Photo of Bexis

Contrary to popular belief – even among some lawyers – off-label use is not necessarily entirely off-label.  That’s a good thing, too.  One of the things we’ve harped on with this blog is that off-label use is legal, common, and in various fields can represent the prevailing medical standard of care.

One thing we’ve never argued on this blog is that off-label use (or any use) of prescription medical products is risk free.  Thus, in cases of widespread off-label uses, by all means the FDA should have the ability to ensure (as is the case with labeled uses) that a product’s labeling informs prescribing doctors of relevant risks.  The alternative view that we encountered back in the Bone Screw litigation (flirted with by the Kessler-era FDA), that warnings pertaining to off-label uses were somehow “promotion” and should be prohibited, always struck us as illogical and counterproductive.

It turns out that the FDA can indeed require warnings about off-label uses.

With prescription drugs, the Agency’s authority to order warnings about off-label uses is pretty well spelled out and straight-forward.  The relevant regulations provide:

A specific warning relating to a use not provided for under the “Indications and Usage” section may be required by FDA in accordance with sections 201(n) and 502(a) of the act if the drug is commonly prescribed for a disease or condition and such usage is associated with a clinically significant risk or hazard.

21 C.F.R. §201.57(c)(6)(i) (emphasis added).

A specific warning relating to a use not provided for under the “Indications and Usage” section of the labeling may be required by the Food and Drug Administration if the drug is commonly prescribed for a disease or condition, and there is lack of substantial evidence of effectiveness for that disease or condition, and such usage is associated with serious risk or hazard.

21 C.F.R. §201.80(e) (emphasis added).  The references in §201.57 to sections 201(n) (21 U.S.C. §321(n)) and 502(a) (21 U.S.C. §352(a)) of the FDCA are to general provisions relating to misbranding.

Thus “[i]n addition to warning about risks from approved uses, the FDA has authority to impose warnings about off-label or unapproved uses when there is evidence of a clinically significant risk.”  Bailey v. Wyeth, Inc., 37 A.3d 549, 556 (N.J. Super. Law Div. 2008) (emphasis added), aff’d, 28 A.3d 1245 (N.J. Super. App. Div. 2011) (citing §201.57 as authority for off-label drug warnings); see Harris v. Amgen, Inc., ___ F.3d. ___, 2013 WL 5737307, at *6 (9th Cir. Oct. 23, 2013) (same authority); Richardson v. Miller, 44 S.W.3d 1, 11-12 (Tenn. App. 2000) (same authority).  Thus, it’s apparent from the face of the FDA’s regulations that the Agency has authority to order prescription drugs to carry warnings relating to off-label uses.

Now for the kicker.  The express language of these regulations – “required by” the FDA – places warnings about off-label uses in the prescription drug context squarely within the realm of “impossibility” preemption under the LevineMensingBartlett line of Supreme Court authority, since off-label use warnings are not something that a drug manufacturer can add to its label unilaterally without prior FDA approval.  We’ve explained that rationale in greater detail here as to other types of label changes that require the Agency’s prior approval.  It’s also the same reasoning that requires preemption in the context of black box warnings, as we’ve discussed here and here.  So, if a plaintiff is demanding warnings about off-label uses, the defense has a preemption defense.

With medical devices, the FDA’s authority to add off-label information to labeling is not as explicit, but nonetheless present, as the FDA itself states:

During its review, FDA may seek a statement in the labeling that there is a lack of evidence that a device is effective for an off-label use or indication.  The [summary of safety and effectiveness data] should contain an explanation of the basis for such limitations.

FDA, ODE, “Summary Of Safety And Effectiveness Data (SSED) − Clinical Section Checklist,” at 6 (June 10, 2010) (emphasis added).

Why devices should be regulated differently from drugs with respect to off-label warnings is unclear, but we speculate that it might be related to 21 U.S.C. §396, which expressly forbids the FDA from regulating off-label use as a medical practice (FDA can’t restrict “the authority of a health care practitioner to prescribe or administer any legally marketed device to a patient for any condition or disease”).  Still, §396 expressly maintains existing FDA authority over device “labeling,” and putting something in a manufacturer’s label hardly amounts to restricting the practice of medicine.  As we’ve discussed before, FDA-approved labels don’t define medical standards of care.  Labeling is one of many sources of information that physicians may, in their discretion, consult.

