Predictive signatures in breast cancer using multiplexed ion beam imaging

描述（由申请人提供）：导管原位癌（DCIS）是一种浸润前病变，每年约占新发乳腺癌诊断的20%。由于DCIS的病理学评估通常依赖于组织学标准，而这些标准在预测哪些病变将进展为浸润性乳腺癌（IBC）方面几乎没有价值，因此许多患者接受了不必要的手术和辅助干预，这通常会导致治疗相关的并发症。迄今为止，大多数检查乳腺癌活检组织中DCIS蛋白表达的临床研究都采用了使用单一一抗的免疫过氧化物酶染色，使得每种蛋白质在单独的连续活检切片中可视化。因此，尽管许多研究暗示了在IBC的发病机制中的关键作用，但尚未描述将乳腺肿瘤细胞的表型与周围微环境相关的多重定量单细胞谱。考虑到这一点，本文提出的工作重点是使用我最近开发的新型成像平台，多路复用离子束成像（MIBI），以定义新的疾病模型，用于映射与发展DCIS和IBC相关的蛋白质表达和组织组织学的进行性扰动。多路离子束成像（MIBI）能够同时分析多达100个目标，并与标准福尔马林固定，石蜡包埋（FFPE）组织标本和全球临床储存库中最常见的样本类型兼容。我将首先扩展最近的工作，以创建一个抗体面板，用于表征成纤维细胞，巨噬细胞和上皮细胞在人类乳腺组织。该面板将用于询问先前构建的组织微阵列，其由正常乳腺、DCIS和IBC的临床注释活检组成。这些实验将首次在亚细胞水平上同时表征正常和肿瘤性人类乳腺组织的上皮和基质成分的表型和组织学特征。然后将对这些特征进行汇总分析，以构建预测性临床分类器， 可用于选择最佳的治疗方案，防止进展为IBC，同时也最大限度地减少治疗相关的发病率。更一般地说，这项工作还将建立一个新的平台，以获得前所未有的疾病发病机制，可以很容易地适应在未来的工作，以风险分层其他侵入前病变。

DESCRIPTION (provided by applicant): Ductal carcinoma in situ (DCIS) is a pre-invasive lesion that comprises approximately 20% of new breast cancer diagnoses each year. Because pathological evaluation of DCIS typically relies on histologic criteria that offer little value in predicting which lesions will progress to invasive breast cancer (IBC), many patients receive unnecessary surgical and adjuvant interventions that often lead to therapy-related complications. To date, most clinical studies examining protein expression of DCIS in breast cancer biopsies have employed immune-peroxidase staining using a single primary antibody, such that each protein is visualized in separate serial biopsy sections. Consequently, despite numerous studies implicating a critical role in the pathogenesis of IBC, multiplexed, quantitative single cell profiles relating the phenotype of breast tumor cells with the surrounding microenvironment have not been described. With this in mind, the work proposed here focuses on using a novel imaging platform that I recently developed, multiplexed ion beam imaging (MIBI), to define new disease models for mapping progressive perturbations in protein expression and tissue histology that correlate with developing DCIS and IBC. Multiplexed ion beam imaging (MIBI) is capable of analyzing up to 100 targets simultaneously and is compatible with standard formalin-fixed, paraffin-embedded (FFPE) tissue specimens, and the most common sample type in clinical repositories worldwide. I will first expand upon recent work to create an antibody panel for characterizing fibroblasts, macrophages, and epithelium in human breast tissue. This panel will be used to interrogate previously constructed tissue microarrays comprised of clinically annotated biopsies of normal breast, DCIS, and IBC. These experiments will, for the first time, simultaneously characterize phenotypic and histologic features of epithelal and stromal components of normal and neoplastic human breast tissue at the subcellular level. These features will then be analyzed in aggregate to construct predictive clinical classifiers that can be used to select optimal therapeutic regimens that prevent progression to IBC while also minimizing treatment-related morbidity. More generally, this work will also establish a novel platform for gaining an unprecedented view into disease pathogenesis that could be easily adapted in future work to risk stratify other pre-invasive lesions.