671 poster The outcome and patterns of failure following postoperative radiation therapy for endometrial cancer

$292

Introduction: Concurrent radiochemotherapy is associated with higher risk for normal tissue toxicity. The tolerance of healthy organs depends on normal tissue volume in irradiated field, total dose, dose per fraction and chemotherapy regimen. The aim of this study was to estimate the risk factors for normal tissue complications in uterine cervix cancer patients treated with concurrent radiochemotherapy. Materials and methods: The normal tissue complication risk factors were estimated in the group of 130 pts treated for uterine cervix cancer from June 1999 to March 2002 with concurrent radiochemotherap~/. Chemotherapy was delivered in weekly doses of 40mg/m of Cisplatin (on average 5,2 cycles). Radiotherapy consisted of teletherapy (9MeV, 15MeV) with conventional fractionation (medium dose to parametria of 49,7Gy) and brachytherapy LDR (2 fraction, medium dose delivered to point A of 48Gy). Total averaged treatment time was 53,7days. Late and early lesions in rectum and bladder according to EORTC/RTOG scale were analysed. The influence of the following factors: a) the total dose in mentioned organs at risk in the reference points according to ICRU, b) patient's age, c) coexisting diseases (eg. diabetes), d) Hb level, e) total treatment time and f) the number of chemotherapy, on toxicity was estimated. Results: The total dose (equal or above 70Gy, in the ICRU reference point) appeared to be the significant risk factor for early reaction in rectum, p = 0,008. Diabetes is the risk factor for early (p = 0,008) and late (p --0,013) complications in rectum. Data - mining and tree regression analysis showed that total dose in the rectum and bladder reference points and overall treatment time are important risk factors for early lesions. Conclusion: The significant risk factors for normal tissue complication in patient treated with concurrent radiochemotherapy appeared to be: total radiation dose in rectum (early reaction), diabetes (early and late reaction) and the total treatment time (early reaction). Anemia and number of chemotherapy cycles have not appeared to be significant risk factors for complications associated with radiation. 670 poster Concurrent chemoradiotherapy using high-dose-rate intracavitary brachytherapy for uterine cervical cancer T. Toita ~, H. Moromizato 2, Y. Kakinohana ~, T. Maehama 2, Y. -Nagai , K. Ogawa 1, W. Tamaki 1, K. Kanazawa 2, S. Murayama ~ 1University of the Ryukyus, Department of Radiology, Okinawa, Japan 2University of the Ryukyus, Department of Obstetrics and Gynecology, Okinawa, Japan Purpose: We retrospectively review our experience with concurrent chemoradiotherapy (CCRT) using high-dose-rate intracavitary brachytherapy (HDR-ICBT) to assess its feasibility and efficacy for patients with uterine cervical cancer. Patients and Methods: Forty patients with uterine cervical squamous cell carcinoma treated with CCRT using HDRICBT were analyzed. The median cervical tumors size assessed by MRI was 63 mm (range: 40-86 mm). Eighteen o patients (45Yo) had enlarged pelvic nodes on MRI ( >1 0 m m ) . Cisplatin (20 mg/m2/day) was concurrently administered with radiotherapy for five days at 21 -day intervals for a median of three courses (range: 1-5 courses). Thirty-eight (95%) patients received EBRT of 40 Gy/20 fractions to the central pelvis, and HDR-ICBT of 18Gy/3 fractions to point A. The cumulative biological effective dose (BED) at point A of this schedule was 77 Gyl0. The median overall treatment time,

Posters

including HDR-ICBT and boost EBRT, was 48 days (range: 37-68 days). The median follow-up period of all 40 patients was 37 months (range: 8-71 months). Results: Grade 3/4 leukopenia was the most common acute side effect (83%). There were no treatment-related deaths. At the time of the analysis, eight patients (20%) were dead, all from cervical cancer. Six patients were alive with disease, and the remaining 26 patients were alive with no evidence of disease. Two patients had pelvic recurrence alone, 11 had distant metastases alone, and one had both pelvic and distant failures. The actuarial 3-year pelvic control rate, disease-free survival rate, and overall survival rate were 91%, 67%, and 79%, respectively. Eight (20%) patients suffered late gastrointestinal complications (all grades). No patient suffered radiation cystitis (all grades). Only one patient experienced > grade 3 complication (enterocolitis). The actuarial 3-year late complication rate (all grades) was 9% for proctitis, and 15% for enterocolitis, respectively. Conclusion: This preliminary study suggested that CCRT using HDR-ICBT was feasible and efficacious for patients with uterine cervical cancer. 671 poster The outcome and patterns of failure following postoperative radiation therapy for endometrial cancer S. Kim 1, H. Wu 2, S. Ha 2

