Regorafenib Before Cetuximab Improves Survival in mCRC

Kohei Shitara, MD

Sequencing regorafenib (Stivarga) before cetuximab (Erbitux) showed superior overall survival (OS) compared with the reverse sequence in patients with metastatic colorectal cancer (mCRC) following failure of standard chemotherapy, according to findings from the phase II REVERCE study presented at the 2018 Gastrointestinal Cancers Symposium.

In the randomized study, the regorafenib-cetuximab (R-C) sequence was associated with significantly longer median OS of 17.4 months compared with 11.6 months with cetuximab followed by regorafenib (C-R). After a median follow-up of 29.0 months with 81 death events, there was a 39% reduction in the risk of death with the R-C sequence versus C-R (HR, 0.61; 95% CI, 0.39-0.96; P = .029).

Lead investigator Kohei Shitara, MD, from National Cancer Center Hospital East, Kashiwa, Japan, said that data from the REVERCE trial suggest “longer survival with regorafenib first treatment followed by cetuximab than that of current standard sequence.” Moreover, he said that longer progression-free survival (PFS) following the second treatment period with cetuximab may have contributed to the improvement in OS with R-C.

The current standard of care is to administer cetuximab followed by regorafenib; however, early stage research has suggested activity for regorafenib in earlier treatment lines for patients with mCRC. Moreover, in preclinical research, downregulation of MAP kinase and Akt with regorafenib was shown to sensitize mCRC cells to anti-EGFR therapies, such as cetuximab.

In the REVERCE study, Japanese investigators evaluated R-C versus C-R in patients previously treated with fluoropyrimidines, oxaliplatin, and irinotecan. Patients were naïve to EGFR inhibitors at study entry. Planned accrual was 180 patients but enrollment was discontinued after randomization of 101 patients from 29 centers because of slow accrual. Analysis of circulating tumor DNA revealed that 90% of patients in the R-C arm and 88% in the C-R arm had RAS wild type mCRC. Bevacizumab was administered previously in 96% and 98% in the R-C and C-R arms, respectively.

After failure of chemotherapy, patients were randomized to R-C or C-R; cituximab was given with or without irinotecan. Patients continued each sequence until disease progression or unacceptable toxicity. Intent to use irinotecan was 59% and 64% in the R-C and C-R arms, respectively. Sequential treatment was successfully administered in 86% of patients.

During treatment period 1, dose reduction of regorafenib from the 160 mg starting dose was required in about 70% of patients, said Shitara, and during treatment period 2, approximately 30% required regorafenib dose reduction. “This is based on the difference of treatment duration,” he said. “The mean duration of regorafenib treatment was very different; 4 months in treatment 1 and 2 months in treatment 2…which might be a contributor to the higher proportion of patients with dose reduction in treatment 1 with regorafenib.”

The primary endpoint was OS. Key secondary endpoints included PFS with initial treatment (PFS1), PFS with second treatment (PFS2), safety, and quality of life (QOL).

The difference in OS between arms in the 86 RAS/RAF wild-type patients who underwent ctDNA analysis at study entry was similar to that of the overall population, at 18.2 months in the R-C arm versus 12.7 months in C-R arm (HR, 0.60; 95% CI, 0.37-0.98; P = .036). Emergent RAS mutations were identified in 26% of patients after cetuximab. In a preliminary analysis, “OS seemed to be worse in patients with emergent RAS mutations,” said Shitara, although the median PFS2 did not seem to differ between the arms in this subgroup compared with the overall population. Biomarker analysis is ongoing.

Median PFS1 (regorafenib vs cetuximab) was 2.4 months in the R-C arm and 4.2 months in the C-R arm (HR, 0.97; 95% CI, 0.62-1.54; P = .91). Median PFS2 favored R-C at 5.2 months compared with 1.8 months with C-R, representing a 71% reduction in the risk of progression or death following the second treatment (HR, 0.29; 95% CI, 0.17-0.50; P <.0001). The time to treatment failure was a median of 7.4 months with R-C and 6.1 months with C-R (HR, 0.60; P = .017).

Among patients with measurable lesions, after treatment period 1, the objective response rate (ORR) was 4.0% with regorafenib and 20.4% with cetuximab, and the disease control rate (DCR) was 46.5% and 77.6%, respectively. After treatment period 2, ORR was 27.9% with R-C and 0% with C-R, and the DCR was 76.7% and 31.0%, respectively.

No unexpected safety signals were detected in either arm. QOL score by the EQ5D was lower during regorafenib than cetuximab during both treatment periods. There was no significant difference in average QOL score between groups during the entire treatment period (P = .65).

Shitara K, Yamanaka T, Denda T, et al. REVERCE: Randomized phase II study of regorafenib followed by cetuximab versus the reverse sequence for metastatic colorectal cancer patients previously treated with fluoropyrimidine, oxaliplatin, and irinotecan. J Clin Oncol. 2018;36 (suppl 4S; abstr 557).

<<< View more from the 2018 GI Cancer Symposium

“Subgroup analysis by clinical factors suggests better outcome with regorafenib first treatment in almost all subgroups,” said Shitara. Specifically, in patients with left-sided primary tumors, OS was 20.5 months in the R-C arm versus 11.9 months in the C-R arm (HR, 0.51; 95% CI, 0.30-0.86; P = .011).