P4‐542: UTILITY OF SPEECH‐BASED DIGITAL BIOMARKERS FOR EVALUATING DISEASE PROGRESSION IN CLINICAL TRIALS OF ALZHEIMER'S DISEASE
Bibliographic record
Abstract
Recent clinical trials in Alzheimer's disease have been overwhelmingly negative, spurring development of novel biomarkers which might capture changes in cognitive function with greater precision. Computational analysis of speech and language represent one such group of biomarkers(1,2). The objective of this study was to examine the utility of a speech-based digital biomarker for tracking disease progression and treatment response to investigational treatment COR388 in a group of patients with Alzheimer's disease. The study was a small, double-blind placebo-controlled Phase 1b trial (NCT03418688) of COR388 with a cohort of nine (9) individuals with Alzheimer's disease. Participants were randomized to receive COR388 or placebo in a 2:1 fashion, BID for 28 days. A tablet-based speech and language assessment was administered at Days 1, 15 and 28. Participants were asked to complete 2 picture description tasks and verbal responses were analyzed. Five aggregate markers, chosen for their previous association to AD(1,2), were computed: discourse, syntactic complexity, lexical complexity, information units and word finding difficulty (WFD). Previous analysis of single outcomes showed significant improvement in the quality of picture descriptions for COR388 patients relative to placebo (increase in unique object content units, p=0.016 and prepositions, p=0.0011). Positive trends, but no significant differences in MMSE scores were observed(3). For aggregate markers, mean baseline to endpoint comparisons showed statistically significant (p<0.05) improvements in syntactic complexity, lexical complexity and information units in those treated with COR388. No significant within-subject differences were observed for placebo. Baseline to endpoint COR388 information unit differences remained significant post Bonferroni correction (p=0.002). Between-group analysis of information unit change scores (Day 28 vs. Day 1) revealed a 10-point increase for COR388 vs. a 5-point change for placebo but this numerical difference was not significant (p=0.21). In this preliminary trial, patients treated with COR388 showed signs of significant improvement relative to placebo as measured by a speech-based, digital biomarker. No significant changes in MMSE were observed suggesting that digital biomarkers may represent sensitive tools for tracking changes in cognition in small trials. The study also highlights the potential therapeutic benefit of COR388, though additional studies of sufficient power are needed.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".