Merck Sharp & Dohme Ltd, R (on the application of) v Licensing Authority & Ors

This is the fourth round in the contest between those who hold marketing authorisations for medicinal products which they have developed, known as innovators, and those who, in applying for marketing authorisation for their own medicinal products seek to rely upon the data already provided by the innovator. (Such companies are called “generic companies” to distinguish them from innovators.)

Merck Sharp and Dohme Limited (“MSD”) hold marketing authorisations for Fosamax 5mg, 10mg and Fosamax Once Weekly, 70mg, medicinal products used to treat osteoporosis. The generic name is alendronate sodium trihydrate. Fosamax 5mg was first authorised in the European Union in July 1993, Fosamax 10mg in May 1995.

Thereafter, MSD developed Fosamax Once Weekly 70mg which was authorised in November 2000.

All three interested parties seek marketing authorisation for generic versions of Alendronate 70mg. Approved Prescription Services Ltd (“APS”) and Generics (UK) Ltd (“Generics”) seek market authorisation from the competent authority in the United Kingdom, the defendant. Arrow Generics Ltd has obtained a marketing authorisation for Alendronate 70mg in Sweden and is apparently preparing a mutual recognition application in the United Kingdom.

Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community Code relating to medicinal products for human use (“the Directive”) requires an applicant for marketing authorisation to supply a full set of data in order to show the safety and efficacy of the product. But in certain circumstances the Directive relieves an applicant of the obligation to supply certain data and permits it to refer to data submitted in respect of a previously authorised product. Where authorisation is sought for a product essentially similar to another product authorised for a period of six or ten years (ten years is the applicable period in the United Kingdom) it may refer to the data already submitted in respect of the other authorised product. It is relieved of any obligation to provide any further data. The procedure for such an application is known as “the abridged procedure”.

Where authorisation is sought for a product which is not essentially similar to one which has been authorised for ten years but differs only in particular respects an applicant may rely upon the original data and submit appropriate additional “bridging” data to cover the respects in which there is a difference between the two otherwise similar products. The procedure for such an application is known as “ the hybrid abridged procedure “.

The European Court of Justice has given judgment in three previous contests between an innovator and generic companies: Case C-368/96 R v The Licensing Authority established by the Medicines Act 1968 ex parte Generics UK Ltd and others (Generics), judgment delivered 3 December 1998,Case C-106/01R (on the application of Novartis Pharmaceuticals UK Ltd) v The Licensing Authority by the Medicines Act 1968 (Novartis) judgment delivered 29 April 2004 and Case C-36/03 Approved Prescription Services Ltd v The Licensing Authority (APS) judgment delivered 9 December 2004. The innovator lost in all three cases. This, the innovator hopes, will be the fourth case to be considered by the Court since it contends that the point in issue has not been decided, is unclear or has been wrongly decided in Novartis and APS.

Although Fosamax Once Weekly has the same active ingredient as Fosamax 5mg and 10mg, it is not “essentially similar” to Fosamax 5mg and 10mg, within the meaning of Article 10 of the Directive. Its posology differs. Posology is not concerned with that art form in which Emma Hamilton excelled. It connotes the schedule of dosage; how much and how often a particular dose of medicine should be taken. In days when a famous conductor was the scion of a pharmaceutical company, producing eponymous pills, a music hall song referred to posology:-

“If you take one pill, you get the Barber of Seville, if you take two more you get the Song of the Toreador but if you take the bloomin’ bottle you’ll hear the angels sing….”

The generic companies rely upon the data submitted in relation to Fosamax 5mg, 10mg and 70mg in seeking marketing authorisation for a generic product which is a copy of Fosamax 70mg, notwithstanding that Fosamax 10mg and Once Weekly 70mg have been authorised for less than ten years. The defendant (“MHRA”) accepts that the generic companies do not have to proffer any further data. The claimant innovator contends that such an approach is unlawful and outwith the relevant Directive. The MHRA and the generic companies contend that the point has already been decided in the three previous decisions of the European Court of Justice and there is no doubt but that they are entitled to authorisation without reliance on any additional data whatever. Thus this case is yet another conflict between rights which an innovator asserts to protect the product of its development and research and those who wish to market a product, without having incurred the expense of development and testing, avoiding repetition of tests already sanctioned by the competent licensing authority.

I shall adopt the approach of the European Court of Justice by calling the product authorised for more than ten years product A (in the instant case Fosamax 5mg), the development or “line extension” of that product, product B (Fosamax 70mg) and the product in respect of which the generic companies seek authorisation, product C. Product A is known as the reference product but since that begs one of the questions which MSD say arises in this case I shall try to avoid referring to product A as the reference product.

The Directive has recently been amended by Directive 2004/27/EC of 31March 2004. Since the wording of the new Directive, required to be implemented by October 2005, differs, I need not refer to its terms.

The recitals in the Directive, relevant to the instant case, provide:

“(2)

The essential aim of any rules covering the production, distribution and use of medicinal products must be to safeguard public health.

