单独的新辅助化疗（NAC）或与靶向治疗联合代表了局部三阴性乳腺癌（TNBC）的护理标准。然而，只有一小部分患者有反应，因此需要更好地了解 TNBC 生态系统中建立连续和多维相互作用的复杂元素。解决这种复杂性需要新的空间定义方法。在这里，我们使用空间转录组学来研究诊断时 TNBC 的多维组织，并探索每个细胞成分对 NAC 反应的贡献。从一系列连续回顾性 TNBC 病例开始，我们设计了一项病例对照研究，包括 24 例 TNBC 患者，其中 12 例在 NAC 后经历了病理完全缓解 (pCR)，12 例无缓解或进展 (pNR)。对 200 多个感兴趣区域 (ROI) 进行了分析。我们的计算方法描述了一个概括临床治疗反应的模型。这些数据在独立的患者队列中得到了验证。比较 pCR 患者和 pNR 患者，在肿瘤、基质和免疫浸润中检测到转录程序的差异。在 pCR 中，观察到肿瘤块和浸润淋巴细胞之间的空间污染，并通过干扰素信号传导的大量激活来维持。相反，pNR 病变显示促血管生成信号和氧代谢增加。在基质中仅观察到适度的差异，揭示了免疫浸润的基于拓扑的功能异质性。因此，空间转录组学提供了有关 TNBC 多维性的基本信息，并可以有效预测肿瘤行为。这些结果为 TNBC 治疗方法的改进和个性化开辟了新的视角。

Neoadjuvant chemotherapy (NAC) alone or combined with target therapies represents the standard of care for localized triple-negative breast cancer (TNBC). However, only a fraction of patients have a response, necessitating better understanding of the complex elements in the TNBC ecosystem that establish continuous and multidimensional interactions. Resolving such complexity requires new spatially-defined approaches. Here, we used spatial transcriptomics to investigate the multidimensional organization of TNBC at diagnosis and explore the contribution of each cell component to response to NAC. Starting from a consecutive retrospective series of TNBC cases we designed a case-control study including 24 TNBC patients of which 12 experienced a pathological complete response (pCR) and 12 no-response or progression (pNR) after NAC. Over 200 regions of interest (ROIs) were profiled. Our computational approaches described a model that recapitulates clinical response to therapy. The data were validated in an independent cohort of patients. Differences in the transcriptional program were detected in the tumor, stroma, and immune infiltrate comparing patients with a pCR with those with pNR. In pCR, spatial contamination between the tumor mass and the infiltrating lymphocytes was observed, sustained by a massive activation of Interferon-signaling. Conversely, pNR lesions displayed increased pro-angiogenetic signaling and oxygen-based metabolism. Only modest differences were observed in the stroma, revealing a topology-based functional heterogeneity of the immune infiltrate. Thus, spatial transcriptomics provides fundamental information on the multidimensionality of TNBC and allows an effective prediction of tumor behavior. These results open new perspectives for the improvement and personalization of therapeutic approaches to TNBCs.