Aristo Pharma GmbH v Takeda Pharmaceuticals USA Inc & Anor

These are two actions relating to the validity of supplementary protection certificate SPC/GB13/052 (‘the SPC’) owned by Takeda Pharmaceuticals USA Inc, a defendant in both proceedings. The two defendant Takeda parties need not be distinguished and they will be referred to collectively as ‘Takeda’.

The two claimants need not be distinguished either. Each in its respective action alleges that the SPC is invalid, advancing the same case.

The SPC protects a drug marketed by Takeda under the name ‘Elvanse’. It is a treatment for attention deficit hyperactivity disorder (ADHD) and has a marketing authorisation in the UK. The ‘product’, i.e. the active ingredient identified in the SPC is lisdexamfetamine (LDX). LDX takes the form of lysine covalently bound to dexamfetamine (DEX). LDX is a pro-drug: it is pharmacologically inactive but upon ingestion it is metabolised to provide the active drug. In this instance the bond between lysine and DEX is hydrolysed, releasing DEX into the patient’s bloodstream. DEX is the pharmacologically active component which treats ADHD. Lysine has no relevant activity of its own.

The claimants’ case is that given the foregoing, the ‘active ingredient’ as that term is defined in the legislation governing SPCs is wrongly identified as LDX in the MA for Elvanse. In reality it is DEX. The relevance is that there is an earlier MA for DEX. The legislation does not permit SPC protection where the MA relied on is not the first MA for the product specified. Accordingly, the claimants contend, the SPC is invalid.

Takeda accept that if the claimants are correct to say that the product and active ingredient in Elvanse is DEX, the SPC is invalid.

Piers Acland KC and Jeremy Heald appeared for the claimant in the first action, Tom Moody-Stuart KC for the claimant in the second action. Thomas Mitcheson KC and Stuart Baran appeared for Takeda.

I congratulate the parties for the efficient presentation of the evidence and arguments at the trial, heard under the shorter trials scheme.

The evidence came from one expert on each side. The claimants’ expert was Professor Jarkko Rautio, Takeda’s was Professor James Brian Houston. Consistently with the effective way these proceedings were conducted, neither expert was cross-examined and indeed there was nothing of technical fact between them. The experts did not agree on points of terminology although in very large part those were matters for the court.

It was not in dispute that the introduction of LDX provided an improved treatment for ADHD when compared with DEX, already on the market. The process of hydrolysing the lysine-DEX bond results in a slower and more prolonged release of DEX than is the case when DEX is administered. This allows once-daily dosing of LDX whereas DEX requires more than one dose per day. Once-daily dosing is more convenient to patients and results in improved dosing compliance. The advantage is particularly marked in the case of school children, who make up a significant proportion of ADHD patients. Once-daily dosing means that the drug can be taken at home before the child leaves for school. The need for a further dose while attending school, necessary when DEX is used, causes complications. These are controlled drugs so a nurse or other individual at the school must be nominated to store the drug safely and to supervise administration. The child’s peers will become aware that he or she is taking treatment for ADHD, which may upset the child and possibly give rise to a stigma.

Immediately before the withdrawal of the UK from the European Union the legislative basis for SPCs was Regulation (EC) 469/2009 of 6 May 2009 (‘the SPC Regulation’). The SPC Regulation has remained part of UK domestic law since withdrawal, albeit in amended form. Although the wording of the articles in the UK version is sometimes different, none of the changes is relevant to the present litigation. Authorities which I will discuss below refer to the SPC Regulation as enacted in 2009 and it is therefore convenient here to refer to that regulation.

Article 1 contains definitions of terms used in the SPC Regulation. The definitions relevant here are those for ‘medicinal product’ and ‘product’:

Thus, the ‘medicinal product’ is the substance or combination of substances administered to the patient. In the present case it is Elvanse, as is common ground. The medicinal product will contain the ‘product’, the latter being the active ingredient or combination of active ingredients which provides the medicinal effect.

Article 3 sets the conditions for the grant of an SPC:

A “valid authorisation … granted in accordance with Directive 2001/83/EC …” means either a national marketing authorisation granted by a Member State under Title III of Directive 2001/83/EC or a centralised marketing authorisation granted by the European Commission under Regulation (EC) No 726/2004. (Directive 2001/82/EC concerns veterinary medicines). The important point is that an SPC can be based only on the first authorisation to place the product on the market as a medicinal product.

Article 8 sets out the information which must be provided by an applicant for an SPC. In relevant part it provides:

Article 8

Content of the application for a certificate

1.

