Reviews/Motor Neurone Disease Treatment/Sodium phenylbutyrate-taurursodiol prolongs survival
DMT · MND-DMT-03 · v2
Every piece of survival evidence for this drug combination comes from one mid-stage trial called CENTAUR and its follow-on analyses [1][2][3]. In that trial the survival measure was a secondary goal, and when the groups were compared as the trial was designed, the difference was not reliable [1]. The encouraging survival numbers come from analyses done after the fact or by comparing the treated patients to an outside database rather than to the trial's own control group [2], and from a longer follow-up in which most placebo patients later switched to the active drug [3]. The direction of all these signals favours the treatment, but none of them is an independent, clean, head-to-head survival result. Because the findings rest on a single small trial and a failed prespecified endpoint, the panel treated the survival claim as not yet carried by the evidence.
All survival data trace to the single CENTAUR phase 2 trial, which randomized 137 participants to active drug or placebo [1]. Its prespecified, randomized survival secondary endpoint did not separate the groups: 'Secondary outcomes did not differ significantly between the two groups' [1]. A post-hoc analysis against a propensity-matched external control cohort from the PRO-ACT database reported a 'hazard ratio, 0.48; 95% CI, 0.31-0.72; p = 0.00048' [2], and the trial's own ITT read gave 'median overall survival (OS) was 4.8 months longer and risk of death 36% lower' [2]. A long-term analysis of the originally randomized groups reported 'Risks of death or tracheostomy/PAV (HR=0.51; 95% CI 0.32 to 0.84; p=0.007)' [3]. The consistent direction favours benefit, but the robust comparisons are post-hoc or non-randomized. The contested point: The division is between the direction of the survival signals, which uniformly favour treatment [2][3], and their provenance: the only randomized, prespecified survival comparison was null [1], the strongest hazard ratio rests on a non-randomized external control [2], and the long-term contrast is confounded because '71% of whom received delayed-start PB/TURSO in the OLE phase' [3]. Whether these post-hoc, single-trial signals can support a survival claim is the point at issue.
All 5 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.
The only prespecified, randomized survival comparison was non-significant: in CENTAUR [1] survival was a secondary endpoint and 'Secondary outcomes did not differ significantly between the two groups.' The positive survival numbers are post-hoc, not the randomized result.
Evidential independence is essentially nil — [1], [2] and [3] all draw on the single CENTAUR trial (n approximately 137). No independent cohort replicates a survival effect.
Where it remains uncertain. The one randomized, prespecified survival comparison was null — 'Secondary outcomes did not differ significantly between the two groups' [1] — so the positive numbers are post-hoc or non-randomized [2][3].
Aleatory
The entire survival record rests on 137 randomized participants [1], so chance variation in a cohort this size materially affects every derived estimate.
Epistemic
[1], [2] and [3] all draw on CENTAUR; no separate cohort tests whether a survival effect reproduces.
Inferential
The HR 0.48 compares the active arm to a propensity-matched PRO-ACT cohort [2]; propensity matching cannot remove unmeasured confounding, and the authors note it 'suggests potentially greater survival benefit... than seen on ITT analysis.'
The load-bearing papers from a corpus of 5, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
The prespecified randomized survival endpoint (time to death, tracheostomy, or permanent ventilation) was NOT significant, undercutting a definitive survival claim.
“Secondary outcomes did not differ significantly between the two groups.”
Panel auditInternal validity: high · scope of inference: Definite ALS, symptom onset within 18 months, over 24 weeks of randomized treatment.Key limitations: survival only a secondary outcome · short 24-week duration · modest sample size · secondary outcomes not significant
ITT survival signal from CENTAUR favouring active drug.
“hazard of death was 52% lower in the former group (hazard ratio, 0.48; 95% CI, 0.31-0.72; p = 0.00048)”
Panel auditInternal validity: low · scope of inference: ALS participants randomized to PB and TURSO in CENTAUR versus matched prior-trial controls.Key limitations: external/historical controls not randomized · post hoc analysis · residual confounding possible · authors frame it as suggestive
Long-term composite survival benefit in originally-randomized groups, though confounded by open-label crossover.
“Risks of death or tracheostomy/PAV (HR=0.51; 95% CI 0.32 to 0.84; p=0.007)”
Panel auditInternal validity: medium · scope of inference: El Escorial Definite ALS, ≤18 months from onset, followed up to 35 months post-randomisation.Key limitations: composite survival endpoint · placebo-to-active crossover in OLE · long-term follow-up of a small cohort
3 of 5 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.
Positive survival signals are real but post-hoc, non-randomized and confined to a single small trial whose prespecified randomized survival endpoint did not reach significance.
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.
No independent replication of the survival signal. Every survival estimate derives from a single phase 2 trial [1][2][3], so the direction of effect has never been confirmed in a separate population. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: An adequately powered, independent randomized controlled trial with survival as a prespecified primary endpoint.
Open question: Does PB/TURSO prolong survival when survival is the prespecified primary endpoint of an adequately powered randomized trial? Why it is still open: In CENTAUR survival was a secondary endpoint and did not differ significantly between groups [1]; no corpus trial tests it as a randomized primary.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19