Mercedeh Ghadessi, Carissa Reid, Andrea Callegro & Toufik Zahaf
PSI Conference session including the following talks – “Historical controls in clinical trials – a Drug Information Association Adaptive Design Scientific Working Group (DIA-ADSWG) roadmap”, “Propensity Score-Integrated Meta-Analytic-Predictive Priors” and “Historical Controls Adaptive design in Vaccines trials: A retrospective evaluation”.


Historical controls in clinical trials – a Drug Information Association Adaptive Design Scientific Working Group (DIA-ADSWG) roadmap - Mercedeh Ghadessi (Bayer U.S. LLC)
Historical controls (HCs) can be used for model parameter estimation at the study design phase, adaptation within a study, or supplementation or replacement of a control arm. Currently on the latter, there is no practical roadmap from design to analysis of a clinical trial to address selection and inclusion of HCs, while maintaining scientific validity. This paper provides a comprehensive roadmap for planning, conducting, analyzing and reporting of studies using HCs, mainly when a randomized clinical trial is not possible. We review recent applications of HC in clinical trials, in which either predominantly a large treatment effect overcame concerns about bias, or the trial targeted a life-threatening disease with no treatment options. In contrast, we address how the evidentiary standard of a trial can be strengthened with optimized study designs and analysis strategies, emphasizing rare and pediatric indications. We highlight the importance of simulation and sensitivity analyses for estimating the range of uncertainties in the estimation of treatment effect when traditional randomization is not possible. Overall, the paper provides a roadmap for using HCs.


Propensity Score-Integrated Meta-Analytic-Predictive Priors - Carissa Reid (Boehringer Ingelheim Pharma GmbH & Co.  KG)
In clinical trials there may be ethical, financial or feasibility constraints that make it difficult to randomize patients to a control group. Historical control groups may address this problem by borrowing information from external patients. Using a Bayesian approach, borrowing techniques rely on finding an appropriate prior to incorporate historical information. Two popular options are the power prior and the Meta-Analytic-Predictive (MAP) prior (Neuenschwander et. al, 2010) approaches. Additionally, propensity scores can be used to select a comparable subset of the historical patients. 
 Bayesian inference and propensity score methods have already been integrated for the purpose of forming historical controls using a propensity score-integrated power prior approach (Wang et. al, 2019). We propose several methods for integrating propensity scores with the MAP prior approach. In the first approach, we stratify patients using propensity scores and then estimate the MAP prior separately in each stratum. The final estimate is a weighted sum of the estimates in each stratum. In the second approach, we derive one MAP prior by replacing the study level of the standard hierarchical model with the strata level. The third MAP prior is derived from a hierarchical model for studies. However, the weight of each study in the estimation of the posterior mean is adjusted based on a summary statistic of the respective propensity scores. 
 
 A simulation study was conducted to compare the performance of the proposed methods and the Wang et al. approach in terms of bias and mean-squared error.

Historical Controls Adaptive design in Vaccines trials: A retrospective evaluation - Andrea Callegro (GSK), Toufik Zahaf (GSK)