SAN FRANCISCO – Adding the oral, investigational multikinase inhibitor regorafenib to best supportive care for metastatic colorectal cancer resulted in a modest but statistically significant increase in median overall survival in the phase III CORRECT trial.
The 760-patient study was stopped early, so that patients in a control group could receive the study drug, researchers reported at the Gastrointestinal Cancers Symposium sponsored by the American Society of Clinical Oncology.
Regorafenib added a median benefit of just 1.4 months, compared with placebo and best supportive care (6.4 months vs. 5.0 months; hazard ratio, 0.77; P = .0052), lead investigator Dr. Axel Grothey said. Though the time was short, he noted, all participants were running out of options, after the failure of standard therapies, including bevacizumab (Avastin) and epidermal growth factor receptor (EGFR) inhibitors in those who had KRAS wild-type tumors.
Regorafenib "identifies itself as a potential new standard of care in this patient population," and will move into earlier lines of therapy, said Dr. Grothey, professor of oncology at the Mayo Clinic in Rochester, Minn. Phase II studies are underway to evaluate the novel Bayer drug in combination with standard chemotherapy backbones such as 5-fluorouracil and irinotecan (Camptosar).
Discussant Dr. Herbert Hurwitz of the Duke Cancer Institute in Durham, N.C., said there is good reason to question why regorafenib worked in CORRECT, as the phase I data leading up to the trial were "relatively modest," and there was no phase II trial. "Rushing from phase I to phase III can work, but I would be cautious not to overinterpret the success of the regorafenib study, particularly when looking at the novel therapeutic classes," he advised.
To be eligible for the international CORRECT trial, patients had to have had progression while on or within 3 months after last receipt of approved standard therapies, which had to include a fluoropyrimidine, oxaliplatin (Eloxatin), irinotecan, bevacizumab, and, if they had KRAS wild-type disease, cetuximab (Erbitux) or panitumumab (Vectibix).
The patients had a median age of 61 years. A total of 60% had received four or more lines of prior therapy. Fifty-seven percent had tumors with a KRAS mutation.
The response rate was similar between regorafenib and placebo (1.0% vs. 0.4%), but regorafenib distinguished itself with a much higher disease-control rate than did placebo (45% vs. 15%). "So the strength of this drug is more in delaying tumor progression than inducing responses," Dr. Grothey said.
The median difference in progression-free survival (PFS) was again small, at just 0.2 months, but this corresponded to a 51% reduction in the risk of progression events (HR, 0.49; P less than .000001), he noted.
The progression-free survival curve "clearly identifies that the median difference in PFS does not fully reflect the efficacy of this drug in this patient population. These curves run together for about 50% of patients but then spread out wide. ... Clearly, the median does not reflect what’s happening," he elaborated.
Thus, "there is clearly a benefit for about 50% of patients, compared to the placebo-control."
The side effect profile was similar to that observed in the drug’s phase I trial and included grade 3 hand-foot skin reactions, fatigue, anorexia, and a class effect of hypertension that was controlled with dose reductions. The proportion of patients experiencing adverse events leading to treatment discontinuation was 8.2% with regorafenib and 1.2% with placebo.
"Efficacy in subgroup analyses are being conducted, with preliminary results that KRAS mutant and KRAS wild-type tumors benefit in a similar extent from the treatment," Dr. Grothey reported. "Further biomarker analyses and quality of life analyses are ongoing."
He suggested that regorafenib may have achieved the results it did in part because it was used as a single agent, whereas trials of other small-molecule kinase inhibitors such as gefitinib (Iressa), sorafenib (Nexavar), and PTK/ZK involved adding the agents to first- or second-line chemotherapy.
"This is not unheard of – that adding a kinase inhibitor to chemotherapy does not produce the desired result," he said, noting that the addition of sorafenib to FOLFOX chemotherapy actually appeared detrimental in patients with metastatic colorectal cancer.
Similarly, in his discussion, Dr. Hurwitz emphasized "that the clean design certainly helped." That is, the trial did not have the confounding effects of chemotherapy or any other anticancer therapy for that matter, did not allow cross-over, and studied a mature therapeutic drug class.
"The benefits in this study with regorafenib were indeed statistically significant. I would suggest that they are clinically meaningful for many, but not necessarily all, patients," Dr. Hurwitz commented. "The toxicity profile is acceptable but does require dose adjustment and monitoring of the patient."