OBJECTIVE

We prospectively evaluated patients with completely resected uterine serous carcinoma (USC) treated with radiation "sandwiched" between carboplatin/paclitaxel (C/T). The primary objective was to determine the safety profile, and the secondary outcome was to evaluate progression-free and overall survival.

METHODS

Surgically staged patients with completely resected USC were enrolled to receive 3 cycles of paclitaxel 175 mg/m and carboplatin (area under the curve, 6-7.5) every 21 days, followed by radiotherapy and an additional 3 cycles of T/C at area under the curve of 5-6 (6 cycles + radiotherapy). Toxicity was graded according to National Cancer Institute Common Toxicity Criteria, version 4.03. Kaplan-Meier and log-rank tests were used to compare survival probabilities.

RESULTS

One hundred forty patients were enrolled, of which 132 were evaluable, completed at least 3 cycles of chemotherapy and radiation. One hundred seven (81%) completed 6 cycles of chemotherapy and radiation. Patients with early-stage (I/II) disease have survival probabilities of 0.96 and 0.81 at 2 and 5 years. Patients with stage I USC and lymphovascular invasion have considerably worse overall survival, with 2.7 times' higher risk of death than those without lymphovascular invasion. Patients with late-stage (III/IV) disease had overall survival probabilities of 0.64 and 0.18 at 2 and 5 years, which is far higher survival than what has been reported in single-modality trials. Interestingly, and different than what is reported in other studies, there is no difference in survival in African Americans versus whites/other races who were evaluable. Of the 779 cycles administered, 22% and 14% of cycles were associated with grades 3 and 4 hematologic toxicities, respectively. Grades 3 and 4 nonhematologic toxicities occurred in 6.9% of cycles.

CONCLUSIONS

The long-term follow-up in this study demonstrates that "sandwich" therapy is an efficacious, well-tolerated treatment approach with acceptable toxicities. Lymphovascular invasion (LVSI) is a significantly poor prognostic factor in stage I USC. Multimodal "sandwich" therapy should be considered in all USC patients who have undergone complete surgical resection and staging.