Notepad: FDA Labeling and Outcome Data on Buprenorphine in OUD

If there is one unanimous tsunami in health policy, it may be the use of Medication Assisted Therapy in opioid use disorder (MAT/OUD).

• In the Annals of Internal Medicine, see Beetham (2019) on buprenorphine access in high OUD mortality areas here, see Martin (2018) on "next stage of buprenorphine care," here, see  Chou (2019) on buprenorphine and opioid tapering here.
• See also in AIM, Barocas (2019) on policy and therapy in OUD, here.  Paired with a cost effectiveness article by Murphy et al. (2019) here.
• In NEJM, see Wakeman (2018) on "myths and realities," primary care and buprenorphine here, see Saloner (2018) moving buprenorphine to mainstream here.
• In JAMA, see Mark (2019) on CMS, FDA, and prior authorization policies for buprenorphine here, Roy (2019) on over the counter non-Rx buprenorphine here, Petz (2019) on nurse practitioner Rx for buprenorphine here.   See Olfson (2019) on trends in buprenorphine prescriptions 2011-2018 here.
• In JAMA Psychiatry, see Fiscella (2019) on the urgent need for buprenorphine deregulation here.
• In Health Affairs, see Harris (2020) on gaps in opioid treatment access for Medicare patients; defines opioid treatment access = MAT.  Here.
• In NEJM, see Haffajee (2020) on delayed competitors to Suboxone here.  Includes citations to November 2019 FDA action to revoke orphan status.  As is my point with this selection of articles, Haffajee takes effectiveness for granted, with one citation to a short-term observational study, Larochelle 2018; for example, you can't assume patients entering methadone clinics for five months and patients declining to start them, are the same.  (For another observational study see Morgan 2019.  Morgan 2018 found drug effectiveness but discontinuation rates are very high.)

The above are for-example-only, easily picking obvious articles from a few top-ranked medical journals.

See similarly a high-profile lead editorial at New York Times, 2019, here.  Although it's about methadone, not buprenorphine, see also Health Affairs, 2019, here.

Pivoting to FDA Labeling

CMS New Policy for OUD Payments.

In rulemaking for new OUD benefits in opioid treatment programs (OTPs; a term of art currently usually meaning methadone centers), CMS proposed separate payment levels depending on the patient's MAT (see CMS, 84 Fed Reg 40529, 8/14/2019, PFS CY2019 rulemaking proposal).  

There would be one code for methadone per week; one code for oral buprenorphine per week; one code for injectable buprenorphine per week, one code for monthly depot injection buprenorphine.   Pricing for drug is found on Table 15 (page 40537) and ranges from $22 for methadone to $4791 for buprenorphine monthly implant insertion.   

FDA Labeling for Buprenorphine Formulations: Clinical Studies  On-Label

Often FDA labeling is taken as a gold standard.  Claims and usage that are "off-label" may be decried in various public policy forums.   

It's easy to find the FDA labeling for buprenorphine sublingual tablets (e.g. SUBUTEX; approved 2002; label updated to 2018) label here.   It's easy to find the FDA labeling for buprenorphine depot injection; (e.g. SUBLOCADE; approved 2002 (original drug), updated 2017); label here.   See the FDA press release for SUBLOCADE, 2017, here.

The FDA-labeled clinical data for SUBUTEX is very weak, and non quantitative, with no charts, graphs, or tables of outcome data (such as percent change in any outcomes.)

The labeled clinical data for SUBLOCADE includes two studies.  One is an observational study of self reported opioid high while on drug (drug-liking analog visual scale) in non-treatment seeking opioid patients.   

The second SUBLOCADE study was an RCT against placebo with a 24 week trial period and follow-on data to 24 months.   Data is shown in Figure 12.   SUBLOCADE depot markedly raised the opioid-free weeks in the first six months of treatment (about 60% instead of about 5-10%).  Good.  However, by two years, the two groups are about the same.   

And note two things: SUBLOCADE was compared to placebo, not methadone or oral buprenorphine; and SUBLOCADE had only a very small differential effect at two years.   If you based your decision-making on this FDA table alone, SUBLOCADE does not look like a very good treatment for 2, 3, or 4 years or longer.   Yet many of the articles above are about the need to get people on buprenorphine for a long time, most often by comparing OUD as a chronically medicated disease like diabetes.

FDA label, SUBLOCADE (for link see text)

Comment

Medicare proposes to pay $115,000 for each two years of SUBLOCADE treatment.  If you used FDA labeling as a gold standard, it's impossible to make a clinical or pharmacologic argument for why this is better than generic buprenorphine.   

FDA itself uses FDA labeling as a gold standard, as in 2019 actions against labs doing pharmacogenetic testing that was not included in FDA labeling.  Here.  If you used FDA labeling as a gold standard, you would not be doing much pharmacogenetic testing, but you would not be giving much depot buprenorphine, either.

___

The SUBLOCADE advisory committee meeting in 2017, here.  See the actual data links for two 2017 meetings here.   On October 31, FDA reviewed depot buprenorphine; on November 1, FDA reviewed injectable buprenorphine.  At the link, which is a long page of many links, look for the actual data at the term "Briefing Materials" for each day.

Per FDA's press release, the SUBLOCADE product had priority review and fast track designations.  INDIVIOR holds the drug license.

___

For a Suboxone marketing probe, 2019, here.

___

For an earlier blog on sublingual buprenorphine alone, here.

___

In its own press release for SUBLOCADE, FDA refers to off-label data.  In stating that buprenorphine-assisted treatment may cut death rates in half, FDA does not refer to its own drug labeling but to NIH data.  Here.

___

For a 2005 report on MAT, still online at ASAM, here.

___

For a general 2019 review,  Bell & Strang, Biol Psychiat here.  See Rosenthal 2017 for advances in buprenorphine delivery formulations, here.

___

For a rare example of medication-assisted therapy getting negative press, see an FDA warning letter to Alkermes in December 2019 regarding improper marketing of Vivotrol (depot naltrexone) - here.

____

One good article on the value of MAT is Wakeman et al., comparative effectiveness, real world evidence, JAMA Network Open 3:e1920622 here.  Effectiveness of MAT vs non-MAT was assessed for 3 months and 12 months; the AHR (adjusted hazard ratio) at 12 months was .74 relative risk in harm.   The authors note, "The most common treatment pathway was nonintensive behavioral health (24 258 [59.3%]), followed by inpatient detoxification or residential services (6455 [15.8%]) and buprenorphine or methadone (5123 [12.5%]). Not receiving any treatment was more common (2116 [5.2%]) than naltrexone (963 [2.4%]) or intensive behavioral health (1970 [4.8%])."
\
____

ICER published a negative cost effectiveness review of advanced MAT drugs, in December 2018, here.

____

THere is a very complex story about the legal marketing status of MAT drugs and special formualations, reviewed in Health Affairs in March 2020 here.