Identification of Leptomeningeal Metastasis-Specific Exosomal Mirna Signature in Cerebrospinal Fluid of Non-Small-cell Lung Cancer Patients.

Ben-Yuan Jiang,Yangsi Li,Xue Wu,Hua Bao,Yan Ding,Jinji Yang,Xuchao Zhang,Xuening Yang,Wenzhao Zhong,Qing Zhou,Haiyan Tu,Shuyu Wu,Yang Shao,Cunyi Gao,Yi-Long Wu
DOI: https://doi.org/10.1200/jco.2018.36.15_suppl.e21024
IF: 45.3
2018-01-01
Journal of Clinical Oncology
Abstract:e21024 Background: Leptomeningeal metastasis (LM) is a devastating complication with poor prognosis in non-small-cell lung cancer (NSCLC) patients. Exosomes are extracellular vesicles in body fluids enriched with microRNAs (miRNAs), which have been implicated in brain metastasis. Here, we aimed to identify exosomal miRNA signatures in NSCLC patients with LM and elucidate the potential LM mechanism. Methods: We prepared exosomes from cerebrospinal fluid (CSF) samples of 29 advanced NSCLC patients with (n = 22, LM+) or without (n = 7, LM-) LM as well as 4 non-cancer individuals. Exosomal miRNA profiles obtained from next-generation sequencing were subjected for differential expression analysis and signature discovery. Results: Unsupervised hierarchical clustering of the miRNA expression profiles clearly separated CSF samples into LM+ and LM- groups. Interestingly, these samples were stratified based on their LM status only, regardless of their intraparenchymal metastasis status. In total, 181 (129 up and 52 down-regulated) miRNAs were identified differentially presented in the LM+ CSF samples compared to the LM- group. Predicted targets of up and down-regulated miRNAs were significantly enriched in the neurotrophin/MAPK and PI3K-AKT signaling pathway, respectively. Top altered miRNAs include dramatically up-regulated miR-200 family members and down-regulated miR-144/451 cluster, which have been previously reported to be involved in metastasis. Using machine learning, we identified a signature of 8 CSF exosome miRNAs (let-7e-5p, miR-142-5p, miR-148a-3p, miR-200b-3p, miR-451a, miR-483-5p, miR-486-5p, and miR-96-5p) for classification of LM+ patients with 97% sensitivity and 100% specificity. Interestingly, one CSF sample, which was taken 321 days before LM was confirmed in clinic, was predicted as LM+ based on our signature. Conclusions: NSCLC patients with LM present a remarkably distinct CSF exosomal miRNA signature, which may involve in the progression of LM, and can be potentially used as diagnostic biomarkers for LM.
What problem does this paper attempt to address?