Opioid REMS: FDA Sees No Clear Verdict On Risk Management Program's Impact

The most recent evaluation of the extended-release/long-acting opioid products REMS reflects an ongoing and frustrating question for both FDA and the biopharma industry: How best to determine whether the risk management programs are truly working?

FDA's analysis of the assessment for the ER/LA Risk Evaluation and Mitigation Strategy suggests that three years after voluntary training for opioids prescribers began under the massive program, the agency still does not know whether the REMS is accomplishing its goals of reducing overdose, addiction and death that results from inappropriate prescribing, misuse and abuse of the powerful pain drugs.

In agency briefing documents released ahead of a May 3-4 advisory committee review, FDA finds shortcomings in various methods used by the opioid product sponsors to assess elements of the shared system REMS.

FDA called the results of prescriber and patient surveys about knowledge and understanding of opioid risks "encouraging" but said they may be biased due to issues of comparability and generalizability.

Surveillance studies suggested encouraging downward trends in some, but not all, clinical outcomes, FDA said. However, these studies had considerable methodological limitations, and the impact of the REMS could not be isolated.

"The lack of studies that directly examine associations between participation in REMS training and changes in clinical practice or patient outcomes limits the ability of these studies to evaluate the effectiveness of the REMS and guide specific changes to the program," the agency said.

FDA will ask members of two advisory committees – Drug Safety and Risk Management, and Anesthetic and Analgesic Drug Products – about the appropriateness of the data sources and methodologies used to evaluate the ER/LA opioid analgesics REMS, alternative measures of success and how the impact of any future changes to the program should be assessed.

The evaluation could affect decisions on whether to expand the ER/LA REMS to immediate-release opioids. The agency recently added a Medication Guide to immediate-release products and said the issue will be on the table during the advisory committee meeting (Also see "Opioids And The Limits Of FDA Power" - Pink Sheet, 22 Mar, 2016.).

Big Group Gets Bigger

Originally approved in July 2012, the REMS currently includes 26 companies that hold NDAs or ANDAs for ER/LA opioids.

However, the group is getting a new member. On April 26, FDA granted final approval for Collegium Pharmaceutical Inc.'s Xtampza ER, an extended-release version of oxycodone with abuse-deterrent properties. The drug had received tentative approval in November, but full marketing approval was held up by patent litigation (Also see "Keeping Track: Approvals For Genentech And AstraZeneca, CR Letter For Neos, And Breakthrough For Teva" - Pink Sheet, 16 Nov, 2015.).

The class-wide REMS consists of elements to assure safe use in the form of voluntary prescriber training, which is conducted through accredited continuing education (CE) programs with funding provided by the ER/LA product sponsors. The REMS also includes a Medication Guide and a timetable for submission of assessments.

The advisory committee review will focus on findings from the fourth assessment conducted under the REMS and the first to address specific numeric goals for REMS-compliant training. Whether the prescriber training should remain voluntary or be made mandatory will be a topic for panel discussion (Also see "FDA's Opioid REMS Review Puts Brakes On Mandatory Education Push" - Pink Sheet, 26 Apr, 2016.).

Provider And Patient Surveys …

The REMS includes eight assessment elements, with prescriber knowledge broken down into two parts (see chart).

The prescriber survey was intended to compare the awareness and understanding of the serious risks and appropriate prescribing practices for ER/LA opioids among prescribers that took the REMS-compliant CE training with those prescribers that did not. It was a follow-up to a 2013 baseline survey of 605 ER/LA opioid prescribers conducted prior to implementation of CE training.

A total of 612 prescribers completed the follow-up survey. Over half of the respondents were recruited from IMS data, while the remainder were invited by CE providers. Approximately 60% reported they had completed REMS-compliant CE training.

"Overall, respondents were knowledgeable about the assessment, management, and counseling requirements for patients being considered for treatment or currently being treated with an ER/LA opioid analgesic," FDA said. "Respondents were less knowledgeable about initiation, modification, and discontinuation of therapy, and general and product-specific information for ER/LA opioid analgesics."

Compared to the baseline survey, overall response rates to 44 items improved, 17 remained the same, and four items decreased, FDA said. Overall awareness of REMS materials was low.

The long-term evaluation survey was intended to evaluate prescribers' knowledge about ER/LA opioids, completion of the REMS processes, and changes in behavior, prescribing and patient-assessment practices for prescribers who completed a CE activity within the last six to 12 months. Prescribers were recruited through CE providers, and a total of 328 completed the survey.

As with the prescriber survey, overall respondents in the long-term evaluation were knowledgeable about patient management and counseling requirements, but less so about patient assessment, initiation and modification of treatment, and general and product-specific information for ER/LA opioids.

"While over half of respondents reported no changes in prescribing behaviors since participating in the CE activity, 22% reported writing prescriptions for ER/LA opioid analgesics less often and 19% reported writing more ER/LA opioid analgesics prescriptions," FDA said. "Thirty-eight percent (38%) of respondents reported prescribing more non-opioid medications since the CE activity while 23% reported limiting which ER/LA opioid analgesics they prescribe. Thirty-two percent (32%) of respondents reported no changes in the types of medications prescribed since the CE activity."

A patient survey was intended to assess knowledge of the safe use of ER/LA opioids following implementation of the REMS. Patients were identified from a medical and pharmacy claims database. A total of 423 patients completed the survey, for a response rate of 17%.

"Overall, respondents had a high understanding of the key risk messages, though the survey respondents were not representative of the drug use population," FDA said, noting that Medicare and Medicaid patients were not included. "There was a lower understanding of aspects of safe storage and using the drug safely."

… Draw Critiques On Comparability, Generalizability

FDA statistical reviewers noted the two prescriber survey efforts "did not attempt to survey the same individual prescribers or to make respondents of these two surveys comparable."

"Comparability is important when assessing a causal question such as the effectiveness of REMS-compliant CE training on prescriber knowledge," the agency said. "In the analyses of follow-up prescriber survey, the applicant holder did not present pairwise comparisons of the demographic and clinical characteristics of respondents in the different groups of interest. … If the two groups in any pairwise comparisons do not have similar characteristics and the results are not adjusted, we cannot attribute any difference in the group scores to REMS-compliant CE training alone."

Agency staff also raised generalizability as a concern for all three surveys, noting for example that prescribers choosing to take the CE training may differ from the ER/LA opioid prescriber population in general.

"All these surveys were convenience samples (respondents were self-selected and therefore may differ with the general population of interest or target population) and had high non-response rates," FDA said. "The results may be biased and cannot be generalized to all ER/LA opioid analgesic prescribers or to the general population of patients who were prescribed ER/LA opioid analgesics."

FDA's statistical review recommended that future analyses should compare characteristics of participants to each survey's target population and propose methods to standardize the results of each survey to its target population.

Surveillance Studies

Studies submitted to satisfy a surveillance monitoring requirement suggested mixed results on adverse outcomes with ER/LA opioid use.

Studies showed decreases in the rate of poison center exposure calls involving opioids and prevalence of self-reported recent abuse of opioids among those entering addiction treatment. However, both of these downward trends began before the REMS was implemented. In addition, self-reported non-medical use of opioids increased among college students, and the incidence of emergency department visits and hospitalizations for prescription opioid overdose did not change significantly after the REMS came online.

FDA concluded the surveillance data provide little evidence that the REMS itself has resulted in changes in safety outcomes of interest.

"The temporal relationships between the REMS implementation and the observed reductions in some outcomes do not suggest that the REMS was the cause of these changes," FDA said. In addition, "it is unknown how many providers would need to be trained, and to what extent the training would need to change clinical practice and prescribing behavior, for a measurable effect on these surveillance outcomes to be plausible."

There are many alternative explanations for the trends seen in the surveillance studies, which were unable to isolate the effects of the REMs from other concurrent efforts to reduce inappropriate prescribing, abuse and misuse, FDA said.

"No study directly evaluates the association between participation in REMS trainings and changes in clinical practice, prescribing behaviors, or patient outcomes," FDA said.

Importantly, however, the surveillance studies also do not demonstrate that the REMS has been ineffective or is failing to make progress toward its goals.

"Even if the REMS educational messages were having meaningful desired effects on behavior in prescribers and patients receiving the education, it is doubtful that these studies would be capable of detecting the impact of these behavior changes on the measured surveillance outcomes," FDA said.

"To assess the effects of the REMS provider trainings directly, pre-post changes in prescriber behavior and/or patient outcomes for a group of providers who participate in the REMS training would need to be compared to those in a group who do not participate in the training," FDA said. "Conducting such a study would be challenging and resource intensive, but the feasibility of this type of investigation should be explored if more rigorous evaluation of the impact of the ER/LA opioid analgesic REMS is needed."

FDA also said better strategies are needed to assess drug use patterns, prescribing behaviors and patient access to the drugs.