Intraperitoneal (IP) therapy was not superior to intravenous (IV) treatment when used in combination with bevacizumab (Avastin) as a regimen for patients with advanced ovarian cancer, according to phase III findings from the GOG-252 study.
The Gynecologic Oncology Group study, which compared three different chemotherapy regimens, failed to build on results of a landmark trial reported more than a decade ago.2Results were reported during the 2016 Society of Gynecologic Oncology Annual Meeting held March 19-22 in San Diego.
In the GOG-252 trial, the median progression- free survival (PFS) was approximately 27 months to 29 months in patients with optimal stage II-III disease treated with regimens consisting of different combinations of IV and IP cisplatin, carboplatin, and paclitaxel, in combination with bevacizumab. An analysis of patients with optimal stage III tumors and no gross residual disease produced median PFS values of 31 months to 34 months. At this time, OS data are immature for the GOG-252 trial.
By comparison, the decade-old GOG 172 trial comparing IP and IV chemotherapy regimens in ovarian cancer showed a median PFS of 23.8 months with IP cisplatin, including 60.4 months in patients who had no visible residual disease after surgery.2
In the IP arm of the GOG-252 trial, patients received a 25-mg/ m2 lower dose of cisplatin and a 5-mg/m2 lower dose of IV paclitaxel compared with GOG-172. Furthermore, the infusion time was also shortened from 24 hours to 3 hours.
Additionally, patients did not receive bevacizumab in the GOG-172 study. “Reduced doses of paclitaxel and cisplatin, as well as crossover, may have compromised the efficacy,” explained lead investigator Joan L. Walker, MD, a gynecologic oncologist at the University of Oklahoma Health Sciences Center in Oklahoma City. “Dose-dense paclitaxel may have improved efficacy well enough to allow us to abandon IP chemotherapy, so should we wait or should we combine both? That’s the question. However, neuropathy might limit that. Bevacizumab interactions also may have clouded our analysis.”
The GOG-252 findings have ignited controversy in clinical and advocacy circles. The Ovarian Cancer Research Fund said in a statement that its team of experts agree that “the results of this trial do not mean that patients and doctors should abandon IP chemotherapy at this time.”
“The recent results, which were in such stark contrast to the previous result, have left both patients and doctors alike scratching their heads,” the group said. “Our scientific advisers don’t believe that the results of this one trial should outweigh the significant prior evidence that suggests that IP chemotherapy is beneficial.
Key Details of GOG-252 Study
GOG-252 involved women with stage II-III epithelial ovarian, fallopian tube, or primary peritoneal carcinoma. All patients underwent optimal surgical debulking to 1 cm or less residual disease by surgeon report. Patients were randomized to one of three chemotherapy regimens (Figure). The primary endpoint was PFS. Data were analyzed from 1560 patients with a median age of 58. Patients with stage III disease accounted for 84% of the study population, and 72% of patients had grade 3 serous ovarian cancer. Walker noted that 57% of the patients had no visible residual disease by surgeon report. An analysis of CT assessment of completeness of surgery is pending.
In both carboplatin arms, 90% of patients completed at least 6 cycles of platinum therapy compared with 84% of patients randomized to the cisplatin-containing regimen. In all 3 arms, 87% to 88% of patients completed at least 6 cycles of the taxane.
The primary analysis showed a median PFS of 26.8 months for patients with optimal stage II-III disease in the IV carboplatin reference arm, 28.7 months among patients who received IP carboplatin, and 27.8 months for patients who received IP cisplatin. Follow-up for overall survival will continue, Walker said.
Two thirds or more patients in each arm developed grade 3 neutropenia. Grade 3 thrombocytopenia occurred in 6.3% and 8.4% the IV and IP carboplatin groups, respectively, as compared with 9% in the cisplatin arm.
Grade 3 hypertension occurred in 12% to 14% of patients in the carboplatin arms and 20.5% of patients randomized to the cisplatin-containing regimen. Grade 3 nausea and vomiting occurred in 11.2% of the cisplatin group and in about 5% in the two carboplatin arms. Grade 2 sensory neuropathy occurred in more than 20% of patients in all three arms of the trial.
“All arms had substantial toxicity,” Walker said. “Neurotoxicity was equally high in all arms. We reserve judgment on recommendations until survival is available. IP cisplatin probably shouldn’t be combined with bevacizumab because it causes severe hypertension.”
An analysis of patient-reported outcomes (PROs)quality of life, neuropathy, nausea, fatigue, and abdominal discomfort—showed consistently lower scores among patients in the cisplatin arm.
PRO scores did not differ appreciably between the two carboplatin arms, with the exception of slower improvement in abdominal discomfort in patients treated with IP carboplatin.