Dr Skoulidis on Divarasib vs Sotorasib or Adagrasib in KRAS G12C–Mutated NSCLC

“Most importantly, in my opinion, the Krascendo 1 clinical trial also met the key secondary end point of overall survival early, at the time of the first interim overall survival analysis.”

Ferdinandos Skoulidis, MD, PhD, MRCP, a professor in the Department of Thoracic/Head and Neck Medical Oncology in the Division of Cancer Medicine at The University of Texas MD Anderson Cancer Center, discussed efficacy findings from the phase 3 Krascendo 1 trial (NCT06497556) evaluating divarasib vs sotorasib (Lumakras) or adagrasib (Krazati) in patients with previously treated KRAS G12C–mutated non–small cell lung cancer, as well as the safety profile of divarasib compared with these first-generation KRAS G12C inhibitors.

At the 2026 ESMO Congress, Skoulidis presented primary results from Krascendo 1, which enrolled 338 patients with previously treated advanced or metastatic KRAS G12C–mutated non–small cell lung cancer (NSCLC). Patients received divarasib at 400 mg once daily (n = 167) or a control KRAS G12C inhibitor (n = 171): sotorasib at 960 mg once daily (n = 156) or adagrasib at 600 mg twice daily (n = 15). The results were simultaneously published in The New England Journal of Medicine.

The trial met its primary end point of progression-free survival (PFS) by blinded independent central review. The median PFS was 9.8 months (95% CI, 8.1-12.9) with divarasib vs 5.7 months (95% CI, 8.1-12.9) with sotorasib or adagrasib (HR, 0.58; 95% CI, 0.44-0.76; P < .0001), resulting in a 42% reduction in the risk of disease progression or death. The PFS benefit was consistent across major clinically and molecularly defined subgroups, Skoulidis noted, including patients with a history of central nervous system metastases, those with liver metastases at baseline and those with KEAP1-mutated tumors, a group with an adverse prognosis.

Divarasib also met the key secondary end point of overall (OS) survival at the first interim analysis, which Skoulidis called the most important finding. The median OS was 17.7 months (95% CI, 8.1-12.9) with divarasib vs 12.0 months (95% CI, 8.1-12.9) with the control (HR, 0.56; 95% CI, 0.41-0.78; P = .0005), translating to a 44% reduction in the risk of death.

Beyond being manageable, the safety profile of divarasib was favorable by many metrics compared with the control, which Skoulidis said surprised him. Grade 3 or 4 treatment-related adverse effects (TRAEs ) occurred in 24.6% of patients who received divarasib vs 34.5% of those who received the control, and TRAEs led to discontinuation in 4.8% vs 11.7%, respectively, even though the median treatment duration was more than twice as long with divarasib (11.3 vs 5.0 months).

The most common TRAEs with divarasib were gastrointestinal, consistent with known class effects: diarrhea (66% vs 38% with the control), nausea (53% vs 20%) and vomiting (39% vs 8%). Most were grade 1 and, according to Skoulidis, manageable with standard supportive measures and reversible. Hepatotoxicity was less frequent with divarasib, and no treatment-related QT prolongation or pneumonitis occurred with divarasib vs 2 and 3 patients, respectively, in the control arm.

Sotorasib and adagrasib hold accelerated FDA approvals in this setting (sotorasib, adagrasib). Divarasib remains investigational; the FDA granted the agent breakthrough therapy designation in 2022 and orphan drug designation in 2026, according to Genentech. Divarasib plus pembrolizumab (Keytruda) is being compared with pembrolizumab plus chemotherapy in the frontline setting in the phase 3 Krascendo 2 trial (NCT06793215), which Skoulidis discussed previously.