Thursday, September 03, 2015

Warning Letters and Jurisfiction

“Jurisfiction” is a word coined by Jasper Fforde, author of the Tuesday Next series, one of the more sophisticated set of children’s works that has come to populate this post-Harry Potter era.  To be very brief, Jurisfiction is the fictional police force for BookWorld, one of Fforde’s fictional universes.  Tuesday is a Jurisfiction agent (sometimes rather more than that).

“Jurisfiction,” unfortunately, is also something we see in our line of work, sometimes making us wonder whether the likes of Emperor Zhark, the Red Queen, and Pinky Perkins may have aliases who serve in the all-too-real judicial branches here in the States.  Jurisfiction is shorthand for a decision that gets a legal issue totally bollixed  – perhaps applying the UltraWord to the issue – allowing the user to control the plot, garbling it, and ultimately making all precedent useless.

We recently ran across a shining example of jurisfiction in the discussion of FDA warning letters found in Mihok v. Medtronic, Inc., ___ F. Supp.3d ___, 2015 WL 4722847 (D. Conn. Aug. 10, 2015).  Here’s what Mihok held on that subject:

The Complaint is rooted in FDA Warning Letters which state that [defendant] failed to comply with the CGMP regulations. . . .  While perhaps not dispositive on the issue, the FDA’s conclusions and interpretations of its own regulations are likely to receive a considerable degree of deference.  See, e.g., Conroy v. Dannon Co., Inc., No. 12 CV 6901(VB), 2013 WL 4799164, at *6 (S.D.N.Y. May 9, 2013) (stating that the FDA’s interpretations of its own regulations promulgated under title 21 “are ‘controlling unless plainly erroneous or inconsistent with the regulations’ or there is any other reason to doubt that they reflect the FDA’s fair and considered judgment”) (citing and quoting PLIVA, Inc. v. Mensing, ––– U.S. ––––, ––––, 131 S.Ct. 2567, 2575, 180 L.Ed.2d 580 (2011)); Dorsey v. Housing Auth. of Baltimore City, 984 F.2d 622, 632 (4th Cir. 1993) (finding district court abused its discretion in refusing to consider regulatory agency’s assessment of defendant’s compliance with agency regulations and noting that the district “court should welcome [the agency’s] appraisal of [the defendant’s] compliance with regulations, given its concern for deference to agency interpretations of its own regulations”). Indeed, it is precisely when a court is called upon to interpret the regulations, i.e., when they are ambiguous, and where their application to facts raises complex issues, that the court is most likely to defer to the FDA’s prior determinations.  See Wilson v. Frito-Lay N. Am., Inc., 961 F. Supp.2d 1134, 1142 (N.D. Cal. 2013) (noting that “an agency’s informal interpretation of its own ambiguous regulation is [typically] controlling” but declining to give “deference to two warning letters that the FDA sent” because neither party to the case “contended that the FDA regulations . . . [w]ere ambiguous, and the Court d[id] not find that they [we]re”); James T. O’Reilly, et al., 1 Food & Drug Admin. §4:56 (4th Ed. 2015) (“The FDA is allowed great deference in the interpretations of its own regulations.... The more complex the issue, the more scope is likely to be given for the FDA to draw the interpretations.”).

As to deference, Defendants cite a non-binding case, Schering-Plough Healthcare Prods., Inc. v. Schwarz Pharma, Inc., 547 F. Supp.2d 939 (E.D. Wisc. 2008), for the proposition that “a warning letter from the FDA is not considered a final agency action,” and contend that, as a result, “Plaintiffs’ allegations ... raise legal questions as to the potential effects of various actions by a federal agency ... [which] should be decided in a federal forum.”  The Second Circuit has not taken a position on whether an FDA Warning Letter is considered a final agency action.  Even if it is not, such letters may still be entitled to deference.  See Cmty. Health Ctr. v. Wilson-Coker, 311 F.3d 132, 138 (2d Cir. 2002) (“[E]ven relatively informal [agency] interpretations, such as letters from regional administrators, warrant respectful consideration” where the statute at issue is complex and the regulatory agency possesses “considerable expertise”) (citations and quotations omitted).  Regardless, they may serve as evidence of regulatory violations.  Gelber v. Stryker Corp., 788 F. Supp.2d 145, 155–56 (S.D.N.Y. 2011) (finding plaintiffs who provided FDA Warning Letters as evidence of violations of FDA regulations stated claims for manufacturing defects).

Mihok, 2015 WL 4722847, at *5-6 (citations not omitted, for once).  Under this analysis, there being no undecided FDA-related issue, Mihok was remanded to state court.

From beginning to end, that analysis is just backwards – or what the republican Brits in Bookworld might call an anagram for “restorative.”  Let’s start with the citations.  None of the citations in the first paragraph have anything to do with warning letters.  Conroy v. Dannon Co., 2013 WL 4799164 (S.D.N.Y. May 9, 2013), deferred to something called an “FDA Q & A,” which interpreted a regulatory stay published in the Federal Register – in other words, something that the FDA intended to be generally followed.  Id. at *6.  No warning letter of any sort there.  Conroy quoted general language from PLIVA v. Mensing, 131 S. Ct. 2567 (2011), about FDA deference.  Mensing was discussing the FDA’s formal litigation position, taken by the Department of Justice in an amicus curiae brief filed in that case at the express behest of the Supreme Court.  Id. at 2575.  Again, that brief took a position that the agency hoped would be accorded general application.  Dorsey v. Housing Auth. of Baltimore City, 984 F.2d 622 (4th Cir. 1993), didn’t deal with the FDA at all, but with “HUD’s 1992 Formal Utility Review.”  Id. at 627.  Note that word – “formal.”

No warning letters so far.  Which brings us to Wilson v. Frito-Lay, Inc., 961 F. Supp.2d 1134, 1142 (N.D. Cal. 2013), which refused to give any deference to two FDA warning letters.  Id. at 1142.  Mihok cited Wilson for dictum that FDA “informal” interpretation of “ambiguous” regulations should be “controlling unless plainly erroneous or inconsistent with the regulation.”  Id.  The “non-binding” (just as binding as Dorsey or Wilson) contrary case, Schering-Plough Healthcare Products, Inc. v. Schwarz Pharma, Inc., 547 F. Supp.2d 939 (E.D. Wisc. 2008), was not dictum, expressly rejecting the “position that the FDA letters express official FDA determination.”  There’s much more to Schering-Plough than the brief paraphrase in Mihok:

The FDA letters cited by [plaintiff] did not provide the defendants the opportunity for a hearing. . . .  Therefore, even if the FDA had purported to make an agency determination through the letters . . . such a determination would not have been in compliance with the explicit statutory requirements, the determination could not be considered an official agency determination to be afforded deference.

