Parkinson's Disease: Diagnostic Criteria, Biological Staging & Clinical Manifestations

The diagnosis of Parkinson's disease (PD) has undergone a remarkable transformation over the past decade. The 2015 MDS Clinical Diagnostic Criteria remain the gold standard for clinical practice, but 2024 marked a pivotal year with the emergence of two biological frameworks — the Neuronal α-Synuclein Disease Integrated Staging System (NSD-ISS) and the SynNeurGe classification — that redefine PD as a biologically anchored entity. Understanding both the clinical and biological diagnostic landscape is now essential for neurologists involved in patient care and clinical trial enrollment.

MDS Clinical Diagnostic Criteria (2015)

The Movement Disorder Society (MDS) Clinical Diagnostic Criteria, published by Postuma, Berg, and colleagues in 2015, represent the current international standard for diagnosing PD in both research and clinical settings⁽¹⁾. They replaced the long-used UK Parkinson's Disease Society Brain Bank (UKBB) criteria and were formally validated in 2018 across 8 centers and 626 patients⁽²⁾.

The criteria retain motor parkinsonism as the prerequisite, defined as bradykinesia in combination with rest tremor, rigidity, or both. Once parkinsonism is established, the diagnosis is refined through three categories of diagnostic features: supportive criteria (positive features increasing diagnostic confidence), red flags (features suggesting an alternative diagnosis, which must be counterbalanced by supportive criteria), and absolute exclusion criteria (which rule out PD regardless of other features).

Two levels of diagnostic certainty are defined: clinically established PD (maximizing specificity — requires ≥2 supportive criteria, no red flags, no exclusion criteria) and probable PD (balancing sensitivity and specificity — red flags allowed if counterbalanced by supportive criteria).

Validation by Postuma et al. (2018) demonstrated that probable PD criteria achieved 94.5% sensitivity and 88.5% specificity (overall accuracy 92.6%), significantly outperforming UKBB criteria, which had 89.2% sensitivity, 79.2% specificity, and 86.4% accuracy (all comparisons P < 0.05). The clinically established PD category captured 59.3% of PD patients with only 1.6% false-positive rate⁽²⁾.

A 2025 autopsy-confirmed study by Fox et al. further corroborated these criteria, finding that an excellent levodopa response and rest tremor were the most useful supportive criteria, while absolute exclusion criteria and red flags were more prevalent in atypical parkinsonism. Notably, supranuclear gaze palsy and rapid wheelchair requirement were found in >5% of pathologically confirmed PD cases — suggesting some red flags may need re-evaluation⁽³⁾.

Hoehn & Yahr Staging

The Hoehn & Yahr scale (1967) remains the most widely used clinical staging system for PD severity. Although not biologically anchored, it provides a practical shorthand for disease severity in clinical practice and trial enrollment:

Limitations: The Hoehn & Yahr scale is purely motor-focused and does not account for non-motor burden, which often drives disability and quality of life. Progression through stages is nonlinear, and the scale does not capture treatment-related complications (fluctuations, dyskinesia). The MDS has called for updated staging that integrates both motor and non-motor features alongside biological markers⁽⁴⁾.

The Biological Revolution: 2024 Frameworks

In February 2024, two landmark publications in The Lancet Neurology proposed fundamentally new ways to define and classify PD using in vivo biomarkers — particularly the α-synuclein seed amplification assay (SAA). A parallel MDS position statement provided institutional context. Together, these publications represent a paradigm shift analogous to the AT(N) framework in Alzheimer's disease.

NSD-ISS: Neuronal α-Synuclein Disease Integrated Staging System

The NSD-ISS, proposed by Simuni, Chahine, Poston, and colleagues (2024), redefines PD and dementia with Lewy bodies under a single biological umbrella: Neuronal α-Synuclein Disease (NSD). The framework rests on two biological anchors⁽⁵⁾:

S (Synuclein): Presence of pathological neuronal α-synuclein detected by CSF seed amplification assay (SAA) — the primary biological anchor. D (Dopaminergic): Evidence of dopaminergic neuronal dysfunction/degeneration, typically via DaT-SPECT — the second biological anchor. Clinical signs and functional impairment then determine staging from 0 to 6:

