March 2026 Research Update

This month added early, mixed-stage evidence that tau-targeting approaches can clearly change tau-related lab measures, while it remains uncertain how much this translates into day-to-day symptoms. In people with PSP, an early-phase antibody study showed a large drop in a tau measure in spinal fluid and was generally tolerable, but it was too small to show clear clinical benefit. In animal and cell models designed to mimic 4R tau diseases like PSP, a different experimental compound reduced tau-related pathology and improved function, but these results do not predict what will happen in humans. [1] [2]

In a small phase 1b trial, bepranemab (a tau-targeting antibody) was compared with placebo for 52 weeks and was reported to be generally well tolerated, with no deaths. The study also reported about an 80% reduction in a targeted "free tau" measure in spinal fluid by week 1, showing that the drug engaged its intended target in the body. Because only 25 people were enrolled and the study was designed mainly for safety and lab readouts, it did not provide firm answers about effects on clinical measures. [1]

In preclinical work, GV1001 was tested in a mouse model and in lab-grown nerve cell models aimed at 4R tauopathy relevant to PSP. GV1001 was reported to reduce several measures of tau buildup and inflammation in the brain, and to improve motor and memory-like test performance in mice. These findings support GV1001 as a candidate for further research, but they are still laboratory and animal results rather than evidence of benefit for people with PSP. [2]

For patients and caregivers, the main update is that one tau antibody (bepranemab) showed strong target engagement in spinal fluid and appeared broadly tolerable in a small early study, which is a step needed before larger trials that can test symptom outcomes. The GV1001 results are earlier-stage and mainly help researchers judge whether a possible treatment is worth moving forward, not whether it works for people yet. What remains uncertain is whether changing these tau-related lab measures will slow progression or improve function, because the human study was not sized to answer that question and the other study was not in humans. [1] [2]

Several studies this month focused on why PSP can be hard to recognize early and how diagnosis might become more accurate with better biomarkers, imaging measures, and symptom-pattern tools. Across these studies, a consistent message was that early PSP can look less "classic" than many people expect, including memory symptoms and even a lack of clear parkinsonism at the first visit. At the same time, blood-based and MRI-based measures showed promise for separating PSP from Parkinson's disease and other look-alike conditions, at least at the group level. [3] [4] [5]

In a large registry analysis of people whose PSP diagnosis was later confirmed at autopsy, memory symptoms were common at the first visit and about half had a cognitive-predominant onset, not just movement-predominant symptoms. The study also reported that 20% to 30% did not show parkinsonian signs at presentation, and nearly one-fifth were initially labeled as Alzheimer's disease. This helps explain why families may experience diagnostic changes over time, especially early in the illness. [3]

A multicenter biomarker study reported that tau-related proteins inside neuron-derived extracellular vesicles (small particles released by nerve cells and measured from blood) were higher in PSP than in healthy controls and Parkinson's disease. An integrated model combining these blood measures showed strong separation of PSP from the comparison groups, including very strong performance in early-stage PSP in this dataset. These results support the idea that a blood test could eventually help support diagnosis, but the authors noted the need for larger and longer-term confirmation. [4]

Two MRI studies added more detail on brain-structure measures that differ in PSP. One study in people with "lower body parkinsonism" reported that MRI planimetry measures such as MRPI and MRPI 2.0 were higher in PSP than in vascular parkinsonism and Parkinson's disease, and MRPI 2.0 separated PSP and normal pressure hydrocephalus from vascular parkinsonism at the group level. Another study using an automated method to measure hypothalamus size found smaller hypothalamic volumes in PSP than in healthy controls and also smaller than in Parkinson's disease, across two separate cohorts. [5] [6]

A machine-learning study used detailed, time-ordered clinical symptoms from a brain bank cohort to predict which underlying brain pathology a person likely had, including categories that can include PSP. The model reached an AUC of 0.83 at 3 years after onset across nine pathology groups and still performed reasonably (AUC 0.80) using a smaller subset of clinical inputs, with restricted eye movement among important predictors. This kind of work is aimed at creating decision-support tools, but it does not yet show how well such tools would perform in routine clinics outside the dataset used to build them. [7]

A separate study looked at depression reporting in several dementias, including PSP, and found that about one-third of patient-caregiver pairs disagreed on whether depression was present. PSP was more likely than Alzheimer's disease to show a pattern where the person endorsed depression but the caregiver did not, and caregiver-reported apathy often appeared in these discordant cases. This does not diagnose depression by itself, but it highlights how symptom reporting can be complicated in PSP and may differ depending on who is asked. [8]

For patients and caregivers, these studies help clarify two things at once: early PSP can present with thinking or memory concerns, and research is actively trying to reduce diagnostic uncertainty with blood and MRI measures. The blood extracellular-vesicle tau findings and the MRI measures may matter in the future if they hold up in larger, real-world testing, because they could support earlier and clearer separation of PSP from similar conditions. Right now, these are research results rather than a change in standard diagnosis, and some tools (like machine-learning symptom models) still need testing in new clinical settings to know how reliable they are. [3] [4] [5] [6] [7] [8]

This month, studies added clearer detail about how PSP affects everyday functions over time and how the biology of PSP may differ from person to person. A longitudinal study showed swallowing problems can worsen measurably and track with overall disease severity and brain shrinkage patterns. [9] Other studies pointed to differences in immune activity [10] , balance-related brain signals [11] , and genetic immune backgrounds [12] , suggesting that PSP is not biologically identical across all patients.

