An oral RAS inhibitor shrank tumours in about a third of RAS-mutant lung cancers

RAS mutations are, as a group, the most common oncogenic drivers of non-small-cell lung cancer (NSCLC), occurring in roughly 30% of patients, and they have been one of the hardest targets in cancer to drug [s1]. A trial published in the New England Journal of Medicine on 2 September reports early results for an oral drug designed to hit them broadly, and the results are promising and preliminary in equal measure [s1].

What the drug is

The drug is daraxonrasib (RMC-6236), described as an oral, multiselective, tri-complex inhibitor of the active (GTP-bound) forms of both mutant and wild-type RAS protein isoforms [s1]. The design point is breadth: rather than targeting a single RAS mutation, it aims at the RAS pathway across variants — which is what makes it potentially relevant to the roughly 30% of NSCLC driven by RAS as a class, not just the narrower slice addressed by mutation-specific drugs [s1].

What the trial was — and was not

This was a phase 1–2, multicenter, dose-escalation and dose-expansion study — the kind of trial designed first to find a safe dose and look for signals of activity, not to prove benefit against a comparator [s1]. Patients had previously treated advanced RAS-mutant NSCLC and received daraxonrasib at doses of 10 to 400 mg once daily in 21-day cycles [s1]. The primary end point was safety; response and its duration were secondary end points, assessed by the investigators using RECIST version 1.1 [s1].

Two features of that design bound what the results can say. There is no control group, so response rates cannot be compared against a placebo or standard of care within the trial. And the responses were investigator-assessed rather than read by a blinded independent committee, which tends to run slightly more generous.

The efficacy signal

As of the 21 July 2025 data cutoff, 136 patients treated at doses of 300 mg or less had been evaluated for safety and efficacy [s1]. The share of patients with an objective response — a complete or partial tumour shrinkage — was 31% at a dose of 120 mg or less, 34% at 160 to 220 mg, and 37% at 300 mg [s1].

That is a coherent, dose-graded signal in a group of patients whose cancer had already progressed through prior treatment, and for a target long considered "undruggable" it is a genuine result. It is also, at 31–37%, a minority of patients, in an uncontrolled trial, with response measured by investigators.

The "previously treated" detail matters to how the number should be read. These were patients whose disease had already advanced through earlier therapy, a setting where response rates to any drug tend to run lower and where new options are genuinely scarce. Much of the recent progress against RAS in lung cancer has been confined to one specific mutation, KRAS G12C; a drug aimed at RAS across variants speaks to the larger group of patients that mutation-specific drugs leave out, which is part of why an early signal here drew attention despite the trial's limits.

The safety cost

The trade-off is not subtle. Adverse events of any grade occurred in 99% of patients at doses of 300 mg or less, with rash, diarrhea, nausea, vomiting, and mucositis or stomatitis each affecting at least 30% [s1]. Adverse events of grade 3 or higher — the serious end — occurred in 54% of patients, including pneumonia in 10%, diarrhea in 9%, rash in 8%, and anemia in 5% [s1]. Four grade 5 adverse events, the category that includes deaths, were reported [s1].

The authors' own conclusion holds both halves together: at doses of 300 mg or less daily, 54% of patients had adverse events of grade 3 or higher, and antitumor activity was reported in more than 30% [s1].

What to watch

The result that would change practice is not in this paper. An early-phase, single-arm response rate tells you a drug does something; it does not tell you whether patients live longer or better than they would on existing options. That requires a randomised comparison, and the natural next step is a phase 3 trial against standard therapy in RAS-mutant NSCLC. The trial was funded by the drug's maker, Revolution Medicines (RMC-6236-001; ClinicalTrials.gov number NCT05379985) [s1].

This article summarises a single early-phase trial and does not offer medical advice. Dosing decisions for any cancer therapy rest with a treating oncologist.

Sources

• Daraxonrasib for Previously Treated RAS-Mutant Non-Small-Cell Lung Cancer, New England Journal of Medicine, 2 September 2026