Endometrial carcinoma--relative effectiveness of adjuvant irradiation vs therapy reserved for relapse.

Fifty-four patients with recurrent endometrial carcinoma were identified from a retrospective review of charts of 304 endometrial cancer patients seen between 1983 and 1989 at our center. A review was undertaken to identify the patterns of relapse, to determine the outcome of salvage treatment, to examine the factors predictive of effective salvage, and, if salvage is effective, to assess an alternative strategy to routine adjuvant postoperative pelvic radiotherapy. Forty percent of the entire recurrent population are long-term survivors. Of the 54 relapsing patients, primary therapy had been surgery alone in 32 and surgery and adjuvant radiotherapy (rt) in 22. Isolated pelvic recurrence was the predominate relapse site in those who had not received adjuvant pelvic RT as primary therapy (23 of 32 or 72%). Distant relapse predominated in those who received adjuvant RT (17 or 22 or 77%). Twenty-eight (54%) failed in the pelvis alone, and 26 (46%) had a component of distant failure. Of the 28 with isolated pelvic relapse, 16 had vaginal mucosal disease involvement only and 12 had disease in the parametrium and/or the pelvic sidewall. With a minimum follow-up for the survivors of 5 years, 21 of the 28 with isolated pelvic relapse received radical radiotherapy and 14 or 67% had maintained pelvic control until death or last follow-up. Eleven of 14 (79%) with disease confined to the mucosa had pelvic control, whereas only 3 of 7 (43%) with extramucosal disease were controlled. No patient experienced major treatment-related toxicity. Tumor size, anatomic extent of pelvic recurrence, RT dose, and disease-free interval were examined for prognostic significance for pelvic control and survival by univariate analysis. Only anatomic extent of pelvic recurrence showed a nonstatistically significant trend as a predictor for control with P = 0.08. In conclusion, a significant proportion of patients with disease recurrence confined to the pelvis can be rendered disease-free long-term with maintained pelvic control. A reexamination of the role of routine adjuvant pelvic RT is therefore undertaken in light of these data.