(3)

However, this objective must be attained by means which will not hinder the development of the pharmaceutical industry or trade in medicinal products within the Community.

(9)

Experiences has shown that it is advisable to stipulate more precisely the cases in which the results of toxicological and pharmacological tests or clinical trials do not have to be provided with a view to obtaining authorization for a medicinal product which is essentially similar to an authorized product, while ensuring that innovative firms are not placed at a disadvantage.

(10)

However, there are reasons of public policy for not conducting repetitive tests on humans or animals without over-riding cause.”

Article 1 defines medicinal product as:-

“Any substance or combination of substances presented for treating or preventing disease in human beings.”

Article 6 provides:-

“1.

No medicinal product may be placed on the market of a Member State unless a marketing authorization has been issued by the competent authorities of that Member State in accordance with this Directive…”

Article 8 provides:-

“1.

In order to obtain an authorization to place a medicinal product on the market …, an application shall be made to the competent authority of the Member State concerned.”

The defendant is the competent authority in the United Kingdom.

“3.

The application shall be accompanied by the following particulars and documents, submitted in accordance with Annex I:

….

(c)

Qualitative and quantitative particulars of all the constituents of the medicinal product in usual terminology, ….

(f)

Posology, pharmaceutical form, method and route of administration and expected shelf life. ….

(i)

Results of:

- physico-chemical, biological or microbiological tests,

- toxicological and pharmacological tests,

-

clinical trials.”

Article 10 provides:-

“1.

In derogation of Article 8(3)(i), and without prejudice to the law relating to the protection of industrial and commercial property:

(a)

The applicant shall not be required to provide the results of toxicological and pharmacological tests or the results of clinical trials if he can demonstrate:

(i)

either that the medicinal product is essentially similar to a medicinal product authorized in the Member State concerned by the application and that the holder of the marketing authorization for the original medicinal product has consented to the toxicological, pharmacological and/or clinical references contained in the file on the original medicinal product being used for the purpose of examining the application in question;

(ii)

or that the constituent or constituents of the medicinal product have a well established medicinal use, with recognised efficacy and an acceptable level of safety, by means of a detailed scientific bibliography;

(iii)

or that the medicinal product is essentially similar to a medicinal product which has been authorized within the Community, in accordance with Community provisions in force, for not less than six years and is marketed in the Member State for which the application is made. This period shall be extended to 10 years…. Furthermore, a Member State may also extend this period to 10 years by a single Decision covering all the medicinal products marketed on its territory where it considers this necessary in the interest of public health. Member States are at liberty not to apply the six-year period beyond the date of expiry of a patent protecting the original medicinal product.

However, where the medicinal product is intended for a different therapeutic use from that of the other medicinal products marketed or is to be administered by different routes or in different doses, the results of appropriate toxicological and pharmacological tests and/or of appropriate clinical trials must be provided.”

Article 11 provides:-

“The summary of the product characteristics shall contain the following information:

….

5.7

posology and method of administration for adults and, where necessary, for children.”

I have already observed that the United Kingdom has extended the period of authorisation, during which data is protected, to the period of ten years allowed in Article 10(1)(a)(iii). Article 10(1)(a)(i) and (iii) refer to medicinal products which are essentially similar to a medicinal product already authorised whereas what become known as the proviso (i.e. that paragraph beginning with “However”) makes no reference to essential similarity.

It should also be observed that whereas posology is referred to in Article 8(3)(f) and point 5.7 of Article 11 it is not referred to in the proviso.

The recitals, to which I have already referred, identify three important objectives which fall to be reconciled. Recital two refers to the paramount objective namely the protection of public health. The third taken with the ninth relate to the need not unduly to impede innovators and the tenth, in contrast to the third and ninth, the objective of avoiding unnecessary tests where they had already been undertaken by another. It should also be noted that whilst the purposes set out in the third and ninth recitals are to safeguard the interests of innovators and thereby foster research, the protection afforded is over and above that which is provided by the law relating to the protection of intellectual property. This, the opening words of Article 10 make clear.

It was in seeking to reconcile these three objectives that the European Court of Justice developed jurisprudence relating to Article 10. Since the essential point in this case is whether the issue advanced by MSD has already been decided, as a matter of principle, by the European Court of Justice, it is necessary to turn to the trilogy.

In Generics the innovator companies held marketing authorisations for not less than ten years in respect of product A. Subsequently they had obtained marketing authorisations for product B, the development of product A, in respect of different therapeutic indications (e.g. a change from an anti-ulcer use to the treatment of severe oesophagitis). Those different indications had been authorised for less than ten years.

The generic companies sought marketing authorisation for products essentially similar to product A, not only for the therapeutic indications authorised for A, but for all the therapeutic indications previously authorised. They sought such authorisation without offering any additional data. The innovator companies contended that the data upon which they had relied to gain authorisation for the additional therapeutic uses were entitled to protection for an additional ten year period and could not be relied upon either by the competent authority or the generic companies.

