Progressive Supranuclear Palsy: How It Differs From Parkinson's Disease, Why Falls Happen Early, and Why There Is No Approved Treatment Yet

Research-informed explainer — last updated April 12, 2026

Progressive supranuclear palsy (PSP) is a rare, progressive brain disease caused by abnormal tau protein accumulation that is consistently misdiagnosed as Parkinson's disease in its early stages — but the two conditions differ fundamentally in prognosis, treatment response, and the safety risks they pose, making accurate diagnosis essential from the very first fall. There is currently no approved disease-modifying therapy for PSP, but understanding the diagnosis improves care planning, fall prevention, and access to clinical trials.

This article draws on research from five movement disorder specialists. Joseph Jankovic, MD, Professor of Neurology and Director of the Parkinson's Disease Center at Baylor College of Medicine, was first author on the original NINDS-SPSP clinical diagnostic criteria for PSP (2,504 citations) — the landmark paper that defined how clinicians diagnose PSP — and has published extensively on Parkinson's disease features and differential diagnosis (5,567 citations). David Irwin, MD, Associate Professor of Neurology at the University of Pennsylvania, published the updated Movement Disorder Society clinical diagnosis criteria for PSP (2,253 citations) and authored a foundational review of tauopathies as clinicopathological entities (198 citations). Irene Litvan, MD, Tasch Endowed Professor at UC San Diego and Director of the Parkinson and Other Movement Disorders Center, published the preliminary NINDS neuropathological criteria for PSP (859 citations), co-authored the SIC Task Force appraisal of parkinsonian diagnostic criteria (1,018 citations), and identified the tau H1 haplotype as a major genetic risk factor for PSP (793 citations). Melissa Armstrong, MD, Professor at the University of Florida and Director of the Mangurian Clinical-Research Headquarters for Lewy Body Dementia, published the diagnostic criteria for corticobasal degeneration — PSP's closest mimic (1,856 citations) — and contributed a major JAMA Parkinson's disease diagnosis and treatment review (2,680 citations). James Bower, MD, Professor of Neurology at Mayo Clinic, published population-based incidence data on PSP and multiple system atrophy in Olmsted County (397 citations) and comprehensive parkinsonism incidence data (510 citations) that frame how rare PSP is relative to Parkinson's disease.

What PSP is and what causes it

PSP is a neurodegenerative disease classified as a primary tauopathy — it is caused by the abnormal aggregation of tau protein in specific brain regions, including the basal ganglia, brainstem, and cerebral cortex. Tau normally stabilizes microtubules inside neurons; in PSP, misfolded tau forms neurofibrillary tangles that disrupt neuronal function and eventually kill cells.

Litvan's preliminary neuropathological criteria established that the diagnosis of PSP at autopsy requires a specific density and distribution of neurofibrillary tangles in the basal ganglia and brainstem. The tau haplotype association study by Litvan and collaborators found that the H1 haplotype of the tau gene is significantly over-represented in PSP patients compared to controls — suggesting that the H1 haplotype creates a permissive environment for tau misfolding, though PSP is not considered a genetic disease in the conventional sense. Most cases are sporadic.

Bower's population-based Olmsted County study found that the incidence of PSP is approximately 5–6 per 100,000 people per year — roughly 40 times rarer than Parkinson's disease — which explains why it is unfamiliar to many primary care physicians and even some general neurologists.

How PSP differs from Parkinson's disease

The early symptoms of PSP — slowness, stiffness, balance problems — are indistinguishable from Parkinson's disease to the untrained eye. Yet the two diseases have fundamentally different trajectories, different examination findings, and different treatment responses.

Jankovic's extensive work on Parkinson's disease clinical features, as described in his widely cited 2008 review, characterizes Parkinson's as typically beginning unilaterally, with resting tremor, good response to levodopa, and a relatively preserved ability to look up and down. PSP, by contrast, presents bilaterally from the start, rarely involves tremor, responds poorly to levodopa, and — most distinctively — impairs voluntary vertical eye movements. The hallmark of PSP is supranuclear gaze palsy, specifically difficulty moving the eyes voluntarily downward, while the eyes can still be moved reflexively by the oculocephalic reflex (the "doll's eye" maneuver). This finding is rare in Parkinson's disease and virtually diagnostic of PSP or a related condition.

