in Parkinson’s disease. Brain J Neurol
90:93–106. https​://doi.org/10.1016/S0074​-7742(10)90007​-X 122(Pt 8):1437–1448
Berg D, Lang AE, Postuma RB et al (2013) Changing the research Davie C (1998) The role of spectroscopy in parkinsonism. Mov
criteria for the diagnosis of Parkinson’s disease: obstacles Disord Off J Mov Disord Soc 13:2–4. https​://doi.org/10.1002/
and opportunities. Lancet Neurol 12:514–524. https​: //doi. mds.87013​0104
org/10.1016/S1474​-4422(13)70047​-4 Davis JW, Grandinetti A, Waslien CI et al (1996) Observations on
Berg D, Postuma RB, Adler CH et al (2015) MDS research criteria for serum uric acid levels and the risk of idiopathic Parkinson’s
prodromal Parkinson’s disease. Mov Disord Off J Mov Disord disease. Am J Epidemiol 144:480–484
Soc 30:1600–1611. https​://doi.org/10.1002/mds.26431​ Del Din S, Godfrey A, Mazzà C et al (2016) Free-living monitoring
Biomarkers Definitions Working Group (2001) Biomarkers and sur- of Parkinson’s disease: lessons from the field: wearable tech-
rogate endpoints: preferred definitions and conceptual frame- nology for Parkinson’s disease. Mov Disord 31:1293–1313.
work. Clin Pharmacol Ther 69:89–95. https​://doi.org/10.1067/ https​://doi.org/10.1002/mds.26718​
mcp.2001.11398​9 Di Battista ME, Cova I, Rubino A et al (2018) Intercepting Parkin-
Boeve BF, Dickson DW, Olson EJ et al (2007) Insights into REM son disease non-motor subtypes: a proof-of-principle study
sleep behavior disorder pathophysiology in brainstem-pre- in a clinical setting. J Neurol Sci 388:186–191. https​: //doi.
dominant Lewy body disease. Sleep Med 8:60–64. https​://doi. org/10.1016/j.jns.2018.03.024
org/10.1016/j.sleep​.2006.08.017 Diederich NJ, Pieri V, Hipp G et al (2010) Discriminative power of
Borgonovo J, Allende-Castro C, Laliena A et al (2017) Changes in different nonmotor signs in early Parkinson’s disease. A case–
neural circuitry associated with depression at pre-clinical, pre- control study. Mov Disord Off J Mov Disord Soc 25:882–887.
motor and early motor phases of Parkinson’s disease. Parkinson- https​://doi.org/10.1002/mds.22963​
ism Relat Disord 35:17–24. https​://doi.org/10.1016/j.parkr​eldis​ Doty RL (2008) The olfactory vector hypothesis of neurodegen-
.2016.11.009 erative disease: is it viable? Ann Neurol 63:7–15. https​://doi.
Braak H, Del Tredici K, Rüb U et al (2003) Staging of brain pathol- org/10.1002/ana.21327​
ogy related to sporadic Parkinson’s disease. Neurobiol Aging Doty RL, Shaman P, Applebaum SL et al (1984) Smell identification
24:197–211 ability: changes with age. Science 226:1441–1443
Busenbark KL, Huber SJ, Greer G et al (1992) Olfactory function in Drossman DA, Dumitrascu DL (2006) Rome III: new standard for
essential tremor. Neurology 42:1631–1632 functional gastrointestinal disorders. J Gastrointest Liver Dis
Caranci G, Piscopo P, Rivabene R et al (2013) Gender differences JGLD 15:237–241
in Parkinson’s disease: focus on plasma α-synuclein. J Neural Duda JE (2010) Olfactory system pathology as a model of Lewy
Transm 120:1209–1215. https​://doi.org/10.1007/s0070​2-013- neurodegenerative disease. J Neurol Sci 289:49–54. https​://
0972-6 (Vienna Austria 1996) doi.org/10.1016/j.jns.2009.08.042
Cersosimo MG (2015) Gastrointestinal biopsies for the diagnosis of Duran R, Barrero FJ, Morales B et al (2010) Plasma alpha-synuclein
alpha-synuclein pathology in Parkinson’s disease. Gastroenterol in patients with Parkinson’s disease with and without treat-
Res Pract. https​://doi.org/10.1155/2015/47604​1 ment. Mov Disord Off J Mov Disord Soc 25:489–493. https​://
Chaudhuri KR, Schapira AHV (2009) Non-motor symptoms of Par- doi.org/10.1002/mds.22928​
kinson’s disease: dopaminergic pathophysiology and treat- El-Agnaf OMA, Salem SA, Paleologou KE et al (2006) Detection of
ment. Lancet Neurol 8:464–474. https​://doi.org/10.1016/S1474​ oligomeric forms of alpha-synuclein protein in human plasma
-4422(09)70068​-7 as a potential biomarker for Parkinson’s disease. FASEB
Chaudhuri KR, Odin P, Antonini A, Martinez-Martin P (2011) Parkin- J Off Publ Fed Am Soc Exp Biol 20:419–425. https​: //doi.