The only thing we found in the device field similar to §§201.57 and 201.80 was 21 U.S.C. §360c(i)(E)(i), which gave the FDA express authority to order off-label warnings for substantially equivalent (§510k) medical devices:

[T]he director of the organizational unit responsible for regulating devices . . . may require a statement in labeling that provides appropriate information regarding a use of the device not identified in the proposed labeling if . . . there is a reasonable likelihood that the device will be used for an intended use not identified in the proposed labeling for the device; and . . . such use could cause harm.

However, by its terms that subsection “has no legal effect after the expiration of the five-year period beginning on November 21, 1997.”  21 U.S.C. §360c(i)(E)(iv).  So, with that provision expired for over a decade, what is the source of the FDA’s (plainly exercised) power to require warnings pertaining to off-label uses of medical devices?

The necessary FDA authority resides in 21 U.S.C. §360j(e) regarding “restricted devices.”  That section creates the concept of prescription-only devices, but does more than that:

(e) Restricted devices

(1) The Secretary may by regulation require that a device be restricted to sale, distribution, or use −

(A) only upon the written or oral authorization of a practitioner licensed by law to administer or use such device, or

Emphasis added.  That’s the prescription-only part, but Part B of the same section of the FDCA also extends the FDA’s device labeling power:

(B) upon such other conditions as the Secretary may prescribe . . . if, because of its potentiality for harmful effect or the collateral measures necessary to its use, the Secretary determines that there cannot otherwise be reasonable assurance of its safety and effectiveness. . . .  A device subject to a regulation under this subsection is a restricted device.

(2) The label of a restricted device shall bear such appropriate statements of the restrictions required by a regulation under paragraph (1) as the Secretary may in such regulation prescribe.

Emphasis added.  Thus under §360j(e) the FDA can impose “other conditions” on medical devices by regulation, and such “conditions” can require “appropriate statements” on the labeling of such devices.

The FDA has just such a regulation for “restricted devices,” which it defines as

a device for which a requirement restricting sale, distribution, or use has been established by a regulation issued under section 520(e) of the act [§360j(e), quoted above], by order as a condition of premarket approval under section 515(d)(1)(B)(ii) of the act, or by a performance standard issued in accordance with sections 514(a)(2)(B)(v) and 514(b) of the act.

21 C.F.R. § 807.3(i) (emphasis added).  Notably, the “restricted device” definition not only includes devices with “requirements” imposed at the time of the PMA process, but also devices subject to FDA “performance standards.”

Now, finally, here is the substantive FDA labeling regulation authorized and required by 21 U.S.C. §360j(e).  As you can see, it’s plenty broad enough – and expressly includes warnings:

(a) FDA may impose postapproval requirements in a PMA approval order or by regulation at the time of approval of the PMA or by regulation subsequent to approval.  Postapproval requirements may include as a condition to approval of the device:…

*          *          *          *

(3) Prominent display in the labeling of a device and in the advertising of any restricted device of warnings, hazards, or precautions important for the device’s safe and effective use … on risks and benefits associated with the use of the device.

21 C.F.R. §814.82(a) (emphasis added).

Thus the FDA’s power to order changes in the labeling of “restricted devices” covers just about any aspect:

(b) In specifying the labeling or change in labeling or change in advertising . . . eliminate or reduce the risk of illness or injury or the danger to the health of individuals, the Commissioner may require the manufacturer . . . responsible for the labeling or advertising of the device to include in labeling for the device, and in advertising if the device is a restricted device, a statement, notice, or warning.  Such statement, notice, or warning shall be in the manner and form prescribed by the Commissioner and shall identify the . . . risk of illness or injury or the unreasonable, direct, and substantial danger to the health of individuals associated with the device as previously labeled.

21 C.F.R. §895.25(b) (emphasis added).  That’s a lot of “shalls” and “requires.”

So, is this just the speculation of defense-oriented DDLaw bloggers?

Nope, we’re not making this up.  Rather, both the FDA and those courts that have addressed the topic of off-label warnings for medical devices rely on these regulations.  For example, in an early medical device preemption case, predating Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), the court extended preemption to labeling claims involving off-label use on this basis.

