Andreas Traschütz

Digital motor outcomes may surpass the sensitivity of clinician-reported outcomes in hereditary spastic paraplegia (HSP), particularly in cases of mild upper limb impairment and early disease stages. We thus validated a trial-ready quantitative motor (Q-Motor) assessment of rapid alternating limb movement tasks against clinician-reported, patient-focused, and performance outcomes in HSP. Exploratory single-center cross-sectional assessment in 41 patients with different HSP genotypes (age: 48 ± 14 years), with speeded foot tapping, diadochokinesia, and finger tapping measured by a high-resolution force transducer. Validation comprised discrimination from 48 age- and sex-matched controls; correlations to the Spastic Paraplegia Rating Scale (SPRS; mean: 18.6 ± 9.1), Friedreich Ataxia Rating Scale- Activities of Daily Living (FARS-ADL), and Nine-Hole Peg Test (9HPT); and stratification by functional stage (FARS stage: mild/moderate/severe = ambulatory/walking aid/wheelchair). Foot tapping best discriminated patients with HSP from controls (e.g., frequency: area under the curve [AUC] = 0.94-0.96), and particularly captured HSP severity and functional impairment by measures of foot elevation and cumulative tap force (across all measures: |rho_SPRS|= 0.33-0.59; |rho_ADL|= 0.35-0.56). Speed measures of diadochokinesia and finger tapping captured functional impairment (|rho_ADL|= 0.32-0.60) and impaired dexterity (|rho_9HPT |= 0.53-0.77) in the upper limbs of patients with HSP-even in the mild stage, and with slowed finger tapping even in patients without upper limb pyramidal signs. Foot tapping measures were most sensitive in discriminating mild-stage impairment (AUC = 0.72-0.95) and predominantly changed between mild and moderate HSP, while upper limb measures of diadochokinesia and finger tapping predominantly changed from moderate to severe HSP. Q-Motor captures motor impairment in HSP, even in the upper limbs and mild disease stages, and should be further evaluated as a possible trial outcome by longitudinal validation in genotypically stratified cohorts and mapping to patient experience.

