Regression analyses based on transformations of cumulative incidence functions are often adopted when modeling and testing for treatment effects in clinical trial settings involving competing and semi-competing risks. Common frameworks include the Fine-Gray model and models based on direct binomial regression. Using large sample theory we derive the limiting values of treatment effect estimators based on such models when the data are generated according to multiplicative intensity-based models, and show that the estimand is sensitive to several process features. The rejection rates of hypothesis tests based on cumulative incidence function regression models are also examined for null hypotheses of different types, based on which a robustness property is established. In such settings supportive secondary analyses of treatment effects are essential to ensure a full understanding of the nature of treatment effects. An application to a palliative study of individuals with breast cancer metastatic to bone is provided for illustration.
Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data
University of Cambridge(剑桥大学);Fred Hutchinson Cancer Center(弗雷德·哈钦森癌症中心);Ghent University(根特大学)
Studying Competing Events with Federated Cumulative Incidence Curves
University of Michigan(密歇根大学);University of Florida(佛罗里达大学);University of Indiana(印第安纳大学);Northwestern University(西北大学);Purdue University College of Pharmacy(普渡大学药学院)
Estimation of separable direct and indirect effects in continuous time
许多研究问题涉及到时间事件的结果,可以防止发生由于竞争事件。在这些情况下,我们必须小心经典统计被估量的因果解释。特别是,估计的危害规模,如比率的原因具体或子分布的危害,是根本难以解释的因果关系。风险量表的估计值(如累积发生率函数的对比)确实具有因果解释,但它们仅捕获治疗对关注事件的总效应;即,通过竞争事件和竞争事件之外的效应。为了理清关注事件和竞争事件的因果处理效应,最近引入了可分离的直接和间接效应。在这里,我们提供了新的结果估计的直接和间接的可分离的影响在连续的时间。特别是,我们推导出连续时间的非参数影响函数,并用它来构造一个估计,具有一定的鲁棒性。我们还提出了一个简单的估计基于半参数模型的两个原因特定的风险函数。我们描述了这些估计量的渐近性质,并给出了模拟研究的结果,表明估计量在有限样本中表现令人满意。最后,我们重新分析了Stensrud et al(2020)的前列腺癌试验。
Inference for competing risks based on area between curves statistics
RWTH Aachen University(亚琛工业大学);Otto von Guericke University Magdeburg(奥托·冯·格里克马格德堡大学);TU Dortmund University(多特蒙德工业大学);Research Center Trustworthy Data Science and Security, UA Ruhr(鲁尔联合大学可信数据科学与安全研究中心)
Cost-effectiveness analysis for therapy sequence in advanced cancer: A microsimulation approach with application to metastatic prostate cancer
目的.晚期癌症患者可能会接受多线治疗,随着疾病进展而转换治疗。受转移性前列腺癌研究的启发,我们开发了一个微观模拟框架来研究治疗序列。方法.我们提出了一个离散时间状态转移模型来研究两种抗癌治疗方案。基于数字化的已发表的无进展生存期(PFS)和总生存期(OS)曲线,我们推断事件类型(进展或死亡),并使用具有竞争风险的累积发生率函数估计转移概率。我们的模型随着时间的推移纳入了患者内的依赖性,这样对一线治疗的反应就可以告知后续事件的概率。控制患者内依赖性程度的参数可用于将基于模型的结果校准为目标试验的结果。我们在转移性前列腺癌的两种治疗序列的研究中证明了这些方法,其中多西他赛(DCT)和醋酸阿比特龙(AA)都适用于一线或二线治疗。我们评估了两种治疗策略的成本、质量调整生命年(QES)和增量成本效果比(ICER):DCT然后AA与AA然后DCT。结果使用相关临床试验的数字化生存曲线,我们确定了8.6-13.9%的PFS时间应归类为死亡,从而可以估计累积发生率函数。假设患者内独立性的模型高估了OS时间,用我们的校准方法进行了校正。校正导致治疗策略(0.07 vs 0.15)和ICER(-76,836美元/QALY vs -21,030美元/QALY)之间的QALY差异发生有意义的变化。结论.微观模拟模型可以成功地用于研究治疗序列的成本效益,注意正确考虑患者内的依赖性。
Capturing Cumulative Disease Burden in Chronic Kidney Disease Outcome Trials: Area Under the Curve and Restricted Mean Time in Favor of Treatment Beyond Conventional Time-to-First Analysis
