Marleen van den Horst

On 23 January 2024, the Provisions Judge of the District Court of The Hague handed down a decision in the PI proceedings initiated by Janssen Biotech Inc. (“Janssen”) against Samsung Bioepis NL B.V. (“Samsung NL”). Janssen claimed that Samsung NL infringed its Stelara SPCs granted in Italy and Denmark. The Provisions Judge denied the PI as according to his preliminary assessment Samsung NL can benefit from the SPC Manufacturing Waiver. Janssen’s SPC- Stelara Janssen holds European patent EP 1 309 692 B1 (“EP 692”) for “ANTI-IL-12 Antibodies, compositions, methods and uses“. The medicinal product of Janssen is marketed under the brand name ‘Stelara’ and contains the active ingredient ustekinumab. EP 692 was granted on 13 May 2009 in, among others, Italy and Denmark. The patent expired on 7 August 2021. After expiration, Janssen was granted a SPC in Italy, Denmark and the United Kingdom (UK – ending on 19 January 2024). Similar Stelara patents in Canada and South Korea have also expired. However, Jansen has filed patent(s) (applications) in these countries that protect the treatment regimen approved for Stelara for ulcerative colitis disease. SPC Manufacturing Waiver Samsung has developed a biosimilar of ustekinumab with Stelara as reference called SB17. Samsung NL issued a notice to the Danish and Italian authorities respectively that it intended to manufacture and stock its biosimilar in Denmark and Italy for the purpose of exporting it to the UK, Canada and South Korea based on Regulation 2019/933 regarding the SPC Manufacturing Waiver “MW Regulation”) and to market the product in the European Union after expiration of the SPC (stockpile exemption). Samsung NL mentioned that it would submit the reference numbers of its market authorisations in said countries as soon as publicly available. Samsung NL additionally undertook towards Janssen that it would not manufacture its biosimilar products for the EU-market until 24 January 2024 in order to avoid discussion on the stockpile exemption in the PI proceedings. The PI Judge’s assessment of infringement According to Janssen Samsung NL infringes its SPCs in Denmark and Italy and cannot benefit from the SPC Manufacturing Waiver because Samsung NL (i) did not specify the reference numbers of the marketing authorisations in the exporting countries, (ii) Janssen’s patent rights are in force in the countries to which Samsung NL intends to export its biosimilar, and (iii) Samsung NL is not allowed under the SPC Manufacturing Waiver to store the biosimilar products. The PI Judge finds that according to the text of MW Regulation the manufacturer is obliged to provide the reference number of market authorisation, or the equivalent of such authorisation in each exporting third country, only as soon as it is publicly available. This entails, according to the preliminary view of the Provisions Judge, that if this number is not yet publicly available, the manufacturer has the possibility to supplement the notification with the reference number of the marketing authorisation as soon as it is publicly available, as reflected in Article 5(5)(e) MW Regulation. The Provisions Judge considers that the MW Regulation is sufficiently clear on this item. It is taken into account that the nature and the content of the provisions in question were extensively debated during the revision of the earlier SPC Regulation. The Provisions Judge is therefore of the opinion that it is not necessary to raise preliminary questions to the Court of Justice of the EU on the application of the SPC Manufacturing Waiver, as suggested by Janssen. The fact that the District Court of Munich in the matter Janssen vs Formycon ruled differently on this matter does alter his view, especially because the Provisions Judge finds the reasoning of the District Court of Munich in said decision not convincing. Janssen further argued that Samsung NL may not manufacture a biosimilar under the waiver because in the countries to which Samsung NL intendeds to export its product, patent rights are existing. The Provisions Judge considers that there is no requirement under the MW Regulation that the intended exporting countries are free of patent rights and/or that the manufacturer must prove this in advance. Such requirement would be contrary to the objective of the MW Regulation of ensuring a level playing field with global competition. If EU based manufacturers were only allowed to manufacture for export to countries that are ‘free of patents’ they would be put in a serious disadvantage compared to non-EU based competitors not subject to such restrictions. In relation to the storage of the products the Provisions Judge states that it follows from the wording of Article 5(a)(a)(ii) of the MW Regulation that under the export exemption, ‘temporary storage’ is allowed. The MW Regulation does not specify a maximum period for such ‘temporary storage’ other than that it must be strictly necessary for the actual export. Conclusion The Provisions Judge is of the preliminary opinion that – contrary to the Janssen vs Formycon decision – a manufacturer of a biosimilar can rely on the SPC Manufacturing Waiver a) even if the market authorisation reference number is not yet publicly available, b) there are possible patents rights in the countries to which the product is exported and c) temporary storage of the products is possible when strictly necessary for the actual export.