son’s disease: the non-motor issues. Parkinsonism Relat Disord org/10.1096/fj.03-1449c​om
17:717–723. https​://doi.org/10.1016/j.parkr​eldis​.2011.02.018 Foulds PG, Diggle P, Mitchell JD et al (2013) A longitudinal study
Chauhan A, Jeans AF (2015) Is Parkinson’s disease truly a prion-like on α-synuclein in blood plasma as a biomarker for Parkinson’s
disorder? An appraisal of current evidence. Neurol Res Int. https​ disease. Sci Rep 3:2540. https​://doi.org/10.1038/srep0​2540
://doi.org/10.1155/2015/34528​5 Freeman WM, Bixler GV, Brucklacher RM et al (2010) A multi-
Chen L, Mo M, Li G et al (2016) The biomarkers of immune dysregu- step validation process of biomarkers for preclinical drug
lation and inflammation response in Parkinson disease. Transl development. Pharmacogenomics J 10:385–395. https​://doi.
Neurodegener 5:16. https​://doi.org/10.1186/s4003​5-016-0063-3 org/10.1038/tpj.2009.60
Cheng H-C, Ulane CM, Burke RE (2010) Clinical progression in Gagnon JF, Bédard MA, Fantini ML et al (2002) REM sleep behavior
Parkinson disease and the neurobiology of axons. Ann Neurol disorder and REM sleep without atonia in Parkinson’s disease.
67:715–725. https​://doi.org/10.1002/ana.21995​ Neurology 59:585–589
Chen-Plotkin AS, Hu WT, Siderowf A et al (2011) Plasma epidermal Gao P, Zhou P-Y, Wang P-Q et al (2016) Universality analysis of
growth factor levels predict cognitive decline in Parkinson dis- the existence of substantia nigra “swallow tail” appearance of
ease. Ann Neurol 69:655–663. https:​ //doi.org/10.1002/ana.22271​

13

1430 I. Cova, A. Priori

non-Parkinson patients in 3 T SWI. Eur Rev Med Pharmacol biomarker cohort. Mov Disord Off J Mov Disord Soc 31:924–
Sci 20:1307–1314 932. https​://doi.org/10.1002/mds.26613​
Gasser T (2009) Genomic and proteomic biomarkers for Parkin- Kawasaki I, Baba T, Takeda A, Mori E (2016) Loss of awareness of
son disease. Neurology 72:S27–S31. https​://doi.org/10.1212/ hyposmia is associated with mild cognitive impairment in Par-
WNL.0b013​e3181​98e05​4 kinson’s disease. Parkinsonism Relat Disord 22:74–79. https​://
Gerhard A (2016) TSPO imaging in parkinsonian disorders. Clin Transl doi.org/10.1016/j.parkr​eldis​.2015.11.015
Imaging 4:183–190. https​://doi.org/10.1007/s4033​6-016-0171-1 Kelly VE, Johnson CO, McGough EL et al (2015) Association of
Greene JG (2014) Causes and consequences of degeneration of the cognitive domains with postural instability/gait disturbance in
dorsal motor nucleus of the vagus nerve in Parkinson’s disease. Parkinson’s disease. Parkinsonism Relat Disord 21:692–697.
