Pharmaceutical patent

In the UK, as well as in many other countries, being the first to market with a generic or biosimilar to a leading branded medicine is a critical strategy that can offer a major advantage to a company in a highly competitive landscape. Being the first company to launch a generic product has several advantages. In particular the company can: (i) capture, and potentially maintain for a substantial period, a larger market share over the second or third generic to the market, and (ii) command a price premium before the inevitable generic price erosion commences. Given the comparatively reduced profit margins on which most generic companies operate, being first to market may represent a key factor for profitability. Therefore, it follows that the loss of this advantage may be substantial, and economically difficult to quantify. This first mover advantage provides a motive to generic companies to launch at risk, i.e., launch before they have cleared the way, or if the patentee has issued proceedings for patent infringement, before that claim has been decided. Importantly, in contrast to e.g. the US, a litigant which clears the way ahead of a launch in the UK obtains no period of exclusivity as against its competitors. The difficulty in quantifying the loss of first mover advantage was one reason why Michael Tappin KC, sitting as a Deputy Judge of the High Court, on 28 March 2025, refused AstraZeneca’s application for an interim injunction against Glenmark from launching generic dapagliflozin (dapag). However, AstraZeneca (AZ) appealed the decision (with the permission of the Court of Appeal) and, following a hearing on 9 April 2025, the Court of Appeal reversed the first instance judgment. The Court of Appeal subsequently handed down its written reasons on 16 April 2025. The ruling, delivered by Arnold LJ (with Warby and Coulson LJJ concurring), provides significant guidance to generic pharmaceutical companies considering launching at risk in the UK. Background Earlier in March 2025, Michael Tappin KC heard the trial on the merits between Glenmark, Teva and Generics (UK) on the one hand, and AZ on the other, concerning the validity of AZ’s patent to dapag. Following a heavily contested trial with seven days in Court taken up largely with extensive oral evidence from expert witnesses from both sides, the Judge reserved judgment in accordance with conventional practice. A few days before the trial, Glenmark informed AZ that it had obtained a marketing authorisation (MA) for its dapag medicine and was planning to launch it at risk. Consequently, on 6 March 2025, AZ issued an application notice requesting the grant of an interim injunction against Glenmark. Given that the trial was scheduled to be heard from 10 to 20 March 2025, the application was heard after the trial, on 27 March 2025, with the judgment being handed down the next day on 28 March 2025. As indicated above, Michael Tappin KC refused the interim injunction on the ground that any damages arising from the refusal of the interim injunction would be an adequate remedy for AZ because they could be calculated to a reasonably high degree of accuracy. By contrast, quantifying any damages payable under the cross-undertaking to Glenmark due to the loss of its first-mover advantage, and to the NHS (to which AZ agreed to give a cross-undertaking) would be significantly more difficult. Accordingly, damages would not be an adequate remedy for Glenmark or the NHS. AZ applied to Judge Tappin for permission to appeal but it was refused. AZ immediately sought, and obtained, permission to appeal from the Court of Appeal. Court of Appeal’s Reversal of the High Court’s Judgment Ahead of the appeal, AZ requested permission to introduce additional evidence which highlighted that two further generic companies (Teva and another unnamed entity, Generic X (although, as the Court observed “it is not hard to work out who it is likely to be”)) were poised to enter the UK market with their generic dapag medicines imminently. Both these generic companies had secured MAs and were ready to distribute substantial volumes. The Court of Appeal acknowledged that this new evidence changed the situation significantly – putting “a different complexion on matters”. It recognised the risk that if the injunction against Glenmark was not granted, at least these two companies might enter the market before the form of order hearing (FOO hearing) following the judgment in the main action which according to the Court rules should take place within 28 days of the date of handing down of the judgment – so potentially in May 2025. The Court found that, without an interim injunction against Glenmark, at least two other generics would enter the market prior to the FOO hearing. As such, the Court of Appeal held that what had been a “real risk” at the first instance had now become a “certainty”. It therefore concluded that such widespread generic entry would likely result in a rapid and irreversible price