Indeed, several courts have recognized that letters such as those cited here do not constitute an official agency determination.  Furthermore, pursuant to 21 C.F.R. §10.85(k), “[a] statement or advice given by an FDA employee . . . is an informal communication that represents the best judgment of that employee at that time but does not constitute an advisory opinion, does not necessarily represent the formal position of FDA, and does not bind or otherwise obligate or commit the agency to the views expressed.”  Id.

547 F. Supp.2d at 946-47 (five citations omitted – discussed below).  The only other warning letter case Mihok cites, Gelber v. Stryker Corp., 788 F. Supp.2d 145 (S.D.N.Y. 2011), is mischaracterized as standing for the proposition that FDA warning letters “may serve as evidence of regulatory violations.”  Gelber, however, is a Rule 12(b)(6) pleading case; all it really stands for is that warning letters may give a defendant “ample notice” of a “plausible” claim under TwIqbal.  788 F. Supp.2d at 156-57.

Finally, there’s Community Health Center v. Wilson-Coker, 311 F.3d 132 (2d Cir. 2002), deferring to a Medicaid administrator’s letter issued under regulations that specifically authorized that sort of “discuss[ion]” and “consult[ation]” on “questions regarding application of Federal policy.”  Id. at 138 (quoting 42 C.F.R. §430.14 (2002)).

In sum, Mihok does not cite a single case giving any sort of deference to an FDA warning letter.  There’s plenty of reasons for that, and we’ll start with the statute – the Food, Drug and Cosmetic Act (“FDCA”) itself.  The FDA is an administrative agency, so its powers have to be delegated by Congress.  The FDCA expressly gives the FDA authority to address “minor violations . . . by suitable written notice or warning.”  21 U.S.C. §336.  That’s where warning letters come from.  If the FDA is content to address some possible violation by letter, by definition it’s a “minor violation.”  It has to be, otherwise, by statute, the FDA has to resort to some other more robust enforcement tool.

What does the FDA itself have to say?  As already discussed, Mensing gave considerable deference to the FDA’s position when appearing in a brief filed by the Department of Justice with the Court.  Just such a Supreme Court brief was filed on behalf of the FDA in 2012 in a case called Holistic Candlers & Consumers Ass’n v. FDA (No. 11-1454).  It’s on Westlaw at 2012 WL 3991471, and those are the citations we’re going to use, but, if you’re looking to save money, you can read the same brief for free on the DoJ website, here.  The case involved, of all things, ear candles, but the appeal concerned precisely the point that Mihok rejected – that FDA warning letters are informal, tentative, and thus completely non-final.  The court of appeals had ruled exactly the opposite of Mihok on the significance of FDA warning letters:

FDA’s warning letters . . . neither mark the consummation of the agency’s decisionmaking process nor determine the appellants’ legal rights or obligations.  The letters plainly do not mark the consummation of FDA’s decisionmaking. . . .  FDA warning letters . . . giv[e] firms an opportunity to take voluntary and prompt corrective action before [the FDA] initiates an enforcement action. . . .  Nor do the letters represent a decision determining rights or obligations, or one from which legal consequences flow. . . .  [A] Warning Letter is the agency’s principal means of achieving prompt voluntary compliance with the [FDCA].  Although a warning letter communicates the agency’s position on a matter, it is only informal and advisory and does not commit FDA to taking enforcement action. . . .  In short, an FDA warning letter compels action by neither the recipient nor the agency.

Holistic Candlers & Consumers Ass’n v. FDA, 664 F.3d 940, 943-44 (D.C. Cir. 2012) (citation to the FDA’s regulatory procedures manual and quotation marks omitted).

The Holistic folks sought certiorari, and the Department of Justice opposed, telling the Supreme Court firmly that FDA warning letters, although providing notice to recipients and giving them the opportunity to respond informally, were not any sort of definitive agency determination of any legal matter:

If FDA believes that a person is violating the FDCA, the agency may issue a warning letter giving the person an opportunity to take voluntary corrective measures before the agency pursues enforcement action.  Warning letters are “the agency’s principal means of achieving prompt voluntary compliance with the [FDCA].”  FDA, Regulatory Procedures Manual, 4-1-1 (July 2012).  A warning letter is “informal and advisory”; it “communicates the agency’s position on a matter” but “does not commit FDA to taking enforcement action.”  Ibid.

FDA Holistic Candlers DoJ br., 2012 WL 3991471, at *5.  That’s one.

[T]he warning letters here did not mark the consummation of FDA’s decisionmaking process. . . .  Violations identified in warning letters “may lead to enforcement action if not promptly and adequately corrected.”  Regulatory Procedures Manual 4-1-1. . . .  Relatedly, the warning letters were not based on a formal and complete administrative record.  At this stage, FDA’s statement that petitioners violated the FDCA was not “final and binding” on the agency or petitioners but rather remained “tentative [and] interlocutory [in] nature.”

Id. at *9-10 (citation omitted) (emphasis original).  That’s two.

Nor did the warning letters finally determine the rights or obligations of petitioners . . . or trigger direct and appreciable legal consequences.  Rather, the letters “request[ed]” that petitioners “take prompt action” . . ., and cautioned that “[f]ailure to promptly correct these deviations may result in regulatory action.”  The letters served only to communicate FDA’s position . . ., and to warn recipients about the possibility of future enforcement action.  See Regulatory Procedures Manual 4-1-1.  The letters nonetheless remained “informal and advisory” and “[did] not commit FDA to taking enforcement action.”  Ibid.

Id. at *10-11 (citations and quotation marks omitted).  That’s three.

FDA warning letters trigger no legal consequences and are subject to further agency “evaluat[ion]” based on the recipient’s response.  Such letters do not trigger any enlarged exposure to penalties for noncompliance with the FDCA, nor does their issuance preclude further agency consideration or review.

