ESTRO 2023 - Abstract Book

S794

Monday 15 May 2023

ESTRO 2023

Data from consecutive patients treated with CIRT from January 2014 to December 2021 for a new diagnosis or a recurrence of unresectable STS were retrospectively collected. Patients were classified as highly radioresistant sarcoma and conventionally radioresistant ones according to the radiosensitivity index previously published. The endpoints of this study were: 5- years local recurrence-free survival (LRFS), the 3 and 5-years overall survival (OS) and acute/late toxicity rates. Actuarial survival outcomes were evaluated using the Kaplan-Meier method. Any correlation between histological type, GTV volume and LRFS was tested using Wilcoxon rank test. Statistical significance was defined as p-value < 0.05. Acute and late toxicities were scored according to the Common Terminology Criteria for Adverse Events (CTCAE) v5.0.C. Results A total of 24 patients with a median age of 52 (range, 13-79) underwent radical CIRT. Eleven (46%) of the analyzed patients were treated for the primary disease and 13 (54%) at time of recurrence. 13 (54%) were males and 11 (46%) were females. None of the patients received previous irradiation at CIRT-target. CIRT was delivered for a thoracic localization in 14 (58%) cases and for abdominal in 10 (42%). The most frequent histology was liposarcoma (20%) followed by leiomyosarcoma (13%). There were 15 (63%) highly radioresistant and 9 (37%) conventionally radioresistant tumors. Median GTV was 263 cc (range 14-3603). Eleven patients (44%) received at least one line of chemotherapy before CIRT. The median prescription dose was 64 Gy(RBE) (range 50.4-76.8), with a median dose per fraction of 3.5 Gy(RBE) (range 3-4.8). After a median follow-up of 31 months (range 1-100), 10 patients (41.6%) experienced an in-field recurrence (local recurrence) and the 5-year LRFS rate was 33.1% (95% CI: 14.0%-77.9%). The radioresistant histology (p=0.7) and the GTV (p = 0.93) were not related with LRFS. The 3-year and 5-year OS was 86% (95% CI: 73.4%-100%) and 72.0% (95% CI: 55.2%-93.8%) (Fig1). GTV volume was not significantly correlated with OS (p = 0.15). With regards to toxicity, no patients needed treatment interruption; the most common acute tocixities were G2 pain and G2 erythema while the most frequent late toxicity was G2 peripheral neuropathy. No G ≥ 3 toxicities were reported.

Conclusion CIRT appeared a safe non-invasive option treatment for unresectable STS, but clinical trials with a longer follow-up and larger cohort are necessary to evaluate the effectiveness and late toxicity. MO-0946 Sacral nerve sparing strategy in carbon-ion irradiation of pelvic sarcomas: the MedAustron approach A. Nachankar 1,2 , E. Hug 1 , M. Pelak 1 , S. Tubin 1 , M. Stock 3,4 , A. Carlino 4 , J. Gora 4 , G. Martino 4 , M. Schafasand 4,5 , P. Fossati 1,6 1 MedAustron Ion Therapy Center, Austria, Radiation Oncology, Wiener Neustadt, Austria; 2 ACMIT Gmbh, Austria, Research & Development in Medical Technology, Wiener Neustadt, Austria; 3 Karl Landsteiner University of Health Sciences, Austria, Medical Physics, Krems an der Donau, Austria; 4 MedAustron Ion Therapy Center, Austria, Medical Physics, Wiener Neustadt, Austria; 5 Medical University of Vienna, Austria, Radiation Oncology, Wien, Austria; 6 Karl Landsteiner University of Health Sciences, Austria, Radiation Oncology, Krems an der Donau, Austria Purpose or Objective Purpose/Objective: Carbon-ion therapy (CIRT) is a potentially curative treatment for inoperable pelvic sarcomas/chordomas. A possibly debilitating side-effect of pelvic hypofractionated CIRT is Radiation-Induced Lumbosacral Neuropathy (RILSN), if doses to nerves >70Gy RBE. At present no clear recommendations are available for sacral nerve sparing optimized (SNSo) CIRT treatment planning. The purpose of this study is to evaluate the feasibility of SNSo-CIRT plans and correlate it with initial, preliminary clinical outcomes data. Materials and Methods Material/Methods: We contoured individual nerve roots at L4–S3 levels until they converge to the sciatic nerves (figure 1a). CIRT plans with pencil beam scanning (PBS) were optimized with local effect model -I (LEM) (Europe) to calculate relative biological effectiveness (RBE)-weighted doses. To validate dose distributions of SNSo-CIRT plans across different RBE models, doses were recalculated using the modified microdosimetric kinetic model (mMKM) (NIRS, Japan-scanning beam). Prescription doses were (LEM) of 73.6 (64-76.8) Gy RBE in 16 fractions of 4-4.8 Gy RBE, (4 fractions/ week). Doses to nerves outside of high dose clinical target volume (HD-CTV) were limited to D2% <69 or <73 Gy RBE and hot spots were avoided on nerves inside HD-CTV. Robustness of SNSo-CIRT plans were evaluated against different range and set-up uncertainty scenarios.