Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jul;49(1):361-70.
doi: 10.3892/ijo.2016.3499. Epub 2016 Apr 25.

Incorporating Epistasis Interaction of Genetic Susceptibility Single Nucleotide Polymorphisms in a Lung Cancer Risk Prediction Model

Free PMC article

Incorporating Epistasis Interaction of Genetic Susceptibility Single Nucleotide Polymorphisms in a Lung Cancer Risk Prediction Model

Michael W Marcus et al. Int J Oncol. .
Free PMC article


Incorporation of genetic variants such as single nucleotide polymorphisms (SNPs) into risk prediction models may account for a substantial fraction of attributable disease risk. Genetic data, from 2385 subjects recruited into the Liverpool Lung Project (LLP) between 2000 and 2008, consisting of 20 SNPs independently validated in a candidate-gene discovery study was used. Multifactor dimensionality reduction (MDR) and random forest (RF) were used to explore evidence of epistasis among 20 replicated SNPs. Multivariable logistic regression was used to identify similar risk predictors for lung cancer in the LLP risk model for the epidemiological model and extended model with SNPs. Both models were internally validated using the bootstrap method and model performance was assessed using area under the curve (AUC) and net reclassification improvement (NRI). Using MDR and RF, the overall best classifier of lung cancer status were SNPs rs1799732 (DRD2), rs5744256 (IL-18), rs2306022 (ITGA11) with training accuracy of 0.6592 and a testing accuracy of 0.6572 and a cross-validation consistency of 10/10 with permutation testing P<0.0001. The apparent AUC of the epidemiological model was 0.75 (95% CI 0.73-0.77). When epistatic data were incorporated in the extended model, the AUC increased to 0.81 (95% CI 0.79-0.83) which corresponds to 8% increase in AUC (DeLong's test P=2.2e-16); 17.5% by NRI. After correction for optimism, the AUC was 0.73 for the epidemiological model and 0.79 for the extended model. Our results showed modest improvement in lung cancer risk prediction when the SNP epistasis factor was added.


Figure 1
Figure 1
Performance of lung cancer risk model with and without the SNP epistatic effect.

Similar articles

See all similar articles

Cited by 2 articles


    1. Field JK, Chen Y, Marcus MW, Mcronald FE, Raji OY, Duffy SW. The contribution of risk prediction models to early detection of lung cancer. J Surg Oncol. 2013;108:304–311. doi: 10.1002/jso.23384. - DOI - PubMed
    1. Bach PB, Kattan MW, Thornquist MD, Kris MG, Tate RC, Barnett MJ, Hsieh LJ, Begg CB. Variations in lung cancer risk among smokers. J Natl Cancer Inst. 2003;95:470–478. doi: 10.1093/jnci/95.6.470. - DOI - PubMed
    1. Cassidy A, Myles JP, van Tongeren M, Page RD, Liloglou T, Duffy SW, Field JK. The LLP risk model: An individual risk prediction model for lung cancer. Br J Cancer. 2008;98:270–276. doi: 10.1038/sj.bjc.6604158. - DOI - PMC - PubMed
    1. Hoggart C, Brennan P, Tjonneland A, Vogel U, Overvad K, Østergaard JN, Kaaks R, Canzian F, Boeing H, Steffen A, et al. A risk model for lung cancer incidence. Cancer Prev Res (Phila) 2012;5:834–846. doi: 10.1158/1940-6207.CAPR-11-0237. - DOI - PMC - PubMed
    1. Park S, Nam BH, Yang HR, Lee JA, Lim H, Han JT, Park IS, Shin HR, Lee JS. Individualized risk prediction model for lung cancer in Korean men. PLoS One. 2013;8:e54823. doi: 10.1371/journal.pone.0054823. - DOI - PMC - PubMed