FDA’s Interesting Off-Label Memo


The FDA recently posted a memorandum spelling out the agency’s views regarding off-label promotions, a document chock full of interesting observations. This discussion is now taking place in a somewhat different political backdrop, however, particularly in light of the dismissal of a high-ranking and well-known member of the Department of Justice. The picture in aggregate is highly fluid – which may or may not portend change – but which nonetheless is almost certainly unsettling to those at the FDA who prefer the status quo where off-label promotion is concerned.

When a solution is not a solution

The January 2017 FDA off-label memo offers the expected rationale for the agency’s current views on commercial speech, including that the requirement to conduct studies of a drug or device for a specific indication does a lot to provide assurance that the therapeutic in question will do more good than harm.

The memo takes a turn for the peculiar, however, when the FDA takes credit for things such as patent protection, which most would say redounds to the credit of the statute and the Patent and Trademark Office. Subsequently, the memo offers alternatives to the current state of affairs, such as a ban on off-label use of drugs and devices, and the application of a higher level of taxation on off-label use.

The problem with these and a couple of others of the FDA’s alternative commercial speech universes is that they are entirely outside the agency’s jurisdiction. For instance, it is quite well known that off-label use is within a physician’s discretion, and doctors are not regulated by the FDA. Not yet, anyway. As for tax policy, there’s no need to explain why this is outside the agency’s statutory authority.

There are several items of interest where context is concerned, not the least of which is that Robert Califf has resigned from the FDA commissioner’s chair, leaving the Trump administration with a very interesting job to fill. A number of names have been floated for that opening, including that of Scott Gottlieb, whose curriculum vitae includes a stint as one of the FDA’s deputy commissioners. Gottlieb has voiced reservations about the agency’s standards regarding off-label speech, but other candidates of record, including venture capitalist Balaji Srinivasan, are unknowns where the off-label discussion is concerned.

All in all, attorneys at the FDA have to assume the commercial speech question is likely to weigh against the agency in the months and years ahead, even if nothing in the way of legislation surfaces.

More than one new face at DoJ

One interesting bit of context here is the dismissal of Sally Quillian Yates, she of the famed Yates memo, which some believe comes up only a little short of the extortive tenor of the Thompson memo. President Trump gave Yates the axe for announcing that DoJ would not mount a legal defense of the administration’s executive order on immigration from six nations said to harbor terrorists. Whatever one believes about that executive order – and it should be noted that recent developments do not necessarily portend a revisitation of the memo that bears Yates’ name – her dismissal sends an unmistakable signal that employees of the federal government can press their luck only so far.

Clearly it’s too early to anticipate what will become of the off-label discussion, but this issue has occupied a tremendous amount of time and effort at the FDA, and these developments can’t help but embolden adversaries of the agency’s current stance on the commercial speech/First Amendment question. Change may or may not be in the offing, but it needn’t start with the executive branch of the federal government. It may come in the courts, and whomever the Trump administration appoints to the job of solicitor general may have some very interesting things to say, indeed, if the matter ever reaches the Supreme Court.

Compliance Benefit-Risk Final Retains ‘Industry-Wide’ Language

The FDA managed to wrap up the guidance for benefit-risk determinations for compliance action in six months, a much shorter turn-around than seen in a number of recent medical device guidances. The agency unveiled the final version just in time for the holiday season, but device makers should be aware that this guidance states that one company’s problem can quickly become a problem for other device makers.

Alignment with ISO 14971 spotty

The agency stated in the June 2016 draft that the intent was to “provide clarity” for the agency and device makers on the factors to be considered regarding compliance and enforcement activities for devices that seem to be emitting problematic safety signals in the postmarket realm. Among the factors taken into consideration is the prospect that a withdrawal would create a shortage of essential medical devices, and the FDA indicated from the outset that both short- and long-term considerations must be taken into account when examining whether the benefit-risk profile of a device seems to suggest a need for compliance and/or enforcement activity.

The final guidance did nothing to align more tightly with ISO 14971 than the draft, a consideration cited by several trade associations as a significant problem with the draft. In both versions of the guidance, the FDA acknowledged the roles of both ISO 14971 and 21 CFR Part 820 in terms of total life cycle management, but the final guidance has little more to say about this other than to concede that alignment of the Quality Systems Regulations with the ISO standard is limited.

Device makers had argued through their trade associations that the failure to provide more specific information on this score makes it difficult at best to anticipate how compliance with the ISO standard would satisfy the agency when using 14971 in determining how to address a problematic or potentially problematic device, all to no apparent avail.

