Roche Discontinues Both Leading Huntington's Disease Programs in Major Gene-Silencing Setback
On July 9, 2026, Swiss pharmaceutical giant Roche announced the simultaneous discontinuation of its two leading clinical candidates for Huntington’s disease (HD)—tominersen and RG6496—marking a major setback for the neurodegeneration research pipeline. The decisions, though announced together, were driven by independent, distinct clinical and preclinical datasets.
The Biomarker Paradox of Tominersen (GENERATION HD2)
Tominersen is an antisense oligonucleotide (ASO) designed to lower levels of "total" (both normal and mutant) huntingtin protein. Delivered via intrathecal lumbar puncture, it was the first therapeutic candidate shown to successfully lower huntingtin in the human brain.
Roche's Phase 2 GENERATION HD2 trial evaluated a 100 mg dose of tominersen in early-stage patients over a 16-month period. The trial's top-line results revealed a stark "biomarker-clinical dissociation":
- Biological Success: Patients receiving tominersen experienced significant, expected reductions in cerebrospinal fluid (CSF) mutant huntingtin, alongside decreases in neurofilament light (NfL)—a key biomarker of neuronal damage.
- Clinical Failure: Despite improved biological markers, patients showed no functional or clinical divergence from the placebo group. The trial failed to meet its clinical endpoints on the composite Unified Huntington's Disease Rating Scale (cUHDRS) and Total Functional Capacity (TFC) scales.
The HDBuzz scientific team interpreted the results as a reflection of the trial's constraints rather than the failure of the huntingtin-lowering hypothesis:
"This study suggests that the degree of lowering of expanded huntingtin and changes to NfL achieved in approximately 16 months in this trial was perhaps not sufficient to meaningfully slow HD progression. But this does not necessarily mean that huntingtin lowering cannot slow HD... Was the amount of huntingtin lowering sufficient in the brain regions that matter most? Was treatment started early enough? Is 16 months long enough to detect changes in a disease that progresses very slowly?"
Because of the lack of clinical benefit, Roche is completely terminating the development of tominersen, with no open-label extensions or compassionate use options.
Preclinical Safety Halts RG6496 (POINT-HD)
In contrast to tominersen, RG6496 was designed as an allele-selective ASO to target and lower only the expanded, mutant huntingtin gene while sparing the wild-type gene. The drug was being evaluated in the Phase 1 POINT-HD safety study, which had successfully dosed only three human participants.
However, Roche ran concurrent long-term animal studies to evaluate chronic repeated dosing. These animal studies surfaced safety findings that led Roche to conclude RG6496 could not be safely administered over the long-term horizons required for treating a chronic neurodegenerative disease. Pointing to this as the cause for immediate termination, Mai-Lise Nguyen, patient partnership leader at Roche, wrote:
"These are independent, data-driven events, which have coincided by chance."
Divergent Path for uniQure's AMT-130 Gene Therapy
The failure of Roche's CSF-delivered ASO program contrasts sharply with the success of uniQure's AMT-130 gene therapy program. AMT-130 utilizes an adeno-associated virus (AAV5) vector to deliver a microRNA directly into the striatum via a one-time neurosurgical infusion.
At the three-year mark in Phase 1/2 trials, AMT-130 demonstrated a 75% slowing of disease progression compared to external controls. In June 2026, the FDA agreed that this three-year data package can serve as the primary basis for an accelerated approval BLA, which uniQure plans to formally submit in Q3 2026.