Targeted pill after surgery delays relapse in RET-driven early lung cancer

Giving the targeted drug selpercatinib as a pill for up to three years after surgery delayed relapse in early-stage lung cancer driven by a RET gene fusion: in the phase 3 LIBRETTO-432 trial, two-year event-free survival was 92% with selpercatinib against 61% with placebo among patients with stage II or IIIA disease [s1]. The trial was small — 151 patients — and it has not yet shown that the drug helps people live longer, so the result is a strong signal rather than a settled case [s1].

RET fusions are a rare driver mutation, found in a small percentage of non-small-cell lung cancers, in which two genes join to switch on uncontrolled growth. Selpercatinib, a selective RET inhibitor that also reaches the brain, was already approved for advanced RET-fusion lung cancer [s1]. LIBRETTO-432 moved it into the adjuvant setting — treatment given after surgery or radiotherapy has removed all visible disease, with the aim of eliminating micrometastases before they grow into a relapse.

What they did

The trial was double-blind and placebo-controlled, an important detail: neither patients nor doctors knew who was taking the drug, which guards against the bias that dogs open-label cancer trials [s1]. It enrolled patients with RET fusion-positive non-small-cell lung cancer who had completed definitive treatment — surgery, or radiotherapy with adjuvant systemic therapy where indicated — and randomly assigned them to adjuvant selpercatinib or placebo for up to three years [s1]. In total 151 patients were assigned, 75 to selpercatinib and 76 to placebo, with a median follow-up of 24 and 27 months in the two groups [s1]. The trial was registered as LIBRETTO-432 and enrolled patients whose tumours were confirmed to carry a RET fusion [s2]. The primary endpoint was investigator-assessed event-free survival in the 109 patients with stage II or IIIA disease — the time until the cancer recurred, progressed or the patient died [s1].

What it showed

Among those stage II or IIIA patients, two-year event-free survival was 92% with selpercatinib and 61% with placebo, a hazard ratio for recurrence, progression or death of 0.17 (95% confidence interval, 0.06 to 0.51; P<0.001) [s1]. A blinded independent review reached a consistent result [s1]. Across the wider group of 151 patients with stage IB, II or IIIA disease, two-year event-free survival was 94% with selpercatinib and 70% with placebo, a hazard ratio of 0.16 (95% confidence interval, 0.06 to 0.48; P<0.001) [s1].

Those hazard ratios are unusually large, but the confidence intervals are wide — a consequence of how few patients and events the trial had — so the true size of the benefit is estimated imprecisely. A hazard ratio of 0.17 implies the rate of recurrence, progression or death was cut by roughly four fifths, but the interval leaves room for a substantially smaller effect [s1]. This is the familiar trade-off in trials of rare mutations: matching a drug to the right target can produce a dramatic effect, but the rarity that makes the target worth hitting also keeps the trials small. The same tension runs through adjuvant trials of other targeted agents, such as belzutifan after surgery for kidney cancer, where an early relapse benefit still has to prove it extends survival. That LIBRETTO-432 was placebo- controlled and double-blind lends its effect estimate more weight than an open-label design would, even at this modest size, because expectations about a known active drug cannot colour when an event is recorded [s1].

The cost side

The most common adverse events were rises in the liver enzymes alanine aminotransferase and aspartate aminotransferase, which reached grade 3 or higher in 17% and 19% of the selpercatinib group, respectively [s1]. Reassuringly on the safety front, the three deaths in the trial all occurred in the placebo group and were due to disease progression, not treatment [s1]. Liver-enzyme elevations of that severity typically require monitoring and dose interruption, and taking a drug for up to three years is a substantial commitment for someone who may already have been cured by surgery.

What to watch

LIBRETTO-432 was funded by Eli Lilly, which makes selpercatinib [s1]. The decisive limitation is that event-free survival, not overall survival, was the endpoint: delaying relapse is valuable, but whether it translates into longer life is unproven, and some adjuvant drugs that delay recurrence never show a survival gain [s1]. The result also depends entirely on molecular testing — it applies only to the small share of lung cancers found to carry a RET fusion, which requires that resected tumours are sequenced for the alteration, still not universal practice. It joins a growing set of driver-matched lung-cancer treatments, from RAS-directed drugs to HER2-targeted therapy, that only reach patients whose tumours have been profiled — and none of which changes the fact that most lung cancer is still caught late, when cure is far less likely.

Sources

• [s1] Wu YL, Hochmair M, Yang Y, et al. Selpercatinib in Early-Stage RET Fusion-Positive Non-Small-Cell Lung Cancer. New England Journal of Medicine. 31 May 2026.
• [s2] ClinicalTrials.gov. A Study of Selpercatinib (LY3527723) in Participants With Early-Stage RET Fusion-Positive Non-Small Cell Lung Cancer (LIBRETTO-432, NCT04819100). U.S. National Library of Medicine.