Orange Book Blog

At the intersection of Patent and FDA Law

Recent Posts

  • Delaware court says 74% is not “about 80%”

    One of the most commonly recurring and fascinating issues in Hatch-Waxman cases is the presence of “about” in claims. Such claims are a challenge for generic defendants, who lack certainty on how they can safely design around such claims, but also presents patentees with challenges as well as they must then demonstrate that “about” covers the claimed range and fend off indefiniteness challenges.

    On its face, an “about” claim seems to run afoul of Section 112’s requirement for “claims particularly pointing out and distinctly claiming the subject matter” which the inventor regards as invention. Of course, there are cases finding such claims invalid for indefiniteness, including most recently for example in  Enviro Tech Chemical Services, Inc., v. Safe Foods Corp., No. 2024-2160 (Fed. Cir. May 4, 2026). But the Federal Circuit has also long held that words like “about” and “approximately” may be appropriately used to “avoid[] a strict numerical boundary to the specified parameter.” OrthoMcNeil Pharm., Inc. v. Caraco Pharm. Lab’ys, Ltd., 476 F.3d 1321, 1326 (Fed. Cir. 2007) (quoting Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1217 (Fed. Cir. 1995)). When a word of approximation like “about” is used, the range must be reasonably certain based on the “technological facts of the particular case.” Ortho-McNeil, 476 F.3d at 1326.

    But patentees still must be able to point to such facts to prove infringement when the accused limitation is covered only by the broadening use of “about.” The challenges of doing so were well demonstrated recently in Vertex Pharms. Inc. v. Lupin Ltd., No. 22-cv-966-SB (D. Del. Aug. 24, 2026). In that case Judge Bibas (of the Third Circuit, sitting by designation) issued a short 10 page-written opinion finding Vertex’s claims to both “80%” and “about 80%” ivacaftor not infringed by Lupin’s drug, which consisted of 74.26% ivacaftor, either literally or under the doctrine of equivalents.

    As a quick background, Vertex was hampered during prosecution by its own previous disclosure that an effective ivacaftor drug would comprise “about 10% by weight to about 80% by weight” ivacaftor. Vertex was able to obtain claims to “80% by weight” and “about 80% by weight” ivacaftor (along with specific excipients) based on arguing surprising results that occurred at 80% by weight. A sample claim is below:

    1. A pharmaceutical composition comprising a single solid dispersion, wherein the solid dispersion comprises:
    i. about 80 wt % of amorphous or substantially amorphous [ivacaftor] by weight of the solid dispersion[];
    ii. about 19.5 wt % of [HPMCAS] by weight of the solid dispersion; and
    iii. about 0.5 wt % of [SLS] by weight of the solid dispersion.

    However, the examiner rejected an attempt to claim “about 72 wt% to about 88 wt%” ivacaftor in a related patent, concluding that such a range was obvious and that there was no proof the unexpected solubility result would hold across such a range.

    During claim construction (which was handled by Judge Andrews), the court found no construction necessary, rejecting Lupin’s argument that “about” should be limited to rounding, i.e. “79.5-80.4” ivacaftor, 19.45-19.54% HPMCAS and 0.45-0.54% SLS, or else the claims would be indefinite. Somewhat strangely, Judge Andrews stated “Defendants provide largely conclusory statements regarding indefiniteness” despite himself noting that “[n]either party here could point to a place in the specification that clearly defined ‘about.’” Nor did Judge Andrews provide any guidance on how a POSA should assess “about,” leaving it entirely to the experts at trial.

    Judge Bibas took over the case and conducted trial. Vertex conceded no literal infringement of the claims to exactly 80/19/0.5%, but the decision on literal infringement for the “about” claims turned entirely on how to apply “about” only for the 80% ivacaftor limitation. “Unfortunately for Vertex,” Judge Bibas found its expert “unauthoritative” because he was “less careful and less scrupulous and more willing to sign off on extreme statements.” Specifically, Judge Bibas was unimpressed with Vertex’s expert’s argument that “about 80% ivacaftor” meant “high drug load” and covered “at least 70 to about 90 percent as a range.” Judge Bibas noted the expert “gave no reason to think that is true” as “high drug load appears nowhere in the patents or their prosecution.”

    In contrast, Judge Bibas found Lupin’s expert “completely credible” as “[s]he was meticulous” and “was admirably careful about what she could or couldn’t say.” Judge Bibas specifically found her opinion that “about” meant “nearly the same as” and limited to less than couple percentage points, noting that Vertex’s patent prosecution “specified the weight of ivactafor down to a tenth of a percentage point.”

    Judge Bibas also found such testimony consistent with the examiner’s views, stated that “about 80%” included 79.9% and “about 0.5%” included 0.55% which “comport[ed] with reading ‘about’ even more narrowly than [Lupin’s] view-and a far cry from [Vertex]’s ‘high drug load’ approach.” Thus, Judge Bibas found no literal infringement.

    Judge Bibas then considered the doctrine of equivalents for both the “80%” and “about 80%” claims together. To begin, Judge Bibas noted that the doctrine of eqvuialents is “an ‘exceptional’ path to proving patent infringement” and “applies only when ‘the nature of language makes it impossible to capture the essence’ and ‘every nuance’ of the patented invention.” Judge Bibas noted further, “many [claim] limitations warrant little, if any, range of equivalents” because of “the inherent narrowness of the claim language” and “prosecution history estoppel.”

    Judge Bibas first held that Vertex’s arguments “would have erased the numerical limits in Vertex’s patents” and Vertex’s use of “claims down to half a percentage point” “belies Vertex’s efforts to assimilate 74.257% to about 80%.” As I’ll discuss below, Judge Bibas’ focus on “half a percentage” point strangely appears to relate to the HPMCAS and SLS limitations, not the ivacaftor limitation, despite Judge Bibas’ myopic focus on the ivacaftor limitation.

    Judge Bibas also considered the prosecution history and that “Vertex emphasized how surprising the result at 80% was, and the examiner viewed about 80% as introducing variance of a tenth of one percent.” Though Judge Bibas appeared to view this as precluding the application of the doctrine based on prosecution history estoppel, he did not do any sort of estoppel analysis or identify any amendment or argument where the patentee surrendered scope. Interestingly, while noting the failure to obtain claims to “about 72 wt % to about 88 wt %” in a different patent in the background, Judge Bibas did not discuss this prosecution history when addressing DOE.

    Judge Bibas continued that even if the doctrine of equivalents could be applied, Vertex could not prove infringement. He first held that the “function, way, result” test “does not fit generic drugs” as “virtually all generic drugs” work the same way as a patented one, citing Voda v. Cordis Corp., 536 F.3d 1311, 1326 (Fed. Cir. 2008) (which does not relate to generic drugs, so I am a bit unclear where he drew this rule from).

    Judge Bibas then held the insubstantial differences test did not help Vertex, because “the intrinsic evidence does not suggest that a person of ordinary skill would view 74.257% and 80% ivacaftor as equivalent drug loads” in light of the “patent examiner read[ing] ‘about 80%’ to introduce a variance of roughly a tenth of a percentage point.” It was not entirely clear what Judge Bibas felt was “substantially different” about 74 versus 80%. Judge Bibas then noted that Lupin utilized a different process (wet versus dry granulation) and was unpersuaded by an argument that “the similar (though not identical) dissolution profiles mean that any difference is insubstantial” because “this is just another way of saying the drugs are bioequivalent.”

    I found this case fascinating, and perhaps an example of a court reaching the right outcome for the wrong reasons.

    First, I think the terms here should have been held indefinite. As I explained above, Judge Andrews’ claim construction order seemed to reflect that there was no guidance on how to be reasonably certain of the scope of the claims. Similarly, Judge Bibas did not credit either side as presenting “reasonably certain” scope of “about 80%” based on “the technical facts of the case.” And from my admittedly cursory review of the record, I’m not sure one could have found one. But it appears Lupin did not continue its indefiniteness arguments to trial. It is possible Lupin made a strategic decision that it would be better to prevail on non-infringement than invalidity, opening up the market for other generic competition.

