'Off-Label Speech' by David A. Simon in (2023) 72 Emory Law Journal 549 comments
This Article argues that the Food and Drug Administration (“FDA”) should regulate drug manufacturer speech about off-label uses based on the evidentiary support for the relevant use. The more evidence that an off-label use is safe and effective, the less restrictive the regulation should be. The less evidence that an off-label use is safe and effective, the more restrictive the regulation should be. Although intuitive, this is not exactly how current regulation of off-label information works. If the FDA approves a drug, the manufacturer can advertise to doctors and patients for the approved indication. Drug manufacturers cannot, however, promote or provide information about an approved drug for an unapproved use—so-called “off-label” use—unless they fall within two narrow safe harbors. Yet many off-label uses are just as safe and effective as on-label (approved) ones. Other off-label uses are supported by quality clinical trial data even though they are not approved.
While the FDA recognizes that not all off-label uses are equally (un)supported by the same level of evidence, it has faced legal and practical challenges regulating information about them in a nuanced way. Courts have held unconstitutional the FDA’s regulations purporting to ban promotional off-label speech by drug manufacturers. And the safe harbors it has constructed are too shallow for much useful speech. To address these challenges, this Article proposes a new approach: working collaboratively with the Centers for Medicare and Medicaid Services, the FDA can use drug compendia—which identify, evaluate, and rate off-label uses—to create a graded system for regulating how drug manufacturers disseminate information about off-label uses that links informational restrictions to the level of evidence supporting the disseminated use. Not only does this system enable a flexible and evidence-based regulatory regime, it also can be easily designed to survive constitutional scrutiny. ...
Doctors need information about drugs. And drug companies are ready to give it to them. When the information concerns a use of a drug the Food and Drug Administration (“FDA”) approved, drug companies can promote their approved drugs to physicians. Drug companies that want to provide information to physicians about an unapproved use of an approved drug, on the other hand, cannot do so except in limited circumstances.
There are two reasons why. The first is that unapproved uses of approved drugs—so-called off-label uses—often pose greater risks, both physically and monetarily, to patients than approved, on-label uses. Because the FDA has not vetted unapproved uses, they may lack the evidentiary support enjoyed by their on-label counterparts. When drug companies promote off-label uses, they increase the probability that a physician will prescribe a drug off-label—and, hence, increase the risk that the patient suffers harm from the unapproved use.
The second is that promoting off-label uses undermines the FDA approval process. Currently, a primary function of FDA approval is to incentivize firms to generate and disclose clinical trial data about a drug’s safety and efficacy. If companies can promote drugs off-label once a drug is approved, they have little incentive to conduct clinical trials to obtain approval for off-label uses. Rather than spend large sums of money to conduct clinical trials for an uncertain result (an FDA approval determining a drug is safe and effective), drug companies can spend much smaller amounts for a sure thing (promote the off-label use and obtain additional sales regardless of safety/efficacy).
Yet many off-label uses are necessary and appropriate.6 For some patients, they are the only available treatment; for others, they represent the medically accepted “standard of care.” Here, dissemination of off-label information can have positive, rather than negative, effects. By providing the physician information about a previously unknown treatment option, dissemination of off-label information increases the chance that a physician will prescribe a drug off- label to a patient who needs it. Limiting promotion of off-label uses, in this case, increases the risk that a physician will not prescribe a needed off-label treatment.
In the FDA’s view, the risk of too much off-label information outweighs the risk of too little. Despite the FDA’s position that off-label drug promotion is illegal, however, courts have not been inclined to agree. Indeed, recent judicial decisions have called into question whether any prohibition on off-label promotion is constitutional.
Responding to these judicial losses, the FDA has carved out safe harbors for manufacturer off-label speech. Unfortunately, these safe harbors are rather wooden and impractical. Drug manufacturers must provide an excessive amount of information in a format that is not useful to physicians. And there are few gradations on the kind, nature, and content of information drug manufacturers can provide if they fall within the safe harbors. All off-label uses that qualify for a safe harbor can be distributed in only the prescribed manner—a manner a physician is unlikely to find helpful.
Both of these challenges—the practical and the legal—are fundamentally about the quality and kind of evidence that supports an off-label use. When there is weak or no evidence supporting an off-label use, the risks posed by dissemination of off-label information are high. Restrictions on speech in such cases are likely to be constitutional because manufacturer statements about potential off-label uses are unlikely to be supported by evidence. When strong evidence supports an off-label use, by contrast, the risks posed by dissemination of off-label information are low—and the risk of not disseminating enough information is high. Here, restrictions on off-label information dissemination are unlikely to be constitutional because manufacturer statements about potential off-label uses are likely to be supported by evidence. Put differently, restrictions on unsupported statements will be more likely to satisfy the constitutional test than will restrictions on supported statements. Since both the legal and practical challenges of off-label promotion relate directly to the evidence supporting off-label uses, this Article argues that the best way to address them is to tie informational restrictions of off-label uses directly to the evidence base for the disseminated use. Uses supported by strong evidence could be disseminated more than those supported by weak or no evidence.