~St. Vincent's Hospital, The Catholic University of, Radiation Oncology, Suwon, Gyeonggi-do, Korea 2Seoul National University College of Medicine, Therapeutic Radiology, Seoul, Korea Purpose: To investigate the outcome and patterns of failure according to FIGO stage in endometrial cancer patients treated with surgery and postoperative radiation therapy (RT). Materials and Methods: Eighty-three patients with endometrial cancer who received postoperative RT between May 1979 and August 2000 and followed up for more than 24 months were included in this retrospective study. Forty-one patients received total abdominal hysterectomy, 41 Wertheim's operation, and 1 vaginal hysterectomy (in one of the stage IB patients). Pelvic lymph node dissection or pelvic lymph node sampling was done in 54 patients and peritoneal cytology in 35. All patients were staged according to 1988 FIGO staging system; 2 stage IA, 24 stage IB, 19 stage IC, 4 stage IIA, 6 stage liB, 9 stage IliA, 2 stage IIIB, and 17 stage IIIC. Histologic diagnoses were adenocarcinoma in seventyfour patients (89%). Histologic grade were Grade 1, 2, and 3 in 21 (25%), 43 (52%), and 10 (12%) patients, respectively. All patients received external beam RT (EBRT) 4500-5075 (median: 5040) cGy to the whole pelvis field. Stage IIIC patients with para-aortic lymph node metastasis received 4500 cGy to the para-aortic field. Fifteen patients received low-dose rate brachytherapy after EBRT. Total 7500-9540 cGy was prescribed to the vaginal surface. Results: With a median follow-up period of 86 months, overall and disease-free survival (DFS) at 5 years was 87% and 85%, respectively. 5-year overall survival rate was 91%, 100%, and 78% for stage I, II and III patients, respectively. Overall, 13 patients (16%) relapsed. Ten relapsed in the extrapelvic sites and 3 in the pelvis. In stage I or II patients, 2 (4%) relapsed in the vagina, 1(2%) in the para-aortic lymph node (pAN), and 2 (4%) in the distant organs. In stage III patients, 1 (4%) relapsed in the vagina, 2 (7%) in the PAN, 3 (11%) in the peritoneum, and 2 (7%) in the distant organs. On univariate analysis, FIGO stage (I, II vs. III, p=0.02), the depth of myometriat invasion (none or <1/2 thickness vs. >1/2 thickness, p=0.01) and parametrial invasion (no vs. yes,

Posters

$293

p=0.02) were significant prognostic factors for DFS. On multivariate analysis, >1/2 thickness myometrial invasion (p=0.03) was the only significant adverse factor for DFS. Three patients experienced significant late RT complication: radiation cystitis in 1 patient, bowel obstruction in 1, and radiation enteritis in 1 who had been treated with para-aortic and whole pelvis field. Conclusions: With surgery postoperative RT, high survival rate could be achieved in stage III patients as well as stage I, II patients. Major patterns of failure in stage Ill endometrial cancer patients were peritoneal seeding and distant metastasis. Postoperative RT was useful in controlling pelvic disease. Development of effective systemic-chemotherapy should be considered in stage III endometrial cancer patients. Site of failure

IStage I, S t a g e III

Vagina

!2155(4%) 1/28(4%)

Paraaortic LN metastasisi1/55(2%)ii2/28(7%) Peritoneal seeding

0/55(0%)

3/28(11%)

Distant metastasis

12/55(4%) 12/28(7%)

672 poster External radiotherapy of cervical cancer after lateral ovarian transposition: data required for ovarian dose estimation

