Tivantinib Holds Promise for MET-Expressing Unresectable Hepatocellular Carcinoma

The oral MET inhibitor tivantinib demonstrated striking efficacy as a single agent in a randomized, placebo-controlled phase II trial conducted in 107 patients with unresectable hepatocellular carcinoma (HCC). The benefits were specifically observed in the subgroup of patients with tumors showing high MET expression (Abstract 4006).

“These findings are the first randomized data in HCC showing an overall survival advantage with a MET inhibitor administered as a single agent and are the fi rst identifi cation of a biologic subgroup of patients responding to a targeted therapy,” said Lorenza Rimassa, MD, of the Humanitas Cancer Center, Italy, who presented the study fi ndings during Saturday’s Gastrointestinal (Noncolorectal) Cancer Oral Abstract Session.

Tivantinib is undergoing evaluation in several cancers. However, the findings in HCC are particularly provocative given the lack of second-line therapies after sorafenib for patients with advanced disease.

Dr. Rimassa reported that tivantinib improved the time to progression—the primary study endpoint—by approximately 1 week compared with placebo in the overall population (Figure).

Although significant, this modest improvement was much more striking when the analysis was restricted to patients harboring HCCs with high MET expression—a factor found to be associated with poor prognosis. Among these individuals, the median time to progression was extended from 6.1 weeks with placebo to 11.7 weeks with tivantinib HR: 0.43; p = 0.03. Moreover, this subgroup demonstrated a significant 3.4-month improvement in overall survival with tivantinib compared with placebo (7.2 months versus 3.8 months, respectively; HR: 0.38; p = 0.01).

No improvements in time to progression or overall survival were observed with tivantinib in patients with tumors exhibiting low MET expression, nor was an overall survival benefit observed in the overall patient population.

The original tivantinib dose of 360 mg was reduced to 240 mg following an unexpectedly high incidence of drug-related neutropenia, which led to sepsis in four individuals. Grades 3 to 5 neutropenia dropped from 21% at the 360-mg dose to 6% at the 240-mg dose. Aside from neutropenia, the most common grade 3 to grade 5 drug-related adverse event was anemia (9% to 16%).

A phase III study of tivantinib in patients with high MET-expressing HCC is currently being planned to confi rm these encouraging findings.

Discussant Jordan Berlin, MD, of the Vanderbilt-Ingram Cancer Center, praised the conviction of the tivantinib investigators to follow the MET biomarker hypothesis in a subsequent trial. However, he suggested that it might be premature to plan a large, randomized phase III trial on the findings of a very small subset analysis. Drawing parallels to pancreatic cancer, he warned, “We have been fooled by subset analyses…. They are hypothesis driving, not trial driving.”