The Court ruled, disagreeing with the Advocate General, that provided that the generic companies’ products were essentially similar to product A, they should be authorised, pursuant to Article 10(1)(iii), for all the therapeutic uses previously authorised without the need for reliance upon any additional data.

The case is of importance not only because the Court defined what is meant by “essential similarity” but also because it demonstrates how the Court reconciled the conflicting interests of the innovating companies and generic companies. In particular, in reconciling those interests it demonstrates the absence of any weight to be given to expenditure in time and costs incurred by the innovating companies in developing their products.

Further, it explains, in seeking to achieve reconciliation between conflicting interests, the importance of protection afforded to innovators by the law relating to intellectual property to which the protection afforded by Article 10 is only an addition. It was described, by Mr Green QC on behalf of APS (now known as Teva) as the third level of protection.

In order to highlight the way in which the Court reconciled the conflicting interests, it is necessary to examine the opinion of Advocate General Ruiz-Jarabo with whose conclusions the Court disagreed. Research into different therapeutic uses with different doses had involved thousands of patients and had cost sums exceeding several tens of millions of US dollars (see paragraph 15 of his opinion). In referring to an innovator’s extension of initial clinical indications, it is interesting to note, changes were made to recommended dosage schedules (see paragraph 26 and footnote 15). The Advocate General did not consider that the requirements of safety and efficacy were relaxed either under the abridged or the hybrid abridged procedures. He concluded, in relation to public health:-

“49.

Finally, I would point out that the protection of public health is compatible with an extension of the marketing authorisation for generic medicinal products so as to include all indications, routes of administration and dosage schedules authorised for the original medicinal product up to the time of issue of that authorisation.”

The Court agreed with that proposition, as I shall relate, but disagreed with his final conclusion based on the importance of protecting research innovation. The Advocate General pointed out that the six or ten year protection period (now laid down in Article 10(1)(a)(iii)) is intended specifically to safeguard the interest of innovators and foster research (see paragraph 52). He pointed out that other Community national and international provisions relating to the protection of intellectual property, in particular patents, provided safeguards for innovation (see paragraph 53). He identified supplementary protection afforded by Community law to compensate for the period extending from the filing of a patent application to the grant of marketing authorisation (see the final paragraph of paragraph 53). But he concluded that in order to protect innovation and pharmaceutical research it was advisable to apply the six or ten year protection to:-

“all new indications of considerable therapeutic importance authorised for an original medicinal product essentially similar to a generic medicinal product.”(Paragraph 54).

However, for the purposes of the argument in the instant case, it is important to recall that the Advocate General confined his views as to the necessity of an additional six or ten year protection to new indications of considerable therapeutic importance. He rejected such protection in relation to new routes of administrations and dosage schedules for the original medicinal product which, he said, did not constitute significant therapeutic innovations and were not therefore covered by the additional protection period (see second paragraph at paragraph 57). The reason why the Advocate General rejected any additional protection period for new routes of administration or dosage schedules was that it was consistent with the rule seeking to avoid repetitious tests on persons and animals, unless strictly necessary:

“The interpretation suggested is also in harmony with the rule requiring non-repetition of tests on persons and animals unless strictly necessary. New therapeutic indications, routes of administration and/or dosage schedules authorised for an original medicinal product are supported by the tests carried out by the innovative undertaking and it is not advisable that they be repeated merely because there has not been a time lapse of more than 6 or 10 years since the authorisation of those modifications.”(Paragraph 59).

As I have said, the Court rejected the Advocate General’s conclusion. Where the generic companies’ product C was essentially similar to product A the fact that the innovator had developed the product for different therapeutic indications afforded the innovator no additional protection period. Generic companies were entitled to market authorisation for all therapeutic indications developed from the original indication, and all developed doses and dosage schedules provided only that product C was essentially similar to product A.

The Court set out criteria for establishing essential similarity, based upon the Commission’s rules governing medicinal products and guidelines on the quality safety and efficacy of medicinal products for human use (see paragraph 31). Product A is essentially similar to product B if it satisfies the criteria of:

“having the same qualitative and quantitative composition in terms of active principles, of having the same pharmaceutical form and of being bioequivalent, unless it is apparent in the light of scientific knowledge that it differs significantly from the original product as regards safety or efficacy.” (See paragraph 36).

The third of those four criteria, bioequivalence, refers to an equivalent bioavailability. Bioavailability is the rate and extent of a medicinal product’s absorption into the body and of its transfer to the site of action (see paragraph 10 in Novartis).

