Abstract
BACKGROUND: Neutropenia is a common adverse reaction of chemotherapy. We assessed whether chemotherapy-induced neutropenia could be a predictor of survival for patients with non-small-cell lung cancer (NSCLC). METHODS: A total of 387 chemotherapy-naïve patients who received chemotherapy (vinorelbine and gemcitabine followed by docetaxel, or paclitaxel and carboplatin) in a randomised controlled trial were evaluated. The proportional-hazards regression model was used to examine the effects of chemotherapy-induced neutropenia and tumour response on overall survival. Landmark analysis was used to lessen the bias of more severe neutropenia resulting from more treatment cycles allowed by longer survival, whereby patients who died within 126 days of starting chemotherapy were excluded. RESULTS: The adjusted hazard ratios for patients with grade-1 to 2 neutropenia or grade-3 to 4 neutropenia compared with no neutropenia were 0.59 (95% confidence interval (CI), 0.36-0.97) and 0.71 (95% CI, 0.49-1.03), respectively. The hazard ratios did not differ significantly between the patients who developed neutropenia with stable disease (SD), and those who lacked neutropenia with partial response (PR). CONCLUSION: Chemotherapy-induced neutropenia is a predictor of better survival for patients with advanced NSCLC. Prospective randomised trials of early-dose increases guided by chemotherapy-induced toxicities are warranted.