629P Neutrophil-lymphocyte ratio (NLR) as a prognostic and predictive biomarker in patients with metastatic castration-resistant prostate cancer (mCRPC) treated with cabazitaxel (CBZ) vs abiraterone or enzalutamide in the CARD study

High NLR as a biomarker of inflammation is associated with poor overall survival (OS) in different malignancies, including mCRPC (Lorente et al, Ann Oncol 2015). The CARD study (NCT02485691) reported improved radiographic progression-free survival (rPFS) and OS with CBZ vs abiraterone or enzalutamide in patients with mCRPC who previously received docetaxel and progressed within 12 months on the alternative androgen receptor-targeted agent (ARTA) (de Wit et al, N Engl J Med 2019). This analysis of CARD evaluated the impact of baseline NLR on outcomes. Multivariable Cox regression analysis with stepwise selection of covariates (stratification factors and pre-planned prognostic factors), including adjustment for treatment, was used to investigate the prognostic association between baseline NLR (as a continuous variable) and OS. The associations between baseline NLR (< vs ≥ median), and OS, rPFS, time to prostate-specific antigen (PSA) progression, and PSA response (confirmed PSA decline ≥ 50% from baseline) were also evaluated. Baseline median NLR in both arms overall was 3.38; higher baseline NLR independently associated with poor OS (HR [95% CI]: 1.05 [1.02–1.08]; p = 0.0003). Additional factors associated with poor OS were lower hemoglobin and high PSA at baseline. Greater clinical activity in terms of rPFS, time to PSA progression and PSA response was seen with CBZ irrespective of baseline NLR, with the benefits of CBZ vs ARTA particularly marked in patients with NLR ≥ median (Table). CBZ also significantly prolonged OS vs ARTA in patients with baseline NLR ≥ median (HR [95% CI]: 0.49 [0.30–0.81]; log-rank p = 0.004).Table: 629PNLR ≥ medianNLR < medianCBZ (n = 63)ARTA (n = 60)CBZ (n = 62)ARTA (n = 61)Median rPFS, months (95% CI)8.5 (4.9–11.4)2.8 (2.7–4.5)7.5 (5.4–8.5)5.1 (3.1–7.0)Median OS, months (95% CI)*15.3 (11.8–20.3)9.5 (9.0–11.8)12.9 (10.5–19.1)13.3 (9.3–17.3)CBZ (n = 56)ARTA (n = 52)CBZ (n = 56)ARTA (n = 50)Time to PSA progression, months (95% CI)*6.9 (3.5–10.3)2.1 (1.7–2.8)5.8 (3.5–8.8)2.1 (1.4–2.8)PSA response, %*37.517.335.712.0* Post-hoc analyses. Open table in a new tab * Post-hoc analyses. This present analysis of CARD confirms that NLR is prognostic for poor outcomes in mCRPC. The superiority of CBZ vs a second ARTA was particularly marked in patients with high baseline NLR.