Reviews/Motor Neurone Disease Treatment/Tofersen lowers neurofilament
DMT · MND-DMT-04 · v2
Across the studies here, tofersen is consistently linked to falling neurofilament levels in people with SOD1 motor neurone disease. The strongest study [1] was a randomised trial that compared the drug against a placebo and still saw neurofilament fall more on the drug. Six further real-world studies [3, 4, 5, 6, 7, 8], from different countries and centres, all point the same way, some with declines holding out to a year or more [3, 5, 7]. The panel noted that the same randomised trial did not show a benefit on how patients actually functioned [1], but the claim here is only about the neurofilament marker, not about clinical improvement. Most of the supporting studies had no comparison group and were small, which can exaggerate how large a fall looks, but the one controlled trial fixes the direction of the effect.
A placebo-controlled phase 3 trial reports that tofersen 'led to greater reductions in concentrations of SOD1 in CSF and of neurofilament light chains in plasma than placebo' [1]. Six independent observational cohorts report the same direction: serum NfL falling from a median 78.0 to 36.0 pg/ml and CSF pNfH from 2226 to 1151 pg/ml, each p=0.02 [3]; serum NfL decreases in D91A carriers [4]; neurofilament light and heavy chain still reduced versus baseline at 12 months [5]; mean serum NfL −57.9% and CSF pNFH −67.6% [6]; CSF and serum NFL 'markedly decreased in the near totality of patients' [7]; and neurofilament falling during treatment while SOD1 activity stayed stable [8]. The direction is unanimous and no study in the corpus reports a null or opposite neurofilament result. Durability to 12 months or beyond is reported in several cohorts [3, 5, 7]. The contested point: The division is between the marker and the patient: the same randomised trial that shows neurofilament falling also found no significant clinical benefit, with the ALSFRS-R difference at 1.2 points and P=0.97 [1]. The panel treats this as a boundary on interpretation, not a contradiction, because the claim is worded only about neurofilament.
All 8 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Motor Neurone Disease Treatment, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
Six of the seven included sources are open-label/observational with no control arm; NfL declines in uncontrolled cohorts are vulnerable to regression to the mean and retention bias, which may inflate magnitude estimates.
Neurofilament is a surrogate biomarker: the VALOR RCT's primary clinical endpoint (ALSFRS-R) was not met (difference 1.2 points, P=0.97), so neurofilament reduction must not be read as clinical efficacy — though this does not falsify the biomarker claim as worded.
Where it remains uncertain. The controlled trial's clinical endpoint was not met (difference 1.2 points, P=0.97) [1], so the neurofilament fall should not be read as proof of clinical efficacy.
Aleatory
Individual effect estimates rest on few patients in [4] (N=11), [6] (N=7) and [7] (N=17), so the size of any one cohort's fall is imprecise.
Epistemic
Six of seven sources lack a control arm [3, 4, 5, 6, 7, 8], so how much of the observed decline is drug effect versus regression to the mean or retention bias is unresolved.
Inferential
A neurofilament fall does not establish patient benefit; the one controlled trial saw no significant clinical effect (difference 1.2 points, P=0.97) [1].
The load-bearing papers from a corpus of 8, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Placebo-controlled phase 3 RCT confirming the neurofilament reduction against the correct comparator.
“Tofersen led to greater reductions in concentrations of SOD1 in CSF and of neurofilament light chains in plasma than placebo.”
Panel auditInternal validity: high · scope of inference: Adults with SOD1 ALS over a 24-28 week randomized period, with open-label extension to 52 weeks.Key limitations: no numeric effect size or p value for neurofilament in abstract · clinical endpoints did not differ significantly · neurologic serious adverse events in 7%
Statistically significant serum NfL and CSF pNfH reduction in a real-world cohort.
“Median serum NfL declined from 78.0 pg/ml (IQR 37.0-147.0 pg/ml; n = 23) to 36.0 pg/ml (IQR 22.0-65.0 pg/ml; n = 23; p = 0.02), median pNfH in CSF from 2226 pg/ml (IQR 1061-6138 pg/ml; n = 18) to 1151 pg/ml (IQR 521-2360 pg/ml; n = 18; p = 0.02).”
Panel auditInternal validity: medium · scope of inference: 24 German SOD1 ALS patients followed up to 12 months in routine early-access care.Key limitations: no control group · small sample · potential autoimmune reaction in two patients
Neurofilament reduction confirmed in a specific SOD1 variant subgroup.
“Tofersen decreases serum neurofilament levels (sNFL), which are associated with the ALS progression rate, in the 6 ALS patients homozygous for SOD1^D91A.”
Panel auditInternal validity: medium · scope of inference: 11 patients homozygous or heterozygous for the SOD1 D91A variant treated up to 16 months.Key limitations: very small sample · single variant only · no control group · possibly reduced effect in heterozygotes
3 of 8 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
An RCT-anchored, unanimous, independently replicated biomarker effect warrants the Established tier, held just below the top of the band by the predominance of uncontrolled designs.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Controlled replication of the biomarker effect. Only one study [1] carries a placebo comparator; the six cohorts [3, 4, 5, 6, 7, 8] cannot separate drug effect from regression to the mean. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A larger or longer placebo-controlled or delayed-start design reporting neurofilament as a prespecified endpoint across the SOD1 population.
Open question: How much of the neurofilament decline in uncontrolled cohorts is attributable to tofersen rather than regression to the mean or retention bias? Why it is still open: Six of seven supporting studies lack a control arm [3, 4, 5, 6, 7, 8]; only [1] provides a placebo comparator.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19