The application for a certificate shall contain:

(a)

a request for the grant of a certificate, stating in particular:

(i)

the name and address of the applicant;

(ii)

if he has appointed a representative, the name and address of the representative;

(iii)

the number of the basic patent and the title of the invention;

(iv)

the number and date of the first authorisation to place the product on the market, as referred to in Article 3(b) and, if this authorisation is not the first authorisation for placing the product on the market in the Community, the number and date of that authorisation;

(b)

a copy of the authorisation to place the product on the market, as referred to in Article 3(b), in which the product is identified, containing in particular the number and date of the authorisation and the summary of the product characteristics listed in Article 11 of Directive 2001/83/EC or Article 14 of Directive 2001/82/EC;

(c)

if the authorisation referred to in point (b) is not the first authorisation for placing the product on the market as a medicinal product in the Community, information regarding the identity of the product thus authorised and the legal provision under which the authorisation procedure took place, together with a copy of the notice publishing the authorisation in the appropriate official publication;

Article 15 states, so far as is relevant:

On 11 April 1990 the European Commission published an Explanatory Memorandum (‘EM’) discussing the then proposed SPC Regulation. The Commission pointed out that SPCs would involve two distinct systems already in place, namely that which authorises the placing of a medicine on the market and the legal system which grants patent protection:

The EM evinces a clear intention that patent offices should find it simple to apply the proposed regulation:

An important task of the patent office concerned is to identify ‘the product’. Paragraph 48 of the EM says:

This implies that all a patent office needs in order to identify the product is the correct marketing authorisation (MA) as provided by the applicant, which will state what the product is.

In Teva BV v Merck Sharpe & Dohme LLC (Joined Cases C-119/22 and C-149/22) EU:C:2024:1039, the CJEU referred to paragraph 16 of the EM in support of its observation that the conditions for obtaining an SPC set out in art.3 of the SPC Regulation are intended to be, ‘in principle, easy to verify’ (at [55]).

In Newron Pharmaceuticals SpA v Comptroller-General of Patents [2024] EWCA Civ 128, Birss LJ (with whom Lewison and Moylan LJJ agreed) said [18]:

The argument in question was addressed at [35]. Birss LJ began by saying:

The appellant in Newron sought to adduce expert evidence on the detail of the contents of the MA, including the Summary of Product Characteristics (SmPC) annexed to it, in support of its contention as to the identity of the product – the active ingredient. Birss LJ said:

The Court of Appeal in Merck Serono SA v Comptroller-General of Patents [2025] EWCA Civ 45 once more emphasised that the SPC system is ‘meant to be a simple transparent one for patent offices to administer’ (at [28]).

Takeda submitted that the importance of keeping the patent office’s task simple is a consequence of the deliberate separation of the functions of the medicines regulator on the one hand and the patent office on the other. Regulatory expertise is required to determine, among other things, the identity of ‘the product’. That is the job of the regulator which, having been carried out, leads to ‘the product’ being identified in the MA. The grant of the SPC is left to the patent office. A key condition to the grant of an SPC imposed by art.3(a) of the SPC Regulation is that the product is protected by a patent in force. Resolution of whether a product falls within the scope of a patent claim may or may not be straightforward and consequently requires expertise in patent law. It is therefore a matter to be dealt with by the patent office. But it is no part of the task allotted to the patent office to revisit the identity of the product concerned, as determined by the regulator.

I agree that consistently with what was said in the EM and what has been said in the judgments I have referred to, it would seem that this is how the SPC Regulation was intended to work. That view is also consistent with art.8(1)(b) of the SPC Regulation which states that the marketing authorisation is a document ‘in which the product is identified’. However, the position is not neatly straightforward. I will discuss this further below after considering the relevant line of CJEU authority.

The authorised product in Massachusetts Institute of Technology (C-431/04) EU:C:2006:291 was a device from which there was controlled release of a known active ingredient, carmustine, from a polymer called polifesprosan. The applicant, MIT, sought an SPC for carmustine in combination with polifesprosan or alternatively for carmustine alone. The application for carmustine alone was rejected because it had an earlier authorisation. The issue with regard to the combination was whether polifesprosan, which enabled the slow release of carmustine, was an active ingredient. The Court ruled that it was not because it did not by itself have any therapeutic effect:

The CJEU applied similar reasoning in GlaxoSmithKline Biologicals SA v Comptroller-General of Patents, Designs and Trade Marks (C-210/13) EU:C:2013:762, ruling that the combination of an adjuvant having no therapeutic effect and a therapeutically active substance did not qualify as an active ingredient.

The difference between those cases and the present proceedings is that Takeda do not argue that the active ingredient – the ‘product’ – is a combination: lysine plus DEX. They say that LDX is a single active ingredient.