To tie dissemination to evidence, this Article argues that the FDA, working with the Centers for Medicare and Medicaid Services (“CMS”), should regulate and use drug compendia: privately produced, publicly regulated publications that collect, evaluate, organize, and rate information about drugs. In this system, off-label uses with higher evidentiary ratings can be disseminated more freely than those with low ratings. The graded approach allows for regulations with a tighter fit to evidence, which, in turn, enables a greater flexibility in information dissemination activities. Because regulations will be tied directly to vetted evidence, the FDA can increase the kind and nature of information dissemination it allows. At the same time, it can limit these activities sufficiently to preserve traditional incentives for FDA approval.
Linking information dissemination to evidence using compendia also enables the FDA to develop and use a new tool: a simplified and uniform disclosure document that can be included in manufacturer communications to physicians about off-label uses. This form—an example of which appears in the Appendix—effectively communicates to physicians complete and relevant information about evidence supporting an off-label use.
This approach not only solves the legal and practical problems with current regulation, but it also balances the two problems faced by public and private solutions posed by other scholars. Public-oriented solutions rely on a centralized authority, usually the FDA, to either conduct its own research or independently review evidence for off-label uses. Determinations about the evidence for a use could then inform decisions about whether promotional activities can occur. Because they fear promotional false positives—allowing promotion of uses that are not safe and effective—these proposals tend to overregulate at great expense, requiring significant government funding, gatekeeping, and administration.
Private-oriented solutions, on the other hand, typically allow off-label dissemination once a private organization determines a particular use meets certain evidentiary requirements. Unlike their public-oriented counterparts, private-oriented proposals fear promotional false-negatives: not allowing promotion of uses that are safe and effective. As a result, these solutions are cheaper but lack enforcement and underregulate the conflicts likely to arise in a private market.
Using drug compendia to link information dissemination to the evidence supporting the disseminated use marries the best of both solutions—leveraging the efficiency benefits of private-oriented solutions with the oversight function of public-oriented ones. Because compendia are privately run, they have low administrative costs. But they are also subject to public regulation by the FDA and the CMS, which ensure that the process by which they evaluate and rate off- label uses is unbiased, reliable, and transparent. To the extent that promotional false positives and negatives exist, they will result from bad evidence, not bad regulation. In short, using compendia to regulate dissemination of off-label promotion is likely to be cheaper than most public-oriented solutions and more effective than most private-oriented ones.
Compendia, though, have their own set of problems. Central among them are opacity, bias, and unreliability. Because these are significant problems, this Article does not propose using compendia in their current form. Instead, the FDA—working with the CMS—should regulate compendia directly and indirectly. Direct regulation specifies conditions, which, if met, would entitle a compendium to “recognized” status under current law. This includes uniform systems of (1) identifying, evaluating, and grading evidence; (2) identifying, evaluating, and acting on conflicts of interest and bias; and (3) publishing all information about (1) and (2). It could also entail additional compliance mechanisms that do not currently exist, including a requirement that compendia apply to “renew” their recognized status periodically and the audit or inspection of compendia by the CMS and/or the FDA. Because the CMS already has the statutory authority to regulate compendia this way—and because compendia are in need of reform—this aspect of the proposal is both practical and desirable.
Direct regulation, however, is not sufficient to link promotion to evidence. To do so, the FDA must regulate compendia indirectly. Currently the CMS indirectly regulates compendia by specifying what evidentiary ratings are sufficient to guarantee reimbursement. This Article proposes that the FDA do the same for purposes of information dissemination of off-label uses: it should specify the level of permissible information dissemination by reference to the evidence grade assigned for the disseminated use. Under this system, uses assigned high evidentiary grades could be disseminated with fewer restrictions than those assigned low evidentiary grades. Because compendia will not produce uniform assessments of evidence, this also gives the FDA flexibility in how it interprets the evidence base for a given use.
This Article proceeds as follows. Part I explains the current framework for regulating manufacturer speech about off-label uses. It highlights how the FDA’s current approach to off-label information faces serious practical and legal challenges. It then argues that the best method for overcoming both of these obstacles is to link information regulation of off-label uses to the evidence supporting the use in question by relying on drug compendia. Part II explains drug compendia and their current weaknesses. Part III then reviews five different methods for using drug compendia to link the level of off-label information dissemination to the evidence base supporting that use. It argues that while drug compendia can provide this linkage, they need to be regulated more closely to do so effectively. After describing the proper nature and scope of this additional regulation, this Article illustrates how this proposal could work in practice using three examples.
