University of Cambridge > > MRC Biostatistics Unit Seminars > Assessing surrogacy using linear mixed models and the surrogate threshold effect

Assessing surrogacy using linear mixed models and the surrogate threshold effect

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Li Su.

A surrogate outcome is a biomarker that is used as a substitute for the true outcome, for the purpose of assessing treatment efficacy in a Phase III clinical trial. In situations where measurement of the most relevant clinical outcome may be difficult, costly, invasive, or require a long follow-up time, use of a more easily or cheaply measured surrogate outcome allows more rapid or less costly evaluation of treatments. On the other hand, inappropriate use of a biomarker as a surrogate can be dangerous, as illustrated by the infamous CAST experience. Hence a potential surrogate needs to be “validated” before use.

This talk will review some of the statistical methods proposed for evaluating potential surrogates, and describe a meta-analytic approach based on a bivariate normal linear mixed model proposed by Buyse et al. (2000). The surrogate threshold effect has been has been suggested as a useful measure of surrogacy that is conceptually meaningful to clinicians. Statistical properties of this measure will be considered, and illustrated via the results of some simulation studies.

This talk is part of the MRC Biostatistics Unit Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity