ESTRO 2020 Abstract book

S643 ESTRO 2020

need for new treatment strategies in stage IIIC cervical cancer. PO-1129 Impact of individualized ITV margins for IMRT of cervical cancer on target coverage K. Wada 1 , E. Nakanishi 1 , M. Toratani 1 , T. Ikawa 1 , N. Kanayama 1 , M. Morimoto 1 , K. Konishi 1 , T. Teshima 1 1 Osaka International Cancer Institute, Radiation Oncology, Osaka, Japan Purpose or Objective Intensity-modulated radiation therapy (IMRT) for cervical cancer has high dose conformality and normal tissue sparing. Daily target motion caused by variation in bladder and rectal filling could cause insufficient target coverage. Generating individualized internal target volumes (iITVs) from CT simulation and pretreatment diagnostic images can manage target motion. However, practical and quantitative research on this approach are lacking. We assessed whether an individualized approach improves target coverage without increasing irradiated bowel volume compared to the conventional approach. Material and Methods We retrospectively analyzed 23 patients who underwent definitive treatment with IMRT for cervical cancer from January 2017 to July 2019. Target volume was defined as the gross tumor volume plus remaining cervix and body of uterus. Various isotropic margins (5, 10, 15, 20, and 25 mm) were added to the target volume, delineated from CT simulation and pretreatment bone registration images, to generate iITVs (iITV-x, x=5, 10, 15, 20, and 25), whereas conventional ITVs (cITV-x, x=5, 10, 15, 20, and 25) were generated solely from CT simulations. ITVs were modified to avoid muscles and bony boundaries and expanded by a 5-mm margin for planning target volumes (PTVs) (iPTV-x and cPTV-x, x=5, 10, 15, 20, and 25). Target coverage was evaluated using cone-beam CT (CBCT) for daily setup. The bowel bag volume overlapping PTVs was evaluated as a surrogate for the volume of irradiated bowel. The correlations between the percentage of off-target fractions and factors such as tumor volume and standard deviations (SD) of rectal and bladder volumes on CT simulation and pretreatment images were analyzed. Results In total, 463 CBCTs were evaluated. The result of per fraction analysis is shown in the table. To obtain >90% target coverage, cITV-20 or iITV-10 were necessary. The mean PTV volumes and the overlapping bowel bag volumes of cPTV-20 vs. iPTV-10 were 979.2 cc vs. 731.1 cc ( p <.01, paired t-test) and 197.7 cc vs. 160 cc ( p =.02, paired t- test), respectively. The mean missed target volume was 4.3 cc. The percentage of off-target fractions and SD of rectal volumes in cITV-5, cITV-10, cITV-15, and cITV-20 were correlated. No other factors were correlated with the percentage of off-target fractions.

Chi-squared test was used to evaluate potential associated risk factors for late G≥2 adverse events. Results Median age was 60 years (33-86). All patients completed the treatment as planned, with acute G2 toxicity rates of 22% and 14% respectively for gastrointestinal (GI) and genitourinary (GI), and no evidence of G3 events. Median follow-up was 19 months (6-47). Late G≥2 toxicity incidence was 1.5% for GI and 3.1% for GU at 1 year, while 2.3% and 2.7% at 2 years. Only one case of G3 late adverse event was observed, consisting of a hemorrhagic cystitis requiring hospitalization after 27 months from the end of RT. At statistical analysis we investigated the following dosimetric parameters as potential predictors of late G≥2 toxicity: bowel V15Gy, V25Gy and V40Gy, and for both rectum and bladder Dmax, Dmean, Davg. No significant correlation was found. Concerning clinical outcomes, our progression free survival rates were both 73% at 2- and 3- years for EC and of 74% and 68% for CC. The most frequent site of progression was the lung. 10 patients died by progressive disease, with 2- and 3-years overall survival (OS) rates of 76% and 72% for EC and of 72% and 62% for CC. For both EC and CC the log-rank test revealed no factors significantly related to PFS or OS. Conclusion Our experience supports the use of IG-IMRT for the post- operative treatment of EC and CC, since it leads to a very low incidence of severe adverse events. Furthermore, our preliminary results in terms of clinical outcomes are promising and long-term updates are warranted. PO-1128 Oncologic outcomes of cervical cancer using FIGO 2018 staging in the era of modern radiation therapy R. Dejean 1 , M. Nourmoussavi 2 , H. Bahig 1 , T. Warkus 2 , V. Samouelian 2 , B. Cormier 2 , M. Beauchemin 1 , M. Barkati 1 1 Centre Hospitalier de l'Université de Montréal, Radiation Oncology, Montréal, Canada ; 2 Centre Hospitalier de l'Université de Montréal, Gynecologic Oncology, Montréal, Canada Purpose or Objective Imaging was recently incorporated in the new FIGO 2018 cervical cancer staging, which now includes nodal disease (IIIC). We sought to evaluate the outcomes and patterns of recurrence using FIGO 2018 staging in a cohort of patients treated in an era of high precision image-guided radiotherapy. Material and Methods We carried out a retrospective database review of 196 patients treated at a single institution from 2011 to 2018. All had pre-treatment MRI and PET scans and were treated with concurrent chemoradiation using intensity-modulated radiotherapy and 3D-image guided brachytherapy. Previously assigned 2009 FIGO staging and all imaging were reviewed for FIGO 2018 reclassification. Kaplan-Meier survival curves and Log-Rank test were performed. Results Five-year actuarial OS, DFS and locoregional control rates were 80%, 82% and 86%,respectively, with a median follow- up time of 51 months (Range 4-96 months). Forty-eight patients experienced a recurrence, 80% of which included distant disease. In-field recurrence rate was 11.7%. Using the new FIGO staging, 48% of patients were upstaged to IIIC. Compared to FIGO 2009, OS and DFS were higher for all 2018 FIGO stages. Five-year OS was 100% for early stage disease (IA1-IIA2) comparatively to 83.0%, 72.7% and 62.5% in node-negative locally advanced disease, IIIC1 and IIIC2 stages, respectively (p<0.0001). Conclusion Integration of state-of-the-art imaging in cervical cancer staging as well as in radiotherapy and brachytherapy planning leads to excellent oncologic outcomes, however distant recurrence remains an important issue. FIGO 2018 staging better reflects patient prognosis, highlighting the