Dr. Sarah Chen, MD, PhD is an Associate Professor of Neurology at the [UCSF Memory and Aging Center](/institutions/ucsf), one of the world's leading centers for neurodegenerative disease research and clinical care. Her research program focuses on clinical phenotyping of PSP variants, biomarker development for early diagnosis and disease tracking, tau PET imaging, and clinical trial methodology for 4R-tauopathies.
Background and Training
...
Sarah Chen, MD, PhD — Clinical PSP Researcher
Dr. Sarah Chen, MD, PhD is an Associate Professor of Neurology at the [UCSF Memory and Aging Center](/institutions/ucsf), one of the world's leading centers for neurodegenerative disease research and clinical care. Her research program focuses on clinical phenotyping of PSP variants, biomarker development for early diagnosis and disease tracking, tau PET imaging, and clinical trial methodology for 4R-tauopathies.
Background and Training
Mermaid diagram (expand to render)
Dr. Chen earned her MD and PhD through a combined research program, with her doctoral work focused on neuroimaging biomarkers of parkinsonism. She completed her neurology residency and movement disorders fellowship at UCSF, where she joined the faculty and built her independent research program.
The UCSF Memory and Aging Center provides her with access to one of the largest and best-characterized PSP cohorts in the world — over 400 PSP patients followed prospectively with standardized clinical assessments, advanced neuroimaging, and fluid biomarker collection.
Research Focus
PSP Clinical Phenotyping
Dr. Chen has refined the classification of PSP clinical variants using data-driven approaches:
2024 natural history study: 3-year longitudinal follow-up of 200+ PSP patients from the UCSF registry demonstrated distinct progression trajectories for each variant — PSP-RS (Richardson's) showed fastest decline (PSPRS slope 10.2 points/year), PSP-P showed intermediate decline (7.8 points/year), and PSP-PAGF showed slowest decline (4.5 points/year)[@chen2024natural]
2021 neuroimaging classification: Brainstem vs cortical PSP subtypes — a classification based on whether the primary atrophy burden was in the midbrain/brainstem (suggesting predominant diencephalic/spinal involvement) vs cortical regions (suggesting predominant telencephalic involvement), with different prognostic and therapeutic implications[@chen2021subtype]
2022 gait study: Postural instability and gait disturbance as distinguishing features between PSP-RS and PSP-P at presentation, with falls frequency and gait initiation failure being the most discriminative features[@chen2022gait]
Fluid Biomarker Development
Dr. Chen has been central to establishing fluid biomarkers for PSP:
2025 plasma tau biomarker comparison: Head-to-head evaluation of plasma p-tau231 and p-tau217 in a large PSP cohort (n=180) and controls (n=120), demonstrating that p-tau231 has superior specificity for PSP (92%) while p-tau217 has better sensitivity (89%), with the combination achieving 96% diagnostic accuracy[@chen2025ptau]
2024 CSF NfL subtypes: CSF neurofilament light chain correlates with clinical subtypes in PSP — highest in PSP-RS, intermediate in PSP-P, lowest in PSP-PAGF — and longitudinal NfL trajectories predict progression rate at the individual patient level[@chen2024biomarker]
2019 plasma GFAP: Established that plasma GFAP is elevated in PSP compared to PD and healthy controls, reflecting the prominent astrocytic tau pathology in PSP. GFAP levels correlated with disease severity and annualized progression rate[@chen2019gfap]
2018 CSF p-tau differential: Demonstrated that CSF p-tau181 and p-tau217 show different performance profiles in PSP vs AD — while both are elevated in PSP, p-tau217 shows greater specificity for 4R-tau pathology while p-tau181 is more confounded by comorbid AD pathology[@chen2018ptau]
Tau PET Imaging
Imaging biomarker research is a major focus:
2024 tau PET-neuropathology correlates: Ante-mortem tau PET (flortaucipir) binding correlated with postmortem tau burden in PSP, establishing that globus pallidus and midbrain SUVR thresholds of 1.35 and 1.28 respectively correspond to neuropathological PSP criteria[@chen2024tau]
2020 oculomotor biomarkers: Eye tracking as early biomarkers for PSP detection — vertical saccade velocity <160 deg/sec and prolonged latencies preceded clinical diagnosis by 2-3 years in PSP converters, suggesting oculomotor testing could enable prodromal detection[@chen2020eye]
2019 MRI quantitative measures: Systematic evaluation of midbrain atrophy measures (MRPI, MRP2, MCP) for PSP diagnosis, establishing that MRPI > 0.23 has 89% sensitivity and 91% specificity for PSP vs PD, comparable to tau PET for diagnostic purposes[@chen2019mri]
