Preclinical Phase III Trials in translational stroke research

Our understanding of basic stroke pathophysiology, risk factors promoting the disease and its sequelae in the central nervous and peripheral organ systems have progressed tremendously for the past decades. Preclinical research has identified a multitude of potential therapeutic targets to replenish our arsenal for fighting ischemic stroke.1 Although the efficacy of numerous therapeutic strategies was demonstrated preclinically, clinical trials failed to confirm this efficacy in patients,2 putting translational stroke research into crisis. However, the translational failure raised our awareness of the pathophysiological complexity of ischemic stroke and underpinned the importance of experimental quality. We also realized that context- and model-dependent efficacy might have confounded preclinical stroke research. Hence, measures to improve study quality, such as blinding, randomization, and attention to systemic physiology,3 are increasingly used and gradually become a prerequisite for publication. The majority of these measures can be easily implemented in most laboratories. However, some more complex but equally important aspects are harder to address by individual groups or even larger centers. For example, it is hardly possible to reproduce the wide range of age and weight, genetic heterogeneity, comorbidities, immunologic conditions, and pre-existing medications typically found in patient populations with stroke (diversity factors).4 Neglecting diversity factors could jeopardize our chances to translate basic research findings successfully into novel stroke treatments and eventually limit the willingness of the public and industry to support preclinical stroke research. In response, Dirnagl et al5 have recently proposed the concept of phase III preclinical trials. It aims to link groups and centers globally, which agree on common quality standards to achieve sufficient power/depth conjointly for adequate preclinical implementation of diversity factors. To resemble the situation of comparable clinical multicenter studies, all research activities may be governed by a supervising committee, featuring central randomization and data analysis. Intercenter comparability, for example, on blinded analysis of functional and imaging data sets, could be ensured by preclinical round robin trials.