Scientific and Regulatory Policy Committee Best Practices: Recommended ("Best") Practices for Informed (Non-blinded) Versus Masked (Blinded) Microscopic Evaluation in Animal Toxicity Studies.

This article describes the Society of Toxicologic Pathology's (STP) five recommended ("best") practices for appropriate use of informed (non-blinded) versus masked (blinded) microscopic evaluation in animal toxicity studies intended for regulatory review. (1) Informed microscopic evaluation is the default approach for animal toxicity studies. (2) Masked microscopic evaluation has merit for confirming preliminary diagnoses for target organs and/or defining thresholds ("no observed adverse effect level" and similar values) identified during an initial informed evaluation, addressing focused hypotheses, or satisfying guidance or requests from regulatory agencies. (3) If used as the approach for an animal toxicity study to investigate a specific research question, masking of the initial microscopic evaluation should be limited to withholding only information about the group (control or test article-treated) and dose equivalents. (4) The decision regarding whether or not to perform a masked microscopic evaluation is best made by a toxicologic pathologist with relevant experience. (5) Pathology peer review, performed to verify the microscopic diagnoses and interpretations by the study pathologist, should use an informed evaluation approach. The STP maintains that implementing these five best practices has and will continue to consistently deliver robust microscopic data with high sensitivity for animal toxicity studies intended for regulatory review. Consequently, when conducting animal toxicity studies, the advantages of informed microscopic evaluation for maximizing sensitivity outweigh the perceived advantages of minimizing bias through masked microscopic examination.

Meta-analysis of blinded and unblinded studies for ongoing aggregate safety monitoring and evaluation.

During the course of clinical development, ongoing aggregate safety monitoring and evaluation are needed to understand the evolving safety profile and to ensure effective risk-management strategies for medicinal products. CIOMS reports and global regulatory guidance (including from ICH, US FDA, and EMA) compel sponsors for assessment of safety based on aggregate data. To identify and characterize the risks of medicinal products at a program level in a more timely and informed manner, aggregate safety evaluations should combine all available information, including from ongoing blinded trials, completed unblinded trials, and other data sources. In this article, we propose two Bayesian meta-analytic approaches for synthesizing blinded and unblinded studies in order to characterize the evolving safety profile of medicinal products at the program level. With the proposed approaches, sponsors can dynamically update knowledge of their product safety profiles as data accrue. Application of the procedures to a real and a hypothetical clinical trial program are provided to illustrate how the proposed approaches can be used to analyze a pre-specified event of interest and to screen for risk-elevated events.

A Single Test to Study Social Behavior and Repetitive Self-grooming in Mice.

The ability to recognize and interact with members of the same species is essential for social communication. Investigating the neural substrates of social interest and recognition may offer insights into the behavioral differences present in disorders affecting social behavior. Assays used to study social interest in rodents include the 3-chamber test, a partition test, and a social interaction test. Here, we present a single protocol that can be used to quantify the level of social interest displayed by mice, the ability to distinguish between different individual mice (social recognition), and the level of repetitive self-grooming displayed. In the first part of the protocol, a social habituation/dishabituation test, the time spent by a test mouse sniffing a stimulus mouse is quantified over 9 trials. In the first 8 interactions, the same stimulus mouse is used repeatedly; on the ninth trial, a novel stimulus mouse is presented. Intact social recognition is indicated by a progressive decrease in the investigation time over trials 1-8, and an increase in trial 9. The interval between each social trial is used to quantify self-grooming, a stereotyped repetitive behavior in mice. We also present a method for randomized, blinded analysis of these behaviors to increase rigor and reproducibility of results. Therefore, this single behavioral test enables ready assessment of phenotypes of both social and repetitive behaviors in an integrated manner in the same animals. This feature can be advantageous in understanding interactions between these behaviors and phenotypes in mouse models with genetic variants associated with autism and other neurodevelopmental or neuropsychiatric disorders, which are often characterized by these behavioral differences.

Bridging blinded and unblinded analysis for ongoing safety monitoring and evaluation.

In order to better characterize the safety profile of investigational new drugs (INDs) during clinical development, more interest and attention have been paid to ongoing safety monitoring and evaluation. The 2015 US FDA IND safety reporting draft guidance compels sponsors to periodically evaluate unblinded safety data. However, maintaining the trial blind is necessary to avoid jeopardizing the validity of study findings. In this article, we propose an innovative new approach which includes analyzing both blinded and unblinded data. The proposed two-stage framework incorporates periodic analyses of blinded safety data to detect and flag adverse events that may have potential risk elevation related to experimental treatment, as well as planned unblinded analyses to quantify associations between the drug and adverse events, and to determine thresholds for referring adverse events for medical review and safety reporting.

Decision on performing interim analysis for comparative clinical trials.

In randomized-controlled trials, interim analyses are often planned for possible early trial termination to claim superiority or futility of a new therapy. While unblinding is necessary to conduct the formal interim analysis in blinded studies, blinded data also have information about the potential treatment difference between the groups. We developed a blinded data monitoring tool that enables investigators to predict whether they observe such an unblinded interim analysis results that supports early termination of the trial. Investigators may skip some of the planned interim analyses if an early termination is unlikely. We specifically focused on blinded, randomized-controlled studies to compare binary endpoints of a new treatment with a control. Assuming one interim analysis is planned for early termination for superiority or futility, we conducted extensive simulation studies to assess the impact of the implementation of our tool on the size, power, expected number of interim analyses, and bias in the treatment effect. The numerical study showed the proposed monitoring tool does not affect size or power, but dramatically reduces the expected number of interim analyses when the effect of the treatment difference is small. The tool serves as a useful reference when interpreting the summary of the blinded data throughout the course of the trial, without losing integrity of the study. This tool could potentially save the study resources and budget by avoiding unnecessary interim analyses.