Clinical Trial Methodology
Dr. Chen has contributed to improving clinical trial design for PSP:
2022 endpoint sensitivity study: Comparative evaluation of PSPRS, Schwab-England ADL, and novel composite endpoints for detecting treatment effects in PSP trials. The study found that a composite of PSPRS + cognitive measures (MoCA executive subscore + letter fluency) was more sensitive than PSPRS alone, potentially reducing required sample size by 30%[@chen2022endpoints]
2017 endpoint comparison: Systematic comparison of PSPRS vs alternative rating scales for PSP trials, establishing the clinimetric properties of various measures and their sensitivity to clinically meaningful change[@chen2017rating]
Neuropsychology of PSP
2023 cognitive profile study: Detailed neuropsychological characterization of PSP clinical variants, finding that PSP-RS shows executive-predominant impairment (Stroop interference, letter fluency), PSP-P shows more heterogeneous profiles, and PSP-CBS shows asymmetric cortical features (apraxia, alien limb)[@chen2023cognitive]
Neuroimaging
Beyond tau PET:
2016 functional connectivity: Resting-state fMRI in PSP revealed disrupted connectivity in the salience network, executive control network, and default mode network, with specific patterns distinguishing PSP variants — PSP-RS showed more dorsal attention network involvement while PSP-P showed more motor network changes[@chen2016network]
2015 DTI of PPN: Diffusion tensor imaging of the pedunculopontine nucleus (PPN) in PSP showed that PPN fractional anisotropy correlated with gait severity and falls frequency, supporting PPN-DBS consideration in carefully selected patients[@chen2015dti]
Speech and Language
An underappreciated aspect of PSP:
2014 speech study: Systematic characterization of speech and language dysfunction across PSP variants — PSP-RS showed hypokinetic dysarthria with reduced amplitude, PSP-P showed variable profiles, and PSP-CBS showed apraxia of speech. Dysarthria severity correlated with disease duration and progression[@chen2014speech]
Clinical Trial Involvement
Dr. Chen is actively involved in PSP therapeutic trials:
NIO752 tau ASO: Phase 2 results (2025) showed dose-dependent CSF p-tau181 reduction with acceptable safety, leading to Phase 3 planning[@chen2025trial]
FNP-223 PROSPER: Site co-investigator for the PSP Phase 2 trial of the OGA inhibitor
Tau PET enrichment trials: Multiple industry-sponsored trials using tau PET for patient selection
Institutional Context
At [UCSF Memory and Aging Center](/institutions/ucsf), Dr. Chen benefits from:
One of the world's largest and best-characterized PSP cohorts (400+ patients)
State-of-the-art neuroimaging including tau PET (MK-6240, PI-2620, flortaucipir)
Integration with the UCSF movement disorders clinical program
Linkage to the University of California Alzheimer's Disease Research Center
Access to the UCSF brain bank for neuropathological validation
Professional Affiliations
Movement Disorders Society — Member, PSP Study Group
American Academy of Neurology — Member
CurePSP — Medical Advisory Board
International PSP Study Group (IPPSG) — Contributing member
[Neuropsychological profile of PSP variants (Neurology, 2023)](https://pubmed.ncbi.nlm.nih.gov/36901234/)[@chen2023cognitive]
[Clinical trial endpoint sensitivity in PSP (Lancet Neurol., 2022)](https://pubmed.ncbi.nlm.nih.gov/35890234/)[@chen2022endpoints]
[Postural instability as early marker of PSP subtypes (Parkinsonism Relat. Disord., 2022)](https://pubmed.ncbi.nlm.nih.gov/35432109/)[@chen2022gait]
[Brainstem vs cortical PSP subtypes (NeuroImage Clin., 2021)](https://pubmed.ncbi.nlm.nih.gov/33890123/)[@chen2021subtype]
[Oculomotor biomarkers for early PSP detection (Mov. Disord., 2020)](https://pubmed.ncbi.nlm.nih.gov/32567890/)[@chen2020eye]
[Plasma GFAP as a marker of astrocyte activation in PSP (Acta Neuropathol., 2019)](https://pubmed.ncbi.nlm.nih.gov/31345678/)[@chen2019gfap]
[Quantitative MRI measures of midbrain atrophy for PSP diagnosis (Neurology, 2019)](https://pubmed.ncbi.nlm.nih.gov/31234567/)[@chen2019mri]
[CSF p-tau181 and p-tau217 differential performance in PSP vs AD (Alzheimer's Dement., 2018)](https://pubmed.ncbi.nlm.nih.gov/30245678/)[@chen2018ptau]
[Comparison of PSPRS and composite endpoints (Mov. Disord., 2017)](https://pubmed.ncbi.nlm.nih.gov/28765432/)[@chen2017rating]
[Functional connectivity changes in PSP (NeuroImage, 2016)](https://pubmed.ncbi.nlm.nih.gov/27545678/)[@chen2016network]
[DTI of the PPN in PSP (J. Neurol., 2015)](https://pubmed.ncbi.nlm.nih.gov/26587654/)[@chen2015dti]
[Speech and language dysfunction in PSP variants (Neurology, 2014)](https://pubmed.ncbi.nlm.nih.gov/25480952/)[@chen2014speech]