ESTRO 38 Abstract book

S880 ESTRO 38

University of Munich TUM - Institute of Innovative Radiotherapy iRT - Department of Radiation Sciences DRS - Helmholtz Zentrum München - Deutsches Konsortium für Translationale Krebsforschung DKTK - Partner Site Mun, , ; 3 Klinikum rechts der Isar - Technical University of Munich TUM - Deutsches Konsortium für Translationale Krebsforschung DKTK - Partner Site Munich, Department of Radiation Oncology, Munich, Germany Purpose or Objective Palliative radiation therapy of painful spinal bone metastases (SBM) triggers a pain response in only two- thirds of treated patients. A reliable prediction model of pain may help to personalize the decision for radiation therapy. In this retrospective study, we investigated the relationship between several pretherapeutic data and pain response in SBM patients. Material and Methods A cohort of 190 patients with SBM treated with radiation was analyzed. Patients with non-solid tumors and previous invasive procedures were excluded. Pretherapeutic information about basic patient characteristics, staging information, treatment modalities, radiation-specific regiments, Karnofsky Performance Index (KI) and Spinal Instability Neoplastic Score (SINS) was collected. SINS was determined using radiotherapy planning CT scans. Pain response was categorized as complete pain response, partial pain response, indeterminate pain response or progression of pain according to previous studies. For statistical analysis, binary variables were created: complete response and any response (complete or partial pain response). Patients with indeterminate pain response were excluded. Univariate and multivariate binary logistic regression (IBM SPSS Statistics 20) was used to test for associations. Results Univariate logistic regression revealed significant associations between complete response and multiple pretherapeutic parameters: SINS (Odds-Ratio [OR] 0.879; 95% confidence interval [CI] 0.784-0.985; p<0.027), KI (OR 1.032; 95% CI 1.006-1.058; p<0.014), neurological dysfunction (OR 0.459; 95% CI 0.232-0.909; p<0.025), tumor type (prostate or mammary carcinoma [OR 0.411; 95% CI 0.235-0.720; p<0.002] vs lung cancer [no significance (ns)] vs others [reference category; p<0.008]), tumor histology (adenocarcinoma [OR 0.127; 95% CI 0.024-0.672; p<0.015], squamous cell carcinoma [ns] and others [ns]), simultaneous systemic therapy (chemotherapy [ns], hormone therapy [ns], targeted [OR 0.344; 95% CI 0.154-0.767; p<0.009], no therapy [reference category; p<0.002] ). There was no significant association between complete response and age, gender, grading, localization, spinal cord compression, soft tissue involvement, metastasis type (blastic/lytic/mixed), opioid/steroid medication and radiation dose. KI (OR 1.053; 95% CI 1.014-1.094; p<0.008) and neurological dysfunction (OR 0.320; 95% CI 0.103-0.997; p<0.049) were also significantly associated with complete response in multivariate logistic regression. Only age (OR 0.974; 95% CI 0.949-0.999; p<0.043) and steroid medication (OR 3.314, 95% CI 1.254-8.757; p<0.016) were significantly associated with any response in univariate and multivariate logistic regression. Conclusion Our results indicate that spinal stability, higher KI, no simultaneous therapy, no neurological dysfunction, specific tumor histologies, and tumor types may be associated with pain response in patients with spinal bone metastases treated with radiation. EP-1633 Profile of patients who die in the first 30 days after palliative radiotherapy in our center D.C. Moreno Santiago 1 , M. Vázquez Varela 1 , M. Altabas Gonzalez 1 , J. Giralt 1

1 Vall D`Hebron Hospital, Radiation Oncology, Barcelona, Spain Purpose or Objective The role of radiotherapy as a supportive treatment in the palliative care of patients with advanced stages of cancer is well known. However, the decision to carry out these treatments is influenced by the benefit that could be obtained in the last days of life. The aim of this study was to analyze the factors that could influence death within 30 days after a palliative radiotherapy (RT) treatment was initiated. Material and Methods We retrospectively collected data from patients treated with palliative intention radiotherapy in our department between January and June of 2018 in order to identify factors related to mortality in the first 30 days after start of the treatment. Clinical features of gender, age, Eastern Cooperative Oncology Group performance status (PS), Karnofsky index, presence of visceral metastases, the indication of radiotherapy treatment (antalgic, haemostatic, whole brain radiotherapy [WBRT], spinal cord compression or any other) and the prescribed doses. Follow-up was made up to 3 months after the RT was performed, but we decided to set our cut-off point in 30 days. Results A total of 353 palliative treatments were performed in 284 patients. The main characteristic of our population are describe in table 1. The mean age was 64 years (5-90). At the end of the follow up period, 27 patients (9.5%) receiving 120 different treatments during the follow up, died within 30 days after the last treatment was performed, with a median of 16 days (13-22 days) after the start of the treatment. In the 27 deads, single doses were administered in 37% of the cases (10), 9 with antalgic intentions. Typical fractionation schemes as 20Gy in 5 fractions and 30Gy in 10 fractions occurred in 7 patientes (25.9%) in both cases, being 60% for pain control and 60% for WBRT, respectively. A higher mortality rate was observed in those who had worse performance status (ECOG 0= 18% vs ECOG 3= 69% p <0.005; KPS> 70= 24.2% vs KPS< 70= 49.3% p <0.005). Likewise, a difference was observed between the group with visceral metastasis and those that did not have it. Regarding the primary tumour, the lowest mortality was observed in the patients with breast, prostate, gynecologist and myeloma without finding a statistical significance. Greater mortality was found in WBRT and the cases of spinal cord compression (46.8% and 51.8% respectively, p NS) and in those who received single dose RT (44.8% p <0.005). Of these patients, 57.5% had a partial or completed response to treatment (p <0.005).