Prognostic value of neutrophil-to-lymphocyte ratio in advanced oesophago-gastric cancer: exploratory analysis of the REAL-2 trial.

ABSTRACT

BACKGROUND: The REAL-2 trial demonstrated that capecitabine and oxaliplatin were effective alternatives to fluorouracil and cisplatin, respectively, when used in triplet chemotherapy regimens for previously untreated oesophago-gastric cancer. The aim of the current analysis was to evaluate the prognostic value of the neutrophil-to-lymphocyte ratio (NLR) in the REAL-2 cohort. MATERIAL AND METHODS: A post hoc exploratory analysis was carried out on REAL-2 patients with the available absolute neutrophil count and absolute lymphocyte count. A high NLR was defined using a cut-off value of >3.0. The NLR was then correlated with clinical outcomes including overall survival (OS), progression-free survival (PFS) and objective response rate. Survival curves were generated using the Kaplan-Meier method and comparison between groups was carried out using Cox regression. RESULTS: Data were available in 908 of the 1002 REAL-2 participants. Of these, 516 (56.8%) were deemed to have a high NLR. In univariate analysis, high NLR was associated with a hazard ratio (HR) for OS of 1.73 (1.50-2.00), P < 0.001, compared with low NLR, equating to median OS values of 9.1 [95% confidence interval (CI) 8.0-9.6] and 12.7 months (95% CI 10.8-14.4), respectively. The NLR remained highly significant for OS (P < 0.001) in a multivariate model including performance status, age, disease extent, presence of liver metastases and presence of peritoneal metastases. For PFS, high NLR was associated with an HR of 1.63 (1.41-1.87), P < 0.001, compared with low NLR in univariate analysis. No significant interaction was found between NLR status and treatment arm, 13% of all patients with low NLR achieving survival beyond 24 months compared with only 6% of patients with high NLR (P < 0.001). CONCLUSION: Our results confirm that high NLR status had a significant negative prognostic effect in the REAL-2 trial population. Based on the multivariate analysis, this effect was independent of other known prognostic factors.