A Thoughtful Response to “A Strategy for Rescheduling Psilocybin” by Dr. Mason Marks
And Why You Should Care About Rescheduling
In “A Strategy for Rescheduling Psilocybin” published in Scientific American, Dr. Mason Marks writes a short but thought-provoking piece about psychedelics law and federal drug scheduling. I enjoyed it. The article makes a valuable contribution by drawing focus onto an often-ignored aspect of the emerging areas of psychedelics and cannabis law—the CSA’s administrative scheduling and rescheduling system.
In so doing, Marks raises an important question that is worth answering. He argues that psilocybin’s Schedule I status stifles research, innovation and access and that rescheduling would address these issues in “one fell swoop.” Whether you agree with this statement or not, with evidence and a colorable argument that psilocybin is misclassified, why has nobody petitioned DEA to reschedule psilocybin?
In this response, I unpack Dr. Marks’ essay, agree with his conclusion, and offer several responses and additions to points he makes, including where I disagree. I also suggest a better avenue for rescheduling psilocybin based on the statutory language.
The central thesis of Marks’ article is that at this point in history, psilocybin is misclassified as a Schedule I drug because of “overwhelming evidence” of controlled studies showing safety and efficacy. That misclassification, he contends, presents significant regulatory obstacles.
Whether Marks proves his “overwhelming evidence” claim is a point best saved for another day; doing so is not necessary to prove his thesis. He identifies solid evidence from Phase I and II FDA-style clinical trials that synthetic psilocybin is safe and effective to treat depression. That’s enough.
From there, Marks describes three pathways to deregulating the drug under the CSA—legislative, administrative, and judicial—and argues psilocybin should be rescheduled.
The Drug Schedules
Marks describes the CSA schedules as a “five-tiered list according to their perceived potential for abuse.” But this is not, strictly speaking, correct. Congress established Schedule I in 1970 as the only schedule for drugs having no currently accepted use in the United States. Hence, marijuana is in Schedule I, while more dangerous drugs, like methamphetamine or cocaine, are in Schedule II. We’ve discussed this issue thoroughly in various briefs we’ve filed in federal court.
This common misconception leads to an important observation about the CSA and its administrative implementation by DEA: in practice, what determines whether a drug is in Schedule I and burdened with draconian regulatory restrictions that go along with it is “currently accepted medical use”—and not abuse potential.
Recognizing that point speaks volumes about the CSA and what would be a far better system. Why should a drug scheduling system supposedly in place to prevent drug abuse and diversion treat “currently accepted medical use” as the determinative criteria for which drugs should be in the most restrictive schedule and hardest to access and research? Shouldn’t the regulatory classification be based on danger? At minimum, shouldn’t drugs that have no currently accepted medical use but lots of promise be easier to research than drugs with a currently accepted medical use, but already known to be dangerous and highly abusive, like methamphetamine (Schedule II).
The indictment of this aspect of the system becomes especially damning now that research has shown beyond cavil that swaths of drugs with “no currently accepted medical use” but relatively low abuse potential in Schedule I, like psilocybin, are promising healing agents for intractable disorders. Put simply, a system that schedules drugs based on potential abuse and harm first makes far more sense.
The deeper one digs into the schedules, the more illogical they get. Shortly after enactment of the CSA, a federal judge explored this illogic in lengthy footnote 4 of the judicial opinion, United States v. Maiden, 355 F. Supp. 743, 748 (D. Conn. 1973):
Section 202 of the Act, in establishing the three findings for each of the five schedules, does not in terms specify whether the findings are cumulative. 21 U.S.C. § 812. In fact they cannot logically be read as cumulative in all situations. For example finding (B) for Schedule I requires that “The drug or other substance has no currently accepted medical use in treatment in the United States.” Finding (B) for the other four schedules specifies that the drug has a currently accepted medical use. At the same time, finding (A) requires that the drug has a “high potential for abuse” for placement in Schedule I, but a “potential for abuse less than the drugs or other substances in schedules I and II” for placement in Schedule III. If the findings are really cumulative, where would one place a drug that has no accepted medical use but also has a potential for abuse less than the drugs in Schedules I and II? According to finding (A) for Schedule III it belongs in Schedule III, but finding (B) for that schedule precludes Schedule III; according to finding (B) for Schedule I it belongs in Schedule I, but finding (A) for that schedule appears to preclude Schedule I. The answer may be that applicability of finding (B) concerning currently accepted medical use should be made first. If the drug has none (and marijuana probably does not, though the testimony indicated some interesting potential uses), then placement in Schedule I may be appropriate whether or not the potential for abuse is higher than for other drugs, so long as the abuse potential is not minimal. But once a substance is precluded from Schedule I because an accepted medical use currently exists, then the comparative potential for abuse may determine its placement in Schedules II, III, IV, or V.
