This page provides a systematic analysis of why amyotrophic lateral sclerosis (ALS) clinical trials fail despite strong preclinical rationale, with recommendations for future trial designs.
Last Updated: 2026-03-13 PT
Why have many neuroprotective phase II/III ALS trials failed despite strong preclinical rationale?[1][2]
Score: 30 (Tier 1)
Rank: #9 in ALS Knowledge Gaps
Despite over 50 clinical trial failures since the 1990s, the translation from promising preclinical results to clinical efficacy remains the central challenge in ALS drug development. This gap encompasses preclinical model limitations, species differences, trial design flaws, patient heterogeneity, and fundamental gaps in understanding disease biology.[3]
The approval of tofersen for SOD1-associated ALS in 2023 demonstrated that careful patient selection, genetic stratification, and biomarker-driven endpoints can yield positive results. However, extending this success to broader ALS populations remains the central challenge.[4]
| Category | Issue | Impact | Mitigation |
|---|---|---|---|
| Preclinical models | SOD1 mice don't capture sporadic ALS | High | Develop C9orf72, TDP-43, FUS models |
| Species differences | Blood-brain barrier, metabolism, physiology | High | Better PK/PD translation |
| Trial design | Endpoint selection, patient heterogeneity | High | Biomarker enrichment, composite endpoints |
| Treatment timing | Animal studies pre-symptomatic; humans post-diagnosis | Medium | Earlier intervention studies |
| Dosing | Maximum tolerated dose may not be optimal | Medium | Biomarker-guided dosing |
| Genetic stratification | Mixed ALS populations in trials | High | Genotype-specific trials |
The majority of early ALS trials focused on glutamate excitotoxicity, reflecting the initial hypothesis that excess glutamate caused motor neuron death:
Why They Failed: The excitotoxicity hypothesis, while partially valid, represents a downstream effect rather than a primary disease driver. Targeting glutamate alone cannot address the upstream molecular triggers.
Multiple antioxidants and mitochondrial protectors have failed:
Why They Failed: Oxidative stress is a secondary response to primary neurodegeneration. Antioxidants may be too late to provide benefit once the cascade is established.
Anti-inflammatory approaches have shown mixed results:
Why They Failed: The role of neuroinflammation in ALS is complex—some inflammatory responses may be protective. Broad anti-inflammatory approaches may remove beneficial immune surveillance.
Despite TDP-43 aggregation being the hallmark pathology in >95% of ALS cases, aggregation-targeting trials have struggled:
Why They Failed: Current approaches may not adequately target the soluble-to-aggregate transition or address loss-of-function from TDP-43 nuclear depletion.
Why They Failed: Protein delivery to motor neurons is constrained by the blood-brain barrier. Local delivery approaches have shown some promise but face surgical risks.
The SOD1 transgenic mouse model, while valuable, has limited translational validity:
| Characteristic | SOD1 Mouse Model | Human ALS |
|---|---|---|
| Disease trigger | Genetic overexpression | Variable (genetic/sporadic) |
| Progression timeline | Weeks to months | Years |
| Pathology | SOD1 aggregation | TDP-43 aggregation (majority) |
| Treatment window | Pre-symptomatic | Post-diagnostic |
Key Insight: Over 200 compounds showed efficacy in SOD1 mouse models but failed in human trials.[9]
ALS involves multiple simultaneous pathological processes:
ALS is increasingly recognized as a collection of molecularly distinct syndromes:
| Subtype | Genetic Basis | Estimated % | Implications |
|---|---|---|---|
| C9orf72 | Hexanucleotide expansion | ~40% familial, ~5-10% sporadic | Most common genetic form |
| SOD1 | Point mutations | ~15-20% familial | Target for ASO therapy |
| FUS | FUS protein mutations | ~5% familial | RNA processing focus |
| TARDBP | TDP-43 mutations | ~5% familial | Proteinopathy focus |
| Sporadic | Unknown | ~50-70% | Most heterogeneous |
Most trials have not stratified by genotype, potentially masking efficacy in specific subgroups.
The ALSFRS-R (Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised) has significant limitations:
Emerging Biomarkers:
A fundamental mismatch exists between preclinical and clinical treatment windows:
| Study Type | Typical Treatment Start | Disease Stage |
|---|---|---|
| Mouse models | Pre-symptomatic | Pre-clinical |
| Human trials | Post-diagnostic | Established disease |
By the time patients present with symptoms, significant motor neuron loss has already occurred.
Design trials for specific genotypes rather than all-comers:
Implement biomarker-driven patient selection and endpoints:
| Biomarker | Use Case | Status |
|---|---|---|
| NfL | Patient enrichment, endpoint | Qualified in some contexts |
| pNfH | Progression marker | Under validation |
| CSF NfL | Pharmacodynamic marker | Emerging |
| Genetic testing | Patient stratification | Required for gene-specific trials |
Target upstream mechanisms rather than downstream symptoms:
Given the multi-pathway nature of ALS, combination strategies may be necessary:
Implement more efficient trial architectures:
The ALS investment landscape reflects both the challenges and opportunities in the field:
| Company | Approach | Status |
|---|---|---|
| Biogen / Ionis | SOD1 ASO (tofersen) | Approved |
| Wave Life Sciences | C9orf72 ASO | In development |
| Denali Therapeutics | Small molecules, transport technology | Multiple programs |
| Prevail Therapeutics | Gene therapy | Acquired by Eli Lilly |
For detailed investment data, see the ALS Investment Landscape page, which tracks:
Based on portfolio analysis, the following mechanisms are underinvested relative to their biological importance:
See ALS vs FTD: Comparative Investment Analysis for cross-disease perspective.
Tofersen (Qalsody) demonstrates that carefully designed trials can succeed:
| Factor | Tofersen Approach | Implication for Future Trials |
|---|---|---|
| Genetic stratification | SOD1 mutation carriers only | Genotype-specific enrollment |
| Biomarker endpoints | NfL reduction as pharmacodynamic marker | Early biological activity |
| Target engagement | Direct ASO delivery to CSF | Ensure target access |
| Extended follow-up | Open-label extension allowed long-term assessment | Capture delayed effects |
Key Learnings:
van den Bosch L. ALS — An accelerating promise. Nat Rev Neurol. 2024. ↩︎
Pugdahl L, et al. Generalised epilepsy in a population-based study of ALS. J Neurol Neurosurg Psychiatry. 2007. ↩︎
Petrov D, et al. Animal models of ALS. Neurobiol Aging. 2017. ↩︎
Miller TM, et al. Trial of antisense oligonucleotide tofersen for SOD1 ALS. N Engl J Med. 2022. ↩︎ ↩︎ ↩︎
Bensimon G, et al. A controlled trial of riluzole in amyotrophic lateral sclerosis. N Engl J Med. 1994. ↩︎
Paganoni S, et al. Ceftriaxone in amyotrophic lateral sclerosis. JAMA Neurol. 2013. ↩︎
Liguori I, et al. Oxidative stress, mitochondrial dysfunction and neurodegenerative diseases. Neurotoxicology. 2013. ↩︎
Gordon PH, et al. Minocycline in amyotrophic lateral sclerosis. N Engl J Med. 2007. ↩︎
Benatar M. ALS: A unifying genetic hypothesis. Neuron. 2007. ↩︎
Benatar M, et al. [Neurofilament as a potential ALS treatment biomarker](https://doi.org/10.1016/S1474-4422(18). Lancet Neurol. 2018. ↩︎