Photo of Michelle Yeary

Today’s case — In re Tepezza Mkt’g, Sales Pracs., & Prod. Liab. Litig., MDL No. 3079, 2025 WL 81338 (N.D. Ill. Nov. Jan. 10, 2025) – doesn’t really break new ground on the substantive law, in fact it is a split decision.  But it comes from an unusual procedural posture worth pointing out. 

The Tepezza MDL was created in June 2023 and shortly thereafter the court ruled on a pending motion to dismiss in a single case.  The court denied the motion as to plaintiff’s pre-market design defect claims.  We were not particularly fond of the ruling and said so here.  It appears that since that time, the court decided to create a group of “bellwether discovery” cases.  The idea is that motions to dismiss get teed up and ruled on in the 12 “bellwether discovery” cases to establish the scope of discovery.  We can’t help but be intrigued with any process which has as a component setting boundaries on discovery.  Realistically, particularized rulings on plaintiffs’ claims could lead to less discovery.  For instance, certain time periods could be deemed out of bounds.  Or, if design defect claims are dismissed, that might limit the amount of research and development discovery that is warranted.  Furthermore, tethering the discovery to the claims of 12 actual plaintiffs, regardless of the rulings on the motions, could also be potentially helpful to defendants.  Often, discovery against the manufacturer in an MDL tosses open the doors to the company at large because plaintiffs claim they are taking discovery for all plaintiffs, for all purposes, for all claims, for all times.  Putting discovery in context, including the context of 12 individual plaintiffs, almost always inures to the benefit of the defense.  For our readers who spend a significant time slogging it out with plaintiffs on discovery disputes, we’ll keep our eyes on whether this discovery bellwether generates any useful precedent or tips going forward.

Now to turn to the substance.  Plaintiffs in the Tepezza MDL allege “hearing loss and/or tinnitus” from the use of an FDA-approved prescription biologic for thyroid eye disease.  Defendants challenged the bellwether plaintiffs’ failure to warn, design defect, and fraudulent misrepresentation.  They won one, lost one, and split one. 

On failure to warn, defendant argued that the claims were preempted because plaintiffs failed to adequately plead that defendant could have used the CBE process to change its label before September 2022—after plaintiffs had stopped their treatment.  First, we should point out that while Tepezza is a biologic, the CBE regulation is substantively identical to that for prescription drugs.  So, to avoid preemption, a plaintiff “must plead that the manufacturer had newly acquired information that showed a causal association between the drug and an effect that warranted a new or stronger warning, precaution, or adverse reaction in the label.”  In re Tepezza, at *4.  In that situation, the manufacturer does not need FDA approval to change its label; therefore, no preemption.  Here, plaintiffs cited to a clinical study that came out in September 2020 that showed a sufficient increase in the incidence rate to arguably satisfy the CBE requirements.  Id. at *5.  Defendant argued that the study could not form the basis for a CBE label change because it was not a “clinical trial,” but the court found “criticism of the quality of [the study] . goes to the sufficiency of the evidence proving the allegations, not the sufficiency of the allegations.  Since all of the bellwether plaintiffs used Tepezza after September 2020, their failure to warn claims were not preempted at the pleadings stage. 

On design defect, plaintiffs failed to meet the TwIqbal pleading standard.  In fact, plaintiff only made boilerplate allegations offering no facts and giving the court no basis from which to infer what aspect of the product makes it defective.  Plaintiffs tried to argue that under at least Florida and New York law, they didn’t need to allege a design defect before discovery because they do not have the technical information required to do so.  However, plaintiffs relied on cases in which the complaints contained many more factual details than they allege here.  Plaintiffs don’t have to prove a specific defect in their complaint, but they have “to do more than assert the conclusion that Tepezza is defective.”  Id. at *7.  Plaintiffs’ allegations about dosage were not supported by the evidence they pointed to in the complaint.  Moreover, simply alleging that dose is too high does not satisfy the need to allege an alternative design.  Id. at *9.  With design defect dismissed, plaintiffs asked for permission to amend their complaints—long after the MDL deadline to do so.  They claimed that they had new evidence that defendant was conducting a clinical trial on a new delivery method.  Not only was the study announced months before the pleadings deadline, but it was also not scheduled to be completed until 2026 and had no safety or efficacy results as of yet.  The ”mere initiation of a study” does not “plausibly suggest” a defect in the current delivery method or indicate that the method being studied will turn out to be safer or more efficacious.  Plaintiff did not offer good grounds for their motion to amend, and it was denied.