spiral (generic price erosion). The Court of Appeal agreed with AZ that if multiple generic entry happened as quickly as indicated by AZ’s new evidence, AZ would be likely to reduce their price prior to the FOO hearing. Given the NHS reimbursement mechanism and the VPAG scheme, AZ would have serious difficulties raising the price again. Nonetheless, this did not imply that damages would not be an adequate remedy for AZ. It would ultimately be contingent on the gap before the FOO hearing – the longer the gap, the more AZ would feel compelled to lower its price. The Court of Appeal also held that in considering the damage to AZ from a refusal of the interim injunction and Glenmark’s market entry before the FOO hearing, the High Court should have considered the damage to AZ not just in the period before the FOO hearing but also afterwards – given that the refusal would inevitably prejudice AZ’s ability to obtain an injunction against not only Glenmark but other generic entrants as well pending appeal (if AZ needed to do so). As a result, it was held that AZ might suffer generic price erosion for an extended duration, increasing the likelihood of needing to reduce its price. If AZ were to succeed in obtaining a final injunction following resolution of the merits, it would be far more difficult for them to raise the price again. Revisiting the American Cyanamid principles, particularly whether damages would be an adequate remedy for AZ, the Court of Appeal held that there was real doubt regarding the adequacy of damages for AZ. Whilst there was some uncertainty as to which side – AZ or Glenmark – was more likely to be inadequately compensated by damages, given that the period until the FOO hearing is rather short, the Court of Appeal held that it would be prudent to preserve the status quo and therefore injuncted Glenmark from entering the market. Clearing the Path The Court of Appeal criticised Glenmark’s strategy of launching during the proceedings without awaiting judgment and remarked that Glenmark had not cleared the path and instead were “jumping the gun”. Arnold LJ commented on the procedural inefficiencies caused by such tactics, which not only caused the parties to expend very considerable costs on the question of what is to happen during the period of about one to three months, but also, in his view, was not the best use of scarce court resources. Key Takeaways This judgment provides timely clarification on how the English courts will approach interim injunctions in the pharmaceutical sector. Like all interim injunction situations, the dapag saga depended on its facts. However some general principles can be taken away as follows: Preservation of the status quo remains a powerful factor, especially where market dynamics may shift rapidly and irreversibly before a final decision can be enforced. The first-mover advantage, while economically significant, does not automatically tip the balance of convenience if the generic has not cleared the path. The Court reiterated the importance of litigation conduct, with Glenmark’s strategy of attempting to launch before the judgment on the merits had been handed down being viewed critically. For patentees, the decision will be welcomed as reinforcement of their ability to seek interim relief even in the narrow window between trial and judgment. For generics, the message is clear: clear the way and don’t jump the gun. The litigation strategy must be aligned not only with regulatory readiness but also with procedural propriety. The ‘race to launch’ should not come at the expense of legal prudence. Also, by adjudicating AZ’s application for an interim injunction and the subsequent appeal in a relatively short period, the High Court and the Court of Appeal have demonstrated remarkable speed and flexibility. Postscript At 2 pm on Monday 28 April, the High Court handed down its decision on the merits following the trial finding AZ’s patent invalid on the basis that the patent did not disclose enough to make it plausible that dapag will have an in vivo effect on blood/plasma glucose (such that it could be used as an experimental tool) or will treat diabetes.   More from our authors: Vissers Annotated European Patent Convention 2024 Edition by Kaisa Suominen, Nina Ferara, Peter de Lange, Andrew Rudge € 105 Annotated PCT by Malte Köllner € 160 AI Governance and Liability in Europe: A Primer by Ceyhun Necati Pehlivan, Nikolaus Forgó, Peggy Valcke € 150