M. Mazonakis ~, J. Damilakis ~, H. Varveris2, A. Tzedakis ~, N. Gourtsoyiannis 3 1University Hospital of Iraklion, Department of Medical Physics, Iraklion, Greece 2University Hospital of lraklion, Department of Radiotherapy, Iraklion, Greece 3University Hospital of Iraklion, Department of Radiology, Iraklion, Greece Purpose: To estimate the radiation dose received by the laterally transposed ovaries during external beam radiotherapy for cervical cancer. Materials and methods: A humanoid Rando phantom was used to simulate radiotherapy in female patients. The phantom was modified to allow the placement of an ionization chamber into points corresponding to the positions of a transposed ovary. Anteroposterior (AP), posteroanterior (PA), right and left lateral field irradiations were generated using a 6 MV photon beam. The dependence of the ovarian dose on the field size and on the distance separating the transposed ovary from the primary irradiation field was investigated. Lead shielding, 8-cm-thick, was used to reduce the ovarian dose. The effect of a universal wedge on the ovarian dose was determined. Results: For a treatment course giving 45 Gy to the tumor, the radiation dose to a laterally transposed ovary was 101-840 cGy depending upon the dimensions of the irradiation fields and the distance from the field edge. The positioning of gonadal shielding on a shadow tray reduced the ovarian dose by 11-21%. The use of wedged lateral beams increased the ovarian dose by a factor of 1.5. For AP, PA and lateral treatment fields, graphs showing dose values to a transposed ovary at all possible ovarian positions relative to the field edge using various field sizes are presented. Conclusion: This study provides the required data to estimate the radiation dose to ovaries for individual patients with cervical cancer undergoing external radiotherapy following lateral ovarian transposition.

673 poster Chemo-irradiation versus radiotherapy alone in locally advanced carcinoma of the uterine cervix - An ongoing phase III trial

E. SaibishKumar, P. Lal, A. Tiwari, S. Kumar, K. Dimri, N. Rastogi, S. Ayyagari Sanjay Gandhi P-G Institute of Medical Sciences, Radiotherapy, Lucknow, India Background: A prospective randomized trial was initiated to test the efficacy of concurrent cisplatin (CDDP) with radiotherapy in patients of carcinoma cervix. Methods: Between December 2000 and January 2004, 101 cases of histology proven previously untreated carcinoma cervix (FIGO stages II to IV) were randomized to either RT alone (n=49) or CT+ RT (n=52). Whole pelvic RT (identical in both arms, 50Gy/25 fractions/5weeks) was followed 1 week later by brachytherapy of 3 fractions of 6Gy HDR to point 'A' spaced a week apart. Chemotherapy consisted of weekly CDDP at 35mg/m 2 (max 50 mg). Results: The median follow up was 19 months. Radiotherapy compliance was seen in 96% cases and scheduled 5 cycles of chemotherapy could be delivered to 82% cases. Nausea and vomiting requiring intervention (Grade II RTOG toxicity) were significantly higher in the CT+RT arm (24% vs. 8%, Z 2 p = 0.00) as were grade II hematological complications (24% vs. 11%, Z 2 p = 0.01). Pelvic control rate was similar for the RT and CT+RT groups at 71% and 69% at 3 years respectively; distant disease free survival 64% and 58% at 3 years, and overall survival identical at 65% at 3 years respectively (all p values non-significant). There was no difference observed in late toxicity in the 2 arms. Conclusions: Concurrently administered CDDP in a weekly schedule with pelvic RT is well tolerated. With the relatively short follow up and limited sample size, no difference in survival is evident in this preliminary analysis.

Head and neck c a n c e r 674 poster Pretreatment volumetric assessment of major salivary glands in head and neck cancer patients

G. Studer 1, J. Waldron ~, D. Jaffray 2, A. Bayley 1, B. Cummings 1, J. Kim ~, J. Ringash 1, B. O'Sulfivan~ IPrincess Margaret Hospital/University of Toronto, Radiation Oncology, Toronto, Canada 2Princess Margaret Hospital/University of Toronto, Radiation Physics, Toronto, Canada Purpose: Intensity modulated radiation therapy (IMRT) offers the opportunity to spare the parotid (PG) and submandibular (SG) salivary glands during the delivery of radiation for head and neck cancer (HNC). In order to optimize salivary sparing a detailed knowledge of the anatomic and volumetric characteristics of these structures is required. The aim of this study was to assess volumetric pre-treatment data for normal PG and SG in HNC patients. Patients and Methods: Normal PG and SG in 100 selected patients (m: f = 1: 1) subjected to curative RT for HNC were assessed. Pre-treatment images (CT or MRI) were electronically imported into the Pinnacle® - V 6.2 b radiation planning system. PG and SG were contoured bilaterally. The volumes of these structures were calculated and analyzed according to gender and age. Intraindividual volume variance of the paired organs were also calculated according to gland (PG vs SG) and laterality.