That left only plaintiffs’ fraudulent misrepresentation claim which was essentially the same as their failure to warn claim.  Plaintiffs’ fraud claim was based on two allegations—that defendant overstated the efficacy of the biologic and defendant misstated the incidence rate.  Applying Rule 9’s heightened pleading requirement for fraud, the court found both allegations were lacking.  The first was simply not supported by any factual allegations (“plaintiffs’ allegations regarding efficacy are highly generalized”).  Id. at *10.  As to the second, plaintiffs had conflicting allegations, including statements that the incidence rate on the label was accurate.  “So, Plaintiffs’ assertion of falsity is not well pled.”  Id. at *11. While the fraud itself was not sufficiently pleaded, the court found plaintiffs’ fraud damages need not be pleaded with particularity.  And defendants raised one more issue—that because the fraud claim is based on the FDA-approved label, it is actually a preempted fraud-on-the-FDA claim.  Here the court concluded it had insufficient information on which to base a decision. Because the fraud claim is substantively the same as the failure to warn claim, a ruling on Buckman preemption would not change the scope of discovery, and therefore the court deferred the issue.  So, on fraudulent misrepresentation the final ruling is defendant’s motion to dismiss is granted in part (fraud allegations), denied in part (damages); and continued in part.  Except as to the Pennsylvania plaintiff because Pennsylvania does not recognize fraud claims based on a failure to warn theory.  Her claim was dismissed with prejudice. 

Like we said, a 50/50 split decision on the claims—but a possible case to watch for discovery. 

Photo of Bexis

Back in 2008, we wrote a post, No, Bu shi, Non, Iie, Nada, Nyet…., collecting a significant body of law holding that overseas defendants are not required to translate, at their expense, documents prepared in their non-English home languages.  The other day we came across Sessoms v. Toyota Motor Sales, U.S.A., Inc., ___ S.E.2d ___, 2024 WL 5249823 (N.C. App. Dec. 31, 2024), reversing yet another order that a foreign defendant pay for translating its own documents:

[W]e conclude the trial court erred by requiring the . . . Defendants to create new documents in English of documents already provided that are in the Japanese language.  Rule 26 of our Rules of Civil Procedure allows a party to seek documents in the possession of the adverse party; it does not generally require the adverse party to pay for any said documents to be translated into the English language.  In other words, there is no duty to produce documents that do not exist.

Id. at *2 (citation omitted).  Sessoms was “persuaded by what we perceive to be the greater weight of authority in the United States that a party producing documents is not required to create new documents consisting of English translations of documents already provided.”  Id. at *3.  Sessoms cited eight cases exemplifying that “authority” – three of which were decided after our 2008 post.  That suggests that the post could use an update.

So here’s that update.

Continue Reading Still Nyet, Defendants Not Required to Pay for Translation of Documents
Photo of Eric Hudson

We write a lot about the learned intermediary rule. There are 50 state surveys and  summaries of helpful decisions, as well as numerous posts on state-specific decisions. We tracked the development of the rule in jurisdictions like West Virginia and Arizona, and we’ve generally been pleased to report positive developments. At the end of 2024, though, we flagged the Himes case from California as one of the ten worst decisions of the year based on its novel approach to warnings causation.  Given our criticisms of Himes, we found it both bizarre and troubling to see the case cited recently by an MDL court in Massachusetts applying Pennsylvania law. Our colleagues in the plaintiffs’ bar are clearly advocating to expand Himes into other jurisdictions, and the defense bar should be ready to counter those efforts.

