Sotorasib Displays Disease Control, Safety in KRAS G12C–Mutant Gastrointestinal Cancers

Sotorasib (AMG 510) was well tolerated and led to disease control in the majority of patients with heavily pretreated KRAS G12C–mutant gastrointestinal (GI) malignancies, according to findings from the phase 1 CodeBreaK100 trial (NCT03600883) that were presented during the ESMO Asia Virtual Congress 2020.

At a median follow-up of 7.7 months (range, 1.2-15.9), no dose-limiting toxicities (DLTs), serious adverse effects (AEs), fatal AEs, or AEs leading to treatment discontinuation were reported.

The confirmed objective response rate (ORR) and disease control rate (DCR) was 7.1% (n = 3) and 76.2% (n = 32), respectively, in patients with colorectal cancer (CRC; n = 42).

Among 15-evaluable patients with other GI malignancies, 1 confirmed partial response (PR) in a patient with appendiceal cancer and 9 cases of stable disease were reported (SD; n = 6 pancreatic; n = 2 appendiceal; and n = 1 bile duct).

“Responses to sotorasib were seen in patients with previously treated CRC, with the majority achieving disease control, resulting in a PFS longer than anticipated with current standard of care,” study author Timothy Jay Price, MBBS, FRACP, DHlthSc, head of clinical oncology research and the combined Hematology and Medical Oncology Service at The Queen Elizabeth Hospital, and clinical professor at the University of Adelaide, said in a virtual presentation during the meeting. “Encouraging antitumor activity was also observed in the subset of patients with other GI tumors.”

KRAS G12C mutations are found in 1% to 3% of all GI malignancies. AMG 510 is a first-in-class KRAS G12C inhibitor. Previously, findings from the phase 1 trial, which were presented during the 2020 ASCO Virtual Scientific Program, demonstrated preliminary safety and antitumor activity in patients with advanced solid tumors harboring KRAS G12C mutations.

To be eligible for enrollment in the multicenter, open-label study, patients had to have a locally advanced or metastatic malignancy, had received prior standard therapies, a KRAS G12C mutation as assessed by molecular testing of tumor biopsies, and no active brain metastases.

In the dose-escalation phase, 2 to 4 patients were each enrolled into 1 of 4 cohorts: 180 mg (cohort 1), 360 mg (cohort 2), 720 mg (cohort 3), and 960 mg (cohort 4). Intra-patient dose escalation was allowed, and additional patients could be added to any cohort if the dose was determined to be safe.

Patients received oral sotorasib daily in 21-day cycles until disease progression, intolerance, or consent withdrawal and underwent radiographic scans every 6 weeks.

In the dose-expansion phase, patients received 960 mg of sotorasib.

DLTs and safety served as the primary end points of the trial. Pharmacokinetics, ORR, duration of response, DCR, progression-free survival (PFS), and duration of SD served as secondary end points.

With regard to baseline characteristics, the median age was 58 years (range, 33-82) and less than half of patients were female (n = 25; 42.4%). The majority of patients had an ECOG performance status (PS) of 0 (n = 21; 35.6%) or 1 (n = 35; 59.3%); 5.1% (n = 3) of patients had an ECOG PS of 2.

All patients had been pretreated with 1 (n = 5; 8.5%), 2 (n = 15; 25.4%), 3 (n = 15; 25.4%), or more than 3 lines of therapy (n = 24; 40.7%). The median number of prior lines of anticancer therapy was 3 (range, 1-4).

As of the data cutoff on January 8, 2020, 59 patients with advanced GI malignancies had been enrolled, spanning CRC (n = 42), pancreatic cancer (n = 10), appendiceal cancer (n = 4), bile duct cancer (n = 1), small bowel cancer (n = 1), and esophageal cancer (n = 1).

Of these patients, 16 remained on treatment and 43 discontinued treatment. Reasons for discontinuation included disease progression (n = 39), patient request (n = 2), and AEs (n = 2). Twenty-eight patients (47.5%) and 10 patients (16.9%) remained on treatment for at least 3 months and 6 months, respectively.

Patients with at least 7 weeks of follow-up were eligible for evaluation.

In the CRC cohort, all 3 objective responses were PRs and were achieved with the highest dose of sotorasib, suggesting a dose-dependent response with the small molecule inhibitor. Among patients with CRC who received the 960-mg dose of sotorasib (n = 25), the ORR and DCR was 12.0% (n = 3) and 80.0% (n = 20), respectively.

Across all doses in the CRC cohort, the median PFS was 4.0 months (range, 0.7-11.0). In the 960-mg cohort, the median PFS was 4.2 months (range, 1.2 to 5.7+).

The patient with appendiceal cancer who achieved a PR received 720 mg of sotorasib; all other patients received the 960-mg dose.

Among the 3 patients with pancreatic cancer who achieved SD, the best tumor burden reductions from baseline were 28.0%, 28.9%, and 31.6%, respectively.

Despite these demonstrations of efficacy, 5 patients experienced progressive disease (n = 2 pancreatic; n = 1 appendiceal; n = 1 small bowel cancer; and n = 1 esophageal cancer).

Three patients were not included in the waterfall plot because of treatment discontinuation due to clinical progression (n = 1 appendiceal cancer; n = 1 pancreatic cancer) and failure to record response data (n = 1 appendiceal cancer achieving SD).

With regard to safety, any-grade treatment-related AEs (TRAEs) occurred in 45.8% of patients (n = 27). The majority of these TRAEs were grade 2 or higher (n = 12; 20.3%); though, 5.1% of patients (n = 3) experienced grade 3 or higher TRAEs.

Any-grade TRAEs occurring in more than 1 patient consisted of diarrhea (n = 10; 16.9%), fatigue (n = 6; 10.2%), nausea (n = 2; 3.4%), blood alkaline phosphatase increase (n = 2; 3.4%), blood creatinine phosphokinase increase (n = 2; 3.4%), anemia (n = 3; 5.1%), and vomiting (n = 2; 3.4%).

Grade 3 TRAEs comprised diarrhea (n = 2; 3.4%) and anemia (n = 1; 1.7%).

“The key message here is the tolerance of this agent,” said Price.

The phase 2 portion of the trial is ongoing, he concluded.

Reference

JH Strickler, Fakih M, Price TJ, Sotorasib, a novel small molecule inhibitor of KRASG12C, for patients with advanced gastrointestinal tumors: results from the CodeBreaK100 phase 1 trial. Presented at: ESMO Asia Virtual Congress 2020; November 20-22, 2020; virtual. Abstract 83M0.