Validation of neutrophil-to-lymphocyte ratio in a multi-institutional cohort of patients with T1G3 non–muscle-invasive bladder cancer

Neutrophil-to-lymphocyte ratio was found associated with worse disease recurrence and progression in patients with T1 nonemuscle-invasive bladder cancer in some single-center studies. We validated high pretreatment neutrophil-to-lymphocyte ratio (cutoff, 3) as an independent predictor of disease recurrence, progression, and cancer-specific survival in patients with primary T1 HG/G3 nonemuscle-invasive bladder cancer treated with intravesical bacillus Calmette-Guérin therapy. Introduction: The aim of this multicenter study was to investigate the prognostic role of neutrophil-to-lymphocyte ratio (NLR) and to validate the NLR cutoff of 3 in a large multi-institutional cohort of patients with primary T1 HG/ G3 nonemuscle-invasive bladder cancer (NMIBC). Patients and Methods: The study period was from January 2002 through December 2012. A total of 1046 patients with primary T1 HG/G3 who had NMIBC on re-transurethral bladder resection (TURB) who received adjuvant intravesical bacillus Calmette-Guérin therapy with maintenance from 13 academic institutions were included. Endpoints were time to disease, and recurrence-free (RFS), progression-free (PFS), overall (OS), and cancer-specific survival (CSS). Results: A total of 512 (48.9%) of patients had NLR 3 prior to TURB. High pretreatment NLR was associated with female gender and residual T1HG/G3 on re-TURB. The 5-year RFS estimates were 9.4% (95% confidence interval [CI], 6.8%-12.4%) in patients with NLR 3 compared with 58.8% (95% CI, 54%-63.2%) in patients with NLR < 3; the 5-year PFS estimates were 57.1% (95% CI, 51.5%-62.2%) versus 79.2% (95% CI, 74.7%-83%; P < .0001); the 10-year OS estimates were 63.6% (95% CI, 55%-71%) versus 66.5% (95% CI, 56.8%-74.5%; P ¼ .03); the 10-year CSS estimates were 77.4% (95% CI, 68.4%-84.2%) versus 84.3% (95% CI, 76.6%-89.7%; P ¼ .004). NLR was independently associated with disease recurrence (hazard ratio [HR], 3.34; 95% CI, 2.82-3.95; P < .001), progression (HR, 2.18; 95% CI, 1.71-2.78; P < .001) and CSS (HR, 1.65; 95% CI, 1.02-2.66; P ¼ .03). The addition of NLR to a multivariable model that included established features increased its discrimination for predicting of RFS (þ6.9%), PFS (þ1.8%), and CSS (þ1.7%). Conclusions: Pretreatment NLR 3 was a strong predictor for RFS, PFS, and CSS in patients with primary T1 HG/G3 NMIBC. It could help in the decision-making regarding intensity of therapy and follow-up.