Objective   To study the correlation of polymorphism of CYP1A1Msp1 and seurm selenium level independently and in combination with the risk of lung cancer.

  Methods   A case-control study was conducted, which included 58 cases of lung cancer and 62 controls. The genotypes of CYP1A1 were detected by PCR-RELP and the serum selenium level was determined by GHAFS.

  Results   There was no significant difference in polymorphism frequencies of CYP1A1 Msp I between the two groups. The genotypes were no significantly associated with the risk of lung cancer; Average serum level of selenium among the lung cancer cases was significantly lower than that among controls(P=0.001);Comparing with those who had higher serum selenium(> 0.109 mg/L)with CYP1A1m₁ m₁, the odds ratios of those who had lower serum selenium(< 0.109 mg/L)with CYP1A1m₂ m₂, with CYP1A1m₁m₂, with CYP1A1m₁ m₁ were 9.00(P=0.006), 3194(P=01195), 5.40(P=0.036)respectively, and those who had higher serum selenium with CYP1A1 m₂m₂, with CYP1A1m₁ m₂ were 1.69(P=0.500), 1.13(P=0.705)respectively.

  Conclusion   Polymorphism of CYP1A1 was not associated with the risk of lung cancer, but there was significant inverse association between the serum selenium level and the risk of lung cancer. A combined effect of polymorphism of CYP1A1 Msp I and serum selenium level on the risk of lung cancer was suggested.