The applicant in Arne Forsgren v Österreichisches Patentamt (C-631/13) EU:C:2015:13 owned a patent for a protein called ‘Protein D’. It was used as a carrier, conjugated to pneumococcal polysaccharide serotypes, which together formed the active part of a vaccine against flu. The vaccine had been granted an MA by the regulator. Mr Forsgren applied for an SPC for Protein D alone. It was accepted that Protein D had a therapeutic effect against Haemophilus influenzae but only when covalently bound with the other substances. The CJEU said:

Thus, a substance which is covalently bound to other active ingredients could qualify for SPC protection, but only if it has a pharmacological, immunological or metabolic action of its own, independently of its being bound to (presumably one or more) other substances. This requirement of a pharmacological, immunological or metabolic action of its own has sometimes since been referred to as the Forsgren test.

The Court explained the origin of the Forsgren test and the connection between the definition of ‘active substance’ in art.1(3a) of the Medicines Products Directive (Directive 2001/83/EC as amended by Directive 2011/62/EU) and ‘active ingredient’ in art.1(b) of the SPC Regulation:

I understand the Court there to be saying that the test for an ‘active substance’ under the Medicinal Products Directive 2001/83/EC, a test applied by a regulator such as the MHRA when considering the grant of an MA, is the same as the test for an ‘active ingredient’ under art.1(b) of the SPC Regulation – the Forsgren test. The requirement under art.8(1)(b) of the SPC Regulation that an application for an SPC shall include a copy of the authorisation to place the product on the market, in which the product is identified, is met where the MA identifies the active substance as determined by the regulator.

In Abraxis Bioscience LLC v Comptroller-General of Patents (C-443/17) EU:C:2019:238 the applicant, Abraxis, sought an SPC for a product called Abraxane. It consisted of nanoparticles of a known and earlier authorised active ingredient, paclitaxel, coated with albumin. The albumin was tightly bound to the paclitaxel by a hydrophobic bond. The composition was referred to in the MA as nab-paclitaxel, ‘nab’ standing for nanoparticle albumin-bound. The Abraxane MA relied on in support of the SPC application was the first authorisation to place nab-paclitaxel on the market.

The reference to the CJEU came from a hearing of an appeal in the English Patents Court from a decision by the UKIPO. Abraxis advanced a first argument in the appeal by reference to art.1(b) of the SPC Regulation. It submitted that nab-paclitaxel was a single active ingredient, a product distinct from paclitaxel because the binding of paclitaxel to albumin has important therapeutic consequences. Abraxis drew a parallel with the practice of the UKIPO in granting SPCs for pro-drugs, in which the known active molecule has been modified, typically by the addition of a covalently bonded polyethylene glycol (PEG) group, leading to a prolonged action after administration. Arnold J rejected the argument, holding that no reference to the CJEU was necessary:

Although he did not expressly say so, the judge appears to have accepted that pro-drugs were different from nab-paclitaxel because pro-drugs are new molecules, i.e. a single new ingredient, albeit derived from an earlier authorised active ingredient, whereas nab-paclitaxel was an earlier authorised active ingredient bound to a non-active carrier.

Abraxis had a separate argument based on art.3(d) of the SPC Regulation. It led Arnold J to refer to the CJEU the question whether art.3(d) of the SPC Regulation permitted the grant of an SPC where the product in question was a new formulation of an old active ingredient. The CJEU ruled that it did not. In the dispositif it said that art.3(d) was to be interpreted to mean:

Pausing there, these authorities appear to create a contradiction. They indicate that the policy of the SPC Regulation is not to burden a patent office considering an application for an SPC with the task of revisiting the assessment of the regulator as to the correct identity of the active ingredient. It would follow that a patent office could seldom, if ever, be challenged on the ground that it has taken the identity of the active ingredient as stated in the MA at face value. Yet patent offices have many times been challenged on exactly that ground. Successive courts have ruled on a contention that the patent office in question, and by extension the court, should reinvestigate the assessment of the regulator. In fact, the CJEU has explained that the Forsgren test provides the correct approach to such reassessments.

Halozyme, Inc v The Comptroller-General of Patents, Designs and Trade Marks [2024] EWHC 3202 (Pat) was an appeal from the decision of the UKIPO in which the Hearing Officer refused two SPC applications by Halozyme. It was only necessary to consider the first since they stood or fell together. Halozyme said that the relevant ‘active ingredient’ was a combination of two substances, each of which had a pharmacological, immunological and/or metabolic action of its own. The central question was whether this was true of one of them, recombinant human hyaluronidase.

