Table1_Identification of a 5-lncRNA-Based Signature for Immune Characteristics and Prognosis of Lung Squamous Cell Carcinoma and Verification of the Function of lncRNA SPATA41.DOCX

Lung squamous cell carcinoma (LUSC) is one of the most lethal cancers worldwide. Traditional tumor-node-metastasis (TNM) staging system has many insufficiencies in predicting immune characteristics, overall survival (OS), and prognosis of LUSC. LncRNA is currently found involved in tumor development and effectively predicts tumor prognosis. We screened potential tumor-related lncRNAs for immune characteristics and constructed a nomogram combining lncRNA and traditional clinical indicators for prognosis prediction. We obtained the large-scale gene expression profiles of samples from 492 LUSC patients in The Cancer Genome Atlas database. SPATA41, AL034550.2, AP003721.2, AC106786.1, and AC078889.1 were finally screened to construct a 5-lncRNA-based signature. The risk score of the signature divided patients into subgroups of high-risk and low-risk with significant differences in OS. Their area under the curve (AUC) reached more than 0.70 in 1, 3, and 5 years. In addition, compared with the high-risk subgroup, the low-risk subgroup exhibited a remarkably favorable prognosis and TME score, along with a higher immune infiltration score and lower TIDE score. The signature also significantly related to chemotherapy response, especially in cisplatin, vinorelbine, and paclitaxel. Importantly, the nomogram we constructed had good reliability with the assessment of the calibration chart and consistency index (c-index). GO and KEGG enrichment analysis indicated that co-expression mRNAs of the 5 lncRNAs were mainly focused on RNA splicing, DNA replication, and protein serine/threonine kinase activity. Functional assays demonstrated that SPATA41, one of the five OS-related lncRNAs, regulated invasion, migration, proliferation, and programmed death in vitro. In summary, our 5-lncRNA-based signature has a good performance in predicting immune characteristics and prognosis of LUSC patients.

肺鳞状细胞癌（Lung squamous cell carcinoma, LUSC）是全球致死率最高的恶性肿瘤之一。传统肿瘤-淋巴结-转移（tumor-node-metastasis, TNM）分期系统在预测LUSC的免疫特征、总生存期（overall survival, OS）及预后方面存在诸多不足。长链非编码RNA（long non-coding RNA, lncRNA）目前被发现参与肿瘤发生发展过程，并可有效预测肿瘤预后。本研究筛选出与免疫特征相关的潜在肿瘤相关lncRNA，构建了联合lncRNA与传统临床指标的列线图用于预后预测。我们从癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库中获取了492例LUSC患者样本的大规模基因表达谱。最终筛选出SPATA41、AL034550.2、AP003721.2、AC106786.1及AC078889.1，构建了基于5个lncRNA的风险特征模型。该模型通过风险评分将患者分为高风险亚组与低风险亚组，两组患者的OS存在显著差异。其在1年、3年及5年的曲线下面积（area under the curve, AUC）均超过0.70。此外，与高风险亚组相比，低风险亚组患者的预后更佳、肿瘤微环境（tumor microenvironment, TME）评分更高，同时免疫浸润评分更高、肿瘤免疫功能异常和排除（Tumor Immune Dysfunction and Exclusion, TIDE）评分更低。该风险特征模型还与化疗响应显著相关，尤其针对顺铂、长春瑞滨及紫杉醇治疗。重要的是，我们构建的列线图经校准曲线及一致性指数（consistency index, c-index）评估后，展现出良好的可靠性。基因本体（Gene Ontology, GO）与京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）富集分析结果显示，这5个lncRNA的共表达mRNA主要富集于RNA剪接、DNA复制及蛋白丝氨酸/苏氨酸激酶活性相关通路。功能实验证实，5个与OS相关的lncRNA之一SPATA41可在体外调控肿瘤细胞的侵袭、迁移、增殖及程序性死亡。综上，本研究构建的基于5个lncRNA的风险特征模型在预测LUSC患者的免疫特征与预后方面表现良好。