    Second, I find Judge Sibas’ emphasis on the examiner’s brief mention of a single example involving “about” to be a bit strange, as the judge repeatedly noted that the examiner “thought that ‘about 80%’ included 79.9% and ‘about 0.5%’ included 0.55%.” That the examiner gave such an example as falling within the scope of the claims does not to me clearly and objectively set an outer bound, especially such a limited range (which even Lupin did not advocate for). But Judge Sibas twice described the examiner as “read[ing] ‘about 80%’ to introduce a variance of roughly a tenth of a percentage point.” While not saying it outright, Judge Sibas seems to have adopted a claim construction of “about 80%” as 79.9-80.1%, even narrower than Lupin’s proposed construction, which Judge Andrews rejected, and then additionally applying that as a limitation on the equivalence range of 80%. But the entire purpose of “about” is to avoid strict numerical boundaries, and instead to focus on the “technological facts of the particular case,” which Judge Sibas does not seem to do. But again, the challenge is in trying to find reasonable certainty on a record that provides none.

    Third, I find Judge Sibas’ analysis of the prosecution history to similarly be a bit strange. There was no dispute that the prior art taught using 10-80% ivacaftor, or that the claims to “80%” and “about 80%” were allowed after pointing to unexpected success in using an 80% formulation based on allegedly surprising solubility. What Judge Sibas’ discussion does not mention, however, is that the argument was that the prior art, while broadly disclosing 10-80% could be used, provided only an example with 50%. As Lupin argued, the “amount of API in the Lupin solid dispersion is much closer to the claimed 80% drug load than the 50% drug load in the prior art.” Judge Sibas did not seem to seriously grasp with this argument. The more problematic part of the prosecution history (strangely given little discussion) was Vertex’s inability to obtain claims to “about 72% to about 88%.”, discussed more below).

    Ultimately, however, I do think Judge Sibas’ decision is correct. Whether applying the “about” broadener to the limitations or applying the doctrine of equivalents (which largely become the same analysis), it seems clear the claims could not cover Lupin’s product.

    The claims require either exactly or about 80% ivacaftor, 19.5% HPMCAS and 0.5% SLS. The parties seemed to agree that Lupin’s generic product was approximately 74.26% ivacaftor, 22.28% HPMCAS, and 3.47% SLS. While I do tend to agree it is a bit difficult to read 74% as “about 80%,” I find some force to Vertex’s argument that a POSA would read “about” to include “the batch-to-batch variability inherent in pharmaceutical manufacturing” and encompass 10% variance (and 80% is not very precise, having only one significant figure), which might explain the ivacaftor limitation. More compelling to me is an argument Vertex made that Lupin’s ANDA would permit an 80% ivacaftor product because it is within 10% of the specified 74.26%, i.e., even under Lupin’s construction its ANDA still literally infringes (and infringes the exact 80% limitation).

    However, Vertex still seems to have a problem in how it can succeed in proving that 3.46% SLS is “about 0.5%” or equivalent to 0.5% no matter how you slice it. I cannot find anywhere where Vertex ever explains how 0.5% SLS could possibly be “about 3.5%” which is well outside of the batch-to-batch variability one would expect. While Vertex attempted to argue the precise weight percentages (especially for SLS for example) were not important, such a broad reading of the claims clearly contrasts with the specificity of the actual limitations, especially the SLS limitation, and Vertex’s argument would clearly seem to vitiate at least the 0.5%/about 0.5% SLS limitation. Thus, I think Judge Bibas reached the correct result.

    What would have been a more interesting question to me is whether a broader claim to 80% ivacaftor or about 80% ivacaftor without specific amounts of HPMCAS and SLS would have read on a 74.26% ivacaftor product (putting aside the argument that the ANDA permits 80% ivacaftor). While clear Judge Bibas believes the answer is no, that to me would have been a much closer call.

    However, Vertex would have had a significant problem in that it attempted to claim ranges in a related patent of about 72-88%, and failed to convince an examiner to allow such claims. And though it didn’t make it much into the opinion, I think the ultimate problem Vertex had is that it was asking for “about 80%” to be viewed as broader than the rejected 72-88% and cover 70-90%, and did not have a compelling argument for why it should be granted claim scope the patent office rejected.

    This is where the arguments get a little confusing, and I think the parties and court somewhat conflated determining the literal scope of the “about” limitation and the equivalence of the non-about claims. In some ways these are the same inquiry, and the patentee cannot apply equivalence to the about claims. See Cohesive Techs., Inc. v. Waters Corp., 543 F.3d 1351, 1372 (Fed. Cir. 2008) (“a patentee has brought what would otherwise be equivalents of a limitation into the literal scope of the claim, the doctrine of equivalents is unavailable to further broaden the scope of the claim.”).

    So there were really two arguably very similar questions: (1) is 74.26% literally “about 80%” and (2) is 74.26% equivalent to 80% (and could Vertex resort to DOE). To properly conduct the analysis requires the use of the oft conflated but analogous doctrines of prosecution history disclaimer and prosecution history estoppel. As a reminder, prosecution history disclaimer is a claim construction issue, and means that the literal scope of the claim is narrowed. Prosecution history estoppel prevents the patentee from recapturing what was disclaimed or surrendered through the doctrine of equivalents.

    Because of the fuzziness of “about,” applying these doctrines to this fact scenario gets quite hairy. For example, if an original claim recited exactly 72-88% and was rejected, and the patentee amended the claim to recite “80%,” it would be clear there was a narrowing amendment and both prosecution history disclaimer and estoppel would apply, with the claim limited to literally 80% and the patentee barred from trying to recapture alleged equivalents. However, when the original claims are “about 72 to about 88%” and amended to “about 80” it is not clear what scope was surrendered and thus what literal scope remains. For example, a patentee could argue that the patentee merely surrendered the literal scope of “about” below 72 and above 88, but still retained 72-88%. Or, a significant digits nerd like myself could take it even further and argue “about 80” is actually broader than “about 72 to about 88” because 80 has fewer significant digits and therefore provides less precision.

    While Vertex didn’t make this specific significant figures argument as best I can tell, it did argue that “about 80” was actually broader than 72-88, covering at least 70-90%. At bottom, however, I think Vertex had no way to explain why it should be given scope either literally (with the about claims) or under DOE (for the exact claims) that the patent office rejected. Digging through the briefing, it appears Vertex attempted to argue the amendment was only “tangential” the alleged equivalent because the concern was directed particularly “at the upper bound of approximately 88%.” But if that were the case, why didn’t Vertex amend the claims to recite “about 72 to about 80%”?

    It is unclear to me why Vertex appears to have argued for a broader range of 70-90% (both for literal scope of the about claims and for equivalence), rather than arguing that while 72% and below or 88% and above might have been surrendered and outside the scope of the claims, 74.3% was still literally covered by “about 80%” and was equivalent to 80%. Vertex made this point, but only when arguing for the tangential exception to PHE, stating “Critically, no specific intermediate value between about 72% and about 80%, and certainly not a composition at 74.3% amorphous ivacaftor, was ever identified, discussed, tested, or negotiated away during prosecution.”

    While I do not think it changes the outcome, I think making this the key point of the argument (rather than buried in the tangential exception) was the only potential path to victory for Vertex. Having a literal range for “about 80%” (and an equivalence range for 80%) that was narrower than “about 72% to about 88%” would have been far more credible argument than asking the Judge to give a broader range than the one rejected by the patent office. The problem I think Vertex had is how can they provide a range that both avoids the 72-88% problem and covers 74.26% that is reasonably certain to a POSA based on the intrinsic record. And I think that’s where I end up full circle that “about” claims are inherently problematic, especially when used to avoid prior art, as it becomes increasingly difficult to justify a reasonably certain, broad claim scope. Indeed, if Vertex’s claims covered e.g., 70-90% ivacaftor, 18-25% HPMCAS, and 0.1-5% SLS, why not seek such claims?