Continue Reading Himes Makes a Sneak Appearance on the East Coast
Photo of Eric Alexander

This time of year, many of us are focused on the NFL playoffs.  For someone who watches the Super Bowl for the commercials or the halftime show, which team wins may not matter much.  For those devoted to a particular team, however, there is one possible result that will be truly satisfying.  The reality is that, the fans of 96.875% of the teams end the season with dissatisfaction and those of 92.857% of the playoff teams end the season with an excruciating loss.  Of course, there are those who can find something positive from a non-championship season.  Maybe the fan’s favorite team showed a big improvement over the prior season or the team finally found a franchise quarterback, either of which could lead to optimism for the next season, even though it will probably result in another non-championship.

In the world of product liability MDLs, way more than 99% of the cases filed end in dismissal.  In any given MDL, very few, or perhaps none, of the cases that the plaintiff lawyers file will result in judgment for the plaintiff.  Of course, the path to and nature of the dismissal will determine how the parties and their counsel feel about it.  A dismissal in connection with a settlement or a dismissal without prejudice will not be as dissatisfying for the plaintiff as a dismissal with prejudice after summary judgment or a defense verdict at trial.  For the defendants in an MDL, the reality is that satisfaction is affected by a range of economic considerations.  Managing an MDL where the vast majority of cases essentially sit around for years waiting for settlement requires a focus on the big picture as well as the case count.

Subject matter jurisdiction in a product liability MDL will almost always be based on diversity and the defendant will usually prefer that any given case within the MDL’s definition proceed in the MDL instead of in state court.  There is a clear gap in the proceeding statements.  A plaintiff with the same state of residence as the defendant(s) cannot stay in federal court based on diversity jurisdiction, and there is no special federal jurisdiction for MDL courts.  In In re Cook Med., Inc., IVC Filters Mkt’g, Sales Pracs. & Prods. Liab. Litig., MDL No. 2570, 2024 U.S. Dist. LEXIS 235780 (S.D. Ind. Nov. 14, 2024) (“Cook”), the defendants succeeded in getting the cases of eleven plaintiffs from Indiana, where the defendants are also based, dismissed without prejudice for lack of subject matter jurisdiction.  That is only part of the story.  The plaintiff lawyers admitted they did not have diversity jurisdiction for these cases when they were directly filed in the MDL and no other basis for keeping the cases in the MDL.  And they refused to dismiss the cases when requested.  So, the defendants sought Rule 11 sanctions in connection with the dismissals.  This was largely symbolic because they sought only the costs of bringing the motions to dismiss in nine cases and later reduced the request to $100 per case.  They apparently did not pursue other avenues for sanctions or costs, probably because Rule 11 allows sanctions to be awarded against an attorney or law firm, not just against a party.

The Cook court denied sanctions.  While the plaintiff lawyers had no argument about jurisdiction, their excuse for refusing to voluntarily dismiss when requested was that Indiana’s savings clause would not provide a year to re-file if the dismissals were voluntary.  The court thus saw the refusals to dismiss as not being sanctionable because they were “in the best interest of Plaintiffs, their clients.”  2024 U.S. Dist. LEXIS 235780, *4.  Here is the problem with that decision.  The cases were directly filed in the MDL in 2018 to 2020, several years after the MDL was established.  [We have tracked various rulings from and related to this MDL over the years, like here, here, here, and here.]  Subject matter jurisdiction is measured at the time the case is filed in or removed to federal court.  Plaintiff lawyers should know the law well enough to ensure that they do not file time-barred cases and/or cases in courts without jurisdiction.  There would have been no need to use the Indiana savings clause if the cases had been timely filed in Indiana state court originally or if the mistake of filing in the MDL had been discovered before the statute of limitations expired.  These cases were pending in the wrong court for four to six years before the savings clause became the excuse for forcing motions to dismiss to be filed.  When the court had to rule on those motions and the attendant obvious sanctions issue, the delay should not have been exculpatory.  After all, the Cook court did find that the plaintiff lawyers “should have realized that Plaintiffs’ cases did not fall within the diversity jurisdiction of this court,” a determination that would have applied at all points from before filing until the ultimate dismissal.  Id. 

