Phase III Trial: Adjuvant Pelvic Radiation Therapy Versus Vaginal Brachytherapy Plus Paclitaxel/Carboplatin in High-Intermediate and High-Risk Early Stage Endometrial Cancer

Abstract

Purpose: The primary objective was to determine if vaginal cuff brachytherapy and chemotherapy (VCB/C) increases recurrence-free survival (RFS) compared with pelvic radiation therapy (RT) in high-intermediate and high-risk early-stage endometrial carcinoma.

Patients and methods: A randomized phase III trial was performed in eligible patients with endometrial cancer. Eligible patients had International Federation of Gynecology and Obstetrics (2009) stage I endometrioid histology with Gynecologic Oncology Group protocol 33-based high-intermediate-risk criteria, stage II disease, or stage I to II serous or clear cell tumors. Treatment was randomly assigned between RT (45 to 50.4 Gy over 5 weeks) or VCB followed by intravenous paclitaxel 175 mg/m2 (3 hours) plus carboplatin (area under the curve, 6) every 21 days for three cycles.

Results: The median age of the 601 patients was 63 years, and 74% had stage I disease. Histologies included endometrioid (71%), serous (15%), and clear cell (5%). With a median follow-up of 53 months, the 60-month RFS was 0.76 (95% CI, 0.70 to 0.81) for RT and 0.76 (95% CI, 0.70 to 0.81) for VCB/C (hazard ratio, 0.92; 90% confidence limit, 0.69 to 1.23). The 60-month overall survival was 0.87 (95% CI, 0.83 to 0.91) for RT and 0.85 (95% CI, 0.81 to 0.90) for VCB/C (hazard ratio, 1.04; 90% confidence limit, 0.71 to 1.52). Vaginal and distant recurrence rates were similar between arms. Pelvic or para-aortic nodal recurrences were more common with VCB/C (9% v 4%). There was no heterogeneity of treatment effect with respect to RFS or overall survival among clinical or pathologic variables evaluated.

Conclusion: Superiority of VCB/C compared with pelvic RT was not demonstrated. Acute toxicity was greater with VCB/C; late toxicity was similar. Pelvic RT alone remains an effective, well-tolerated, and appropriate adjuvant treatment in high-risk early-stage endometrial carcinomas of all histologies.

Trial registration: ClinicalTrials.gov NCT00807768.

Figures

FIG 1.

CONSORT diagram. RT, radiation therapy.

FIG 2.

Intention-to-treat analysis of recurrence-free survival (RFS) by randomized treatment. There were 130 events reported as of December 11, 2016, with a median follow-up time of 53 months. There was insufficient evidence to reject the null hypothesis of no superiority of vaginal cuff brachytherapy plus three cycles of carboplatin and paclitaxel chemotherapy (VCB/C) over pelvic radiation therapy (RT). The log-rank test statistic for a true intention-to-treat analysis was 2.75 (one-tailed test P = .31). The estimated treatment hazard ratio (HR) was 0.92 (regimen II relative to regimen I). The (1-α) × 100% Wald CI was 0.651 to 1.296 for a two-sided α = 0.05 (0.025 in each tail) and 0.688 to 1.226 for a two-sided α = 0.10 (0.05 in each tail). Analysis was repeated to assess the sensitivity of results to different patient groups. When all ineligible patients were removed or when all ineligible or untreated patients were removed, the results were similar.

FIG 3.

Intention-to-treat analysis of overall survival (OS) by randomly assigned treatment. As of December 11, 2016, 76 deaths were reported. The median follow-up time was estimated to be 53 months. There was insufficient evidence to reject the null hypothesis of no superiority of vaginal cuff brachytherapy plus three cycles of carboplatin and paclitaxel chemotherapy (VCB/C) over radiation therapy (RT) with respect to OS. The log-rank test statistic was −0.756 (one-tailed test P = .57). The estimated treatment hazard ratio (HR) was 1.04 (regimen II relative to regimen I). The (1-α) × 100% Wald CI was 0.664 to 1.632 for a two-sided α = 0.05 (0.025 in each tail) and 0.713 to 1.518 for a two-sided α = 0.10 (0.05 in each tail). An effect size of 0.51 (49% decrease in hazard) was not contained in these CIs.

FIG 4.

Forest plot of recurrence-free survival (RFS) by treatment of selected subgroups. RFS treatment hazard ratio (HR) estimates on the basis of a Cox proportional hazards model for selected subgroups are displayed in a forest plot and plotted on the log scale with a 95% CI. The HR estimate is represented graphically by a vertical dash; the CI is represented by a horizontal line. The relative HR estimates and variance of the log HR are listed. The HRs are relative to the radiation therapy (RT) arm and vary around 1.0. The vertical line at 1.0 represents no difference in the hazard rates; to the right favors WPRT and to the left favors the arm with vaginal cuff brachytherapy plus chemotherapy (VCB/C). None of the CIs exclude 1.0. LCL, lower confidence limit; UCL, upper confidence limit.

FIG 5.

Forest plot of overall survival (OS) by treatment of selected subgroups. OS treatment hazard ratio (HR) estimates on the basis of a Cox proportional hazards model for selected subgroups are displayed in a forest plot and plotted on the log scale with a 95% CI. The HR estimate is represented graphically by a vertical dash; the CI is represented by a horizontal line. The relative HR estimates and variance of the log HR are listed. The HRs are relative to the radiation therapy (RT) arm and vary around 1.0. The vertical line at 1.0 represents no difference in the hazard rates; to the right favors WPRT and to the left favors the arm with vaginal cuff brachytherapy plus chemotherapy (VCB/C). None of the CIs exclude 1.0. LCL, lower confidence limit; UCL, upper confidence limit

FIG 6.

Cumulative incidence of pelvic or para-aortic recurrence (competing event is death before recurrence of interest). Three competing risk analyses were carried out for three different types of recurrences: any vaginal, any pelvic or any para-aortic nodes, and any distant. More than one recurrence type could be reported for each patient. Death before a specific type of recurrence was considered a competing event. There was a significant differential in the cumulative incidence of pelvic or para-aortic node recurrences between the treatment arms (hazard ratio [HR] of radiation therapy [RT] relative to vaginal cuff brachytherapy plus chemotherapy [VCB/C], 0.472; 95% CI, 0.24 to 0.94). The 52-month cumulative incidence proportion of these types of recurrences was 0.044 for the RT arm and 0.092 for the VCB/C arm. There were no differences in the incidence of vaginal or distant recurrences between the two treatment arms. Approximately 2.5% and 18% of patients will have a vaginal or distant recurrence, respectively, within 5 years of treatment.