FAP介导的肿瘤相关成纤维细胞功能在三阴性乳腺癌免疫微环境及治疗应答中的作用

Role of FAP-mediated cancer-associated fibroblast function in the immune microenvironment and therapeutic response of triple-negative breast cancer

摘要:

背景与目的 三阴性乳腺癌（TNBC）是一种临床行为高度恶性的乳腺癌亚型，免疫检查点抑制剂（ICB）联合化疗已成为其新辅助治疗的重要策略。然而，ICB疗效差异显著，相关免疫微环境机制尚未明晰。肿瘤相关成纤维细胞（CAF）通过调控肿瘤微环境（TME）影响免疫治疗反应，其中成纤维细胞活化蛋白（FAP）为CAF的关键标志物，其在TNBC中的免疫功能尚缺乏系统研究。本研究旨在揭示FAP在TNBC中的表达特征、对免疫微环境及ICB疗效的影响，并探讨其可能的免疫抑制机制及临床应用潜力。方法 综合利用TCGA和I-SPY2临床试验数据，分析FAP表达与TNBC患者预后、免疫细胞浸润及免疫检查点分子表达之间的关系。通过CIBERSORT算法、GSEA通路富集分析及差异表达分析（DESeq2），评估FAP相关免疫微环境特征，并在转录组层面筛选其下游相关基因。构建FAP过表达与敲低的CAF模型，联合CD8⁺ T细胞共培养，检测其对CD8⁺ T细胞活性和凋亡的调控效应。进一步通过qPCR与Western blot验证转录组分析中发现潜在靶基因的表达变化。结合I-SPY2数据与湘雅医院TNBC病例的多重免疫荧光病理样本与临床资料，分析FAP表达水平与新辅助免疫治疗病理完全缓解率（pCR）之间的关系及其预测价值。结果 FAP在TNBC肿瘤组织中明显高于正常组织，高表达患者总生存期（OS）明显缩短。多因素Cox分析确认FAP为TNBC患者OS的独立不良预后因子。FAP高表达与CD8⁺ T细胞、自然杀伤细胞、滤泡辅助性T细胞浸润减少明显相关，同时伴随CD276、TIM-3、PD-L2等免疫检查点分子上调。CAF模型中，FAP过表达可抑制CD8⁺ T细胞活性并促进其凋亡，敲低FAP则产生相反效应。转录组分析显示FAP高表达组中COL1A1等多种胶原蛋白基因明显上调，且与FAP表达呈正相关；qPCR与Western blot实验证实FAP可正向调控COL1A1表达。I-SPY2数据分析表明，FAP低表达TNBC患者在接受帕博利珠单抗联合化疗后pCR率明显高于FAP高表达组，湘雅临床病理队列亦验证FAP高表达患者CD8⁺ T细胞浸润减少，且pCR率更低，受试者工作特征分析曲线下面积达0.857。结论 FAP高表达CAF通过促进COL1A1分泌、抑制CD8⁺ T细胞功能及上调免疫检查点分子，塑造免疫抑制性TME，导致TNBC患者预后不良及免疫治疗应答下降。FAP不仅可作为TNBC患者生存及疗效的预测生物标志物，亦为未来联合靶向治疗的潜在干预靶点。

Abstract:

Background and Aims Triple-negative breast cancer (TNBC) is a clinically aggressive subtype of breast cancer with limited treatment options. Immune checkpoint blockade (ICB) combined with chemotherapy has emerged as a key neoadjuvant therapeutic strategy for TNBC. However, significant variability in ICB efficacy exists among patients, and the underlying mechanisms related to the tumor immune microenvironment (TME) remain unclear. Cancer-associated fibroblasts (CAFs), as major stromal components, regulate TME and influence immunotherapy responses. Fibroblast activation protein (FAP), a key marker of CAFs, has been associated with poor prognosis in multiple solid tumors, yet its immunological role in TNBC has not been systematically investigated. This study aims to elucidate the expression pattern of FAP in TNBC, its impact on the immune microenvironment and ICB efficacy, and to explore its potential immunosuppressive mechanisms and clinical implications.Methods The data from TCGA and the I-SPY2 clinical trial were integrated to assess the association of FAP expression with prognosis, immune cell infiltration, and immune checkpoint molecule expression in TNBC. Immune landscape profiling was conducted using CIBERSORT, GSEA enrichment analysis, and differential gene expression analysis (DESeq2) to characterize the immune features associated with FAP expression and to identify downstream genes at the transcriptomic level. CAF models with FAP overexpression or knockdown were constructed and co-cultured with CD8? T cells to evaluate FAP's regulatory effects on CD8? T cell activity and apoptosis. The expression of COL1A1, a potential FAP-regulated gene identified from transcriptomic analysis, was validated using qPCR and Western blot. Finally, clinical data and multiplex immunofluorescence pathology samples from TNBC patients at Xiangya Hospital were analyzed alongside I-SPY2 data to determine the predictive value of FAP expression for pathological complete response (pCR) following neoadjuvant immunotherapy.Results FAP was significantly upregulated in TNBC tumor tissues compared to normal tissues and associated with shorter overall survival. Multivariate Cox regression analysis identified FAP as an independent adverse prognostic factor. High FAP expression was correlated with reduced infiltration of CD8? T cells, NK cells, and Tfh cells, as well as upregulation of immune checkpoints including CD276, TIM-3, and PD-L2. In CAF models, FAP overexpression suppressed CD8? T cell activity and promoted apoptosis, while FAP knockdown had the opposite effect. Transcriptomic analysis showed that COL1A1 and other collagen-related genes were significantly upregulated in the FAP-high group and positively correlated with FAP expression; qPCR and Western blot confirmed that FAP positively regulates COL1A1 expression. Analysis of I-SPY2 data revealed that FAP-low patients receiving pembrolizumab plus chemotherapy had significantly higher pCR rates compared to FAP-high patients. Consistently, clinical data from the Xiangya cohort showed reduced CD8? T cell infiltration and lower pCR rates in FAP-high patients, with a ROC AUC of 0.857 for predicting treatment response.Conclusion FAP-high CAFs contribute to the formation of an immunosuppressive TME in TNBC by promoting COL1A1 secretion, inhibiting CD8? T cell function, and upregulating immune checkpoint molecules. High FAP expression is associated with poor prognosis and reduced response to immunotherapy, highlighting FAP as both a prognostic biomarker and a potential therapeutic target for stratified and combination treatment strategies in TNBC.