Speaker
Patrick Lawler, MD, MPH
Cardiology and Critical Care Divisions
McGill University and University of Toronto
Keywords
REMAP-CAP; adaptive platform trials; critical care; pneumonia
Key Points
-
Clinical trials face challenges that were amplified during the COVID-19 public health emergency, including avoiding shortcomings of prior data, equipoise, heterogeneity of treatment effect, making trials happen quickly, the complexity of studying a single intervention in isolation, and the sustainability and efficiency of trial conduct.
-
REMAP-CAP is an adaptive platform trial simultaneously evaluation multiple treatments in domains using a master protocol. In adaptive trials, trial design features are modified during the trial to maximize statistical efficiency or achieve better patient outcomes. Examples include response-adaptive randomization, sample size reassessment, multi-arm designs, group sequential stopping, and others.
-
Most diseases have multiple component therapies, and several questions must be evaluated for each component, including whether there is heterogeneity of treatment effect. Clinical trials generally assume homogenous treatment effects; in reality there are heterogeneous treatment effects.
-
In therapeutic anticoagulation domain of REMAP-CAP, Bayesian techniques allowed for greater statistical efficiency in arriving at conclusions about effectiveness. A post hoc analysis evaluated strategies for further studying heterogeneity of treatment effect.
-
REMAP-CAP offered operational and statistical efficiencies, addressing challenges associated with the COVID-19 pandemic. The trial shortened the divide between clinical care and research and demonstrated a capacity for “parabiosis” with other trials in a multiple-platform trial. Adaptive designs increasingly empowered the trial to hunt for heterogeneity of treatment effect.
Discussion Themes
– When an adaptive design enables researchers to control for variations in usual baseline care for complex conditions, there are benefits for statistical efficiency and power and possibly an ability to isolate heterogeneity of treatment effect.
– There were advantages and disadvantages to the multi-platform approach. The approach definitely generated evidence more quickly. Prospective harmonization with other platform trials also brought operational efficiencies, including accelerated recruitment and enrollment.
– In applying this approach to studies of chronic conditions, different monitoring guidelines may be needed to evaluate more modest treatment effect. For example, futility criteria could be lowered. Two-stage designs could also be helpful, such as first evaluating safety outcomes before scaling up to a phase 3 study, or evaluating surrogate outcomes in a phase 2 component and then graduating the treatments that appear more likely have a greater benefit in phase 3, or other ways of mitigating the need for a large sample size.
– An adaptive platform can place additional burdens on the DSMB. REMAP-CAP has a single DSMB overseeing all domains, with its membership having both clinical expertise and experience adaptive trials, and the committee needing to review both primary outcome data and secondary endpoints and other information.
#pctGR, @Collaboratory1