Id. at *14.  That’s four.  After the DoJ/FDA told the Supreme Court all this, the Court denied the Holistic folks’ petitioni for certiorari.  See 133 S.Ct. 497 (2012).

So, if what was really going on in Mihok was Mensing-style deference to the FDA’s position, then “deference” should have required treating warning letters as:  “informal,” “advisory,” “not commit[ting] the FDA,” “not based on a formal and complete administrative record, “not final and binding,” “tentative,” not “finally determin[ing]” “rights or obligations,” “trigger[ing] no legal consequences,” “subject to further agency “evaluat[ion]” − a far cry from how they were characterized in Mihok.  Critically, that’s the FDA’s own characterization of a warning letter.  The Mensing deference language traces back to the “considered judgment” standard in Auer v. Robbins, 519 U.S. 452, 462 (1997), in which the Court was also dealing with a position taken in a legal brief filed on behalf of an agency.  We strongly doubt that Auer deference, or any other sort of deference, can legitimately be given to anything, such as FDA warning letters, when the very agency that issued them states so clearly and repeatedly that they are not binding, even (indeed, especially) on itself.

The FDA’s position is underscored by the regulatory procedures manual that the DoJ cited several times in its brief in Holistic Candlers.  Turns out that, unlike the citations in Mihok, the manual says exactly what the DoJ told the Supreme Court it did.  “[I]t is the Food and Drug Administration’s (FDA’s) practice to give individuals and firms an opportunity to take voluntary and prompt corrective action before it initiates an enforcement action.  Warning Letters are issued to achieve voluntary compliance and to establish prior notice.”  FDA Regulatory Procedures Manual §4.1.1 (“Warning Letter Procedures”).  Warning letters “are based on the expectation that most individuals and firms will voluntarily comply.”  Id.  “A Warning Letter is informal and advisory.  It communicates the agency’s position on a matter, but it does not commit FDA to taking enforcement action.  For these reasons, FDA does not consider Warning Letters to be final agency action.”  Id. 

Furthermore, the FDA’s Regulatory Procedures Manual makes pretty clear is that warning letters are “not appropriate” if alleged violations are “repeated,” “continual,” “intentional,” “flagrant,” or “willful”; or if “[t]he violation presents a reasonable possibility of injury or death.”  Id.  This confirms the first point we made above – that warning letters are only authorized by the FDCA for “minor violations,” not the kind of things that, in tort litigation, might bring punitive damages into play.

This is hardly all.  Mihok also cited (2015 WL 4722847, at *6) a book section, James T. O’Reilly, et al., 1 Food & Drug Admin. §4:56 (4th Ed. 2015).  That’s the treatise’s general FDA deference section, and like many of the cases Mihok cites, that section doesn’t say “boo” about warning letters.  However, other sections of the same treatise do.  Section 6:2 of the O’Reilly treatise, entitled “Warning letters—What they are and how to respond,” refutes practically everything stated in Mihok:

[A] warning letter is not “final agency action,” it is not a “finding” that warrants negligence per se liability action. . . .  A trial court denied a defendant’s request for judicial notice of an FDA warning letter sent to it regarding its Internet marketing.  Warning letters, even as supplemented by FDA’s Web site and the appellants’ conversations with FDA officials, do not constitute final agency action, and so a judicial review complaint is not cognizable under the APA and must be dismissed for failure to state a claim.  An FDA warning letter is not a final decision by the FDA and its position may change after further investigation.  A warning letter does not constitute evidence in the record that the FDA has determined or will determine that a product’s mark is deceptive. . . .  FDA warning letters have not been judicially reviewable “final action” but some cases will suggest a basis for such a review.

O’Reilly, et al.,1 Food & Drug Admin. §6:2 (2015) (various footnotes omitted) (emphasis added).  Thus the same treatise states that warning letters are:  (1) “not final agency action,” (2) not anything that “warrants negligence per se,” (3) not subject to “judicial notice,” (4) subject to “change after further investigation,” and (5) not “evidence in the record that FDA has determined” anything.  And that’s just the most pertinent section of this treatise.  See also Id. §4:45 (“Statements of lower-level agency officials do not rise to the level of final agency action, even when they are contained in warning letters”); §10:17 (“FDA warning letters are insufficient evidence of how reasonable consumers would view the labels”); §15:73 (“an FDA warning letter did not support an inference of the requisite scienter,” quoting Avon Pension Fund v. GlaxoSmithKline PLC, 343 F. Appx. 671, 674 (2d Cir. 2009)); §26:130 (“a warning letter from the FDA is informal and advisory”).  Don’t look at us; we know O’Reilly has appeared as a plaintiff-side FDA expert, but that’s the treatise that Mihok chose to cite.

Finally, we turn to case law.  As we suspected from Mihok’s inability to come up with any warning letter precedent that actually stood for the propositions being advanced, courts are close to unanimous that warning letters aren’t “final” in any meaningful sense.  “[E]very court to consider the question has held that an FDA warning letter does not constitute ‘final agency action.’”  Cody Laboratories, Inc. v. Sebelius, 446 F. Appx. 964, 969 (10th Cir. 2011).  While warning letters may have helped push some complaints past the TwIqbal “plausibility” standard on Rule 12 motions to dismiss (see Gelber, supra), that’s about it.  We’ll start with Holk v. Snapple Beverage Corp., 575 F.3d 329 (3d Cir. 2009), where the defendant relied upon “several” FDA warning letters on a particular topic.  The court refused to defer to them:

We do not think these letters are sufficient to accord the policy the weight of federal law. . . .  Congress likely intended to give administrative action the effect of law when the agency adhered to a relatively formal administrative procedure tending to foster the fairness and deliberation that should underlie a pronouncement of such force. . . .  In this case, the deficiencies inherent in the process by which the FDA arrived at its policy . . . are simply too substantial to be overcome by isolated instances of enforcement. . . .  Both lack the formal, deliberative process.

