ESTRO 2023 - Abstract Book

S120

Saturday 13 May

ESTRO 2023

Conclusion SABR patients can receive high doses to cardiac substructures. Further multivariate analysis is being performed to assess impact on survival. PD-0159 Association of Statin Intensity With Cardiac Events and Mortality After Radiotherapy for NSCLC G. Walls 1 , J. O'Connor 1 , E. McCarron 2 , V. Giacometti 3 , F. Duane 4 , M. Harbinson 5 , C. McCann 5 , P. McKavanagh 5 , A. Gavan 6 , J. McAleese 3 , A. Cole 3 , K. Butterworth 1 , C. McGarry 3 , G. Hanna 1 , S. Jain 1 1 Queen's University Belfast, Patrick G Johnston Centre for Cancer Research, Belfast, United Kingdom; 2 Belfast Health & Social Care Trust, Department of Clinical Biochemistry, Belfast, United Kingdom; 3 Belfast Health & Social Care Trust, Northern Ireland Cancer Centre, Belfast, United Kingdom; 4 St Luke’s Radiation Oncology Network, Department of Radiation Oncology, Dublin, Ireland; 5 Belfast Health & Social Care Trust, Department of Cardiology, Belfast, United Kingdom; 6 Queen's University Belfast, Northern Ireland Cancer Registry, Belfast, United Kingdom Purpose or Objective Mechanisms by which cardiac radiation dose leads to death are purported to include acute cardiac events and subclinical impairment of cardiac function that impedes recovery from other acute medical issues. Whilst historically prescribed to lower serum lipids, cardiology guidelines now recommend statins for a myriad of cardiovascular risk factors (CVRFs), owing to their anti-inflammatory, antifibrotic and antioxidant properties. Herein the association of statin intensity with survival in patients completing NSCLC radiotherapy (RT) is examined. Materials and Methods Patients treated with curative-intent RT between 2015-2020 at a regional centre were identified. Clinical notes were interrogated for baseline patient, tumour and CVRF details and both follow-up cancer control and CV events. The latter were verified by a cardiologist. Death certificates were accessed for the primary cause of death. RT planning scans were retrieved for application of a deep learning-based auto-segmentation tool to generate cardiac chamber and great vessel structures. The coronary arteries and conduction nodes were added manually and the former were verified by the atlas lead author. A composite cardiac base structure was created. All plans were re-calculated with the same AAA algorithm. Three Cox proportional hazards regression models for overall survival were generated: unadjusted, adjusted for tumour factors and CVRFs only, and a model fully adjusted with all of the available parameters including previous cardiac events. Results A total of 478 patients were included, with a median age of 70, Charlson Comorbidity Index of 5 and BMI of 26.5. The median heart base dose was 9.6Gy (IQR 5.5–15.1Gy). The median lung V20 was 20.0% (IQR 14.8–27.1). CVRFs were common,