Cervical tumor control evaluated with ICRU 38 reference volumes and integrated reference air kerma

Purpose: The purpose of this investigation was to evaluate the relationship of ICRU 38 reference volumes and integrated reference air kerma to primary cervical tumor control. Materials and methods: This retrospective study includes 1253 women with carcinoma of the uterine cervix treated with radiotherapy. Patients were treated from 1959 to 1993 at the Mallinckrodt Institute of Radiology, There were 39 women with stage Ia disease, 211 with stage Ib1, 88 with stage Ib2, 580 with stage 2, and 335 with stage 3 disease. Most patients with stage Ia disease were treated with low dose rate brachytherapy alone. All other patients were treated with external irradiation and low dose rate brachytherapy. External irradiation doses ranged from 8.2-63.8 Gy to the whole pelvis and an additional parametrial boost to deliver a total of 65-75 Gy to Point P depending on tumor stage. Low dose rate brachytherapy was delivered to the primary tumor with doses ranging from 40x10(4)-60 x 10(4) Gy cm(2) (integrated reference air kerma; IRAK). Isodose reference volumes (60-160 Gy) and point A doses were determined for all patients. Tumor control was defined as control of the primary cervical cancer with or without failure at other pelvic or distant sites. Results: The mean 60 Gy volume was calculated to be 139.5 cm(3) for stage Ia, 200.4 cm(3) for stage Tol, 270.9 cm(3) for stage Ib2, 235 cm(3) for stage 2, and 293.4 cm(3) for stage 3. No patient with stage Ia disease had a failure in the cervix. For stage Ib1 tumors the mean 60 Gy volume was 219.1 cm(3) for those who failed in the cervix and 199.9 cm(3) for those who did not fail (P = 0.73). For stage Ib2 tumors the mean 60 Gy volume was 354.4 cm(3) for those who failed and 260.2 cm(3) for those who did not fail (P = 0.004). The mean 60 Gy volume was 249.3 cm(3) for those with stage 2 disease who failed and 233.8 cm(3) for those who did not fail (P = 0.02). For patients with stage 3 cancer the mean 60 Gy volume was 321.6 cm(3) for those who failed and 287.3 cm(3) for those who did not fail (P = 0.20). Reference volumes from 70-160 Gy were not statistically different within each stage for those who failed in the cervix compared to those who did not fail. Cervical recurrences by clinical stage did not have statistically significant differences for mean IRAK except for stage IT disease (P = 0.001). Conclusion: Analysis of the 60-160 Gy reference volumes and IRAK failed to demonstrate a consistent positive increasing correlation of these values to primary cervical tumor control. (C) 2001 Elsevier Science Ireland Ltd. All rights reserved.