Id. at 341-42 (citations and quotation marks omitted).  Thus, when the FDA “took steps to address the allegations by issuing warning letters . . . it did not take final action.”  Perez v. Nidek Co., 711 F.3d 1109, 1120 (9th Cir. 2013).  FDA “letters are not final agency action . . ., as there has been no final agency action, let alone action that has been or could be judicially reviewed.”  Schering-Plough Healthcare Products, Inc. v. Schwarz Pharma, Inc., 586 F.3d 500, 508 (7th Cir. 2009) (affirming the “nonbinding” decision that Mihok shrugged off).  “[T]he type of informal letter issued by the FDA . . . does not constitute . . . formal or final agency action.”  Dietary Supplemental Coalition, Inc. v. Sullivan, 978 F.2d 560, 562-63 (9th Cir. 1992).  FDA warning letters are “informal communications.”  Professionals & Patients for Customized Care v. Shalala, 56 F.3d 592, 599 (5th Cir. 1995).  “Warning letters merely establish a dialogue between the FDA and the [addressee] and do not necessarily lead to further sanctions.”  Biotics Research Corp. v. Heckler, 710 F.2d 1375, 1378 (9th Cir. 1983).  A warning letter “d[oes] not represent the FDA’s final word,” thus there is no “per se” rule requiring their disclosure in securities litigation.  Fire & Police Pension Ass’n v. Abiomed, Inc., 778 F.3d 228, 243 & n.9 (1st Cir. 2015).  Add the Holistic Candlers case quoted extensively above, and that’s a lot of directly-on-point court of appeals decisions discounting the adjudicative value of FDA warning letters.

The West Virginia Supreme Court of Appeals (not normally pro-defense in our experience), when faced with a verdict based on the purported collateral estoppel effect of “false and misleading” language contained in an FDA warning letter, decided that there was no such thing – because such letters had no finality.  That court said so nearly as many times as the FDA did in its Holistic Candlers brief:

The Appellants contend that the circuit court erred by giving preclusive effect to DDMAC’s determination that their [drug related] statements were false and misleading.  They point out that, pursuant to the FDA’s own guidelines, warning letters are merely “informal and advisory” and do not constitute a final judgment of the FDA.  The Appellants further assert that, despite the circuit court’s finding to the contrary, they did not have the ability to administratively appeal the allegations contained in those warning letters and, thus, were never afforded the opportunity to defend against the FDA’s informal determinations.  This Court agrees.

State ex rel. McGraw v. Johnson & Johnson, 704 S.E.2d 677, 687 (W. Va. 2010) (footnote omitted).  That’s one.

Here, no preclusive effect can be given to the FDA’s determination that the statements and omissions in the [defendant’s statements] are false and misleading, because the FDA did not render a “final adjudication on the merits” on this issue, nor did the Appellants have an opportunity to fully and fairly litigate the question.

Id. at 688.  That’s two.

Clearly, in issuing warning letters, the FDA is not acting pursuant to any adjudicatory authority, nor does it employ any due process procedures similar to those accorded defendants in courts of law.  Rather, in issuing a warning letter, the FDA, acting pursuant to its regulatory authority, attempts to remedy a perceived violation through informal means.  No hearing is provided prior to the issuance of the letters, nor is the recipient notified of the alleged violations.  Indeed, the purpose of the warning letters is to provide such preliminary notification, thereby giving the alleged violator an opportunity to resolve the problem in an informal manner before actual adjudication takes place.  Accordingly, the warning letters cannot be considered quasi-judicial determinations by the FDA and, thus, are not subject to collateral estoppel under West Virginia law.

Id. at 689.  That’s three.

Plenty of federal district court cases reach the same holdings.  In Von Koenig v. Snapple Beverage Corp., 713 F. Supp.2d 1066 (E.D. Cal. 2010), the court agreed with Holk, supra, concluding that an “informal” FDA policy, “arrived at without the benefit of public input or formal procedures” and enforced by warning letters, should not “be accorded the weight of federal law.”  Id. at 1075.

The fact that [defendant] eventually received the Warning Letter from the FDA does not change things.  The receipt of the Warning Letter . . . does not establish [defendant] “knew” that it was out of compliance with CGMPs. . . .  [E]ven a warning letter from the FDA is “informal and advisory.”  While it reflected the FDA’s position on the matter, it did not commit the FDA to taking enforcement action.

In re Genzyme Corp., 2012 WL 1076124, at *10 (D. Mass. March 30, 2012) (citations omitted).  Just because a warning letter states that a product is “adulterated” doesn’t mean that it is, because such letters are “an informal and advisory method that does not commit the FDA to taking enforcement action.”  In re Digitek Products Liability Litigation, 821 F. Supp.2d 822, 829 (S.D.W. Va. 2011).

“FDA warning letters are informal and advisory, and do not amount to an FDA action.”  CytoSport, Inc. v. Vital Pharmaceuticals, Inc., 894 F. Supp.2d 1285, 1294 (E.D. Cal. 2012).  “[T]he FDA warning letter is not a final decision by the FDA and its position may change after further investigation.”  Id. at 1296.  “There is nothing magical about [an FDA] warning letter. Although the language sounds ominous, it really is rather boilerplate.”  Anderson v. Abbott Laboratories, 140 F. Supp.2d 894, 902 (N.D. Ill. 2001).  “[A]n FDA warning letter does not necessarily compel action by the recipient nor the agency, and does not represent a decision from which legal consequences would flow.”  Craig v. Twinings, Inc., 2015 WL 505867, at *7 (W.D. Ark. Feb. 5, 2015).  A “warning letter is not a ‘final agency action,’ but rather is “is a tentative or interlocutory action.”  Regenerative Sciences, Inc. v. FDA, 2010 WL 1258010, at *7 (D. Colo. March 26, 2010).  It “is correct that FDA warning letters are not binding.  As the FDA Regulatory Procedures Manual states, ‘A Warning Letter is informal and advisory.’”  Gitson v. Trader Joe’s Co., 2013 WL 5513711, at *10 (N.D. Cal. Oct. 4, 2013).  “[R]egardless of any warning letters that the FDA may have sent . . ., it is clear that the FDA has not completed this investigation.”  “FDA’s warning letters do not represent the consummation of FDA’s process, determine any legal rights or obligations, or affect plaintiffs’ legal rights.”  Cody Laboratories, Inc. v. Sebelius, 2010 WL 3119279, at *12 (D. Wyo. July 26, 2010), aff’d in pertinent part, 446 F. Appx. 964 (10th Cir. 2011).  “[R]egardless of any warning letters that the FDA may have sent to Defendants, it is clear that the FDA has not completed this investigation.”  Summit Technology, Inc. v. High-Line Medical Instruments Co., 922 F. Supp. 299, 306 (C.D. Cal. 1996).  “It is well-settled . . . that an agency’s investigatory activity does not constitute final agency action.”  Genendo Pharmaceutical N.V. v. Thompson, 308 F. Supp.2d 881, 884 (N.D. Ill. 2003).  Thus, an “FDA letter is insufficient to establish a violation.”  Chan v. Orthologic Corp., 1998 WL 1018624, at *20 (D. Ariz. Feb. 5, 1998).  “Regulatory letters . . . do not amount to final agency action” and “do not bind the agency to the views expressed in them.”  Clinical Reference Laboratory, Inc. v. Sullivan, 791 F. Supp. 1499, 1503-04 (D. Kan. 1992), aff’d in part & rev’d in part on other grounds, 21 F.3d 1026 (10th Cir. 1994).