Two sub-issues arose. The first was how that question should be approached, specifically what materials were relevant in the assessment. The second was to answer the central question on the facts, having determined the correct materials to be considered.

On the first sub-issue Meade J set out three possible approaches to the assessment. The first was that taken by the Hearing Officer: he had said that two documents were primarily to be relied on when assessing a candidate active ingredient. One was the SmPC annexed to the MA. The other was the European Public Assessment Report (EPAR). The EPAR is not formally part of the MA but is derived from the scientific report produced for the European Medicines Agency by the Committee for Medicinal Products for Human Use. The Hearing Officer took the view that identification of the active ingredient should be ‘SmPC/EPAR-led’, although other materials may additionally be considered.

The second approach, supported by Halozyme, was that in addition to the SmPC and the EPAR, other materials should be taken into account, in that instance the basic patent and three scientific papers it had filed in evidence.

In a respondent’s notice the UKIPO argued for a third, hard-line approach: only the SmPC should be taken into account, a view consistent with keeping an application to a patent office for an SPC simple.

Meade J considered the evidence and arguments based on all the materials that Halozyme had proposed and dismissed the appeal. It was therefore not necessary for him to make any ruling on which of the three approaches he had identified was the correct one. He gave two reasons why he should not express a view. First, there was a relevant pending reference before the CJEU from a Czech court based on similar facts. Secondly, he said that he was not at all confident that he knew whether the regulator asks the question required by the Forsgren test when it considers whether a substance is an excipient.

The reference to the CJEU is from Nejvyšší správní soud (the Supreme Administrative Court of Czechia) in Halozyme, Inc v Úřad průmyslového vlastnictví [the Czech IPO] (C-456/24) EU:C:2026:348. On 23 April 2026 Advocate-General Emiliou delivered his Opinion. It begins with the Advocate-General observing that the reference ‘offers the Court an opportunity to revisit the intricate regime governing the grant of supplementary protection certificates’. As he implies, an intricate reality has long overtaken the early ideal of simplicity.

Four questions were put to the Court. The Advocate-General thought it necessary only to answer the first. He said:

Although the central issue was confined to whether a substance found by the regulator to be an excipient can be an active ingredient within the meaning of art.1(b) of the SPC Regulation, the Advocate-General considered the matter in broader terms:

He reiterated that an active substance within the meaning of the Medicinal Products Directive and an active ingredient as defined in the SPC Directive are the same thing:

The Advocate-General went on to firmly state his view regarding the role of the SPC procedure:

He noted the rigour and duration of the MA procedure:

The Advocate-General further noted the EM’s goal of a simple, transparent system and observed that it is consistent with art.8 of the SPC Regulation. He continued:

The Advocate-General faced up to the apparent difficulty presented by the judgment in Forsgren. Halozyme had submitted that since the national courts must assess whether a substance qualifies as an active ingredient on the basis of whether it produces a pharmacological, immunological or metabolic action of its own which is covered by the therapeutic indications of the MA, the court can only perform that assessment after consideration of all the relevant facts. The Advocate-General said:

The Advocate-General proposed that the Court should answer the first question in this way:

As I understand that proposed answer, although the second sentence is confined to the consequences of the regulator designating a substance as an excipient, the first sentence means, more generally, that the patent office should look no further than the MA when deciding whether a substance is an active ingredient – subject presumably to the exceptional circumstances the Advocate-General considered in his paragraph 74.

Of course this Opinion is not the last word. The judgment of the Court is yet to come and even when it is delivered it will not be binding on this court. Nonetheless, I see the force in the Advocate-General’s reasoning, apparently in line with the submissions of the Czech, Irish, French, Dutch and Finnish governments and those of the European Commission. There is evidently an appetite for simplification.

The claimants’ pleadings state that Takeda’s application for an SPC relied on three MAs granted on 1 February 2013 and three on 11 June 2015, which Takeda admit in their Defences. Attention was focused on the first three MAs and in particular the SmPC for those three.

This is section 2 of the SmPC:

As was common ground, this information is presented in conformity with the European Commission’s Guidelines on SmPCs, of which this is the relevant paragraph:

The claimants submitted that section 2 of the SmPC gives the game away by identifying the active moiety of LDX as DEX, an active substance of an already approved medicinal product.

I disagree. I do not read the two sentences of the guidelines as mutually exclusive alternatives. Both apply and the SmPC complies with both. The first sentence states that where the active substance is a pro-drug, the quantity of that active substance should be stated in the SmPC. In compliance, section 2 of the SmPC says that for the three capsules concerned, respectively 30, 50 and 70mg of the active substance, identified as LDX, is present. But this is an instance where the active substance has an active moiety: DEX. The second sentence from the guidelines requires the equivalent quantities of the active moiety to be stated and they are. On a reasonable reading, section 2 identifies LDX as the active substance, albeit containing the active moiety DEX.