On 9 January 2025 I reported on Parts 1 and 2 of a three-part article in EPI Information by Tamaris Bucher, a Principal Patent Attorney at Novartis Pharma AG, on the current approach to antibody patents at the EPO. In Parts 1 and 2, Bucher argued that Part G.II.6.2 of the EPO Guidelines, which starts with a presumption that antibodies are prima facie non-inventive, creates an artificial barrier to obtaining patent protection for antibody inventions and sets a higher standard for the assessment of antibody claims that does not have clear basis in Art. 56 EPC or EPO case law. Bucher’s submission in Part 3 of the article is that the EPO Guidelines on inventive step for antibody patents are based on an unfounded prejudice that antibody generation work is routine and do not reflect the fact that antibody development and engineering for diagnostic or therapeutic purposes is a highly complex art requiring inventive skill. Scientists do not simply vaccinate an animal with the target and apply the product of the Köhler-Milstein hybridoma technique in a diagnostic or therapeutic context. Rather, the development of commercially relevant antibodies follows an unpredictable path and involves the combination of multiple techniques with the goal of obtaining an antibody with appropriate affinity, biological activity, selectivity, cross-reactivity, immunogenicity, stability, solubility, viscosity, aggregation, purity and biophysical criteria and the absence of undesirable post-translational modifications. Therapeutic and diagnostic antibodies are not simply products of nature, as the EPO appears to view them. Part 3 presents a proposed solution. The starting point, according to Bucher, is to bring the Guidelines on antibodies into consistency with the principles applied to small molecule compounds and to step back from the use of the term “antibody” in the formulation of the objective technical problem. Typically, the objective technical problem for antibody claims is formulated as (i) the provision of an antibody to the target X that exhibits the relevant improvement (where there is an improvement in a particular property) or (ii) the provision of an alternative antibody to the target or for use in the same purpose (where no improvement is identified). However, in Bucher’s view the use of the term “antibody” may have the inadvertent effect of causing EPO examiners to overlook the variability in antibody complementarity-determining regions (CDRs, the regions of antibodies that are responsible for binding to the target). Unlike small molecule compounds, for which structures are claimed as graphical representations, antibodies are represented by seemingly generic terms (“SEQ ID No. Z” or “SEQ ID No. ZZ”, which each represent a series of amino acids) so differences between 3D structures are less apparent. This makes it easier for examiners to overlook the different structures in the paratope (the part of the antibody involved in binding to the target) in claims defined by specific CDR or VH/VL sequences. The issue is further compounded by the discouragement of the assessment of structural information in Part G.II.6.2 of the Guidelines. Bucher suggests reformulating the objective technical problem as the provision of a specific alternative molecular structure (or sequence) for use in achieving the functional activity Y to reflect the complex and varying structures that can be represented by sequences in different SEQ IDs. For the final step of the problem-solution approach – in which it is considered whether or not the claimed solution, starting from the closest prior art and in light of the objective technical problem, would have been obvious to the skilled person – Bucher argues that the correct question should be whether the claimed antibody with its specific amino acid sequences would have been made by the skilled person by following the prior art and with a reasonable expectation of success of achieving the desired activity, not whether it could have been made using unlimited time and resources. The key question that arises is what the impact would be of such a change of approach in relation to antibody claims. Bucher argues that it would be limited: claims to antibodies defined by SEQ ID are relatively narrow in scope so the grant of additional antibody patents would not be problematic for the antibody field or detrimental to the public in general. Whether readers agree or not it is clear that this is an area ripe for consideration given the importance of biologics in modern healthcare. Bucher’s article series has already generated significant discussion amongst patent practitioners so there is clearly momentum for further debate – and potentially a change in approach – in this critical area. Ms Bucher’s article can be found at the following link: epi Information | The Barrier Around Antibody Inventions at the European Patent Office More from our authors: Vissers Annotated European Patent Convention 2024 Edition by Kaisa Suominen, Nina Ferara, Peter de Lange, Andrew Rudge € 105 Annotated PCT by Malte Köllner € 160 AI Governance and Liability in Europe: A Primer by Ceyhun Necati Pehlivan, Nikolaus Forgó, Peggy Valcke € 150

