New Patient Questionnaire Catches Ataxia Before Symptoms Show

For people living with spinocerebellar ataxias, a group of rare inherited disorders in which the cerebellum slowly degenerates, the future of drug development hinges on a deceptively simple question: how do you measure the disease? A new study from Radboud University Medical Center in the Netherlands, published in the Journal of Neurology, offers one of the most thorough answers yet. Researchers followed 76 carriers of mutations linked to three subtypes of the disease, SCA1, SCA3, and SCA7, over a full year, putting a 70-item patient questionnaire called the PROM-Ataxia through a battery of demanding statistical tests. Their verdict is nuanced but encouraging: the questionnaire captures the full weight of the disease from the patient’s own perspective, and remarkably, it can even detect the burden of illness in people who do not yet show any measurable gait problems.

Spinocerebellar ataxias are caused by toxic expansions of CAG repeats in the DNA, a molecular stutter that produces abnormally long stretches of the amino acid glutamine in disease proteins. Larger expansions typically trigger earlier and more severe disease, but even among carriers of the same mutation, the clinical picture varies enormously. What unites all subtypes is progressive ataxia, the loss of coordination that makes walking, speaking, and fine hand movements increasingly difficult. Yet the disease does not stop at the motor system. Many patients develop cognitive changes, mood disturbances, and difficulties with everyday tasks that no single clinician-rated scale fully captures. With no disease-modifying treatments currently available but nucleotide-based gene silencing therapies now advancing through development pipelines, the race is on to find outcome measures that can detect meaningful change within the compressed timeframe of a clinical trial.

The instrument of choice for decades has been the Scale for the Assessment and Rating of Ataxia, or SARA, a bedside examination in which a neurologist scores gait, stance, sitting, speech, and limb coordination on a scale from 0 to 40. It is quick, well validated, and widely used, but it has blind spots. It reflects only what the clinician observes during a brief examination, ignores non-ataxia symptoms entirely, and some of its items may have questionable relevance to patients’ daily lives. The PROM-Ataxia was developed to fill this gap through a multi-phase process that actively involved ataxia patients themselves. It spans three domains, physical, activities of daily living, and mental health, with each of its 70 items rated on a five-point scale, yielding a total score from 0 to 280 points. A condensed 10-item short form also exists for settings where brevity matters.

The Dutch team, led by Kristofoor Leeuwenberg and senior authors Roderick Maas and Bart van de Warrenburg, pooled data from three single-center observational cohorts at Radboud University Medical Center. The 76 mutation carriers, 20 with SCA1, 40 with SCA3, and 16 with SCA7, spanned all four stages of disease severity defined by Klockgether and colleagues, from people with no walking difficulties to those confined to a wheelchair. Seventeen of the participants were classified as pre-ataxic, meaning their SARA score was below 3 despite carrying the mutation. Sixteen healthy adults served as controls, and 68 of the carriers returned for a one-year follow-up visit. Alongside the PROM-Ataxia, participants completed or underwent eight additional assessments, including the SARA, the Inventory of Non-Ataxia Signs, the SCA Functional Index, the Cerebellar Cognitive Affective Syndrome Scale, depression and quality-of-life questionnaires, and the activities of daily living subscale of the Friedreich Ataxia Rating Scale.

The first major finding concerned convergent validity, the degree to which scores on the new instrument track scores on established measures of the same underlying construct. The results were striking. The PROM-Ataxia total score correlated moderately to strongly with all eight comparison measures, with Spearman correlation coefficients ranging from 0.46 to 0.89, all statistically significant. The strongest associations appeared with the SARA, the functional subscale of the Unified Huntington’s Disease Rating Scale, and the Friedreich Ataxia Rating Scale daily living subscore, coefficients between 0.84 and 0.89, indicating that the questionnaire rises in lockstep with worsening clinical ataxia and growing disability as judged by both clinicians and patients. Equally telling was the pattern among the domains: the psychosocial section of the mental domain correlated strongly with the depression questionnaire PHQ-9 at 0.71, confirming that each domain captures a distinct facet of the disease rather than a single undifferentiated burden.

