Operable Stages IB and II cervical carcinomas: a retrospective study comparing preoperative uterovaginal brachytherapy and postoperative radiotherapy.

PURPOSE: To evaluate our data concerning prognostic factors and treatment toxicity in a series of operable cervical carcinomas. METHODS AND MATERIALS: Between May 1972 and January 1994, 414 patients with cervical carcinoma, staged according to the 1995 FIGO staging system (286 Stage IB1, 38 Stage IB2, 56 Stage IIA, and 34 Stage IIB with 1/3 proximal parametrial involvement), underwent radical hysterectomy with (n = 380) or without (n = 34) bilateral pelvic lymph node dissection (N+: n = 68). Group I included 168 patients who received postoperative radiation therapy (RT): 64 patients had low-dose-rate vaginal brachytherapy with a median total dose (MTD) of 50 Gy; 93 patients had external beam pelvic RT (EBPRT) with an MTD of 45 Gy over 5 weeks, followed by low-dose-rate vaginal brachytherapy (MTD: 20 Gy); and 11 patients had EBPRT alone (MTD: 50 Gy over 6 weeks). Group II included 246 patients treated with preoperative low-dose-rate uterovaginal brachytherapy (MTD: 65 Gy); 32 of these 246 patients also received postoperative EBPRT (MTD: 45 Gy over 5 weeks) delivered to the parametria and pelvic nodes. Mean follow-up from the beginning of treatment was 106 months.
RESULTS: First events included isolated locoregional recurrences (35 patients), isolated distant metastases (27 patients), and locoregional recurrences with synchronous metastases (13 patients). The 10-year disease-free survival (DFS) rate was 88% for Stage IB1, 44% for Stage IB2, 65% for Stage IIA, and 48% for Stage IIB. Multivariate analysis showed that independent factors influencing the probability of DFS were as follows: cervical site (exocervical or endocervical vs. both endo- and exocervical, relative risk [RR]: 1.77, p = 0.047), vascular space invasion (no vs. yes, RR: 1.95, p = 0.041), age (>51 years vs. <or=51 years, RR: 1.90, p = 0.013), 1995 FIGO staging system (IB1 vs. IIA, RR: 2.95, p = 0.004; IB1 vs. IB2, RR: 3.49, p = 0.0009; and IB1 vs. IIB, RR: 4.54, p = 0.00002), and histologic pelvic lymph node involvement (N- vs. N+, RR: 2.94, p = 0.00009). The sequence of adjuvant RT did not influence the probability of DFS (Group I vs. Group II, p = 0.10). In Group II, after univariate analysis, DFS was significantly influenced by histologic residual cervical tumor in the hysterectomy specimen (yes vs. no: 71% vs. 93%, respectively, p < 10(-6)) and by the size of the residual tumor (<or=1 cm vs. >1 cm: 83% vs. 41%, respectively, p = 0.001). The overall postoperative complication rate was 10% in Group I and 9% in Group II (p = 0.7). The rate of postoperative ureteral complications requiring surgical intervention was lower in Group I than in Group II (0.6% vs. 2.3%, respectively, p = 0.03). The overall 10-year rate for Grade 3 and 4 late radiation complications was 10.4%. Postoperative EBPRT significantly increased the 10-year rate for Grade 3 and 4 late radiation complications (yes vs. no: 22% vs. 7%, respectively, p = 0.0002).
CONCLUSION: The prognosis for patients with cervical carcinoma was not influenced by the sequence of adjuvant RT (preoperative uterovaginal brachytherapy vs. postoperative RT) for Stages IB, IIA, and IIB with 1/3 proximal parametrial involvement. However, postoperative EBPRT increased the risk of late radiation complications.