Antioxid Redox Signal 21:649–667. https​://doi.org/10.1089/ https​://doi.org/10.1016/j.parkr​eldis​.2015.04.002
ars.2014.5859 Kim J-M, Jeong H-J, Bae YJ et al (2016) Loss of substantia nigra
Haehner A, Boesveldt S, Berendse HW et al (2009) Prevalence of smell hyperintensity on 7 T MRI of Parkinson’s disease, multiple
loss in Parkinson’s disease—a multicenter study. Parkinsonism system atrophy, and progressive supranuclear palsy. Parkin-
Relat Disord 15:490–494. https​://doi.org/10.1016/j.parkr​eldis​ sonism Relat Disord 26:47–54. https​://doi.org/10.1016/j.parkr​
.2008.12.005 eldis​.2016.01.023
Hall S, Surova Y, Öhrfelt A et al (2015) CSF biomarkers and clinical Knudsen K, Krogh K, Østergaard K, Borghammer P (2017) Consti-
progression of Parkinson disease. Neurology 84:57–63. https​:// pation in Parkinson’s disease: subjective symptoms, objective
doi.org/10.1212/WNL.00000​00000​00109​8 markers, and new perspectives: constipation in Parkinson’s
Haqparwar J, Pepe A, Fassbender K et al (2016) Reduced MIBG accu- disease. Mov Disord 32:94–105. https​: //doi.org/10.1002/
mulation of the parotid and submandibular glands in idiopathic mds.26866​
Parkinson’s disease. Parkinsonism Relat Disord. https​://doi. Koehler NKU, Stransky E, Meyer M et al (2015) Alpha-synuclein
org/10.1016/j.parkr​eldis​.2016.10.011 levels in blood plasma decline with healthy aging. PLoS One
Hawkes CH, Shephard BC, Daniel SE (1999) Is Parkinson’s disease 10:e0123444. https​://doi.org/10.1371/journ​al.pone.01234​44
a primary olfactory disorder? QJM Mon J Assoc Physicians Kori M, Aydın B, Unal S et al (2016) Metabolic biomarkers and neu-
92:473–480 rodegeneration: a pathway enrichment analysis of Alzheimer’s
Heimer L, Van Hoesen GW, Trimble M, Zahm DS (2008) Cooperation disease, Parkinson’s disease, and amyotrophic lateral sclero-
and competition of macrosystem outputs. Anatomy of neuropsy- sis. Omics J Integr Biol 20:645–661. https​://doi.org/10.1089/
chiatry. Elsevier, Amsterdam, pp 101–139 omi.2016.0106
Heintz-Buschart A, Pandey U, Wicke T et al (2017) The nasal and Lee A, Gilbert RM (2016) Epidemiology of Parkinson disease. Neu-
gut microbiome in Parkinson’s disease and idiopathic rapid eye rol Clin 34:955–965. https​://doi.org/10.1016/j.ncl.2016.06.012
movement sleep behavior disorder: nose and gut microbiome in Lee PH, Lee G, Park HJ et al (2006) The plasma alpha-synuclein
PD and iRBD. Mov Disord. https​://doi.org/10.1002/mds.27105​ levels in patients with Parkinson’s disease and multiple sys-
Higgins PDR, Johanson JF (2004) Epidemiology of constipation in tem atrophy. J Neural Transm 113:1435–1439. https​: //doi.
North America: a systematic review. Am J Gastroenterol 99:750– org/10.1007/s0070​2-005-0427-9 (Vienna Austria 1996)
759. https​://doi.org/10.1111/j.1572-0241.2004.04114​.x Lee JM, Derkinderen P, Kordower JH et al (2017) The search for
Hobson DE, Lang AE, Martin WRW et  al (2002) Excessive day- a peripheral biopsy indicator of α-synuclein pathology for
time sleepiness and sudden-onset sleep in Parkinson disease: Parkinson disease. J Neuropathol Exp Neurol. https​: //doi.
