Analyzing mRNAsi-Related Genes Identifies Novel Prognostic Markers and Potential Drug Combination for Patients with Basal Breast Cancer.
Analyzing mRNAsi-Related Genes Identifies Novel Prognostic Markers and Potential Drug Combination for Patients with Basal Breast Cancer.
复制标题
DOI：
10.1155/2021/4731349
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Liu H
中科院分区：
文献类型：
--
作者：
Huang K;Wu Y;Xie Y;Huang L;Liu H
基底层乳腺癌亚型是所有乳腺癌亚型中预后最差的亚型。近年来，一种新的肿瘤干度指数mRNAsi被发现能够衡量组织的致癌分化程度。参与多种癌症过程的mRNAsi源自一类逻辑回归（Oscillation）机器学习算法对各种干细胞和肿瘤细胞的全基因组表达的创新应用。然而，基底乳腺癌中的mRNAsi在很大程度上是未知的。在这里，我们发现基底乳腺癌在所有四种乳腺癌亚型中携带最高的mRNAsi，特别是385个mRNAsi相关基因与基底乳腺癌的高mRNAsi值呈正相关。这种高mRNAsi也与基底乳腺癌中的活跃细胞周期、DNA复制和代谢重编程密切相关。有趣的是，在385个基因中，TRIM59，SEPT3，RAD51AP1和EXO1可以作为独立的保护性预后因素，但CTSF和ABHD4B可以作为基底乳腺癌患者的独立不良预后因素。值得注意的是，我们建立了一个强大的预后模型，包含6个mRNAsi相关基因，可以有效地预测基底乳腺癌亚型患者的生存率。最后，药物敏感性分析显示，某些药物组合可能通过靶向mRNAsi相关基因而有效地治疗基底乳腺癌。综上所述，我们的研究不仅确定了基底乳腺癌的新的预后生物标志物，而且通过建立mRNAsi相关的预后模型提供了药物敏感性数据。
Basal breast cancer subtype is the worst prognosis subtypes among all breast cancer subtypes. Recently, a new tumor stemness index-mRNAsi is found to be able to measure the degree of oncogenic differentiation of tissues. The mRNAsi involved in a variety of cancer processes is derived from the innovative application of one-class logistic regression (OCLR) machine learning algorithm to the whole genome expression of various stem cells and tumor cells. However, it is largely unknown about mRNAsi in basal breast cancer. Here, we find that basal breast cancer carries the highest mRNAsi among all four subtypes of breast cancer, especially 385 mRNAsi-related genes are positively related to the high mRNAsi value in basal breast cancer. This high mRNAsi is also closely related to active cell cycle, DNA replication, and metabolic reprogramming in basal breast cancer. Intriguingly, in the 385 genes, TRIM59, SEPT3, RAD51AP1, and EXO1 can act as independent protective prognostic factors, but CTSF and ABHD4B can serve as independent bad prognostic factors in patients with basal breast cancer. Remarkably, we establish a robust prognostic model containing the 6 mRNAsi-related genes that can effectively predict the survival rate of patients with the basal breast cancer subtype. Finally, the drug sensitivity analysis reveals that some drug combinations may be effectively against basal breast cancer via targeting the mRNAsi-related genes. Taken together, our study not only identifies novel prognostic biomarkers for basal breast cancers but also provides the drug sensitivity data by establishing an mRNAsi-related prognostic model.