Neutrophil-to-lymphocyte ratio (NLR) as a prognostic indicator of treatment response to novel hormones in metastatic castrate-resistant prostate cancer: Real-world data from a single institution.

Abstract

52 Background: Prostate cancer is the most common solid malignancy in men. Despite advances in radical management, a proportion of patients develop castrate resistant metastatic disease (mCRPC). Management options in mCRPC include the novel hormones Enzalutamide (E) or Abiraterone Acetate (AA). There is limited knowledge of clinical factors suggesting improved response to either E or AA to guide clinician choice. Systemic inflammation caused by tumor activity may predict response to E or AA. The NLR is a marker of systemic inflammation which has been assessed previously. We present findings from a single institution evaluation of NLR for E and AA. Methods: Men receiving E or AA for mCRPC between March 2018 - February 2020 at our institution were identified. Baseline characteristics and pre-specified metrics including hematology, age and disease burden as well as time on treatment and survival data were retrospectively taken from hospital systems and collated into a single dataset. Time on treatment was used as a surrogate for progression free survival (PFS). Patients could switch therapies if they developed dose limiting toxicity within 3 months of treatment as per UK guidelines. AUROC curve analysis was applied to determine NLR cut off for E and AA cohorts specific for PFS. Results: 128 patients are included in the final analysis; 65 patients received E, 63 AA. Cohorts were well balanced with a median age at initial treatment of 75 (range 46-102) for AA and 74 (range 48-91) for E. The median Gleason score for AA was 8 and 9 for E. Patients receiving AA were more likely to have greater than 3 bone metastases (85% vs 74%). Patients receiving E were more likely to have visceral disease, 13% vs 10%. NLR cutoff, determined by AUROC curve analysis was 3.09 for E with an AUC of 0.782 (p <0.001). For AA, NLR cutoff was 3.1 with a non-statistically significant AUC of 0.588 (p=0.704). Median PFS was 239 days (range 13-1997) for E, 175 days (range 22-2037) for AA. Conclusions: The data does not support the hypothesis that NLR can be used to determine between the therapeutic option of E or AA in individual patients. Regardless of NLR, both E and AA remain good therapeutic options for men with mCRPC. This single institution real world data provides evidence of the role of NLR in determining response to novel hormone agents in the mCRPC setting for patients commencing E. No statistically significant NLR could be calculated for patients receiving AA. Although well balanced, in this non-randomized population, patients with multiple bony metastases were more likely to receive AA and those with visceral metastases, E. It is possible that this imbalance explains the lack of statistical significance in the AA cohort. We aim to expand our study cohort and evaluate OS and other potential metrics including the effects of tumor heterogeneity on the prognostic value of NLR.