On 30 October 2024, the District Court of the Hague handed down two merits decisions on the widely litigated apixaban patent of Bristol-Myers Squibb (see here and here – Dutch language versions). EP 1 427 415 B1 (EP 415), which lapsed in 2022, and the corresponding supplementary protection certificates (SPCs) have been subject to litigation throughout Europe (see e.g. the previous posts on the UK, France and Spain). In 2023 the Dutch Court of Appeal already granted preliminary injunctions against Sandoz and Teva and Stada as it found that there was not a serious chance that the apixaban patent and SPC would be held invalid in merits proceedings. Both Sandoz and Teva appealed this decision to the Supreme Court. In these merits proceedings Sandoz and Teva had filed invalidity claims to pave the way for generic apixaban in the Netherlands. The Dutch court found the patent and SPC (no 300500) to be valid and granted an injunction against Sandoz. In its decisions the Dutch court addressed two pivotal aspects: (formal) priority of the patent in light of the G 1/22 and G 2/22 decisions of the Enlarged Board of Appeal (EBA); and inventive step in light of the EBA’s G 2/21 decision to determine whether BMS could rely on the purported technical effect. Priority Teva had argued that BMS Company, applicant of the patent, was not entitled to the priority right when it filed the application. The inventors had initially transferred the priority document to BMS Pharma, but the subsequent PCT application, from which EP 415 stems, named BMS Company as applicant. Only after the application was filed, the priority document was transferred from BMS Pharma to BMS Company. The court dismisses the argument. As was held by the EBA in G 1/22 and G 2/22 and now repeated by the court, there is a rebuttable presumption that the subsequent applicant is entitled to the priority right. The presumption can only be rebutted in exceptional cases. It is for the party challenging the entitlement to priority to prove that this entitlement is missing; merely raising speculative doubts is not sufficient. BMS Company is therefore presumed to have been entitled to the priority right, which presumption was not rebutted by Teva. According to the court, there can be no doubt that the intention within the BMS group was to file the application with a valid priority claim. Referencing the EBA in G 1/22, the court considers that the applicant should be protected from the situation like the one at hand, where their own intermediate prior art is published under the assumption of the applicant’s entitlement to the priority right. The court further considers that the validity of the priority claim is confirmed by the nunc pro tunc transfer (i.e. a later, confirmatory transfer) of the priority document from BMS Pharma to BMS Company, which transfer is deemed effective as of the moment of the PCT application. Inventive step and G2/21 To determine if the patent is inventive, the court applies the EPO’s problem-solution approach. Parties agree that apixaban is a so-called selection invention in which case an inventive step may be accepted only if the selection is connected to a particular technical effect and if no hints existed which lead the skilled person to the selection. Thus, the selection of apixaban from the compounds disclosed in the closest prior art should be coupled to a particular (new or improved) technical effect. Parties assume that this effect (in any case) is that of Factor Xa inhibition. Since the same effect is attained by the compounds disclosed in the prior art, the effect must be present to an unexpected degree. Consequently, the question that lies before the court is whether BMS can rely on the purported technical effect of improved Factor Xa inhibition. The court thereto recalls the standard set out by the EBA in G 2/21: the patentee may rely upon a technical effect for inventive step if the skilled person, having the common general knowledge in mind, and based on the application as originally filed, would derive said effect as being encompassed by the technical teaching and embodied by the same originally disclosed invention. If the G2/21 test is met, the patentee may still present evidence during the examination that the alleged technical effect actually occurs, i.e. post-filed evidence can be taken into account. An assessment of the patent application and the common general knowledge leads the court to conclude that the technical effect on which BMS wishes to rely for inventive step, namely improved Factor Xa inhibition, will be considered by the skilled person as encompassed by the technical teaching. According to the court, the skilled person would derive from the application that it not only aims to provide new compounds as Factor Xa inhibitors, but also improved Factor Xa inhibitors over those in the prior art. The specific technical means that the application provides to solve this problem comprise at least the synthesized compounds disclosed in the examples of the application (among which apixaban). Moreover, the skilled person will assume that these compounds have been tested but they will also be able to synthesize the compounds themselves and easily test the compounds for their efficacy as Factor Xa inhibitors. It is then not required for the application to contain test results or other evidence, nor is it required