Defending FDA’s Off-Label Policy: What Are The Alternatives?

As part of FDA’s defense in a First Amendment case filed by Amarin, the agency’s Janet Woodcock submitted a declaration outlining the government interests at stake – and explicitly responding to alternative approaches that a federal appeals court suggested in the landmark Caronia ruling that opened the door for Amarin’s suit. Here is an excerpt from the declaration responding to those alternatives.

FDA has examined alternative approaches that have been suggested, for example by the US Court of Appeals for the Second circuit in US v. Caronia, and determined, based on FDA’s technical and policy expertise, that they are impractical, ineffective, unrealistic, or based on inaccurate assumptions.

Janet Woodcock is Director of FDA’s Center for Drug Evaluation & Research. This excerpt is from a declaration submitted by the agency on June 23 to the Manhattan federal court as part of the Amarin v. FDA litigation.

Many of these proposed approaches are blunt tools that may address one or two of the interests identified above, but do not take into account or attempt to navigate the complex mix of numerous and sometimes competing interests that FDA’s approach addresses. The proposed alternatives, and the reasons they would be inadequate substitutes for FDA’s current regulatory authorities and enforcement policies in protecting the public health, include the following:

• Prohibiting altogether the use and/or prescribing of an unapproved new use of an approved product

An outright prohibition on the use of approved products for unapproved uses would be effective in protecting the government interests in motivating scientifically robust research into unapproved uses and ensuring that new uses of an approved drug are proven to be safe and effective before they are used to treat patients.

However, the prohibition would substantially restrict the discretion and independence of healthcare providers, and would fail to take into account the interests behind allowing healthcare providers to determine the best treatment options for individual patients in specific circumstances, such as in treating diseases for which there are no approved treatments or in treating individual patients for whom all approved treatments have failed.

Thus, this more restrictive alternative would likely have an adverse impact on the public health.

• Creating ceilings or caps on the number of prescriptions for an unapproved use.

This proposed alternative is similar to the total prohibition above, except that it would allow some amount of prescribing before a ceiling or cap was reached. Once the prohibition was operative, it would present the same problem of limiting healthcare provider discretion in determining treatments geared toward the needs of the individual patient.

However, before that ceiling was reached, firms could encourage the use of a product for an unapproved use with none of the safeguards of rigorous FDA review—just as if there were no requirement of premarket review for a second intended use.

Thus, an arbitrary cut-off of this type is not aligned with any discernible government interest and would adversely affect the public health.

This alternative would also be impractical to administer and enforce because, in many cases, it may be difficult to determine for what specific use a product is being prescribed.

Prescriptions written by healthcare providers do not ordinarily reflect whether a drug was prescribed for an approved or unapproved use. With certain limited exceptions (for example, in the case of drugs with significant risks or very high costs where prior authorization is required prior to dispensing), the reasons for which a drug was prescribed is not available in the data provided to the Government in claims for reimbursement under Medicare or Medicaid.

• Limiting Medicare and Medicaid reimbursement to approved uses.

This alternative—having the government limit its Medicare and Medicaid reimbursement to approved uses – would again limit healthcare provider discretion in determining treatments geared toward the needs of individual patients under Medicare and Medicaid. T

here would be no governmental interest in virtually eliminating the prescribing of unapproved uses for one subset of the population but having it continue for the remainder of the population. And, as in the previous alternative, this alternative would be impractical to administer and enforce.

• Prohibiting specific unapproved uses that are exceptionally concerning or developing tiers based on level of safety concerns with greater regulatory controls for the relatively more dangerous products.

These alternatives would tie the regulatory controls to the degree of safety concerns about the drug.

Under the first alternative, the government would prohibit specific unapproved uses for drugs that were exceptionally concerning from a safety perspective. The second alternative would similarly tie the applicable regulatory control to the level of safety concern, with stronger controls applied to more dangerous products.

Both alternatives would be inadequate to protect the public safety because the required safety assessment would depend on the generation of data regarding product dangers before any controls can be applied. It would essentially allow a product to be distributed for unapproved uses without the development of data to support such use and with the submission of the data to FDA, and therefore would allow the product to be in distribution before FDA could assess its safety for that use.

It would therefore return drug regulation to the era before the 1962 amendments, when the government was limited to using post-marketing remedies after the product had injured members of the public at large.

With respect to the less exceptional or lower tier drugs, both alternatives would undermine the other interest advanced by premarket review by diminishing the incentive to engage in premarket review.

• Requiring firms to list all potential indications for a product in the initial premarket application

Another proposal is to require manufacturers to list all potential uses in the first application to enable physicians, the government and patients to track a drug’s development. However, science is currently not capable of divining all potential uses of a medical product from an initial study; data and information develop over time through scientific study before and after product approval.

If a firm’s listing of one or more potential indications, submitted at the same time as the data supporting the primary indication, were the only requirement necessary before firms were allowed to market their product for the claimed indications, this would undermine several government interests…including incentivizing robust research, requiring premarket safety and effectiveness review for each use, and developing appropriate instructions for use.

Potential variations on this proposed alternative are likewise unworkable.