On 1 November 2023, the District Court of The Hague handed down a decision in the invalidity proceedings instituted by Sandoz B.V. (“Sandoz’’) against Bayer Intellectual Property GmbH. (“Bayer’’). The Court held the once-daily dosing regimen of rivaroxaban, claimed in EP 1 845 961 B1 (“EP 961”), inventive. The use of rivaroxaban and the anticipated launch of Sandoz’s generic product Rivaroxaban is used as an anticoagulant for the treatment of thromboembolic disorders (TEDs). Bayer markets its rivaroxaban product under the name Xeralto. Bayer was the owner of EP 1 261 606 B1 that protects the active ingredient rivaroxaban. Based on said patent, Bayer was granted SPC no. 300370 that will expire on 1 April 2024. Sandoz is planning to launch a generic product containing rivaroxaban in the Netherlands after expiration of the SPC. Sandoz started invalidity proceedings claiming that EP 961 is invalid due to lack of inventive step. Bayer instituted a counterclaim requesting: 1) an injunction against Sandoz for (likely) future infringement of the NL part of EP 961 and 2) a declaration that any offering or placing on the market of Sandoz’ product (e.g. listing the product in the ‘G-Standaard’), or importing or storing the product for those purposes, is an infringing act. What preceded EP 961 was granted on 22 April 2015. Opposition was filed by fifteen parties. In 2018 the Opposition Division revoked EP 961 for lack of inventive step. In appeal, the Technical Board of Appeal (“TBA’’) maintained the patent as granted. Since then, multiple national suits have been filed to invalidate EP 961 e.g. in Germany, Norway and in the United Kingdom. In its decision of 9 June 2023 the Court of Oslo concluded that the patent was inventive. Difference between the TBA and NL proceedings The main difference is that the TBA found EP 961 inventive based on Kubitza I & II and the Harder Abstract, whereas in the Dutch (and Norwegian) proceedings Sandoz introduced the ‘’Harder Poster’’ as new prior art, which was not introduced in the opposition. The Harder Poster reveals parameters in the pharmaco-dynamics that suggest suitability for a once-daily dosing. After summarising the reasoning of the TBA, the District Court of The Hague discusses the Norwegian decision in the parallel proceedings, where the Harder Poster has been taken into account in assessing inventive step. The District Court’s assessment of inventive step Following the parties, the District Court applies the PSA and takes the Harder Poster as the closest prior art. The difference between the Harder Poster and the claimed invention is that the latter concerns the medical use of rivaroxaban in the claimed dosage regimen. Harder is a phase I study on the pharmacodynamic effects of rivaroxaban on eight healthy subjects, but not a study on what the effects and therapeutic efficacy are in patients. The technical effect of the distinguishing feature is that rivaroxaban in the dosage regimen (once daily for at least five days with a rapid release dosage form) is effective and safe in TED patients. According to the Court the objective technical problem is: “finding an oral dosage regimen of rivaroxaban that is safe and effective against TEDs”. This is almost the same as defined by the TBA, albeit that the Court does not concur with the TBA’s (additional) objective of finding a ‘convenient’ dosage regimen as part of the problem to be solved. To assess inventive step, it must be assessed whether, based on the technical problem, the invention was obvious for the average skilled person on the priority date. According to the Court an inventive step is lacking if the person skilled in the art, based on the relevant state of the art, would have solved the problem as claimed in the patent. As the invention is the result of further research, the invention is not only obvious if the skilled person had carried out that research and the results are clearly predictable, but also if there is an incentive in the state of the art that there is a reasonable expectation of success. This means that the skilled person is able to predict a reasonable successful outcome of the research project within an acceptable time period. The hope to succeed is insufficient. The Court points out that the Harder Poster does not provide a reasonable prediction that rivaroxaban would be safe and effective in patients in a once-daily dose. It does not provide convincing results to support this, as the study was conducted in only eight healthy subjects, lacks data on half-life time, whereas the type of tests used were of an experimental nature. Moreover, the skilled person would not have a reasonable expectation of success, because of numerous negative pointers, like 1) the high (and life-threatening) risk of over- and under-dosing patients in anticoagulant research would make him cautious to extrapolate results in phase I studies to a phase II study, 2) Harder does not provide half-time data, which is an important indicator for determining dosing frequency, 3) the negative research experience with a comparable Xa inhibitor, razaxaban, leading to severe bleeding in patients and 4) some medical-ethics committees rejected Bayer’s proposal to test once daily dosing of rivaroxaban in clinical trials. The District Court also assesses Sandoz’ argument that the skilled person, starting to test two- or three-day doses in a phase II study would, through routine steps, ultimately end up with a once-daily dosing regimen. But the Court rejects this view as Sandoz failed to describe how the skilled person would have overcome the negative pointers and would endeavor to also try a once-daily doses with a reasonable expectation of success. Conclusion Based on its own assessment and with reference to the Norwegian judgment and the TBA decision, the District Court rejects Sandoz’ claim for invalidity and holds EP 961 inventive. Both counterclaims of Bayer are granted.

