KRAS-Targeted Therapy

Can KRAS-mutant pancreatic cancer be treated with a drug that targets KRAS?

It can now be targeted, but not yet reliably held. KRAS was considered undruggable for decades, and pancreatic cancer is the tumour where that mattered most, since mutant KRAS is present in over 90% of cases. Two things have changed. A G12D inhibitor added to standard chemotherapy produced a confirmed response in 63.3% of treatment-naive patients in an early-phase trial, against a historical benchmark of roughly 23% for the chemotherapy backbone alone. And in a separate cohort treated with a RAS(ON) inhibitor, the escape routes have now been mapped rather than merely suspected: most tumours that progressed had acquired a RAS pathway alteration, most often amplification of the mutant KRAS allele itself. Neither trial was randomised, and no survival benefit has been demonstrated for either drug.

What the evidence shows

• In advanced KRAS-G12D-mutant pancreatic ductal adenocarcinoma, the KRAS-G12D inhibitor HRS-4642 added to gemcitabine and nab-paclitaxel met its primary endpoint with a confirmed objective response rate of 63.3% (95% CI, 43.9-80.1) in 30 treatment-naive patients, at a median follow-up of 12.3 months. Nature Medicine
• That figure is set against a historical objective response rate of roughly 23% for the gemcitabine and nab-paclitaxel backbone alone — a cross-study benchmark rather than a randomised comparison within the trial. Nature Medicine
• Targeting KRAS did not come cheaply: grade 3 or worse treatment-related adverse events occurred in 90.3% of patients, predominantly haematologic, though none led to treatment discontinuation or death. Nature Medicine
• Resistance is already characterised for this drug class. Among 44 patients with RAS-mutant metastatic pancreatic cancer who progressed on the RAS(ON) inhibitor daraxonrasib, paired circulating tumour DNA profiling found acquired RAS pathway alterations in 26 (59%). Nature Medicine
• Amplification of the mutant KRAS allele was the dominant escape route, in 16 of those 44 patients (36%) — the tumour making more of the target rather than altering it. Nature Medicine
• No patient developed a secondary KRAS mutation, which distinguishes resistance to a RAS(ON) tri-complex inhibitor from the resistance patterns reported with mutant-selective KRAS G12C(OFF) inhibitors. Nature Medicine
• A pretreatment TP53 mutation was associated with acquired KRAS amplification at progression, in 47% of TP53-mutant tumours versus 8% of TP53 wild-type tumours (P = 0.03). Nature Medicine
• No genomic driver of resistance could be identified in roughly 40% of progressing tumours, leaving non-genomic mechanisms that circulating tumour DNA sequencing does not capture still to be characterised. Nature Medicine

What this evidence cannot settle

The response data come from an open-label, single-arm phase 1b/2 trial that treated 31 patients, 30 of whom formed the efficacy population, with follow-up still maturing. Its comparator is a historical cross-study figure, not a randomised arm, and a response rate is not a survival benefit; the randomised phase 3 trial NCT07232875 is the test of whether one follows from the other. The resistance analysis describes a different drug in a different trial, so it does not establish how long a response to HRS-4642 lasts or whether the same escape routes apply to it. Both studies are confined to pancreatic cancer with these specific alterations, and neither speaks to KRAS-mutant disease arising in other organs.

This summary was assembled by an automated editorial process from the articles listed below. Every statement above links to the article that establishes it, and each of those articles is checked against its primary source before publication. No clinician has reviewed this page.