Nearly six years ago, in 2015, the FDA attempted to slip a change to its “intended use” regulations (21 C.F.R. §§201.128, 801.4) – which had not been updated since the 1950s – through the administrative process by hiding it in a Federal Register notice about electronic cigarettes.  80 Fed. Reg. 57756 (FDA Sept. 25, 2015).  We caught them, and the FDA then faced the prospect of responding to public comments about the First Amendment and truthful “promotion” of off-label use.

Although the method raised our eyebrows, the proposed change itself was (and is) quite necessary.  The FDA proposed to delete problematic (albeit largely unenforced) language from those regulations, which purported to make mere knowledge of off-label use a basis for changing a product’s “intended use,” setting off a cavalcade of regulatory consequences, including the product being declared misbranded and adulterated.

As we’ve pointed out elsewhere, these long-outdated “intended use” regulatory provisions are the linchpin in the agency’s prohibition of any “promotion” of off-label uses – even truthful and scientifically valid off-label information.  As we’ve discussed many times, a flat prohibition of a particular class of speakers’ (regulated product manufacturers) truthful speech on a particular topic (off-label use of their products) has become increasingly problematic under the First Amendment, given a string of Supreme Court decisions, most notably:  Reed v. Town of Gilbert, 576 U.S. 155 (2015) (content-based and speaker-based governmental restrictions are subject to strict scrutiny); Sorrell v. IMS Health, Inc., 564 U.S. 552 (2011) (pharmaceutical detailing is First Amendment protected); and Thompson v. Western States Medical Center, 535 U.S. 357 (2002) (striking down similar FDA prohibition of a different category of truthful speech by regulated entities).  Ever since the FDA lost United States v. Caronia, 703 F.3d 149 (2d Cir. 2012), the government has avoided appellate review of truthful off-label promotion cases, with good reason.

Back to the administrative front.

On January 9, 2017, on the eve of a change of administrations, the FDA tried a bait-and-switch “clarification” of its proposed amendments.  Instead of simply deleting the problematic language (“with the knowledge of such persons or their representatives”), the FDA retained mere knowledge and folded it into a brand new – and never commented on – “totality of the evidence” test for deeming the creation of a new intended use.  82 Fed. Reg. 2193, 2206 (FDA Jan. 9, 2017).  This would-be final rule ignored the First Amendment problems with continued prohibition of truthful, non-misleading off-label information by claiming the First Amendment would be dealt with in parallel proceedings.  Id. at 2209.  Simultaneously, however, the FDA released a “memorandum” enunciating essentially the same crabbed view of the First Amendment that had failed in Western States and Caronia (and elsewhere).  We discussed that, here.

Those parallel proceedings never amounted to anything.

The 2017 proposed rule was delayed, and then withdrawn, before it ever went into effect, as we discussed here.  Thereafter, the FDA did essentially nothing for the rest of TFG’s administration, until September 23, 2020, when FDA reissued a proposed rule that was similar to the original 2015 proposed rule, in that the “totality of the evidence” was dropped.  In the 2020 proposal, a regulated entity’s knowledge of off-label use would not be used against it in determining a new intended use.  85 Fed. Reg. 59718 (FDA Sept. 23, 2020). We discussed that proposal here.

Well, the 2020 proposed rule has now been finalized, but unfortunately, totality of the evidence is still there – just in the comments, rather than the actual language of the amended regulations.  We note the response to comment two.

[I]n determining a product’s intended use, the Agency may look to any relevant source of evidence, including a variety of direct and circumstantial evidence.  FDA’s position is reflected in the notices issued in this rulemaking over the past 5 years, and has been noted in court decisions.  This position has also been explained in numerous litigation briefs and other FDA pronouncements. . . .  One of the purposes of this rulemaking is to put to rest any dispute about FDA’s interpretation of its statute and regulations, and its policy—as embodied in this rule as well as in the precedent cited above—regarding evidence that may be relevant to establishing intended use.

Id. at 41388-89 (citations and bullet points omitted).

