Abstract Book

S891

ESTRO 37

survival (OS) is usually still measured in months. To offer an optimal palliative treatment to these patients, especially towards end of life, a reliable estimation of OS is indispensable. Therefore, the aim of this study was to find new clinical factors related to OS. Material and Methods Prospectively, we collected 17 clinical factors in 168 patients with lung cancer who were referred to palliative radiotherapy between 09/2015 and 12/2016. We aimed to assess possibly relevant predictive factors on survival, including patient and tumour characteristics, social factors and medication (like morphine equivalent dose). The Kaplan-Meier method and Cox`s proportional hazards model were used. Results Median overall survival was 2.95 months (95%CI, 2.37- 3.54). Median age was 66 years (range 32-91 years), 57% of patients were male. Most patients were diagnosed with adenocarcinoma (44%). Reason for referral were mainly bone metastases (49.4%), followed by brain metastases (33.3%). Out of 17 factors, 4 factors showed a significant relation with OS in univariate analysis: 1) Karnofsky performance score (KPS) <70 (HR: 2.13 [95%CI, 1.48- 3.08], p<.001), 2) age ≤66 (HR: 1.45 [95%CI, 1.02-2.05], p=0.037), 3) total number of metastases ≥4 at time point of treatment (HR: 3.92 [95%CI, 1.23-12.45], p=0.020), and 4) morphine equivalent dose >100mg/day (HR: 1.78 [95%CI, 1.15-2.75], p=0.010). Finally, in the multivariable analyses 3 of these factors remained statistically significant in relation to OS: 1) KPS (HR: 2.04 [95%CI, 1.37-3.03], p<.001), 2) age (HR: 1.65 [95%CI, 1.15-2.38], p=0.007), and 3) total number of metastases at time point of treatment (HR: 5.72 [95%CI, 1.39-23.55], p=0.016). Median OS of patients with a KPS <70 was only 1.8 months (range 1.13-2.47 months). Conclusion The present study has identified poor performance status (KPS <70), age ≤66 and ≥4 metastases at time point of treatment as strong negative prediction factors on OS. All 3 factors are easily applicable in the clinic. Among these factors KPS shows the highest impact and should be definitely be taken into account to identify patients who are most at risk of adverse outcomes after treatment. EP-1657 A Study of re-radiation rate after 8 gy single fraction radiotherapy to symptomatic bone metastases S. Juneja 1 , K.L. Bradley 1 1 Queen Alexandra Hospital, Hematology and Oncology, Portsmouth, United Kingdom Purpose or Objective Metastases to bones are frequently observed in advanced malignancies, leading to debilitating symptoms, affecting the quality of life adversely. Not uncommonly, metastatic vertebral deposits pose a threat to the spinal cord, and, if unaddressed, can lead to permanent neurological impairment. There are recent conflicting reports questioning the efficacy and longevity of response to 8 Gy / single fraction radiotherapy to bone metastases. Re- radiation rates have been reported to be as high as 25% in some studies. This real life study evaluates clinically observed re-radiation rates following this regimen in our institution, to find a cause for discrepancy of results, if any, and to establish an updated protocol for management of bone metastases, keeping patient and tumour factors into account. Material and Methods This is a single arm, retrospective study. Records of 100 patients and 154 sites of symptomatic bone metastases,

treated in 2016 with 8 Gy/1 fraction of radiotherapy were studied. Re-radiation rate, median time to re-radiation, symptomatic improvement, tumour, patient and physician factors were examined. Kaplan Meier survival curves were plotted, median overall survival and survival from first treatment with 8 Gy/1 fraction was calculated for the cohort. Regression analysis and appropriate tests of significance are being applied to establish factors governing re-radiation. Available literature is being reviewed, with a view to propose an algorithm for management of bone metastases, with appropriate radiotherapy fractionation schedule. Results Median age for patient at the time of the first decision to treat a site with 8 Gy single fraction was 67 years. 71% of patients had reached the survival end point at the time of data collection. Median overall survival for the cohort was 786 days (26.2 months). Median site-specific survival since first radiation was 214.5 days (7.2 months). Of the irradiated sites, 33.8% were metastases from a breast primary. Most common histology was adenocarcinoma (68.2%). Data indicating response to treatment was available for 127 sites, of which 82.7% reported to have achieved good subjective symptomatic control. 28 of 154 (18.2%) sites were re-radiated, either due to poor primary response, or symptom relapse. No adverse events were reported after re-irradiation. Conclusion 8 Gy single fraction radiotherapy to bone metastases is a cost-effective treatment, which leads to good results in terms of symptom control and quality of life. EP-1658 A scoring system to predict bone metastasis after radical resection of colorectal cancer L. Käsmann 1 , A. Li 2 , D. Rades 1 , C. Fu 2 1 University of Lübeck, Department of Radiation Oncology, Lubeck, Germany 2 Changhai Hospital- The Second Military Medical University, Department of Colorectal Surgery, Shanghai, China Purpose or Objective Colorectal cancer is the third most common type of cancer and the fourth leading cause of cancer-related death in the world. The development of bone metastasis is a severe burden for patients´ prognosis and quality of life. The aim of this study is to develop a scoring system to predict bone metastasis after radical resection within 5 years to identify patients at risk for continuous monitoring and optimize surveillance. Material and Methods We evaluated the patient records of 1,749 patients treated during 2001-2010 at one cancer center. 50 patients developed bone metastasis within 5 years after treatment. Univariate and multivariate analysis were performed to find the risk factors (age, gender, localization, histology, preoperative carbohydrate antigen 199 (CA-199) level, T-stage, lymph node metastasis and pulmonary metastasis) using Chi square test and logistic regression. Results Univariate analysis showed tumor site (P=0.026), preoperative carbohydrate antigen 199 (CA199) level (P=0.039), lymph node metastasis (P<0.001) and pulmonary metastasis(P<0.001) were significantly associated with the development of bone metastasis. On multivariate analysis three factors, namely rectal cancer (p=0.038), lymph node metastasis (p=0.006) and pulmonary metastasis (p<0.001) were identified as