On 3 October 2023, the Provisions Judge of the District Court of The Hague rendered its decision in the PI proceedings between Grünenthal GmbH. et al. (“Grünenthal’’) and Teva B.V. et al. (“Teva’’). The Provisions Judge (“PI Judge”) is of the preliminary opinion that EP 1 457 208 B9 (“EP 208”) makes the technical effect plausible and is inventive. The likelihood of EP 208 being held valid in final relief proceedings is high and therefore a PI against Teva is granted. Making this case particularly interesting is that in parallel final relief proceedings, courts in Germany and the United Kingdom have found EP 208 invalid. The PI Judge (who is also a Judge in the Dutch local division of the UPC) is of the preliminary opinion that the plausibility standard in the Netherlands is different from that in the UK, referring to the recent PI decision of the Court of Appeal in the Apixaban case. Grünenthal is since July 2022 the owner of the Dutch part of EP 208 that is in force in the Netherlands until 14 March 2024. It markets the patented drug, which is a long-acting testosterone drug, under the name ‘Nebido’. What preceded  After Teva received a Dutch marketing authorisation for its generic product called ‘Testosterone Teva’ in December 2022, Grünenthal sent warning letters to Teva in February and April 2023, Teva did not respond until 24 August 2023 when it notified Grünenthal that it planned to enter its generic product in the G-Standaard of October 2023. Grünenthal initiated PI proceedings against Teva before the District Court of The Hague, accusing Teva of infringing EP 208. The interim claim in the PI proceedings  A first, interim hearing was held on 12 September 2023 dealing with the incidental (provisional) claim of Grünenthal, requesting the PI Judge to order Teva to withdraw the listing of its generic product from the October G-standard in order to avoid it being published.  The PI Judge handed down its interim decision on 13 September 2023, ordering Teva to withdraw its product from the October 2023 G-Standaard. According to the PI Judge, Grünenthal has a profound interest in maintaining the status quo as Teva is not yet on the market (nor any other generic) nor listed in the G-Standard, so no price erosion has taken place yet. The fact that Teva chose to ignore the warning letters is explicitly taken into account in the ultimate balance of interest in favour of Grünenthal. The main claims in the PI proceedings  On 26 September, the second oral hearing on the main claims was held. Grünenthal claimed that EP 208 is (about to be) infringed and is seeking a PI prohibiting manufacturing and distributing the generic product and an injunction against unlawful conduct.  The PI Judge rendered its decision on the main claims on 3 October 2023, finding the Dutch part of EP 208 provisionally valid and infringed. According to the PI Judge the Dutch part of EP 208 and the Dutch PI proceedings differ from the German and English final relief proceedings because: i) the claims of Dutch part of EP 208 have been substantially limited by Grünenthal during the Dutch PI proceedings, ii) the statements presented, the expert evidence and substantive debate were not identical and iii) the Dutch plausibility standard in the light of inventive step, after G2/21, differs from the plausibility standard in the UK. In formulating the Dutch plausibility standard, the PI judge refers to the G2/21 decision and to the recent decision of the Court of Appeal in the Apixaban case (section 4.35-4.43). which we have discussed in our Pharma Update of 25 August. The PI judge mentions that the plausibility criterion formulated by the Enlarged Board of Appeal in G2/21 is an abstract criterion. It must be filled in based on the specific circumstances of the case. This leads the PI Judge in section 4.41 to consider: “In applying the G2/21 test when assessing the inventive step of a claim in which the technical effect is not included as a feature, unlike the referral from Teva to the TBA case Agrevo, it is not necessary for the patentee to provide proof, or make it plausible in PI proceedings, that the technical effect works with at least the majority of the claimed compositions.”  The PI Judge observes that the outcome could have been different if Teva had made it plausible that the person skilled in the art, based on the application, could not derive, or has reasons to question that the technical effect would occur over the full breadth of claim 1 as limited in the Dutch part of EP 208.

