Clinical and molecular characteristics of multiple primaries identified by whole-exome sequencing-based genetic divergence in non-small-cell lung cancer.

Abstract

e20558 Background: With the recommendation of high-resolution chest computed tomography scan in lung cancer screening, multiple primary lung cancers (MPLC) is gaining more attention. Identifying MPLC from intrapulmonary metastases (IPM) is crucial to the clinical management. The current guidelines for the identification are always based on expert consensus rather than clinical evidence. We aim to develop a reliable identification method and further figure out the clinical and genetic features of MPLC. Methods: We retrospectively screened 305 lung lesions from 131 patients who underwent radical resection for multiple lung carcinomas. To minimize the effect of therapeutic intervention on prognostic outcomes, patients with each single tumor size smaller than 4cm, free of nodal or systemic metastases, and had high-quality whole-exon sequencing (WES) data were included. We selected 11 patients with confirmed intrapulmonary metastases for training. All patients’ follow-up information was collected until disease recurrence or metastases. A WES genetic divergence method for identifying MPLC from IPM was developed. The disease-free survival (DFS) was used as a validation tool for identification reliability. Results: A total of 61 eligible multi-lesion patients with 128 resected non-small cell lung cancer (NSCLC) lesions were finally analyzed. Based on WES genetic divergence identification, 51 MPLC patients were identified, whose DFS was significantly prolonged (HR, 0.27; 95% CI, 0.08 to 0.91; P = 0.02), while other recommended methods for identifying MPLC patients failed to show a DFS advantage. Among MPLC patients, 43.1% (22/51) had a first-degree relatives cancer family history, of whom 54.5% (12/22) had a lung cancer family history. EGFR was still the most common mutation gene (72.5%, 37/51); however, TP53 mutations was less frequent in MPLC (21% vs. 48%, P < 0.05), and tumor mutation burden (TMB) was lower (median, 0.73 muts/Mb vs. 1.15 muts/Mb, P < 0.01). Notably, despite the lack of shared alterations, unique germline variants in cancer-predisposing genes, particularly nonsynonymous SNV in solute carrier gene (SLC) family, were observed in most MPLC patients, rarely overlapping with IPM patients. Conclusions: WES-based genetic divergence clearly distinguishes independent primary tumors from IPM in NSCLC patients. MPLC patients represent a unique population which deserves further study.