Early Tofersen Slows SOD1-ALS Progression Over 2+ Years — Neurofilament as a Biomarker of Benefit

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease that destroys upper and lower motor neurons, causing progressive paralysis and death — median survival is 2–5 years from symptom onset. Approximately 20% of familial ALS cases carry mutations in the SOD1 gene, which encodes superoxide dismutase 1. In SOD1-ALS, the mutant protein accumulates in motor neurons and generates toxic gain-of-function effects. For three decades, this mechanism was identified but untreatable. Tofersen changed that: it is an antisense oligonucleotide (ASO) delivered intrathecally that binds SOD1 mRNA and silences it, reducing SOD1 protein production in the spinal cord.[1]

The original VALOR trial randomized 108 SOD1-ALS patients to tofersen 100 mg intrathecally every 4 weeks or placebo for 28 weeks. The primary endpoint — ALSFRS-R decline (ALS Functional Rating Scale–Revised, a 48-point functional measure) — did not meet statistical significance in the overall population, but a prespecified faster-progressing subgroup showed meaningful slowing. Critically, tofersen substantially reduced plasma NfL (neurofilament light chain, a biomarker of neuronal injury) — a finding that predicted subsequent benefit in the open-label extension. FDA granted accelerated approval based on the NfL reduction as a reasonably likely surrogate biomarker.[1, 2]

The JAMA Neurology report presents 2+ year data from the open-label extension, comparing participants who received tofersen throughout (early-start group) versus those who crossed over from placebo after 28 weeks (delayed-start group). The early-start group showed sustained and superior functional preservation compared to the delayed-start group — the gap in ALSFRS-R trajectories widened over time rather than narrowing, consistent with disease-modifying rather than merely symptomatic benefit. The delayed-start group never fully caught up, even after 2 years of tofersen treatment, suggesting irreversible motor neuron loss during the initial placebo period. NfL levels normalized toward normal in the early-start group and showed the same improvement trajectory after crossover in the delayed group, confirming it as a reliable pharmacodynamic marker.[1]

Clinical Context

Tofersen's development exemplifies the promise and complexity of antisense oligonucleotide (ASO) therapeutics in neurodegeneration. The VALOR trial (2022, NEJM) randomized 108 patients with SOD1-ALS to intrathecal tofersen or placebo for 28 weeks. The primary endpoint — ALS Functional Rating Scale-Revised (ALSFRS-R) slope — was not met: a 1.2-point difference in favor of tofersen did not reach statistical significance (p=0.97 against a pre-specified threshold of p<0.01).[2] Yet tofersen substantially reduced plasma neurofilament light chain (NfL) — a biomarker of ongoing axonal injury — by 60% versus placebo, and a post-hoc analysis of the fastest-progressing patients showed a statistically significant functional benefit. The FDA approved tofersen in April 2023 under the accelerated approval pathway based on plasma NfL reduction, making it the first disease-modifying therapy approved for a genetic form of ALS.

The critical scientific question tofersen raises is whether NfL reduction as a surrogate endpoint translates to durable functional and survival benefit — and whether earlier initiation in presymptomatic mutation carriers prevents neuronal loss before it manifests clinically. The ATLAS trial — ongoing at the time of VALOR publication — tests tofersen in presymptomatic SOD1 mutation carriers, guided by rising NfL as a biomarker trigger. The open-label extension data now reported in JAMA Neurology follow patients from VALOR through extended treatment, directly addressing whether the NfL signal translates to preserved function over a longer horizon than the 28-week pivotal trial could capture.[3]

Why It Matters Clinically

Tofersen represents the first precision medicine approach to ALS — targeting a specific molecular mechanism in a genetically defined subpopulation. The broader significance is the validation of ASO technology in a disease where drug delivery to the CNS is inherently difficult. The intrathecal route bypasses the blood-brain barrier, and monthly lumbar puncture-based dosing is feasible in an ALS population. The same ASO technology platform is being applied to other familial ALS causes (FUS-ALS, TDP-43-ALS) and to Huntington's disease, where a similar silencing approach is in advanced trials. The VALOR long-term data make the case that early gene silencing before irreversible injury is the right disease-modification strategy — a lesson likely to generalize.[3]

Limitations

SOD1-ALS accounts for only ~2% of all ALS. The 108-patient VALOR trial and its extension are not powered for mortality endpoints. The original trial missed its primary endpoint in the overall population, and the accelerated approval rests on surrogate marker (NfL) data — confirmatory clinical benefit trials are required. Not all SOD1 mutations respond equally; some variants have faster progression trajectories and may respond differently to tofersen.