The impact of fraudulent and irreproducible data to the translational research crisis - solutions and implementation

Abstract

One of the aims of basic neuroscience research is ultimately the development of therapeutics to cure diseases. Funders granting money to research institutions increasingly express interest into how their financial resources are used and look for successful translation in clinical practice. Disappointingly, many findings that started out promising in basic research projects and phase I trials did not live up to the promise of therapeutic efficacy in later phase II or III trials. An inordinately high amount of time and money is thus spent on research that does not always have the required human impact. Potential reasons for these problems are numerous. Although research misconduct occurs and contributes to this shortcoming, it is not the only important factor. Frequently, basic science results turn out to be irreproducible. Irreproducibility, outside of malfeasance, is multifactorial and can include poor experimental design, conduct, statistical analysis, reporting standards, and conceptual flaws. Further confounding problems include an insufficient transferability of animal to human physiology, as well as intersubject group variability, for example, sexual dimorphisms. While the causes of poor data reproducibility are therefore numerous, equally there are many groups that can contribute to improvements in how basic science is reported. Here, we will review how the Journal of Neurochemistry can contribute to increasing the value of preclinical and translational research. Despite a vast amount of very promising basic research findings, these failed to successfully translate into the clinical practice so far. The reasons for this 'data reproducibility crisis' are numerous, for example, rooting in insufficient experimental design, conceptual flaws, incorrect statistical planning and evaluation, incomplete model system that do not adequately reproduce the human physiology, and further reasons discussed in this Review with the aim to present practical solutions that can be implemented by researchers, journals editors, and reviewers. We will also explain measures the Journal of Neurochemistry have implemented to overcome these issues and weaknesses in preclinical research. These includes adherence to the ARRIVE ( www.nc3rs.org) guidelines, NINDS standards (doi: 10.1038/nature11556), and The Transparency and Openness Promotion Guidelines (TOP) Committee guidelines (https://cos.io/top/#TOP). This article is part of the 60th Anniversary special issue.

Keywords: 60th Anniversary Journal of Neurochemistry; clinical trials; data reproducibility; scholarly publishing; statistics.