    So the clear takeaway to me is that while “about” can be used to attempt to broaden scope, it is no substitute for actually claiming the broader range and subjecting that broader range to prosecution.

  • Revisiting Teva v. Lilly

    I shared my thoughts on the Teva v. Lilly decision earlier this year, and it appears I continue to be in the minority of those who view this decision as a fact-intensive application of existing law giving deference to a jury verdict, rather than a sea-change in the law creating a separate standard for method of treatment claims.

    However, I can’t blame anyone who takes the latter position, as numerous statements in the panel opinion can very reasonable be read in such a way.

    Lilly petitioned for rehearing en banc. Unsurprisingly, Lilly framed the issue entirely as the panel setting out a distinct legal standard for method of treatment claims. In a bold and cleverly worded statement, Lilly framed the question as this: “Whether adding functional method-of-use limitations to otherwise non-enabled genus claims renders the claims enabled, as the panel held.”1

    The answer to Lilly’s question plainly must be “no.” While I’m sure prosecutors are already ensuring they are writing method of use genus claims, I don’t think anyone could reasonable defend that distinction as an end-around the holding of Amgen v. Sanofi and the established enablement and written description requirements.  

    The problem I think Lilly will have is that there was no finding here that the genus of antibodies was not enabled or not adequately described—those claims were eliminated in IPR on obviousness grounds. Thus, Lilly is presenting something of a strawman argument.

    While admittedly much of the language of the panel opinion focuses on the method of use aspect of the claims and certain sentences out of context seem to suggest that was the sole basis of its holding, the opinion also very clearly lays out why a reasonable jury could have found the genus of antibodies as “well-known” based on Lilly’s own IPR arguments. That language certainly seems to suggest that the panel did not conclude the genus of antibodies itself suffered from a lack of written description or enablement.

    Indeed, nearly every page of the opinion described how the genus was “well-known.” See Panel Opinion at e.g., 4 (“Lilly maintained that, by November 2006, anti-CGRP antagonist antibodies ‘were well known in the art’—indeed, that the prior art was ‘replete with exemplary disclosures of anti-CGRP antagonist antibodies.’”), 9 (“circumstances like those here—where a claim pertains to a well-known genus”); 12-13 (“A reasonable jury could have found that anti-CGRP antagonist antibodies themselves and methods of making them were well known, replete, or extensively described in the prior art—based on Lilly’s own statements that they were ‘well known,’ ‘replete,’ or ‘extensively described’ in the prior art.”); 13 (“against a backdrop of antiCGRP antagonist antibodies (and methods of making them) being well known”); 14 (“As already discussed, a reasonable jury could have found that, by the priority date, (1) anti-CGRP antagonist antibodies and methods of making them were well known….”); 17 (“That is not the case here, where anti-CGRP antagonist antibodies were well known in the prior art and disclosed in the specification”); 17 (“a well-known genus”); 19 (“In both cases, the relevant point remains: a reasonable jury could have found that a skilled artisan reading the specification would have understood that anti-CGRP antagonist antibodies were well known (and disclosed)….”); 20 (“AbbVie did not involve the circumstances relevant here and in that precedent—i.e., a well-known genus…”); 23 (“In light of the well-known status of anti-CGRP antagonist antibodies… Given that anti-CGRP antagonist antibodies (and methods of making them) were already well known….”).

    Lilly’s petition argued that the panel “invoked one perceived difference” from the Supreme Court’s recent decision in Amgen v. Sanofi: the claims reciting a method of treatment. While I would agree the panel’s first sentence distinguishing Amgen supports such a statement:

    The asserted claims here are unlike the claims in Amgen and Baxalta, however, because they do not claim humanized anti-CGRP antagonist antibodies themselves; instead, they claim only the use of such antibodies for the different, limited purpose of treating headache. 

    the panel’s opinion continues:

    In light of the well-known status of anti-CGRP antagonist antibodies and the routine nature of humanization, the more relevant “research assignment” in this case would have been determining which humanized anti-CGRP antagonist antibodies treat headache.  See Amgen, 987 F.3d at 1084 (observing that the specification’s teachings must be “at least commensurate with the scope of the claims”).  That assignment was completed; the specification disclosed that all such antibodies work for that purpose.

    Thus, the panel seemed to find a second key distinction as the Amgen case did not involve any factual finding that the genus was already well-known. Why the panel did not lead with this distinction is certainly puzzling, as it is the far more important distinction.

    Lilly’s en banc petition also downplays another important point and distinction from Amgen: a jury heard the evidence and found Lilly had not proven lack of written description or enablement by clear and convincing evidence. Indeed, the entirety of Lilly’s acknowledgment of this fact is a single statement “Following a jury verdict for Teva, Appx4554-62, the district court granted JMOL for Lilly on enablement and written description.” Lilly’s petition never grapples with the standard it needed to meet to obtain JMOL, that “a reasonable jury could not have returned the verdict.” The petition does not even use the term reasonable, let alone explain how a reasonable jury could not credit Lilly’s own arguments regarding the well-known status of anti-CRGP agonists and conclude that there was therefore not clear and convincing evidence of a lack of enablement and a lack of written description.

    As a side note, I have always found the invocation of JMOL following a jury verdict to be interesting, as by definition it requires a conclusion that the seated jury was “not reasonable.” It also seems that the Seventh Amendment guarantees a right to a jury trial, not necessarily a reasonable jury…. But that’s a topic for another time.

    In sum, while I think Lilly makes excellent points about how one plausible reading of the Federal Circuit’s opinion would lead to an absurd result, I am not sure it shows that the ultimate conclusion of the panel was erroneous. Indeed, it is precisely because Lilly’s proposed reading of the opinion is so absurd that I think objectively the opinion must be read to be limited to the facts that the genus was described as “well-known” being sufficient for a reasonable jury to conclude there was not clear and convincing evidence of invalidity as the key holding.

    It will be interesting to see if the Federal Circuit decides to clarify the opinion, as some amici have invited it to do. For example, in a joint brief filed by Amgen and Sanofi (the adverse litigants from the Supreme Court), the parties acknowledged “the Panel indicated that Lilly’s admissions could dictate the outcome of this case without distinguishing at all between composition of matter and method of treatment claims” and argued “the Court should limit its decision to these unique circumstances.”

    I agree such clarification would be helpful. We’ll see if the Federal Circuit takes the invitation. On Monday it invited Teva to respond, and I expect we’ll see a very strong brief emphasizing the “well-known” aspect of the genus, which was already laid out in Teva’s opening brief to the court, which included sections such titled “Precedents concerning novel classes of antibodies or methods of using novel small-molecule drugs are not controlling with respect to novel methods of treatment using known classes of antibodies” and “The district court erred by treating this case as if it involves claims to a novel class of antibodies.”

    1 Lilly also raised a second question regarding whether a specification’s disclosure of species outside the scope of a claimed genus constitutes a disclosure of representative species that can provide written-description support, but that is a topic for a different article.

  • Federal Circuit reiterates a bond is required for preliminary injunctions of generic launches

    In Otsuka America Pharm., Inc. v. Hetero Labs Ltd., the Federal Circuit today affirmed the District of Delaware’s decision to enjoin an at-risk generic launch, but reversed the court’s decision to waive the bond requirement under FRCP 65(c). This note focuses on the bond requirement, which I have some personal history with and have always felt is a somewhat underdeveloped area of the law.

    The Court’s opinion on the bond issue was relatively brief. That is unsurprising as the language of FRCP 65(c) is quite unambiguous: a court “may issue a preliminary injunction . . . only if the movant gives security in an amount that the court considers proper to pay the costs and damages sustained by any party found to have been wrongfully enjoined or restrained.”

