Impact of dose in outcome of irradiation alone in carcinoma of the uterine cervix: analysis of two different methods

This is a retrospective analysis of 1211 patients with histologically proven invasive carcinoma of the uterine cervix with a minimum follow-up of 3 years treated with irradiation alone. The pelvic failure rates by stage were 9.6% for IB, 18.6% for IIA, 23% for IIB, 41% for III, and 75% for Stage IVA disease. External beam and intracavitary irradiation doses to point A and pelvic lymph nodes were calculated. In patients with Stage IB and IIA disease there was no significant correlation between doses to these points and pelvic tumor control. In Stage IIB doses of less than 6000 cGy to point A correlated with a high pelvic failure rate (8 of 12, 66.7%) in contrast to doses of 6000 to 9000 cGy (61 of 261, 23.4%) or higher than 9000 cGy (10 of 74, 13.5%) (P 0.01). In Stage III the pelvic failure rate with doses below 6000 cGy to point A was 72% (18 of 25) compared to 39% (71 of 180) for 6000 to 9000 cGy or 35% (27 of 77) with doses above 9000 cGy (p ≤ 0.01). TDF calculation of doses was carried out. In Stage IB and IIA there was no significant correlation between TDF to point A and probability of pelvic recurrence. In Stage IIB with TDF below 135, the pelvic recurrence rate was 41.6% (20 of 48) compared to 20% (61 of 305) with higher TDF (p ≤ 0.01). In Stage III the pelvic failure rate was 51% with TDF below 160 (70 of 136) in comparison with 29.5% (46 of 156) with higher TDF (p ≤ 0.01). Grade 2 sequelae of therapy were noted in about 10% of the patients and grade 3 in 4.7% of patients with Stage IB (18 of 384), 10.2% (12 of 128) with Stage IIA, 9.3% (33 of 353) with Stage IIB, and 8.2% (24 of 293) with Stage III disease. Doses from external beam and intracavitary irradiation to the rectum or the bladder neck were calculated. The actuarial incidence of major rectal or rectosigmoid sequelae was 2% to 4% with doses to the rectum of 6000 to 8000 cGy, 7 % to 8 % with 8000 to 9500 cGy, and 13% with doses higher than 9500 cGy (p <- 0.01). A correlation was also found between the TDF to the rectum and frequency of severe rectum/rectosigmoid sequelae (2.5 % with TDF of 160 or less, 5.5 % with 160 to 185, and 9.9 % with TDF above 185). The incidence of severe urinary sequelae was less than 2% with TDF below 135 and about 4% with higher TDF. The incidence of severe small bowel injury was 1% (3 of 328) with doses below 5000 cGy, 2.5% (6 of 231) with doses of 5000 to 6000 cGy, and 3.6% (23 of 629) with higher doses. We conclude that TDF calculations are accurate predictors of probability of pelvic tumor control and may be more discriminating than "physical dose" computations. TDF also correlates well with incidence of sequele of therapy. Phase II randomized studies are strongly recommended to establish optimal tumor doses for various stages/volumes of carcinoma of the uterine cervix.