On 15 August 2023, the Court of Appeal of The Hague (”CoA”) rendered its decision in the PI proceedings between Bristol-Myers Squibb Holdings Ireland Unlimited Company (“BMS’’) and Sandoz B.V. (“Sandoz’’), Centrafarm B.V. et al. (“Centrafarm”) & Teva B.V. et al. (“Teva’’). The CoA is of the preliminary opinion that EP 1 427 415 B1 (“EP 415”) makes the technical effect plausible, allowing post-published evidence to be taken into account. The CoA comes to its finding after the Enlarged Board of Appeal (“EBA’’) handed down its decision in G2/21. The likelihood of EP 415 being held valid in the pending final relief proceedings is high and therefore a PI against the generic companies is granted. Making this case of particular interest is that two out of the three CoA Judges are also Appeal Judges in the UPC. This case may show the position the UPC may take on plausibility. BMS is the owner of EP 415 and of SPC no. 300500 based on this patent (“SPC”). EP 415 was in force in the Netherlands until 16 September 2022. The SPC relates to apixaban or pharmaceutically acceptable salt forms thereof (marketed by BMS as Eliquis®) and is in effect from 17 September 2022 until 19 May 2026. What preceded  The first apixaban case in the Netherlands was the PI proceedings between BMS and Sandoz. The Provisions Judge of the District Court of The Hague (“PI Judge” rendered his decision on 10 May 2022. According to the PI Judge, there is no inventive step due to lack of plausibility and therefore a good chance that the patent (and the SPC) will be held invalid in final relief proceedings. A PI against the sale of generic apixaban by Sandoz was denied. No appeal was filed. On 23 March 2023 the EBA decided case G2/21 on plausibility. It considered three questions about the circumstances in which it is permissible to rely on post-published evidence of a technical effect in support of inventive step. The EBA provides guidelines on whether or not to consider post-filed evidence. The new PI proceedings After the EBA ruling, BMS filed another interim relief proceedings against Sandoz and against Centrafarm. BMS requested the PI Judge to grant a PI against both generic companies, alleging infringement of the NL part of EP 415 and the related SPC. The key question was whether the G2/21 decision changed the position taken in the earlier preliminary judgement given on 10 May 2022 and what the standard of plausibility should be in the Netherlands for taking into account post-published evidence. According to the Provisions Judge, the G2/21 decision does not change the application of the concept of plausibility, as the EBA embraces the consistent line of case law of the Technical Boards of Appeal. Therefore, the PI request of BMS is dismissed. BMS also filed interim relief proceedings against Teva before the same Provisions Judge, requesting a PI against Teva. In this case the Parties agreed that a judgement should be rendered identical to the judgment of 17 May 2023, so that all cases could subsequently be heard in the same appeal proceedings. The Provisions Judge rejected the PI against Teva on the same grounds as expressed in his  decision of 17 May 2023. The main considerations of the CoA The key question in the interim relief appeal proceedings is which criteria should be applied in order to decide whether or not post-published evidence can be accepted to hold a patent claim inventive (ab initio plausibility or ab initio implausibility) in light of the decision of the EBA in G2/21. Unlike the Provisions Judge, the CoA holds the patent inventive. It finds that the post-published evidence can be taken into account. According to the CoA, the sole criterion in G2/21 for taking into account a technical effect when applying the PSA is if for the skilled person, having the common general knowledge in mind, and based on the application as originally filed, said technical effect is derivable as being encompassed by the technical teaching and embodied by the same originally disclosed invention. This entails that it is not obliged that the patent application always needs to provide ‘’proof’’ or needs to make plausible that the technical claimed effect actually occurs. This needs to be filled in on a case-by-case basis. The CoA refers to par. 77 of G2/21 to stress that the admission of post-published evidence is different in assessing inventive step when compared to assessing sufficiency of disclosure. The circumstances of the apixaban case imply that on the priority date the average person skilled in the art could derive from the application, WO 652, that apixaban has the most advantageous effect as an fXa inhibitor. Therefore, the criterion as described in G2/21 has been met. The patent application disclosed more than a mere verbal statement of the technical effect so there is no reason for the skilled person in the art to doubt that the function of apixaban (an improved fXa inhibitor) could not be achieved (ab initio implausibility). There is no requirement that any evidence to that effect should be included in the application. The CoA stated that in the final relief proceedings in France and Norway, the Courts came to a similar conclusion based on the same reasoning. It is remarkable that the CoA explicitly considers the decisions of these countries as highly persuasive to the detriment of the UK High Court and Appeal Court decisions, which it distinguishes as being made on the basis of assessing sufficiency of disclosure. The CoA points out (sections 3.21-3.24) that in many countries invalidity proceedings have been instituted, some of which have been decided, others are pending, two of which in NL.