ESTRO 2021 Abstract Book

S1014

ESTRO 2021

(6.1%). G ≥2 and G ≥3 late toxicity occured in 11 (18.6%) and 9 (5.7%) out of 59 pts treated with SIB, respectively, significantly higher (p<0.01) than the corresponding values for pts treated without SIB (9 and 6/185, 4,8% and 3.2% respectively). Median OS, PFS, DPFS and LPFS, assessed in stage III, were 26.0, 11.9, 12.8 and 15.1 months, respectively. At univariate analysis, gender, and CA 19.9 value before RT were the only predictive variables for OS. Ca 19.9 value before RT was confirmed at multivariate analysis: median OS was 20.0 and 29.3 months in patients with GICA value pre-RT higher or lower to the median value (91), respectively (p=0.05). Gender (p= 0.03) and GICA value pre RT (p=0.04) were also the only predictive variables for DPFS while none of the considered variables were associated to LPFS. Conclusion Hypofrationated RT is feasible after induction Cht with acceptable acute and late toxicity rate. A dose > 44.25 Gy is significantly more toxic even when delivered on small sub-volumes. With respect to literature, outcome of present study compares favourably with results obtained with standard fractionation. PO-1226 Inhomogeneous dose escalation in pancreatic SBRT:feasibility and impact of anatomical configuration M. Loi 1 , R. Doro 2 , S. Lucidi 1 , P. Bonomo 1 , G. Simontacchi 1 , D. Greto 1 , A. Allegra 1 , V. Di Cataldo 1 , G. Francolini 1 , I. Bonucci 2 , L. Livi 3 , L. Masi 2 1 Azienda Ospedaliero-Universitaria Careggi, Radiation Oncology Unit, Florence, Italy; 2 IFCA, Department of Medical Physics, Radiation Oncology, Florence, Italy; 3 Azienda Ospedaliera Universitaria Careggi, Radiation Oncology Unit, Florence, Italy Purpose or Objective In non-resectable locally advanced pancreatic cancer, adjunction of Stereotactic Body Radiotherapy (SBRT) following chemotherapy has been proposed to improve outcome. However, prescription of more intensive schedules has been traditionally limited by risk of severe injury to nearby organs at risk (OARs), particularly in case of substantial overlap between planning volume and critical structures. To achieve higher tumor control probability, a biologically effective dose (BED) of 100 Gy 10 , corresponding to a dose 50 Gy in 5 fractions of 10 Gy assuming α/β=10, is desirable. The aim of our in-silico study is to assess the feasibility of dose escalation using inhomogeneous dose prescription for pancreatic SBRT and to determine which patients may be eligible for this strategy based on anatomical proximity between target volumes and OARS. Materials and Methods Data from 14 locally advanced pancreatic cancer patients treated at our Institution were collected. For each patient, a CyberKnife (CK) Synchrony treatment plan was optimized for fiducial-guided pancreatic SBRT aiming at a planned dose of 50 Gy and 40 Gy in 5 fractions to the gross tumor volume (GTV) and the planning target volume (PTV), respectively. PTV was created by 5 mm GTV isotropic expansion. Acceptable target coverages were: 1) a dose of 50 Gy and 47.5 Gy to at least 90% and 95% of the GTV, and 2) a dose of 40 Gy to 95% of the PTV. Planned doses to the target regions and OARs (duodenum, stomach and bowels) were evaluated and statistically analyzed. For each plan, the intersection volume between the PTV and OARs expanded by 5 mm was defined as Expansion-Intersection Volume (EIV), as reported by Tomatis et al. Results Median GTV and PTV volumes were 40.8 (range 22.3-205.3) cc and 73.7 (range 36.1-266.7) cc , respectively. V35 to duodenum, stomach and bowel was maintained below 0.5 cc in all cases. Median EIV was 12.9 (3.9- 25.1) cc. Median V50 and V47.5 for GTV was 91.0% (range 82.4%-97.8%) and 96.8% (range 92.5%-99.9%), respectively: GTV coverage was acceptable in 10 out of 14 cases. Median V 40Gy for PTV was 96.8% (range 90.0%- 99.8%): PTV coverage was acceptable in 11 out of 14 cases. Spearman correlation showed a significant association between EIV and V 47.5Gy for GTV (rho -0,77228, p<0.001) and V 40Gy for PTV (rho -0,68352, p<0.001), respectively.

Conclusion Inhomogenous dose escalated prescription using fiducial-based SBRT with Cyberknife respiratory tracking is a feasible strategy in selected patients with locally advanced pancreatic cancer. EIV, but GTV size, is significantly correlated with target coverage probability and may provide a simple tool to identify patients eligible for dose escalation.