Scientific Rigor Recommendations for Optimizing the Clinical Applicability of Translational Research

Abstract

The approval of new therapies to treat neurodegenerative disease conditions by the Food and Drug administration (FDA) has been hindered by many failed clinical trials, which were based upon “significant” efficacy in preclinical or translational studies. Additional problems during drug development related to significant adverse events and unforeseen toxicity have also hampered drug development. Recent reviews of preclinical data suggests that many studies have over-estimated efficacy due to poor or inadequate study design, exclusion of important data (negative or neutral) and lack of study randomization and blinding. This article describes in detail a set of recommendations to improve the quality of science being conducted in laboratories worldwide, with the goal of documenting in the peer-reviewed literature, including Journal of Neurology and Neurophysiology, the scientific basis for the continued development of specific strategies to treat neurodegenerative diseases such as Stroke, Alzheimer's disease, Huntington's disease, Parkinson's disease, Spinal cord injury, and Amyotrophic lateral sclerosis. The minimum recommendations for effective translational research include the need for model justification, study group randomization and blinding, power analysis calculations, appropriate statistical analysis of all data sets, and a conflict of interest statement by investigators. It will also be beneficial to demonstrate reproducible efficacy in multiple species and in studies done by independent laboratories.

Translational Stroke Research Opportunities and a Strategy to Develop Effective Cytoprotection

“…This assumes that the level of current research is inadequate and researchers must improve research study design, data handling, and interpretation. This recommendation is consistent with previous recommendations and publications, including STAIR [20], RIGOR [21][22][23][24] and CAMARADES [25,26], guidelines that are now mainstays in the field. However, based upon manuscript submission to this Journal, there is still less than 30% compliance by research laboratories worldwide primarily because high quality fully transparent research can be costly and there is insufficient funding available (see [30]).…”

“…However, based upon manuscript submission to this Journal, there is still less than 30% compliance by research laboratories worldwide primarily because high quality fully transparent research can be costly and there is insufficient funding available (see [30]). For the benefit of the stroke research community worldwide, we must first reiterate the basic recommendations resulting from the 2012 NINDS transparency in research workshop, which were published in the Journal of Neurology and Neurophysiology, Stroke, Translational Stroke Research and Nature among other journals [21][22][23][24].…”

“…& Dr. Malcolm Macleod and colleague, Dr. Emily Sena from the University of Edinburgh, summarized the CAMARADES effort and emphasized that researchers need to define the method of randomization, conduct power analysis and conduct studies blinded [see RIGOR guidelines above [21][22][23][24]. & Committee members recommended that aged animals be used for stroke research, comorbid conditions need to be represented (see [31,32,33]), chronic stroke models (months of survival) should be developed because rodents display adaptive plasticity, and stroke studies should incorporate atrial fibrillation, hypoperfusion, transient ischemic attack (TIA), and white matter damage.…”

In Vivo Animal Stroke Models

“…The RIGOR guidelines were proposed by the NIH National Institute of Neurological Disorders and Stroke to ensure worldwide uniformity in stroke research practices. Many similarities abound in the guidelines recommended by STAIR, STEPS, and RIGOR, foremost is the adherence to good laboratory (GLP) practices, including the need for all animal modeling studies to pursue treatment condition blinding, randomization, and complete power analysis and statistical analysis [120]. Moreover, these GLP practices should be clearly stated when submitting translational grant applications and manuscripts [120].…”

“…Many similarities abound in the guidelines recommended by STAIR, STEPS, and RIGOR, foremost is the adherence to good laboratory (GLP) practices, including the need for all animal modeling studies to pursue treatment condition blinding, randomization, and complete power analysis and statistical analysis [120]. Moreover, these GLP practices should be clearly stated when submitting translational grant applications and manuscripts [120]. Of note, Translational Stroke Research and the Journal of Neurology and Neurophysiology have established this policy in their manuscript submissions.…”