Challenges of conventional and novel approaches to clinical trial designs in cardiovascular medicine.

Abstract

Randomized clinical trials (RCT) are the gold standard for guiding treatment recommendations. Their results reflect the average treatment effect generated in samples from a selected patient population. Observational studies may serve as external validation of RCT findings, but are influenced by various biases. Conducting clinical trials in the cardiovascular field faces many challenges, including financial constraints, end-point selection including determining their value for doctors and patients, and assessment of long-term treatment effects. In addition, there is renewed debate about the optimal statistical approach for the evaluation of trial outcomes. These factors occur in parallel to efforts to develop novel clinical trial designs that address the above challenges. Pragmatic trials, for instance, may use data already collected during patient care. As alternative to fixed sample size, two-arm designs, adaptive trial designs have been introduced, that allow for pre-specified adaptations throughout a trial. Master protocols are used to govern platform, umbrella and basket trial designs all allowing to address more than one aspect at a time (e.g., the impact of multiple treatments on a single condition or a single treatment on multiple conditions). Cluster designs may allow the evaluation of new treatments into clinical routine. Here, we review strengths and limitations of conventional and novel trial designs. We provide a general description of current topics with a focus on treatment comparisons in the field of cardiovascular research.