    Despite that language, the district court honed in on the Third Circuit’s exceptions to that requirement, including when (1) “there is no risk of monetary loss to the defendant”; (2) “in non-commercial cases . . . upon considering the possible loss to the enjoined party together with the hardship that a bond requirement would impose on the applicant”; or (3) in “suits to enforce important federal rights or public interests, arising out of comprehensive federal health and welfare statutes.”

    The district court waived the bond requirement under the second of those exceptions, finding “the risk to financial harm to Hetero ‘speculative at best’ and expressed concern regarding ‘a chilling effect on access to justice’ if a multi-million-dollar bond were required in this case.” The Federal Circuit quickly found this inconsistent with Third Circuit law, as it had “never excused a district court from requiring a bond where an injunction prevents commercial, money-making activities” and “Hetero’s attempt to enter the market with its generic pharmaceutical product is clearly a commercial, money-making activity.” The panel also noted that “With regard to the district court’s concern about imposing a large expense for a bond on the plaintiffs,” Delaware’s local rules allow for such expenses to be taxed as costs.

    While not surprising, the Federal Circuit’s holding is an important reminder that a bond is required unless very narrow exceptions apply. And for good reason: the bond is typically the cap that an enjoined party may recover for a wrongful injunction absent exceptional circumstances. See, e.g., Par Pharm., Inc. v. TWI Pharms., Inc., No. CCB-11-2466, 2016 WL 5820211, at *1 (D. Md. Oct. 4, 2016) (“When a party is found to have been wrongfully enjoined, that party may recover proven damages, up to the full amount of the bond, that naturally and proximately resulted from the wrongful injunction.”).

    There is surprisingly little caselaw in the pharmaceutical context on the setting and recovery of injunction bonds. What is clear, however, is that courts have been loathe to apply any exception to the bond-cap rule. For example, in AstraZeneca v. Breath, the Federal Circuit rejected Apotex’s arguments that “it would be unjust to limit its recovery to the bond because it will have suffered significantly more damages” because “AstraZeneca agreed to post the original bond amount, understanding that—absent ‘rare exceptions’—its liability during that time period would be limited to that amount.” Id. Apotex did not even argue that one of the “rare exceptions” recognized by the Third Circuit applies—likely because they are very difficult to prove. See, e.g., Takeda Pharms., U.S.A., Inc. v. W.-Ward Pharm. Corp., No. CV 14-1268-RGA, 2018 WL 6529289, at *3 (D. Del. Nov. 12, 2018) (“Wrongfully enjoined parties may only recover in excess of the bond where the party seeking the injunction engaged in bad faith or fraud.”); see also Miche Bag, LLC v. Thirty One Gifts LLC, No. 2:10-CV-781-TS, 2011 WL 13803, at *2 (D. Utah Jan. 4, 2011) (“In the Tenth Circuit, the general rule is that absent proof of malice in obtaining injunctive relief, a party cannot be liable in damages resulting from a wrongfully or erroneously granted injunction beyond the limits or maximum amount of the bond or bonds.”).

    But this creates a bit of a conundrum: if the enjoined party is capped by the bond, then it understandably must request a large bond that cautiously exceeds the highest possible damages amount. Recognizing this, some courts have held that the bond should be set on the high side. For example, the 7th Circuit recognized “When setting the amount of security, district courts should err on the high side.” Mead Johnson & Co. v. Abbott Lab’ys, 201 F.3d 883, 888 (7th Cir. 2000). That is because the bond does not entitle the enjoined party to that amount, which still must be proven and thus “an error in setting the bond too high thus is not serious” while “an error in the other direction produces irreparable injury, because the damages for an erroneous preliminary injunction cannot exceed the amount of the bond.” Id. The 7th Circuit recognized this increases the fees of the bond, but noted they are typically small in comparison to the amount of money at issue. And as the Federal Circuit recognized, they may be recoverable.

    Finally, I was curious why there was preliminary injunction briefing at all so early in the case. It took a little digging through the record, but while not mentioned in the Federal Circuit opinion, Otsuka alleges that it never received Hetero’s Notice Letter because of defective service and thus was deprived of the ability to timely file suit and obtain the 30-month stay.

    It’s a reminder that the 30-month stay is in effect a free 30-month preliminary injunction with no protection of a bond under 65(c), and that absent the stay a party must post such bond. Which inevitably leads to the question: should brands have to compensate generics for lost sales during the 30-month stay? As far as I’m aware, the only available remedy for a “wrongful” 30-month stay lies in antitrust, which of course requires far more than simply prevailing in the underlying patent lawsuit. And while Attorney’s Fees should be theoretically possible, that too is a high bar which courts have been reluctant to award, and in any event would likely fail to fully compensate the wrongfully enjoined.

    So I’ll end with a question, should Congress create a remedy for wrongfully stayed generics to encourage them to fight through paragraph IV litigation and enter the market? What do you think?

  • Hikma v. Amarin: Opinion Analysis

    By Rocco Screnci

    Another patent case at the Supreme Court. Another unanimous opinion. Another narrow ruling that changes seemingly little about the current state of the law. See, e.g., Helsinn v. Teva; Amgen v. Sanofi. This time, however, a reversal, which could make it harder for brands to get past the pleading stage in skinny label cases.

    Today, the Supreme Court decided Hikma v. Amarin, a case about what it takes to allege induced infringement based on a generic maker’s marketing statements when generic otherwise “carves out” patented uses from its label.

    For a more detailed discussion about the case’s background, be sure to check out our oral-argument recap. To make a long story short: Amarin claimed that Hikma’s generic product infringed patents for methods of reducing cardiovascular risk (the “CV Indication”) listed in the Orange Book for Amarin’s Vascepa. Even though Hikma filed a section viii statement that sought to limit the generic’s approved uses to a non-patented indication, Amarin argued that certain marketing statements, when coupled with side effects listed on Hikma’s generic label, were enough to show that Hikma actively induced healthcare professionals to prescribe its generic product for the patented CV-indicated use. The district court rejected this argument and dismissed Amarin’s complaint, but the Federal Circuit reversed. And today, the Supreme Court reversed the Federal Circuit, holding that Amarin indeed failed to state a claim for induced infringement under 35 U.S.C. § 271(b). Rather than articulate any new test, the Supreme Court restated its longstanding precedent on “active inducement” and held that the Federal Circuit had strayed from that precedent by wrongly shifting focus from the whether the generic maker encouraged infringement to how doctors may interpret a generic maker’s statements. More than anything, however, it seems the Court simply felt that Amarin’s allegations established a mere possibility of active inducement, not the required plausibility of inducement. This is a line that is blurry at best.

    The unanimous decision, authored by Justice Jackson, proceeds in three parts.

    In Part I, the Court explains the legislative backdrop, regulatory scheme, and factual history that led us here. It describes the FDA’s role in approving new and generic drugs before they can go to market; the function of the Orange Book and how the section viii pathway that lets generic makers carve out patented uses to market generic drugs for unpatented uses; and how state and federal drug-substitution laws permit or sometimes “require” that prescribers and manufacturers provide “the cheaper generic version” of a drug to patients, meaning that generic manufacturers “surely know (and perhaps even expect) that their products will be put to infringing use.” Part I also explains why that knowledge (or expectation, as the case may be), is not enough to show induced infringement: Besides needing to show that Hikma knew that some third party directly infringed Amarin’s patents, Amarin needed to show that Hikma took “active steps to encourage direct infringement.” And finally, Part I summarizes the development of Amarin’s Vascepa and Hikma’s generic and the legal dispute that followed Hikma’s launch.

    Part II turns to the applicable legal standards. The Court began with procedure: Because this case reached the Court on a motion to dismiss, the Court reiterated the “well-established pleading standards” that applied to Amarin’s complaint. This will be familiar to anyone who has taken civil procedure. In short, a plaintiff must plead allegations that, if true, plausibly state a claim for relief. And to cross the line between “possible” and “plausible,” the plaintiff must plead facts that let “the court draw the reasonable inference that the defendant is liable for misconduct alleged.” Part and parcel of that requirement is ruling out the “obvious” and innocent alternatives. Next, the Court addressed substance: Because induced infringement under § 271(b) requires “active steps to encourage direct infringement,” the Court emphasized that it is not enough to allege that Hikma made statements that could lead a healthcare provider to prescribe the generic product for a patented use. Put differently, statements designed to cause providers to prescribe the generic for a patented use are different than statements that could cause providers to prescribe that patented use, and inducement requires the former, not the latter.