The claimants next rely on section 5.1 of the SmPC:

The claimants argued that the first sentence could not be clearer: LDX is pharmacologically inactive. It therefore cannot be an active ingredient.

I think this is to confuse usages. The experts were agreed that according to general usage, pro-drugs are generally inactive compounds that are transformed into an active compound upon administration.

However, the term ‘active ingredient’ has a very specific meaning in the context of the SPC regime. Plainly ‘active ingredient’ can encompass pro-drugs as appears from the Commission guidelines. It must follow that what is generally regarded as an inactive compound, such as a pro-drug, can also be an ‘active ingredient’ within the meaning of the SPC Regulation.

In my view a reasonable reader would have no difficulty in understanding that in the first sentence of section 5.1 of the SmPC ‘pharmacologically inactive’ is being used in the generally understood sense – to mean that Elvanse is (strictly, contains) a pro-drug which is, as is to be expected, inactive in vitro. The reader would not mistake this for an assertion that Elvanse is not an ‘active ingredient’ in the specific sense of the term as used in art.1(b) of the SPC Regulation. The second sentence would be understood to mean that hydrolysis of LDX by red blood cells releases the active moiety of LDX, i.e. DEX, and that the latter is responsible for LDX’s activity.

Section 4.2 of the SmPC states that Elvanse capsules may optionally be opened and the contents dissolved in a glass of water. It continues:

The claimants conceded that ‘active ingredient’ would there undoubtedly be understood to mean LDX.

In my view, if an assessment of whether LDX or DEX is the active ingredient were to be confined to the MA, a reasonable reader would understand that it is LDX.

The parties also referred to the Public Assessment Report (PAR) for Elvanse. A PAR is a document published by the regulator, in this case the MHRA, giving in some detail its reasons for the grant of the MA and which also supports the matters set out in the shorter SmPC.

The claimants relied on this passage in section III.2:

Again, I think that a reasonable reader would understand what was being said in context. LDX is pharmacologically inactive in vitro, but that does not of itself prevent LDX from being an ‘active ingredient’ within the meaning of art.1(b) of the SPC Regulation. Its biological actions are due to the DEX released following hydrolysis,

In two passages of the PAR LDX is referred to as the parent drug and in one of them DEX is referred to as the active substance. Both passages are part of a scientific analysis of how LDX operates as a drug. A reasonable reader would understand them as such and not interpret them to be discussing whether either LDX or DEX falls within the concept of an ‘active ingredient’ as used in the SPC Regulation.

As Takeda pointed out, in three places the PAR expressly states that LDX is the ‘active ingredient’.

I conclude that on a reasonable reading of the PAR, LDX would be understood to be the ‘active ingredient’ of Elvanse within the meaning of art.1(b).

If the question whether LDX or DEX is the active ingredient is to be determined by what is said in the MA and the PAR, the answer would be LDX. However, the claimants said that the court should go further and consider whether such an answer is consistent with the scientific facts – the properties of LDX and DEX.

In a further document issued by the MHRA as part of the regulatory procedure leading to the grant of the MA, the MHRA listed differences in relevant properties of LDX and DEX. This was the Day 210 Report, a final assessment report, published on 16 December 2012. Takeda summarised the points made in the Day 210 Report, which are not challenged as technical facts and which I will set out in abbreviated form:

The Day 210 Report refers to the ‘CHMP reflection paper’, a document published by the European Medicines Agency on 18 October 2012 dealing with the differences between what it calls a reference active substance and variants of in form of a single enantiomer, a complex, a derivative or a different salt or ester. It states that the paper is intended to answer the questions ‘when should such a [variant] be regarded as a new active substance (NAS) in relation to the relevant reference active substance and what level of evidence would be required to confirm the designation as a new active substance?’

The CHMP reflection paper provides a list of evidence likely to be sufficient to show that the variant is a new active substance and another list of evidence unlikely to be sufficient. These are items listed:

As can be seen, if a variant such as a derivative of the reference active substance has a PK profile different to that of the reference active substance, this by itself is not enough to confer on the variant the status of a new active substance under the Medicinal Products Directive. On the other hand, if the change in PK profile means that there is a significant change to the dosing frequency, such as ‘bd to od’ – meaning twice daily to once daily – the variant is likely to qualify as a new active substance. All the more so if this means that there is an improvement in clinical efficacy, such as improved compliance when administered to schoolchildren.