Why your problem becomes my problem

Perhaps of greater interest to device makers, however, was the section appearing in both versions stating that any hazards associated with a specific device could be extrapolated to other devices made by that manufacturer, and to any similar devices made by other manufacturers. The FDA did not specifically state that any compliance and enforcement activities mandated by the agency would necessarily extend to similar devices made by other companies, but those in the life science industries are all too aware that bad news announced by a federal agency can quickly become a contagion in the court of public opinion. In some circumstances, the FDA may feel pressured to act on an individual company’s offerings even in the absence of a clear safety signal specific to that company’s device.

It might be pertinent to point out that members of the Senate and the House of Representatives occasionally dive into the med tech regulation pool, such as when Sen. Patty Murray (D-Washington) introduced the Preventing Superbugs and Protecting Patients Act of 2016 despite that the agency had mandated postmarket surveillance studies of duodenoscopes in late 2015. The savvy device manufacturer might consider the possibility that the FDA would prefer to undertake more vigorous – and potentially more sweeping – enforcement action rather than deal with further congressional meddling when adverse events reports begin to spike for a competitor’s offerings.

Legal Standing Aside, FDA Inks Emerging Signals Final Guidance

The FDA needed a year to move a draft guidance for emerging medical device safety signals to final form, assuring industry along the way that such disclosures would be handled deftly. At least one trade association raised the question of whether the agency enjoys the statutory authority to issue such advisories in the first place, but as a physician might say, the best medicine for this kind of issue is still prevention.

Statutory authority question gains no traction

The Center for Devices and Radiological Health published the draft “emerging signals” guidance in December 2015 with the acknowledgement that such advisories could lead to a problematic degree of under-utilization. The agency’s argument was that the benefits of such a disclosure would outweigh the risks, perhaps a nod to the benefit-risk draft guidance for regulatory action, which the agency may believe is a necessary companion piece to the emerging signals guidance.

One of the device associations that commented to the docket for the draft guidance argued that the Medical Device amendments do not authorize the disclosure of potential device safety information in the same manner as is found in the provisions governing drug regulation. Another device association asserted that any such announcements by the FDA would be arbitrary and capricious – which by precedent would seem to invite a lawsuit – or at the very least that such a move on the FDA’s part is more appropriately handled via comment and rulemaking rather than by guidance development.


A significant change made for the final guidance is the explicit promise that the FDA would not act on any information it deemed unreliable, unconfirmed, or not backed by sufficient strength of evidence. Industry may find itself flashing back to the drop in utilization of drug-eluting stents seen in 2006 and 2007 after disclosures regarding late-stent thrombosis at a medical society meeting in Barcelona, but the power morcellation question might be seen as a more obvious example of regulatory overreach with or without the help of physician input.

A second difference between the draft and final versions of the emerging signals guidance is that the latter discusses the formation of a group of subject-matter and regulatory experts charged with arriving at some understanding of the meaning of the initial signal. One of the options the agency could exercise is to require that the sponsor conduct additional postmarket studies of the device, while another option would be to suggest the manufacturer make a change to the device label. The exercise of this second option would presumably come up short of a removal of an indication for use, if only because a preliminary signal is not generally accorded sufficient weight to trigger a rescission of a 510(k) or a withdrawal of a PMA or PMA supplement.

More data, more need for compliance

An obvious subtext to the emerging signals guidance is the ongoing stand-up of the National Evaluation System for health Technology, known as NEST, which will take several years and several hundred million dollars to erect. The NEST system would presumably generate large volumes of data about device performance, but such a system might carry the signal-to-noise problem often seen in “big data” concepts. It is impossible for the FDA to guarantee it will not over-interpret a signal, particularly as it will undoubtedly be bombarded with allegations that the 21st Century Cures program will turn the FDA premarket programs into a series of rubber-stamp processes.

Perhaps now more than ever, device makers would do well to toe the line on the recent medical device reporting final guidance, if only to avoid pinging any patella tendons inside and outside the agency. Corrective and preventive action is still a big theme in warning letters, too, and any device maker that is seen as dropping the ball here will curry no favor with the FDA.

There is no guarantee the FDA will not overreact to an “emerging signal,” but companies can do a lot to keep investors and FDAers on their side – and to fend off would-be litigants – by running a tight regulatory ship.

