ESTRO 2024 - Abstract Book

S1997

Clinical - Mixed sites, palliation

ESTRO 2024

prescribed to the gross tumor volume (GTV) (D99% ≥ 95%). OARs within 3 cm from the PTV (axial plane) were delineated. The stomach, duodenum, small bowel, large bowel, and rectum were contoured as separate structures and, in the following, referred to as GI OAR. GI toxicity included anorexia, enteritis/colitis, diarrhea, dyspepsia, fistula, hemorrhage, ileus, nausea, perforation, stenosis, ulcer, and vomiting. TRAEs were evaluated by NCI Common Terminology Criteria for Adverse Events version (CTCAE) at baseline and weeks 2, 6, 12, 36, and 52. The median follow-up time was calculated using the inverse Kaplan-Meier method [4]. The 1-year cumulative SABR-related toxicity was analyzed based on the highest toxicity grade registered per patient and assessed using the one minus Kaplan-Meier method reported with the 95% confidence intervals (95% CI). The statistical significance level was set at 0.05. Model performance assessment with calibration, Brier score, and Receiver Operating characteristics curve was assessed using the RiskRegression package of R. Between October 2019 and February 2022, 121 patients were enrolled and treated for 147 targets (139 treatment plans). The median follow-up time for toxicity was 11.8 months (IQR 9.8 – 12.7 months). The most frequent diagnoses were prostate-, GI-, lung-, and kidney cancer. Targets were mainly located in the liver (41%), lymph nodes (35%), or adrenal glands (14%). Most patients had an ECOG performance status (PS) of 0 (56%), 36% had a PS of 1, and 7% had a PS of 2. Nearly half of all targets (48%, n = 71) were within 10 mm of a radiosensitive OAR defined as the esophagus, intestines, spinal cord, cauda equina, or sacral plexus. In total, 70% of the patients (n = 85) experienced at least one TRAE (any grade). Overall, no grade ≥ 4 TRAEs, 3.5 % grade 3 TRAEs (0.1% – 6.9%), and 43.7% grade 2 TRAEs (34.1% – 51.9%) were registered within the first year of follow-up. Specifications of TRAEs by CTCAE category are presented in Table 1. Fifty patients (41%) experienced at least one treatment-related GI toxicity (Grade ≥ 1). We found a significant association between grade ≥ 2 GI toxicity and the clinical variables PS and target location in the adrenal glands. No associations were found between grade ≥ 2 GI toxicity and other target locations, primary tumor types, or treatment with systemic therapy within 28 days of SABR onset. For 110 of the 121 patients, at least one GI OAR structure was available for evaluation. Significant associations were found between grade ≥ 2 GI toxicity and GI OAR D 0.1cc , D 1cc , and D 20cc . Our multivariable model included PS and D 0.1cc , and the association between grade ≥ 2 GI toxicity and GI OAR D 0.1cc remained significant. The D 1cc and D 20cc yielded approximately the same quality fits. The predicted risk of GI toxicity as a function of dose and PS is presented in Figure 1. The area under the 12- month receiver operating characteristic curve for the prediction of grade ≥ 2 toxicity was 78.5% (67.8% – 89.2%). Results:

Table 1 Specifications of TRAEs by CTCAE category