The Court then considered the second question, namely what therapeutic indications may be authorised under the abridged procedure (now Article 10(1)(a)(iii)). It ruled that where a medicinal product is essentially similar to a product which has been authorised for not less than six or ten years the applicant is not required to provide any data (see paragraph 39). The competent authority may use all the documentation relating to tests for the original medicinal product and for more recent therapeutic indications (paragraph 40). The Court observed that similarity in therapeutic indication is not one of the criteria according to which two medicinal products may be regarded as essentially similar. Accordingly, an applicant for marketing authorisation for a medicinal product essentially similar to a product authorised for six to ten years is not required to provide any data whatever the therapeutic indication (see paragraph 43). An applicant may receive marketing authorisation under the abridged procedure for all therapeutic indications including those indications which had been authorised for less than six or ten years (see paragraph 44). Thus, the Court rejected the Advocate General’s conclusion that such therapeutic innovation should be given independent protection (see paragraph 46.)

The reasoning for rejecting the view that new therapeutic indications should be given an additional period of protection is relevant in the instant case. The Court rejected the idea that such innovations should be cloaked in a further period of six to ten years protection firstly because of the wording of what is now Article 10(1)(a)(iii) (paragraph 47) and secondly because the concept of a major therapeutic innovation lacks precision and therefore undermines the principle of legal certainty. There is no precise way of distinguishing between simple and greater changes of therapeutic indications (see paragraphs 47 to 49).

Thirdly, the Court took the view that the same answer should be given to the question in relation to dosage form, doses and dosage schedules so long as they do not preclude essential similarity between the medicinal products (see paragraphs 55 to 56).

The decision in Generics left open the question whether an applicant could rely upon data supplied by an innovator where the generic product C was essentially similar to a development of the original product, product B, but not to product A. In Novartis the innovator had developed the original product Sandimmun (product A), authorised for more than ten years, into another product Neoral (product B) which lacked bioequivalence to product A. It was supra-bioavailable, in other words its rate and extent of absorption into the body and transfer to the site of action was greater than that of Sandimmun. In those circumstances, having regard to the criteria for essential similarity, B was not essentially similar to A.

The generic companies sought authorisation for Sang Cya, product C, which was essentially similar to B and accordingly not essentially similar to A. The Court ruled that although bioavailability was not explicitly referred to within the proviso it was implicit. In those circumstances the generic applicant could rely upon the bridging data on which the innovator had relied in relation to product B and the original data in relation to product A and was not required to proffer any separate data, notwithstanding that Neoral (Product B) had been authorised for less than ten years.

Novartis is of significance in the instant case not least because it is said that the wording of the Court’s judgment fully disposes of MSD’s application. Apart from the wording, it is argued that the reasoning of both the Court and Advocate General Jacobs, whose conclusion was accepted by the Court, apply a principle which concludes the matter against MSD. It is the controversy as to whether that is a fair reading or, alternatively, whether the Court in Novartis was wrong, which founds the submissions advanced by MSD.

The innovator’s first submission was that cross-referencing was never to be permitted lest innovative firms be placed at a disadvantage (paragraph 43 of the Advocate General’s Opinion). Alternatively, Novartis suggested that cross-reference should be permitted only where Products A and B meet in full the requirements of essential similarity (paragraph 44). The Advocate General concluded that Novartis’ first submission was inconsistent with the Court’s conclusions in Generics which, he said, were founded on the essential similarity of the original reference product and its subsequent variants. He said that any conclusion to the contrary would allow applicants to gain additional data protection merely by creating a new “designation” notwithstanding essential similarity. That would be to “elevate form over substance” (see paragraph 57).

The Advocate General rejected the alternative submission for three reasons. Firstly, the cost of the development of product A into product B has no correlation with the issue whether A is essentially similar or not (see paragraph 59). No doubt the Advocate General bore in mind the expense and difficulty of developing products for different therapeutic indications, notwithstanding that they were essentially similar, as emerged in Generics. Secondly, to limit the possibility of cross-reference only to products which were essentially similar, as defined in Generics, would be to confine such cross-reference to new therapeutic indications because a change in dose will affect quantitative composition, a change in dosage form will effect pharmaceutical form and both such changes will have an impact on bioequivalence (see paragraph 60). Thirdly, the ability to cross-refer provides the correct balance between the conflicting objectives of data protection and the avoidance of unnecessary testing on humans and animals because it reserves additional data protection:-

“for the most significant modifications to an original product, namely those which involve the introduction of a new active substance.” (Paragraph 61)

The Advocate General concluded that essential similarity was not a pre- requisite for the use of the hybrid abridged procedure under the proviso. His reasoning is germane to the instant case. If one product (C) is essentially similar to another (A) which has received authorisation then the safety and efficacy of product C may be assumed. A similar assumption may be made where product C is essentially similar to product B. The safety and efficacy of product C may be assumed notwithstanding that B’s authorisation has existed for less than six or ten years. As the Advocate General put it:-

“The purpose of the proviso is to allow an applicant whose product is essentially similar to an existing product except in-so-far as it differs in one or more of the respects stipulated by the proviso to submit additional or bridging data only with regard to that difference. The relaxation of the criteria of essential similarity in respect of the differences specified in the proviso is possible precisely because the proviso then requires additional bridging data to be submitted, thereby assuring that the safety and efficacy of the new product can none the less be assessed”. (See paragraph 67)

The Advocate General found support in the Notice to Applicants (see paragraph 68).