a survey by the Canadian Movement Disorders Group. JAMA org/10.1093/jnen/nlw10​3
287:455–463 Lehéricy S, Bardinet E, Poupon C et al (2014) 7 T magnetic reso-
Hong Z, Shi M, Chung KA et al (2010) DJ-1 and alpha-synuclein nance imaging: a closer look at substantia nigra anatomy
in human cerebrospinal fluid as biomarkers of Parkinson’s dis- in Parkinson’s disease. Mov Disord Off J Mov Disord Soc
ease. Brain J Neurol 133:713–726. https​://doi.org/10.1093/brain​ 29:1574–1581. https​://doi.org/10.1002/mds.26043​
/awq00​8 LeWitt P, Schultz L, Auinger P et al (2011) CSF xanthine, homo-
Iranzo A, Lomeña F, Stockner H et al (2010) Decreased striatal dopa- vanillic acid, and their ratio as biomarkers of Parkinson’s dis-
mine transporter uptake and substantia nigra hyperechogenicity ease. Brain Res 1408:88–97. https​://doi.org/10.1016/j.brain​
as risk markers of synucleinopathy in patients with idiopathic res.2011.06.057
rapid-eye-movement sleep behaviour disorder: a prospective Li Q-X, Mok SS, Laughton KM et al (2007) Plasma alpha-synuclein
study [corrected]. Lancet Neurol 9:1070–1077. https​://doi. is decreased in subjects with Parkinson’s disease. Exp Neurol
org/10.1016/S1474​-4422(10)70216​-7 204:583–588. https​://doi.org/10.1016/j.expne​urol.2006.12.006
Iranzo A, Stockner H, Serradell M et al (2014) Five-year follow-up of Liepelt-Scarfone I, Behnke S, Godau J et al (2011) Relation of risk
substantia nigra echogenicity in idiopathic REM sleep behavior factors and putative premotor markers for Parkinson’s disease.
disorder. Mov Disord Off J Mov Disord Soc 29:1774–1780. https​ J Neural Transm 118:579–585. https​://doi.org/10.1007/s0070​
://doi.org/10.1002/mds.26055​ 2-010-0553-x (Vienna Austria 1996)
Jellinger KA (2012) Neuropathology of sporadic Parkinson’s disease: Lin X, Cook TJ, Zabetian CP et al (2012) DJ-1 isoforms in whole
evaluation and changes of concepts. Mov Disord Off J Mov Dis- blood as potential biomarkers of Parkinson disease. Sci Rep
ord Soc 27:8–30. https​://doi.org/10.1002/mds.23795​ 2:954. https​://doi.org/10.1038/srep0​0954
Jennings D, Siderowf A, Stern M et al (2014) Imaging prodromal Par- Loddo G, Calandra-Buonaura G, Sambati L et al (2017) The treat-
kinson disease: the Parkinson associated risk syndrome Study. ment of sleep disorders in Parkinson’s disease: from research
Neurology 83:1739–1746. https​://doi.org/10.1212/WNL.00000​ to clinical practice. Front Neurol 8:42. https​://doi.org/10.3389/
00000​00096​0 fneur​.2017.00042​
Jiang H, Huang J, Shen Y et  al (2016) RBD and neurodegenera- Luan H, Liu L-F, Tang Z et al (2015) Elevated excretion of biopyrrin
tive diseases. Mol Neurobiol. https​://doi.org/10.1007/s1203​ as a new marker for idiopathic Parkinson’s disease. Parkinson-
5-016-9831-4 ism Relat Disord 21:1371–1372. https​://doi.org/10.1016/j.parkr​
Kang UJ, Goldman JG, Alcalay RN et  al (2016) The BioFIND eldis​.2015.09.009
study: characteristics of a clinically typical Parkinson’s disease

13
Diagnostic biomarkers for Parkinson’s disease at a glance: where are we? 1431

Lucio CG, Vincenzo C, Antonio R et al (2013) Neurological applica- Perlmutter JS, Norris SA (2014) Neuroimaging biomarkers for Parkin-
