Abstract
BACKGROUND: Surrogate endpoints, such as those of interest in chronic kidney disease (CKD), are often evaluated using Bayesian meta-regression. Trials used for the analysis can evaluate a variety of interventions for different sub-classifications of disease, which can introduce two additional goals in the analysis. The first is to infer the quality of the surrogate within specific trial subgroups defined by disease or intervention classes. The second is to generate more targeted subgroup-specific predictions of treatment effects on the clinical endpoint. METHODS: Using real data from a collection of CKD trials and a simulation study, we contrasted surrogate endpoint evaluations under different hierarchical Bayesian approaches. Each approach we considered induces different assumptions regarding the relatedness (exchangeability) of trials within and between subgroups. These include partial-pooling approaches, which allow subgroup-specific meta-regressions and, yet, facilitate data adaptive information sharing across subgroups to potentially improve inferential precision. Because partial-pooling models come with additional parameters relative to a standard approach assuming one meta-regression for the entire set of studies, we performed analyses to understand the impact of the parameterization and priors with the overall goals of comparing precision in estimates of subgroup-specific meta-regression parameters and predictive performance. RESULTS: In the analyses considered, partial-pooling approaches to surrogate endpoint evaluation improved accuracy of estimation of subgroup-specific meta-regression parameters relative to fitting separate models within subgroups. A random rather than fixed effects approach led to reduced bias in estimation of meta-regression parameters and in prediction in subgroups where the surrogate was strong. Finally, we found that subgroup-specific meta-regression posteriors were robust to use of constrained priors under the partial-pooling approach, and that use of constrained priors could facilitate more precise prediction for clinical effects in trials of a subgroup not available for the initial surrogacy evaluation. CONCLUSION: Partial-pooling modeling strategies should be considered for surrogate endpoint evaluation on collections of heterogeneous studies. Fitting these models comes with additional complexity related to choosing priors. Constrained priors should be considered when using partial-pooling models when the goal is to predict the treatment effect on the clinical endpoint.