Further, echoing paragraph 60, he pointed out that to limit use of the proviso to products which are essentially similar would render two of the categories of difference, namely change in dose or in route of administration, inapplicable. As he pointed out:-

“A change to the dose of a medicinal product will preclude essential similarity, given that it will constitute a change to the quantative composition of the product. Similarly, an alteration to the route of administration will in many instances amount to a modification of pharmaceutical form.” (Paragraph 69).

The Court agreed with the Advocate General’s reasoning and conclusion. In the light of MSD’s argument that the proviso is not concerned with the ability of a generic company to cross-refer to the data of an innovator, the structure of the Court’s judgment is of significance. The Court first considered whether two products could be essentially similar if they were not bioequivalent (questions four and five); secondly, it considered whether essential similarity was a pre-requisite for the use of the proviso (the second part of the third question), thirdly, it considered whether a national competent authority was ever entitled to cross-refer without consent to data submitted in support of product B authorised for less than six or ten years (the first and second questions) (see paragraph 32 in the Court’s preliminary observations). In those observations, the Court recalled that the primary purpose of the rules under the Directive is to safeguard public health.(paragraph 30).

The Court concluded that products cannot be regarded as essentially similar where they are not bioequivalent (see paragraph 35).

The Court concluded that the application of the proviso was not confined to products which were essentially similar. It reasoned that such restriction would render the proviso largely ineffective. Essential similarity requires bioequivalence. But products administered by routes or doses different from those of the reference product do not “generally” have the same bioavailability and thus the proviso would be largely ineffective where medicinal products were administered by different routes or in different doses since they would generally lack bioequivalence (see paragraphs 51 and 52).

The Court also relied on the Notice to Applicants which expressly stated that the proviso could be applied absent essential similarity (see paragraph 53). It stressed that the purpose of the bridging data under the hybrid abridged procedure, pursuant to the proviso, is to prove safety and efficacy in relation to the differences between Products A and B. It concluded that an application for marketing authorisation for product B may be made where the differences fall within the proviso even though product B is not essentially similar to reference product A. The wording in paragraph 55 lost something in translation but its meaning is plain. The wording was important to the defendant and the generic companies’ arguments since it refers to more than one difference under the proviso;

“… an application for marketing authorisation for a medicinal product may be made under the proviso with reference to an authorised medicinal product provided that the medicinal product in respect of which marketing authorisation is sought is essentially similar to the authorised medicinal product, unless one or more of the differences set out in the proviso apply, as the case may be.” (Paragraph 55).

In answer to the first and second questions the Court concluded that the competent authority of a Member State was entitled, in considering marketing authorisation of product C, to refer to the data submitted in support of product B authorised for a period of less than six or ten years, notwithstanding that product B lacks bioequivalence to product A. The Court first observed that in Generics new therapeutic indications for a medicinal product already authorised could not be afforded a further period of protection of six or ten years (see paragraphs 57 to 58). Nor can such protection be afforded where a medicinal product is administered by different routes or in different doses. The Court then, in a manner which is of significance to the arguments in this case, identified what is meant by a development of an original or reference medicinal product. A product which is administered by routes or in doses different from those of a medicinal product already authorised:-

“is a development of the original or reference medicinal product in the same way as a medicinal product intended for a different therapeutic use from that of the original or reference medicinal product.” (See paragraph 60)

It is not decisive that the product to be administered by a different route or in different doses does not satisfy the criteria for essential similarity, unlike a medicinal product intended for a different therapeutic use (see paragraph 61).

The Court continued:-

“It should be noted in that connection that whether or not the product resulting from the development of the reference medicinal product satisfies all the criteria for essential similarity to the latter product does not necessarily bear any relationship to the cost or difficulty involved in that development.” (Paragraph 62).

The Court reasoned that differences in route of administration or in the dose “generally imply a lack of bioequivalence”. If cross-referral is permissible where there is a difference in route of administration or in the dose it must follow that such cross-referral is also permitted where there is a difference in bioavailability, even though the difference is not attributable to a change in the route of administration or dose:-

“If, as stated at paragraph 64 of the present judgment, the applicant for marketing authorisation for product C may refer to pharmacological, toxicological and clinical documentation in respect of product B, which is the product of the development of the reference product A and essentially similar thereto, apart from the route of administration or the dose, as the case may be, since the differences in those two factors generally imply that products A and B are not bioequivalent (see paragraph 51 of the present judgment), it must, a fortiori be able to do so where products A and B are distinguishable only by their different bioavailability, even though the route of administration and dose remain unchanged.” (See paragraph 66).