tions for myocardial MIBG scintigraphy. Nucl Med Rev Cent son disease: facts and fantasy. Ann Neurol 76:769–783. https​://
East Eur 16:35–41. https​://doi.org/10.5603/NMR.2013.0007 doi.org/10.1002/ana.24291​
Magdalinou N, Lees AJ, Zetterberg H (2014) Cerebrospinal fluid bio- Pilotto A, Yilmaz R, Berg D (2015) Developments in the role of tran-
markers in parkinsonian conditions: an update and future direc- scranial sonography for the differential diagnosis of parkinson-
tions. J Neurol Neurosurg Psychiatry 85:1065–1075. https​://doi. ism. Curr Neurol Neurosci Rep 15:43. https​://doi.org/10.1007/
org/10.1136/jnnp-2013-30753​9 s1191​0-015-0566-9
Mahlknecht P, Gasperi A, Willeit P et al (2016) Prodromal Parkin- Poewe W, Mahlknecht P (2012) Combined assessment of midbrain
son’s disease as defined per MDS research criteria in the general hyperechogenicity, hyposmia and motor asymmetry improves
elderly community. Mov Disord Off J Mov Disord Soc 31:1405– diagnostic accuracy in early Parkinson’s disease. Expert Rev
1408. https​://doi.org/10.1002/mds.26674​ Neurother 12:911–914. https​://doi.org/10.1586/ern.12.75
Mahlknecht P, Krismer F, Poewe W, Seppi K (2017) Meta-analysis of Polivka J, Polivka J, Krakorova K et  al (2016) Current status of
dorsolateral nigral hyperintensity on magnetic resonance imaging biomarker research in neurology. EPMA J 7:14. https​://doi.
as a marker for Parkinson’s disease: DNH on MRI as a marker for org/10.1186/s1316​7-016-0063-5
PD. Mov Disord 32:619–623. https:​ //doi.org/10.1002/mds.26932​ Postuma RB, Berg D (2016) Advances in markers of prodromal
Marras C, Rochon P, Lang AE (2002) Predicting motor decline and Parkinson disease. Nat Rev Neurol 12:622–634. https​://doi.
disability in Parkinson disease: a systematic review. Arch Neurol org/10.1038/nrneu​rol.2016.152
59:1724–1728 Postuma RB, Gagnon JF, Vendette M et al (2009) Quantifying the
Masters JM, Noyce AJ, Warner TT et al (2015) Elevated salivary risk of neurodegenerative disease in idiopathic REM sleep
protein in Parkinson’s disease and salivary DJ-1 as a potential behavior disorder. Neurology 72:1296–1300. https​: //doi.
marker of disease severity. Parkinsonism Relat Disord 21:1251– org/10.1212/01.wnl.00003​40980​.19702​.6e
1255. https​://doi.org/10.1016/j.parkr​eldis​.2015.07.021 Postuma RB, Berg D, Stern M et al (2015) MDS clinical diagnostic
McGhee DJM, Royle PL, Thompson PA et  al (2013) A sys- criteria for Parkinson’s disease. Mov Disord Off J Mov Disord
tematic review of biomarkers for disease progression Soc 30:1591–1601. https​://doi.org/10.1002/mds.26424​
in Parkinson’s disease. BMC Neurol 13:35. https ​ : //doi. Priori A, Foffani G, Pesenti A et al (2004) Rhythm-specific pharma-
org/10.1186/1471-2377-13-35 cological modulation of subthalamic activity in Parkinson’s
McKeith IG, Boeve BF, Dickson DW et  al (2017) Diagnosis and disease. Exp Neurol 189:369–379. https​://doi.org/10.1016/j.
management of dementia with Lewy bodies: fourth consensus expne​urol.2004.06.001
report of the DLB Consortium. Neurology 89:88–100. https​:// Qian L, Flood PM (2008) Microglial cells and Parkinson’s disease.
doi.org/10.1212/WNL.00000​00000​00405​8 Immunol Res 41:155–164. https​: //doi.org/10.1007/s1202​
Mehta SH, Adler CH (2016) Advances in biomarker research in Par- 6-008-8018-0
kinson’s disease. Curr Neurol Neurosci Rep 16:7. https​://doi. Qiang JK, Wong YC, Siderowf A et al (2013) Plasma apolipoprotein
org/10.1007/s1191​0-015-0607-4 A1 as a biomarker for Parkinson disease. Ann Neurol 74:119–
Merello M, Nouzeilles MI, Arce GP, Leiguarda R (2002) Accuracy 127. https​://doi.org/10.1002/ana.23872​
of acute levodopa challenge for clinical prediction of sustained Rajput AH, Voll A, Rajput ML et al (2009) Course in Parkinson
long-term levodopa response as a major criterion for idiopathic disease subtypes: a 39-year clinicopathologic study. Neurology
Parkinson’s disease diagnosis. Mov Disord Off J Mov Disord Soc 73:206–212. https​://doi.org/10.1212/WNL.0b013​e3181​ae7af​1
17:795–798. https​://doi.org/10.1002/mds.10123​ Rango M (2015) Parkinson’s disease: in vivo brain metabolomics by
Miller DB, O’Callaghan JP (2015) Biomarkers of Parkinson’s dis- MRS. Int Rev Neurobiol 122:81–94. https​://doi.org/10.1016/
ease: present and future. Metabolism 64:S40–S46. https​://doi. bs.irn.2015.05.007
org/10.1016/j.metab​ol.2014.10.030 Redenšek S, Dolžan V, Kunej T (2018) From genomics to omics
Moschos MM, Tagaris G, Markopoulos I et al (2011) Morphologic landscapes of Parkinson’s disease: revealing the molecu-
changes and functional retinal impairment in patients with Par- lar mechanisms. Omics J Integr Biol 22:1–16. https​: //doi.
kinson disease without visual loss. Eur J Ophthalmol 21:24–29 org/10.1089/omi.2017.0181
Mosley RL, Hutter-Saunders JA, Stone DK, Gendelman HE (2012) Reijnders JSAM, Ehrt U, Weber WEJ et  al (2008) A systematic
Inflammation and adaptive immunity in Parkinson’s dis- review of prevalence studies of depression in Parkinson’s dis-
ease. Cold Spring Harb Perspect Med 2:a009381. https​://doi. ease. Mov Disord Off J Mov Disord Soc 23:183–189. https​://
org/10.1101/cshpe​rspec​t.a0093​81 doi.org/10.1002/mds.21803​(quiz 313)
Mulak A, Bonaz B (2015) Brain-gut-microbiota axis in Parkinson’s Rickards H (2005) Depression in neurological disorders: Parkinson’s
disease. World J Gastroenterol 21:10609–10620. https​://doi. disease, multiple sclerosis, and stroke. J Neurol Neurosurg Psy-
org/10.3748/wjg.v21.i37.10609​ chiatry 76:i48–i52. https​://doi.org/10.1136/jnnp.2004.06042​6
Noyce AJ, Bestwick JP, Silveira-Moriyama L et al (2012) Meta-analy- Rietdijk CD, Perez-Pardo P, Garssen J et al (2017) Exploring Braak’s
sis of early nonmotor features and risk factors for Parkinson dis- hypothesis of Parkinson’s disease. Front Neurol. https​://doi.
ease. Ann Neurol 72:893–901. https:​ //doi.org/10.1002/ana.23687​ org/10.3389/fneur​.2017.00037​
O’Callaghan C, Lewis SJG (2017) Cognition in Parkinson’s disease. Rizzo G, Copetti M, Arcuti S et  al (2016) Accuracy of clinical
International review of neurobiology. Elsevier, Amsterdam, diagnosis of Parkinson disease: a systematic review and meta-
pp 557–583 analysis. Neurology 86:566–576. https​: //doi.org/10.1212/
Ogawa I, Saito Y, Saigoh K et al (2014) The significance of oxidized WNL.00000​00000​00235​0
DJ-1 protein (oxDJ-1) as a biomarker for Parkinson’s disease. Savica R, Carlin JM, Grossardt BR et al (2009) Medical records
Brain Nerve Shinkei Kenkyu No Shinpo 66:471–477 documentation of constipation preceding Parkinson disease:
Pagel JF (2009) Excessive daytime sleepiness. Am Fam Physician a case–control study. Neurology 73:1752–1758. https​://doi.