Background: Neuropsychological deficits have been observed in patients with cerebellar damage, but never thoroughly investigated in autosomal recessive spastic ataxia of Charlevoix‐Saguenay (ARSACS). Objectives: The goal is the characterization of presence, severity, and profile of neuropsychological deficits in ARSACS using the cerebellar cognitive‐affective syndrome (CCAS) scale. Methods: Prospective study including a discovery cohort from Saguenay/Canada (n = 31, median [inter‐quartile range] age: 57 [54–62] years), and a validation cohort from Tübingen, Germany (n = 17, 35 [21–43] years) with matched controls (n = 19). Results: All ARSACS patients failed in multiple CCAS‐related subtests and exceeded cutoffs for “definite CCAS.” Even the younger validation cohort failed more subtests than controls (5 [3–7] vs. 1 [1–2], P < 0.001) and had lower CCAS total scores (81 [67–86] vs. 101 [91–106], P < 0.001). Total scores worsened in the older discovery cohort (40 [25–52], P < 0.001) and correlated with age/disease duration (ρ = −0.575, P < 0.001) and ataxia severity (Scale for the Assessment and Rating of Ataxia: ρ = −0.527, P = 0.003). Conclusions: Neuropsychological deficits consistent with CCAS are consistent in ARSACS, present early, and progress in the disease course. © 2026 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Background: Digital‐motor outcomes promise better responsiveness than clinician‐reported outcomes in ataxia trials. However, their patient meaningfulness and sensitivity to change remain to be demonstrated, particularly in the upper limb domain. Objectives: Validation of quantitative motor (Q‐Motor) assessment for upper limb ataxia against patient‐reported outcomes and regarding sensitivity to both longitudinal and treatment‐induced change, the latter in n‐of‐1 treatment settings. Methods: Single‐center longitudinal assessment of finger tapping, diadochokinesia, grip‐lift, spiral drawing, and target reaching in (1) 36 cross‐genotype ataxia patients and 20 controls, validating digital measures for correlations with patient‐reported outcome measure (PROM)‐ataxia, 2‐weeks test–retest reliability, and sensitivity to change within a trial‐relevant 1‐year follow‐up, anchored in Patient Global Impression of Change (PGI‐C); and (2) two patients with spinocerebellar ataxia type 27B (SCA27B) on versus off treatment with 4‐aminopyridine. Results: Twenty‐four digital measures correlated with the PROM‐ataxia upper‐limb composite (|ρ| = 0.4–0.7) and had excellent test–retest reliability (ICC = 0.91–0.99). Correlations to individual PROM‐ataxia items were specific for functional impairment the respective measure was hypothesized to capture. Speed of finger tapping and diadochokinesia, and smoothness of target reaching (spectral arc length of movement in three dimensions [SPARC _3D ]) captured 1‐year progression in ataxia patients (| r _prb | = 0.38–0.51), and specifically in patients with worsening PGI‐C. Estimated sample sizes to detect longitudinal change were lower for digital than clinical outcomes (SPARC _3D : n = 33, Scale for the Assessment and Rating of Ataxia (SARA): n = 79, nine‐hole peg‐test: n = 214). Speed of diadochokinesia, stability of grip‐lift, and variability of target reaching captured treatment responses to 4‐aminopyridine in SCA27B, exceeding minimal detectable and minimal important change. Conclusion: Digital upper limb measures capture patient‐meaningful 1‐year longitudinal and treatment‐induced change, and are therefore promising outcomes for upcoming ataxia trials. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Background: Current clinical scales that track disease progression are more tailored to spasticity or ataxia, with limited sensitivity to change. Objectives: The aim was to develop a sensitive and valid scale specifically geared towards optimized sensitivity to change and adapted to patients presenting with both spasticity and ataxia. Methods: Longitudinal data from 127 spastic paraplegia type 7 (SPG7) and 112 autosomal recessive spastic ataxia Charlevoix‐Saguenay (ARSACS) patients were collected within the multicenter PROSPAX study. Sensitivity to change over 2 years of 30 items from the Scale for the Rating and Assessment of Ataxias (SARA), Spastic Paraplegia Rating Scale (SPRS), and the Activities of Daily Living subscale of the Friedreich's Ataxia Rating Scale (FARS‐ADL) was evaluated. Items that demonstrated the highest sensitivity to change were used to build the Spastic Ataxia Composite scale (SPAXCOM). Results: With seven items, the SPAXCOM showed an effect size of 0.71, higher than reference scales (SARA: 0.43, SPRS: 0.42, FARS‐ADL: 0.27). The SPAXCOM had a similar sensitivity to change for both genotypes and was more sensitive in participants with a SARA lower than 10 and within the intermediate disease stage (FARS‐Disease Staging: 2–3.5). The SPAXCOM showed a high correlation with disease duration ( r = 0.67 [0.60; 0.72]) and disease stage ( r = 0.86 [0.83; 0.89]). Perception of improvement, stagnation, and worsening were associated with a mean yearly SPAXCOM change of 0.44 (−0.14; 1.01), 0.61 (0.19; 1.03), and 1.22 (0.96; 1.49), respectively. Conclusion: The SPAXCOM is more sensitive to change and homogeneous across genotypes than the reference scales, allowing a reduction of the required sample size in future clinical trials. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Background: An intronic (GAA)•(TTC) repeat expansion in FGF14 was recently identified as the cause of spinocerebellar ataxia 27B (SCA27B), a disorder presenting with both chronic cerebellar ataxia and episodic symptoms. The phenotype of SCA27B overlaps with that of CACNA1A spectrum disorders. Objective: The objective of this work was to investigate the prevalence of GAA‐ FGF14 repeat expansions in patients with ataxia so far considered to be related to underlying CACNA1A variants. Methods: This is a cross‐sectional multicenter study. Results: GAA‐ FGF14 testing showed pathogenic expansions (≥250 repeats) in 6/67 (9%) patients carrying CACNA1A variants. All patients with a pathogenic GAA‐ FGF14 expansion had a disease onset >40 years and carried variants of uncertain significance (VUSs) in CACNA1A . Genetic reevaluation led to the reclassification of CACNA1A VUSs as likely benign in four of six patients, who were ultimately diagnosed with SCA27B. Conclusions: Late‐onset ataxia cases previously considered as CACNA1A ‐related disorder should be reevaluated and tested for SCA27B, particularly if related to a VUS in CACNA1A . © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Robust and highly sensitive outcomes are crucial for small trials in rare diseases. Combining different outcome types might improve sensitivity to identify disease severity and progression, yet innovative methodologies are scarce. Here we develop an Item Response Theory framework that allows integrated modeling of both continuous and categorical outcomes (ccIRT). With degenerative ataxias, a group of rare neurological coordination diseases, as a showcase, we developed a ccIRT model integrating two ataxia outcome types: a clinician‐reported outcome (Scale for the Assessment and Rating of Ataxia; SARA; categorical data) and digital‐motor outcomes for gait and limb coordination (continuous data). The ccIRT model leveraged data from 331 assessments from a natural history study for spastic ataxias. The model describes SARA items and digital‐motor outcomes data as functions of a common underlying ataxia severity construct, evaluating 9 gait and 17 limb coordination digital‐motor measures for their ability to add to SARA in estimating individual ataxia severity levels. Based on our proposed workflow for assessing digital‐motor outcomes in ccIRT models, the final model selected three digital gait and three limb coordination measures, reducing average uncertainty in ataxia severity estimates by 49% (10% SD) compared to the SARA‐only IRT model. Trial simulations showed a 49% and 61% reduction in sample sizes needed to detect disease‐modifying effects in two genotypes. Overall, our ccIRT framework for combining multiple outcome domains, even of different variable types, facilitates a more precise estimation of disease severity and a higher power, which is particularly relevant for rare diseases with inherently small and short trials. Trial Registration: ClinicalTrials.gov : NCT04297891

