基于单细胞水平RNA指纹谱的肺腺癌软脑膜转移患者脑脊液循环肿瘤细胞检测及种植机制研究

软脑膜转移（LUAD-LM）是肺腺癌严重的并发症之一，患者预后极差。LUAD-LM患者脑脊液中的肿瘤细胞存在高度异质性导致细胞学检测敏感性差，早期诊断困难；肿瘤细胞在中枢神经系统种植的分子机制不清。我们前期采用单细胞RNA测序发现5例患者中1776个脑脊液循环肿瘤细胞（CSF-CTC）的高度异质性转录组特征，其高表达肺腺癌和上皮标记，且细胞代谢、黏附相关通路显著活跃。通过比较LUAD-LM患者CSF-CTC和外周血CTC测序结果，发现CSF-CTC中存在CLDN7特异性表达，且CLDN7表达增加了细胞的侵袭性。我们拟采用留一交叉验证法构建多基因RNA指纹谱用于CSF-CTC检测，评估该方法在LM诊断和预后中的价值；同时探索CLDN7促进CTC表面整合素锚定表达，启动LUAD-LM种植的分子机制。本项目将为LUAD-LM的早期诊断和预后评估提供新方案，并为LUAD-LM的精准干预提供新靶点。

Leptomeningeal metastasis is a major complication of advanced lung adenocarcinoma with unfavored prognosis. Tumoral heterogeneity is the major obstacle for the diagnosis and treatment of lung adenocarcinoma leptomeningeal metastases (LUAD-LM) and elucidating the mechanism of tumor LM colonization. By single cell sequencing we obtained the unique transcriptome from thousands of cerebrospinal fluid (CSF) circulating tumor cells (CTC), normal CSF white blood cells, and peripheral blood-CTC, and identified CLDN7 as the CSF-CTC specifically expressed gene, which is associated with tumor cell invasion. In this study, we will specify a RNA fingerprint panel for CSF-CTC using the single cell level transcriptome and further establish a CTC scoring system by leave-one-out cross validation based machine learning to improve the diagnosis and prognosis of LUAD-LM. We will also explore the mechanism behind CLDN7/integrin overexpression induced CSF-CTC colonization. This project will provide a novel for the early diagnosis of LAUD-LM and spotlight the driver molecular for LUAD-LM development.