Modeling ovarian cancer screening for CEA

描述（由申请人提供）：对卵巢癌筛查的兴趣正在增长。被诊断患有卵巢癌的女性的五年生存率总体为50%，但癌症局限于卵巢的女性则为95%。然而，只有25%的卵巢癌在早期阶段被诊断出来，这表明通过早期检测有机会获得显着改善。我们的目标是通过考虑疾病和筛查人群的异质性来提高先前开发的卵巢癌筛查微观模拟模型的准确性，以评估使用标记物组纵向检测发展中疾病的成本效益。我们将扩大该模型的范围，以适应1）一组血清标记物用于筛查，和2）基于风险的筛查。我们还将纳入筛查和疾病对生活质量的影响，并根据筛查和治疗成本更新模型。这些努力将使我们能够确定卵巢癌筛查的潜在最有效的策略，并报告其成本效益。 本研究的具体目的有两个：一是开发一个最先进的卵巢癌筛查微观模拟模型，二是使用该模型探索卵巢癌筛查替代策略的成本效益。目标2有两个组成部分：使用一组血清标志物和成像来确定卵巢癌筛查的潜在成本效益策略，并估计这些策略在不同人群中的成本效益。 两项卵巢癌筛查的随机对照试验（RCT）正在进行中，一项在美国，一项在英国，但结果还要再过几年才能出来。无论随机对照试验的结果如何，关于成本效益、更频繁筛查的有效性以及多种标志物和成像的创新使用的问题仍将存在。分子发现可能很快就会产生一组标记物，这些标记物可以在涉及成像的多模式策略中一起用作一线筛选。因为进行新的随机对照试验来测试每一种潜在的更好的筛查策略将是非常昂贵的，因此需要一个准确的模拟模型来制定合理的医疗保健政策，并指导未来的基础和应用水平的研究。 为了提高模型预测的准确性，我们将对其进行改进，以考虑疾病的异质性（组织学，分级）和筛查人群的异质性（风险水平）。此外，我们将改进模型的检测组件，以更好地表示成像，并更新模型中使用的成本估计，使用来自华盛顿州Regence Blue Cross和Medicare的保险索赔数据。将进行广泛的验证，以评估该模型的预测与从试验中获得的估计数的一致性。 最终，一个增强的微观模拟模型，结合新的发展，未来的创新以及我们今天认识到的，将有助于指导政策和研究投资决策。

DESCRIPTION (provided by applicant): Interest in screening for ovarian cancer is growing. Five-year survival in women diagnosed with ovarian cancer is 50% overall, but in women with cancer confined to the ovaries, it is 95%. Only 25% of ovarian cancer is diagnosed in this early stage, however, suggesting that there is an opportunity for significant improvement through early detection. Our goal is to improve the accuracy of a previously developed microsimulation model of ovarian cancer screening by accounting for heterogeneity in the disease and in the population screened, in order to evaluate the cost-effectiveness of using a marker panel longitudinally to detect developing disease. We will expand the scope of the model to accommodate use of 1) a panel of serum markers for screening, and 2) risk-based screening. We will also incorporate QOL effects of both screening and disease, and update the model with respect to screening and treatment costs. These efforts will enable us to identify the potentially most efficient strategies for ovarian cancer screening and to report their cost-effectiveness. The specific aims of this study are twofold: one, to develop a state-of-the-art microsimulation model of ovarian cancer screening and two, to use the model to explore the cost-effectiveness of alternative strategies for ovarian cancer screening. There are two components to aim 2: to identify potentially cost-effective strategies for ovarian cancer screening using a panel of serum markers and imaging and to estimate the cost-effectiveness of the strategies in various populations defined by risk level. Two randomized controlled trials (RCT) of ovarian cancer screening are underway, one in the U.S. and one in the U.K., but results will not be available for several more years. Regardless of the outcomes of the RCTs, questions about cost-effectiveness, the efficacy of more frequent screening, and innovative use of multiple markers and imaging will remain. Molecular discoveries are likely soon to yield a panel of markers that can be used together as a first-line screen in a multimodal strategy involving imaging. Because it would be prohibitively expensive to conduct new RCT to test each potentially better screening strategy, an accurate simulation model will be necessary to develop sensible health care policy as well as to direct future research at both the basic and applied level. To improve the accuracy of the model's predictions, we will refine it to account for heterogeneity in the disease (histology, grade) and heterogeneity in the population screened (risk level). In addition, we will refine the detection component of the model to better represent imaging, and update the cost estimates used in the model, using insurance claims data from Regence Blue Cross of Washington State and Medicare. Extensive validation will be undertaken to assess the consistency of the model's predictions with estimates obtained from trials. Ultimately, an enhanced microsimulation model that incorporates new developments, innovations of the future as well as those we recognize today, will help guide policy and research investment decisions.