For example, if firms were required to obtain approval at one time for all intended uses, the initial application might be significantly delayed while new indications were explored. If a firm were unable to seek approval later for uses that were not identified at the time of an initial application, there would be no incentive to continue scientific exploration that could lead to the development and approval of new medical treatments.

Thus, this alternative would negatively impact the public health.

For example, Imbruvica (ibrutinib) was approved for Mantle Cell Lymphoma in 2013, then for Chronic Lymphocytic Leukemia in 2014. In 2015, it was approved through the breakthrough therapy designation for Waldenstrom’s Macroglobulinemia, a rare form of cancer. It is the only product currently approved for that disease.

Similarly, Rapamune (sirolimus) which was initially approved in 1999 as an immunosuppressive agent to help prevent organ rejection. In 2015, it became the first drug to receive approval to treat lymphangioleiomyomatosis (LAM), a rare, progressive lung disease that primarily affects women of childbearing age.

• Allowing firms to actively promote an unapproved use as long as they disclose that the use is unapproved and include other appropriate warnings.

This proposed alternative would allow firms to promote an unapproved use limited only by the need to provide certain disclosures.

While warnings and disclosures can help to provide material information necessary to assist in understanding data and their value, they are not always effective in curing misleading impressions. Nor do they protect all of the public health interests advanced by premarket review.

This alternative would permit firms to bypass the premarket review process for new intended uses once the product was approved for just one use by disclosing that the use is unapproved or including certain warnings. It would therefore undermine government interests…including incentivizing robust research, requiring premarket safety and effectiveness review for each use, and developing appropriate instructions for use.

In addition, a statement that a product is unapproved does not necessarily convey that the speaker does not consider it suitable for that use. Nor do such statements or other warnings substitute for the labeling that is developed after an assessment of the data and information relating to the new use, which labeling would include directions and warnings specifically applicable to that use.

Furthermore, studies have shown the limitations of disclosures in terms of consumer perception and understanding in certain contexts – limitations that may exact too great a cost when lives and health hand on their success.

Thus, this alternative could return the regulation of medical products to the days before the enactment of current premarket review requirements with only after-the-fact remedies that were insufficient to prevent public health tragedies, and would be detrimental to the public health.

• Educating healthcare practitioners and patients to differentiate false and misleading promotion from truthful and non-misleading information.

Although FDA does have several educational resources in this area, it is unrealistic to suggest that this type of program can be conducted on the scale necessary to effectively combat the adverse impact of false and misleading promotion.

But, even assuming that such a program were feasible, this alternative, like the previous proposal immediately above, would allow firms to bypass the premarket review process by marketing or promoting a product for an unapproved use and thereby undermine the substantial government interests in incentivizing robust scientific research, requiring premarket review, and developing required labeling that provides appropriate information for safe and effective use.

It would replace the FDA’s thorough and rigorous scientific review process with a potentially far more cursory review by individual prescribing healthcare providers and patients. Individual healthcare providers and patients should not be expected to acquire the tools, background, and specialized expertise in statistics, pharmacokinetics, biomedical engineering, and other fields to conduct a thorough evaluation of the risks and benefits of a new intended use that even roughly approaches that provided by FDA review, and it is unrealistic to suggest that a government-sponsored education campaign would provide this kind of multi-disciplined expertise.

And such an education campaign would do nothing to provide each practitioner or patient with the time needed to do so for every use of hundreds of drugs.

This suggested alternative also does not appear to take into account the possibility that firms may present truthful but incomplete information, and that the individual healthcare provider or consumer would not be well position to uncover or weigh the significance in the absence of a full disclosure of all relevant data.

• Reminding healthcare providers of potential malpractice liability

This proposed alternative appears to be suggested as a way of making healthcare providers more cautious regarding prescribing medical products for unapproved uses.

To the extent it discourages all prescribing of unapproved uses, this alternative would not advance the interests behind allowing healthcare providers to determine the best treatment options for individual patients in specific circumstances, such as in treating diseases for which there are no approved treatments or in treating individual patients for whom all approved treatments have failed.

In addition, like the previous example, it would allow firms to bypass the premarket review process for new intended uses and thereby undermine the significant government interests advanced by that process. Furthermore, this alternative would do nothing to deter firms from developing biased presentations with the potential to mislead the listener.

• Taxing firms more heavily for sales of products for unapproved uses than for approved uses

This proposed alternative would allow unrestricted sharing of information about unapproved uses of approved products, but purports to retain the incentive for seeking FDA approval by taxing sales for unapproved uses more than sales for approved uses.

The proposal does not align with the government interests in part because it would affect all prescribing of an unapproved use equally – whether or not there were special circumstances that warranted prescribing for unapproved uses.

Moreover, it would allow companies to substitute a tax payment for the cost of the robust scientific research needed to prevent the public from injuries associated with inadequately studied and tested products.

It is not apparent how such tax payments would remedy or deter, let alone prevent, the significant public health harms that premarket review is designed to avert. Financially robust entities could, essentially, fund the continued distribution of products with unknown safety and effectiveness profiles while imposing risks on the public.

This alternative would also likely be impractical to administer and enforce because it may be difficult to determine in many cases the particular use for which a product is being prescribed.