ESTRO 2022 - Abstract Book

S1084

Abstract book

ESTRO 2022

tended to deteriorate OS but difference was not significant (p=0.157), while that in the late phase had significant influence on OS (p<0.001).

Conclusion CIRT can be safe and effective for HCC even with poor hepatic function.

PO-1286 Skeletal muscle measured at T12 is a prognostic biomarker in oesophageal cancer patients

D. McSweeney 1 , G. Radhakrishna 2 , A. Green 1 , P.A. Bromiley 3 , M. van Herk 1 , A. McWilliam 1

1 University of Manchester, Division of Cancer Sciences, Manchester, United Kingdom; 2 The Christie NHS Foundation Trust, Department of Medical Oncology, Manchester, United Kingdom; 3 University of Manchester, Division of Informatics, Imaging and Data Sciences, Manchester, United Kingdom Purpose or Objective Sarcopenia is emerging as a prognostic factor for multiple patient groups treated with radiotherapy (RT) where it is associated with increased toxicity and decreased overall survival. Sarcopenia is typically assessed via the skeletal muscle index (SMI): skeletal muscle area at L3 normalised by patient height. Therefore, cohorts where routine imaging does not include L3 are often neglected. Patients with oesophageal cancer are known to experience malnutrition and weight loss, associated with poorer outcomes. In this work, we explore the utility of SMI, measured at T12, as a prognostic factor in patients with oesophageal cancer treated with concurrent chemoradiotherapy (CCRT). We then compare sarcopenia with other measures of patient frailty: performance status (PS) and body mass index (BMI). Materials and Methods 103 patients with oesophageal cancer treated with CCRT, 50Gy in 25 fractions, were retrospectively collected. Patient characteristics are shown in Table 1 . T12 was manually identified on RT planning scans. An in-house artificial intelligence algorithm was then used to segment the skeletal muscle compartment at T12 and segmentations were visually assessed for accuracy. Muscle area was extracted and SMI at T12 calculated for all patients with successful delineations. Prognostic value was investigated using Kaplan-Meier curves (split on sex-specific median SMI) and a multivariable Cox model controlling for biological sex, age, tumour volume, PS and BMI. The primary endpoint was overall survival. Results After removing segmentation failures, 98 patients were available. Kaplan-Meier curves did not show significant differences in overall survival when split on sex-specific median SMI ( Figure 1 ; log-rank p=0.074 ). However, in multivariable analysis, SMI was significantly associated with survival (HR=0.73, p=0.044 ), where a higher SMI seems to be protective ( Table 2 ). The results suggest that there is an interaction between SMI and other factors in the multivariable model, showing that SMI provides additional prognostic information beyond PS and BMI.

Conclusion We show that SMI, evaluated at T12 using routine planning scans, is a prognostic factor in patients with oesophageal cancer treated with CCRT, with increased muscle mass being protective. Results from our multivariable Cox model show that SMI provides additional information beyond PS and BMI. Although our results require further validation, SMI shows promise for patient treatment stratification.