Abstract 2893: Incorporation of the SEZ6L SNP into The Liverpool Lung Project Risk Prediction Model

Abstract

Abstract Lung cancer risk prediction models will be essential tools for the identification of high-risk individuals if ongoing lung cancer CT screening trials report positive findings. The Liverpool Lung Project (LLP) risk model estimates an individual's 5-year absolute risk of lung cancer based on five epidemiological risk factors; smoking duration, prior diagnosis of pneumonia or another cancer; family history of lung cancer including age at onset; and occupational exposure to asbestos (Cassidy et al. The LLP Risk Model: an individual risk prediction model for lung cancer. British J Cancer; 98 (2):270-6, 2008). DNA genotyping indicated a role of the SEZ6L, a Met430IIe polymorphic variant found on 22q12.2 region to have increased risk of lung cancer (Gorlov et al. Cancer Research 7: 8406-11, 2007). Therefore, we quantified the improvement in risk prediction attributed to addition of SEZ6L in the LLP risk model. SEZ6L was genotyped in 200 lung cancer patients and 188 controls from the Liverpool Lung Project, and combined with previously identified epidemiological risk factors. Bivariate association between the SNP and the epidemiological risk factors was studied using the Pearson's chi-squared test; multivariable conditional logistic regression modelled individual risk profile, which was combined with local incidence data for prediction of absolute risk of lung cancer in the next 5-year. The improvements in the models associated with the SNP was assessed through a pairwise comparison of the area under the receiver operating characteristic curve (ROC-AUC) and the Net Reclassification Improvements (NRI). A higher proportion of homozygote mutant genotype was observed among patients (10%) compared to controls (3%). No interaction was observed between any of the epidemiological risk factors and the SEZ6L genotype, in combined or separate analysis of patients and controls. The extended model (model including the SNP genotype) showed better calibration compared to the baseline model (model with only epidemiological factors). There was a modest, but statistically significant increase when SEZ6L was incorporated into the risk model, the ROC-AUC increased from 0.72 (0.66-0.77) for the baseline model to 0.75 (0.71-0.80) for the extended model. The NRI also revealed a significant improvement of around 12% for the extended model; this improvement was even better for individuals classified into the two intermediate risk categories by the baseline model (NRI=27%). Our results suggest addition of SEZ6L improved the performance of the LLP risk model particularly for individuals whose initial absolute risks were unable to discriminate them into ‘low’ or ‘high’ risk group. However, given the small number of patients and controls included in this analysis, further validation of the observed improvement is essential. Note: This abstract was not presented at the AACR 101st Annual Meeting 2010 because the presenter was unable to attend. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2893.