    Lastly, Part III applies the law to the facts and concludes that Amarin’s complaint did not cross the plausibility threshold.

    First, the Court found that several of Hikma’s statements “have an ‘obvious alternative explanation.’” For instance, Hikma’s allegedly inducing label statements were most obviously there because they were legally required, not because Hikma intended that prescribers would rely on them for infringing uses. So too with Hikma’s describing its product as “generic Vascepa.” This is normal industry practice and consistent with how the law defines a “generic version” of a drug. So the Court refused “to put generic manufacturers between a rock and a hard place by turning adherence to the law and industry standards into building blocks for illegal conduct.”

    Second, the Court reiterated that mere inaction is not grounds to infer active inducement. Hikma had no duty to further disclaim the un-approved indications for its generic in various press releases.

    And third, the Court reprised its discussion in Part II about the line between encouraging infringement and understanding it could happen. To this end, the Court characterized Amarin’s assertions as “implausibly roundabout ways to induce medical providers to infringe.” For example, a press release that described Hikma’s generic as a “hypertriglyceridemia” drug (rather than a drug for the unpatented indication for severe hypertriglyceridemia) was no different from characterizing a “leukemia drug” as a “cancer drug.” According to the Court, describing a drug using a “broad category” (e.g., “cancer drug” or drug for “hypertriglyceridemia”) is “not an instruction to prescribe the drug” for a particular use. Similarly, when Hikma issued a press release that discussed Vascepa’s aggregate sales (i.e., for patented and unpatented uses), the causal chain between those statements and direct infringement was far too weak to plausibly believe that Hikma made that statement to induce providers to prescribe the generic for patented uses. In essence, Amarin would somehow need to plausibly plead that Hikma intended that a healthcare provider would read a press release intended for investors, know enough about drug sales to understand that sales figures to encompass both patented and unpatented sales, and would become motivated to prescribe the drug for the patent uses simply because of those aggregate sales. Because this chain, on its face, is merely possible, not plausible, it failed to give rise to an active-inducement claim.

    As I mentioned in the Oral Argument Recap, I saw two likely outcomes for this case: Either Amarin would have a very narrow victory, or Hikma would win in a procedure-heavy opinion. I saw the former option as more likely because the justices seemed to lament taking the case to engage in mere “error correction.” But, as it turns out, the Court took the latter approach here. As I discussed, this outcome is not all too surprising, especially in view of some pointed questions about the pleading standard from Justices Kagan, Gorsuch, and the opinion author, Jackson.

    So where does this leave us?

    In terms of the big picture, this can be seen as a win, though potentially a narrow one, for generic makers. The decision confirms that label statements required by law cannot support induced infringement, and it requires that generic makers take more specific steps to promote infringing uses that characterizing its drug as a “generic,” “generic equivalent,” or “AB rated” version of a brand drug. But this case is about the pleading standard, so it is unlikely that it will pose insurmountable barriers for brands to state claims for induced infringement. In all likelihood, the response will be more detailed complaints and more creative theories of inducement, not fewer complaints.

    What the opinion did clarify, however, is that “that the key question is whether a defendant actively encouraged infringement through its statements, not merely how others may understand those statements.” This seems to give leeway to generics to rely on implication, though the Court was careful to clarify that active does not necessarily mean express.

    In any event, District courts will need to scrutinize a complaint’s allegations in deciding whether statements actively encourage infringement.  And courts now may be more willing to grant motions to dismiss.

    This case may have post-pleading effects, too. By refocusing the analysis on the defendant’s conduct, today’s decision may limit the need for expert testimony from physicians on how they would interpret the label or various post-marketing statements.  That said, because it is hard to decouple a party’s actions from that party’s intent, some courts may still find expert testimony from physicians probative on the issue of active inducement

    As for the parties, the path forward is less clear. During argument, Amarin’s lawyer suggested that the proper procedure would be for the Federal Circuit to remand to the district court where Amarin would seek leave to file an amended complaint to cure the pleading defects. Hikma’s lawyer, however, noted that Amarin conceded to a judgment of dismissal with prejudice, thus limiting the opportunity to re-open the pleading window. Whether the district court will nevertheless entertain a motion that seeks an opportunity to amend the complaint is uncertain.

  • A note on ANDA settlement agreements

    By Alex Menchaca

    Settlement agreements in ANDA litigation are typically subject to confidentiality obligations and not publicly disclosed. The terms of those settlement agreements can be of particular interest to competitor generic drug companies that are trying to get the best deal in a later settlement. Occasionally, one party to the agreement must file the agreement with the U.S. Securities and Exchange Commission. The obligation to file the agreement with the SEC arises when the SEC registrant enters into a “material definitive agreement.” (SEC Form 8-K, Instruction Item 1.01.) Those SEC filed agreements can provide useful guidance sometimes related to a particular formulation, but more generally to the kinds of terms one can expect to see in such ANDA settlement agreements. I was able to find a number of such agreements by doing a search at this link – https://www.sec.gov/edgar/search/.

    As with any contract, those ANDA settlement agreements will include the standard boilerplate terms and conditions. This note will not address those standard terms.

    More interesting are the terms addressing the kinds of rights the parties in ANDA litigation negotiate, such as entry date, pre-marketing activity, and most favored nations treatment.

    Entry Date – The entry date (or “Generic Entry Date” / “License Effective Date”) represents the point at which the generic manufacturer is permitted to commercially enter the market under a patent license. Rather than a single fixed date, the agreements typically adopt a multi-trigger structure – often a fixed future date, a favorable or adverse court decision, or a third-party launch. Such a multi-trigger structure is often desirable because it provides (i) the brand some continued period of exclusivity, (ii) the generic an entry date prior to patent expiry, and (ii) some level of comfort to the generic should some external development permit a different generic to launch earlier (e.g., litigation invalidation of the patents).

    Pre-Entry Market Activity – One purpose of the Hatch-Waxman Act is to enable the parties to work out the patent issues so that generic competition can start when all exclusivities expire. Similarly, pre-entry market activity enables the generic to undertake some marketing activities in advance of the entry date so that the entry date is actually meaningful (i.e., the generic has contracts in place and product on hand to enter the market). Such pre-entry market activity can include

    • Manufacturing and importing product prior to entry
    • Notifying customers of future availability
    • Making non-binding offers
    • Entering into binding contracts shortly before launch.

    The main goal of these terms is to enable the generics to be “launch-ready” to ensure immediate competition at entry, but it also enables the brand company to understand how and when its exclusivity in the market will be changing.

    Most Favored Nation Clauses – Most favored nation (MFN) provisions ensure that if the brand company later grants more favorable terms to another generic entrant, the settling generic may receive those same improved terms. For instance, an agreement may require the brand to notify the generic of any more favorable license granted to a third party and allows that generic to adopt those terms or that the agreement is automatically amended to incorporate more favorable entry dates or pre-commercialization rights granted to others. MFN clauses are of particular interest to first or early-settling generics to ensure that they are not disadvantaged relative to later settling generics. Such clauses also permit the brand companies to harmonize conditions across multiple generics so that they are all treated similarly.

    Other Terms – As noted above, ANDA settlement agreements include many other standard terms – e.g., license (typically non-exclusive and may be royalty-free or royalty bearing); at-risk launch provisions (typically triggered by other unauthorized generic launch); covenants not to sue; releases; patent no-challenge clauses. The terms discussed above, however, seem to be the terms that are given the most attention in settlement negotiations.

