ESTRO 38 Abstract book

S752 ESTRO 38

(Median Area, MA), and we also divided the MA for the median intensity in Hounsfield Units (MA/I). We analyzed overall survival (OS) with these parameters (using the median value as cut-off) and the known prognosticators (age, ECOG, stage, previous surgery, number of metastases), with Kaplan Meier method (univariate) and Cox Regression Analysis (multivariate). Results We included 138 patients (97 males and 41 females), with a median age of 67 years (mean 64 years, range 30-84 years). At univariate analysis of OS, the significant parameters were the ECOG (p<0,001), the MA (p:0,001), MA/I (p:0,005), previous surgery (p:0,001), number of metastases (p<0,001). At multivariate analysis, only ECOG (p:0,004), MA (p:0,005), number of metastases (p:0,001) resulted significant. Conclusion MA and MA/I in NSCLC patients could represent an independent prognosticator of survival, and could help to stratify the patient’s prognosis. EP-1378 Brain imaging pre/post treatment in cN2 non- small cell lung cancer treated with chemo-radiation C.W.E. Chan 1 , H.W.G. Gan 1 , P. Lyons 1 , P. Mcloone 1 , N. Mohammed 1 1 Beatson West of Scotland Cancer Centre- Glasgow- UK, Clinical Oncology, Glasgow, United Kingdom Purpose or Objective The incidence of brain metastases(BM) ranges from 10-47% and increases with more advanced non-small-cell lung cancer(NSCLC). Stage 3 NSCLC has 20% of patients diagnosed with BMs, and hence British Thoracic Society(BTS) guidelines state that brain imaging should be performed as part of pre-treatment staging. Asymptomatic BM identified on imaging may alter treatment. Therefore, the aims of this study are to assess pre-treatment imaging in radically treated cN2 NSCLC patients, the incidence of BM after definitive treatment and outcomes. Material and Methods Patients with stage cN2 NSCLC diagnosed between January 2013 to December 2016 and treated with chemo-radiation therapy(CRT) in the West of Scotland were analysed retrospectively using electronic patient records. Information of pretreatment investigations, treatment given, subsequent follow-up and outcomes were collated and analysed. Results 107 patients with stage cN2 NSCLC treated with CRT were identified and analysed. The median age was 65 years(26- 84 years), 56(52%) female and 51(48%) male. ECOG performance status was recorded as 0, 1 and 2 in 55, 47 and 3 patients. TNM staging was T1(11%), T2(25%), T3(27%), T4(34%), Tx(2%) and all patients were cN2, M0(96%) and M1(4%). Pathology was squamous cell in 50%, adenocarcinoma 42% and other in 8%. 7 patients also demonstrated ALK or EGFR variants. Radiotherapy dose was 55Gy in 20 fractions over 4 weeks using volumetric modulated arc radiotherapy (VMAT) or 3DConformal radiotherapy. Nine patients received neoadjuvant chemotherapy and all patients had concurrent platinum doublet. All patients underwent a positron emission tomography(PET) scan, but only 45 patients(41%) had a pre-treatment brain scan, of which 27 were CT scans and 18 were MRI scans. The proportion of brain scans performed increased over time:

Medically inoperable patients due to age or other comorbidities with biopsy proven peripheral Stage I NSCLC and a minimum follow up of 7 years were evaluated. PET/CT imaging was completed for staging. Three to five gold fiducial markers were implanted in or near tumors under CT guidance or via electromagnetic navigational bronchoscopy to serve as targeting references. Gross tumor volumes were contoured using lung windows. The margins were expanded by 5 mm to construct the planning treatment volume (PTV). Doses delivered to the PTV ranged from 45-60 Gy in 3 or 5 fractions. Treatments were delivered daily utilizing the CyberKnife system with fiducial tracking. Pulmonary function testing was completed a median 49 days prior to the start of SBRT and 1 year after completion. Clinical examination and PET/CT imaging were completed at 6 month follow up intervals. Results From August 2005 to August 2011, forty patients (median age 76 years) with a median maximum tumor diameter of 2.7 cm (range, 1.4-5.0 cm) were treated and followed for a minimum of 7 years or until death. A median dose of 50 Gy was delivered over a median 7 days (range, 3-13 days). The median follow up for all and alive patients was 4.6 and 8.9 years, respectively. There were 2 primary tumor failures, 1 involved lobe failure, and 1 nodal failure. The 5 year Kaplan-Meier primary tumor control, LC, LRC, distant control, and DFS estimates were 95%, 91%, 88%, 82%, and 70%, respectively. Overall survival (median 4.5 years) and cancer specific survival estimates at 5 and 8 years were 43% and 23%, and 79% and 62%, respectively. Nine patients, all whom had fiducials placed under CT guidance, developed a pneumothorax requiring chest tube placement: no pneumothoraces were observed following bronchoscopic fiducial placement. The 5 year cumulative incidence of radiation induced rib fracture (RIRF) was 53%, with a median time to fracture of 39 months. There was no difference in percent predicted FEV1 one year following SBRT; there was, however, a modest decline of 7.4% (p=0.02) in the percent predicted DLCO at one year. Conclusion Our institution’s medically inoperable Stage I NSCLC outcomes compare favorably with conventional SBRT outcomes from RTOG 0618. Bronchoscopic fiducial placement minimizes the risk of pneumothoraces requiring chest tube placement. Although the RIRF rate was high, the majority of patients were asymptomatic, diagnosed via surveillance PET/CT imaging, and required only conservative management. EP-1377 Role of psoas volume in locally advanced or metastatic NSCLC patients undergoing radiation therapy V. Nardone 1 , S. Falivene 1 , Giugliano, Francesca Maria 1 , F. Scala 2 , Tini, Paolo 3 , Pirtoli, Luigi 3 , C. Guida 1 , S. Cappabianca 2 1 Ospedale del Mare, Radiation Therapy Unit, Napoli, Italy ; 2 University of Campania “Luigi Vanvitelli”, Department of Precision Medicine, Naples, Italy ; 3 Unit of Radiation Therapy, University Hospital of Siena, Siena, Italy Purpose or Objective To evaluate the volume and median intensity of psoas muscle on CT imaging at the time of diagnosis as a prognosticator in patients with locally advanced or metastastic NSCLC undergoing chemotherapy or We included patients with NSCLC undergoing palliative radiation therapy at our Department betweeen january 2010 and december 2017. We contoured on CT made at the diagnosis the volume of right and left psoas volume from the cranial border of L4 till the caudal border of L5. We calculated the median volume between the sides and we divided the median volume for the height of the muscle radio/chemotherapy. Material and Methods