We don’t deal with FDA regulatory matters that often, and with the criminal side even less, but the recent decision out of Colorado in U.S. v. Bader, slip opinion here, was too juicy to ignore.
It points out a serious flaw in the FDA’s regulatory scheme that we think needs a formal regulatory fix – and soon.
The issue in Bader is the limits on pharmacy compounding.
In Bader a pharmacist purchased bulk human growth hormone (“HGH”) – although it could just as easily been some other type of drug (maybe steroids, often used for similar purposes?) – a drug that’s subject to severe restrictions when distributed as a drug in the USA.
Anyway, the defendant bought HGH in bulk form and from our point of view, didn’t appear to do much to it. As stated in the opinion, he:

inspected the bulk HGH powder he had received for quality and potency, measured out the proper amount of bulk HGH powder into a single-dose container, separately packaged an appropriate quantity of saline for the consumer to mix with the powder, labeled and packaged the HGH dosage and the saline together, and supplied the two items to the customer. (The customer was responsible for actually mixing the saline and HGH before administering the drug.)

Slip op. at 4.
After that the defendant pharmacist was open for business, able to fill prescriptions in avoidance of otherwise strict regulations, because “the FDA rigorously regulates the importation and distribution of finished drugs that are ready for distribution to consumers, but exercises relatively little regulatory oversight over the importation of drug ingredients to be used by pharmacists to create ‘compounded’ drugs.” Slip op. at 3.
So the FDA busted this pharmacist, claiming that what he was doing wasn’t “compounding,” but merely “repackaging” of a drug that was effectively in consumer-usable form. Id,
Problem is, after all these years, the FDA still doesn’t have any formal definition of “compounding” in its regulations. So it had a hard time distinguishing “compounding” from “repackaging.”
Here’s why. Over the years the FDA has deferred to state definitions of “compounding.” Slip op. at 6 (referring to “the FDA’s traditional deference to state regulation of the practice of compounding”). Well, it just so happens that, in Colorado anyway, the state has defined “compounding” to include repackaging. Slip op. at 5.
Oops. The FDA’s failure to have its house in order on defining “compounding” comes back to bite it (you decide where) in a big way.
Time to change gears. The Agency next attempts to fit pharmacy compounding (at least in this one instance) into the definition of “manufacturing”. Slip op. at 8. Its basis: compounding technically falls into the definition of a “new drug” – subjecting its maker to regulation as a manufacturer – even though the FDA hasn’t enforced the Act in this fashion for half a century:

[P]racticality aside, the FDCA contains no exception to the definition of a new drug for one that is compounded. Thus, on its face, the FDCA is susceptible to an interpretation that would require all compounded drugs to go through the NDA process. For half a century, practicality prevailed – the FDA simply declined to exercise any authority it might have exerted over compounded drugs, electing to defer to the state regulation of pharmacists who engaged in the practice of compounding.

Id.
Congress mucked things up further by adding pharmacy compounding provisions to the FDC Act in 1997 in something called FDAMA. Those provisions exempted compounded drugs from the Act’s new drug requirements if:

• compounded by a licensed pharmacist pursuant to a valid prescription issued to an individual patient (or is prepared in limited quantities in anticipation of receiving a prescription based on prior business history)
• the pharmacist complies with industry standards set forth in the U.S. Pharmacopeia or other sources (such as state law)
• the compounded drug is not one that appears on an FDA list of specific drugs deemed to be unsafe (maybe steroids are on this list)
• the pharmacist doesn’t churn out “inordinate amounts” of a copy of a commercially available drug
• the drug is not one that the FDA has specifically designated as being unsuitable for compounding (maybe steroids are on this list)
• the pharmacy is not producing “inordinate amounts of compounded drug products” generally, and
• the pharmacy doesn’t advertise or promote the compounding of any particular drug or class of drugs.