The updated Movement Disorder Society criteria for PSP, developed by Irwin and collaborators, define several clinical phenotypes of PSP beyond the classic Richardson syndrome presentation:

• PSP-Richardson syndrome (PSP-RS): The classic form — early falls, supranuclear gaze palsy, postural instability
• PSP-Parkinsonism (PSP-P): More asymmetric, with some levodopa response early on, making it especially likely to be confused with Parkinson's disease
• PSP-Progressive Gait Freezing (PSP-PGF): Predominantly gait freezing with minimal other features
• PSP-Corticobasal Syndrome (PSP-CBS): Overlapping with corticobasal degeneration features
• PSP-Frontal (PSP-F): Predominantly behavioral and cognitive

Armstrong's diagnostic criteria for corticobasal degeneration define the most important PSP mimic. Corticobasal syndrome involves asymmetric apraxia, alien limb phenomenon, and cortical sensory loss not characteristic of PSP, though some patients have pathological overlap. Getting this distinction right matters because prognosis and symptomatic treatments differ.

Why falls happen early and are so dangerous

In Parkinson's disease, falls typically occur years into the disease as postural reflexes gradually deteriorate. In PSP, falls — often backward — are a presenting symptom that can occur within the first year. This happens because PSP damages the brainstem nuclei and cerebellar connections that control automatic postural adjustments, causing a loss of righting reflexes before patients are aware they have a serious neurological condition.

Jankovic's NINDS-SPSP criteria identify early unprovoked falls — particularly falls in the first year of symptom onset — as a required feature for probable PSP, distinguishing it from Parkinson's disease where such falls would be unusual early in the course. Patients with PSP are at extremely high risk for serious fall-related injury: hip fracture, subdural hematoma, and skull fractures are common because the postural instability is severe and patients fall at full height without protective arm movements.

Physical therapy focused on fall prevention, use of a weighted walker (to reduce backward falls), and home modification are critical interventions as early as diagnosis.

Why levodopa usually does not help

The dopamine pathway damage in Parkinson's disease is relatively selective to the substantia nigra, and dopamine replacement with levodopa restores function. In PSP, the damage is more widespread — including basal ganglia circuits beyond the substantia nigra, brainstem nuclei, and cortical regions — and the tau pathology is not responsive to dopamine.

Litvan's SIC Task Force criteria document that at most 25–30% of PSP-Parkinsonism patients show meaningful transient improvement on levodopa, typically early in the course, and this response fades. This limited response to levodopa is both diagnostically informative and practically important: patients who are prescribed levodopa for a PSP diagnosis may attribute lack of improvement to insufficient dosing rather than the wrong diagnosis.

The treatment gap and current research

There are no approved disease-modifying therapies for PSP, and no drugs that reliably reduce symptom burden over the long term. Multiple clinical trials targeting tau accumulation have been conducted:

• Davunetide (intranasal neuropeptide): Failed in phase III trial
• Tideglusib (GSK-3 inhibitor): Did not meet primary endpoints
• Anti-tau antibodies (gosuranemab, tilavonemab): Large phase II trials completed without meeting efficacy endpoints

The failure of anti-tau strategies in PSP is actively studied, and researchers are examining whether tau seeding, spreading, and different tau isoform ratios in PSP require different therapeutic approaches than those used in Alzheimer's disease.

Irwin's tauopathy review notes that PSP is characterized by accumulation of 4-repeat tau isoforms in neurons and astrocytes — a distinct pattern from the 3-repeat and 4-repeat mix seen in Alzheimer's disease — which may explain why anti-tau strategies developed for Alzheimer's have not transferred to PSP.

Symptomatic treatments that provide modest benefit include antidepressants for frontal lobe behavioral changes, speech therapy for dysphagia (swallowing impairment is a major source of morbidity and aspiration risk), and botulinum toxin for eyelid-opening apraxia and blepharospasm. Eye drops are often needed for dry eyes resulting from reduced blink rate.

Questions to ask your doctor

• Given that I have early falls, difficulty looking down, and my levodopa trial has not helped much, could this be PSP rather than Parkinson's disease?
• Have you seen the MRI results, and is there evidence of midbrain atrophy — specifically the "hummingbird sign" or "penguin sign" that suggests PSP?
• Should I see a movement disorder specialist at an academic center that has specific PSP expertise?
• What fall prevention strategies — including physical therapy, home modification, and walker type — are most appropriate for my level of postural instability?
• Am I eligible to participate in clinical trials of tau-targeting therapies?
• When should I start planning for swallowing evaluation, and is a speech-language pathologist who specializes in neurodegenerative disease available?

The bottom line

PSP is a rare tauopathy that is routinely misdiagnosed as Parkinson's disease, sometimes for years. The clinical clues that distinguish it — backward falls in the first year, vertical gaze palsy, symmetric presentation, and poor response to levodopa — are well defined by the NINDS-SPSP and MDS criteria developed by the researchers cited in this article. While no approved therapy slows PSP's progression, an accurate diagnosis changes everything: fall prevention becomes urgent, levodopa dose escalation stops, appropriate expectations are set, and patients gain access to specialized care and clinical trials. Evaluation by a movement disorder specialist with PSP experience is the most important step any patient or family facing this diagnosis can take.