[T]he FDA can require a manufacturer to provide additional labeling that addresses potential off-label uses.  21 C.F.R. §895.25.  Consequently, the fact that [defendant’s] implant might have been used for an off-brand purpose is not sufficient to distinguish this case from [cases not involving off-label use].

Reeves v. AcroMed Corp., 44 F.3d 300, 305-06 (5th Cir. 1995) (emphasis added); see In re Orthopedic Bone Screw Products Liability Litigation, 1996 WL 221784, at *6 (E.D. Pa. Apr. 8, 1996) (“[t]hrough [§895.25(a)] the FDA regulates off-label uses of medical devices”); McGuan v. Endovascular Technologies, Inc., 106 Cal. Rptr.3d 277, 281-82 (Cal. App. 2010) (FDA’s power under §360j(e) to “condition its approval on adherence to various requirements” resulted in preemption of all claims, including for “off-label promotion”); see generally Carson v. Depuy Spine, Inc., 365 F. Appx. 812, 814 n.1 (9th Cir. 2010) (FDA “is free to impose device-specific restrictions by regulation.  §360j(e)(1).”); Caplinger v. Medtronic, Inc., 921 F. Supp.2d 1206, 1211 (W.D. Okla. 2013) (same); Wilhite v. Howmedica Osteonics Corp., 833 F. Supp.2d 753, 756 (N.D. Ohio 2011) (same).

In 2011, well after the expiration of 21 U.S.C. §360c(i)(E), the FDA cleared a §510k device, but imposed a black box warning pertaining to an off-label use.

FDA has determined that in order to provide reasonable assurance of safety and effectiveness, it is necessary to restrict the [device] to sale, distribution, and use with labeling, advertising, and promotional material that bears a warning statement in a black box that alerts users to the risk associated with off-label use. . . .  However, FDA believes it is necessary to require this warning in labeling and advertising by restricting the device under section 520(e) of the Federal Food, Drug, and Cosmetic Act (FD&C Act) (21 U.S.C. §360j(e)).

“Medical Devices; Ovarian Adnexal Mass Assessment Score Test System; Labeling; Black Box Restrictions,” 2011 WLNR 26903209 (FDA Dec. 30, 2011) (sorry, no internal pagination) (emphasis added).  There it is again – §360j(e) – used as authority for requiring that medical device labeling include warnings (in this case a boxed warning) about off-label use.  See also “Guidance For Industry And FDA Staff – Class II Special Controls Guidance Document: Ovarian Adnexal Mass Assessment Score Test System,” 2011 WL 1427005, at *19-20 (F.D.A. March 23, 2011) (containing the specific contents of the boxed off-label use warning in question).

Why do we care?  Number one, the FDA’s authority to require warnings and other information pertaining to off-label use demonstrates that off-label use isn’t some sort of unregulated and anarchic black hole.  The FDA still retains its usual authority to require the addition of risk information, subject to its established scientific standards.

Number two, implied impossibility preemption under LevineMensingBartlett should apply generally to prohibit all plaintiffs from making any demands for warnings pertaining to off-label uses against all prescription medical devices because, for both drugs and devices, such warnings explicitly and repeatedly require FDA pre-approval.  The FDA has always maintained tight control over the information that regulated manufacturers may provide that pertains to off-label uses.  Whether or not we like that as a First Amendment proposition, it’s a fact, and it’s reflected in the FDA’s regulations pertaining to off-label use warnings for both prescription drugs and medical devices.

Number three, in the specific context of PMA medical devices, the FDA’s power to require off-label information in labeling means that a plaintiff’s demand for off-label warnings “different from or in addition to” what the FDA has required (or chosen not to require).  That, in turn, mandates that such demands be expressly preempted under Riegel.

[R]egulations issued under section 520(e) . . .  of the act may impose restrictions on the sale, distribution or use of a device beyond those prescribed in State or local requirements.  If there is a conflict between such restrictions and the State and local requirements, the Federal regulations shall prevail.

Dunlap v. Medtronic, Inc., 47 F. Supp.2d 888, 895 (N.D. Ohio 1999).

Federal preemption is the strongest defense available to manufacturers of prescription drugs and medical devices.  Where available it should be employed to the maximum extent possible.