In 61 people followed over time, swallowing difficulty (dysphagia) worsened, with notable changes in a swallowing score and in tongue-pressure measures. Faster worsening in one swallowing score was seen in those with a predominant speech-language disorder presentation. Changes in swallowing measures also tracked with changes in a standard PSP severity scale. Baseline swallowing scores were related to how quickly some swallowing measures later changed, while brain atrophy rates were separately associated with baseline swallowing scores. [9]

Immune-related findings were reported both in blood and in brain tissue. In a small blood study, PSP participants had a lower frequency of natural killer (NK) cells, particularly loss of a CD56bright subset, along with differences in NK-cell gene activity that highlighted pathways related to blood vessel growth signals like VEGF-A. In a separate autopsy study of 32 PSP brains, HLA haplotype groups showed regional differences in microglia, CD8 T cells, and phosphorylated tau pathology, and symptom sequences differed by haplotype group. [10] [12]

Studies also examined brain and sleep signals related to movement and cognition. In an EEG balance study, PSP participants had reduced fronto-medial theta activity compared with Parkinson's disease, and lower theta related to more postural sway in parkinsonism groups, which may relate to how the brain responds to balance challenges. In a sleep-center cohort of people with REM sleep behavior disorder, the small subgroup whose RBD occurred with a tau-related brain disease (6 people with Alzheimer's disease, PSP, corticobasal degeneration, or frontotemporal dementia, analyzed together) had more cognitive impairment and a lower percentage of REM sleep than other RBD groups, suggesting that RBD linked to tau diseases may come with a distinct cognitive and sleep profile. [11] [13]

Two broader biology studies focused on tau itself and on large-scale protein patterns across diseases. In a lab study using tau filaments from PSP and other tau disorders, researchers reported that tau "seeds" could be amplified without added cofactors under certain conditions and that the shape/state of tau monomers affected propagation, which helps refine how researchers model tau spread in the lab. Separately, a brain proteomics atlas spanning multiple neurodegenerative diseases, including PSP with tau pathology, reported both shared and disease-specific protein signatures and described molecular subtypes within some diseases, offering a resource for future PSP comparisons. [14] [15]

For patients, the swallowing study most directly relates to daily life because it shows that swallowing can decline over time in PSP and can be tracked with repeat assessments, which may help anticipate needs as the disease changes. The immune, EEG, and lab tau studies mainly add to understanding of why PSP varies between people and which body systems may be involved, but they are not yet tests that guide routine care. What remains uncertain is how these biology signals connect to individual symptoms or to future treatments, because many of these studies were small, cross-sectional, or laboratory-based. [9] [10] [12] [11] [14] [15]

This month’s case reports highlighted how PSP-like symptoms can sometimes come from a different cause, and how individual patients may show unexpected strengths or respond to supportive procedures. One report described a potentially treatable brain lymphoma that initially looked like PSP until repeat imaging and biopsy clarified the diagnosis. Other reports described a person with PSP who could sing familiar songs despite severe speech difficulty, and a person with PSP who had meaningful pain relief and functional improvement after shoulder replacement surgery without reported complications. [16] [17] [18]

In the PSP-mimic report, a man developed rapid cognitive decline and parkinsonism that resembled PSP, but repeat MRI later showed features suggestive of lymphoma, confirmed as primary CNS lymphoma by biopsy. After corticosteroid treatment, the report described clinical stabilization. The main message is that when symptoms progress quickly or do not fit perfectly, repeating imaging can sometimes reveal a different diagnosis, including rare but treatable causes. [16]

In the speech case, a man with PSP and progressive non-fluent aphasia could sing a familiar song fluently but struggled with an unfamiliar song unless he had lyrics, and he read aloud more accurately than he sang. This single case illustrates that some automatic or well-learned communication pathways can remain relatively preserved even when everyday speech is difficult, though this pattern will not apply to everyone. [17]

In the orthopedic case, a woman with PSP underwent staged reverse shoulder arthroplasty for severe rotator cuff problems and was reported to have meaningful pain relief and improved shoulder motion and function, with no perioperative complications described. The report also noted that rehabilitation was adapted to PSP-related cognitive and movement challenges. This does not prove outcomes for most people with PSP, but it shows that carefully selected supportive procedures can sometimes improve comfort and function in an individual patient. [18]

For patients and caregivers, these cases mainly add awareness rather than general rules. The lymphoma case is a reminder that a PSP-like picture can rarely have a different, potentially treatable cause, especially with rapid change, and that repeat testing sometimes matters. The singing and shoulder surgery cases show that PSP symptoms can be uneven, with some abilities or quality-of-life improvements still possible in specific situations, but single-patient reports cannot predict what will happen for others. [16] [17] [18]