QOLP-26. PATIENT REPORTED OUTCOMES MEASUREMENT INFORMATION SYSTEM (PROMIS) SCREENING FOR ANXIETY & DEPRESSION IN CENTRAL NERVOUS SYSTEM(CNS) CANCER: LARGE COHORT REPORT FROM THE NEURO-ONCOLOGY BRANCH NATURAL HISTORY STUDY (NOB-NHS)

Abstract BACKGROUND Emotional distress, including anxiety and depression, is associated with significant morbidity. Studies in CNS tumor patients have been limited by small non-diverse samples. The NOB-NHS includes screening for depression and anxiety patients undergoing longitudinal care in the NOB. In a previously reported pilot study, anxiety(11%) and depression(15%), and association with performance status and use of psychotropic medications were reported. This report attempts to validate these findings in a larger more diverse cohort. METHODS All patients(n= 269) enrolled in the NOB-NHS were screened for depression and anxiety using PROMIS measures. Descriptive statistics and standardized classification of moderate-severe depression and anxiety are used to describe the sample characteristics. Emotional Distress was defined as PROMIS T scores-> 60 on depression, anxiety, or both scales. Chi-square and Fishers exact tests were used to identify associations. Significance level was set at p< 0.05. RESULTS The sample was primarily white(82%), males(61%), 54 months(0–398) from diagnosis, with a median age of 49(21–81), good performance status(KPS 90, 75%) and a high-grade neoplasm(63%) with glioblastoma(30%) the most common diagnosis. Significant emotional distress was present in 19% of the sample. Poor performance status(28%) and use of psychotropic medications(32%) were both significantly associated with emotional distress(X2(1)= 4.8, p< 0.03 and X2(1)= 7.4,p< 0.01), respectively. Past recurrence(25%), progression/pseudo-progression on MRI(26%), female gender(23%), corticosteroid use(28%) and non-glioblastoma(21%) had higher incidence of distress, although not significant. Tumor location, age and disease status was not associated with emotional distress in this sample. CONCLUSION This is one of the largest reported cohorts using screening measures for emotional distress in this population and confirms earlier findings of association with functional status and psychotropic medications. Future studies will investigate longitudinal trends, diagnostic referrals, and emotional distress phenotypes, including genomic predisposition, to improve individual patient/population care.