Slip op. at 8-9, citing various parts of 21 U.S.C. §353a.
The last bullet point – prohibiting all advertising, truthful or otherwise, was the kicker. It got this part of FDAMA declared unconstitutional in Thompson v. Western States Medical Center, 535 U.S. 357 (2002), on First Amendment grounds. We’ve discussed these First Amendment issues here, in the context of off-label use. So Congress, trying to clarify, just made things worse.
Bader, of course, has nothing to do with the First Amendment. Trouble is, the Ninth Circuit, in a part of its ruling that the Supreme Court didn’t touch in Thompson, declared the whole kit and caboodle of FDAMA’s compounding provisions not just unconstitutional, but unseverable – which means all of it fell together. Western States Medical Center v. Shalala, 238 F.3d 1090, 1096-98 (9th Cir. 2001). This is all discussed, slip op. at 9.
After Thompson, the FDA rejiggered its informal guidance concerning compounding, but although it initiated some administrative action on the First Amendment in response to Thompson, but never followed through on that. The Agency did state, in 2002, “that it intended to proceed as if FDAMA was void in its entirety.” Slip op. at 14-15 (quoting FDA guidance).
So the FDA’s administrative sloth in not wanting to confront the First Amendment implications of Thompson comes back to haunt the Agency in Bader. It’s stuck with the Ninth Circuit’s unconstitutionality ruling, and only an informal guidance – not having force of law – that addresses what constitutes compounding. And what’s worse, that guidance “emphasizes that the FDA will continue defer to state regulation of the traditional practice of compounding.” Slip op. at 11.
It gets screwier. There’s a second decision, Medical Center Pharmacy v. Mukasey, 536 F.3d 383, 394 (5th Cir. 2008), that according to the Bader court: (1) disagreed with the Ninth Circuit’s severability ruling, and (2) construed the statute/regulations to mean that compounded drugs that complied with the provisions of the now-severable FDAMA, were not subject to the NDA process at all. Slip op. at 13.
Colorado, of course, isn’t in either the Ninth or the Fifth Circuits, but rather is in the Tenth.
So what is the status of a severely restricted drug, imported as a bulk ingredient, and then reduced to individual-sized units, without really changing its composition, in a way that avoids those legal restrictions?
There were three possible positions: (1) that FDAMA was unconstitutional, so there was no exemption and compounded drugs were “new drugs,” thus requiring compounders to be regulated as “manufacturers” (the FDA’s position); (2) that FDAMA was unconstitutional, so compounding was relegated to the FDA’s prior position for half a century of deferring to the states (the defendant’s position); and (3) FDAMA was severable, and the unconstitutional parts created an exemption, provided certain criteria were met. That’s the Fifth Circuit’s view, but not advocated by either party in Bader. Slip op. at 13-14.
The court chose option (3), slip op. at 17, even though that effectively meant the government can renege on its 2002 public statement that it considers the compounding aspects of FDAMA void in their entirety.
After all that, what’s the definition of “compounding”?
Good question.
The FDA started out arguing that what the defendant did was mere “repackaging.” But the upshot of the court’s ruling on the vampire-like status of FDAMA (rising from the dead) means that what the defendant pharmacy made could be considered a “new drug” and the defendant would be regulated as a manufacturer.
But there’s still the uncomfortable fact that nowhere – not in the pre-1997 FDCA, not in FDAMA, and not in the FDA’s guidances – is any attempt made to define “compounding.” Slip op. at 19-20. In the absence of any definition provided by Congress or formally embodied in an FDA regulation, the court wasn’t about to let the Agency gin something up ad hoc on the basis of warning letters and other informal actions. Slip op. at 20.
In default of any federal definition, the states had their own, varying definitions. Slip op. at 20-21.
So the court decides that the Colorado definition of “compounding” – which includes simple “repackaging” – applies. Slip op. at 21-22.
But….
The FDA can superimpose its FDAMA definition of what’s a “new drug” on top of this patchwork of state regulations. Thus it can regulate some compounding pharmacies as “manufacturers.” Slip op. at 22-23.
That means, in the circumstances of the Bader case that the defendant first has to meet the Colorado definition of “compounding” that includes “repackaging” (which even the court thinks he can do, slip op. at 24), but then must also meet the previously bullet pointed FDAMA requirements – except for the last one, which was declared unconstitutional in Western States. Slip op. at 24-25.
So what do we think?
First, it’s a morass. The FDA should undertake formal rulemaking to define “compounding” in a way that’s not unconstitutional under Western States. The state standards are designed for other purposes and potentially allow not only HGH but other highly restricted drugs to be prescribed freely through “compounding” in a way that nullifies the otherwise applicable restrictions, which presumably are imposed for weighty public health reasons.
Second, can compounding possibly be considered “manufacturing” – especially in the Bader case? As non-regulatory lawyers, that seems completely counterintuitive, since “manufacturing” connotes whipping the stuff up from scratch – the opposite of what happened here. The defendant pharmacist did nothing to the HGH except transfer it from bulk form to patient-ready doses. It makes more sense to consider it “repackaging” and regulate it as such.
Third, we have serious doubts whether the court’s analysis in Bader, regardless of the correctness of its lengthy discussion of the administrative antecedents, can fly in a criminal prosecution. The upshot is that the the government, as prosecutor, gets to apply a different definition to the defendant’s conduct after the fact than it had enunciated at the time. That sounds like ex post facto to us – although, again, we’re not criminal lawyers, and maybe there’s some nuance or exception that we don’t know about.
Thus, it all comes back to number one. The FDA needs to clarify the law applicable to compounding so that it can enforce it in something other than the Rube-Goldberg manner that’s going on in Bader.