Abstract A4: SNP-SNP interaction and nongenetic risk factors combine to identify lung cancer susceptibility: Validation in a prospective study

Abstract

Abstract Introduction: Key risk indicators for lung cancer include a smoking history, genetic factors, increasing age, family history and impaired lung function (COPD). In a case-control study, 20 genetic markers associated with lung cancer were combined with non-genetic risks to derive a lung cancer susceptibility score (LCSS). The aim of this prospective follow up study was to compare the LCSS in a cohort of very high risk smokers, with and without lung cancer. Methods: A cohort of 728 high risk individuals (smoker/ex-smokers, >40 yrs old, >20 pack years, with spirometric confirmed COPD) were recruited and followed for a mean of 5 years. All volunteers underwent spirometry, completed a modified ATS respiratory questionnaire and gave blood for DNA. These volunteers were recruited from the community (N=420) and a COPD outpatient clinic (N=308). To date, we have identified 53 cases of lung cancer (histology confirmed) in this cohort (24 community and 28 clinic). The 20 SNP panel was genotyped using the iPLEX system. Each subjects' genotypes were combined, in a predefined algorithm, to derive the lung cancer susceptibility score. Results: COPD patients subsequently diagnosed with lung caner were comparable to those with no lung cancer for age, gender, smoking exposure. The mean LCSS, was higher in those who developed lung cancer than those who did not (LCSS 7.0 vs 5.7, P=0.06), and significantly higher than those recruited from the community with mild-moderate COPD (LCSS 7.0 vs 4.7, P<0.001). The performance characteristics of the LCSS reported here, confirms its utility in high risk smokers, to target lung cancer screening and/or diagnostic approaches (using molecular, cytological and/or radiological based tools). Conclusion: We show, in a prospectively recruited high risk cohort, that our LCSS identifies those with mild-moderate COPD at greatest risk of lung cancer. Such a group might benefit from aggressive screening/diagnostic approaches, so that their lung cancer is diagnosed in stage 1 when survival has been shown to be dramatically improved. Citation Information: Cancer Prev Res 2010;3(1 Suppl):A4.