MicroRNA-Related Genetic Variants As Predictors of Early Stage Non–Small Cell Lung Cancer Clinical Outcomes
X. Pu,C. Lu,D. J. Stewart,J. Gu,M. A. T. Hildebrandt,J. Lin,S. M. Lippman,W. Xifeng
DOI: https://doi.org/10.1158/1055-9965.epi-11-0095
2011-01-01
Abstract:Abstract Purpose: MicroRNAs (miRNA) can function as oncogenes or tumor suppressors. In this study, we test the hypothesis that genetic variation within miRNA processing genes or miRNA binding sites in cancer-related genes may alter clinical outcomes for non–small cell cancer (NSCLC) patients. Methods: We genotyped 72 single nucleotide polymorphisms (SNPs) from eight miRNA processing genes and 168 SNPs from 133 predicted miRNA binding sites in 535 patients with early stage (stages I, IIA, and IIB) NSCLC to determine the effect of these variations on progression risk and overall survival. We further conducted a subset analysis to exclude potential bias due to different treatment regimens. Results: In the analysis of overall survival, an FAS SNP remained significant for decreased risk after multiple comparisons in all early patients (HR: 0.59, 95% CI: 0.44–0.77; GG: MST = 58 mos, GA + AA: MST = 118 mos) and surgery plus chemotherapy subgroup (HR: 0.19, 95% CI: 0.07–0.46, P = 1.84 × 10−5; GG: MST = 65 mos, GA + AA: MST = 137 mos). In the surgery-only patients, an FZD4 SNP (HR: 0.46, 95% CI: 0.32–0.65; GG: MST = 59 mons, GA + AA: MST = 117 mos) had a significant protective effect after multiple comparisons adjustment. For progression risk, following multiple comparison, an SP1 SNP (HR: 2.19, 95% CI: 1.45–3.32; GG: MST > 270 mos, GA + AA: MST = 45 mos) and an MBD1 SNP (HR: 2.41, 95% CI: 1.45–4.0; AA: MST > 270 mos, C + CC: MST = 90 mos) remained significant increased progression risk in the entire and surgery only population respectively. All the associations significant after adjusting for multiple comparisons were validated by Bootstrap resampling method. A strong cumulative effect of these variant genotypes to increase risk and dramatically decrease median event-free time was consistently observed. We also identified potential higher-order gene–gene interactions among these SNPs. Conclusions: We identified significant association between miRNA-related genetic variants and early stage NSCLC patient clinical outcomes. With validation, our result can be used in the prognosis of clinical outcomes for early stage NSCLC patients.