The investigational targeted therapy divarasib extends overall survival in patients with previously treated, advanced KRAS G12C-mutated non-small cell lung cancer (NSCLC), outperforming two existing KRAS inhibitors, according to findings from the phase 3 Krascendo 1 trial.

Among 338 patients with locally advanced or metastatic disease, second-line divarasib cut the risk for progression or death by 42% and boosted overall survival by almost 6 months vs sotorasib and adagrasib.

Those two drugs, both first-generation KRAS G12C inhibitors, are approved in the US and Europe as second-line therapies for patients with advanced G12C-mutated NSCLC. But neither has been shown to improve overall survival compared with docetaxel, said Krascendo investigator Ferdinandos Skoulidis, MD, PhD, of The University of Texas MD Anderson Cancer Center in Houston.

Based on the new findings, divarasib “represents a new standard of care” for this patient population, Skoulidis said during a presentation at the European Society for Medical Oncology (ESMO) Annual Meeting 2026 in Madrid, Spain.

The findings were simultaneously published online on October 25 in The New England Journal of Medicine.

KRAS G12C is the most common KRAS mutation in NSCLC, occurring in up to 14% of patients with nonsquamous disease. Divarasib, a next-generation oral KRAS G12C inhibitor, has been shown to be more potent and selective than sotorasib and adagrasib in the lab. Krascendo 1 looked at whether that translates into better outcomes for patients.

The open-label trial involved 338 patients who’d previously received one to three lines of systemic therapy, including platinum-based chemotherapy, a PD-(L)1 inhibitor and, potentially, docetaxel.

About half of patients were randomly assigned to receive divarasib at 400 mg once daily. In the control group, 91% vs 9% received sotorasib (960 mg once daily) vs adagrasib (600 mg twice daily). Treatment continued until disease progression, unacceptable adverse events, loss of clinical benefit, withdrawal of consent, death, or trial termination.

After a median of 14.3 months, median progression-free survival, the primary endpoint, was 9.8 months vs 5.7 months with divarasib vs the control group (hazard ratio [HR], 0.58; P < .0001). Median overall survival was 17.7 months vs 12 months, respectively (HR, 0.56; P = .0005).

The overall survival benefit was generally consistent across patient subgroups, Skoulidis said, including patients with a history of brain metastasis and those with KEAP1-mutant tumors.

Patients in the divarasib arm were able to stay on treatment significantly longer than those receiving the older agents (11.3 months vs 5 months). The rate of grade 3 or 4 treatment-related adverse events was lower with divarasib (24.6% vs 34.5% with the comparator drugs).

However, divarasib came with more gastrointestinal toxicity, including diarrhea, nausea, and vomiting of any grade. Diarrhea occurred in about three quarters of patients receiving divarasib, and grade 3 or 4 diarrhea was twice as common with the new agent (9.6% vs 4.1% with the comparator drugs).

Liver toxicity, in contrast, appeared less frequent with divarasib, based on ALT and AST measurements. On balance, Skoulidis said, “divarasib exhibited a manageable and, I would argue, favorable safety profile compared to sotorasib or adagrasib.”

However, study discussant Benjamin Besse, MD, PhD, of Paris-Saclay University in Paris, France, pointed out that information on the duration of adverse events was lacking. 

“Is the diarrhea there for 10 months, or is the diarrhea here for just 2 days and then the patient has a normal life?” Besse said. He added that clinical trials broadly need better ways to report adverse events, including scores that consider grade and duration (a call that prompted applause from the audience).

Besse also noted that many patients in the real world would not be eligible for Krascendo 1 and similar trials, and outcomes might therefore be less impressive outside of studies.

Still, speaking at an ESMO press conference, Besse agreed that the new results are likely to change the standard of care for patients with advanced G12C-mutated NSCLC.

The “big question” going forward, Besse said, is whether divarasib could have a role in first-line treatment.

The trial was funded by F. Hoffmann-La Roche. Skoulidis reported financial ties to the company. Besse reported indirect relationships with the company through his institution.