Sanctionable conduct by a lawyer should not get a retroactive blessing if it persists long enough to hurt the lawyer’s clients.  Nor is the consideration that sanctions “can affect the reputation and creativity of counsel” usually going to be sufficient for an MDL to refrain from sending the message that filing frivolous cases and refusing to dismiss them is sanctionable.  Id. (quoting from Hartmarx Corp. v. Abboud, 326 F.3d 862, 867 (7th Cir. 2003)).  Indeed, because the savings clause would not resurrect a case that was untimely when filed, if any of these cases were filed in the wrong court after the statute of limitations had expired, then the much later refusal to dismiss served no interest of the lawyer’s client.

By contrast, just about any unsupportable claim or position advanced by a plaintiff lawyer could be said to be done in the “best interest” of the lawyer’s client, who presumably wants to win a big award or secure a big settlement.  That may be why Rule 11 has no provision that excuses unsupported representations to the Court—such as a jurisdictional allegation in a complaint—based on the interest of the party on whose behalf the representation is made.  It does specify, however, that any “sanction imposed under this rule must be limited to what suffices to deter repetition of the conduct or comparable conduct by others similarly situated.”  Fed. R. Civ. P. 11(c)(4).  In the context of an MDL, trying to deter the direct filing of cases without subject matter jurisdiction makes sense.

Photo of Stephen McConnell

Blair v. Abbvie Inc., 2025 WL. 57198 (W.D. Pa. Jan. 9, 2025), is, from the defense perspective, a favorable opinion dismissing (some with prejudice, some with leave to amend) all counts of the plaintiff’s complaint.   The opinion is a bit odd, in a semi, unintentionally-ironic sort of way, because it faults the plaintiff for not supplying enough information, while the court’s opinion does not tell us what the product is, except it is some kind of implant.  We’re guessing that the implant had something to do with the plaintiff’s eye, because the court observes in a footnote that the plaintiff included among the named defendants an eye care subsidiary … that did not actually exist.  Call it foreshadowing. In most of the complaint, there was no there there.

The Blair opinion follows Pennsylvania law to the effect that Restatement (Second) of Torts section 402A, Comment k requires dismissal of strict liability design claims.  Some W.D. Pa decisions bounced strict liability design defect claims on this ground and some did not. The Blair court sided with the better W.D. Pa. decisions, logic, truth, the American way, and an E.D. Pa. decision (Smith v. Howmedica Osteonics Corp., 251 F. Supp. 3d 844, 847-51 (E.D. Pa. 2017) – it is always delightful to see a sound decision from our home district) by applying comment k to bar the strict liability claims.  That same logic ended up also barring the implied warranty design defect claims, because the elements of the strict liability and warranty claims were “coextensive.” Dismissal was with prejudice because there is nothing the plaintiff could do to fix those claims. 

Sadly for the defendants, comment k did not bar the negligent design defect claim, nor any species of the manufacturing defect claims.  But sadly for the plaintiff, all the other claims in the complaint were inadequately pleaded. What the complaint said was purely formulaic. The allegations were “too broad and conclusory.” All the plaintiff tells us is that the product was “manufactured and sold by” the defendants, “that it was defective when it was sold, that it reached him without changes in its condition, and that he was injured after it was implanted.” Not good enough. The manufacturing claim did not state what went wrong or even how the product allegedly failed.  Such gaps/omissions meant that the claim did not even rise to the level of a dreaded “malfunction theory” claim.  

This inadequate pleading carried over to the plaintiff’s negligence-based claims, all of which were also dismissed. The plaintiff did not “identify the design defect, anything about the manufacturing process, or what information” the defendants should have given to the plaintiff’s medical providers. The Blair court helpfully gathers some good W.D. Pa. precedents, and they will be useful if you find yourself defending a case in the Pittsburgh area.  The judges will likely take the time to read such precedents, as they will not be distracted by the local professional football team, which has made its annual first round exit from the playoffs.

Meanwhile, the plaintiff will get the opportunity to try to do better.  So, next year, will the Steelers.  We predict that neither will succeed.

