New Trial Suggests Maintenance Chemotherapy Does Not Aid Overall Survival

Maintenance chemotherapy after debulking surgery and adjuvant platinum-based chemotherapy did not improve overall survival (OS) rates in women with ovarian cancer who had achieved complete remission, according to results from a randomized trial presented at the 2017 Society of Gynecologic Oncology Annual Meeting.

After a median follow-up of almost six years, patients treated with conventional paclitaxel and a paclitaxel conjugate had a median OS of 51 and 60 months, respectively, compared with 55 months for patients assigned to surveillance with standard follow-up care.

The trial did show a modest improvement in progression-free survival (PFS) with either of the taxane-based maintenance therapies, Larry Copeland, M.D., a medical oncologist at The Ohio State University.

“The trial was designed to give us a definitive answer, which is why we chose OS as the primary endpoint,” Copeland said. “As often occurs in the trials we do, the PFS was positive, but the OS did not show a benefit. In addition, the additional therapy was not without some toxicity that we consider significant.”

The trial design addressed a common problem in the management of ovarian cancer. Upfront surgery and chemotherapy leads to a clinical complete response (CCR) for many patients, but a substantial proportion of patients who achieve a CCR eventually have recurrent disease, which often proves fatal.

Investigators in the Gynecologic Oncology Group (now part of the National Research Group) 212 trial conducted a three-arm randomized trial to determine whether maintenance therapy for patients in CCR would improve survival. The trial evaluated a novel formulation of paclitaxel linked to a polyglutamate polymer, which achieves higher tissue concentrations of the cytotoxic agent.

Patients who attained CCR with surgery and a first-line platinum-taxane combination chemotherapy were randomized to surveillance, conventional paclitaxel (135 mg/m²), or the paclitaxel conjugate CT-2103. The trial had 90 percent statistical power to detect a 25 percent decrease in the hazard for death.

The data and safety monitoring committee recommended termination of the trial after a third planned interim analysis showed the results had crossed predefined statistical criteria for futility. The trial ended after a 12-year run, including a median follow-up of 71 months.

The results provided a suggestion of a trend toward worse survival in the patients randomized to conventional paclitaxel, which was associated with a hazard ratio of 1.104 versus the surveillance group, although the difference did not reach statistical significance (97.5 percent CI, 0.884-1.38). The five-month difference in the CT-2103 arm also did not reach statistical significance versus the surveillance arm (HR, 0.979; 97.5 percent CI, 0.781-1.23).

Both paclitaxel arms had significantly better PFS versus surveillance, which was associated with a median PFS of 13.4 months. Patients randomized to CT-2103 had a median PFS of 16.3 months (HR, 0.847; 95 percent CI, 0.721-0.995) and those receiving conventional paclitaxel had a median PFS of 18.9 months (HR 0.783, 95 percent CI 0.666-0.921).

Adverse events, all grades and grade 3/4, occurred substantially more often in the paclitaxel arms. Grade 3/4 events occurred in 50.7 percent of patients treated with CT-2103, 39.3 percent of the paclitaxel group, and in 12.5 percent of the surveillance group. Copeland singled out grade 3/4 neurotoxicity, which occurred in 12.4 percent of the CT-2103 arm and 7.2 percent of the paclitaxel arm versus 1.9 percent of the surveillance group. Grade 2 alopecia occurred in 24.5 percent of the CT-2103 group, 44.9 percent of the paclitaxel group, and 13.9 percent of the surveillance group.

Patient-reported quality-of-life data suggested no difference, at best, and possibly a small decrement in the paclitaxel groups at various time points.

Investigators performed a couple of exploratory analyses in an attempt to gain more insight in the findings. The trial design did not evaluate the impact of complete surgical resection (R0) on outcome. Overall, patients whose surgery resulted in the absence of residual disease had a median OS of 70.0 versus 43.6 months for patients with gross residual disease.

The addition of maintenance therapy did not extend the benefit of R0 surgery. Patients with optimal debulking had a median OS of 72.6 months with CT-2103, 64.3 months with paclitaxel, and 61.6 months with surveillance. The difference between the CT-2103 and surveillance group did not achieve statistical significance (HR, 0.856; 95 percent CI, 0.729-1.006).

Investigators also investigated whether maintenance therapy might induce resistance in patients with residual disease after surgery. The analysis showed a median OS of 48.7 months with surveillance versus 39.9 months in each of the paclitaxel groups. Though suggestive, the difference did not achieve statistical significance.