This response strikes us as highly unusual, particularly given the long and convoluted history of this regulatory change.  (A) The FDA proposes a relatively simple change.  (B) The FDA springs “totality of the evidence” on everyone.  (C) The agency retreats.  (D) The FDA deletes “totality of the evidence” from the regulatory proposal.  (E) Finally while omitting the concept from the language of the regulation itself, the FDA resuscitates totality of the evidence in its response to comments.  To us, all this seems like playing fast and loose with the whole notice and comment procedure.  Having explicitly proposed a “totality” test in 2017, and then withdrawn it under fire, we don’t think the new regulatory language can properly be interpreted to include by implication what was explicitly deleted.  Nor do we think the FDA is likely to find much judicial deference, given this regulatory history.  Recall that the FDA was slapped down by the District of Columbia Circuit on another off-label issue not long ago.

We also think that it’s high time that the FDA joined the rest of the world and start calling off-label use “off-label use.”  The phrase this rule uses is a clumsy mouthful – “unapproved use of a medical product that is approved, cleared, granted marketing authorization, or exempted from premarket notification.”  85 Fed. Reg. at 41385.

Finally, with respect to the First Amendment itself, the FDA doesn’t seem to have learned anything from the last decade.  The great majority of the cases they cite involve illegal distribution of drugs and devices for recreational or other non-therapeutic purposes.  Id. at 41387.  Heck, those are the cases the FDA should win – we want it to win − and there’s something wrong if the agency doesn’t.  However, those aren’t the cases that concern us.  The off-label promotion this Blog has supported is truthful, therapeutic, and supported by scientific evidence.  When is the last time the FDA has won one of those?  Or even prosecuted one of those (Maybe here)?

The FDA discusses using “truthful” speech as “evidence” of intent.  Id. at 41391-91.  In the situations we’re concerned with, that distinction is well nigh meaningless.  Conducting, say, medical education concerning off-label uses, is solely speech.  There is no need for “evidence” of intent in these situations.  Intent is admitted.  Such speech is intended to inform medical personnel about off label uses.  In these situations, it is likely that the FDA – if and when challenged – will lose the First Amendment argument.

We were particularly interested in how the FDA would address the Reed line of First Amendment cases, since the agency’s ban on truthful speech about off-label uses is unquestionably both content and speaker based.  That ban is not directed against any subject except off-label use, and it is not directed against any speakers other than those the FDA regulates.  We find the FDA’s position in its response to comment 15.  Id. at 41393-94.  There are four relevant paragraphs.

In paragraph one, the FDA points out that neither Sorrell nor any of several circuit court opinions have “overruled” Central Hudson Gas & Electric Corporation v. Public Service Commission, 447 U.S. 557 (1980).  85 Fed. Reg. at 41394.  That’s a strawman.  Even we don’t think the Gilbert line of cases overruled Central Hudson.  Rather, the plain implication of these cases is that they limit Central Hudson – excluding from its reach truthful topic-based and speaker-based speech.  The Central Hudson test, after all, does not ask who the speaker is or the subject of the speech:

In commercial speech cases, then, a four-part analysis has developed.  At the outset, we must determine whether the expression is protected by the First Amendment.  For commercial speech to come within that provision, it at least must concern lawful activity and not be misleading.  Next, we ask whether the asserted governmental interest is substantial.  If both inquiries yield positive answers, we must determine whether the regulation directly advances the governmental interest asserted, and whether it is not more extensive than is necessary to serve that interest.

447 U.S. at 566.  There’s nothing in this test inherently contradicting the Reed line of cases reaffirming strict scrutiny of commercial speech restrictions based on subject matter or who is speaking.

Paragraph two admits, as it must, that Reed itself “applied strict scrutiny to content-based restrictions,” which the FDA characterizes as “noncommercial.”  85 Fed. Reg. at 41394.  That’s whistling past the graveyard if we’ve ever heard it.  Just as we reviewed Central Hudson, we also reviewed Reed.  One of the subject-matter restrictions struck down in Reed was precisely a distinction between commercial and noncommercial speech.  576 U.S. at 160 (“first” substantive category of struck-down ordinance was a “sign communicating a message or ideas for noncommercial purposes”).  Further, when the Reed Court discussed what kind of speech restrictions would be permissible, it did not draw any distinction between commercial and noncommercial subjects:

The Town has ample content-neutral options available to resolve problems with safety and aesthetics.  For example, its current Code regulates many aspects of signs that have nothing to do with a sign’s message: size, building materials, lighting, moving parts, and portability.  And on public property, the Town may go a long way toward entirely forbidding the posting of signs, so long as it does so in an evenhanded, content-neutral manner.

