¶ ALS Progression Rate Heterogeneity — Mechanism and Biomarker Predictors
This experiment addresses ALS Knowledge Gap #3 (32 points, Critical): "What determines rapid versus slow progression trajectories across ALS phenotypes?"[@westeneng2018][@van2019] Despite similar clinical presentations, ALS patients show dramatically different progression rates — some lose ambulation within 12 months while others remain functional for 5+ years. Understanding the molecular drivers of this heterogeneity could enable precision medicine approaches and dramatically improve clinical trial efficiency through enrichment strategies.
ALS progression rate is determined by a combination of: (1) genetic modifiers (e.g., UNC13A, ATXN2 polyQ repeats), (2) immune landscape composition at diagnosis, (3) metabolic state (BMI, lipid profiles, glucose metabolism), and (4) initial pattern of regional involvement. These factors can be captured in a composite biomarker score that predicts progression trajectory at diagnosis.
Cohort: 2,000 ALS patients from established biobanks (PRO-ACT, ENCALS, answer ALS)
- Inclusion: Diagnosed ALS, ≥2 longitudinal ALSFRS-R assessments, available plasma/CSF
- Data: Demographics, genetics (known modifiers), baseline clinical, longitudinal functional scores
- Endpoint: ALSFRS-R slope (ΔALSFRS-R/month)
Biomarker panels:
- Neurofilaments (NfL, pNfH) — established progression markers
- Inflammatory cytokines (IL-6, TNF-α, IL-1β, CXCL10)
- Metabolic markers (HbA1c, lipid panel, vitamin D)
- Genetic modifiers (UNC13A, ATXN2, C9orf72, TARDBP)
Analysis:
- Machine learning to identify multi-marker signatures predicting fast vs slow progression
- Validation in independent cohorts
- Feature importance to identify mechanistic drivers
Cohort: 500 newly diagnosed ALS patients at 20 sites
- Design: Prospective observational with standardized biomarker collection
- Primary endpoint: ALSFRS-R slope at 12 months
- Secondary: Ventilatory decline, survival, cognitive progression
Biomarker sampling:
- Plasma NfL at baseline, 3, 6, 12 months
- CSF metabolomics at baseline
- PBMC transcriptomics at baseline
Validation metrics:
- Sensitivity/specificity of composite score for fast progression prediction
- Calibration curves across progression subgroups
Design: Retrospective application of enrichment strategy to completed Phase 2/3 trials
Analysis:
- Simulate trial outcomes with vs without progression-based enrichment
- Calculate required sample size reduction
- Identify optimal cutoffs for "fast progressor" definition
- Patient-derived data (primary): PRO-ACT database (n=10,000+), ENCALS registry
- iPSC motor neurons: Isogenic lines with progression-associated genetic variants
- SOD1/G93A mice: Characterization of fast vs slow progressing sublines
- Primary: Validated composite biomarker score predicting progression rate at diagnosis
- Secondary: Mechanistic insights into drivers of progression heterogeneity
- Tertiary: Clinical trial enrichment strategy reducing required sample size by 30-50%
| Dimension |
Score |
Rationale |
| Technical |
9/10 |
Established biomarker platforms; large existing datasets |
| Timeline |
8/10 |
36 months for full validation; retrospective phase faster |
| Cost |
7/10 |
Estimated $4.5M total; leveraging existing cohorts reduces cost |
| Interpretability |
9/10 |
Clear endpoints; validated clinical relevance |
| Phase |
Cost |
| Phase 1 (Discovery) |
$1.2M |
| Phase 2 (Validation) |
$2.0M |
| Phase 3 (Simulation) |
$0.8M |
| Total |
$4.0M |