Modeling prostate cancer control: prevention, screening, treatment, followup care

描述（由申请人提供）：该提案的广泛长期目标是将疾病建模定位为临床决策和政策制定的可行工具。具体目标是利用CISNET前列腺组开发的三个互补监测模型来解决前列腺癌中一些最持久和最紧迫的政策问题。 在拟议的工作中，我们将扩展模型，以捕获诊断后的下游结果，如疾病复发和二次治疗。扩展将由一些最大和最丰富的基于人群的数据源提供信息，并将使用这些数据集和最近发表的美国和欧洲前列腺癌筛查试验的结果进行广泛验证。 扩展模型将用于预测不同筛查和治疗政策的预期成本和效益，以确定那些可能在实践中最有价值的政策。筛查政策将考虑开始和停止筛查的不同年龄、筛查间隔、基于PSA的活检转诊标准以及纳入新型筛查生物标志物的组合政策。治疗策略将包括生化失败后立即与延迟初级治疗以及立即与延迟二级治疗。我们将广泛调查包括激素治疗在内的政策对疾病特异性和其他原因死亡率的影响，激素治疗是疑似或确诊转移性疾病最常见的全身治疗。认识到不同的亚群可能需要不同的政策，我们还将调查在已知影响前列腺癌风险和结果的因素（即年龄，合并症，种族和肥胖）定义的亚群中是否需要有针对性的政策。拟议的工作全面涵盖了癌症控制问题的连续性，适合建模，今天面临的前列腺癌研究人员。我们计划使用协调的比较建模方法进行研究，其中独立模型被标准化，并通过使用共同的输入和实施共同的“基本情况”情景进行比较。 .

DESCRIPTION (provided by applicant): The broad, long-term goal of this proposal is to position disease modeling as a viable tool for clinical decision making and policy development. The specific goal is to utilize the three complementary surveillance models developed by the CISNET prostate group to address some of the most persistent and pressing policy questions in prostate cancer. In the proposed work we will extend the models to capture downstream outcomes following diagnosis, such as disease recurrence and secondary treatment. The extensions will be informed by some of the largest and richest population-based data sources available and will be validated extensively using these datasets and results from recently published US and European prostate cancer screening trials. The extended models will be used to project the expected costs and benefits of different screening and treatment policies in order to identify those likely to be of most value in practice. The screening policies will consider different ages to start and stop screening, inter-screen intervals, PSA-based criteria for biopsy referral, and combination policies that incorporate novel screening biomarkers. The treatment policies will include immediate versus delayed primary treatment and immediate versus deferred secondary treatment following biochemical failure. We will extensively investigate the ramifications for both disease-specific and other-cause mortality of policies that include hormonal therapy, the most common systemic treatment for suspected or confirmed metastatic disease. Recognizing that different policies may be called for in different subgroups, we will also investigate the need for targeted policies within subpopulations defined by factors known to affect prostate cancer risk and outcomes, namely age, comorbidity, race, and obesity. The proposed work comprehensively covers the continuum of cancer control issues amenable to modeling that face prostate cancer investigators today. We plan to conduct the research using a coordinated comparative modeling approach in which independent models are standardized and made comparable by the use of common inputs and the implementation of common "base case" scenarios. .