Search

Objective: To determine the outcome, pattern(s) of failure, and optimal treatment volume in Stage IIIC endometrial carcinoma patients treated with surgery and postoperative radiation therapy (RT). Methods: Between 1983 and 1998, 30 Stage IIIC endometrial carcinoma patients were treated with primary surgery and postoperative RT at the University of Chicago. All underwent total abdominal hysterectomy, bilateral salpingo-oophorectomy, sampling of pelvic lymph nodes (PLN), and peritoneal cytology. All were noted to have PLN involvement. Para-aortic lymph nodes (PALN) were sampled in 26 cases, and were positive in 14 cases (54%). Twenty women received whole-pelvic RT (WPRT) and 10 (WPRT), plus paraortic RT (extended-field RT, EFRT). One EFRT patient also underwent concomitant whole-abdominal RT (WART). Adjuvant vaginal brachytherapy (VB) was delivered in 10, chemotherapy in 5, and hormonal therapy in 7 patients. Results: At a median follow-up of 32 months, the actuarial 5-year disease-free and cause-specific survivals of the entire group were 33.9% and 55.8%, respectively. Overall, 16 women (53%) relapsed. Sites of failure included the pelvis (23%), abdomen (13%), PALN (13%), and distant (40%). Of the 7 pelvic failures, 4 were vaginal (3 vaginal only). Patients treated with VB had a trend to a lower vaginal recurrence rate (0/10 vs. 4/20, p = 0.12) than those not receiving VB. All 4 PALN failures were in women treated with WPRT (2 negative, 1 unsampled, and 1 positive PALN). None of the 10 EFRT patients (2 negative, 8 positive PALN) recurred in the PALN. No patient developed an isolated abdominal recurrence. Two patients developed significant RT sequelae: chronic diarrhea in 1 patient treated with WPRT and VB, and small bowel obstruction in 1 patient treated with EFRT. Conclusion: FIGO Stage IIIC disease comprises a small percentage of endometrial carcinoma patients but carries a poor prognosis. Our failure pattern suggests that the optimal adjuvant RT volume is EFRT, even in women with negative PALN sampling. VB should also be administered to improve local control. The low rate of abdominal recurrence does not support the routine use of WART in these women. Given the predominance of failure in distant sites, attention should be focused on the development of systemic chemotherapy protocols.

To evaluate the risk of pelvic recurrence (PVR) in high-risk pathologic Stage I--IV endometrial carcinoma patients after adjuvant chemotherapy alone.Between 1992 and 1998, 43 high-risk endometrial cancer patients received adjuvant chemotherapy. All patients underwent primary surgery consisting of total abdominal hysterectomy and bilateral salpingo-oophorectomy. No patients received preoperative radiation therapy (RT). Regional lymph nodes and peritoneal cytology were sampled in 62.8% and 83.7% of cases, respectively. Most patients had Stage III--IV disease (83.7%) or unfavorable histology tumors (74.4%). None had evidence of extra-abdominal disease. All patients received 4-6 cycles of chemotherapy as the sole adjuvant therapy, consisting primarily of cisplatin and doxorubicin. Recurrent disease sites were divided into pelvic (vaginal, nonvaginal) and extrapelvic (para-aortic, upper abdomen, liver, and extra-abdominal). Median follow-up was 27 months (range, 2--96 months).Twenty-nine women (67.4%) relapsed. Seventeen (39.5%) recurred in the pelvis and 23 (55.5%) in extrapelvic sites. The 3-year actuarial PVR rate was 46.5%. The most significant factors correlated with PVR were cervical involvement (CI) (p = 0.01) and adnexal (p = 0.05) involvement. Of the 17 women who developed a PVR, 8 relapsed in the vagina, 3 in the nonvaginal pelvis, and 6 in both. The 3-year vaginal and nonvaginal PVR rates were 37.8% and 26%, respectively. The most significant factor correlated with vaginal PVR was CI (p = 0.0007). Deep myometrial invasion (p = 0.02) and lymph nodal involvement (p = 0.03) were both correlated with nonvaginal PVR. Nine of the 29 relapsed patients (31%) developed PVR as their only (6) or first site (3) of recurrence. Factors associated with a higher rate of PVR (as the first or only site) were CI and Stage I--II disease.PVR is common in high-risk pathologic Stage I-IV endometrial cancer patients after adjuvant chemotherapy alone. These results support the continued use of locoregional RT in patients undergoing adjuvant chemotherapy. Further studies are needed to test the addition of chemotherapy to locoregional RT.