Objective: To screen out the potential gene biomarkers to predict responses to neoadjuvant chemoradiotherapy (CRT) in patients with rectal cancer and to explore the main downstream pathways of resistance. Methods: The gene expression profiles (GSE35452) of locally advanced rectal cancer undergoing neoadjuvant chemoradiotherapy from 46 specimens (24 responders, TRG 0/1, and 22 non-responders, TRG 2/3) were downloaded from the GEO database. The differentially expressed genes were identified to screen out the potential biomarkers by use of the GCBI platform. GO and KEGG pathways enrichment analysis were performed to integrate enrichment results of differentially expressed genes. Signal-signal interaction network was constructed and analyzed to screen out potential main downstream pathways. Results: A total of 1079 differentially expressed genes were screened, including 657 up-regulated and 422 down-regulated ones. Among these genes, REG4 had the maximum fold change value of -6.029 491. In GO term, these differentially expressed genes were mainly enriched in molecule metabolic process, cell cycle, DNA-dependent transcription, signal transduction and apoptotic process. The KEGG pathways enrichment analysis showed that the differentially expressed genes were enriched in 65 KEGG pathways, including metabolic pathways, cell cycle and metabolism pathways. Signal-signal interaction network analysis showed that MAPK signaling pathway and cell cycle pathway might play a determinant role in the development of neoadjuvant chemoradiotherapy resistance. Further analysis showed that CDKN1B, CDKN2A, RBL1, TFDP1, CCND2, CCNE2, CDC6 and CDK6 in cell cycle might induce chemoradiotherapy resistance by blocking G1/S phase cell cycle arrest, decreasing the apoptosis of tumor cells and increasing S phase ratio of chemoradiotherapy resistance. Conclusion: G1/S phase cell cycle arrest blocking plays an important role in the development of chemoradiotherapy resistance in patients with rectal cancer. Moreover, the key genes, such as REG4, may be useful in predicting responses to neoadjuvant chemoradiotherapy.
目的： 筛选直肠癌新辅助放化疗（CRT）疗效预测的分子标记物，分析调节放化疗抗性的下游核心信号通路。 方法： 从GEO数据库下载一组直肠癌新辅助放化疗患者的芯片数据（GSE35452），其中，反应良好组（TRG 0/1级）24例，反应不良组（TRG 2/3级）22例。运用GCBI平台，分析两组间的差异表达基因，进行分子标记物筛选。进一步运用GO分析对差异基因进行显著性的功能分析，并对差异基因所参与的信号通路（pathway）进行显著性的信号通路分析。然后基于图论的方法，以信号通路为研究单元，基于KEGG数据库中的相互作用关系，构建信号通路间相互作用网络。 结果： 共筛选出1 079个差异表达基因。其中657个基因上调，422个基因下调，差异倍数最大的基因为REG4基因。GO富集分析结果提示，这些基因富集在分子及物质代谢、细胞周期、DNA转录、信号转导、细胞凋亡等生物学进程的分子功能上。信号通路分析结果提示，差异基因共富集在65条信号通路上，包括物质代谢、细胞周期、细胞增殖等代谢通路。信号通路间相互作用网络分析提示：MAPK信号通路和细胞周期信号通路不仅是影响直肠癌放化疗抗性的核心通路，同时也是主要下游效应通路。进一步比对发现，共13个差异基因富集于细胞周期通路，其中8个基因（CDKN1B、CDKN2A、RBL1、TFDP1、CCND2、CCNE2、CDC6、CDK6）参与了G1/S期的调节，对上下调关系进行梳理，推断这些基因可能通过抑制G1/S期阻滞，减少肿瘤细胞凋亡，并通过增加放疗抵抗的S期比率，导致CRT抗性的发生。 结论： 通过生物信息分析，首次提出G1/S期去阻滞在直肠癌CRT抗性研究方面的潜在价值，并发现了REG4等一系列潜在新型CRT抗性预测分子标记物。.
Keywords: Biomarkers; Neoadjuvant chemoradiotherapy; Pathways; Rectal neoplasms.