ESTRO 2021 Abstract Book

S996

ESTRO 2021

2 University College London NHS Foundation Trust, Department of Clinical Oncology , London, United Kingdom; 3 University College London NHS Foundation Trust, Department of Clinical Oncology, London, United Kingdom; 4 University of Liverpool, Department of Clinical Oncology, Liverpool, United Kingdom Purpose or Objective Chemoradiotherapy or radical radiotherapy is the standard of care for locally advanced NSCLC. The RTOG 0617 trial showed an association between higher heart radiation dose and poorer survival. Radiation related heart disease (RRHD) may manifest earlier in patients with NSCLC than in other tumour types. As ischaemic heart disease (IHD) and NSCLC share risk factors, some patients may be more susceptible to RRHD due to pre- existing heart disease (HD). We aimed to assess the burden of HD and the interaction between the location of the gross tumour volume (GTV), in relation to anatomical landmarks, cardiac dose and the subsequent impact on survival. Materials and Methods A retrospective cohort analysis was performed on patients with NSCLC treated with radical radiotherapy (60- 66 Gy in 2Gy/fraction or 55 Gy in 2.75Gy/fraction) between November 2014 and October 2019 at a single centre. Electronic patient record was used to collect: demographic, comorbidity, staging, outcome information and radiotherapy dosimetry including mean heart dose (MHD), converted to an EQD2 using α/β ratio of 3Gy. Patients were grouped according to whether the GTV (nodal or primary whichever was most caudal) extended below the superior border of T7 level (“unfavourable” location) or not (“favourable” location), as this has been shown to correspond to the superior heart border. Groups were compared using an unpaired t-test. Kaplan Meier survival curves were compared using the log rank test. Results 116 patients were included in our analysis: 62 males and 54 females with median age of 71.0 years (31.0 – 96.0). 67% (78) of these patients had Stage III disease. 54% (63) received a 2Gy/fraction regime while 46% (53) received 2.75Gy/fraction. 27% (31) underwent concurrent and 27% (31) sequential chemotherapy, 47% (54) were treated with radiotherapy alone. 28% (33) of patients had documented HD (ischaemic, structural and arrhythmias) while 57% (66) of patients had risk factors for IHD (smoking, hypertension, hypercholesterolaemia and diabetes). 29% (34) of patients had favourable GTV location and 71% (82) had unfavourable GTV location. The mean MHD (2Gy/fraction equivalent) in the favourable group was 8.2 Gy vs. 16.1 Gy in the unfavourable group (p < 0.0001). In those with Stage III disease, the mean MHD was 9.1 Gy vs 17.0 Gy in favourable and unfavourable groups respectively (p < 0.0001). Figure 1 demonstrates the impact of existing HD or risk factors, and tumour location on survival. 2 year survival probability was 0.89, 0.72, 0.61 and 0.48 in those with favourable/no HD, unfavourable/no HD, favourable/HD and unfavourable/HD respectively (p=0.015).

Fig 1: 2 year Kaplan Meier survival curves by GTV location and HD risk factors

Conclusion Poorer overall survival was seen in patients with known HD or cardiac risk factors with a GTV that extends caudally to T7 (in the same axial plane as the heart). Patients with these identifiable risk factors could potentially derive benefit from cardiac-sparing radiotherapy.

PO-1200 Development and validation of two Australian models to predict 2-year survival in stage I-III NSCLC S. Vinod 1 , N. Lee 2 , J. Shafiq 2 , M. Field 3 , C. Fiddler 4 , S. Varadarajan 5 , S. Gandhidasan 4 , E. Hau 5