Photo of Michelle Yeary

Defendant in Beavan v. Allergan U.S.A., Inc., 2014 N.J. Super. Unpub. LEXIS 2898 (N.J. App. Nov. 21, 2024) made two solid arguments for summary judgment – preemption based on the FDCA’s recall regulations and plaintiff’s lack of admissible expert testimony.  The trial court rejected both.  The appellate court, however, saw the merit in the expert argument and that was all defendant needed to secure a dismissal. 

Defendant manufactures a prescription drug that is used in the treatment of various eye diseases.  The drug is administered by injection.  A routine product inspection revealed that approximately 2.2% of the units of a particular lot of the drug contained a silicone particulate which could be dispensed when the drug was injected.  Defendant sought FDA approval to send out a Dear Health Care Provider letter (“DHCP letter”), but the FDA concluded the issue was not a “safety concern.”  Instead, and with the FDA’s approval, defendant issued a drug recall and a notice to physicians that there was a low risk of a mild reaction.  Id. at *2-4.  Plaintiff received an injection of defendant’s drug from the lot that was recalled.  After which she was diagnosed with retinal detachment.   Plaintiff filed her complaint alleging claims for design, manufacturing, and warning defects under New Jersey’s Products Liability Act. 

Defendant’s preemption argument centered on the FDA’s recall regulations which it contended preempted the state from regulating prescription drug recalls.  The court disagreed finding that the voluntary nature of the recall defeated defendant’s implied preemption argument.  Because a manufacturer can add a warning to its labeling, which includes via DHCP letters, without FDA approval, defendant was not foreclosed from warning physicians about the manufacturing defect and therefore, there was no preemption under Wyeth v. Levine, 555 U.S. 555 (2009), and its progeny.   Further, while the NJ PLA affords defendants a presumption that their FDA-approved labeling is adequate, the presumption is rebuttable with allegations that the manufactured acquired knowledge of harmful effects after the sale of the product and failed to disclose them.  In other words, NJ recognizes a post-sale duty to warn.  Here, defendant’s DHCP letter was sufficient to rebut the adequacy presumption and therefore a sufficient basis on which to assert a non-preempted failure to warn claim.

But that is where the bad news ends because plaintiff’s experts took a leap of faith and assumed that the injection plaintiff received contained the defect.  You know what they say about people who assume.  While both of plaintiff’s experts testified that the silicone particulate caused plaintiff’s injuries, neither offered any evidence that the “injection plaintiff received was defective and no evidence of a particulate in her eye.”  Id. at *34.  Therefore, plaintiff’s theory of causation was “based on evidence that does not exist and would leave a jury to speculate whether there was ever a particulate in the applicator or particulate injected into plaintiff’s eye.”  Id.   New Jersey’s net opinion rule prohibits the admission of an expert’s conclusions that are not supported by “factual evidence or other data.”  Id. at *32. 

Not only did Plaintiff’s experts based their conclusions on an unsupported assumption, they failed to rule out that plaintiff’s injuries could have been caused by various other factors—such as a different silicone insert used in her treatment that became dislocated at the same time she suffered her injury or that retinal detachment is a risk of some of plaintiff’s pre-existing and underlying eye diseases.  Id. at *34-35.  The court also acknowledged that New Jersey has recently adopted the Daubert factors for expert admissibility and that plaintiff’s experts would not pass under that standard either.  Id. at *35-36. 

The case was reversed and remanded for summary judgment to be entered on the grounds of no expert evidence of specific or general causation.  We would have liked a better preemption outcome, but we are not greedy.  A lack of causation is good enough for us any day.

Photo of Bexis

This blog has long encouraged defendants in prescription medical product liability litigation to seek relevant ediscovery from plaintiffs.  We even have an ediscovery cheat sheet with almost 250 favorable decisions either allowing defense-side ediscovery in personal injury cases or else sanctioning plaintiffs for spoliating sought-after electronic data.  But we confess, we’ve been focused so firmly on social media and smartphones, where ediscovery from plaintiffs originated, that we have ignored the rising popularity of fitness trackers, Fitbits, smart watches, smart rings and similar devices (even clothing) being marketed to people who may eventually become plaintiffs.  These products create a great deal of health-related (and other) information that is of obvious relevance in mass (and other) tort litigation.