We acknowledge that a city might reasonably view the general regulation of signs as necessary because signs take up space and may obstruct views, distract motorists, displace alternative uses for land, and pose other problems that legitimately call for regulation.  At the same time, the presence of certain signs may be essential, both for vehicles and pedestrians, to guide traffic or to identify hazards and ensure safety.  A sign ordinance narrowly tailored to the challenges of protecting the safety of pedestrians, drivers, and passengers − such as warning signs marking hazards on private property, signs directing traffic, or street numbers associated with private houses − well might survive strict scrutiny.

Id. at 172-73 (citations and quotation marks omitted).  We don’t think it’s possible to tease a commercial/noncommercial distinction out of this discussion.

Paragraphs two and three parse two more recent Supreme Court decisions that did not command a majority on First Amendment issues:  Matal v. Tam, ___ U.S. ___, 137 S.Ct. 1744 (2017), and Barr v. American Ass’n of Political Consultants, Inc., ___ U.S. ___, 140 S.Ct. 2335 (2020).  85 Fed. Reg. at 41394.  As to Matal, the FDA raises the same strawman about overruling Central Hudson outright.  Yes, only Justice Thomas would have taken that step in Matal.  137 S. Ct. at 1769.  But only four justices would apply the Central Hudson test, as the FDA’s description admits.  The others reserved the question of strict scrutiny for a content-based commercial speech restriction.  Id. at 1764.  It should also be noted that the four justices in Matal whose opinion the FDA claims as support for its position included now former Justices Kennedy and Ginsburg, who are no longer on the Court.

Barr did not involve a content- or speaker-based restriction on speech at all.  Rather, Barr involved a form of conduct – use of robocall devices to make certain telephone calls.  140 S.Ct. at 2344 (“In plain English, the [statute] prohibited almost all robocalls to cell phones.”) (footnote omitted).  Justice Kavanaugh – who had not yet joined the Court in Matal – wrote the plurality opinion applying Reed in Barr.

As relevant here, a law is content-based if “a regulation of speech ‘on its face’ draws distinctions based on the message a speaker conveys.”  Reed, 576 U.S., at 163, 135 S.Ct. 2218.  That description applies to a law that “singles out specific subject matter for differential treatment.”  ” Id. at 169. . . .

Under [the statute], the legality of a robocall turns on whether it is “made solely to collect a debt owed to or guaranteed by the United States.”  A robocall that says, “Please pay your government debt” is legal.  A robocall that says, “Please donate to our political campaign” is illegal.  That is about as content-based as it gets.  Because the law favors speech made for collecting government debt over political and other speech, the law is a content-based restriction on speech.

140 S. Ct. at 2346 (emphasis added).  Thus, “[i]n short, the robocall restriction with the government-debt exception is content-based.  Under the Court’s precedents, a ‘law that is content based’ is ‘subject to strict scrutiny.’  Reed, 576 U.S., at 165.”  Id. at 2347.

Even without considering recently-elevated Justice Barrett’s views, it seems pretty clear after Justice Kavanaugh’s opinion in Barr (and we would say after Reed and Sorrell) that strict scrutiny of content-based restrictions on commercial speech commands a majority of the Court.

And what is the FDA’s First Amendment backup plan?

It doesn’t seem to have one.  Paragraph four of its Reed analysis, to use another metaphor, sticks the agency’s head in a hole in the ground – simply refusing to consider that its off-label promotion ban may be barred by the First Amendment:

Accordingly, given that the Supreme Court has not overruled Central Hudson or Wisconsin v. Mitchell and given that the laws being reviewed in the cited cases were quite different from the premarket review provisions of the [FDCA], we believe it would be wrong to conclude that the Supreme Court has implicitly but sweepingly reversed these long-standing precedents to invalidate the regulatory regime under the [FDCA].  And even if some form of heightened scrutiny were applicable to reliance on speech as evidence of intended use, FDA believes that the public health necessity of the premarket review provisions discussed in this preamble, including its references, justifies and necessitates this regime under any standard.

85 Fed. Reg. at 41394.

We have to assume that at least some of the commentators whose views the FDA just rejected have been giving some thought to their next legal steps.  We hope and expect that someone will challenge the administrative bona fides of the FDA’s on-again-off-again-on-again “totality of the evidence” approach to squelching truthful off-label speech.  We also expect to see test cases in the reasonably near future presenting, on favorable facts, the position that the FDA is acting unconstitutionally in imposing a facially content- and speaker-based ban on such speech.

To the FDA − stealing one of King George’s lines in “Hamilton” – “Good luck.”