Thus the Court concluded that even though difference in bioavailability was not expressly identified in the proviso since it was implicit in the differences expressed, cross-referral to data submitted in support of product B, which differed in bioavailability to product A, was permissible in an application for marketing authorisation for product C, essentially similar to product B but not to product A.

Although the judgment in APS, dated 9 December 2004, followed that of the Court in Novartis (April 2004) the reference to APS was made in December 2002, before the opinion of the Advocate General and the judgment of the Court in Novartis. The only distinction between APS and Novartis was that the difference between product B and product A was not in bioavailability but rather in pharmaceutical form. It concerned Prozac. Prozac in capsule form (product A) had been authorised for over ten years, since 1988. Prozac Liquid (product B) was subsequently authorised in October 1992 under the hybrid abridged procedure pursuant to the proviso. In 1999 a generic company, APS, sought authorisation for a generic liquid containing the active ingredient Fluoxetine under the abridged procedure using Prozac capsules as a reference product but also seeking to rely on the clinical data in relation to Prozac Liquid. The Court applied the same reasoning as that which it applied in Novartis. Pharmaceutical form is not referred to in the proviso. But a difference in the route of administration between product B and product A normally entails a change of pharmaceutical form. In those circumstances, where products A and B are distinguishable only by their different pharmaceutical form, a new applicant may cross-refer to the data in respect of product B even though the route of administration remains the same. Just as a difference in bioavailability is implicit within the proviso, so is a change in pharmaceutical form. In both cases an applicant for authorisation of product C which is essentially similar to product B may refer to the data in respect of product B notwithstanding that that product had been authorised for less than six or ten years (see paragraphs 25 to 26 of the judgment).

The additional feature of APS which founded an argument advanced by Mr Gordon Q C, on behalf of MSD, related to the procedure under which an applicant for authorisation of product C could seek authorisation on the basis that product C was essentially similar to product B, a development of product A. The United Kingdom contended that the applicant must apply under the proviso although it accepted that no more data would be required (see paragraphs 74 to 75).

Advocate General Jacobs concluded that the application for authorisation of product C need not be made under the proviso. He said:-

“Indeed, there are good reasons against requiring an application in respect of product C to proceed under the proviso. The proviso operates in circumstances where bridging data are required because of a difference between the new product and the earlier product or products to whose data reference is made. Where product C claims essential similarity to product B which is a variant of product A, no additional data are required. There is therefore no need to proceed under the proviso.” (See paragraph 83).

In those circumstance he concluded that the applicant was required to show essential similarity to “one or other form of the reference product and that the product in question has been authorised within the Community for not less than six or ten years.” (See paragraph 84).

The Advocate General pointed out that there was no requirement that the data to which reference was made in an application must have been submitted to an authority at least six or ten years previously. It is the product which must have been authorised for that period (paragraph 85). Thus, the Advocate General in construing Article 10(1)(a)(iii) elided the original reference product, product A, authorised for six or ten years, and the development of that product, product B, referring to that development as being “a form of the reference product”.

The Court followed the Advocate General in relation to the procedure for marketing authorisation for product C. It concluded that an application for marketing authorisation for product C “may” be made under Article 10(1)(a)(iii) as product C is essentially similar to product B in circumstances where:-

“- product B is a new pharmaceutical form of product A, and

-

product A but not product B, has been authorised for marketing in the Community for at least the six or ten year period stipulated therein.” (Paragraph 30).

Both the Advocate General and the Court referred to the Notice to Applicants: procedures for Marketing Authorisations which supported their conclusion (see e.g. paragraph 27 of the Court’s judgment).

MSD challenged the decision of the competent authority, MHRA, to allow generic companies to submit marketing authorisation applications for Alendronate 70mg relying on the data submitted by MSD for Fosamax Once Weekly. It is, as MHRA pointed out, not so much a challenge to the policy but to its application. MSD seeks, in this application, a reference to the European Court of Justice. It does not go so far as to contend that the jurisprudence of the European Court of Justice dictates refusal of authorisation but rather that the question is as yet unresolved. It relies on three distinctions from the previous decisions. First, the difference between Fosamax Once Weekly 70 mg (Product B) and Fosamax 5mg and 10mg (A) is a difference in posology, a difference not hitherto the subject of decision. Second, Fosamax Once Weekly involves multiple differences between the reference product Fosamax 5mg and 10mg. It involves a change both in the strength of the dose and in its bioavailability.

Third, MSD contends that it sought and received authorisation for Fosamax Once Weekly following a full “stand-alone” application not an application under the hybrid abridged procedure; it did not rely merely upon bridging data.

Underlying those three contentions were more fundamental arguments. In writing and in oral submission, Mr Gordon QC, at first, suggested that the Court had left undecided the circumstances in which one product may be regarded as so distinct from another as to be entitled to an additional data protection period. In developing these submissions Mr Gordon QC contended that the proviso to Article 10(1)(a)(iii) was not concerned with data protection.