What we found is that surprisingly few defendants seem to be seeking this type of information – at least there are very few decisions involving discovery of these devices.

Continue Reading Ediscovery for Defendants – The New Frontier
Photo of Bexis

For any of our loyal readers looking to start the New Year with a healthy helping of CLE credit, a friendly reminder that four of your bloggers – Bexis, Steven Boranian, Stephen McConnell, and Lisa Baird – will be presenting a free 90-minute CLE webinar on “The Good, the Bad and the Ugly: The best and worst drug/medical device decisions of 2023” on Thursday, January 16th at 12 p.m. EST. The webinar will provide further insight and analysis on the cases featured in our posts on the worst decisions and best decisions of the past year.

This program is presumptively approved for 1.5 CLE credit in California, Connecticut, Illinois, New Jersey, New York, Pennsylvania, Texas and West Virginia. Applications for CLE credit will be filed in Delaware, Florida, Ohio, and Virginia. Attendees who are licensed in other jurisdictions will receive a uniform certificate of attendance but Reed Smith only provides credit for the states listed. Please allow 4-6 weeks after the program to receive a certificate of attendance.

Additionally, please note that CLE credit for on-demand viewing is only available in California, Connecticut, Illinois, New Jersey, New York, Pennsylvania, Texas, and West Virginia. Credit availability expires two years from the date of the live program. In order to receive CLE credit, you will need to notify Learning & Development CLE Attendance once you have viewed the program on-demand. 

The program is free and open to anyone interested in tuning in, but you have to sign up in advance here.

Photo of Steven Boranian

Every time we read an order about “pre-approval” defects in prescription drugs, we stop to scratch our heads.  It is not because we are easily confused, at least not on most days.  No, we pause to ponder what exactly a “pre-approval” defect is and how a company can be held to answer for an alleged defect in a product’s design or in its warnings before the product is approved, let alone marketed and used to treat a patient. 

“Pre-approval” defect is a legal construct invented by plaintiffs’ lawyers to dodge federal preemption.  When it comes to design defect, the signpost is Mutual Pharma v. Bartlett, where the Supreme Court held that federal law preempts state law design defect claims involving a generic drug because the manufacturer could not have changed the design without the FDA’s pre-approval.  That makes it impossible to comply with state and federal requirements at the same time—thus preemption.  Cases like the Sixth Circuit’s opinion in Yates v. Ortho-McNeil extended that to innovator products, on the ironclad reasoning that a manufacturer of an innovator product cannot change the product’s design without FDA approval any more than a generic manufacturer can. 

The latest plaintiffs to use claims of “pre-approval” defect to avoid preemption (successfully, for now) are the plaintiffs in the Suboxone Film MDL, where they allege that suboxone film which dissolves under the tongue caused dental issues.  Under Bartlett, federal law should preempt any design defect claim, since the defendant could not have altered the product design without FDA approval.  To get around that, though, the plaintiffs argued that federal law did not prevent the defendant from proposing a different design before FDA approval.  In re Suboxone Film Prods. Liab. Litig., No. 24-md-3092, 2024 WL 5264278, at *8 (N.D. Ohio Dec. 31, 2024).  As an example of how the defendant could have designed a safer product, the plaintiffs cited a drug with the same active ingredient, but as an injectable gel—i.e., a different product.  Id. at *11. 

This is a “pre-approval” defect claim, and you can see why it makes no sense.  This “pre-approval” claim is merely a claim that, under state law, the manufacturer should never have sold the product in its approved form.  But, the FDA, under federal law, said that it could. 

State law should have given way.  But it didn’t, at least not for now.  The district court dedicated an entire section of its order to the presumption against preemption (you can see the smoke coming out of Bexis’ ears now).  It also rejected Yates as precedent because the Yates opinion noted that results “might differ ‘in some circumstances’” and because the court found older opinions that Yates left untouched to be more persuasive.  Id. at *9-*10.  The district court also said that no other circuit has followed Yates, an assertion that we have not verified.  Regardless, we reported just over a year ago that preemption of “pre-approval” design defect claims was a majority position.  We doubt that has changed. 

