Clinical trials aimed at identifying effective treatment and prevention strategies to reduce the risk of diabetes often require long term follow-up of patients in order to observe a sufficient number of diabetes diagnoses to precisely estimate treatment effects. In such settings, the availability of a surrogate marker that could be used to estimate the treatment effect and could be observed earlier than the occurrence of a diabetes diagnosis would allow researchers to make conclusions regarding the treatment effect with less required follow-up time. That is, validated surrogate markers could enable shorter randomized clinical trials and require smaller sample sizes, thus accelerating acquisition of clinical information. Identifying such surrogate markers and determining when these markers should be collected would contribute significantly to research aimed at identifying effective treatments in diabetes. Research on identifying useful surrogate markers focuses on estimation of the proportion of treatment effect explained by a surrogate marker since a valid surrogate marker should capture a large proportion of the true treatment effect on the primary outcome. However, current methods to estimate the proportion of treatment effect explained usually require restrictive model assumptions that may not hold in practice. In this study, we aim to shift current research practice on surrogate marker identification away from restrictive model-based approaches towards a robust estimation approach by proposing novel methods that allow for more flexible model assumptions. Specifically, we propose to develop novel statistical methods to estimate the proportion of treatment effect explained by one or more potential surrogate markers and to identify when such surrogate marker information should be collected using a landmark approach such that a test for treatment effectiveness using the information will have desirable power, thereby allowing for less required follow-up time. We will apply these methods to data from the Diabetes Prevention Program Outcomes Study (DPPOS) in order to understand if such new methods can identify surrogate markers of diabetes and complications associated with diabetes that adequately capture the preventive treatment effect of Metformin and a lifestyle intervention, two forms of prevention examined in DPPOS.
In this project, we develop novel statistical methods to estimate the proportion of treatment effect explained by one or more potential surrogate markers of diabetes and identify when such surrogate marker information should be collected using data from the Diabetes Prevention Program Outcomes Study. The results of this study have the potential to inform and improve the design, content, and analysis of future studies aimed at diabetes prevention and provide new insight describing the complex relationship between treatment, potential surrogate markers, diabetes, and complications associated with diabetes.