胶质母细胞瘤（GBM）是一种不可治愈的脑癌，从诊断开始到平均存活时间少于两年。GBM的标准治疗是多模治疗，包括手术切除、放疗和化疗。然而，预后仍然不佳，迫切需要有效的抗癌药物。由于单个GBM的不同区域包含多种癌症亚群（“肿瘤内异质性”），这很可能是治疗失败的原因，因为某些癌细胞可以逃避免疫监视和治疗威胁。在此，我们使用轨道阱次级离子质谱（OrbiSIMS）技术生成代谢组学数据，以研究高度异质性的肿瘤微环境中的脑肿瘤代谢。我们的结果表明，基于OrbiSIMS的非靶向代谢组学方法能够从福尔马林固定石蜡包埋组织档案中区分形态不同的区域（活细胞、坏死细胞和非癌变细胞）。具体而言，基于一组代谢物包括胞嘧啶、磷酸、嘌呤、黄嘌呤和8-羟基-7-甲基鸟嘌呤，我们可以将坏死区的癌细胞与活细胞的GBM细胞分离出来。此外，我们将坏死和活区域中的普遍代谢物映射到代谢途径中，从而发现色氨酸代谢在GBM细胞的存活中可能是必不可少的。总之，这项研究首次证明了OrbiSIMS在原位研究GBM肿瘤内异质性的能力，所获得的信息有助于改善我们对癌症代谢的理解，并发展出能够有效地针对肿瘤内多个亚群的新疗法。

Glioblastoma (GBM) is an incurable brain cancer with a median survival of less than two years from diagnosis. The standard treatment of GBM is multimodality therapy comprising surgical resection, radiation, and chemotherapy. However, prognosis remains poor, and there is an urgent need for effective anticancer drugs. Since different regions of a single GBM contain multiple cancer subpopulations ("intra-tumor heterogeneity"), this likely accounts for therapy failure as certain cancer cells can escape from immune surveillance and therapeutic threats. Here, we present metabolomic data generated using the Orbitrap secondary ion mass spectrometry (OrbiSIMS) technique to investigate brain tumor metabolism within its highly heterogeneous tumor microenvironment. Our results demonstrate that an OrbiSIMS-based untargeted metabolomics method was able to discriminate morphologically distinct regions (viable, necrotic, and non-cancerous) within single tumors from formalin-fixed paraffin-embedded tissue archives. Specifically, cancer cells from necrotic regions were separated from viable GBM cells based on a set of metabolites including cytosine, phosphate, purine, xanthine, and 8-hydroxy-7-methylguanine. Moreover, we mapped ubiquitous metabolites across necrotic and viable regions into metabolic pathways, which allowed for the discovery of tryptophan metabolism that was likely essential for GBM cellular survival. In summary, this study first demonstrated the capability of OrbiSIMS for in situ investigation of GBM intra-tumor heterogeneity, and the acquired information can potentially help improve our understanding of cancer metabolism and develop new therapies that can effectively target multiple subpopulations within a tumor.