The district court therefore denied the defendant’s motion to dismiss on pre-approval design defect, but it granted the motion on “post-approval” design, under which the plaintiffs alleged that the product’s design was defective after the FDA approved the product.  This of course is a claim that the defendant should “stop selling” a federally approved drug in order to comply with state law, which is obviously preempted under Bartlett.  The district court correctly so held.  Id. at *11. 

That brings us to the plaintiff’s warnings based claims, where they alleged that the defendant did not adequately warn of adverse dental effects until a label change in June 2022.  If you thought “pre-approval” design defect claims were questionable, then “pre-approval” failure-to-warn claims are downright nonsense.  There is no duty to warn of risks in products that are not yet being sold, and the FDA is the final arbiter of what prescription drug labeling can and cannot say once products enter the market.  Under Supreme Court precedent (in Levine v. Wyeth, Pliva v. Mensing, Mutual Pharma v. Bartlett, and Merck v. Albrecht), there is only one exception to federal preemption under these circumstances—when the plaintiff can show that “newly acquired information” would have allowed the defendant to unilaterally strengthen its label under the FDA’s Changes Being Effected (or “CBE”) regulation. 

And the CBE regulation does not apply pre-approval.  So how can there be a “pre-approval” failure to warn claim?  There can’t be. 

This district court could have said that, but instead it wrote that because the plaintiff did not use the product “before its approval,” the court “need not address whether FDA approval of the label forecloses any warning claim arising before then.”  Id. at *12.  We are not sure why the court felt compelled to say this.  No one used the product before it was approved, and no failure-to-warn claim could have arisen before approval. 

The real controversy was on the garden-variety, “post-approval” failure-to-warn claim under which the plaintiffs argued that a December 2022 publication provided “newly acquired information” on adverse dental effects that could have supported a unilateral label change, thus avoiding preemption.  The defendant countered that it had already updated the label to include adverse dental effects in June 2022 and that the December publication revealed nothing new.  The district court, however, ruled that the plaintiffs had alleged newly acquired information sufficient to survive a motion to dismiss.  Perhaps the outcome will be different on a more-developed record. 

The concept of “pre-approval” claims still makes our heads hurt, but we have written so much on them by now that our views are firm.  A “pre-approval” design defect claim is just a claim that the defendant never should have sold the approved product.  And a “pre-approval” failure-to-warn claim is merely a claim that the drug labeling, as approved by the FDA, was not adequate.  No mental gymnastics are required to find such claims preempted. 

Photo of Lisa Baird

“Off label use” is a bit of an odd thing.  The FDA does not regulate the practice of medicine, but it does get involved in the labeling of medical devices and pharmaceuticals through the applicable pre-marketing review processes, and those labels identify the use (or uses) intended for that medical product.  That use (or those uses) are ones for which there is reasonable evidence of safety and efficacy, in the FDA’s estimation, but the practice of medicine moves faster than the regulatory process.  Once an approved device or drug is available, doctors can prescribe it for any purpose they believe is warranted—regardless of what the label says.  Indeed, sometimes it is the medical standard of care to prescribe a medical product for uses other than what is stated on the label.

But when a use is off-label, what role can or should a manufacturer have in discussions with physicians or medical debate about those off-label uses?  The idea of manufacturers paying for ads that say “use our ProductTM for condition X” when the FDA has only reviewed the clinical data and authorized the ProductTM to be labeled for condition Y makes many uneasy.  But what if using Product for condition Y has become the standard of care—can the manufacturer communicate with doctors about that?  What if there are quality, published scientific articles about that unapproved use?  Can the manufacturer unilaterally send out copies?  Can the manufacturer provide copies of that quality, published scientific article, but only if a physician asks for it? 