FDA Suspends LDT Regulation, But Liability Still in Play

After two years and seemingly endless sturm und drang, the FDA’s device center has announced it will suspend its proposed framework for regulation of lab-developed tests. The agency made it clear this is not the end of its interest in the matter, but Congress has an eye on this issue as well, which makes the outcome of all this activity difficult to forecast. If there is anything resembling certainty in all this, it may be that lab-developed tests will soon be subject to more liability under state tort law.

PMA Preemption Likely to Hold

The FDA said in its Nov. 18 announcement that the agency is aware of the importance of working with “stakeholders, our new administration, and Congress” to forge the appropriate approach to regulation of lab-developed tests (LDTs). However, the statement includes the remark that the FDA will publish an outline of what the agency sees as an appropriate risk-based paradigm for LDT regulation “in the near future,” which could be used to “help guide continued discussions” on the subject. Clearly, the FDA intends to stay in the game as this controversy evolves in 2017, although it is interesting to ask whether the agency would have pulled back on the draft had the presidential election yielded a different result.

The House Energy and Commerce Committee had floated a discussion draft of a new framework for regulation of LDTs in 2015, which calls for the establishment of a new center at FDA that would regulate these tests, and which would report to the FDA commissioner “in the same manner as the other agency centers.” This passage would seem to put LDTs on the same regulatory footing as therapeutic devices, which in turn would suggest a similar legal status where preemption of tort law for PMA devices is concerned. The draft also cited preemption with the statement that the states would not be allowed to sustain any existing or establish any new requirements for LDTs that would be “different from, or in addition to” the mandates set out in the discussion draft, a seemingly familiar passage that presumably would not pertain to tests cleared under the 510(k) program.

Regardless of how the preemption question evolves, there are a number of parties who have expressed concern about the potential for an LDT-specific regulatory regime to increase legal exposure. The American Society for Human Genetics said in a Feb. 2, 2015, letter to the FDA that the agency’s proposed LDT regulation framework would subject tests for genomic purposes “to the states’ strict product liability tort regimes.” The American Clinical Laboratory Association raised a similar set of concerns in a June 2013 citizen’s petition to the FDA requesting the agency scupper the LDT regulation effort.

Lawsuits Already a Risk

There have already been several liability cases for LDTs, including the wrongful birth case in the state of Washington that cost LabCorp – and the hospital where the test was conducted – a total of $50 million. LabCorp was on the receiving end of another lawsuit, Khadim v. LabCorp, a case the company managed by having itself identified in legal terms as a provider, which in Virginia limits the damages available to plaintiffs. It does not seem a stretch to imagine that provider status under state law would suffer in a formal federal regulatory environment.

There are other ways the FDA can boost its enforcement activities in this area, such as the issuance of a safety alert, a move that would lead to increased media scrutiny and possibly media coverage-driven lawsuits. There is one inescapable fact to consider, however: Modern medicine’s reliance on LDTs will continue to increase and grow increasingly visible, particularly as gene sequencing technologies become less expensive and more commonly used to determine a course of treatment that may or may not achieve the desired result.

Obviously there are several moving parts to this predicament, but the net effect is unavoidably that the stakes for makers of these tests will grow, regardless of how Congress, the FDA and the White House answer the LDT regulation question.

New Guidance from FDA: When to Submit a 510(k) for a Change to a Cleared Medical Device

Courtney A. Stevens, Esq. |Senior Attorney, Medmarc Loss Control

FDA’s newest guidance for medical device manufacturers, Deciding When to Submit a 510(k) for a Change to an Existing Device, issued August 8, addresses a question manufacturers commonly face,—when a 510(k) is necessary for a change to an already cleared device. Manufacturers’ failures to submit 510(k)s are frequently cited in warning letters as rendering a device adulterated. As such, it’s an issue medical device companies can’t be too careful in scrutinizing. Thankfully, this guidance does provide such much-needed clarity on exactly when a 510(k) is necessary, and when processing the change in accordance with Quality System (QS) requirements (e.g., documentation of changes and approvals in the master record, verification and revalidation, etc.) is sufficient.

The confusion over whether a 510(k) is necessary is largely due to the subjective, relative language in the regulations, requiring device-makers to submit a 510(k) when a change “could significantly affect the safety or effectiveness of the device.” (21 CFR 807.81 (a)(3)). The Agency tried to clarify its interpretation of that language in its first guidance document on this issue, published in 1997, but clearly, as evidenced by the frequency with which the manufacturers’ determination of “significant” changes differed from the Agency’s, greater clarity was needed still. (Once finalized, this draft guidance will supersede the 1997 Guidance on the subject.)

This guidance document improves upon its predecessor by providing a number of exacting flow chart-decision trees to guide manufacturers through the determination of a 510(k)s’ necessity with regard to different types of changes.