    If you would like copies of the ANDA settlement agreements that we collected from the SEC site, please let us know.

  • Hikma v. Amarin: Oral Argument Recap

    By: Rocco Screnci

    The Supreme Court heard oral argument yesterday in Hikma Pharmaceuticals USA Inc. v. Amarin Pharma, Inc. This is a case and an issue that we (and many others) in the pharma-patent world have been watching for a while. But for those who haven’t followed this case, here is a recap of what led us to the Supreme Court:

    • Amarin obtained a New Drug Application (NDA) for Vascepa, a drug that was originally indicated for lowering triglycerides in patients who have severe hypertriglyceridemia. (the “SH Indication”)
    • Seven years later, Amarin obtained approval for a new indication: To reduce risk of certain cardiovascular issues when used as an adjunct to maximally tolerated statin therapy in patients with elevated triglycerides. (the “CV Indication”)
    • In 2016, before Amarin received the CV Indication, Hikma filed an Abbreviated New Drug Application (ANDA) with a Paragraph IV certification challenging Amarin’s patents that covered the SH Indication.
    • Hikma prevailed on its Paragraph IV certification. But because Amarin obtained the CV Indication before Hikma prevailed on its Paragraph IV challenge, Hikma needed to address the CV Indication before the FDA could approve the ANDA.
    • Hikma thus filed a section viii statement, seeking approval only for the uses not covered by Amarin’s patents covering the CV Indication. This led to a so-called “skinny label” for Hikma’s generic product, meaning that the label would omit the CV Indication, and Hikma could only market its generic for the SH-indicated use.
    • Though the skinny label carved out the CV indication under the “Indications and Usage” section, the label still referenced potential side effects for people with cardiovascular disease. It also referenced the clinical trial that led to the CV Indication.
    • Upon receiving FDA approval of the ANDA, Hikma began marketing its generic product as a “generic version” of Vascepa, Vascepa’s “generic equivalent,” or “generic Vascepa.” Hikma also described its product as being a generic for “hypertriglyceridemia” and referenced Vascepa’s total sales (i.e., sales for both SH-indicated and CV-indicated uses).
    • Amarin filed suit, alleging that Hikma’s label and post-approval communications induced infringement of Amarin’s CV Indication patents for several reasons:
      • Press releases described the “generic version” of Vascepa; A press release in which Hikma referenced Vascepa’s total sales (i.e., sales for both SH-indicated and CV-indicated uses); Hikma’s removal of a statement from the label regarding the unknown effects of the active ingredient on cardiovascular mortality and morbidity; The label’s inclusion of potential side effects for people with cardiovascular disease and mention of the clinical trials related to the CV Indication; and
      • Website materials that described Hikma’s product as therapeutically equivalent to Vascepa for treatment of hypertriglyceridemia, which Amarin described as being broader than the permitted SH Indication.

    The district court dismissed Amarin’s complaint under Federal Rule of Civil Procedure 12(b)(6), noting that the allegations in the complaint did not rise to the level required to allege induced infringement under 35 U.S.C. § 271(b). To that end, the district court ruled that Amarin’s allegations fell short of the necessary “encourage[ing], recommend[ing], or promot[ing]” of the patented use (”taking Hikma’s generic for the reduction of CV risk.”).

    The Federal Circuit reversed on appeal. The focus of the analysis turned not on the allegations concerning the label, but on the allegations about the label plus the allegations of Hikma’s other actions. As the Federal Circuit noted, “this is not a section viii case in which the patent owner’s claims rest solely on allegations that the generic manufacturer’s proposed label is ‘not skinny enough,’ such that the label alone induces infringement.” Indeed, the Federal Circuit expressed that it would very likely agreed with Hikma if Amarin’s inducement claims were based solely on the label. Even so, the Federal Circuit held that, at the pleading stage, the additional allegations about Hikma’s press releases and website were enough to plausibly allege that Hikma actively induced infringement of the CV Indication patents.

    Hikma petitioned for certiorari, and the Supreme Court of the United States accepted to hear the case on two questions presented:

    (1) Whether, when a generic drug label fully carves out a patented use, allegations that the generic drugmaker calls its product a “generic version” and cites public information about the branded drug (e.g., sales) are enough to plead induced infringement of the patented use; and

    (2) whether a complaint states a claim for induced infringement of a patented method if it does not allege any instruction or other statement by the defendant that encourages, or even mentions, the patented use

    Argument this morning began, as it does in every Supreme Court case, with arguments from petitioner’s counsel. Hikma’s lawyer started his argument by framing the Federal Circuit’s decision as one that requires generic makers to monitor and actively discourage infringement, which is more than the law requires. Such a requirement, he urged, would disrupt the delicate balance struck by the Hatch-Waxman Act and effectively nullified the section viii pathway by subject generic makers to, at best, paying the hefty legal fees required to defend a lawsuit beyond the pleading stage.

    The justices’ questioning of Hikma’s counsel was relatively brief. The most interesting questions were asked by Justice Sotomayor and Justice Jackson. Justice Sotomayor emphasized her concern with announcing bright-line rules in the case because, at bottom, the case was resolved on a motion to dismiss, which presents a case-specific and ultimately fact-specific inquiry. As she noted during her questioning, the Supreme Court usually does not engage in error correction. Justice Jackson, for her part, sought clarification about Hikma’s labeling, asking whether FDA would even permit Hikma to add a disclaimer. This questioning seemed mostly rhetorical, which is a tactic justices may use to signal their position and potentially persuade their colleagues to join them.

    Following Hikma’s opening argument, an attorney from the Solicitor General’s office argued. The SG Office’s participation in oral argument is a somewhat common occurrence—albeit slightly rarer in IP cases—when the case involves important issues involving national policy. Deputy Solicitor General Malcolm Stewart, a career attorney in the SG’s office, argued on behalf of the United States in support of Hikma.

    A slightly livelier bench asked Deputy SG Stewart questions on topics ranging from the broad policy concerns to purely legal questions. On the former issue, Justices Jackson and Kavanaugh chimed in, both inquiring about the economic consequences for generic manufacturers if the Court set too low a bar for induced infringement in skinny-label cases and how that may disrupt the balance struck by Hatch-Waxman. On the latter, Justice Sotomayor reiterated her concerns about setting bright-line rules. This is a concern the Chief Justice seemingly shared, as he expressed skepticism about the government’s position and the adoption of a bright-line rule that made it too easy for generic companies to avoid induced infringement. Deputy SG Stewart countered this point by noting that the high burden of pleading inducement was “by design” and consistent with the traditional standards for induced infringement. Finally, Justice Alito and Justice Kagan used their razor-sharp styles of questioning to home in on the contours of what legal ruling was at stake in this case. Justice Kagan questioned whether and how the FDA’s determinations about the generic’s label should factor into the analysis. And Justice Alito, in typical fashion, deployed hypotheticals to find the limits of the government’s proposed legal test: What if Hikma not only described its product as “generic Vascepa,” but also mentioned that Vascepa is approved for CV risk? What if Hikma said that ”studies show” that the active ingredient reduces CV risk? Both would, as the government conceded, likely qualify as induced infringement.

    After Deputy SG Stewart finished his argument, Amarin’s counsel began to argue his case. He began by noting that the case was not a Hatch-Waxman or skinny-label case, but rather a run-of-the-mill induced infringement case. As he explained, the ultimate issue of inducement is fact intensive and thus not something a court should ordinarily resolve on a Rule 12(b)(6) motion to dismiss, where the Court must accept all well-pleaded allegations as true and draw all reasonable inferences in the non-moving party’s favor.

    After counsel for Amarin’s opening remarks, the Justices began asking questions. The bench was notably more interested in asking questions, with each justice (except for the Chief Justice and Justice Sotomayor) interjecting at least once. During this round of argument, two interesting threads emerged.

