There is little doubt that cigarette smoking remains a major environmental health risk that humans are facing in the twenty-first century. Cigarette smokers are more likely to develop many forms of diseases than nonsmokers, including cancers and vascular diseases. With the availability of the human genome sequence, we become more aware of the genetic contributions to these common diseases, especially the interactive relations between environmental factors (e.g., smoking) and genes on disease susceptibility, development, and prognosis. Although smoking is responsible for up to 30% of pancreatic cancers and about 10% of cases are ascribed to genetic reasons, some genetic variants do not predispose carriers to disease development unless they are exposed to a specific adverse environment such as smoking. This smoke-gene interaction could potentially be responsible for most of the cases. Certain polymorphisms in genes such as CYP1A1 have been shown particularly sensitive to smoking-induced pathogenesis, including pancreatic cancer and atherosclerosis. We found that individuals with CYP1A1 CC genotype had a more than three fold increase in risk for severe coronary atherosclerosis when they smoked. Patients with endothelial nitric oxide synthase (eNOS) intron 4 27 repeat homozygotes were more likely to develop severe coronary stenosis when they smoked. On the other hand, DNA variants at the eNOS gene also dictate how smoking affects the expression of eNOS. We showed that GSTM1 deficiency was not involved in smoking-induced vascular diseases, but p53 polymorphisms tended to modify the disease severity in smokers. We are still at an early stage of defining the pairs and mechanisms of smoke-gene interaction, and this etiologic mechanism may hold great potential for risk assessment, treatment strategy, and prognostic predictions.