It begins by setting out the guidance principles to be first considered in determining the propriety of a 510(k), which I briefly summarize here:

  • Modifications made with intent to significantly affect safety or effectiveness of a device. This is the same language as is found in the regulations, and its meaning is fleshed out in the remainder of the document.
  • Could “significantly affect” evaluation and the role of testing. In order to determine significance of the effect, manufacturers must conduct risk-based assessments.
  • Unintended consequences of changes. One component deemed to make up a “significant effect” is if the change would result in unintended consequences or effects. The draft guidance provides sterilization as an example which may affect device materials, thereby affecting performance of the device.
  • Use of risk management. Here, the draft refers to ISO 147981: Medical devices – Application of risk management to medical devices, and instructs manufacturers to utilize an assessment combining the probability of occurrence of harm and the severity of that harm in determining “significant effect.”
  • Evaluating simultaneous changes. Even though changes may occur simultaneously, each change should be assessed individually and in combination.
  • Appropriate comparative device and cumulative effect of changes. In making the determination of a 510(k)’s propriety, manufacturers need to consider (1) how different a change makes the device from its initial or most recent iteration as described in their most recently cleared 510(k); and (2) the cumulative effect of all changes since the last 510(k) cleared for this device. That is, though previous changes did not require a 510(k) when made in isolation, does the cumulative effect of this change with those previously made nor warrant a 510(k), even if it, by itself, would not?
  • Documentation required. Even if a manufacturer determines a 510(k) is appropriate for a particular change, this does not alleviate them from compliance with all existing QS requirements, including all documentation, verification, and validation duties.
  • 510(k) submission for modified devices. When a 510(k) is submitted for a device with multiple modifications since its last cleared 510(k), the 510(k) should describe not only the most recent change that warranted the 510(k), but also all previous modifications even though they did not merit the submission of 510(k)s in and of themselves.
  • Substantial equivalence determination. Manufacturers need understand that submission of a 510(k) for a change pursuant to everything outlined in the regulation and this guidance document does not assure that a substantial equivalence determination will be provided.


With these considerations in mind, manufacturers may proceed to the different parts of the guidance instructing them on decision-making for different kinds of changes—labeling, control mechanisms, operating principles, etc. In each of these, manufactures will find the aforementioned decision trees to guide them through submission criteria.

An example of the flow charts included in this draft guidance:

8.16 - guidance flowchart

In addition to charts guiding decision making, the guidance also provides examples of documenting changes and written regulatory change assessments.

This should go a long way in facilitating manufacturers’ understanding of when 510(k)s for changes are necessary, and reduce the number of warning letters for companies’ failure to submit them, accordingly.

FDA Issues Draft Guidance: Postmarket Cybersecurity for Medical Devices

Jordan Lipp | Partner Davis Graham & Stubbs LLP

A little over a year after issuing final guidance on premarket submissions for management of cybersecurity in medical devices, discussed here, the FDA issued draft guidance on postmarket cybersecurity (available here).  The FDA’s stated purpose of this draft guidance, which it just issued, is to clarify “FDA’s postmarket recommendations and [to] emphasizes that manufacturers should monitor, identify and address cybersecurity vulnerabilities and exploits as part of their postmarket management of medical devices.”  As cybersecurity threats are continually evolving, the FDA explains that it is not possible to completely mitigate cybersecurity risks solely through premarket controls.  Recognizing that “medical device cybersecurity is a shared responsibility between stakeholders,” the draft guidance addresses both risk management and remediation of cybersecurity threats.  It also discusses the interplay of cybersecurity issues and medical device companies’ reporting requirements, setting forth several examples of what should or should not be reported. Continue reading “FDA Issues Draft Guidance: Postmarket Cybersecurity for Medical Devices”

Regulatory Roundup: FDA Reports and Guidances in August

Courtney A. Stevens, Esq. | Senior Attorney, Medmarc Loss Control

Recommendations for a National Medical Device Evaluation System

On August 20, the FDA released, for public comment, a report from the Medical Device Registry Task Force and the Medical Devices FDA logo.jpgEpidemiology Network. The report described the inadequacies in the existing framework for medical device evaluation, what a more effective system would look like, what devices would be particularly ripe for evaluation via such a system, and how the system could be implemented. The full report can be found here.

FDA Issued Several New Drug and Device Guidances

The Agency issued several new guidance documents in August, including the following. Continue reading “Regulatory Roundup: FDA Reports and Guidances in August”