ESTRO 2020 Abstract Book

S227 ESTRO 2020

Groningen, The Netherlands ; 2 University Medical Center Groningen- University of Groningen, Department of Cardiology, Groningen, The Netherlands ; 3 Radboud University Medical Center, Department of Radiation Oncology, Nijmegen, The Netherlands ; 4 Beatson West of Scotland Cancer Centre, Department of Clinical Oncology, Glasgow, United Kingdom ; 5 University Medical Center Groningen- University of Groningen, Department of Radiology, Groningen, The Netherlands Purpose or Objective Growing evidence indicates that overall survival after radiotherapy for lung and oesophageal cancer is related to radiation dose to heart and lungs, suggesting the occurrence of severe toxicity. However, the exact toxicity responsible for this mortality has not been identified. Animal studies showed that thoracic irradiation can damage the pulmonary microvasculature (1), leading to pulmonary hypertension (PH) within the first months after irradiation. The general aim of the ongoing multicenter CLARIFY study is to quantify the incidence and impact of radiation induced PH in patients treated for lung and oesophageal cancer. Aim of this present analysis of the first 29 evaluable patients included in the study, is to test the impact of radiation dose on right ventricular function 6 weeks after radiotherapy, as a first indication of PH. Material and Methods In the CLARIFY study we will assess cardiac damage in 320 patients using echocardiography and the blood biomarkers, NT-proBNP and Troponin-T, before and at 6, 14, 26 and 52 weeks after radiotherapy (RT). All patients were treated for locally advanced oesophageal (18) or lung cancer (11) with curative-intent (neo-adjuvant) (chemo-) radiotherapy to a total dose of 41.4-60 Gy. Mean heart dose (MHD), mean lung dose (MLD), lung volume receiving 5 Gy (lung V5) and lung volume receiving 20 Gy (lung V20) were extracted from the treatment planning system. To test the relation between endpoints and RT, the association between dose parameters and cardiac function parameters from echocardiograms and blood biomarkers was assessed by Spearman correlation analysis. Results Table 1 summarises the associations between RT parameters, biomarkers and cardiac function parameters. RT led to a dose dependent increase in cardiac wall stress, as evident from the significant associations of NT-proBNP with both lung V5 and MHD. In line with the hypothesis that lung irradiation may cause signs of PH, right-ventricle function reduced dose-dependently; the contraction of the right ventricle (TAPSE) is negatively affected by lung dose (figure 1). Interestingly, the strongest associations of NT- proBNP and TAPSE with dose were observed with the lung V5, indicating a critical role for low doses in the induction of right-ventricle function changes. In addition to changes in right-ventricle parameters, a lung dose dependent decrease in deceleration time over the mitral valve was observed.

(D 2% / 50% / 98% / 50 Gy (V ..Gy

) and volume receiving 2 / 5 / 10 / 20 / 30 / 40 ) was derived of the DVHs, all expressed in an

equivalent dose of 2 Gy per fraction (EQD 2 , using α/β-value of 3 Gy). Of each patient the clinical factors age, gender, Karnofsky performance status (KPS), cardiac comorbidities, forced expiratory volume in 1 second (FEV 1 ), COPD GOLD score, tumor diameter and histology were collected. Cox regression was used for univariate (UVA) and multivariate (MVA) analyses. A multistep approach was used to select the best variables for UVA and MVA ( Figure 1 ). Eighty-four percent of the patients was diagnosed with a peripheral lesion (versus 16% central) and almost 70% had a tumor diameter inferior to 30 mm.

Results The analysis of the heart revealed the V 2Gy as prognostic factor (HR 1.001, p < 0.001). Within the dosimetric analysis of the cardiac substructures, the MVA showed the ascending aorta D mean (HR 1.05, p = 0.008), the inferior vena cava D max (HR 1.01, p = 0.024) and left atrium V 2Gy (HR 1.004, p = 0.026) as being prognostic. The combined MVA of heart V 2Gy with the prognostic clinical factors showed heart V 2Gy (HR 1.001, p = 0.039), KPS < 80 (HR 0.57, p = 0.038) and tumor diameter (HR 1.02, p = 0.012,) as being significant. In the combined analysis of the substructures with the clinical factors, none of the dosimetric parameters of the substructures remained significant. The median OS was 21.3 months (range: 0.5- 72.5 months), and the OS at 2 years, 3 years and 5 years was 76%, 68% and 59%, respectively. Conclusion Overall survival in patients with early stage NSCLC treated with SBRT is influenced by the volume of the heart receiving 2 Gy in combination with the KPS and tumor diameter.

Poster discussion: CL: Upper and Lower GI 1

PD-0419 Thoracic radiotherapy decreases right ventricle function: first results of the CLARIFY study E. Slager 1 , R. Wijsman 1 , J.P. Van Melle 2 , Y.M. Hummel 2 , J. Bussink 3 , N. Mohammed 4 , T.P. Willems 5 , E.S. Hoendermis 2 , J.A. Langendijk 1 , C.T. Muijs 1 , P. Van Luijk 1 1 University Medical Center Groningen- University of Groningen, Department of Radiation Oncology,