Smoking cessation decreases the risk of lung cancer. However, little is known about how smoking cessation affects lung cancer survival. We examined the association between smoking cessation and overall survival (OS) and recurrence-free survival (RFS) in 543 early stage non-small cell lung cancer (NSCLC) patients. The data were analyzed using log-rank test and Cox proportional hazard models, adjusting for age, gender, stage, and smoking intensity. The median follow-up time was 57 months (range 0.2-140 months). There were 191 recurrences and 285 deaths. The 5-year OS rates were 50% (95% confidence interval (CI), 43-58%) for current smokers, 54% (44-65%) for ex-smokers who quit 1-8 years, 59% (49-70%) for ex-smokers who quit 9-17 years, 58% (47-69%) for ex-smokers who quit > or =18 years prior to diagnosis, and 76% (63-90%) for never smokers (P=0.09, log-rank test). The adjusted hazard ratios for ex-smokers who quit 1-8, 9-17, > or =18 years, and never smokers were 0.82 (95% CI, 0.59-1.13), 0.69 (0.49-0.97), 0.66 (0.45-0.95), and 0.54 (0.29-0.996), respectively, when compared with current smokers (P(trend)=0.004). Similar associations were found among ever smokers-only, when smoking cessation time was treated as a continuous variable, and for RFS. The significantly beneficial effects of smoking cessation on OS and RFS were observed among women only, while not among men (P=0.01 for interactions between gender and smoking cessation). In conclusion, smoking cessation is associated with improved survival in early stage NSCLC patients. The longer the time since cessation of smoking, the better the survival outcome.