Keywords: abuse potential, abuse liability, statistical analysis, CNS drugs, FDA Guidance for Industry
The FDA has finalized its guidance documents on the assessment of abuse potential of new chemical entities and abuse-deterrent opioids. The primary objective of a human abuse potential (HAP) study is to provide information on the relative abuse potential of a test drug in humans. In January 2017, the FDA finalized their guidance on the assessment of the abuse potential of new chemical entities. As stated in the guidance (section V. C. 5.), the statistical model that should be used is a linear mixed-effects model, which includes period, sequence, and treatment as fixed effects and subject as a random effect. The primary analyses should be based on testing the differences between the means from the primary measure at the peak of drug response effects (Emax) produced by the test drug, the positive control, and placebo, using proper tests and appropriate statistical methods at a significance level of 0.05 (1-sided). The null hypotheses for the test drug should be constructed based on the presumption that the test drug produces abuse potential similar to the positive control and therefore differentiates from placebo. In order to demonstrate that the drug has no abuse potential, the null hypotheses should be rejected. However, if the null hypothesis is not rejected, the 1-sided test does not allow testing for significance in the other direction. Furthermore, on some subjective measures related to drug pharmacology, the above hypotheses may not be applicable and may need to be reversed. The guidance provides the 3 hypotheses to be tested in order to answer the 3 questions listed above, including the delta margins. This is in contrast to the draft guidance (January 2010), which offered no such details on the statistical analysis of HAP studies. The FDA also has a guidance on the evaluation of abuse-deterrent opioids (April 2015). This guidance lays out the methods for assessing abuse potential outcome measures in the potentially abuse-deterrent test product relative to a formulation of the drug without abuse-deterrent properties (control) or a newly formulated opioid product (positive control). The purpose of this session is to evaluate and discuss the statistical methodological approaches outlined in the guidance documents, with a comparison of the recommendations for abuse-deterrent opioids vs new chemical entities.