作者: Jiyoun Yeo , Erin L. Crawford , Xiaolu Zhang , Sadik Khuder , Tian Chen
DOI: 10.1186/S12885-017-3287-4
关键词:
摘要: Annual low dose CT (LDCT) screening of individuals at high demographic risk reduces lung cancer mortality by more than 20%. However, subjects selected for based on criteria typically have less a 10% lifetime cancer. Thus, there is need biomarker that better stratifies LDCT screening. Toward this goal, we previously reported test (LCRT) comprising 14 genome-maintenance (GM) pathway genes measured in normal bronchial epithelial cells (NBEC) accurately classified (CA) from non-cancer (NC) subjects. The primary goal the studies here was to optimize LCRT specificity and ease clinical implementation. Targeted competitive multiplex PCR amplicon libraries were prepared next generation sequencing (NGS) analysis transcript abundance 68 sites among 33 GM target NBEC specimens collected retrospective cohort 120 subjects, including 61 CA cases 59 NC controls. Genes contribution and/or prior evidence association with risk. Linear discriminant used identify most accurate classifier suitable stratify After cross-validation, model expression values 12 (CDKN1A, E2F1, ERCC1, ERCC4, ERCC5, GPX1, GSTP1, KEAP1, RB1, TP53, TP63, XRCC1) factors age, gender, pack-years smoking, had Receiver Operator Characteristic area under curve (ROC AUC) 0.975 (95% CI: 0.96–0.99). overall classification accuracy 93% CI 88%–98%) sensitivity 93.1%, 92.9%, positive predictive value 93.1% negative 93%. ROC AUC significantly (p < 0.0001) best features alone. displayed implementation throughput, quality-controlled targeted NGS platform. As such, it optimized validation ongoing blinded prospective study. Following validation, expected utility stratifying annual compared current