The etiology of colorectal cancer is complex and multifactorial. Tobacco smoke has been found to be associated both with colorectal adenoma and cancer development. It was hypothesized that tobacco smoking could interact with genetic factors, providing different risk estimates according to different genetic predisposition. We reviewed and summarized by a meta-analytic approach the evidence from the literature on the interaction of smoking with the five most studied gene polymorphisms (GSTM1, GSTT1, mEH3, mEH4, NAT2). We calculated pooled Odds Ratios for each gene-smoking combination by random effects models, and provided a pooled P-value for gene-smoking interaction. Heterogeneity among studies was evaluated by the Q statistic and I(2). Overall, 27 case-control studies or nested case-control studies are included in this review: 12 presented data on GSTM1 polymorphism, eight on GSTT1, seven on mEH3, mEH4, and 10 on NAT2. We found a weak suggestion of an antagonistic effect of mEH3 low or medium metabolizer with smoking on colorectal adenoma risk (pooled P-value for the interaction: 0.02): smokers carriers of mEH3 low or medium metabolizer had slightly lower risk (Odds Ratio; 95% Confidence Interval: 1.6; 1.2-2.1) than smokers with mEH3 high metabolizer (1.8; 1.4-2.4). A non-significant positive interaction between GSTT1 null genotype and smoking was suggested for colorectal adenoma risk. None of the other common genetic polymorphisms involved in tobacco carcinogens metabolism seemed to modify the smoking-related risk of colorectal adenoma or cancer.