In reply Mr Gordon QC adopted a more striking pose. He contended that the Court in Novartis and APS was wrong. The real point of the instant case, so he said, was revealed in the discussion as to whether a statutory route, under the Directive, could be discerned for the grant of a marketing authorisation for product C, essentially similar to product B but not to product A. This discussion, so Mr Gordon QC contended, revealed either the error or at least the inadequacy of the Court’s rulings in Novartis and APS. On a proper construction, the generic companies cannot rely either on Article 10(1)(a)(iii) or on the proviso. The absence of any statutory route for authorisation of product C, essentially similar to product B but not to product A, demonstrates and confirms that the European Court of Justice has not hitherto sufficiently analysed the difficulties which arise where product C is not essentially similar to product A. That failure of analysis itself underlines and emphasises a failure properly to strike the balance between protecting the interest of the innovators and avoiding unnecessary testing.

MSD’s arguments should be tested against the following principles, derived from the Directive and the Court’s jurisprudence:-

1.

The primary objective of the Directive is to safeguard public health (see the Second Recital of the Directive and e.g. Novartis judgment at paragraph 30).

2.

Article 10, as interpreted by the Court, provides a complete code as to the circumstances in which an applicant may cross-refer to data relied upon in support of a previous authorisation (it will be necessary to discuss this second principle in greater detail, see paragraphs 67, and 73-77 of this judgment).

3.

The identity of the applicant for authorisation is not a feature of the provisions of Article 10. It is irrelevant whether the applicant is an innovator which holds marketing authorisation for the original product or its development or a generic company which seeks authorisation (see the wording of the Directive at Articles 8 and 10).

4.

Cross-reference to data relied upon in support of the authorisation of a product authorised for at least six or ten years or its development is permissible where product C is essentially similar to product A (Generics) or to product B (Novartis and APS).

5.

Product B is a development or line extension of product A if the differences between product B and product A are expressly identified in the proviso or “generally entail” or “generally imply” the difference in question between product A and product B (see Novartis at paragraph 66 of the judgment and APS at paragraph 26).

6.

The objective of ensuring that innovative firms are not placed at a disadvantage, identified in Recitals 3 and 9 of the Directive, is achieved by providing protection for a period of not less than six to ten years, a protection which is additional to that which is afforded by the domestic and Community laws of intellectual property and the additional supplementary protection afforded by Council Regulation 1768/92/EEC (Generics judgment paragraphs 73-76).

7.

The expense and difficulty in producing and testing a product which is a development of the original authorised product is no ground for permitting a further period of data protection for the developed product (see Generics at paragraphs 46 to 48).

Mr Gordon contends that the instant case is distinguishable because none of the previous cases concerned a difference in dose schedule. I shall assume that that is correct although examination of the facts in Generics, particularly paragraph 26 and footnote 15 of the Advocate General’s opinion, show that, in developing new therapeutic indications, Glaxo changed the dose schedules.

To my mind the principles, particularly that which I have identified at principle 5, in relation to the interpretation of the proviso, explained in Novartis and APS, apply with equal force in the instant case.

In Novartis a change in route of administration generally entailed a difference in bioavailability. In APS a change in route of administration normally entailed a change of pharmaceutical form (a pill differs from a suppository in pharmaceutical form). In the instant case a change in the dose or strength of the dose will generally entail a change in the timing or regularity when the dose should be taken. Indeed, Mr Sales argued that the reference to different doses in the proviso to Article 10(1)(a)(iii) could refer specifically to posology. If Fosamax 10mg is taken daily it may be said that it is taken in a different dose to that which is taken weekly (in the form of Fosamax 70mg). This led to controversy, since Mr Gordon pointed out that posology finds specific reference in Article 8(1)(f) and point 5.7 of Article 11 but not in the proviso. I am prepared to assume that posology is not specifically referred to in the proviso. But it does not matter. It seems to me clear that a change in dose, even if “dose” is given the restricted meaning of strength, falls implicitly within the proviso. A change in dose generally entails or normally entails a change in posology. Mr. Gordon protested that there was no evidence of this. But even a judge must, occasionally, take his medicine. I asked Mr Gordon, and through him, those behind him, to give me an example of a medicinal product a change in the strength of which does not lead to a change in the frequency with which it is administered. He could give me none.

The conclusion is supported by both the May 2001 and February 2004 Notice to Applicants in relation to the Procedures for Marketing Authorisations which give guidance on the hybrid abridged procedure under the proviso. This advice is relied upon by both the Advocate General and the Court in Novartis and APS. Both specifically refer to differences in posology in Appendix IV.

It is also of significance that the difficulty and expense of developing Fosamax 70mg from Fosamax 5mg or 10mg is not an element upon which MSD relies. It does not contend that it was any more difficult or expensive to develop Fosamax Once Weekly than the new therapeutic indications referred to in Generics. The Court made it clear that expense and difficulty of development is no warrant for extending a further period of protection (see my principle 7). Even the Advocate General in Generics, who relied upon that feature in reaching the conclusion that, when a new therapeutic indication had been developed a further period of protection should be afforded, rejected a further period in relation to new routes of administration and dosage schedules (see second part at paragraph 57 of his opinion).