that the application explicitly mentions the technical effect that is relied on. What is relevant is that the effect does not change the nature of the claimed invention. According to the court, BMS’s post-filed evidence demonstrates that apixaban is a better inhibitor of Factor Xa than the compounds disclosed in the prior art. The court concludes that apixaban would not be obvious to the skilled person. Foreign rulings The court notes that its assessment of the inventive step aligns with that of the French, Norwegian and Swedish courts. In contrast, the English High Court of Justice and Court of Appeal came to a different finding, which the Dutch court says is the result of the application of the test formulated by the Supreme Court in Warner-Lambert v Generics [2018] UKSC 56. The Irish court applied a test similar to that of the English court, resulting in a different outcome there as well. The District Court follows the reasoning of the Dutch Court of Appeal in 2023 (see this earlier post here). The Court of Appeal at the time considered the Warner-Lambert test, applied by the English courts, to be developed specifically for sufficiency of disclosure and not for assessing inventive step. The District Court – like the Court of Appeal – emphasizes that the G2/21 threshold for inventive step is distinct from and easier to meet than the standard for sufficiency. Accordingly, both the Court of Appeal and the District Court could justify the differing outcomes. A clear Dutch approach thus appears to be emerging. More from our authors: Vissers Annotated European Patent Convention 2024 Edition by Kaisa Suominen, Nina Ferara, Peter de Lange, Andrew Rudge € 105 Annotated PCT by Malte Köllner € 160 AI Governance and Liability in Europe: A Primer by Ceyhun Necati Pehlivan, Nikolaus Forgó, Peggy Valcke € 150

This morning, the Barcelona Appeal Court has announced a judgment of 18 July 2024, reversing the judgment of 15 January 2024 from Commercial Court number 4 of Barcelona, which had found patent EP 1,427,415 (“EP ‘415”), protecting apixaban, to be invalid. The main highlights of the decision may be summarized as follows: The first interesting aspect is the finding in relation to the right to priority. As in other European countries, Teva had challenged BMS’s right to priority on the grounds that the right to priority had not been correctly assigned from the BMS entity that filed the priority patent application, to the BMS entity that filed the application that resulted in the granting of EP ‘415. In short, the Court has denied Teva’s locus standi to question the right to priority on this formalistic ground. In particular, it has found that it does not make sense for a third party to discuss the legal relationship between the first applicant and the second applicant when there is no conflict between them. So, without citing decisions G 1/22 and G 2/22, the Court has followed the same logic. The Court has added that, even if one were to accept a third party’s locus standi in the circumstances described, the nullity attack should likewise be rejected, as it would be based on an extremely formal and, therefore, abusive, argument based on the different legal personality of the two companies of the same group. Finally, the Court has noted that BMS filed two legal opinions from two U.S. law experts which, at least, cast doubts on the strength of Teva’s arguments on the merits (i.e. whether or not the assignment had taken place correctly under U.S. law). Taking into account that, according to Spanish procedural law, in case of doubt, the petition of the party having the burden of proof must be rejected, the Court concluded that Teva’s lack of right to priority attack should have been dismissed in any event. The second aspect addressed by the judgment is inventive step. As patent aficionados will no doubt be well aware, the main theme of Teva’s inventive step attack around Europe has been that the patent application, as filed, did not make it plausible that apixaban had the alleged technical effect (factor Xa inhibition, etc.). In particular, in Spain, Teva based its inventive step attack on T 488/16 (Dasatinib) which, as is well known, was a classic example of the “ab initio plausibility” test. The Barcelona Appeal Court has first considered whether, in contrast to the European Patent Office (“EPO”) Technical Boards of Appeal (“TBAs”), national courts must examine “plausibility” in the first place. In this regard, the Court has noted that, unlike the EPO’s administrative organs, the function of national courts is to review whether a granted patent falls within any of the nullity grounds enshrined in article 138 of the European Patent Convention (“EPC”). In particular, the Court has noted that when, as in the case at hand, during prosecution the patent application was limited to one compound only (i.e., apixaban), the only question to be answered is whether or not the claim as granted is affected by any of such nullity grounds. In this regard, the