The FDA has—shall we say—fluctuated over time about what is, or is not, permissible for a manufacturer to say or do with respect to off-label uses (or, in its parlance, “unapproved uses”).  There was an earlier final guidance issued in 2009, “Good Reprint Practices for the Distribution of Medical Journal Articles and Medical or Scientific Reference Publications on Unapproved New Uses of Approved Drugs and Approved or Cleared Medical Devices.”   

That was replaced in 2014 with a draft guidance “Distributing Scientific and Medical Publications on Unapproved New Uses–Recommended Practices.” 

That March 2014 draft guidance was replaced by a revised draft guidance in October 2023 (“2023 Draft Guidance”). 

And now, on Monday, January 6, 2025, the FDA issued its new final guidance, “Communications From Firms to Health Care Providers Regarding Scientific Information on Unapproved Uses of Approved/Cleared Medical Products:  Questions and Answers” (“2025 Final Guidance”) to describe its

enforcement policy regarding certain firm-initiated communications of scientific information on unapproved use(s) of the firm’s approved/cleared medical products to health care providers (HCPs) engaged in prescribing or administering medical products to individual patients.  

The Federal Register notice is here.

Overall, the framework of the 2025 Final Guidance will be familiar.  It describes certain types of manufacturer-initiated communications with health care providers about off-label uses that should not result in enforcement action by the FDA, which clearly suggests that other types of communications about off-label uses (even if truthful and not misleading) are potentially fair game for FDA enforcement.

In particular, the 2025 Final Rule attempts to articulate how a manufacturer of an approved/cleared medical product can stay out of regulatory trouble by providing a Q&A series about what to do, or not to do.  For example, any manufacturer-initiated communications to physicians about off-label uses should be “scientifically sound” and of a certain type (published scientific or medical journal articles, clinical practice guidelines, reference texts, or digital clinical practice resources).  They also must be accompanied by the current approved labeling, additional statements (about the approved use, that the unapproved use has not been approved by FDA, and that the safety and effectiveness of the medical product for the unapproved use(s) has not been established); and relevant warnings and contraindications; and so on.

There are, of course, First Amendment questions raised by any attempts to regulate speech about off-label use, and those questions remain present with the 2025 Final Guidance as well.  If the manufacturer’s speech about the use of ProductTM for Condition Y is not false or misleading in any respect, can a regulation restricting that speech be consistent with the First Amendment?  That is a topic near-and-dear to the heart of Bexis, as covered in numerous posts. 

The 2025 Final Guidance does acknowledge, at least to some degree, that there are First Amendment implications raised by a regulatory guidance directed at speech.  But there is nothing really new from the Agency on the First Amendment issue in the 2025 Final Guidance.  Mostly, the 2025 Final Guidance cites the FDA’s own January 2017 Memorandum, “Public Health Interests and First Amendment Considerations Related to Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products” which was pretty well picked apart in this post.

More recently, Bexis has suggested that Loper Bright provides a pathway for courts to independently examine the statutory and regulatory framework (house of cards?) that the FDA traditionally has used to justify enforcement of its policies about off-label promotion, and decide for themselves whether they provide the authority the FDA says it has.  The 2025 Final Guidance takes a stab at articulating the relevant “FDA Authorities” that empower its oversight of information about off-label uses, but it is no more compelling than the Agency’s earlier justification attempts.  Compare 2025 Final Guidance at 7 & fn. 13-16 with 2023 Draft Guidance at 2 & fn. 3.

Bexis also has posited that the FDA exceeds its authority in attempting to regulate information about off-label uses, and thus is vulnerable to a legal challenge to its regulatory guidances, in Is the FDA’s Off-Label Speech Proposal Infected with the Ultra Vires?  Nothing in the 2025 Final Guidance seems to address or anticipate Loper Bright-based challenges, however.

Off label will never be off topic on this blog, so undoubtedly there will be more to come as we ruminate some more now that the 2025 Final Guidance finally is here.

Drug & Device Resources

Bexis’ Book

FDA-regulated products now account for an estimated one-fifth of overall economic activity in the U.S. They have also been the focus of a litigation explosion. This timely guide covers all aspects of litigation involving drugs, medical devices, vaccines and other FDA-regulated prescription products.
Learn More About This Book