Correlations Between Depression According Dsm-Iv-Tr (Dsm) Criteria, Three Validated Scales, Oncologist Assessment, And Clinical Psychiatric Interview In Elderly Advanced Ovarian Cancer (Aoc) Patients (Pts): A Gineco Study

9080 Background: Depression is a major outcome in cancer pts. Clinicians typically rely on their clinical impression of depression rather than pts self-reports. Our aim was to explore the association between patient-reported depression, oncologist assessment (OA) and a clinical psychiatric interview (CPI) in elderly AOC pts. Methods: This analysis was a secondary endpoint of theElderly Women AOC trial, designed to assess the impact of geriatric covariates, notably depression, on survival in pts over 70 receiving 6 courses of carboplatin. Depression was assessed using the Geriatric Depression Scale-30 (GDS; cut off score of 10/30), the Hospital Anxiety Depression Scale (HADS; cut off score of 15/42), the distress thermometer (DT; cut off score of 4/10) and OA (yes/no). CPI was conducted by psychologists within 10 days after inclusion. The interview guide for CPI (yes/no) was constructed and adapted from three validated scale: GDS, Hamilton Depression Rating Scale Hamilton (HDRS), Montgomery Asberg Depression Rating Scale (MADRS) and the DSM criteria. DSM was considered as the gold standard. Results: Out of 111 pts, 100 (90.1 %) completed all the assessment (OA, GDS, HADS, DT, CPI). Patients characteristics were: mean age 78, performance status ≥2: 48 (55%). Thirty six pts (36%) were identified as depressed by the CPI versus 17 (17%) by OA, 32 (32%) by the GDS, 36 (36%) by the HADS, 58 (58%) by the DT and 16 (16%) according to DSM. We found a significant correlation between DSM and GDS (r=0.58; p<0.001), DSM and CPI (r=0.53; p<0.001). We did not find any significant correlation between DSM and OA (r=-0.05;p=0.733), DSM and DT (r=-0.07;p=0.583), and between DSM and HADS (r=0.127;p=0.258). Identification according to OA (yes/no) resulted in 87% false negatives and 18% false positives rates. The best sensitivity and specificity as a screening tool was found for GDS, 94% and 80% respectively. Conclusions: The use of validated tools such as GDS and a collaboration between psychologists and oncologists are warranted to better identify emotional disorders in elderly women with AOC.