Further, it’s hard to see how something so tentative could constitute admissible evidence.  Forget hearsay; most states have rules of evidence equivalent to Fed. R. Evid. 404(6) (“[e]vidence of other crimes, wrongs, or acts is not admissible” except for enumerated exceptions), and Rule 609 (governing when witnesses can be impeached with criminal “convictions”).  Thus, we believe that, in most jurisdictions, juries don’t get to hear about mere arrests or indictments for purported “crimes” in the absence of any conviction.  By contrast, an FDA warning letter doesn’t even reach the formality of an indictment.  Rather, in the words of the O’Reilly treatise, “warning letters . . . pointedly tell the offending violator to ‘knock it off.’” 1 Food & Drug Admin. §6:1.  A mere warning is not substantive evidence of an “other wrong.”

For these reasons, a number of cases have excluded FDA warning letters and other informal FDA enforcement activity:  Last year in Ortho-McNeil-Janssen Pharmaceuticals, Inc. v. State, 432 S.W.3d 563 (Ark. 2014) (which we discussed here), the court reversed a substantial verdict in part because an FDA warning letter was improperly admitted:

“A Warning Letter is informal and advisory.  It communicates the agency’s position on a matter, but it does not commit FDA to taking enforcement action.  For these reasons, FDA does not consider Warning Letters to be final agency action on which it can be sued.” . . .  [F]or evidence to be admissible, it must be more probative than prejudicial. . . .  Here, the “Warning Letter” was highly prejudicial.  Reports issued by governmental agencies, because of their “official” nature, may well carry inordinate weight in the minds of jurors.

Id. at 579 (quoting FDA Regulatory Procedures Manual, other citations and quotation marks omitted) (note, the link on WL is broken).  See King v. Danek Medical, Inc., 37 S.W.3d 429, 441-42 (Tenn. App. 2000); Newman v. McNeil Consumer Healthcare, 2013 WL 4460011, at *17-18 (N.D. Ill. March 29, 2013); In re Viagra Products Liability Litigation, 658 F. Supp.2d 950, 966 (D. Minn. 2009); In re Seroquel Products Liability Litigation, 2009 WL 223140, at *4-5 (M.D. Fla. Jan. 30, 2009); Chan, 1998 WL 1018624, at *5; Summit Technology, Inc. v. High-Line Medical Instruments, Co., 933 F. Supp. 918, 934 & n.9 (C.D. Cal. 1996).  The Second Circuit, in a case not cited in Mihok, excluded an “interim” governmental report from evidence, concluding that non-final agency action would mislead a jury due to its having “an aura of special reliability and trustworthiness . . . not . . . commensurate with actual reliability.”  New York v. Pullman, 662 F.2d 910, 915 (2d Cir. 1981).  Ditto FDA warning letters.

In light of all of this precedent, we think that the funhouse-mirror view of FDA warning letters expressed in Mihok – that they are entitled to significant “deference” – is an outstanding example of jurisfiction – a fanciful view of the law unsupported even by the cited materials and contrary to both the FDA’s own position and to overwhelming precedent on all levels.

Wednesday, September 02, 2015

Plaintiffs’ Daubert Challenge Rejected

            This post is from the non-Reed Smith side of the blog.

            If you skim through our Daubert posts, they are usually celebrating successful defense challenges to plaintiffs’ experts.  We have posts on repeat offenders like Blume and Parisian.  We have posts on unqualified experts – those with no educational or professional experience on the topics they wish to opine.  We have lots of posts on unsupported opinions, insufficient data and unreliable methodology.  And there are posts on experts who seek to invade the province of the jury (those who want to interpret corporate documents) or the court (those whose vocabulary include things like negligent and reckless). 

            It is less common for plaintiffs to file Daubert motions seeking to exclude defense experts.  But that is just what happened in In re Zimmer Nexgen Knee Implant Prods. Liab. Litig., 2015 WL 5050214 (N.D. Ill. Aug. 25, 2015).  Which means today we get to celebrate the denial of a Daubert motion – a Daubert win from a slightly different angle.

            On of the issues in the Zimmer Nexgen litigation is whether defendant’s high-flex knee implants have an artificially higher failure rate than their equivalent counterparts.  Id. at *1.  To dispute such a conclusion, defendant proffered testimony from an expert epidemiologist, Dr. Vitale, who conducted a literature review from which he concluded that the studies relied on by plaintiffs were outliers.  Id.  Plaintiffs challenged both the qualifications and methodology of defendant’s expert.

            On qualification, plaintiffs focused on the fact that defendant’s expert is a pediatric spine surgeon; that he does not specialize in knee replacements.  Plaintiffs, however, overlooked that Dr. Vitale also has a master’s degree in public health with an emphasis on clinical outcomes research, an area in which he has also published.  Id. at *4.  His testimony and conclusions were not based on his experience as a pediatric spine surgeon, but rather his “epidemiological and general clinical research expertise.”  Id. at *5. 

            The analysis of Dr. Vitale’s research methodology is more complex and multi-faceted.  We’ll walk through the important aspects, but the overall conclusion of the court is that his methodology was reliable and admissible.  Those aspects of his research and conclusions plaintiffs take issue with do not warrant exclusion, but rather can be tested through cross-examination.

