Previous studies have suggested that GST genotypes may play a role in determining susceptibility to lung cancer, though the data are often conflicting. In this study we investigated GSTM1, GSTT1 and GSTP1 status in relation to lung cancer risk in patients attending a Manchester bronchoscopy clinic. Cases were all patients (n=94) currently with, or with a history of, tumours of the lung, trachea or bronchus. The control group were all other patients (n=165) who were free of benign and malignant tumours both at the time of, or prior to, diagnosis. All patients were interviewed for information on lifestyle risk factors, and DNA extracted from bronchial lavage and blood samples was used for genotyping. GSTM1 null genotype was associated with decreased lung cancer risk (odds ratio (OR) 0.50, 95% confidence interval (CI) 0.29-0.87), particularly among men (OR 0.43, 95% CI 0.21-0.87) and those above the median age (OR 0.33, 95% CI 0.15-0.70). No difference in GSTT1 and GSTP1 genotype distribution was seen between cases and controls. The GSTM1 null genotype was associated with a decreased risk of squamous cell carcinoma: the OR, adjusted for age, sex and pack years was 0.32 (95% CI 0.12-0.82). As previous studies have reported that the GSTM1 null genotype is associated with an increased lung cancer risk, further work is required to determine whether the observed association is true, or whether it arises from bias or confounding factors.