ESTRO 2021 Abstract Book

S1033

ESTRO 2021

Conclusion Patients with adenocarcinoma of upper rectum and recto-sigmoid having adverse prognostic features treated with NACTRT experienced down staging of the disease and a trend towards better disease free survival compared to early stage patients undergoing upfront surgery. PO-1252 Chemoradiation for anal cancer: correlation of hematologic toxicity and dose to pelvic bone marrow. E. Ferrara 1 , I. Luciani 1 , M. Zannetti 1 , F. Mastroleo 1 , A. Brignoli 1 , B. Mattio 1 , L. Turri 1 , L. Vigna 2 , E. Mones 2 , G. Loi 2 , M. Krengli 1,3 1 AOU "Maggiore della Carità" University Hospital, Radiation Oncology, Novara, Italy; 2 AOU "Maggiore della Carità" University Hospital, Medical Physics, Novara, Italy; 3 University of Piemonte Orientale (UPO), Department of Translation Medicine, Novara, Italy Purpose or Objective Hematologic toxicity (HT) in patients (pts) undergoing chemoradiation (RCT) for anal cancer still represents one of the major causes of treatment interruption, leading to an increase of overall treatment time and worsening of quality of life. The aim of this work is to evaluate a dosimetric correlation between the dose received by specific pelvic bone marrow subsites and the HT. Materials and Methods We retrospectively selected 65 pts who received concurrent RCT for anal cancer from 2011 to 2020. Pts underwent pelvic intensity-modulate radiotherapy (IMRT) or volumetric-modulated-arc radiotherapy (VMAT) with total dose of 45 Gy to prophylactic nodes and 50.4 Gy to intermediate risk volume with a boost to gross tumor volume up to 59.4 Gy. We prescribed concomitant chemotherapy (5- fluorouracil 1000 mg/m2 day 1-4 + mitomycin bolus 10 mg/m2 day 1, q28) for 2 concomitant cycles. To assess a correlation between dosimetric parameters (mean dose, V5, V10, V15, V20, V30, V40, V45) of bone marrow (BM) and HT, we identified 4 sub- volumes in the pelvic bones: iliac crests (IC), lumbar sacral BM (LSBM), lower pelvis (LP), composed by pubis + ischia + acetabula + femoral heads, and the sum of all subsites defined as total pelvic bone marrow (TPBM). HT was evaluated according to RTOG scoring scale; endpoints included white blood-cells count (WBC), neutrophil-count (NC), hemoglobin (Hb) and platelet (PTL) nadirs. Results 46/50 (70,7%) of pts were women; 3 pts (4,6%) were HIV positive. Clinical stage was as follow: 27,7 % T2, 41,5 % T3 and 24,6 % T4; 67,7 % of pts had positive nodes. RCT was interrupted for more than 3 days in 10 pts due to HT. 40 % of pts showed G3 HT toxicity: 27,7% WBC, 23,1 % NC and 10,8 % PTL. A multivariated linear regression model assessed a statistically significant correlation between the HT and the following dosimetric parameters: V40 LSBM (p = 0.01); V45 LSBM (p = 0.008); V40 TPBM (p = 0.03); V45 TPBM (p = 0.047). These correlations were confirmed by non-parametric Spearman analysis in particular for V45 TPBM (p=0.0001). A statistically significant correlation between V45 TPBM and the risk of HT > grade 3 was found at the logistic regression (OR 1.128, SE 0.079, p = 0.085, 95% CI 0.983 / 1.296). The ROC curve analysis showed that V45 TPBM was the best classifier for toxicity in this patient cohort. V45 TPBM > 17% resulted to predict HT > grade 3 with a sensitivity of 100%. Conclusion With the introduction of more conformal techniques such as IMRT and VMAT there is an increase of low-dose to pelvic bone structure; however, there is also the possibility to model treatments in order to spare BM and reduce HT toxicity. From our analysis, V45 TPBM dosimetric parameter could be able to predict HT > grade 3 (V45 TPBM > 17%). This parameter could be used to improve toxicity profile by treatment optimization.