Perhaps the most clinically consequential result emerged from the comparison between pre-ataxic carriers and healthy controls. Even before any measurable gait ataxia appeared, the questionnaire distinguished the two groups clearly. Total scores averaged 34.1 points in pre-ataxic carriers versus just 10.4 in controls, a highly significant difference, and the physical domain, activities of daily living domain, cognitive section of the mental domain, and short form all showed significant separations as well. Only the psychosocial section of the mental domain failed to reach statistical significance. This matters because earlier studies using generic patient-reported measures of quality of life, depression, and sleep quality have repeatedly struggled to detect differences between pre-ataxic individuals and healthy people. The PROM-Ataxia, by contrast, appears sensitive enough to register the subtle, pre-clinical burden that mutation carriers already experience, a property that could prove invaluable in prevention trials aimed at people who are still years away from overt symptoms.

The longitudinal picture was more complicated. Over one year, the average PROM-Ataxia total score rose by 3.9 points, a statistically significant increase consistent with the degenerative course of the disease, and the activities of daily living domain rose by 1.7 points. But the variability between individuals was enormous, with a standard deviation of 22.0 points for the total score change. This heterogeneity dragged down the standardized response mean, a measure of responsiveness calculated as the mean change divided by the standard deviation of that change, to just 0.17, far below the values of 0.82 for the SARA and 0.47 for the daily living scale. Notably, the variability was consistent across all disease stages and unrelated to baseline severity, age, sex, or subtype. It was not entirely random, however: in the SCA3 cohort, patients’ self-rated global impression of change in quality of life emerged as a significant predictor of questionnaire change, suggesting that fluctuations in subjective well-being, rather than disease progression alone, partly drive the scatter.

An illustrative case from the study captures this tension. The largest outlier in the cohort improved by 60 points on the questionnaire after adopting a new balance-focused physical therapy regimen, even though the objective SARA score worsened by a single point over the same period. The questionnaire, in other words, registers more than cerebellar degeneration; it registers the lived experience of the disease, including the benefits of rehabilitation, adaptation, and life circumstances. For a researcher designing a drug trial with one or two years of follow-up, that sensitivity to external influences is a double-edged sword. It makes the instrument richly informative about what matters to patients, but it also inflates the noise against which a treatment effect must be detected, potentially requiring larger sample sizes in a field where patient populations are inherently small.

To help interpret individual scores, the team estimated the minimal clinically important difference, the smallest change a patient would consider meaningful, using two complementary approaches. In the anchor-based method, participants in the SCA3 cohort rated their perceived change over the past year on seven-point scales tailored to gait, daily activities, mental health, cognition, and quality of life, and these ratings were linked to corresponding questionnaire domains. The anchor-based estimate for the total score came out at 9.8 to 12.7 points per year, with domain-specific values of 3.8 points for the physical domain, 3.2 for daily living, 1.8 to 2.9 for psychosocial symptoms, 1.8 for cognition, and 1.6 to 3.7 for the short form. A distribution-based method, defined as half the standard deviation of change, produced closely matching values, and the estimates aligned well with those recently reported by an American research group studying a mixed polyglutamine ataxia cohort, lending the numbers considerable robustness.

The study’s authors are candid about its limitations: some outcome measures were missing in certain cohorts because the three protocols were not identical, the anchor-based analysis was confined to the SCA3 subgroup, and the findings cover only three of the many known SCA subtypes. Even so, the overall message for the field is clear. The PROM-Ataxia is a valid, patient-centered companion to the SARA rather than a replacement for it, capable of revealing disease burden even in the pre-ataxic stage where other patient-reported measures fall silent. The authors recommend that its results be interpreted alongside objective clinical assessments, the patient’s global impression of change, and systematic documentation of contextual factors that shape well-being. As gene-silencing therapies edge closer to the clinic, tools that measure what patients actually feel, not just what neurologists observe, may become the difference between a trial that succeeds and one that misses the benefit it was designed to find.

Subject of Research: Validation of the PROM-Ataxia patient-reported outcome measure in spinocerebellar ataxia

Article Title: Convergent validity, responsiveness, and meaningful within-subject change of the PROM-Ataxia in spinocerebellar ataxias

Article References: Leeuwenberg, K. E., van Prooije, T. H., Reniers, C., Kapteijns, K. C. J., Maas, R. P. P. W. M., & van de Warrenburg, B. P. (2026). Convergent validity, responsiveness, and meaningful within-subject change of the PROM-Ataxia in spinocerebellar ataxias. Journal of Neurology, 273(10), Article 583. https://doi.org/10.1007/s00415-026-14108-1

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14108-1

Keywords: spinocerebellar ataxia, PROM-Ataxia, patient-reported outcomes, SARA, minimal clinically important difference, convergent validity, responsiveness, pre-ataxic, clinical trials, cerebellar degeneration, gene silencing, trial readiness