ESTRO 2020 Abstract Book

S288 ESTRO 2020

additional RT. We collected baseline characteristics, PD1i and RT details, as well as toxicities. Patient characteristics and adverse events were compared by Kruskal-Wallis-test for continuous variables and Pearson's chi-square test for categorical variables. Results 201 patients with different types of malignancies (62.5% malignant melanoma; 18.5% non-small cell lung cancer; 7% renal cell carcinoma; 19% others) could be analyzed. We obtained a median follow up period of 22.3 months until 12/2017. Patients received at least 3 cycles of either pembrolizumab (n=112) or nivolumab (n=89). 153 patients received additional RT. Regarding the incidence of toxicities the most frequent adverse events in the RIT group were fatigue (47.2%) and skin toxicities (45.5%) followed by abdominal (33.3%) and lung (21.8%) problems. In the IT group, most frequent side effects were lung toxicities (19.6%) followed by skin toxicities (18.4%), fatigue (17.6%) and abdominal problems (16.3%). In total, more toxicities were reported when adding RT (8.5% vs 19.4%). The expected IT-related adverse events like lung, thyroid and liver toxicities did not differ statistically significantly in between both treatment groups. At least 1 maximum-grade 3 or 4 adverse event of any cause occurred in 11.5 % of the patients in the IT group and in 6.9 % of those in the RIT group, including 1 grade 4 event (2.1%) in the IT group and 6 grade 4 events (3.9%) in the Patients being treated with radiotherapy and PD-1 inhibitors tend to have more side effects when compared to those receiving PD-1 inhibitors alone. Regarding immunotherapy-related adverse events we observed no statistically significant higher numbers when patients are also treated with RT. The reported higher numbers are most likely due to the side effects of the RT itself. With a rare incidence of grade 3 or 4 toxicities the combination treatment of RIT can be applied safely in clinical routine. PH-0526 Socioeconomic status does not affect survival in patients with brain metastases S. Nagtegaal 1 , A. Claes 1 , S.G. Elias 2 , T.J. Snijders 3 , H.M. Verkooijen 4 , J.J.C. Verhoeff 1 1 umc Utrecht, Radiation Oncology, Utrecht, The Netherlands ; 2 umc Utrecht, Julius Center For Health Sciences And Primary Care, Utrecht, The Netherlands ; 3 umc Utrecht, Neurology, Utrecht, The Netherlands ; 4 umc Utrecht, Imaging Devision, Utrecht, The Netherlands Purpose or Objective In most major cancer sites a low socioeconomic status (SES) is consistently associated with poorer survival. However, the generalizability of these results may be limited due to differences in local healthcare systems. Ideally, a patient’s SES should not affect survival. Therefore, we studied the effect of SES on survival in a cohort of brain metastasis patients. Material and Methods We used a consecutive retrospective cohort of 401 patients treated with stereotactic radiosurgery for brain metastases (BM) in our institute between 2012 and 2017. Patients had one of the five most common primary tumour types: non-small cell lung carcinoma (NSCLC), breast cancer, renal cell carcinoma (RCC), melanoma and gastro- intestinal (GI) tumours. Baseline tumour-specific prognostic factors, taken from the Graded Prognostic Assessment (GPA), were collected. These were age, RIT group . Conclusion

Karnofsky performance status, number of BMs, presence of extracranial metastases and tumour genotype. The SES is represented by a score, based on education level, income and employment, and is determined periodically by the Netherlands Institute for Social Research. The scores are published as averages per postal code. The individual SES scores used in analysis were those most current at baseline. A log-rank test was used to compare SES quartiles. Cox proportional hazard models were made for each tumour group, as well as for the cohort as a whole. Included were the SES score, together with the prognostic factors from the GPA. Results In total, 401 patients were included. Median follow-up was 9.6 months, and 345 deaths were observed. The SES score in this cohort ranged from -2.69 to 2.62, with a mean and standard deviation comparable to that of the general population (0.41 and 0.91, respectively). Baseline characteristics per SES quartile are shown in Table 1. Individual adjusted hazard ratios for SES per primary tumour are shown in Figure 1, along with the HR for the entire cohort. The HR for SES in all tumour types is 1.00 (95% CI 0.89-1.13). Similar HRs are seen in NSCLC and GI, with broader confidence intervals for the smallest subgroups, with the one in the RCC group not including 0. However, in the full model an interaction between SES and tumour type was not significant, suggesting the prognostic effect of SES is not dependent on the primary tumour. Unadjusted Kaplan-Meier survival curves are shown in Figure 2, showing largely comparable survival across the SES quartiles during the first months of follow-up. After 20 months the curves start to deviate. However, the log-rank test found no significant difference.