Only 15% of heavy smokers go on to develop lung cancer, indicating the existence of individual predisposition.Using a case-control study, we ascertained whether there were differences between cases and controls in the frequency of GSTM1 and GSTT1 gene polymorphisms, and whether their effect might be linked to smoking habit.While a risk of lung cancer of 1.7 (1.0-3.0) was observed for GSTM1 gene deficiency, there appeared to be no such association for absence of the GSTT1 gene. The interaction observed with tobacco use indicated an excess risk of 24.5% (IC 95%-59.8-109.0) for the GSTM1 gene, with no such interaction in evidence for the GSTT1 gene. The highest risk for a combination of the different haplotypes was for subjects with the GSTT1 gene present and the GSTM1 gene deleted, with an Odds Ratio of 2.19 (1.18-4.07).The absence of the GSTM1 gene is, while that of the GSTT1 gene is seemingly not, implicated in susceptibility to lung cancer.