Validation in the PPMI, PASADENA, and SPARK cohorts (Dam et al., 2024) demonstrated that 93% of clinically diagnosed sporadic PD patients were SAA-positive (S+), consistent with NSD. Among sporadic PD, the stage distribution at diagnosis was: Stage 2B (25%), Stage 3 (63%), and Stage 4 (9%). Importantly, NSD-ISS stage at baseline predicted progression: median time to a clinically meaningful outcome was 8.3 years for Stage 2B, 5.9 years for Stage 3, and 2.4 years for Stage 4⁽⁶⁾.

SynNeurGe: A Biological Classification

The SynNeurGe classification, proposed by Höglinger, Adler, Berg, and colleagues (2024), takes a complementary approach: a three-component biological classification that accommodates the heterogeneity of PD⁽⁷⁾:

S (Syn): Presence or absence of pathological α-synuclein in tissues or CSF (via SAA, skin biopsy, or tissue IHC). N (Neur): Evidence of neurodegeneration via neuroimaging (DaT-SPECT, FDG-PET, cardiac MIBG). G (Ge): Documentation of pathogenic gene variants that cause or strongly predispose to PD (e.g., SNCA, LRRK2, GBA1, PRKN, PINK1). These three biological components are linked to a clinical component (C), defined by either a single high-specificity clinical feature or multiple lower-specificity features.

A key strength of the SynNeurGe system is its accommodation of S-negative genetic PD — some LRRK2 and GBA1 carriers develop clinical PD without detectable α-synuclein aggregates on current assays, and certain PRKN/PINK1 mutation carriers lack Lewy body pathology at autopsy. By incorporating a genetic component (G), SynNeurGe captures these biologically distinct forms of PD that would be missed by a synuclein-only definition⁽⁷⁾.

MDS Position Statement (Cardoso et al., 2024)

The MDS Position Statement, authored by Cardoso, Goetz, Mestre, and colleagues, provides an official institutional framework for interpreting these biological advances. Key principles include: a biological definition of PD should precede any staging system; staging should integrate both motor and non-motor clinical features with biological markers; any system must be validated against clinical outcomes and neuropathology; global consensus — including clinician, scientist, and patient input — is essential before adoption; and the clinical diagnosis of PD as currently practiced remains valid and necessary⁽⁴⁾.

Motor Manifestations

The cardinal motor features of PD — bradykinesia, rigidity, rest tremor, and postural instability — remain the clinical bedrock of diagnosis. The MDS-UPDRS Part III provides standardized motor assessment.

Bradykinesia is the mandatory feature, defined as slowness of movement with progressive reduction in speed and amplitude (decrement) on repetitive actions. It is the most specific motor feature and must be distinguished from simple slowness due to weakness, pain, or stiffness. Rigidity is increased resistance to passive movement (lead-pipe or cogwheel type), present throughout the range of motion. It should be distinguished from spasticity (velocity-dependent) and paratonia. Rest tremor — the classic 4–6 Hz "pill-rolling" tremor — is present in approximately 70% of PD patients at diagnosis and is the most specific supportive criterion. However, ~30% of patients are "atremulous," and tremor-predominant PD has a distinct prognosis (see below). Postural instability typically emerges later in disease and is assessed by the pull test. Early postural instability (<3 years) is a red flag for atypical parkinsonism.

Motor Subtypes

PD is clinically heterogeneous, and two major motor subtypes are widely recognized:

Tremor-dominant (TD): Characterized by prominent rest tremor with relatively mild bradykinesia and gait impairment. TD PD is associated with slower motor progression, later onset of cognitive decline, less dopaminergic degeneration on DaT-SPECT, and overall better prognosis. Postural instability/gait difficulty (PIGD): Characterized by prominent axial features — gait impairment, postural instability, freezing of gait, and falls. PIGD subtype is associated with faster motor and cognitive decline, earlier autonomic dysfunction, and greater cortical and cholinergic involvement. The ELLDOPA trial — which demonstrated dose-dependent motor improvement with levodopa in early PD — also captured the importance of subtype in levodopa response patterns⁽⁸⁾. The CALM-PD trial, comparing pramipexole versus levodopa as initial therapy, showed that while pramipexole reduced motor complications (28% vs 51%), levodopa provided superior UPDRS improvement (9.2 vs 4.5 points)⁽⁹⁾ — differences that may also relate to subtype distribution.

