PREDICTIVE MODELING IN LUNG CANCER

描述（由申请人提供）：2012年美国预计将诊断出201,000例肺癌新病例。肺癌是癌症相关死亡的最常见原因，每年导致近15万人死亡。据估计，照顾肺癌患者的费用每年超过400亿美元。尽管肺癌对社会造成了巨大的影响，但目前还没有可靠的、临床适用的方法来预测肺癌患者治疗后的预后。研究表明，肺癌诊断后的生存取决于患者和肿瘤的特征。治疗或手术的类型也会影响生存。由于缺乏详细的患者信息、方法学问题和缺乏验证，先前试图建立肺癌治疗后生存预测模型的尝试受到严重限制。该职业发展计划旨在为申请人提供培训和支持，使其成为一名独立的临床研究人员，专注于评估和模拟胸部肿瘤学的结果。本计划书的职业发展目标是：1. 在负责任的研究行为、研究设计、统计、建模方法、决策分析以及与患者和提供者的风险沟通方面，获得教学培训，打下坚实的基础。2. 发展专业知识，建立预测模型，评估常见胸部癌症的竞争疗法，并进行成本效益分析。3. 培养必要的技能，以沟通和传播研究结果，在实践中实施研究成果，并影响政策和医疗保健服务的变化，以改善结果。短期职业发展目标将通过在华盛顿大学完成临床研究硕士学位来完成。为了培养实践技能，申请人将利用决策分析模型来评估和预测早期肺癌患者手术或放疗后的长期生存。类似的方法将用于研究局部晚期肺癌治疗方案的有效性和成本效益。临床目标是开发和推广能够预测肺癌治疗后生存的工具，并评估治疗方案的成本效益。临床医生和公众可以在华盛顿大学网站上通过一个电子的、用户友好的界面获得这些模型。这些模型将支持研究人员对患者的预后进行评估。我们的结果也将作为评估基因研究等新兴测试价值的基线，这些测试可以与模型进行比较并纳入模型。该提案的职业目标是将候选人发展成为一名独立的研究者，他可以实施建模方法来评估干预措施对肿瘤临床护理的影响。

DESCRIPTION (provided by applicant): An estimated 201,000 new cases of lung cancer will be diagnosed in the United States in 2012. Lung cancer, the most common cause of cancer-related mortality, causes nearly 150,000 deaths annually. It is estimated that the costs of taking care of patients with lung cancer exceed $40 billion annually. Despite the tremendous impact of lung cancer on society, there are no reliable, clinically applicable methods to predict post-treatment outcomes in lung cancer patients. It has been shown that survival after diagnosis of lung cancer depends on both patient and tumor characteristics. Type of treatment or operation also impacts survival. Previous attempts at creating predictive models for survival after treatment for lung cancer have been severely limited by lack of detailed patient information, methodologic issues, and lack of validation. This career development proposal is designed to provide training and support for the applicant to become an independent clinical researcher focused on evaluating and modeling outcomes in thoracic oncology. The career development goals of this proposal are; 1. Obtain didactic training for a strong foundation in responsible conduct of research, research design, statistics, modeling methodology, decision analysis, and communication of risk to patients and providers. 2. Develop expertise in creating predictive models to assess competing therapies for common thoracic cancers and performing cost-effectiveness analyses. 3. Develop the skills necessary to communicate and disseminate results of the studies, implement research findings in practice, and influence change in policy and healthcare delivery to improve outcomes. The short-term career development goals will be accomplished by completing a Master of Science in Clinical Investigation degree at Washington University. To develop the practical skill set, the applicant will utilize decision analytic modelig to evaluate and predict long-term survival after surgery or radiation therapy for patients with early-stage lung cancer. Similar methods will be used to study the effectiveness and cost- effectiveness of treatment options for locally advanced lung cancer. The clinical objective is to develop and disseminate tools that can predict survival after treatment for lung cancer and to evaluate the cost-effectiveness of treatment options. The models will be made available to clinicians and the public on the Washington University website via an electronic, user-friendly interface. The models will support investigators seeking to assess prognosis for patients. Our results will also serve as baseline for assessing the value of new and emerging tests like genetic studies, which could be compared to and incorporated into the models. The career objective of the proposal is to develop the candidate into an independent investigator, who can implement modeling approaches to assess the impact of interventions on clinical care in oncology.