It is not clear why this solution proposed by the judge in Maiden is correct as a matter of statutory construction. Why is the second listed factor (“currently accepted medical use”) in the schedules more important than the first or third? And in Nat'l Org. for Reform of Marijuana L. (NORML) v. DEA, 559 F.2d 735, 748 (D.C. Cir. 1977), the D.C. Circuit Court of Appeals all but rejected that approach:
To treat medical use as the controlling factor in classification decisions is to render irrelevant the other “findings” required by Section 202(b). The legislative history of the CSA indicates that medical use is but one factor to be considered, and by no means the most important one.
Indeed, in the legislative history, abuse potential is considered the most important scheduling factor. And the D.C. Circuit in NORML also explained that in 1977 DEA’s own scheduling practices supported the conclusion that substances lacking medical usefulness need not always be placed in Schedule I. There, DEA’s Chief Counsel testified that several substances listed in CSA Schedule II, including poppy straw, had no currently accepted medical use and even acknowledged that marijuana could be rescheduled to Schedule II without a currently accepted medical use. And yet today, DEA says otherwise. Contrary to the D.C. Circuit’s holding in NORML and along the lines of the solution proposed the judge proposed in Maiden, DEA treats accepted medical use as dispositive. Hence, drugs with relatively low abuse potential like psilocybin remain in Schedule I.
Or at least sometimes. As Shane just wrote, just recently, FDA wanted to create a exception to this paradigm for, of all things, fentanyl related substances. The key lines are on pages 5 and 6 of Dr. Throckmorton’s testimony:
[I]f the [aforementioned] process leads HHS to conclude that the substance has some potential for abuse, but less than that of substances in schedules I and II, HHS would likewise convey the conclusion to DOJ and provide its analysis, and DOJ would be required, within the same time period, to remove the substance from schedule I and reschedule it in schedule III.
Let me put this in starker terms: the administration appears willing to bend the matrix for fentanyl related substances but not for psilocybin (or marijuana). If the above standard applied to psilocybin—which (1) has a potential for abuse but less than that of substances in schedules I or II and (2) has infinitely more promise that fentanyl related substances—it should be rescheduled. There is no good reason I can think of to make it easier to reschedule fentanyl related substances over psilocybin.
But I digress. If you think there is a viable and meritorious lawsuit or three (as Marks suggests in claiming the judicial path might be most effective) challenging DEA’s approach to scheduling, you are right. It’s an absolute mess, and we’ve just scratched the surface of impropriety.
FDA Approval Is Not Required For Rescheduling
Next, Marks explains that “federal lawsuits on cannabis rescheduling provide guidance and suggest that FDA approval is not a requirement for meeting that criterion.” This observation is important and should be noted. In conversations with practitioners and industry, it is routinely misunderstood. But Marks is right, and his point bears repeating: FDA approval is not a precondition to rescheduling.
But the issue is a bit more nuanced than described by the article, which is understandable: it is a Scientific American piece, not a law review article. Marks does not discuss the standard DEA uses to evaluate “currently accepted medical use,” and that standard, unfortunately, undermines the ultimate conclusion he reaches.
