ESTRO 2021 Abstract Book

S1232

ESTRO 2021

Psychiatric disorders are common in the general population (10%). The overall cancer incidence for patients with a psychiatric disorder (PwPD) compared to the general population appears the same, but the outcome might be different. It is thought that cancer related outcomes could be worse because of the effect of psychiatric disorders on cancer diagnosis (patient and/or doctors delay) and cancer treatment (treatment choice, -shared decision making, -interruptions or -refusal). In this study, we evaluated if cancer patients with schizophrenia spectrum disorder (SSD), bipolar disorder (BD) or borderline personality disorder (BPD) receive comparable radiotherapy (RT) compared with a control population of patients with the same cancer type and staging but without a psychiatric disorder in their medical history (PwoPD) and if there is a difference in overall survival after RT between these two groups. Materials and Methods Patients with SSD, BD, and BPD were included through a text-based search of the electronic health records of all patients that received radiotherapy between 2015-2019 in the University medical Center Utrecht (UMCU). Each PwPD received a matched PwoPD from the database. This match was based on tumor location, type, staging (after neo-adjuvant treatment), fitness, age, and gender. Outcomes were (i) the amount of RT fractions received, (ii) RT dose in equivalent dose in 2 Gray fractions (EQD2) and (iii) death. Kaplan Meier survival curves were constructed of the PwPD- and control cohort. Uni- and multivariable hazard ratios (HR) for death during follow-up were calculated for all baseline characteristics in relation to death. Results 89 PwPD were included, of them 44 patients with SSD, 35 with BD, and 10 with BPD. Eighty-nine pairwise matched PwoPD were included for the control cohort. We found no statistically significant differences between PwPD and PwoPD for the average amount of RT fractions received, average dose in EQD2, or in the number of PwPD that received a higher or lower RT dose compared to their match. Multivariable HR for death during follow up were not significant for patients with SSD or BD. BPD had a multivariable HR of 2.88 (1.00- 8.32, p 0.05).