CHICAGO – Adding the anti-angiogenic agent bevacizumab to chemotherapy improves outcomes in women with platinum-sensitive recurrent epithelial ovarian cancer, the randomized Gynecologic Oncology Group 213 trial has shown.
Results reported at the annual meeting of the Society of Gynecologic Oncology showed that addition of bevacizumab reduced the risk of death (one of the trial’s co-primary endpoints) by a near-significant 17%, which amounted to 4.9 more months of life.
Dr. Robert L. Coleman
“Paclitaxel, carboplatin, and bevacizumab extended overall survival in patients with platinum-sensitive recurrent ovarian cancer, but narrowly missed that statistical upper limit of significance. The combination was associated with a significant improvement in progression-free survival as well as objective response,” commented lead investigator Dr. Robert L. Coleman, professor and vice chair of clinical research in the department of gynecologic oncology and reproductive medicine, University of Texas MD Anderson Cancer Center in Houston.
“Although toxicity was more commonly observed on the concomitant bevacizumab arm, no new safety signals were detected, and no negative impact on a patient-reported global functioning scale was observed,” he added.
Session attendee Dr. Brad Monk of the University of Arizona Cancer Center in Phoenix wondered if the overall survival results could be interpreted as positive in light of information gleaned since the trial’s inception.
“I think we all accept that 5 months is 5 months of overall survival. I accepted 8 years ago, when the study was developed, that [the statistical test used] was a two-sided test, looking for both superiority and inferiority,” he said. “But today we would never expect that bevacizumab would make survival worse. So if you were to apply a one-tailed test, the P value would go from .056 to .028, and then you would say it’s positive. I know that we can’t change the statistical design, but is this study really negative statistically? Is this a practice-changing study?”
“You are right, we have learned a lot,” Dr. Coleman agreed. “I think one of the most striking differences in this trial versus what our expectations were was the performance of the control arm,” wherein survival far exceeded the 22 months assumed in the trial’s planning; meanwhile, a progression-free survival benefit of bevacizumab was preserved. “I think that this is the strongest signal for that, and the fact that we were able to achieve that knowing that the environment has changed so much, to me, does feel like it is practice changing,” he said.
Two previous trials — CALYPSO (Caelyx Plus Carboplatin Versus Paclitaxel Plus Carboplatin in Patients With Epithelial Ovarian Cancer in Late Relapse) and OCEANS (Ovarian Cancer Study Comparing Efficacy and Safety of Chemotherapy and Anti-Angiogenic Therapy in Platinum-Sensitive Recurrent Disease) — tested variations on a chemotherapy backbone exclusively among platinum-sensitive patients, he noted, giving some background to the research. Neither found a significant survival benefit (Br J Cancer. 2012;107:588-91 and J Clin Oncol. 2012;30:2039-45).
Participants in the current trial were 674 women with ovarian cancer from the United States, Japan, and South Korea who had a complete response to primary therapy but experienced a recurrence at least 6 months later. They were randomized twice: first to receive secondary cytoreductive surgery or not, and then to receive carboplatin plus paclitaxel, with versus without concurrent and maintenance bevacizumab (Avastin). Genentech provided the agent and supported the National Cancer Institute/Cancer Therapy Evaluation Program for the trial.
In terms of adverse events of special interest, the bevacizumab group had significantly higher rates of grade 3 or worse thromboembolism (4% vs. 1%), infection (13% vs. 6%), hypertension (12% vs. less than 1%), proteinuria (8% vs. 0%), and joint pain (15% vs. 5%), reported Dr. Coleman, who disclosed that he had no conflicts of interest.
“However, arterial thrombosis, neutropenia, and febrile neutropenia were not different between the two arms. And in addition, while gastrointestinal events were more common in the experimental arm, the grade 3/4 events were no different, and no patient died as a result of this adverse event, “ he noted. Furthermore, the allergy (mainly carboplatin hypersensitivity) and grade 3 or worse neuropathy were essentially equally common across groups.
Median overall survival was 42.2 months with bevacizumab and 37.3 months without it (hazard ratio 0.83; P = .056). In subgroup analyses, benefit was similar of whether patients had undergone randomization for cytoreductive surgery, had a progression-free interval of more than 12 months, and had received bevacizumab before.
Progression-free survival was 13.8 months with bevacizumab and 10.4 months without it, a significant difference (hazard ratio 0.61; P less than .0001). The overall response rate was also better with the anti-angiogenic agent (79% vs. 59%, P less than .0001).