Prognostic role of inflammatory biomarkers from peripheral blood and clinical factors in metastatic castration-resistant prostate cancer (mCRPC) patients (pts) treated with radium-223 (Ra-223) (BIO-Ra-223 study).

Abstract

e17026 Background: Ra-223 is a treatment option for mCRPC pts with bone metastases according to the survival benefit observed compared to placebo in the ALSYMPCA trial. In the last years, many studies showed this benefit in the real-life pts is lower than that reported in the trial, probably due to a suboptimal selection of pts with poor prognostic characteristics. Therefore, the identification of prognostic factors to select mCRPC pts most likely to benefit from Ra-223 is needed. The multicentre retrospective BIO-Ra-223 study has investigated the prognostic role of peripheral blood immune cells and clinical factors to develop a novel prognostic score for mCRPC pts treated with Ra-223. Methods: Complete blood count was assessed before Ra-223 treatment calculating neutrophil-to-lymphocyte ratio (NLR), derived NLR (dNLR), lymphocyte-to-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), systemic inflammation index (SII). Clinical factors included pre-treatment Eastern Cooperative Oncology Group performance status (ECOG PS), Gleason Score (GS) group, number of bone metastases, alkaline phosphatase (ALP), line of therapy, previous chemotherapy and the presence of lymphadenopathies. Statistical analyses included survival ROC curves for biomarkers’ cutoffs, univariable and multivariable Cox analyses, internal validation, c-index calculation and Schneeweiss scoring system. Results: From September 2013 to July 2020, 519 mCRPC pts received Ra-223 as 1st-2nd, 3rd-4th and further-line in 48%, 38% and 14% of pts. The median overall survival (mOS) of the entire cohort was 19.9 months. All biomarkers and clinical factors (except for GS group) significantly predicted OS at the univariable analyses. In the multivariable ones, all biomarkers, ECOG PS, number of bone metastases and ALP significantly correlated with OS. The multivariable model with NLR (< 3.1 vs ≥3.1), ECOG PS (0-1 vs 2-3), number of bone metastases (< 6, 6-20, > 20) and ALP (< 220 vs ≥220) showed the highest c-index (0.711), which was maintained after internal validation (bootstrap re-sampling) (c-index: 0.707). Using the Schneeweiss scoring system, ten categories were identified in 494 pts with complete data and merged in two prognostic groups with distinctive OSs: group 1 (score 0-4, 337 pts) with a mOS of 27.8 months and group 2 (score 5-10, 157 pts) with a mOS of 9.7 months (HR 4.03, p < 0.001). Conclusions: The obtained score, composed of NLR, ECOG PS, number of bone metastases, and ALP identifies two distinctive prognostic groups of mCRPC pts. Moreover, this score is easily and widely applicable for clinical practice and trials at no additional costs. Although external validation is needed, these preliminary results showed that this novel prognostic score is promising and could help the patients’ selection for Ra-223 treatment.