Advanced statistical modeling and analytics for prostate cancer interventions

项目摘要 制定前列腺癌控制的合理策略需要整合来自大量收集的信息 临床试验和观察性研究，并开发统计和计算机模型，以填补不可避免的 证据缺口前列腺癌的治疗是特别有争议的，因为缺乏一个 就最佳筛查和治疗方法达成共识。本申请的主要研究者是 该团队开发统计方法和计算机模型，以推进合理的前列腺癌政策。这 这项工作只能由癌症监测和统计建模方面的熟练专家来完成。它要求 复杂统计计算方面的专业知识，必须在内部和 外部的它需要出色的写作技巧，使项目从调查阶段通过出版物 它需要经过验证的跨多学科团队沟通和协作的能力。的PI 本申请的作者利用这些技能和其他技能建立了他的研究小组，成为前列腺领域的领导者。 癌症模型和政策。在这个应用程序中，他建议使用这些技能来支持新的建模 分析包括（1）高风险患者亚组的靶向筛查，（2）决策框架 围绕用于筛查和治疗的新型生物标志物，以及（3）推进估计过度诊断的方法- 通过筛查来检测癌症，否则这些癌症永远不会被诊断出来。此外，本申请 将支持（1）开发在线计算器和决策辅助工具，以协助临床医生和决策者， 选择适当的建议和（2）在多项目中提供高质量的生物统计服务 前列腺癌转化研究计划。总体目标是优化干预措施， 发病率和死亡率在男性中最常见的癌症。

PROJECT SUMMARY Developing sound strategies for prostate cancer control requires integrating information from a vast collection of clinical trials and observational studies and developing statistical and computer models to fill in inevitable evidence gaps. Management of prostate cancer is particularly controversial because of the absence of a consensus about optimal screening and treatment approaches. The PI of this application is a member of a team that develops statistical methods and computer models to advance sound prostate cancer policies. This work can only be done by a skilled specialist in cancer surveillance and statistical modeling. It demands expertise in complex statistical computing, which must be meticulously executed and validated, internally and externally. It requires outstanding writing skills to bring projects from the inquiry phase through the publication process, and it requires a proven ability to communicate and collaborate across multidisciplinary teams. The PI of this application has used these skills and others to establish his research group as a leader in prostate cancer modeling and policy. In this application he proposes to use these skills in support of new modeling analyses including (1) targeted screening in high-risk patient subgroups, (2) frameworks for decision making around novel biomarkers for screening and treatment, and (3) advancing methods to estimate overdiagnosis— the detection by screening of cancers that would never be diagnosed otherwise. Additionally, this application will support (1) developing online calculators and decision aids to assist clinicians and policymakers in selecting appropriate recommendations and (2) providing high-quality biostatistical service in a multi-project translational prostate cancer research program. The overarching aim is to optimize interventions to reduce morbidity and mortality from this most common cancer in men.