Data-Driven SuStaIn Model of Disability Progression in Amyotrophic Lateral Sclerosis.
Аннотация
OBJECTIVE
To determine whether ordinal Subtype and Stage Inference (SuStaIn) applied to routine ALSFRS-R item scores can identify reproducible disability progression patterns in amyotrophic lateral sclerosis (ALS) and provide clinically meaningful staging.
METHODS
We analysed baseline ALSFRS-R item responses from 866 PRO-ACT participants. Ordinal SuStaIn inferred subtype-specific sequences of functional deterioration and assigned each participant to a subtype and a SuStaIn-derived functional stage. Longitudinal stability was assessed across 3625 consecutive follow-up visit pairs from 697 participants. Structural reproducibility was evaluated in an independent cohort of 301 consecutive ALS patients. Associations of baseline subtype and stage with survival and subsequent ALSFRS-R decline were examined using Cox and piecewise-linear models.
RESULTS
A three-subtype solution identified fine motor-, gross motor- and bulbar-predominant patterns of early disability. Within each subtype, SuStaIn reconstructed ordered multidomain sequences of functional deterioration and assigned each participant a SuStaIn-derived stage. Subtype assignment was stable across 90.4% of consecutive visit pairs, and stage was non-decreasing in 98.8%. Subtype-specific event ordering was reproduced in the validation cohort. Higher baseline stage was associated with increased mortality risk in the fine motor- and gross motor-predominant subtypes, but not clearly in the bulbar-predominant subtype. Baseline stage showed subtype-dependent, non-linear associations with subsequent functional decline, with acceleration up to mid-stage breakpoints in the fine motor- and gross motor-predominant subtypes and less evident stage-dependent acceleration in the bulbar-predominant subtype.
INTERPRETATION: Routine ALSFRS-R item-level data can define clinically interpretable ALS progression subtypes and latent SuStaIn-derived functional stages. Joint subtype-stage modelling may refine prognostic stratification and support prognosis-informed trial enrichment.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.