For drugs not approved by FDA, to have “a currently accepted medical use” in the United States, DEA requires that all of the following must be demonstrated: The drug's chemistry is known and reproducible; there are adequate safety studies; there are adequate and well-controlled studies proving efficacy; the drug is accepted by qualified experts; and the scientific evidence is widely available. 57 Fed. Reg.10499 (1992). We briefed this five-part test extensively in Sisley v. DEA and showed it to be complete hogwash and unlawful.1 While we lost on a procedural point, one judge wrote separately to agree with us, noting that “in an appropriate case, the Drug Enforcement Administration may well be obliged to initiate a reclassification proceeding for marijuana, given the strength of petitioners’ arguments that the agency has misinterpreted the controlling statute by concluding that marijuana.” When a case is dismissed on a procedural point, and a respected federal judge and potential Supreme Court judge nominee goes out of his way to write a concurring opinion and comment on the merits (when he did not have to), that is pretty powerful evidence that something in the way DEA makes this sausage isn’t kosher.2
In any case, until this five-part test is declared unlawful, DEA will use it. And while safety and efficacy (what Marks highlights) supply evidence for two of the criteria, DEA requires all five elements must be shown under the five-part test.
These other criteria are not meaningless, particularly the nebulous “consensus of the national community of experts” where “a material conflict of opinion among experts precludes a finding of consensus.” It is easy to see how this undermines Marks’ conclusion. Marks concludes that “qualifying evidence may come from adequate and well-controlled clinical trials” and therefore, that Phase I or II trials showing safety and/or efficacy should be sufficient for rescheduling. But this isn’t the case because of the other parts of the test that have to be met. Even if some clinical evidence showed safety and efficacy, DEA would likely argue that “a material conflict of opinion” exists among qualified experts—the agency refuses to recognize the opinions of medical practitioners who are not experts in evaluating drugs—and prevail under its five-part test and substantial evidence review if challenged in court. Substantial evidence is a highly deferential standard that is hard to surmount.
Next, Marks then describes what he calls the three pathways to rescheduling: legislative, administrative, and judicial. But really there aren’t three—just two—because the judicial pathway isn’t distinct from the administrative pathway. But one can only sue the DEA in court after it denies a rescheduling petition. And the remedy for such a lawsuit is not a judicial order rescheduling a substance or instructing DEA to reschedule the substance but vacating an administrative decision of DEA and remanding the case back to DEA. This process can take years, maybe decades. As we discussed in briefing in Sisley v. DEA, the NORML cannabis rescheduling petition filed in the early 70s took over 20 years to adjudicate, bouncing between the second and third pathways. Marks suggests the “judicial path may be the most effective,” and he very well might be right. But historically, the administrative/judicial path involves significant time and effort.
Psilocybin Has A “Currently Accepted Medical Use With Severe Restrictions”
In my view, the biggest oversight in Marks’ article is what it understandably does not discuss: statutory language that supports rescheduling psilocybin and explains why psilocybin (and MDMA) is no longer properly classified as a Schedule I drug. This language has never really been the focus of any administrative or judicial proceeding. It is this statutory language, however, that goes to the heart of his thesis: promising evidence of safety and efficacy obtained from early clinical trials ought to result in removing a drug from Schedule I. I agree.
Generally, what separates Schedule I from Schedule II-V drugs is that Schedule I drugs have “no currently accepted medical use” while drugs in Schedule II-V do have currently accepted medical use. But there is an exception: Schedule II also houses drugs that have a “currently accepted medical use with severe restrictions.” 21 U.S.C. § 812(b)(1)(2). This phrase has to mean something or else it would not be in the statute. But what is that? Other than being something more than “no currently accepted medical use” but less than “currently accepted medical use in treatment in the United States,” it isn’t clear.
There isn’t much in the statute or legislative history to tell us what this means. But over the years, FDA and DEA have made several statements suggesting that drugs that have sufficiently progressed through the FDA process and are available for expanded access—which applies to psilocybin and MDMA—are drugs with “currently accepted medical uses with severe restrictions” that should be in Schedule II. That comports with the plain meaning of the text, and it also makes sense: as drugs show medical usefulness through Phase I and II trials, they should be more available for research by moving those drugs to Schedule II. All of this talk about loosening Schedule I research controls to be more like Schedule II is, in some senses, already in the CSA and could be used by the administration now without congressional intervention, if it really wanted to.
