FDA Proposes a Tier-Based Approach to Evaluate “Bioequivalence” of Abuse Deterrence of Generic SODF Opioids

March 24, 2016

By Kurt R. Karst

Ever since FDA Commissioner Dr. Robert M. Califf's February 2016 call for a sweeping review of FDA policies concerning opioids (in response to growing pressure from Congress), the Agency has been on a tear to meet that call with new policies and requirements.  There's the recent announcement that the Agency will require class-wide safety labeling changes for immediate-release opioid pain medications (as well as a related citizen petition response – Docket No. FDA-2014-P-0205).  And now FDA is tackling generic opioids. 

On March 24, 2016, FDA announced (Docket No. FDA-2016-D-0785) the availability of a highly-anticipated draft guidance document concerning abuse-deterrent Solid Oral Dosage Form (“SODF”) generic opioids: “General Principles for Evaluating the Abuse Deterrence of Generic Solid Oral Opioid Drugs.”  Publication of the draft guidance is now one of the items we can check off of the Calendar Year 2016 Guidance Agenda FDA published in January 2016. (This blogger is still sitting on the edge of his seat watining for other guidances identified in the “Generics” category of the Guidance Agenda, including “180 Day Exclusivity: Guidance for Industry;” “Three-Year Exclusivity Determinations for Drug Products;” “Submission of ANDAs for Certain Highly Purified Synthetic Peptide Drug Products;” and “Determining Whether to Submit an Application Under 505(b)(2) or 505(j).”)

The draft guidance, which is a nice complement to the brand-side guidance – “Abuse-Deterrent Opioids: Evaluation and Labeling” (see our previous post here) – lays out the principles for ANDA applicants to evaluate the abuse-deterrence of a proposed generic version of a brand-name Reference Listed Drug (“RLD”) with labeling that describes properties that are expected to deter misuse or abuse. We’ve seen things here and there over the past few years giving us some vague sense of FDA’s thinking on how an ANDA applicant might demonstrate that a generic SODF opioid drug product is no less abuse-deterrent than the RLD with respect to potential routes of abuse. . . but nothing solid . . . just bread crumbs. Consider, for example, FDA’s Draft Bioequivalence Guidance for generic EMBEDA (morphine sulfate/naloxone) Extended-release Capsules (NDA 022321), which recommends a “fasting, crushed drug product” study to allow “for the assessment of Naltrexone bioequivalence in a potential abuse situation.”

For now, the draft guidance provides testing recommendations for four of the seven categories of abuse-deterrent technologies described in FDA’s “Abuse-Deterrent Opioids: Evaluation and Labeling” guidance: solid oral opioid drug products formulated to incorporate physical or chemical barriers, agonist/antagonists, aversive agents, or combinations of two or more of these technologies (but not products in the delivery system, NME/prodrug, or novel approaches categories). For these technologies, and where the RLD labeling describes abuse-deterrent properties, FDA recommends that ANDA applicants conduct a comparative evaluation of the abuse potential of the proposed generic test product (“T product”) to the RLD product (“R product”) – sometimes including a control product (“C product”) – taking into consideration all potential routes of abuse (i.e., injection, ingestion, insufflation, and smoking) according to the following five principles:

  • Tier-based approach to testing. FDA recommends that potential ANDA applicants follow a tier-based approach to efficiently compare a T product to its R product and limit the number of tests required for evaluating the abuse deterrence of T product. This tier-based approach allows for hierarchical testing, starting with simple and gentle manipulations of the product in in vitro studies (Tier 1) and progressing to more destructive mechanical and chemical manipulations until R product’s abuse deterrence is defeated or compromised, or T product is shown to be less abuse-deterrent than R product.
  • Evaluation of Abuse Deterrence. The evaluation of the abuse deterrence of the T product should be based on its performance relative to R product. The proposed generic product need not have the same formulation design as the R product. In order to adequately compare R and T products, a potential ANDA applicant should identify the R product’s abuse deterrence for all routes of abuse using the tier-based approach described in this guidance. If R product has not been found by the potential applicant to have any abuse deterrence for a particular route of abuse, the potential applicant should summarize the studies conducted and the results to support the applicant’s assessment that the RLD has no abuse deterrence with respect to that route and explain why there is no need to test its T product in comparative in vitro or other studies for that route. The evaluation of the abuse deterrence of T product should be based on the potential applicant’s best understanding of the abuse deterrence of R product, the potential routes of abuse, and specific measures meaningful to the evaluation of abuse by those routes. . . .
  • Use of control.  Manipulation of an opioid product is a function of several factors including, but not limited to, tampering skills, time, and tampering resources available. The abuse-deterrent properties of currently approved drug products are not absolute, and can eventually be compromised or defeated. Therefore, it is important to identify appropriate discriminatory study conditions to compare R and T products. For certain comparative studies (e.g., extractability studies), such discriminatory study conditions should be identified by including a [C product] and comparing it to R product in order to identify the abuse deterrence of R product. Potential ANDA applicants should select an appropriate C product for their proposed T product. When available, C product should be a non-abuse-deterrent version of the opioid R product that contains the same active pharmaceutical ingredient (API) as the R product.
  • Identification of discriminatory study conditions. The parameters for the discriminatory study conditions should lie within the range specified in this guidance for different routes of abuse (Appendices 2-5). In order to determine the abuse deterrence of T product by, for example, the injection route, a potential ANDA applicant should first identify the in vitro discriminatory study conditions under which the % extraction of opioid from R product is statistically less than the % extraction of opioid from C product, i.e., the conditions under which R product is statistically superior to C product. The potential applicants should then compare the % extraction of opioid of T product to R product under the same discriminatory study conditions.
  • Comparison of R and T products. Once the in vitro discriminatory study conditions have been identified, a potential ANDA applicant should perform the recommended statistical comparisons for each of the different routes of abuse as recommended [elsewhere in the draft guidance].

The tier-based approach to testing is illustrated in a decision tree for evaluation of the extractability of opioid from an intact product for ingestion:


Although in vitro (mechanical and chemical manipulation) studies are the default studies FDA expects ANDA applicants to conduct, the Agency recognizes that in some instances there may not be a reliable in vitro testing methodology. In those cases, FDA recommends pharmacokinetic studies (subject to FDA’s review of a protocol). Other types of studies are not recommended except in cases involving an excipient. For example, writes FDA, “in comparing the abuse deterrence potential of an excipient that functions as an aversive agent, FDA may recommend that applicants conduct pharmacodynamics studies with drug liking as a comparative endpoint between the R and the T product to permit FDA to evaluate formulation equivalence.” As new technologies emerge, FDA will use an adaptive approach and revise recommendations and develop new ones as needed.