79:391–396 org/10.1212/WNL.0b013​e3181​c34af​5
Pereira PAB, Aho VTE, Paulin L et al (2017) Oral and nasal microbiota Scalzo P, Kümmer A, Bretas TL et al (2010) Serum levels of brain-
in Parkinson’s disease. Parkinsonism Relat Disord 38:61–67. derived neurotrophic factor correlate with motor impairment
https​://doi.org/10.1016/j.parkr​eldis​.2017.02.026 in Parkinson’s disease. J Neurol 257:540–545. https​: //doi.
org/10.1007/s0041​5-009-5357-2

13

1432 I. Cova, A. Priori

Scheperjans F (2017) The prodromal microbiome: the prodromal disease: does dopamine replacement therapy play a role? Cortex
microbiome. Mov Disord. https​://doi.org/10.1002/mds.27197​ 73:80–105. https​://doi.org/10.1016/j.corte​x.2015.08.005
Scheperjans F, Pekkonen E, Kaakkola S, Auvinen P (2015) Link- Takeda A, Baba T, Kikuchi A et al (2014) Olfactory dysfunction and
ing smoking, coffee, urate, and Parkinson’s disease—a role dementia in Parkinson’s disease. J Parkinson’s Dis 4:181–187.
for gut microbiota? J Parkinson’s Dis 5:255–262. https​://doi. https​://doi.org/10.3233/JPD-13027​7
org/10.3233/JPD-15055​7 Taymans J-M, Mutez E, Drouyer M et  al (2017) LRRK2 detec-
Schneider SA, Boettner M, Alexoudi A et al (2016) Can we use periph- tion in human biofluids: potential use as a Parkinson’s dis-
eral tissue biopsies to diagnose Parkinson’s disease? A review of ease biomarker? Biochem Soc Trans 45:207–212. https​://doi.
the literature. Eur J Neurol 23:247–261. https​://doi.org/10.1111/ org/10.1042/BST20​16033​4
ene.12753​ Torbey E, Pachana NA, Dissanayaka NNW (2015) Depression rating
Schwarz ST, Afzal M, Morgan PS et al (2014) The “swallow tail” scales in Parkinson’s disease: a critical review updating recent
appearance of the healthy nigrosome—a new accurate test of Par- literature. J Affect Disord 184:216–224. https:​ //doi.org/10.1016/j.
kinson’s disease: a case-control and retrospective cross-sectional jad.2015.05.059
MRI study at 3T. PLoS One 9:e93814. https​://doi.org/10.1371/ Tremlett H, Bauer KC, Appel-Cresswell S et al (2017) The gut micro-
journ​al.pone.00938​14 biome in human neurological disease: a review: gut microbiome.
Schwarzschild MA, Schwid SR, Marek K et al (2008) Serum urate as Ann Neurol 81:369–382. https​://doi.org/10.1002/ana.24901​
a predictor of clinical and radiographic progression in Parkinson van Dijk KD, Bidinosti M, Weiss A et al (2014) Reduced α-synuclein
disease. Arch Neurol 65:716–723. https​://doi.org/10.1001/archn​ levels in cerebrospinal fluid in Parkinson’s disease are unrelated
eur.2008.65.6.nct70​003 to clinical and imaging measures of disease severity. Eur J Neurol
Shi M, Zabetian CP, Hancock AM et al (2010) Significance and con- 21:388–394. https​://doi.org/10.1111/ene.12176​
founders of peripheral DJ-1 and alpha-synuclein in Parkinson’s Vivacqua G, Latorre A, Suppa A et al (2016) Abnormal salivary total
disease. Neurosci Lett 480:78–82. https:​ //doi.org/10.1016/j.neule​ and oligomeric alpha-synuclein in Parkinson’s disease. PLoS
t.2010.06.009 One 11:e0151156. https:​ //doi.org/10.1371/journa​ l.pone.015115​ 6
Slotnick S, Ding Y, Glazman S et al (2015) A novel retinal biomarker Wang Y, Shi M, Chung KA et al (2012) Phosphorylated α-synuclein
for Parkinson’s disease: quantifying the foveal pit with opti- in Parkinson’s disease. Sci Transl Med 4:121ra20. https​://doi.
