时间：2019年07月05日（星期五）9:00-10:00

地点：学院南路，图配楼514

报告题目：Progression-free Survival Analysis in Cancer Trials:  StatisticalIssues due to Intermittent Observation Scheme

报告人：曾蕾蕾，加拿大滑铁卢大学统计与精算系副教授。

报告摘要：A composite endpoint called progression-free survival has beenroutinely used to evaluate the treatment effect in cancer trials.

When the non-terminal event such as progression is only assessedperiodically, progression-free survival is subject to a dual censoringscheme involving interval censoring for progression and right censoring fordeath. We highlight statistical issues associated with the conventional approachof using right endpoint imputation in this setting, explore the determinants ofthe asymptotic bias and point out the loss of power in detecting treatmenteffect. We also consider the design of cancer trials directed at the compositeevent of progression-free survival that insures the desired power. Inparticular, we derive sample size criteria based on an illnessÐdeath model thatconsiders cancer progression and death jointly while accounting for the factthat progression is assessed only intermittently.

报告人简介：LeileiZeng received her Ph.D. degree from the University of Waterloo in 2005. From2006 to 2011, she worked as an assistant professor at Simon Fraser Universityand held the University Faculty Award from the Natural Sciences andEngineering Research Council of Canada. Dr. Zeng joined the University ofWaterloo in 2011 to take up a position as an Associate Professor and the GrahamTrust Chair in Health Statistics (2011–2016). She was the Associate Chair ofUndergraduate Studies of the Department of Statistics and Actuarial Sciencefrom 2013 to 2015. Dr. Zeng currently serves on the Board of Directors of theStatistical Society of Canada (SSC) as the Ontario Representative. Dr.Zeng’s research interest lies in the development of statistical methodologiesfor life history and longitudinal studies, incomplete observation and missingdata problems, the design of clinical trials and epidemiological studies, andmodel misspecification and evaluation. Much of this work has been motivated by direct applications toimportant practical problems in medical and public health research. Her currentmain research applications are in modelling progression and comorbidity inrheumatic diseases, early life risk factors for cognitive deterioration anddementia/Alzheimer’s, and the association between stress and femalereproduction.

本次活动受威尼斯432888cam2019专题学术讲座项目资助。