INDIVIDUALIZING COLON CANCER THERAPY USING HYBRID RNA AND DNA MOLECULAR SIGNATURE

描述（由申请人提供）：预测哪些癌症患者对治疗反应最好是一个巨大的卫生保健问题。根据对全外显子组测序数据的早期解读，最近有人提出，可能只有约12种分子途径驱动癌症的发生和进展。现在许多实验性疗法都针对这些途径。我们认为，基因表达特征可能是判断特定分子途径激活的最佳方法之一，通过提供许多基因活性的“分子总结”。此外，我们认为选择性增加突变评估可能会提高混合多分析物（DNA + RNA）测试的分辨能力，可用于指导患者选择最有效的药物。我们最近开发了基因表达特征来测量结肠癌中RAS和PI3K两种最重要的通路的激活，对于这两种通路靶向治疗的可用性越来越高。由于这些途径的复杂性，对典型单基因突变的简单分析只能识别出一部分（<30%）人群的反应特征。在这里，我们建议从技术上验证现有的RAS和PI3K签名，并通过新的突变评估来完善它们的活性。多分析物特征/算法将在CLIA环境中与接受西妥昔单抗治疗的结直肠癌患者队列进行临床验证。该方法将在不久的将来为临床应用准备签名。

DESCRIPTION (provided by applicant): Predicting which cancer patients will best respond to therapy is an enormous health care issue. It has been recently suggested, based on early reads of whole exome sequencing data, that there may only be ~12 molecular pathways that drive the development and progression of cancer. Many experimental therapies are now targeting these pathways. We believe that gene expression signatures may be one of the best ways to judge the activation of a particular molecular pathway, by providing a "molecular summary" of the activity of many genes. Moreover, we believe that the selective addition of mutational assessment may improve the resolving power of a hybrid, multi-analyte (DNA + RNA) test that may be used to guide patients to the most effective drugs. We have recently developed gene expression signatures to measure the activation of two of the most important pathways in colon cancer, RAS and PI3K, for which there is an increasing availability of pathway targeted therapeutics. Due to the complex nature of these pathways, simple analysis of canonical single gene mutations only identifies the response characteristics of a proportion (<30%) of the population. Here, we propose to technically validate the existing RAS and PI3K signatures and to refine their activity through novel mutational assessment. Multi-analyte signatures/ algorithms will be clinically validated in a CLIA environment with a cohort of colorectal cancer patients treated with cetuximab therapy. This approach will prepare signatures for clinical application in the near future. PUBLIC HEALTH RELEVANCE: This proposal seeks to identify a reliable means to identify the right drugs for the right cancer patients. Current approaches over treat many patients to help an unknown few, with substantial costs and toxicities. The application of molecular signatures to individualize therapy holds significant promise to personalize cancer care, improve response rates, reduces toxicity and associated costs. Here we will combine RNA gene expression signatures with gene mutation assessments to identify responders and non-responders to cetuximab therapy.