    First, an important legal issue raised in response to questions from Justices Kagan, Gorsuch, and Jackson involved the pleading standard. As Justice Kagan observed, press releases and promotional websites are typically done to influence investors, not prescribers. This creates a problem for Amarin because there would be a disconnect between Hikma’s intent and the unlawful activity. Put differently, Hikma’s intent in releasing the at-issue press releases was to influence investors to buy Hikma’s stock, not to persuade medical professions to prescribe Hikma’s product for off-label, patented uses. Or as Justice Jackson explained: Amarin’s framing of the case improperly focused on how prescribers interpreted the press releases rather than on what Hikma intended in making those press releases. Justice Gorsuch pointed out that this disconnect could pose problems because the pleading standard requires plausibility, not mere possibility. So Amarin would have needed to allege facts indicating that Hikma’s intent in those marketing materials was more likely unlawful (i.e., to promote infringement) than lawful (i.e., to influence investors). In response, Amarin’s counsel noted that Hikma did not challenge the adequacy of the allegations on the “intent” element, so it should not be a basis for reversing the Federal Circuit’s decision.

    Second, counsel for Amarin repeatedly emphasized that Amarin had not sued other generic companies. Those companies, unlike Hikma, had not marketed their products in a way that Amarin construed as promoting the patented CV-indicated use. This emphasis seemed to quell some concerns about the impact of the case on generic pharmaceuticals and the viability of the section viii pathway.

    So where does this leave things? For now, it is tough to say. My impression is that the Supreme Court is unlikely to write a sweeping opinion that drastically broadens or narrows what qualifies as induced infringement in cases involving skinny labels. I think Amarin can expect a narrow decision that states that, under the generous pleading standard, it has alleged enough facts to support a finding of induced infringement. That said, I would not be surprised if the Court issued a civil-procedure-heavy opinion that rules against Amarin and holds that Amarin’s allegations do not cross the line of plausibility, as Amarin’s allegations make it equally likely that Hikma’s intent was entirely lawful as it was unlawful.

  • Teva v. Lilly should not be read as creating a new 112 rule for method of use claims

    Last week the Federal Circuit decided Teva v. Lilly, restoring the jury’s verdict that Teva’s claims to methods of using CRGP antagonists to treat headache were not proven invalid under 112.

    The facts are at a high level, relatively straightforward. Teva’s patent family claimed CGRP antagonists as well as methods of using such antagonists to treat headaches. Lilly challenged those patents through IPRs, with the Board finding the product claims unpatentable based on Lilly’s arguments that the genus of antagonists was well-known and routine to achieve, but upholding the method of use patents finding no reasonable expectation of success in using the genus to treat headaches. The Federal Circuit affirmed both decisions. The parties went to trial, and Lilly argued that the method of use claims were invalid under 112 because they claimed the use of every possible humanized anti-CGRP antibody with no limitation on structure, while disclosing only a single working example within the scope of the claims. The jury found in favor of Teva, but the district court granted Lilly’s JMOL, concluding that the single representative species was not enough under the circumstances of claiming a broad genus. The Federal Circuit reversed, finding substantial evidence could support the jury verdict.

    The most controversial aspect of the opinion is that it can be read as drawing a bright-line distinction between claims to a product and claims to method of using the product for the purposes of 112 and applying Amgen. Indeed, the court wrote “Lilly takes issue with the distinction we have drawn between (1) claims to a method of using humanized anti-CGRP antagonist antibodies to treat headache and (2) claims to such antibodies themselves” rejecting Lilly’s argument that this is a “semantic distinction without a difference.” And the court later stated “we note that Lilly’s argument might be more persuasive if the asserted claims were to the genus of humanized anti-CGRP antagonist antibodies themselves. If that were so, this case would resemble Amgen….The asserted claims here are unlike the claims in Amgen[], however, because they do not claim humanized anti-CGRP antagonist antibodies themselves; instead, they claim only the use of such antibodies for the different, limited purpose of treating headache.”

    I certainly can’t disagree with anyone who reads this as the holding, and criticizes that holding as nonsensical and emphasizing semantic differences in claiming with unfair results. And I’ll certainly agree the Federal Circuit could have better addressed Lilly’s argument and explained why it was not creating a semantic difference. But I think given the opinion as a whole, that is now how the holding should be read. For example, when distinguishing the cases Lilly cited regarding semantic differences, the court said that neither addressed “a well-known genus used as part of a different invention.” Similarly, when distinguishing Amgen, the court emphasized “In light of the well-known status of anti-CGRP antagonist antibodies and the routine nature of humanization, the more relevant ‘research assignment’ in this case would have been determining which humanized anti-CGRP antagonist antibodies treat headache. [] That assignment was completed; the specification disclosed that all such antibodies work for that purpose.”

    In other words, while this point should have been made more clear, I read the holding as turning on the factual determination of the number of representative species and whether there was sufficient support for the genus, not whether it was a method claim. When the court emphasized that the invention was the method of treating headaches, not merely the antibodies themselves, I believe they were inartfully trying to explain why the claims could be novel and non-obvious while relying on the prior art to provide written description and enablement support. But the 112 holding itself was based on finding a sufficient disclosure and enablement of the full genus of antibodies and the belief that the district court too narrowly analyzed the disclosure of the specification in light of the prior art.

    Specifically, when analyzing the number of representative species, the district court recognized that “the jury could have found that (1) a POSA would have known methods for making murine (mouse) anti-CGRP antibodies, (2) a POSA could generate anti-CGRP antagonist antibodies, (3) a mouse immunized with CGRP would generate anti-CGRP antibodies, (4) antibodies that antagonize CGRP could be identified using tests that could analyze millions of potentially relevant antibodies in days, and (5) a POSA could confirm an antibody’s ability to antagonize CGRP in animals.” Teva Pharms. Int’l GmbH v. Eli Lilly & Co., No. 18-CV-12029-ADB, 2023 WL 6282898 (D. Mass. Sept. 26, 2023). However, the district court concluded that a “reasonable jury could not have found that the Patents-in-Suit disclosed more than one humanized anti-CGRP antagonist antibody within the scope of the Asserted Claims” and noted that the “jury heard uncontroverted evidence that no humanized anti-CGRP antagonist antibodies were known in the prior art.” Id. at *11.

    The Federal Circuit focused on different evidence stating a “reasonable jury could have found that anti-CGRP antagonist antibodies themselves and methods of making them were well known, replete, or extensively described in the prior art—based on Lilly’s own statements that they were ‘well known,’ ‘replete,’ or ‘extensively described’ in the prior art. A reasonable jury could have also found that humanization was a well-established and routine procedure by the priority date—again, based on Lilly’s own statements that it ‘was a well-established and routine procedure’ by the priority date.”

    The court elaborated “Although the specification disclosed just one humanized anti-CGRP antagonist antibody, it also disclosed several murine versions and prior-art methods of humanization—against a backdrop of antiCGRP antagonist antibodies (and methods of making them) being well known and humanization being routine.” In other words, the murine antibodies were still relevant and the jury could properly have considered them.

    Thus, I do not read the court as holding that the claims pass 112 muster merely because they are method of use claims, but pass muster because based on the evidence (including Lilly’s arguments in the PTAB) a reasonable jury could have found written description support for the full genus. Indeed, the court stated the “jury therefore could have reasonably found that the specification disclosed a representative number of species of humanized anti-CGRP antagonist antibodies for purposes of the claimed invention, thus rendering JMOL of no written description improper.” In other words, in my opinion the court would have reached the same conclusion on 112 if the claims recited the genus of antibodies themselves (those claims simply would have been unpatentable over the prior art).

    The court also stated that “critically, a skilled artisan would have understood from the specification that all humanized anti-CGRP antagonist antibodies treat headache.” It also noted elsewhere that Lilly did not dispute that “a reasonable jury could have found that a skilled artisan would have understood from the specification that all humanized anti-CGRP antagonist antibodies treat headache.” In other words, the court found there was support for the genus of antibodies, and support for their use in treating headaches. I think properly read, the holding shows that claiming a method of use requires an additional showing for 112, not an artificially lower burden.