            What was his research?  Dr. Vitale conducted a “formal systematic review of the literature.”  (He reviewed other data as well that plaintiffs did not challenge).  In the simplest terms, Dr. Vitale used search terms to find relevant literature on the topic at issue or to create a “neutral snapshot of the existing research on a particular question.”  Id. at *3.  Irrelevant studies – those that don’t address the research topic – are omitted, as are studies that have other flaws.  Id. at *4.  The remaining literature is reviewed and scored and the research usually concludes with summary statistics and qualitative findings.  Id. at *3.  Plaintiffs did not challenge that a formal systematic literature review is an invalid methodology, but rather that the way Dr. Vitale conducted his review made it unreliable.

            Plaintiffs’ first argument was that Dr. Vitale failed to comply with internationally recognized guidelines for systematic reviews known as PRISMA.  Id. at *6.  The court was skeptical of plaintiffs’ argument from the outset considering plaintiffs’ own expert neglected to cite PRISMA, “which casts some doubt on Plaintiff’s claim of widespread acceptance.”  Id  at *7.  More importantly, the court noted that PRISMA is really about the “reporting” of systematic reviews, not their “conduct.”  Id. While reporting is important for publication, it isn’t really the cornerstone of a Daubert analysis.  So, failing to report his results graphically, for example, isn’t something that calls into question the reliability of Dr. Vitale’s research.  Id. at *8.  Nor was it fatal that Dr. Vitale’s report did not include an explicit statement of his research question.  Again, this is important for publication so that subsequent researchers can determine if the analysis is applicable to their interests.  In the context of litigation, Dr. Vitale’s research objective was clear – both from his deposition testimony and the totality of his report.  Id. at *7.  Would Dr. Vitale’s research and conclusions have to be reported differently to pass peer review for publication?  Maybe.  But that’s not Daubert. 

            Plaintiffs next challenged Dr. Vitale’s opinion on the ground that he commingled heterogeneous studies – studies of varying lengths, size, follow-up, etc.  Id. at *9.  While heterogeneity can be problematic, Dr. Vitale’s report acknowledged that “the variability among study cohorts precludes aggregate analysis.”  Id.  It is up to the plaintiffs to cross-examine Dr. Vitale regarding this inclusion and exclusion of certain studies, it is not the court’s role “to determine whether Dr. Vitale’s ultimate conclusion is the right one.”  Id.  That’s for the jury.

            Plaintiffs also alleged that Dr. Vitale “cherry-picked” the studies he included in his review.  Maybe a little pot calling the kettle on this one.  There is no shortage of instances where defendants have challenged plaintiffs’ experts on this basis.  Here, however, the court did not find evidence of inappropriate selection or exclusion by Dr. Vitale.  First, Dr. Vitale did not ignore the studies as plaintiffs contend.  All were cited and discussed in the report, including the reasons why they were excluded.  The court took note of Dr. Vitale’s even-handed exclusion, for objective scientific reasons, of studies that were both favorable and unfavorable to the defense position.  Id. at *10-11.  While there may have been oversights, those should be the topic of cross-examination.  There was no basis to conclude that Dr. Vitale attempted to distort the data such that his results should be excluded as unreliable.  Id.

            Finally, plaintiffs challenged Dr. Vitale’s subsequent analysis on range of motion.  In response to Dr. Vitale’s initial report, plaintiffs claimed that he had failed to consider the range of motion achieved by the patients in the studies (plaintiffs contend that failure rates may be higher for those who achieve high flexion as opposed to those who do not).  So, Dr. Vitale went back to look at that issue.  Plaintiff then claimed that Dr. Vitale “is required to conduct a completely new systematic review with a search driven by this new question.”  Id. at *12.  If we understand correctly, plaintiffs’ position is that because the defense expert did a subsequent analysis to respond to plaintiffs’ criticism, he has to start from scratch and do his analysis based on plaintiffs’ stated objective.  The court understood that too:

Dr. Vitale simply revisited those studies to determine whether [plaintiffs’] attack had merit and updated his data accordingly.  The court does not agree with Plaintiffs that this subsequent analysis was invalid or unreliable.


            So, when the bellwether trials in the Zimmer NexGen litigation begin, the jury will be able to hear from Dr. Vitale that in his expert epidemiological opinion, the studies relied on by plaintiffs are outliers.  He’ll be allowed to present his methodology and his conclusions with a stamp of approval from the judge that he is both qualified to do so and that he used a reliable methodology.  Sure, plaintiffs can pick around the edges – but most of their arguments have been rejected.    It is slightly odd to be lauding the denial of a Daubert motion, but Zimmer’s expert report in this case seems to stand head and shoulders above what we often see from plaintiffs. While we may be the ones more often challenging things like cherry-picking and commingling, it’s nice to see that when plaintiffs tried to turn the table – the defendant still came out on top. 

Tuesday, September 01, 2015

New Jersey Federal Baked Goods Fraud Class Actions Are Toast

Almost on this date in 1901 (tomorrow actually), Teddy Roosevelt for the first time uttered in public the immortal phrase, "Speak softly and carry a big stick". It is hard to find people who do not admire that statement.  It is harder to find people who actually practice it. These days we are accustomed to wimpy parents and blowhard politicians who talk tough but do nothing.   As a result, you get teenagers gobbling up the cheese-of-the-month delivery and Russia gobbling up Crimea.  We also encounter some soft-hearted and -headed judges who speak loudly and wield no stick – at least when it comes to clamping down on bogus lawsuits.  Corporate defendants hauled/haled/but-definitely-not-hailed into court are held to ludicrously high standards, while plaintiffs dwell in a world of do-overs and feckless flexibility. 

A couple of weeks ago we discussed CD Cal Judge Wilson, who is as demanding as he is smart.  One could say (and we have said) the same about Judges Posner, Boggs, Kozinski, and Rakoff.  One could also say that about D NJ Judge Irenas who sits in the Camden federal courthouse, which we can see out our window if we crane our heads just so.  If you read the Robing Room evaluations of Judge Irenas (which you must necessarily take with a grain of salt, since they are peppered with the comments of sore losers) a picture emerges of intelligence and rigor.  

