Ursula A. Matulonis, MD
Weekly paclitaxel for recurrent ovarian cancer prevailed over the VEGF inhibitor cabozantinib in a randomized comparison of the two drugs.
Both agents led to median progression-free survival (PFS) of about 5 months. However, secondary analyses of overall survival, response rate, and event-free survival suggested that cabozantinib was inferior to paclitaxel, Ursula A. Matulonis, MD, reported at the 2016 Society of Gynecologic Oncology meeting.
“The analysis for the primary objective revealed insufficient evidence to reject the null hypothesis that cabozantinib was equal to or worse than weekly paclitaxel,” said Matulonis, medical director of the Medical Gynecologic Oncology Program at Dana-Farber Cancer Institute. “Event-free survival showed that cabozantinib did worse than weekly paclitaxel.”
“The dose and schedule of cabozantinib examined in this study was not worthy of further investigation,” she added.
Cabozantinib targets both VEGF receptor and the hepatocyte growth factor receptor MET. A preliminary clinical trial of cabozantinib resulted in an objective response rate of 29% in platinum-resistant/refractory ovarian cancer and 40% in patients with platinum-sensitive disease (EORTC/NCI/AACR Symposium on Molecular Targets and Cancer Therapeutics, 2010, abstract 407).
In a subsequent phase II trial, treatment with cabozantinib resulted in response rates of 18%, 22%, and 28% in patients with platinum-refractory, resistant, and sensitive ovarian cancer, respectively (J Clin Oncol
. 2011;29[suppl]:abstract 5008).
The preceding results provided the basis for the Gynecologic Oncology Group (GOG) 186K trial, reported by Matulonis. Investigators in the multicenter phase II trial enrolled patients with persistent or recurrent ovarian cancer treated with as many as 3 prior systemic regimens, no more than one of which could have been non-platinum/taxane.
Eligible patients received cabozantinib at 60 mg daily or paclitaxel at 80 mg/m2/week for every 3 weeks out of 4. CT/MRI assessment for treatment response occurred every 8 weeks.
The primary objective was PFS, as assessed after 3.68 months (pre-cycle 5/week 16) and 7.36 months (pre-cycle 9/week 32). Secondary objectives included overall response, frequency and severity of toxicities, and translational research objectives (still ongoing).
The trial accrued a total of 111 patients, followed for a median of 14.5 months in the paclitaxel arm and 13.9 months in the cabozantinib arm. The 2 treatment arms did not have any substantive imbalances in baseline characteristics. About a third of patients were 50 to 59 years of age and about 40% were aged 60 to 69. About 70% had performance status 0, and more than 80% had serous histology. More than 80% of patients had received 1 or 2 prior regimens, and fewer than 15% had prior exposure to bevacizumab.
The primary analysis showed no significant difference in median PFS: 5.3 months with cabozantinib and 5.5 months with paclitaxel. A preliminary survival analysis resulted in a median of 19.4 months for the cabozantinib group, whereas the median had yet to be reached in the paclitaxel arm.
An unadjusted analysis suggested no difference in overall survival (HR, 1.66; 90% CI, 0.93-2.97; P
= .15). However, a stratified survival analysis yielded a hazard ratio of 2.27 for cabozantinib versus paclitaxel, which did achieve statistical significance (90% CI, 1.17-4.41; P
Overall response rates by RECIST criteria were 8.3% in the cabozantinib arm and 28.3% in the paclitaxel arm. The difference translated into an odds ratio of 0.23 for the likelihood of response with cabozantinib versus paclitaxel.
“The dose and schedule of cabozantinib examined in this study was not worthy of further investigation,” said Matulonis.
Matulonis UA, Sill M, Thaker PH, et al. Nrg/GOG 186K: A randomized phase II study of NCI-supplied cabozantinib versus weekly paclitaxel in the treatment of persistent or recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer—final results. Presented at: 2016 SGO Annual Meeting. March 19-22, 2016. San Diego, CA. Late-breaking abstract.
<<< View more from the 2016 SGO Annual Meeting