    It appears to me that this was likely a very close call on whether the facts truly show full written description support given a single working example within the scope of the claims, but I would view Teva v. Lilly as being limited to its facts and not representing a broad shift in 112 law for method of use claims.

    It’s also a good reminder of the interplay between 112 and obviousness and how arguments for one can come back to bite you on the other. For the defense side specifically, keep in mind that obviousness requires only a single embodiment be obvious whereas 112 support must enable the full scope of the claims, so defendants can thread the needle and argue that one embodiment was obvious over the prior art while other embodiments in the scope of the claim are not enabled in light of the same prior art.

  • Should ANDA filers be using the PTAB to mount early challenges to OB patents?

    I wrote previously about the difficulty challengers to pharmaceutical patents face in proving standing at the Federal Circuit to appeal unfavorable decisions under the Federal Circuit’s view of what constitutes “concrete plans.” Thus, pharma companies may justifiably be hesitant to file petitions for PTAB review too early, lest they have no chance for appeal.

    However, the discretionary denial practice under Acting Director Coke Morgan Stewart has focused heavily on “settled expectations” and made it difficult for petitioners to raise challenges more than a few years after issuance, especially if the party waits until litigation has begun. And new Director John Squires has delegated discretionary denial authority to Ms. Stewart, thus I expect this practice to continue. Thus, potential challengers may have an incentive to file early challenges if they want to be able to take advantage of the PTAB.

    As others have written, “the PTAB’s role in adjudicating OB patents has been modest, both as an absolute matter and relative to its role for non-OB patents.” Rai, A.K. et al., Post-Grant Adjudication of Drug Patents: Agency and/or Court?, 37:139 BERKELEY TECH. L. J. 139, 166 (2022), available here. As the authors note, this likely is at least in part because the invalidation of a patent in the PTAB does not lift the automatic stay, therefore “the PTAB route is unlikely to be faster than the district court route, and may even be slower.” Id. at 167.

    However, this doesn’t account for the benefits ANDA filers could obtain by knocking out patents in advance of filing and the ensuring Hatch-Waxman litigation. The authors noted that biosimilar patents were much more often challenged prior to any litigation (53% v. <10%). Id.

    While the potential inability to appeal adverse decisions may dissuade some challengers, the Federal Circuit’s standing doctrine could create an interesting loophole strongly incentivizing early challenges. If a challenger is precluded from appealing the PTAB decision, there is a strong argument collateral estoppel does not apply due to traditional collateral estoppel rules. While Federal Circuit dicta suggests that it would not, the court has not fully answered the question. See AVX Corp. v. Presidio Components, Inc., 923 F.3d 1357, 1363 (Fed. Cir. 2019) (“this court has not decided…whether § 315(e) would have estoppel effect even where the IPR petitioner lacked Article III standing to appeal the Board’s decision to this court. For this court to so hold … we would also have to consider whether § 315(e) should be read to incorporate a traditional preclusion principle—that neither claim nor issue preclusion applies when appellate review of the decision with a potentially preclusive effect is unavailable.”).

    Early challengers may be able to have their cake and eat it too – challenge the patents in the PTAB while reserving the ability to raise those challenges again later in district court.

    Of course, as I’ve written about previously, challengers likewise cannot benefit from collateral estoppel, thus potentially mitigating the benefits of PTAB challenges in the first place, especially for large patent portfolios.

    It will be interesting to see if there are any Congressional efforts to increase the use of PTAB for resolving validity of OB patents, perhaps by giving collateral estoppel effects to related patents, allowing PTAB decision to lift the automatic stay, or creating other incentives for early challenges.

  • Switching to WordPress

    TypePad has shut down on us, so we’re moving the blog to WordPress. Thank you for your patience as we learn how to navigate this new blog!

  • Judge Hughes concurrence highlights post grant review appeal standing issue for pharma cases

    In Incyte Corp. v Sun Pharmaceutical Indus., Inc., the Federal Circuit found a PGR petitioner lacked standing to appeal because its “development plans amount to an expression of
    intent to create a product that runs a substantial risk of infringement if it is able to clear all development hurdles, secure FDA-approval, and bring its product to market” which was “too speculative to show concrete plans to develop a” product that “will be administered at the claimed dosage.” (emphasis in original). 

    The majority opinion is hardly surprising to those who have followed this issue. What is far more interesting to me is Judge Hughes’ concurrence. While agreeing the majority decision “is the result compelled by [the Federal Circuit’s] precedent,” Judge Hughes wrote separately because that he believes that “precedent on whether parties have standing to appeal to this court from an adverse administrative post-grant review is too restrictive and creates a special standing rule for patent cases.” Judge Hughes further noted that the “existence of this narrower special rule is even more pronounced in the pharmaceutical space.” I agree with both points.

    Judge Hughes then went through several cases, noting the focus on a lack of certainty of when or if a product would get approval and be marketed. But as Judge Hughes rightly points out, “[a]s a practical matter, pharmaceutical drugs generally have long development timelines, which inherently means there is more uncertainty about whether a drug will ever reach the market or infringe a given patent.” As he further noted, precluding appeal is especially harmful because a “party seeking to develop a drug that may infringe an existing patent has a significant interest in trying to invalidate that patent before making the large financial and time investments such development efforts demand.” In other words, challengers are stuck in a Catch 22 – challenge a patent before significant investment and run the risk of no appellate review, or wait until after investment when costs are already sunk (or design around the patent, which may be unnecessary).

    I agree with Judge Hughes’ policy concerns, and am interested to see if Incyte takes the invitation to seek en banc or Supreme Court review. What the concurrence is light on, however, is legal analysis of what is required for standing under Supreme Court jurisprudence. Judge Hughes cites a single case on standing, stating “Article III standing requirement is not meant to be a high barrier; the Supreme Court has characterized it as setting ‘the irreducible constitutional minimum.’” citing Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992) (emphasis added by Judge Hughes).

    Noticeably absent, for example, is any citation to MedImmune, Inc. v. Genentech, Inc., 549 U.S. 118 (2007), where the Supreme Court addressed standing in the analogous situation of declaratory judgments, albeit in the context of whether a licensee was required to terminate the license prior to bringing an action. As Justice Scalia noted, the Supreme Court cases on standing “do not draw the brightest of lines between those declaratory-judgment actions that satisfy the case-or-controversy requirement and those that do not.” Justice Scalia quoted an earlier 1941 Supreme Court case, where the Court “summarized as follows: ‘Basically, the question in each case is whether the facts alleged, under all the circumstances, show that there is a substantial controversy, between parties having adverse legal interests, of sufficient immediacy and reality to warrant the issuance of a declaratory judgment.’”

    I’m not sure if Incyte is the best vehicle for a challenge. For example, the majority noted that the petitioner’s “declarations show Incyte allocated a small amount of funds one month before filing this appeal for initial development of two topical drugs to treat alopecia areata: one with the active ingredient implicated by the claims at issue, deuterated ruxolitinib, and the other
    with non-deuterated ruxolitinib, which undisputably would not be covered by the claims at issue” and did not explain why it would likely infringe the dosage limitations other than by providing conclusory testimony. Thus, reasonable minds can certainly differ about whether there was “sufficient immediacy and reality.” However, the majority also made statements such as “The testimony shows that, when it filed this appeal, Incyte faced significant manufacturing, formulation, testing, and regulatory hurdles to bring either product to market” which Judge Hughes noted seems to require facts “that would only happen on the eve of FDA approval or commercial launch.” Requiring a party reach that stage to find “sufficient immediacy” would be a high bar indeed.

    As Judge Hughes noted in other cases standing was found lacking where a drug “had already been developed and completed Phase II clinical trials.” citing Allgenesis Biotherapeutics Inc. v. Cloudbreak Therapeutics, LLC, 85 F.4th 1377, 1380–81 (Fed. Cir. 2023). A case with such indisputably concrete investments might make a better vehicle.