CHICAGO – For the first time ever, a new drug has shown a survival advantage – albeit a small one – in metastatic soft tissue sarcomas, diseases that are almost without exception unresponsive to medical therapies and rapidly fatal.
In a randomized, phase III trial, median overall survival for patients with metastatic leiomyosarcoma or liposarcoma treated with eribulin (Halaven) was 13.5 months compared with 11.5 months for patients treated with dacarbazine.
“For me as a clinical oncologist, this a clinically meaningful result given the unmet need in this rare, hard-to-treat family of diseases,” said lead investigator Dr. Patrick Schöffski, professor of medicine and head of experimental oncology lab at University Hospitals in Leuven, Belgium, at the annual meeting of the American Society of Clinical Oncology.
“I’ve been an oncologist for over 20 years, and there has never been a positive sarcoma study for survival – ever,” said ASCO expert Dr. Gary K. Schwartz, chief of hematology and oncology at Columbia University, New York, who was not involved in the study. “This is the first positive sarcoma study showing a survival benefit with any chemotherapy in the history of medical oncology.”
The findings represent “a small step for cancer, and a giant step for sarcoma.” Dr. Schwartz said at a press briefing.
In an interview, he added that he was “shocked” when he found out that the survival benefit with eribulin was seen in patients with liposarcoma as well as leiomyosarcoma. “I thought that the benefit would only be in patients with leiomyosarcoma,” he said.
Leiomyosarcoma and liposarcoma, also called adipocytic sarcoma, are two of the most common histologic subtypes of highly uncommon soft tissue sarcomas. Patients with advanced metastatic sarcomas generally have a very poor prognosis and precious few treatment options.
Eribulin is a synthetic analogue of halichondrin B, a macrolide antibiotic isolated from a sea sponge that has strong cytotoxic properties. In preclinical studies, it was shown to kill cancer cells via anti-mitotic effects. It is approved in the United States for treatment of patients with metastatic breast cancer for whom at least two other lines of therapy have failed.
Dr. Schöffski and colleagues in centers in the United States, Europe, and Korea randomly assigned patients with metastatic intermediate or high-grade leiomyosarcoma or liposarcoma to receive either eribulin 1.4 mg/m2 on days 1 and 8 of 21-day cycles (228 patients), or conventional therapy with dacarbazine 850, 1,000, or 1,200 mg/m2 on day 1 of every 21 day cycle (224 patients).
The majority of patients had undergone at least two prior regimens for advanced disease.
The trial met its primary endpoint of an overall survival advantage, with median OS of 13.5 months for patients on eribulin, compared with 11.5 months for those on dacarbazine. The hazard ratio (HR) was 0.768 favoring eirbulin (P = .0169).
The median progression-free survival was 2.6 months in each group.
In all, 26% of patients on eribulin and 14% on dacarbazine required dose reductions, and 8% and 5%, respectively, discontinued therapy because of treatment-emergent adverse events.
The most common grade 3 or greater toxicities in each arm were neutropenia, fatigue, and nausea. The adverse events in general were in line with known safety profiles for each agent, Dr. Schöffski said.
There were 10 deaths among patients on eribulin, two of which were attributed to the drug (1 neutropenic sepsis and one sepsis). There were three deaths among patients on dacarbazine, none attributed to therapy.
In an interview, Dr. Schöffski noted that he has treated one patient, a man with dedifferentiated liposarcoma, with 46 cycles of eribulin, “so we see activity in a tumor type that is notoriously resistant to established drugs.”
He said that moving forward, investigators may wish to test eribulin in combination with other drugs, such as therapies targeted to genetic mutations, immunotherapies, or other approaches.
The study was sponsored by Eisai, Inc. Dr. Schöffski and Dr. Schwartz reported having no relevant conflicts of interest.