Integrative prediction models for metastasis risk in colon cancer

描述(申请人提供)：虽然第三阶段结肠癌的辅助化疗可带来总体生存效益，但42%至44%的第三阶段患者即使没有辅助治疗也不会在五年内复发。相反，临床试验未能证明辅助化疗对II期结肠癌的整体益处；然而，部分高危患者可能从辅助治疗中受益。因此，迫切需要一种准确可靠的方法来确定复发风险，以及相应的全身治疗受益的可能性。我们最近根据小鼠结肠癌细胞的转移生物学特性开发了一种新的基因表达标记(“34基因标记”)，能够将结肠癌患者分为低复发风险组和高复发风险组。这样的基因标记可能允许更适当地选择接受或不接受辅助化疗的患者，从而使那些风险较低的III期患者能够避免系统治疗的潜在发病率、偶尔死亡和确定的经济费用，同时提高II期患者的存活率。为了准确识别个人风险，临床、人口统计学、病理学和体细胞突变数据都必须与基因表达相结合；为了结合这些不同的数据，我们建议进行研究，以开发一个综合的转移风险预测模型，允许整合不同类型的数据。我们将验证这一模型的预测能力，并在未来五年探索平台，前瞻性地应用这一工具作为指导结肠癌患者治疗决定的方法。这项重点研究将在临床试验之前，在相对较短的时间内将我们的分子研究成果转化为临床应用。这项建议的长期目标是从不同类型的数据中开发出一个临床上有用的转移评分，该评分可以应用于II期和III期结肠癌患者，以降低结肠癌和结肠癌治疗的死亡率、发病率和相关成本。为此，我们的具体目标是：1)建立结肠癌综合转移风险预测模型；2)确定34基因转移评分临床测试的最佳平台；3)在档案组织上盲法测试优化的预后标志，以确定测试的稳健性，以及是否应该将转移风险评分推进到前瞻性临床试验，以预测II期和III期患者的预后。

DESCRIPTION (provided by applicant): While adjuvant chemotherapy for stage III colon cancer results in an overall survival benefit, 42 to 44% of stage III patients will not recur in five years even without adjuvant treatment. Conversely, clinical trials have failed to demonstrate an overall benefit of adjuvant chemotherapy for stage II colon cancer; however, a subset of high-risk patients may benefit from adjuvant treatment. Thus, an accurate and reliable method of determining risk of recurrence, and corresponding likelihood of benefit of systemic therapy, is greatly needed. We have recently developed a novel gene expression signature ("34-gene signature") based on the metastatic biology of mouse colon cancer cells, capable of segregating colon cancer patient groups into low and high risk of recurrence. A gene signature such as this may allow for more appropriate selection of patients to receive or not receive adjuvant chemotherapy, thus enabling those phase III patients at low risk to avoid the potential morbidity, occasional mortality, and definite financial expense of systemic therapy, while improving the survival of phase II patients. To accurately identify personal risk, clinical, demographic, pathologic, and somatic mutation data all must be incorporated with gene expression; to incorporate these various data, we propose to conduct research to develop an integrative metastatic risk prediction model that allows for integration of diverse types of data. We will validate the predictive power of this model and, over the next five years, explore platforms to apply this tool prospectively as an approach to guide treatment decisions in colon cancer patients. This focused study will translate our molecular findings to clinical application in a relatively short time, in advance of clinical trials. The long-term goal for this proposal is to develop a clinically useful metastasis score from diverse type of data that can be applied to stage II and III colon cancer patients for the purpose of reducing mortality, morbidity, and the cost associated with colon cancer and colon cancer treatment. To this end, our specific aims are as follows: 1) develop an integrative metastasis risk prediction model for colon cancer; 2) determine the optimum platform for the 34-gene metastasis score clinical test; and 3) test the optimized prognosis signature in a blinded fashion on archival tissue, to determine the robustness of the test and whether the metastasis risk score should be advanced to a prospective clinical trial to predict outcomes in stage II and III patients.