The Day 210 Report refers to the differences in PK profile as between LDX and DEX:

The PAR indicates that a further advantage of LDX is relevant, namely that it was less prone to abuse in the form of non-medical use:

The foregoing supported the ultimate conclusion reached by the MHRA:

The claimants contended that this conclusion was wrong. Initially it had five arguments.

In support of the first of these they referred to the evidence of their expert, Professor Rautio. He and Professor Houston both said that LDX and DEX have a different PK profile, but in Professor Rautio’s opinion it was wrong to view a difference in PK profile as a difference in pharmacological or metabolic action. In his view the pharmacological action of a drug is its therapeutic effect on the body. Professor Houston was of the opposite view. Similarly, while the experts agreed that LDX and DEX have different metabolic properties, Professor Rautio stated that only their metabolic actions, what caused their therapeutic effect, were relevant; they were the same. Professor Houston took the opposite view: the difference in PK profiles meant that LDX has a metabolic action different from that of DEX.

The claimants’ first argument was that the evidence of Professor Rautio should be preferred. Since LDX had neither a pharmacological nor a metabolic action distinct from those of DEX, it could not be a new active substance.

Secondly, the claimants contended that Professor Rautio’s view is supported by the judgments in MIT, GSK and Abraxis in which, they said, the CJEU ruled that suitability, sufficiency and/or efficacy do not provide a basis on which an active ingredient can be identified.

Thirdly, if, contrary to their primary case, the difference in PK profiles was a relevant consideration, the MHRA had based its finding of that difference on the wrong comparison. There were available extended-release reformulations of DEX, marketed as Dexedrine Spansules and Adderall XR. The experts were agreed that these had a PK profile comparable to that of LDX. When compared with the extended-release DEX formulations LDX provides no new pharmacological or metabolic advantage and therefore cannot be said to be a new active substance.

Fourthly, with regard to abuse resistance, the advantage of LDX claimed was the product of the slower release profile of LDX. For the reasons given in the first argument, this does not confer either a pharmacological or metabolic action. Further, the Forsgren test is concerned with therapeutic actions, not what happens when a drug is abused.

Fifthly, the Forsgren test for an active ingredient in accordance with the SPC Regulation was not the same as the test for a new active substance under the Medicinal Products Directive. The Forsgren test requires the patent office to consider whether the substance in issue has a pharmacological, immunological or metabolic action of its own. Those last three words form no part of the test to be applied by the regulator. It is an irrelevant question for the regulator since it is simply considering one candidate substance, or combination of substances, so the words ‘of its own’ can have no operative meaning. Even where the regulator is rightly satisfied that the substance has met the requirements of the Medicinal Products Directive, once the process moves forward to the patent office considering the grant of an SPC, the substance must be examined more rigorously – specifically as to whether it has a pharmacological, immunological or metabolic action of its own. Even if LDX satisfies the test for a new active substance, it does not qualify as an active ingredient when the Forsgren test, with that further requirement, is applied.

In oral argument the claimants’ case was presented a little differently. They said that it was no part of their case that what the MHRA did was wrong when granting an MA for LDX. The claimants accepted that the MHRA correctly characterised LDX as the new active substance within the meaning of the Medicinal Products Directive. The point emphasised was that LDX did not, however, satisfy the Forsgren test because it has no pharmacological, immunological or metabolic action of its own.

Takeda argued that I should adopt the simple approach recommended by Advocate General Emiliou in Halozyme. Since the MHRA has approved LDX as a new active substance in the MA, that is the end of the matter. LDX is ipso facto an active ingredient. Failing that, LDX is shown to be the active ingredient in the MA and PAR. Finally, if the scientific facts – the properties of LDX and DEX – are to be taken into account, LDX satisfies the Forsgren test.

As I have said, it became apparent during oral submissions that it was common ground between the parties that LDX satisfies the test for a new active substance in accordance with the Medicinal Products Directive. It seems to me that in consequence the claimants’ third argument falls away, as must the first and fourth arguments subject only to the fifth. I will consider the first, third and fourth arguments anyway, along with the second argument which is on the law, before turning to their key fifth argument.

I did not find the evidence of the experts helpful, save with regard to peripheral matters. I do not doubt that well informed individuals such as Professors Rautio and Houston can reasonably differ about the precise meaning of the term ‘pharmacological’ or ‘metabolic’ action. But the debate would concern the general understanding of persons in their field, if (though it appears not to be the case) there is a general understanding. The issue here is the meaning of ‘pharmacological action’ and ‘metabolic action’ in the Forsgren test. The underlying facts are not in dispute. The application of those facts to the correct interpretation of the Forsgren test is a matter for the court.

