Atezolizumab and Cobimetinib Regimen Shows Activity in CRC

Johanna C. Bendell, MD

In patients with advanced colorectal cancer (CRC), the combination regimen of cobimetinib (Cotellic) and atezolizumab (Tecentriq) was found to demonstrate clinical activty as well as have a good safety profile, according to phase Ib findings presented at the 2016 ASCO Annual Meeting.¹

Additionally, the combination was found to elicit a higher clinical response rate in patients with microsatellite-stable (MSS) disease compared with either cobimetinib or atezolizumab alone. The investigator-assessed observed response rate (ORR) was 17% and the 6-month overall survival (OS) was 72% among a cohort of 23 patients in the dose-escalation and cohort expansion trial

“These results suggest that cobimetinib can sensitize tumors to atezolizumab by increasing MHC I expression on tumor cells and promoting intratumoral CD8 T-cell accumulation,” said lead author Johanna C. Bendell, MD, director of the GI cancer research program and associate director of the Drug Development Program at the Sarah Cannon Research Institute, and an associate with Tennessee Oncology in Nashville. “This is consistent with the hypothesis that tumors with high mutation burden are more responsive to combined immunotherapy.” She called the pairing of an anti—PD-L1 agent and a MEK inhibitor, such as that of the experimental combination regimen, a "rational combination."

The investigators escalated cobimetinib among patients with MSS CRC, increasing the dose from 20 mg daily to 40 mg daily and then to 60 mg daily in a cycle of 21 days on/7 days off. They combined this with atezolizumab (800 mg IV every 2 weeks).

Upon determining the maximum tolerated dose (MTD), the investigators opened tumor-specific expansion cohorts, including non—small cell lung cancer (NSCLC), metastatic melanoma, solid tumors, and CRC, which focused on the KRAS mutation. They evaluated safety, tolerability, and ORR using RECIST v1.1.

In the CRC group (n = 23), the median age was 57, females comprised 52% of the cohort, and nearly two-thirds of patients were white and had an ECOG performance status of 0 (61% each). Twenty of the 23 patients in the cohort had KRAS-mutated tumors.

The median number of prior systemic therapies was 3. All patients had received prior oxaliplatin and irinotecan, and 52% had received prior adjuvant therapy. Seventy percent were of unknown microsatellite instability (MSI) status.

As of data cutoff, 4 patients (17%) were taken off cobimetinib. No patients were withdrawn from atezolizumab. The median follow up for safety was 3.8 months (range 1.1-15.1 months).

In this cohort, the most common adverse events (AEs) were diarrhea (69.6%), fatigue (52.2%), dermatitis acneiform (43.5%), and rash (34.8%). Maculopapular rash, nausea, and pruritis were also observed (26.1% each).

The total incidence of serious treatment-related AEs (grade 3) was 34.8%. Diarrhea was the only treatment-related serious AE that occurred in 2 or more patients (8.7%). No patients experienced grade 5 or treatment-related grade 4 AEs.

Confirmed responses per RECIST v1.1 revealed 20% ORR in the KRAS-mutant subgroup (95% CI, 5.7-43.7). Of these patients, 20% showed a partial response (PR), 20% had stable disease (SD), 50% had progressive disease (PD), and 10% were not able to be evaluated. In the overall CRC cohort, the ORR was 17% (95% CI, 5.0-38.8). Seventeen percent showed a PR, 22% had SD, 52% had PD and 9% were not evaluable. Bendell noted that these responses did not correlate with PD-L1 status.

Four patients experienced PRs in tumor size reduction from baseline. Of these, 3 patients were mismatch-repair proficient (not microsatellite instability-high); 1 patient had unknown MSI status and was not evaluable. Tumor volume reduction was not associated with PD-L1 status [TC3 (n = 1, PD), TC0 (n = 18), NA (n = 4)]. Median duration of response was not reached (range 5.4-11.1 months). According to Bendell, responses are ongoing in half of the responding patients.

For duration of treatment and response, the investigators reported that the median time to first response was 3.7 months (range 1.8-4.1 months). The median duration of response was not reached (5.4-11.1+ months). Stable disease can be durable, Bendell said, at six months or more.

In the KRAS cohort, median progression-free survival (PFS) was 2.3 months (95% CI, 1.8-9.5) but the median OS was not evaluable. (95% CI, 6.5-NE). The 6-month PFS was 39% (95% CI, 0.16-0.61) and the 6-month OS was 77% (95% CI, 0.57-0.97).

When considering all of the 23 patients with CRC, the median PFS and median OS were unchanged. The 6-month PFS declined slightly, to 35% (95% CI, 0.14-0.56). Accordingly, the 6-month OS declined, too, to 72% (95% CI, 0.52-0.93).

In response to these findings, discussant Neil H. Segal, MD, PhD, of Memorial Sloan Kettering Cancer Center, said, “This study demonstrates encouraging preliminary activity and supports continued investigation of cobimetinib and atezolizumab in MSS colorectal cancer,” he said. “But this trial does not yet support a change in standard practice.”

Based on these preliminary results, trial expansion is ongoing in patients with MSS metastatic CRC (NCT01988896).

1. Bendell JC, Kim TW, Goh BC, et al. Clinical activity and safety of cobimetinib (cobi) and atezolizumab in colorectal cancer (CRC). J Clin Oncol 2016;34 (suppl; abstr 3502).