Predicting colon cancer metastasis through spatial molecular characterization of the tumor immune microenvironment

结直肠癌（CRC）是第三大常见癌症，也是美国癌症相关死亡的原因。 美国的在手术切除时检查腋窝淋巴结对于淋巴结转移是必要的 虽然最大化评估的淋巴结数量很重要，但最近基于人群的研究 显示对淋巴结受累的评估通常是不完整或不充分的。这可能会影响准确性 肿瘤分期和下游疾病管理选项，例如患者是否应该接受辅助治疗 化疗开发通过间接评估淋巴结受累的替代评估方法 在切除不充分的情况下，机制将是有启发性的。肿瘤浸润淋巴细胞（TIL）和 其他免疫细胞类型是CRC的重要预后指标。TILs的类型、密度和位置 关于肿瘤，除了肿瘤特异性体细胞改变谱之外，可以确定TIL对预后的影响。 此外，免疫细胞内的空间依赖性、免疫细胞特异性、蛋白质组学和转录组学表达模式， 肿瘤免疫微环境（Tumor Immune Microenvironment，TIME）是一种免疫系统，可以识别肿瘤周围的协调免疫反应， 肿瘤转移使用高度多重空间组学可以对TILs进行全面表征 技术，但高成本和低吞吐量阻止了它们的临床部署。虚拟染色可以推断分子 在形态学允许的情况下，以低成本从组织组织学获得信息。我们的目标是设计一个低成本的虚拟 从高度多重空间分子信息中提取的染色试验，可以补充手术淋巴液 用于复发风险评估的淋巴结解剖并与其他新兴预测因子（例如，循环肿瘤 DNA）。在一组有或没有淋巴结和/或远处转移的III期肿瘤中，我们将鉴定空间蛋白质组， 转移的全转录组标志物与数字空间分析和Visium空间转录组学， 免疫细胞我们还将评估上游细胞类型特异性DNA甲基化改变伴随空间 建筑时间的变化。将通过成本较低的多重免疫荧光技术验证已识别的标记物 染色最后，我们将建立与已确定的空间转移标志物的组织学对应关系， 虚拟染色算法，将H& E染色的组织转换为经验证的多重免疫荧光和完整的 转录组标记。分子标记物的空间和细胞类型特异性模式表明患者是否 已经或可能发生转移的患者将在此框架下进行鉴定。从组织中推断出这些信息 形态学可以为淋巴结切除提供低成本和高度可解释的辅助分子评估， 预测复发风险和对辅助化疗的反应。我们希望我们的发现能提供 R 01临床试验的初步数据，以前瞻性地比较识别的标志物与独立转移 预测器（例如，液体活组织检查）评估患者预后和治疗选择的能力。

Colorectal Cancer (CRC) is both the third most common form of cancer and cause of cancer-related deaths in the United States. Examination of axillary lymph nodes at the time of surgical resection is essential for prognostication and while it is important to maximize the number of lymph nodes assessed, recent population-based studies have shown that evaluation of lymph node involvement is usually incomplete or inadequate. This can impact the accuracy of tumor staging and downstream disease management options, such as whether the patient should receive adjuvant chemotherapy. Developing alternative assessment methods which assess lymph node involvement through indirect mechanisms would be illuminating in cases where resection is inadequate. Tumor-infiltrating lymphocytes (TIL) and other immune cell types are important prognostic indicators in CRC. The type, density, and location of TILs with respect to the tumor, in addition to tumor-specific somatic alteration profiles, can determine TIL's effect on prognosis. Furthermore, spatially dependent, immune cell specific, proteomic and transcriptomic expression patterns inside and around tumor – the Tumor Immune Microenvironment (TIME) – can discern the coordinated immune response to tumor metastasis. The comprehensive characterization of TILs is possible using highly multiplexed spatial omics technologies, but high cost and low throughput prevent their clinical deployment. Virtual staining can infer molecular information at low cost from tissue histology where the morphology allows. We aim to design a low-cost Virtual Staining test, distilled from highly multiplexed spatial molecular information, that could complement surgical lymph node dissection for recurrence risk assessments and compete with other emerging predictors (e.g., circulating tumor DNA). In a set of stage III tumors with or without nodal and/or distant metastases, we will identify spatial proteomic and whole transcriptomic markers of metastasis with digital spatial profiling and Visium spatial transcriptomics of immune cells. We will also assess upstream cell-type specific DNA methylation alterations concomitant with spatial architectural TIME changes. Identified markers will be validated through lower-cost multiplexed immunofluorescence staining. Finally, we will establish histological correspondence to identified spatial metastasis markers and develop virtual staining algorithms to convert H&E-stained tissue into validated multiplexed immunofluorescent and whole transcriptomic markers. Spatial and cell-type specific patterns of molecular markers that indicate whether a patient has or is likely to develop metastasis will be identified under this framework. Inferring such information from tissue morphology can provide a low-cost and highly interpretable adjunct molecular assessment for lymph node resection, to predict recurrence risk and response to adjuvant chemotherapy. We expect that our findings will provide preliminary data for an R01 clinical trial to compare identified markers prospectively to independent metastasis predictors (e.g., liquid biopsy) for their ability to assess patient prognosis and treatment options.