NEW YORK – A Phase III trial of Janssen's Yondelis (trabectedin) in BRCA1/2-mutated, platinum-sensitive ovarian cancer failed to improve survival compared to chemotherapy.
Results from a randomized, investigator-initiated trial showed ovarian cancer patients with BRCA1/2-mutations or a BRCAness phenotype had lower overall survival when treated with the alkylating drug Yondelis versus those treated with physician's choice chemotherapy. Domenica Lorusso, a gynecologic oncologist at Fondazione IRCCS National Cancer Institute of Milan, presented the data from the MITO23 trial at the American Society of Clinical Oncology's annual meeting this week.
Yondelis is currently approved in Europe and in 70 countries for the treatment of relapsed, platinum-sensitive ovarian cancer in combination with pegylated liposomal doxorubicin chemo. The Phase III trial that supported the European approval evaluated that combination in platinum-sensitive, recurrent ovarian cancer and reported a median overall survival in BRCA-mutant patients of 34.2 months and a progression-free survival of 10.1 months, showing a benefit over chemo alone.
However, in the US, Yondelis is only approved for previously treated, unresectable or metastatic liposarcoma or leiomyosarcoma.
Lorusso and colleagues previously conducted a Phase II study exploring Yondelis monotherapy in recurrent ovarian cancer patients with a BRCA mutation or BRCAness phenotype, and found the drug had "huge clinical benefit" in this early study, she said. They hoped to replicate this in the latest Phase III trial presented at the meeting.
The trial enrolled 244 patients with recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancers, and, at the time of this analysis, 208 patients were evaluable for response and overall survival. The patients needed to harbor a germline BRCA1/2 mutation or have a BRCAness phenotype, clinically defined as patients who had received and responded to two or more lines or platinum-based chemo.
There is a need for better treatments for recurrent epithelial ovarian cancer, Lorusso said, and deeper knowledge of the drugs already available.
However, in the Phase III MITO23 trial, single-agent Yondelis did not improve outcomes compared to chemotherapy. Median overall survival among Yondelis-treated patients was lower at 15.8 months compared to 17.9 months for those on chemo. Progression-free survival was also not improved with a median of 4.9 months on Yondelis and 4.4 months on chemo.
The clinical benefit rate was higher in the Yondelis arm, 60 percent versus 51 percent in the chemo arm, in the MITO23 trial. But there was also a higher proportion of patients who had adverse events on Yondelis, and more patients discontinued Yondelis treatment due to adverse events compared to chemo.
The researchers also reported on several exploratory and subgroup analyses, looking at whether Yondelis performed better in BRCA1- or BRCA2-mutated patients, those with or without measurable disease at study entry, and comparing the drug to platinum or non-platinum chemo. However, they found no specific group benefited more from Yondelis in any of these exploratory analyses.
Interestingly, the overall response rate and survival outcomes with Yondelis was much lower in the MITO23 trial than in the previous Phase II study Lorusso conducted. In MITO23, 17 percent of patients on Yondelis responded and median overall survival was 15.8 months. In the earlier trial involving a similar patient population, the response rate was 39 percent on Yondelis and median overall survival was 18 months.
Lorusso noted that the researchers are continuing translational studies to try to identify predictive biomarkers of response, including genomic and transcriptomic analysis of homologous recombination-related genes, microRNA expression, and tumor infiltrating lymphocytes.
"In this heavily chemotherapy and PARP inhibitor pretreated recurrent ovarian cancer population, nevertheless, trabectedin retained clinical activity, and the ongoing translational studies hopefully will help in identifying predictive biomarkers of response and the patients who benefit from this strategy," Lorusso said.
One explanation for the negative turnout may be the proportion of patients who received PARP inhibitors before randomization, Lorusso said. About one-third of patients enrolled had received prior PARP inhibitor treatment, which may have bolstered patients' ability to respond to chemo.
"The trial was negative because the control arm overperformed with respect to our expectations and with respect to literature data," she said. "We will address the efficacy of chemo after PARP, and we will explore how trabectedin performed in this subgroup."
Roisin O'Cearbhaill, a medical oncologist at Memorial Sloan Kettering, noted that a more refined definition of recurrent ovarian cancer patients is needed, taking into consideration biomarker-defined groups like those with BRCA1/2 mutations, rather than using the traditional platinum-sensitive or platinum-resistant categories. This might help the field identify better treatment strategies.
"We can consider combining existing treatments with novel targeted therapies. We can consider immune combinations [to benefit this group]," O'Cearbhaill said. "It is essential that we develop reliable biomarkers that are able to select patients who are most likely to derive benefit from our treatments and also to identify those patients who have higher risk of toxicity."