I do not agree that the claimants’ case is supported by the judgments in MIT, GSK or Abraxis. In MIT and GSK the claim was that the active ingredient took the form of a combination of substances, where one of them had an earlier authorisation and the other had no therapeutic effect of its own. Takeda’s case is that LDX is a single substance which qualifies as a new active ingredient. I accept that a candidate active ingredient is not a single substance just because the applicant for the SPC gives it one name, see Abraxis. But here the experts were agreed that LDX is a single molecule derived from DEX and the claimants did not submit that there should be any different understanding in the context of the SPC Regulation. These are their respective formulae:

In Abraxis there was an unchallenged finding of fact that the candidate active ingredient was an earlier approved active substance bound to a carrier which had no therapeutic effect of its own. By contrast LDX is just a single substance, a derivative of DEX.

The claimants formally disclaimed any reliance on an earlier authorisation for any extended-release formulations of DEX in a Response to a Notice to Admit Facts, presumably because there are none. They have not advanced any evidence to show that either of the extended-release formulations of DEX have the same pharmacological and metabolic properties as LDX. The issue must therefore be whether the new compound LDX has a pharmacological or metabolic action of its own when compared with the earlier authorised product, regular DEX.

The claimants’ argument that LDX does not offer better resistance to abuse depends on the proposition that the difference in PK profiles is irrelevant, a proposition I have not accepted.

Finally there is the claimants’ argument that the words ‘on its own’ as used by the CJEU in Forsgren create a relevant difference between that test and the test for an active substance in the Medicinal Products Directive. I disagree.

In Forsgren the issue was whether the candidate for an SPC, Protein D, qualified as an active ingredient under the SPC Regulation. It was protected by a patent and was not the subject of an earlier authorisation. However, there had been a finding of fact that Protein D has a therapeutic effect only when bound to other substances. It was no more than a carrier. In formulating the Forsgren test the CJEU was therefore at pains to emphasise that to qualify as an active ingredient a substance has to display a requisite therapeutic effect – a pharmacological, immunological or metabolic action – of its own. In making that ruling, the Court did not qualify its premise that ‘active substance’ in art.1(3a) of the Medicines Products Directive and ‘active ingredient’ in art.1(b) of the SPC Regulation have the same meaning. Thus, if a regulator were to deal with an application for an MA for a carrier protein it would be duty bound to refuse the application because the carrier has no pharmacological, immunological or metabolic action of its own. As the Court indicated in Forsgren and as the Advocate-General said in Halozyme, ‘active substance’ and ‘active ingredient’ mean the same thing. The test for the two is identical. The recent judgment of Meade J in Accord Healthcare Limited v Novartis AG [2026] EWHC 2127 (Pat) went forward on the basis that the active ingredients in issue were what the regulators concluded them to be in granting the relevant Mas although this was apparently common ground between the parties (at [543]).

Moreover, the words ‘of its own’ are not really engaged in the present case. There is no other substance which falls to be considered when deciding whether LDX has a pharmacological and/or metabolic action. As I have found, it is a single new substance and therefore can only act by itself.

For the foregoing reasons I find that on the agreed scientific facts and having regard to the guidance provided by the European Medicines Agency, LDX has an immunological and/or metabolic action (of its own) due to its PK profile when compared with that of DEX. I am less certain about the relevance of the lesser danger of abuse of the drug but need not decide that point.

It follows that LDX is not only the active substance within the meaning of the Medicinal Products Directive, but also the active ingredient as defined by art.1(b) of the SPC Regulation.

For completeness I should mention that in order to avoid what Takeda characterised as burdensome disclosure, at the case management conference Takeda agreed that (a) the fact that the MHRA found that LDX is a new active substance and (b) whether or not the MHRA should have made that finding should be treated as irrelevant at this trial. Nonetheless, the arguments advanced by the parties required the ground covered by the MHRA during the regulatory process to be covered in the same or a similar way at the trial. And as I have said, the correctness of the decision of the MHRA fell away as an issue.

In addition, Takeda advanced an argument based on the Plant SPC Regulation, Regulation (EC) 1610/96. Given my finding stated above and particularly given the way in which the claimants put their case in their oral argument, it adds nothing.

That leaves Takeda’s contention that a short cut should be taken as recommended by Advocate-General Emiliou in Halozyme. I find his reasoning compelling, though with the qualification that if correct in law it has the potential to rule out a challenge to an unambiguous finding by the regulator even if (in principle) the regulator’s reasoning were to be irrational. In this jurisdiction there may be the possibility of a judicial review of a finding by the MHRA, but I have not had full argument on the scope and flexibility of that potential remedy in such a case. I infer from Advocate-General Emiliou’s Opinion that to the extent that he considered the possibility of an irrational decision by a regulator which is difficult to correct, he believed that it is a price worth paying for a simpler procedure. Because of the finding I have made it is not necessary for me to take this further.

