Divarasib Displays Early-Phase Activity in KRAS G12C–Mutant Solid Tumors

Treatment with the KRAS G12C inhibitor divarasib (formerly GDC-6036) led to durable responses with a tolerable safety profile in patients with a variety of solid tumors harboring a KRAS G12C mutation.

Treatment with the KRAS G12C inhibitor divarasib (formerly GDC-6036) led to durable responses with a tolerable safety profile in patients with a variety of solid tumors harboring a KRAS G12C mutation, according to findings from a phase 1 trial (NCT04449874) published in the New England Journal of Medicine.

At the November 21, 2022, data cutoff, patients with non–small cell lung cancer (NSCLC; n = 58) achieved a confirmed response rate of 53.4% (95% CI, 39.9%-66.7%) with a median progression-free survival (PFS) of 13.1 months (95% CI, 8.8-not estimable [NE]). Among patients with colorectal cancer (CRC; n = 55), these figures were 29.1% (95% CI, 17.6%-42.9%) and 5.6 months (95% CI, 4.1-8.2), respectively.

Notably, dose-limiting toxicities and treatment-related deaths were not observed relating to divarasib treatment. Dose reduction or discontinuation due to treatment-related adverse effects (TRAEs) occurred at rates of 14% and 3%, respectively, in the safety population (n = 137). Thirty-one percent of patients experienced serious adverse effects (AEs), leading to dose modifications and discontinuation in 17% and 3% of patients, respectively.

Divarasib is a selective, highly potent covalent KRAS G12C inhibitor that binds to cysteine residue and locks the protein into its inactive state. In vitro, the agent has displayed potency 5 to 20 times greater than that of other KRAS G12C inhibitors, such as sotorasib (Lumakras) and adagrasib (Krazati), with as much as 50-times selectivity.

The open-label, multicenter, dose-escalation, and dose-expansion phase 1 trial enrolled adult patients with locally advanced or metastatic solid tumors with a KRAS G12C mutation and an ECOG performance status of 1 or less. Eligible patients had disease that progressed following treatment with at least 1 standard therapy, disease for which standard therapy was ineffective for or led to intolerable AEs, or disease for which a clinical trial of an investigational agent is the standard of care. Those who previously received a KRAS G12C inhibitor, were treated withanother anticancer therapy within 3 weeks or 5 half-lives prior to the start of study treatment, were treated with radiation within 4 weeks before divarasib initiation, or who had active, untreated central nervous system metastases were excluded from the study.

Oral divarasib was given once daily at dose levels ranging from 50 mg to 400 mg in 21-day cycles until unacceptable toxicity, disease progression, or patient withdrawal. The single-patient dose-escalation cohorts received the agent at the 50-mg and 100-mg dose levels, then a 3+3 design was used for the 200-mg and 400-mg cohorts, which enrolled 3 additional patients to each. Thereafter, patients were enrolled to the 400-mg dose-expansion cohorts, which included a cohort of patients who underwent a biopsy before treatment and approximately 1 to 2 weeks following the initiation of treatment—these patients received divarasib at a dose of 100 mg to 400 mg.

The primary end point was safety. Additionally, pharmacokinetics, preliminary antitumor activity, and identification ofbiomarkers of response and resistance were assessed.

In the overall population, the median age was 65 years (range, 30-85). Most patients were female (56%), White (80%), and had an ECOG performance status of 1 (58%). Patients received a median of 2 (range, 0-8) prior lines of systemic therapy, including 0 (1%), 1 (24%), 2 (28%), 3 (20%), and 4 or more (27%) prior lines.

Additional findings revealed that responses were also observed among patients with other solid tumors (n = 22), all of whom received divarasib at a dose of 400 mg. Thirty-six percent of patients achieved a partial response (PR), stable disease (SD) was reported in 50%, and progressive disease (PD) as best response occurred in 1 patient. Two patients discontinued treatment before first tumor assessment.

In the NSCLC cohort, patients with measurable disease at baseline experienced a complete response (CR; 2%), PR (59%), SD (29%), and PD (7%) as their best response; 3% discontinued treatment before the first tumor assessment. The median time to response was 1.3 months (range, 1.2-8.3) and the median duration of response (DOR) was 14.0 months (95% CI, 8.3-NE). Patients who received divarasib at a dose of 400 mg (n = 39) experienced a confirmed response rate of 56.4% (95% CI, 39.6%-72.2%) and a median PFS of 13.7 months (95% CI, 8.1-NE); the median DOR was 11.9 months (95% CI, 6.9-NE).

Among patients with CRC, best responses included CR (2%), PR (35%), SD (49%), and PD (11%). Two patients discontinued treatment before the first tumor assessment. The median time to response was 2.2 months (range, 1.2-6.8) and the median DOR was 7.1 months (95% CI, 5.5-7.8). The confirmed response rate for patients who received divarasib at a dose of 400 mg (n = 39) was 35.9% (95% CI, 21.2%-52.8%) with patients experiencing a median DOR of 7.7 months (95% CI, 5.7-NE). The median PFS was 6.9 months (95% CI, 5.3-9.1).

Pharmacokinetic data collected from 4 patients showed that the mean half-life of 400 mg of divarasib was 17.6 hours (standard deviation ± 2.7 hours). The median time to maximum concentration was 2.0 hours (range, 0.5-8.0), the mean maximum concentration was 657 ng/mL (standard deviation ± 185 ng/mL), and the mean area under the curve was 9130 ng x hours/mL (standard deviation ± 3160 ng x hours/mL) in 76 patients across tumor types with evaluable pharmacokinetic data.

In terms of safety, any grade TRAEs occurred in 93% of patients, with 11% experiencing a grade 3 event and 1% reporting a grade 4 event. Any grade and grade 3 to 5 TRAEs were reported in the 60-patient NSCLC cohort (93% and 18%, respectively) and in the CRC cohort (96%; 7%). Common any grade and grade 3 to 5 TRAEs overall included nausea (74%; 1%), diarrhea (61%; 4%), and vomiting (58%; 1%).

The median treatment duration was 6.9 months (range, 0-24.1) in the overall population, 8.3 months (range, 0-24.1) in the NSCLC cohort, and 5.5 months (range, 0.2-15.6) in the CRC group. Overall, 69% of patients discontinued treatment with divarasib, primarily due to progressive disease (50%), clinical progression (7%), and AEs (5%).

Reference

Sacher A, LoRusso P, Patel MR, et al. Single-agent divarasib (GDC-6036) in solid tumors with a KRAS G12C mutation. N Engl J Med. 2023;389(8):710-721. doi:10.1056/NEJMoa2303810