ESTRO 2022 - Abstract Book

S178

Abstract book

ESTRO 2022

MO-0220 Dose-volume constraints for gastric and duodenal toxicity in pancreas moderately hypofractionated RT

S. BROGGI 1 , P. Passoni 2 , P. Tiberio 1 , A. Cicchetti 3 , B. Longobardi 4 , N. Slim 2 , M. Reni 5 , A. Del Vecchio 1 , N.G. Di Muzio 2 , C. Fiorino 1 1 San Raffaele Scientific Institute, Medical Physics, Milano, Italy; 2 San Raffaele Scientific Institute, Radiotherapy, Milano, Italy; 3 San Raffaele Scientific Institute, Medical Phyiscs, Milano, Italy; 4 San Raffaele Scientific Institute, Medical Phyisics, Milano, Italy; 5 San Raffaele Scientific Institute, Oncology, Milano, Italy Purpose or Objective Hypofractionated radiotherapy (RT) of pancreatic adenocarcinoma is challenging, due to the tolerance of adjacent normal tissues. Aim of the study was to assess dose-volume histogram (DVH) related predictors of acute and late gastric and duodenal toxicities for locally advanced pancreatic (LAP) patients. Materials and Methods 200 patients with histologically proven pancreatic adenocarcinoma treated in the period 2004-2019 were considered. All patients received induction chemotherapy (ChT) followed by concurrent chemoradiotherapy (CRT). Delivered dose was 44.25 Gy in 15 fractions to pancreatic tumor and lymphonodes radiologically involved. 28 pts received a simultaneous integrated boost with doses in the range 48-55 Gy to a tumor sub-volume infiltrating the vessels. RT was delivered with Helical Tomotherapy. Duodenum and stomach absolute DVHs were collected and analyzed. Gastric/duodenal CTCAEv5 Grade ≥ 2 toxicities were considered. First, absolute DVHs of patients with/without toxicities were compared through two- sided t-test; the DVH points corresponding to the lowest p-values and the near D max value (D 0.03 ,dose received by 0.03 cc of OAR) were included and tested in univariate and multivariate logistic regressions, testing the association between these parameters and the risk of toxicity e. The best cut-off values discriminating between patients with/without toxicity were also assessed through a ROC analysis. Results Grade ≥ 2 toxicity occurred in 18 patients (6 acute and 12 late): 11 duodenal (ulcer, duodenitis) and 16 gastric (ulcer, stomatitis) toxicities were reported. Both for stomach and duodenum, V44 and V15 were selected as the most promising discriminating DVH parameters (figure 1). At univariate analysis, duodenum V44 (p= 0.01; OR= 1.07) and D 0.03 (p=0.03; OR=1.2) were found the most predictive parameters for duodenal toxicity; V44 (p= 0.0033) was confirmed at multivariate analysis. Regarding gastric toxicity, stomach D 0.03 (p= 0.03; OR=1.20) was the only significant predictor. The best duodenum V44 and D 0.03 cut-off values were 9.1 cc and 46.7 Gy; for both parameters a negative predictive value equal to 97% was found. Concerning stomach, D 0.03 > 45 Gy was the best cut-off value with a negative predictive value of 94%. The crude rates of duodenal toxicity were 4/172 (2.3%) vs 5/23 (21.7%) if duodenum V44 < 9.1 cc and ≥ 9.1 cc respectively. The crude rates of gastric toxicities were 9/144 (6.2%) and 7/56 (12.5%) if stomach D 0.03 < 45 Gy and ≥ 45 Gy respectively.

Conclusion In a large population of patients treated with hypofractionated radiotherapy for LAP cancer, the risk of duodenal and gastric toxicities was found to be related to the high dose region of duodenum and stomach DVH. The suggested constraints