ESTRO 2020 Abstract Book

S536 ESTRO 2020

of LC showed HR tumor size 1.04 (95% CI 1.02-1.06, p = 0.001) and HR FEV 1 of 0.97 (95% CI 0.95-1.00, p = 0.031). LC was 88% at 2 years and 81% at 3 years. Twenty-seven patients (12%) developed a local recurrence. Disease progression was reported in 75 patients (34%). DFS was 66% at 2 years and 56% at 3 years. UVA results for DFS are listed in Table 1. MVA of DFS showed HRs for tumor size 1.03 ( p < 0.001), FEV 1 0.98 ( p = 0.004) and localization of the tumor 1.87 ( p = 0.017).

Conclusion CPAP was tolerated well and increased lung volume significantly. LC was independent of tumor type, origin, location, CPAP technique or dose. Patients treated for primary lung cancer had better PFS and OS than patients with metastatic disease Additional follow up studies are warranted. PO-1005 Influence of PTV underdosage on local control in patients with central NSCLC treated with SBRT M. Duijm 1 , N. Van der Voort van Zyp 2 , P. Granton 1 , P. Van de Vaart 2 , M. Mast 2 , E. Oomen - de Hoop 1 , M. Hoogeman 1 , J. Nuyttens 1 1 Erasmus MC Cancer Institute, Radiation Oncology, Rotterdam, The Netherlands ; 2 Haaglanden MC, Department of Radiation Oncology, The Hague, The Netherlands Purpose or Objective Stereotactic Body Radiation Therapy (SBRT) of central lung tumors is characterized by risk-adapted fractionation schedules with accompanying Organs At Risk (OAR) dose- constraints. PTV underdosage is often accepted to respect these dose-constraints and to avoid high rates of toxicity. The purpose of this study was to analyze the effect of PTV underdosage on local tumor control and to determine whether other factors influence local- or disease control. Material and Methods Two hundred and twenty patients with centrally located NSCLC treated with ≤12 fractions of SBRT were included. A central tumor was defined as a tumor <2cm from the esophagus and/or main bronchus or first branches of the bronchial tree. PTV underdosage was allowed in order to meet the OAR constraints. Because of variations in treatment schedules, doses were converted into a biologically equivalent dose using an α/β-value of 10Gy (BED 10 ). The following dosimetric PTV parameters were derived from individual DVHs: maximum and minimum point dose (D max , D min ), mean dose (D mean ) and dose to 2% / 50% / 98% of the PTV (D 2% / D 50% / D 98% ). PTV underdosage was defined as the (percentage of) PTV receiving less than 100Gy BED 10 ; (%)V <100 BED 10 . Potential dosimetric- and patient- prognostic factors for local tumor control (LC) and disease free survival (DFS) were evaluated using univariate (UVA) and multivariate (MVA) Cox Regression analysis. The majority of the patients had stage I (38%) or stage II (52%) NSCLC. The most commonly used fractionation schedules were 5 x 9-12Gy (37%), 8 x 7.5Gy (31%) and 12 x 5Gy (15%). Results The median V <100 BED 10 was 4.2cc in patients with a local recurrence compared to 1.2cc in patients without a local recurrence ( p = 0.593). The median %V <100 BED 10 was the same in patients with and without local recurrence (2%, p = 0.127). When the prescribed dose was <100 Gy BED 10 , patients were more likely to develop a local recurrence (HR 2.24; p = 0.045). None of the PTV parameters derived of the DVHs appeared to be a prognostic factor for LC. The UVA results of the other factors are shown in Table 1 . MVA

Conclusion Underdosage of the PTV is not affecting the local recurrence probability, however a prescribed dose of <100 Gy BED 10 doubles the chance on a local recurrence. Larger tumor size and lower FEV 1 were both independently associated with local recurrence and disease recurrence, aditionally a tumor located in the lower lobe was associated with disease recurrence. PO-1006 Immunotherapy related pneumonitis correlates with radiomics in NSCLC patients treated with Nivolumab I. Mihaylov 1 , G. Lopes 2 , D. Saravia 2 , D. Kwon 3 , R. Yechieli 1 , A. Dal Pra 1 , L. Freedman 1 , T. Diwanji 1 , B. Spieler 1 1 University of Miami, Radiation Oncology, Miami, USA ; 2 University of Miami, Medical Oncology, Miami, USA ; 3 University of Miami, Biostatistics and Bioinformatics, Miami, USA Purpose or Objective Immune checkpoint inhibitors (ICI) offer a new paradigm in cancer treatment. For patients with advanced non- small cell lung cancer (NSCLC), immunotherapy (IT) has led to significant improvements in survival and quality of life. IT also has been associated with debilitating and sometimes life-threatening immune-related adverse events (irAEs) such as pneumonitis, colitis, hypophysitis, inflammatory arthritis and thyroiditis. Methods that predict normal tissue toxicity to IT could inform clinical decision-making and further personalize patient care. Radiomics, the study of diagnostic imaging for patterns of intensity indiscernible to the naked eye, has been used to construct predictive models for tumors. We hypothesized that for patients with advanced NSCLC treated with Nivolumab after frontline chemotherapy, radiomics of pre- IT computed tomography (CT) lung imaging correlate with the incidence of checkpoint inhibitor pneumonitis (CIP). Material and Methods Forty-nine patients were selected for preliminary analyses from an IRB-approved database of 159 patients with advanced NSCLC treated with Nivolumab after frontline chemotherapy. Nine of those patients had confirmed CIP, the primary endpoint. The last pre-IT positron emission