Comprehensive genetic characteristics of multiple primary lung cancer in patients in China.

Abstract

e20597 Background: With the development of molecular biology and next-generation sequencing (NGS), researchers have been exploring the use of genomic technologies for the classification of lung cancers. The aim of this study was to investigate genetic characteristics of multiple primary lung cancer (MPLC) to facilitate precise diagnosis and therapy by NGS. Methods: A total of 271 lesions from 92 patients with MPLC (225 lesions) and 20 intrapulmonary metastasis (IM, 46 lesions) which were diagnosed by American College of Chest Physicians (ACCP) guidelines, were sequenced by Acornmed panel with 808 cancer-related genes. Results: The most frequent mutation in 225 lesions was EGFR (46.7%), followed by RBM10 (19.6%), BRAF (16.4%) and KRAS (15.1%) among MPLC while IM lesions showed high TP53 (63.0%) mutation frequency. Further analysis showed that MPLC patients had higher proportion of private mutations, and lower proportion of shared mutations and branch shared mutations than IM (p < 0.0001). In our cohort, 95.7% (88/92) of MPLC patients had at least one driver alteration, and EGFR and KRAS comprised 57.4% and 18.0% of 183 driver mutations, respectively. Further analysis showed that only 19.6% (18/92) of MPLC shared the same driver mutation in each patient and 61.1% (11/18) of co-driver mutation was EGFR L858R which was commonly detected in lung cancer. In contrast, 95.0% (19/20) of IM shared the same driver mutation in each patient. MPLCs were associated with a high frequency of T > C transitions (p < 0.001), while T > G transversions (p < 0.001) were more frequent in tumors from IM. MPLC had significantly higher intratumor heterogeneity (ITH) (17.1 vs 4.05, p < 0.001) and lower weighted-genomic integrity index (wGII) than IM (0.01638 vs 0.18321, p < 0.001). Besides, MPLC had lower percentage of clonal mutations (p < 0.001) than IM. However, IM had higher arm-level somatic copy number variations fraction (p < 0.001) than MPLC. Conclusions: Our analysis reveals that patients with MPLC have distinct molecular characteristics different from IM, which may increase the understanding of MPLC. In addition, the molecular features of MPLC may provide new opportunities to distinguish MPLC from IM by broad panel NGS.