There remains one further matter. The claimants relied on the judgment of the German Federal Patent Court (FPC) dated 4 June 2024, 3Ni 22/22 (EP), in proceedings concerning the German equivalent of the SPC. The FPC ruled that Takeda’s German equivalent of the SPC is invalid. The claimants submitted that the FPC’s reasoning is instructive. (Takeda have appealed and the appeal is pending.)

The German SPC was granted by the German Patent and Trade Mark Office on the basis of marketing authorisations from the German Federal Institute for Drugs and Medical Devices and also those of the UK MHRA. The FPC was shown the PAR and the Day 210 Report published by the MHRA.

The FPC does not appear to have disputed the correctness of the finding by both the MHRA and the German marketing authorisation that LDX is an active substance which is new relative to DEX within the meaning of art.1(3a) of the Medicines Products Directive.

However, the FPC said that the finding under the Medicines Products Directive was irrelevant because the criterion in law for such a finding was not the same as that for a finding that a product is an active ingredient within the meaning of art.1(b) of the SPC Regulation:

I am respectfully unable to agree with the FPC on its view of the law. For the reasons I have explained, particularly by reference to the judgment of the CJEU in Forsgren and the opinion of Advocate-General Emiliou in Halozyme,I have no doubt that the criteria for assessing whether a substance is a new active substance under the Medicines Products Directive and whether it qualifies as an active ingredient under the SPC Regulation are identical.

Freed from having to investigate the reasoning of the MHRA or the German regulator, the FPC considered afresh whether LDX is an active ingredient. The FPC referred to the German MA which, in accordance with the European Commission’s Guidelines on SmPCs, stated the equivalence of LDX to DEX in terms of drug quantity. The claimants in the present case raised a similar point, discussed above. For convenience I reproduce here one of the examples from the UK MA:

The FPC said that these statements of equivalence made it ‘immediately clear’ that LDX and DEX are identical active substances. I am unable to agree for the reasons I have given in relation to the claimants’ argument along similar lines.

The FPC pointed to the explanation in the regulatory documents that upon oral administration LDX is hydrolysed to release DEX which has a therapeutic effect. The FPC drew from this that LDX therefore has no pharmacological or metabolic effect of its own as required by the Forsgren test. As I have said, in my view the European Medicines Agency’s CHMP paper clearly indicates otherwise. I would assume that the FPC thought that the CHMP paper was irrelevant as it was concerned only with the test of a new active substance under the Medicines Products Directive, but in my view it is entirely relevant.

The FPC referred to an earlier judgment by another Senate of the FPC in Paliperidone (judgment of 5 September 2017, 14 W (Pat) 25/16 GRUR Int 2017, 961). The FPC noted that in this earlier judgment the court found that an SPC for a pro-drug was valid on the grounds that it exhibited its own pharmacological and metabolic effects ‘due to a significantly reduced dosing frequency, an exceptionally slow and continuous release, and a significantly prolonged, therapeutically desired mean duration of action.’ The FPC distinguished the Paliperidone judgment because ‘lisdexamfetamine does not exhibit such distinct and extensive new pharmacological and metabolic properties’.

This last observation of the FPC appears to be consistent with the view that LDX does exhibit new pharmacological and metabolic properties of its own, just not enough. Certainly the CHMP paper indicates that the differences in properties between an authorised substance and, for instance, a pro-drug derivative, have to be significant. But it also says that a change from the need to dose twice daily to just once daily is likely to be significant, a sufficient difference in properties, and I have no reason to believe that it is not significant in the present case. Again, I assume that the FPC did not feel the need to consider the CHMP paper because it is concerned with an assessment under the Medicinal Products Directive.

The FPC referred to a formulation of DEX sold as Attentin, which appears to have been a slow-release formulation. For the reasons I have stated, I do not agree that this is relevant.

Finally, the FPC said the reduced risk of abuse was not related to the medicinal effect of LDX and was therefore not to be taken into account. If correct, it makes no difference to the PK profile of LDX conferring on LDX a pharmacological and/or metabolic actions of its own.

With great respect to the FPC, I am not persuaded by its judgment that the conclusions I have reached should be qualified in any way.

I was also shown judgments in parallel proceedings in Denmark, Sweden and Switzerland. These were all in Takeda’s favour, so I need not discuss them.

The SPC was validly granted. The claims are dismissed.