ESTRO 2020 Abstract book

S242 ESTRO 2020

different chemotherapy protocols in patients treated in the phase III PET Plan trial (ARO 2009-09; NCT00697333). Material and Methods The prospective randomized controlled phase-III PET-Plan trial was conducted in 24 centers. Patients with inoperable locally advanced NSCLC suitable for radio-chemotherapy were randomized at a 1:1 ratio concerning the target volume (TV) definition and received quality-assured isotoxically dose-escalated RT (60–74 Gy, 2 Gy per fraction) was applied to the respective TVs. Concurrent platinum-based chemotherapy was mostly conducted according to Vokes et al.(Vokes, II et al. 2002), using cisplatin 80mg/m² (day 1 and 22) and vinorelbin 15mg/m² (day1+8 and 22+29) (protocol 1, P1) or cisplatin 20mg/m² (day1-5 and 29-33) and vinorelbin 12.5 mg/m² (day 1,8,15 and 29,36,43) (protocol 1, P2) or carboplatin AUC1 (day1- 5 and 29-33) and vinorelbin 12.5 mg/m² (day 1,8,15 and 29,36,43) (protocol 1, P3) according to Semrau et al.(Semrau, Klautke et al. 2008), at the discretion of the treating physician. Results Between 05/2009 and 11/2016, 205 patients were randomized and 172 patients in the per-protocol analysis (A: n=84, B: n=88). 30 patients were treated according to P1, 92 according to P2 and 28 according to P3. One patient did not receive any treatment in Arm A (1%) and 21 patients received different treatments. In the PP analysis of 79% in arm A and 82% in arm B received the 1st cycle per protocol (pp) and 73% (Arm A) and 72% (Arm B) received the 2nd cycle pp but there was no statistical significant difference in regards of overall survival (OS) and progression free survival (PFS). Causes of deviation were mostly renal or hematologic toxicity. All 3 protocols were well tolerated. Patients in P1 (17%) und P2 (15%) developed more grade 3 leucopenia than P3 (4%) but less grade 3 renal toxicities 0% (P1,2) vs 4%(P3). Only patients in P1 developed grade 4 hematologic toxicities (7% leuco- , 3% thrombocytopenia and 3% anemia ).Patents treated with P1 or P2 had a better survival as opposed to P3 (p=0.008 and p=0.009 respectively) but there was no difference between P1 and P2 (p=0.6). Furthermore there was no difference observed concerning PFS between the different protocols. Patients treated with cisplatin (n=122) had a better OS as opposed to carboplatin (n=28) Vinorelbin in combination with cisplatin either as a single dose of 80mg/m² per cycle or daily doses of 20mg/m² (5 days per cycle) for concurrent chemoradiation show similar efficacy in this patient cohort. Patients treated with carboplatin had a significant worse survival. This may be due to the selection of the drug due to restriction in normal tissue function, which may also cause an impaired prognosis. PD-0414 Correlating dosimetry with the outcomes - A analysis of UK’s largest cohort of lung SABR patients. A. Saha 1 , M. Beasley 1 , N. Hatton 1 , K. Franks 1 , P. Jain 1 , M. Teo 1 , K. Clarke 1 , P. Dickinson 1 , P. Murray 1 , J. Lilley 2 1 St. James Institute of Oncology- Leeds Teaching Hospitals- NHS trust., Clinical Oncology, Leeds, United Kingdom ; 2 St. James Institute of Oncology- Leeds Teaching Hospitals- NHS trust., Medical Physics, Leeds, United Kingdom Purpose or Objective SABR is an established treatment option for early stage medically inoperable peripheral lung cancer. A recent study have shown that tight conformity of radiotherapy (p=0,003). Conclusion

plans might worsen loco-regional control and can predict distant metastases. The aim of the study is to report local control (LC), progression free survival (PFS), overall survival (OS) and dosimetry of early stage peripheral lung cancer patients treated with SABR and to try to explore any dosimetric predictors of outcome. Material and Methods 1266 patients treated in our institute (May 2009 - August 2018) were included. Electronic medical records were reviewed for baseline characteristics, treatment details and outcomes. Dosimetric data was extracted from XIO and Monaco software. Patients were treated according to the UK SABR consortium guidelines. Kaplan Meier analysis with log rank test was used for survival analysis. Univariate and multivariate cox regression model was used for exploring the correlation between the dosimetric variables and outcomes, using IBM SPSS statistics version 21 software. Results Median age was 75 years with 47.4% male. Median follow up was 56 months. Median OS was 36 months with 1, 2 and 5 years OS rates of 84.2%, 64.5%,31.5%, respectively. Median for LC and PFS wasn’t reached. 1,2 and 5 years LC rates were 98.2%,95.1%,92.5%, respectively.1,2 and 5 years PFS rates were 87.4%, 78.4%,72.5%, respectively. PTV volume, dose to 99% volume of PTV (D99) and R50 were significantly associated with OS in univariate analysis. On multivariate analysis PTV volume and minimum dose to PTV (DPTV min) were significantly associated with OS (Table 1). D99 of >53 Gy and R100 of >1.1 was significant associated with better OS(Table 2).Nothing was associated with LC on univariate analysis but on multivariate analysis, dose fractionation were found to be associated with LC. R100 of >1.1 was associated with better local control. PTV volume, mean lung dose (MLD) , V20 , V12.5 and dose fractionation were significantly associated with PFS on univariate analysis. On multivariate analysis PTV volume and dose fractionation were significantly associated with PFS. 60Gy in 8 fractions regimen was better compared to 50Gy in 8 fractions regimen in terms of PFS but not for OS or local control. PTV volume ≤30 cc, R100 >1.1, R50 >6 and MLD ≤4Gy was associated with better PFS but on multivariate analysis only R100 was statistically significant.