We can be an oath-helper on that fact.  After our first year of law school, we summered at McCarter & English in Newark, NJ. Irenas was a partner there, and was universally acknowledged to be the smartest lawyer in the building.  In our second or third week as a summer associate, we received an assignment from Irenas.  The other summer associates cackled with glee. That is because in addition to having a reputation for being smart, Irenas had a reputation for not suffering fools gladly.  And all of the summer associates were fools.  An Irenas assignment was an opportunity, but a frightening one. 

The assignment involved the inevitable research memo.  After we turned it in to him and were grilled to a nice medium-well, we got on a conference call.  At one point, the party on the phone voiced some trepidation and reservations, whereupon Irenas thundered a withering critique of said reservations. Loud dysphemisms filled the air.  He must have seen us shaking with terror, because he smiled and gestured to show us that his finger, which was resting on the microphone (back then speakerphones were separate from the actual telephone, and were connected by a wire) was also pushing down on something we did not know existed - a mute button.  He then released the button and rendered a much more temperate, but still piercing, dissection of the reservations.  What he said was about a hundred times more insightful and felicitously expressed than our pathetic research memo.  Lesson learned.  

Our encounter with Irenas was all too brief that long-ago summer.  In the decades since we have never encountered a smarter lawyer.  Eventually, he became a judge, and his reputation for intelligence followed him to the federal courthouse.  Of course, there are plenty of smart judges out there.  But not all are willing to apply that intelligence to winnow away weak litigation.  A delicious example of a brilliant judge taking a close look at a case and dismissing it for not passing muster is Mladenov v. Wegman's Food Markets, Inc., 2015 U.S. Dist. LEXIS 112740 (D.N.J. August 26, 2015). 


Mladenov was actually three cases – three purported diversity class actions in which the plaintiffs alleged that three different grocery stores misrepresented bread and bakery products as being baked fresh in the store, when they were actually frozen, processed, or baked in another location, supposedly in violation of New Jersey consumer fraud statutes as well as express warranties.  The class was defined as all individuals and entities within New Jersey who purchased bread and/or bakery products advertised and sold as “made in house” and/or “freshly baked” and /or "freshly boiled” and/or “fresh” in those three grocery stores located in New Jersey on or after December 14, 2008. 

Judge Irenas tested the ingredients of the lawsuits layer by layer.  The court initially asked the parties to address whether the classes would be ascertainable.  The proposed classes consisted of consumers who purchased individual allegedly-misrepresented products and the proposed method of identifying class members would rely on retail records.  The class definitions also required that the purchased products were specifically advertised at the time as based in store or baked fresh.  But here is a big problem:  the retail records, such as receipts, would not reflect such information.  Moreover, the plaintiffs did not dispute that cash purchasers of the bread and bakery products who did not use loyalty cards could not be identified in any reliable way.  So much for ascertainability.  The court also decided that certification under Fed. R. Civ. P. 23(b)2) – which applies to actions for injunctive or declaratory relief – was unavailable here because even though such claims were included in the prayer for relief, it was obvious that the lawsuits were primarily for money damages. 

Getting beyond the class certification issue, the court went on to dismiss the substantive claims because the plaintiffs failed to plead unlawful conduct, ascertainable loss, or a causal relationship between the alleged unlawful conduct and loss.  The plaintiffs alleged that the defendants displayed in-store signs such as “STORE BAKED ROLLS.” But the complaints did not allege that the plaintiffs actually saw such signs, in which store that occurred, or when the plaintiffs saw it.  As the court observed, the defendants’ stores contain numerous bread and bakery products and the signs advertising such products change often.  It was simply not the case that all relevant items were stamped “freshly baked.” 

Further, nowhere in their complaints or anywhere else did the plaintiffs allege facts supporting an out-of-pocket loss, i.e., that the products they purchased were worthless.  Wherever the bread was baked, it was presumably edible.  The plaintiffs claimed to be particularly health conscious consumers, but they did not allege that the relevant products were somehow less nutritious due to their not being made from scratch in the store.  As the court saw it, the real allegation underlying the case was that the plaintiffs paid an unnecessary premium for what they believed to be store-made bread.  That would be a benefit-of-the-bargain theory of ascertainable loss.  But the plaintiffs “gave no basis for valuing the products they received as opposed to the products they were promised.”  Rather, the plaintiffs alleged only that they purchased various bread and bakery products at premium prices over the years and would not have done so “in absence of Defendant’s misleading advertisement.”  Easy to say.  The problem with this theory ties in with the class action ascertainability issue:  the plaintiffs did not specify any instance in which they even saw the defendant’s advertisements, either in the stores or on websites.  The plaintiffs also failed to allege which food products they purchased as a result of viewing the advertisements.  Thus, the court could not, without more, “infer from Plaintiffs’ pleadings a link between an affirmative misrepresentation and an ascertainable loss.” 

The court slices open the plaintiffs’ allegations and finds more puff than filling.  We then get to watch each legal theory fall, one by one.  Thus, the court tartly dispenses with the misrepresentation claims because the “conclusory allegations regarding numerous potential purchases of various products over a substantial period of time with the mere specter of supposedly misleading advertisements generally existing in Defendants’ stores and websites will not suffice under Rule 9(b)’s heightened pleading standard.”

The warranty claims disintegrated into mere crumbs.  An express warranty claim “requires a plaintiff to allege that she brought a product based on a particular promise regarding that product, which ultimately proved false.”  But, again, the plaintiffs cannot successfully plead such a claim without identifying in the complaints any specific sign or advertisement they saw and the products they purchased as a result.  Again, the plaintiffs simply did not satisfy the recipe.     

The claims for injunctive and declaratory relief also found their way to the waste bin.  There was no plausible claim of threat of immediate harm.  How could there be?  After all, the plaintiffs disclaimed any intention to continue to purchase the defendants’ bread and bakery products.  Instead, the plaintiffs argued that they were entitled to injunctive relief based on the threat of future harm to other consumers. How thoughtful.  How inadequate.   At this stage, when no class has been or ever will be certified, the court considered “Plaintiffs’ claims as they apply to Plaintiffs alone, not the putative class.”     

