PO.BCS01.15 · 生物信息与计算
通过一种新型数据缩放方法增强对HER2阳性乳腺癌中免疫和致癌信号的检测
Enhanced detection of immune and oncogenic signals in HER2-positive breast cancer by a novel data-scaling method
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景：HER2阳性乳腺癌是一种临床上侵袭性的亚型，占所有病例的15-20%，其特征为ERBB2基因扩增和HER2受体过表达，通过PI3K/AKT/mTOR和MAPK/ERK通路驱动致癌信号传导。准确地对肿瘤组织与癌旁正常组织（NAT）进行转录组分析，是发现这些不同机制并指导治疗策略的关键。然而，传统的差异表达（DE）工作流程常因离群的高读段样本以及标准化的局限性而出现数据失真，可能掩盖具有生物学意义的差异。
方法：本研究重新分析了HER2阳性乳腺癌患者的大体RNA-seq数据（GSE292823），以比较传统DE分析与一种新型的基于缩放的预处理方法。
结果：缩放后的数据方法显著改善了PCA中的样本聚类，并增强了对差异表达基因的检测，包括与曲妥珠单抗（trastuzumab）耐药相关的CCL5。IL18、CD86和NOD2的上调提示炎症信号传导、抗原提呈通路和免疫细胞募集，这些都是有助于肿瘤细胞逃避免疫反应的关键因素。此外，若干其他免疫调控基因下调，包括CD3G和GZMA，提示细胞毒性T细胞活性受到抑制且效率降低。SATB-AS1、PICSAR和ROR1-AS1的下调（已知与肺癌及其他类型癌症中染色质重塑改变和长链非编码RNA介导的调控相关）提示HER2阳性乳腺癌中存在类似的调控机制。这些发现证明了这种新型预处理方法在揭示塑造HER2+乳腺癌肿瘤微环境和免疫反应的激活型和抑制型基因网络方面的价值。此外，基因本体（GO）富集分析显示，与原有方法相比，该预处理方法增强了对具有生物学意义通路的检测。传统方法识别出诸如白细胞黏附和T细胞抑制等一般性免疫相关通路，而新方法揭示了若干特定的免疫功能，包括免疫效应过程的调控、免疫突触形成，以及T细胞受体复合物活性的进一步下调。此外，新方法还捕获到囊泡介导信号传导和颗粒相关组分的上调，提示免疫细胞与肿瘤之间的接触增强。
结论：缩放预处理改善了对基因表达差异和显著通路的检测，从而使我们能够更全面地了解HER2阳性乳腺癌中的肿瘤-免疫相互作用。
查看英文原文 English abstract
Background: HER2-positive breast cancer, a clinically aggressive subtype accounting for 15-20% of all cases, is characterized by ERBB2 gene amplification and overexpression of the HER2receptor, which drives oncogenic signaling through PI3K/AKT/mTOR and MAPK/ERKpathways. Accurate transcriptomic profiling of tumor tissue versus normal adjacent tissue (NAT)is key to discovering these distinct mechanisms and guiding therapy strategies. However,traditional differential expression (DE) workflows often suffer from data distortions due tooutlier, high-read samples, and normalization limitations, likely obscuring biologically relevantdifferences.
Methods: This study reanalyzed bulk RNA-seq data from HER2-positive breast cancer patients(GSE292823) to compare traditional DE analysis with a novel scaling-based preprocessingmethod.
Results: The scaled data method significantly improved sample clustering in PCA and enhanceddetection of differentially expressed genes, including CCL5, which is associated with resistanceto trastuzumab. Upregulation of IL18, CD86, and NOD2 indicated inflammatory signaling,antigen presentation pathways, and immune cell recruitment, all key factors that can help tumorcells in evading an immune response. Furthermore, several other immune-regulated genes weredownregulated, including CD3G and GZMA, suggesting the suppression and reduced efficiencyof cytotoxic T-cell activity. The downregulation of SATB-AS1, PICSAR and ROR1-AS1,known to be associated with altered chromatin remodeling and long non-coding RNA-mediatedregulation in lung cancers and other types, pointed to similar regulatory mechanisms in HER2-positive breast cancer. These findings demonstrate the value of the novel preprocessing methodin uncovering both activated and suppressed gene networks that shape the tumormicroenvironment and immune response in HER2+ breast cancer. Additionally, Gene ontology(GO) enrichment analysis revealed that the preprocessing method enhanced the detection ofbiologically relevant pathways compared to the original approach. While the traditional methodidentified general immune-related pathways such as leukocyte adhesion and T cell suppression,the new method uncovered several specific immune functions, including regulation of immuneeffector processes, immunological synapse formation, and further downregulation of the T cellreceptor complex activity. Moreover, the new method captured upregulation of vesicle-mediatedsignaling and granule-associated components, suggesting heightened immune cell-tumorengagement.
Conclusion: The scaled preprocessing improves the detection of gene expression differences andsignificant pathways, thereby allowing for a more comprehensive view of tumor-immuneinteractions in HER2-positive breast cancer.
利益披露 Disclosure
C. Gunadi, None..
E. Xue, None..
A. Lei, None..
Q. Wang, None.