cal coherence tomography. Mov Disord Off J Mov Disord Soc org/10.1126/scitr​anslm​ed.30025​66
30:1692–1695. https​://doi.org/10.1002/mds.26411​ Wang N, Gibbons CH, Lafo J, Freeman R (2013) α-Synuclein in cuta-
Stebbins GT, Goetz CG, Burn DJ et al (2013) How to identify tremor neous autonomic nerves. Neurology 81:1604–1610. https​://doi.
dominant and postural instability/gait difficulty groups with the org/10.1212/WNL.0b013​e3182​a9f44​9
movement disorder society unified Parkinson’s disease rating Wang DD, de Hemptinne C, Miocinovic S et al (2016) Subthalamic
scale: comparison with the unified Parkinson’s disease rating local field potentials in Parkinson’s disease and isolated dystonia:
scale: PIGD and the MDS-UPDRS. Mov Disord 28:668–670. an evaluation of potential biomarkers. Neurobiol Dis 89:213–
https​://doi.org/10.1002/mds.25383​ 222. https​://doi.org/10.1016/j.nbd.2016.02.015
Stern MB, Doty RL, Dotti M et al (1994) Olfactory function in Parkin- Weil RS, Schrag AE, Warren JD et al (2016) Visual dysfunction in
son’s disease subtypes. Neurology 44:266–268 Parkinson’s disease. Brain J Neurol. https:​ //doi.org/10.1093/brain​
Stewart T, Sui Y-T, Gonzalez-Cuyar LF et  al (2014) Cheek cell- /aww17​5
derived α-synuclein and DJ-1 do not differentiate Parkinson’s Welge-Lüssen A (2009) Ageing, neurodegeneration, and olfactory and
disease from control. Neurobiol Aging 35:418–420. https​://doi. gustatory loss. B-ENT 13(5 Suppl):129–132
org/10.1016/j.neuro​biola​ging.2013.08.008 Wenning GK, Shephard B, Hawkes C et al (1995) Olfactory func-
Strafella AP, Bohnen NI, Perlmutter JS et al (2017) Molecular imag- tion in atypical parkinsonian syndromes. Acta Neurol Scand
ing to track Parkinson’s disease and atypical parkinsonisms: new 91:247–250
imaging frontiers: molecular imaging in parkinsonian disorders. Xiao Q, Chen S, Le W (2014) Hyposmia: a possible biomarker of
Mov Disord 32:181–192. https​://doi.org/10.1002/mds.26907​ Parkinson’s disease. Neurosci Bull 30:134–140. https​://doi.
Swann NC, de Hemptinne C, Aron AR et al (2015) Elevated synchrony org/10.1007/s1226​4-013-1390-3
in Parkinson disease detected with electroencephalography: Zhan A, Mohan S, Tarolli C et  al (2018) Using smartphones and
elevated synchrony in PD. Ann Neurol 78:742–750. https​://doi. machine learning to quantify Parkinson disease severity: the
org/10.1002/ana.24507​ mobile Parkinson disease score. JAMA Neurol. https​://doi.
Swanson CR, Li K, Unger TL et al (2015) Lower plasma apolipoprotein org/10.1001/jaman​eurol​.2018.0809
A1 levels are found in Parkinson’s disease and associate with Zhu K, van Hilten JJ, Marinus J (2016) Course and risk factors for exces-
apolipoprotein A1 genotype: plasma A po A1 and APOA1 geno- sive daytime sleepiness in Parkinson’s disease. Parkinsonism Relat
type in PD. Mov Disord 30:805–812. https​://doi.org/10.1002/ Disord 24:34–40. https​://doi.org/10.1016/j.parkr​eldis​.2016.01.020
mds.26022​
Tahmasian M, Bettray LM, van Eimeren T et al (2015) A systematic
review on the applications of resting-state fMRI in Parkinson’s

13