We’ve briefed this point as well in a case pending in the First Circuit that Shane will argue in January 2022. In 1982, FDA recommended that THC remain in schedule I as a drug with “no currently accepted medical use in treatment in the United States” because it could not be lawfully marketed under the FDCA. But importantly, FDA also concluded that THC could be placed in schedule II as a drug with “a currently accepted medical use with severe restrictions.” It did so because drugs “in the later stages of the investigational process may fall within this statutory language.” In the case of THC, development had progressed sufficiently far to be termed “currently accepted medical use with severe restrictions” because the National Cancer Institute’s included THC in a “group C distribution scheme,” making THC available to an estimated 4,000 cancer specialists for use in combating nausea and vomiting in cancer patients undergoing chemotherapy. FDA authorized this broader distribution plan because, among other reasons, the close supervision required by the study protocol appeared to provide adequate safeguards for patient safety and sufficient evidence of effectiveness existed to support broader availability for treatment of patients. Today, FDA’s expanded access program, which psilocybin is eligible for, squarely fits the bill. 21 C.F.R. § 312.300 et seq.3
The argument that a drug that has progressed far enough in FDA clinical trials to be eligible for expanded access has a “currently accepted medical use with severe restrictions” has substantial merit. And psilocybin passes that hurdle and easily meets the other Schedule II factors. So I agree with Marks’ thesis, but for a different reason.
Finally, Marks mildly overstates and understates the effect of rescheduling psilocybin. Certainly, Marks hits the Schedule I stigma and public funding nails on the head. But the degree to which rescheduling from I to II loosens regulatory controls is overstated. We’ve written about that issue in another court filed brief around page 16. A move to Schedule III, in contrast, is more significant. The understatement? With careful design and calibrated controls, rescheduling to Schedule II could permit intrastate psilocybin therapy regimes to proceed without violating federal law, even without FDA approval. More on that issue for another day. At minimum, rescheduling psilocybin would remove the drug from the ambit of idiotic laws like this that requires arms of the federal government to actively oppose attempts to legalize Schedule I substances.
Administrative Rescheduling Deserves More Attention
While I can nitpick at the edges, I enjoyed this article and its attention to the rescheduling process. It raises deeper questions worthy of conversation that have been ignored for too long.
Assume Marks is right and that rescheduling psilocybin would solve even one significant issue relating to psilocybin. If so, why is nobody putting resources and willpower behind rescheduling pushes and/or the administrative process? If there is a chance that we could loosen research restrictions and make psychedelic research easier on a national scale, shouldn’t we be putting at least some resources into that? This doesn’t cost much and the payoff could be huge. And by the way, submitting a petition to DEA along these lines is really easy and takes no time at all. I wrote a complete, ready-to-submit rescheduling petition for psilocybin in three hours. And if you are a paid subscriber of the newsletter, I will send it to you—no strings attached.
The other important takeaway is not just that psilocybin ought to be moved out of Schedule I, but that how administration of the schedules over the years is profoundly broken. We tried to fix one the problems in Sisley v. DEA—overturning DEA’s unlawful interpretation of “currently accepted medical use in treatment the United States”—and nearly succeeded. But at a core level, the system doesn’t really work well. And hopefully, articles like what Marks writes can get the community focus on a problem that needs fixing: the backwards administrative process that underlies drug control in the United States.
In conclusion, in my view, the greatest contribution of Marks’ article is something it never expressly says but certainly suggests: decoupling the CSA rescheduling regime from the FDA approval process could accelerate research with controlled substances, and ensure a future with sensible and equitable drug policies. This important issue has gotten little attention and support from the community, but it is deserving of far more.
Our arguments were and still are almost certainly correct. The federal government did not even bother defending against our assault. This legal issue is very important, not just because DEA uses the five-part test to refuse to reclassify marijuana, despite widespread medicinal use in the states. As states rapidly enact their own medical psilocybin initiatives, the five-part test will be the basis for refusing to reclassify psilocybin. As we explain in our briefing, the “currently accepted medical use” requirement was put in place in 1970 to avoid a Schedule I classification where that drug that was being used by doctors in medical practice. The notion that “currently accepted medical use” has anything to do with FDA approval or FDA criteria is completely untethered to the text and legislatively history.
I am aware that most sausages are not kosher.
Years later, HHS told DEA in passing that GHB was “far enough along in the development process to meet the standard under Schedule II of a drug or substance having a “currently accepted medical use with severe restrictions.” 65 Fed. Reg. 13,237 (2000).