Parallel designs with an end-of-treatment analysis are commonly used for randomised trials, but they remain challenging to conduct in rare diseases due to small sample size and heterogeneity. Model-based approaches can be more powerful to detect treatment effects. We investigated the performance of longitudinal modelling to evaluate disease-modifying treatments in rare diseases a simulation study, leveraging as showcase a model of the SARA score progression in Autosomal Recessive Cerebellar Ataxia. We compared the power of parallel, crossover and delayed start designs, investigating several trial settings: trial duration (2 or 5 years); disease progression rate (slower or faster); magnitude of residual error ( σ =2/ σ =0.5); number of patients (100 or 40); method of statistical analysis (longitudinal analysis with non-linear or linear models; standard statistical analysis). In our simulations, using non-linear mixed effect models resulted in higher power than a rich, sparse linear mixed effect model or standard statistical analysis. Parallel and delayed start designs performed better than crossover designs. Our analysis showed that since disease progression is slow and residual variability is high (for the standard clinician-reported outcome), longer durations are needed for power to be greater than 80%, up to 5 years for slower progression and 2 years for faster progression ataxias.

With targeted treatment trials on the horizon, identification of sensitive and valid outcome measures becomes a priority for >100 spastic ataxias. While digital-motor measures, assessed using wearable sensors, are considered prime outcome candidates for spastic ataxias, genotype-specific validation studies are lacking. We here aimed to identify candidate digital-motor outcomes for spastic paraplegia type 7 (SPG7)—one of the most common spastic ataxias—that (1) reflect patient-relevant health aspects, even in mild, trial-relevant disease stages; (2) are suitable for a multicenter setting; and (3) assess mobility also during uninstructed walking simulating real life. This cross-sectional multicenter study (7 centers, 6 countries) analyzed defined laboratory-based walking and uninstructed “supervised free walking” in patients with SPG7 and healthy controls using 3 wearable sensors (Opal APDM). For the extracted digital gait measures, we assessed effect sizes for the discrimination of patients and controls (Cliff δ) and Spearman correlations with measures of functional mobility and overall disease severity (Spastic Paraplegia Rating Scale [SPRS], including mobility subscore SPRS^mobility; Scale for the Assessment and Rating of Ataxia [SARA]) and the activities of daily living subscore of the Friedreich Ataxia Rating Scale (FARS-ADL). Gait was analyzed in 65 patients with SPG7 and 50 healthy controls. Among 30 hypothesis-based gait measures, 18 demonstrated at least moderate effect size (δ > 0.5) in discriminating patients from controls and 17 even in mild disease stages (SPRS^mobility ≤ 9, n = 41). Spatiotemporal variability measures such as spatial variability measure SPcmp (ρ = 0.67, p < 0.0001) and stride time CV (ρ = 0.67, p < 0.0001) showed the largest correlations with functional mobility (SPRS^mobility)—as with overall disease severity (SPRS, SARA) and activities of daily living (FARS-ADL). The correlations of variability measures with SPRS^mobility could be confirmed in mild disease stages (e.g., SPcmp: ρ = 0.50, p < 0.0001) and in “supervised free walking” (e.g., stride time CV: ρ = −0.57, p < 0.0001). We here identified trial-ready digital-motor candidate outcomes for the spastic ataxia SPG7 with proven multicenter applicability, ability to discriminate patients from controls, and correlation with measures of patient-relevant health aspects—even in mild disease stages. If validated longitudinally, these sensor outcomes might inform future natural history and treatment trials in SPG7 and other spastic ataxias.

Degenerative cerebellar ataxias comprise a heterogeneous group of rare and ultra‐rare genetic diseases. While disease‐modifying treatments are now on the horizon for many ataxias, robust trial designs and analysis methods are lacking. To better inform trial designs, we applied item response theory (IRT) modeling to evaluate the natural history progression of several ataxias, assessed with the widely used scale for assessment and rating of ataxia (SARA). A longitudinal IRT model was built utilizing real‐world data from the large autosomal recessive cerebellar ataxia (ARCA) registry. Disease progression was evaluated for the overall cohort as well as for the 10 most common ARCA genotypes. Sample sizes were calculated for simulated trials with autosomal recessive spastic ataxia Charlevoix–Saguenay (ARSACS) and polymerase gamma (POLG) ataxia, as showcased, across multiple design and analysis scenarios. Longitudinal IRT models were able to describe the changes in the latent variable underlying SARA as a function of time since ataxia onset for both the overall ARCA cohort and the common genotypes. The typical progression rates varied across genotypes between relatively high in POLG (~ 0.98 SARA points/year at SARA = 20) and very low in COQ8A ataxia (~ 0.003 SARA points/year at SARA = 20). Smaller trial sizes were required in case of faster progression, longer trials (~ 75–90% less with 5 years vs. 2 years), and larger drug effects (~ 70–80% less with 100% vs. 50% inhibition). Simulating under the developed IRT model, the longitudinal IRT model had the highest power, with a well‐controlled type I error, compared to total score models or end‐of‐treatment analyses. The established longitudinal IRT framework allows efficient utilization of natural history data and ultimately facilitates the design and analysis of treatment trials in rare and ultra‐rare genetic ataxias.