Background: Smoking may affect the efficacy of chemotherapy and the incidence of adverse events. We investigated the correlation between smoking history and gemcitabine-induced neutropenia.
Patients and methods: Data on smoking history and incidence of grade 3-4 neutropenia were retrospectively gathered for 103 chemo-naive patients treated with gemcitabine monotherapy (59 patients with pancreatic, 41 with hepatobiliary and three with other cancers).
Results: There was a significantly higher incidence of grade 3-4 neutropenia among patients without a history of smoking (55.7%) than among those with a history of smoking (including current and ex-smokers; 23.6%) [odds ratio (OR) 0.244, 95% confidence interval (CI) 0.105-0.569; P < 0.001]. After adjustment for age, gender, platelet and baseline neutrophil counts, history of surgery for primary cancer, creatinine concentration, hemoglobin concentration, aspartate aminotransferase concentration, alanine aminotransferase concentration and total bilirubin concentration, logistic regression analysis identified a history of smoking as an independent inverse predictor of gemcitabine-induced neutropenia (OR 0.188, 95% CI 0.057-0.618; P = 0.006).
Conclusion: Patients without a history of smoking may be at higher risk of developing gemcitabine-induced neutropenia. The mechanism underlying this phenomenon is unclear at this point.