position followed by this judgment resembles the German position. The Court has then moved to consider a second reason for rejecting Teva’s inventive step attack. As mentioned earlier, in Spain, Teva mounted its lack of inventive step attack on T 488/16 (Dasatinib). After G 2/21 was published, Teva relied on par. 72 of G 2/21, where the Enlarged Board of Appeal tried to protect the EPO from the disarray brought by the three divergent lines of case law discussed in G 2/21 (“[…] the Enlarged Board is satisfied that the outcome in each particular case would not have been different from the actual finding of the respective board of appeal“), to allege that the “ab initio plausibility” test applied in T 488/16 continued to be good law (i.e. nothing had changed). The Barcelona Appeal Court has noted that, contrary to Teva’s position, one cannot extract from G 2/21 the teaching that “everything has changed so that everything remains the same.” The Court has added that, for this reason, it cannot embrace Teva’s argument in the sense that G 2/21 has not left behind the “ab initio plausibility” test applied in T 488/16. This has been relevant for reversing the first instance decision which, after discussing G 2/21, ultimately upheld the inventive step attack applying Dasatinib (i.e., the “ab initio plausibility” test). In short, the Barcelona Appeal Court has considered that the Enlarged Board of Appeal, in G 2/21, introduced a more nuanced test. According to par. 5.24 of the judgment, the patent owner may rely on a technical effect if the person skilled in the art, departing from the application as filed and the common general knowledge, first, may conclude that such technical effect derives from the original technical teaching and, second, represents an embodiment of the same, meaning by technical teaching the invention claimed in the application. Against this background, the Court has noted that, once the patent has been granted, in the context of litigation before a national Court, the burden of proving that the conditions laid down by G2/21 are not fulfilled lies with the third party questioning the validity of the patent granted. Building from here, the Court has noted that Teva’s case had been built on the premise that T 488/16 (Dasatinib) required the application as filed to include information that made the technical effect plausible. It has then added that the new test introduced by G 2/21 does not require this. To sum-up, the Court has concluded that the legal test on which Teva based its inventive step attack has been left behind by G 2/21. Finally, following the order of Teva’s complaint, the Court has dealt with sufficiency. In this context, it should be clarified that the arguments used by Teva to question sufficiency were exactly the same lack of “plausibility” arguments used to combat inventive step. Like in the case of inventive step, the Court started its analysis highlighting that, once the patent has been granted, a national Court must examine the sufficiency of the claim as granted. In particular, in par. 8.8 of the judgment, the Court has noted that, contrary to Teva’s assertion, one does not have to look for a needle in a haystack because during prosecution, to overcome the examiner’s objections, “[…] the applicant simply cleaned-up the haystack and kept the needle.” Also, in par. 8.15-8.16 of the judgment, the Court has walked a very fine line in drawing a very important distinction between the reflections developed by the Enlarged Board of Appeal in par. 73-77 of G 2/21, meant to apply to second medical use claims, and the claims of the case at hand, which are not second medical use claims but medical use claims contained in the patent that disclosed apixaban for the first time. In par. 8.15, the Court has noted that since, in the case of second medical use claims, the patentability can only be justified by having invented a new medical use of a compound that was already known, the test must be stricter than in the latter case. The Court has added that, in any event, even if one were to apply to the medical use claims of EP ‘415, the test applied to second medical use claims, Teva had not established that the person skilled in the art would not have considered the technical effect of apixaban credible. All in all, this important judgment has aligned Spanish case law on this lively area with the position on inventive step and sufficiency taken by all other Courts in Continental Europe in the same case. More from our authors: Vissers Annotated European Patent Convention 2024 Edition by Kaisa Suominen, Nina Ferara, Peter de Lange, Andrew Rudge € 105 The Unitary Patent and the Unified Patent Court, Second Edition by Pieter Callens & Sam Granata € 176 Artificial Intelligence and Patents: An International Perspective on Patenting AI-Related Inventions by Jonathan P. Osha € 174 Enforcement of Decisions of the Unified Patent Court: A Survey of Participating Member States by Christof Augenstein and Johanna Flythstrom € 123