Merck stopped its sotatercept PAH trial early — for benefit, not failure

A terminated Phase 3 trial usually signals a drug in trouble. HYPERION is the opposite: Merck stopped its study of sotatercept (Winrevair) in pulmonary arterial hypertension (PAH) precisely because the drug was working — and keeping patients on placebo could no longer be justified.

The ClinicalTrials.gov record lists NCT04811092 as TERMINATED, with the stated reason being “loss of clinical equipoise based on robust evidence of clinical benefit of sotatercept demonstrated in previous studies.” Under Amendment 11, the registry notes, the study was closed so eligible participants could receive sotatercept via the SOTERIA extension study (NCT04796337) or through commercial access. Sotatercept is an already-approved PAH therapy that is added to background treatment and is initiated and managed by PAH specialists — not a stand-alone or self-started medication.

Who was studied

This was not “all PAH.” Per the registry’s eligibility criteria, HYPERION enrolled adults with WHO Group 1 PAH (idiopathic, heritable, drug/toxin-induced, connective-tissue-disease-associated, or post-shunt-correction subtypes), diagnosed within 12 months of screening, classified WHO functional class II or III, and at intermediate-to-high risk of progression. All participants were already on stable doses of double or triple background PAH therapy (plus diuretics if needed) for at least 90 days before screening. Sotatercept or placebo was added on top of that regimen.

What the closed trial showed

HYPERION was a randomized, double-blind, placebo-controlled study that enrolled 321 such participants. The primary endpoint was median time to clinical worsening (TTCW) — time from randomization to a first adjudicated morbidity event or death.

Per the posted results table, the primary endpoint favored sotatercept with a hazard ratio of 0.24 (95% CI 0.14–0.41; p<0.0001). Median TTCW was not reached in the sotatercept arm versus 23.0 months (95% CI 17.3–NA) on placebo. Serious adverse events occurred in 39 sotatercept-treated participants and 45 placebo-treated participants.

A hazard ratio of 0.24 implies a roughly 76% lower risk of clinical worsening — the kind of margin that erases equipoise.

Because the trial was halted by design rather than run to its planned ~35-month readout, these figures should be read as the data available at early termination, not a completed event-driven analysis. An association from a curtailed trial is not the same as a full survival result, and this is not medical advice.