Non-Motor Manifestations

Non-motor symptoms (NMS) are now recognized as integral to PD, often more disabling than motor features, and frequently present years before motor onset. The prodromal non-motor phenotype — hyposmia, REM sleep behavior disorder, constipation, depression, and anxiety — is a cornerstone of biological staging and disease-modification trial enrollment.

Psychosis and hallucinations represent a critical non-motor milestone. Visual hallucinations (typically well-formed, non-threatening) affect 20–40% of PD patients and often precede cognitive decline. Pimavanserin, a selective 5-HT_2A inverse agonist, was the first drug FDA-approved specifically for PD psychosis based on the HARMONY trial, which demonstrated a significant reduction in psychosis relapse compared to placebo⁽¹⁰⁾. Quetiapine and clozapine remain alternatives when pimavanserin is insufficient, though clozapine requires regular CBC monitoring.

PD dementia develops in a substantial proportion of patients over the disease course. The EXPRESS trial established rivastigmine as the standard of care, demonstrating significant improvement on the ADAS-cog (2.1 vs −0.7 points, P<0.001) in 541 patients with PD dementia. GI side effects (nausea 29%, vomiting 17%) and worsening tremor (10%) were more common with rivastigmine⁽¹¹⁾.

Prodromal PD & Biomarkers

The concept of prodromal PD recognizes that neurodegeneration — and α-synuclein pathology — is present years to decades before the cardinal motor features emerge. The MDS Research Criteria for Prodromal PD (Berg et al., 2015; updated by Heinzel et al., 2019) use a Bayesian likelihood ratio approach, combining risk markers and prodromal markers to calculate the probability of having prodromal PD^(12,13).

Risk markers include: age, male sex, pesticide exposure, caffeine non-use, family history of PD, and genetic variants (LRRK2, GBA1 mutations). Prodromal markers include: polysomnography-confirmed RBD (likelihood ratio >100), hyposmia (LR 4–6), constipation, depression, excessive daytime somnolence, symptomatic hypotension, urinary dysfunction, erectile dysfunction, and abnormal DaT-SPECT. When the combined posterior probability exceeds 80%, prodromal PD is considered "probable."

Key Biomarkers

α-Synuclein Seed Amplification Assay (SAA): This is the transformative biomarker enabling the biological definition of PD. CSF SAA detects pathological α-synuclein aggregates with >95% sensitivity and >95% specificity for manifest PD in the PPMI cohort. The assay identifies PD pathology even in prodromal stages — including isolated RBD and asymptomatic LRRK2/GBA1 carriers who are SAA-positive. However, CSF collection is required (lumbar puncture), and blood-based SAA assays are under active development but not yet validated for clinical or research staging^(5,6).

DaT-SPECT (Dopamine Transporter Imaging): Detects nigrostriatal dopaminergic deficit. A normal DaT-SPECT is an absolute exclusion criterion in the MDS criteria. However, DaT-SPECT becomes abnormal only after ~50% of dopaminergic neurons are lost, limiting its utility in the earliest disease stages. In the NSD-ISS, DaT-SPECT serves as the "D" anchor^(1,5).

Cardiac MIBG Scintigraphy: Reduced myocardial MIBG uptake reflects cardiac sympathetic denervation and is a supportive criterion in the MDS-PD criteria. It helps distinguish PD/DLB from MSA and PSP, which typically have preserved cardiac innervation⁽¹⁾.

Skin Biopsy (Phosphorylated α-Synuclein): Dermal nerve fiber analysis for phosphorylated α-synuclein deposits is an emerging less-invasive alternative to CSF SAA. Preliminary data suggest comparable sensitivity, but standardized protocols and multi-center validation are still needed. The SynNeurGe classification includes skin biopsy as an endorsed method for the "S" component⁽⁷⁾.

Framework Comparison Table