We will end with the icing on the cake, the aspect of the case that is most relevant to drug and device law.  The plaintiffs’ claims under the New Jersey consumer protection statute appeared to rely on regulations (specifically, 21 C.F.R. § 101.95) under the Food Drug and Cosmetic Act.  But it is “well settled … that the FDCA creates no private right of action.”  The plaintiffs in this bakery fraud case could not use the New Jersey statute “to bootstrap a FDCA claim they could not otherwise bring.”  That is a tasty result.  (We cannot resist wondering why California courts resist this sort of logic and rigor, and why they are willing to entertain so much silly food litigation premised on nothing more than a hunger for attorney fees.) 

Mladenov may not be a drug or device case, but what it says about no FDCA private right of action, or class ascertainability, or connecting specific representations to injury, are all pertinent to our field.  Plus there’s this:  all lawyers should know that one cannot walk into Judge Irenas's courtroom with legal arguments that are half-baked. 

Monday, August 31, 2015

Guest Post - Game of Thrones, the FDA Under Attack, An In-House View of Amarin

Today's guest post is a first for this blog in two ways, first, our guest poster is anonymous.  Second, our guest post is from an in-house source.  The two are, of course, related.  While it is important to us as lawyers that our clients not be charged with anything we say on our blog, that goes double (or triple) for someone actually working in-house.  So you won't know exactly who to credit, or blame, for what follows.

What follows is our in-house source's thoughts on the recent Amarin First Amendment victory over the FDA.


We are witnessing a strange chapter in the hegemony of FDA control over drug promotion.  For while the OIG and DOJ continue to make good use of their Swiss army knife of an Anti-Kickback Statute, the FDA has been taking body blows from First Amendment litigators.  First, the Supreme Court’s Sorrell decision, Sorrell v. IMS Health Inc., 131 S. Ct. 2653 (2011), protecting promotional communications as commercial speech; next, the Second Circuit in United States v. Caronia, 703 F.3d 149 (2d Cir. 2012), vacating a misbranding conviction on First Amendment grounds; and finally, the Southern District of New York in Amarin Pharma, Inc. v. FDA, ___ F. Supp.3d ___, 2015 WL 4720039 (S.D.N.Y. Aug. 7, 2015), declaring that a manufacturer may legally engage in off-label promotional communications.


Any doubts that this may indeed be a watershed event should be assuaged by the FDA’s public relations campaign to marginalize Caronia and the agency’s anxious efforts to moot the Amarin case or controversy before decision.


After all, the attack here is not a flanking maneuver.  It is aimed straight at the FDA’s weakened center and threatens to scatter the spoils of decades of doctrinal toil spent erecting the divide between on- and off-label communications, an article of faith in the industry.  And, to add insult to injury, and in marked contrast to the restraint shown in Caronia, the opinion in Amarin deliberately sabotages a key justification for FDA’s land grab on this subject −  that the FDA must be allowed to regulate truthful, non-misleading speech to protect the integrity of the drug approval process – by reasoning that the regulatory scheme “predates modern First Amendment law respecting commercial speech” and is thus vulnerable to “frontal assault” from manufacturers.  2015 WL 4720039, at *25.


Of course the victory is not conquest, and needs to be tempered by the acknowledgement that Caronia and Amarin reflect Second Circuit jurisprudence with limited precedential value outside of New York, Connecticut and Vermont.  The decision is also not a blanket license to promote off-label.  Different facts, and in particular different safety risks or less reliable scientific evidence, may result in a different balancing and warrant greater restrictions to protect the public health.  And there is the court’s admonition that the First Amendment protects speech, not conduct, 2015 WL 4720039, at *27, so promotional tactics designed to increase off-label use may still receive the full treatment from OIG and DOJ.  That being said, manufacturers are ahead of the FDA 3-0 in cases where the federal courts have applied Central Hudson’s commercial speech balancing test (see Central Hudson Gas & Electric Corp. v. Public Service Communication, 447 U.S. 557 (1980)). to drug promotion (4-0 if you include the venerable Washington Legal Foundation decision, WLF v. Friedman, 13 F. Supp.2d 51 (D.D.C. 1998)).


Given these caveats, the end game that will emerge from whatever chaos attends the demise of the FDA’s position is difficult to handicap at this point in time, which calls to mind Littlefinger’s apt observation, “Chaos is not a pit.  Chaos is a ladder.”  Interested observers, and forward thinking advisors, might be well-advised to climb that ladder and consider what this new world order would look like if this view from Manhattan were fully realized, at least for the sake of discussion.  For one thing, the heretofore binary world, divided between scientific exchange (unregulated) and promotion (heavily regulated), turns into a tripartite world, with a third category one could label “protected speech”:


So while the distinction in 21 C.F.R. § 312.7, cordoning off communications that are made outside of “a promotional context,” still applies to emancipate scientific exchange, promotional speech would now be separated into communications subject to traditional FDA controls and a new category of truthful and non-misleading promotion that is neither necessarily on-label nor based upon “substantial evidence.”  In effect, the dichotomy between on- and off-label communications disappears, since none of these three categories relies on the distinction as a criterion.


Most importantly, the authority for what is “truthful and not misleading” is not inevitably the FDA − which, if asked, is likely to take the position that the standard is coterminous with the existing substantial evidence requirements.  Put another way, the Second Circuit decisions stand for the proposition that “truthful and not misleading” data does not need to meet the FDA’s heightened standard, or require FDA approval, before it can be communicated to healthcare providers.


This change, should it take hold, would be significant, but probably not as momentous as the hysterics among us would have you think.  For one thing, one would expect that the drill for updating safety information would remain unchanged, since the interests of industry and the FDA are closely aligned on this issue.  There is every incentive for manufacturers to continue to work through the existing FDA processes to bring about label revisions to ensure that accurate and up-to-date contraindications, adverse reactions and interactions are reflected in the label (and failure to warn defenses in product liability cases preserved).  There is also much that remains on the FDA’s plate in terms of policing efficacy claims, even if the new standard is changed to truthful and not misleading.  If anything, the change should presage more, rather than less work for FDA staffers, now stripped of the thick armor provided by a blanket off-label prohibition.


Would this be a good thing for patient health?  It is hard to see how it would not be.  Manufacturers have always been the best source for detailed scientific information about their products and have clamored for many years against seemingly arbitrary restrictions on their ability to communicate this information.  Earlier efforts to fix the problem have fallen victim to what appeared to be kneejerk conservatism from the FDA – a condition that got the agency into this battle in the first place.