Global Statistical Science(全球统计科学);Eli Lilly and Company(礼来公司)
tteICE: An R Package for Estimating Treatment Effects on Time-to-Event Outcomes with Intercurrent Events in Two-Arm Trials
Kobe University(神户大学);Shanghai University of International Business and Economics(上海对外经贸大学);Chinese Academy of Medical Sciences & Peking Union Medical College(中国医学科学院 北京协和医学院)
Best for which estimand? A known-truth benchmark of longitudinal-matching and target-trial-emulation methods for time-varying treatments
School of Population and Public Health, University of British Columbia(不列颠哥伦比亚大学公共卫生与人口健康学院);Centre for Advancing Health Outcomes, St. Paul’s Hospital(圣保罗医院健康结果促进中心)
Pseudo-value Based Mean Cumulative Count Regression
University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校);Novartis Pharma AG(诺华制药股份有限公司);Department of Statistical Sciences, University of Toronto(多伦多大学统计科学系);Hoffman La Roche Ltd(罗氏制药有限公司)
Nonparametric Efficient Estimation of Dynamic Treatment Regimes with Competing Risks
University of Minnesota(明尼苏达大学);Edward Hines Jr. VA Hospital(爱德华·海恩斯少将退伍军人医院);Augusta University(奥古斯塔大学);Veterans Affairs Health Care System(退伍军人保健系统);HealthPartners Institute(健康伙伴研究所)
Authentic Multinational Federated Time-to-Event Analyses Among People with HIV in Latin America
Vanderbilt University Medical Center(范德堡大学医学中心);Pontificia Universidad Católica de Chile(智利天主教 Pontificia 大学);Instituto Nacional de Infectologia Evandro Chagas, Fundação Oswaldo Cruz (INI-Fiocruz)(Evandro Chagas 国家传染病研究所,奥斯瓦尔多·克鲁兹基金会);Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán(萨尔瓦多·祖比兰国家医学与营养科学研究所);Johns Hopkins University(约翰斯·霍普金斯大学);Vanderbilt University(范德堡大学);Fundación Huésped(Huésped 基金会);University of Pennsylvania(宾夕法尼亚大学)
Survival analysis for AdVerse events with VarYing follow-up times (SAVVY): summary of findings and a roadmap for the future of safety analyses in clinical trials
Evaluating Dynamic Conditional Quantile Treatment Effects with Applications in Ridesharing
DiDi Chuxing(滴滴出行);University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校);Shanghai University of Finance and Economics(上海财经大学);London School of Economics and Political Science(伦敦政治经济学院)
Estimating Pathway Treatment Effects in the Presence of Intermediate Events with Multi-State Data
Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Center(弗雷德·哈钦森癌症中心疫苗与传染病部门);Department of Economics, University of California San Diego(加州大学圣地亚哥分校经济学系);Department of Biostatistics, University of Michigan(密歇根大学生物统计学系);Amazon Inc.(亚马逊公司)
Generalized interpretation and identification of separable effects in competing event settings
在竞争事件背景下,原因特异性累积发生率的反事实对比量化了治疗对关注事件的总体因果效应。然而,治疗对比赛项目的影响可能间接地促成了这一总效应,使其解释复杂化。我们之前提出了可分离效应(Stensrud et al,2019),以定义治疗对关注事件的直接和间接效应。该定义的前提是将治疗分解为两个成分,沿着两个独立的因果路径起作用,一个完全在竞争事件之外,另一个完全通过竞争事件。与以前的直接和间接效应定义不同,在对治疗成分进行单独干预的研究中,可以对可分离效应进行实证审查。在这里,我们扩展和推广的概念,可分离的影响,在几个方面,允许解释,识别和估计下相当弱的假设。我们提出并讨论了一个定义的可分离的效果,适用于一般的时变结构,其中的可分离的效果仍然可以有意义地解释,即使它们不能被视为直接和间接的影响。我们进一步推导出在尚未提供分解治疗的观察性研究中识别可分离效应的较弱条件;特别是,这些条件允许感兴趣事件、竞争事件和失访的常见原因随时间变化。对于这些一般设置,我们提出了半参数加权估计,是直接实现。作为一个例子,我们应用估计量来研究强化血压治疗对急性肾损伤的可分离影响,使用随机临床试验的数据。