e17550 Background: Skeletal metastases are seen in over 90% of metastatic castrate resistant prostate cancer (mCRPC) patients during the course of their disease and is a major cause for significant morbidity and mortality in them. Radium 223 (Ra-223) is a bone seeking calcium mimetic alpha emitter that has demonstrated significant improvement in survival and reduction in skeletal related events (ALSYMPCA trial). Ra-223 treatment was well tolerated and showed benefit before or after chemotherapy. In this study we aimed to assess efficacy of Ra-223 in mCRPC patients treated in our centre and their survival in relation to some common clinical variables that may assist patient selection in routine clinical practice. Methods: We conducted a retrospective audit of mCRPC patients treated with Ra-223 from June 2015 – December 2019. We analysed treatment outcomes and use of Haemoglobin (Hb), Albumin (Alb) and Alkaline Phosphatase (ALP) as potential prognostic factors. Results: There were 46 patients in total with a median age of 76 years (range; 56-90) and 26% were ≥80 years of age. The median follow-up was 10 months (range; 2-38 months). Median baseline PSA was 103.7 (range; 7.2 - 2534) and median baseline ALP was 188 U/L (range; 51 - 2262). Ra-223 was given as 1st or 2nd line treatment in 65% of our patients. Median number of cycles of Ra-223 administered was 4 and 41% of patients completed 6 cycles of treatment. Biochemical response (drop in PSA > 50%) was only seen in 11% of patients but good response in ALP levels (drop in ALP > 50%) was seen in 35%. Median overall survival (OS) was 10 months. Median OS for patients who received Ra-223 as 1st or 2nd line vs subsequent line was 13 and 8.5 months respectively. Median OS for patients with Haemoglobin (Hb) ≥12gm/dl) and lower Hb were 11 and 7 months respectively. Patients with normal albumin at baseline had median OS of 15 months compared to 6 months for those with low albumin. Favourable prognostic group patients (Normal Hb, Normal albumin at baseline and normal ALP after treatment) had a median OS of 31 months compared to 9 months for those in less favourable group. Patients who completed at least 5 cycles had a better survival (13 vs 8.5 months). Fractures were reported in 3 patients (6.5%). Conclusions: Ra-223 treated patients in our centre had a reasonable survival despite having a larger proportion of elderly population compared to the registration trial. Ra-223 treatment seems to be more effective when given at earlier lines in mCRPC setting. Pre-treatment Albumin lower than normal seems to be a predictor for poor overall survival.