Thirdly, it would undermine the purpose of Article 10 to provide a further period of protection because an innovating company has changed the frequency with which the dose should be taken merely by increasing its strength. It would be so easy for such a company to claim further protection merely by doubling the dose and thereby halving the frequency at which it should be taken. This case, in that respect, is far weaker than the case of the innovating company in the three previous decisions reached by the European Court of Justice. I conclude that same reasoning applied by the Court in Novartis and APS applies with equal or even greater force to the instant case.

Mr Gordon’s second point of distinction is that Fosamax Once Weekly differs not only in the strength of the active substance but also in its posology, difference in bioavailability and an “improved tolerability and compliance profile”.

In Novartis the Court referred to more than one difference in its concluding paragraph 55, answering the second part of the third question. I repeat the quotation:-

“(an application may be made under the proviso) … provided that the medicinal product in respect of which marketing authorisation is sought is essentially similar to the authorised medicinal product, unless one or more of the differences set out in the proviso apply, as the case may be”(my emphasis).

Mr Gordon contended that that passage referred only to the question of the scope of the proviso, whether essential similarity was a pre-requisite for the use of the proviso. The submission is in my view incorrect. The structure of the judgment in Novartis shows that the essential question of the right of cross-referral, at the very heart of the dispute, was determined by the meaning of the proviso interpreted in the light of the objectives of the Directive. Thus, when the Court went on to consider the right of the competent authority to cross-refer to data in its possession it did so by referring back to the question of the scope of the proviso. There is simply no basis for seeking to sever the question of cross-referral from the question of the scope of the proviso. That is why the Court chose a particular order in which to answer the questions. The scope of the proviso determined the right of cross-referral. There is no distinction to be drawn between the right of the competent authority and the right of an applicant to cross-refer. That right, which exists irrespective of the identity of the person seeking to invoke it, is determined by the scope of the proviso. ( see principles 2 and 3 )

In any event, the very process of implication of differences which are not expressly identified in the proviso demonstrates that the proviso applies in relation to more than one difference. If a difference in route of administration implies a difference in pharmaceutical form then generally product B will have more than one difference from the reference product A. If product A is to be taken by mouth and product B by another route product B will differ not only in relation to route of administration but also in pharmaceutical form and probably bioavailability. Yet the fact that product B has more than one difference does not lead to the conclusion that an applicant for marketing authorisation must provide all the data specified in Article 8 as opposed to merely bridging data, covering, for example, the difference in route of administration and pharmaceutical form. Nor is there any warrant for requiring a generic company seeking to market product C, identical to product B, to adduce any fresh data at all. If product C is identical to product B, which has been authorised despite the fact that it is different in more than one respect to product A, there can be no safety or efficacy reasons for requiring more data.

It seems to me that on a plain reading of Novartis and, as a matter of principle, the fact that Fosamax 70mg differs in more than one respect to Fosamax 5mg and 10mg does not take it outwith the proviso.

The third ground of distinction on which Mr Gordon relies is that the application for Fosamax Once Weekly was not made under the hybrid abridged procedure pursuant to the proviso but was a full application. He points out that the defendant accepted that the stand-alone application was validly made. He relied upon the fact that the defendant, as late as July 2003, asserted that an application under the proviso could not be accepted in relation to Fosamax, since it differed not only in strength but also in posology (see letter dated 9 July 2003). This letter was, however, written before the decision in APS. So too was the announcement of the change of policy dated 26 May 2004 which suggested that whilst a generic company did not need to produce its own bridging data it should make its application pursuant to the proviso.

In my view, the fact that the application for Fosamax Once Weekly was made under the full procedure is immaterial. Firstly, as the claimant accepted in its amended statement of grounds (paragraph 47) and in a letter dated 4 February 2000, the need to make a full application had nothing to do with the terms of Article 10(1)(a) but was because of the need for later recognition in other Member States where it would be necessary to demonstrate that the dossier data was identical in all those Member States. Secondly, it would be absurd if MSD could cloak itself with a further period of 10 years protection merely by the form of the application. All it would have to do, and did, as the correspondence makes clear (see e.g. letter dated April 2000), was to photocopy pages of data which had earlier been submitted. Indeed, it only had to pay the fees for a hybrid abridged application. Thirdly, the Advocate General and the Court’s reasoning in APS demonstrates that the procedure pursuant to which authorisation of developed product B is sought is not determinative (see paragraph 26 of the Court’s judgment and paragraph 75 in the Advocate General’s opinion).

As a matter of principle, it would be wrong if an innovator could acquire a further period of protection merely because the competent authority had sanctioned a full as opposed to a hybrid abridged application. Genericcompanies have no say in the manner in which authorisation is sought for a developed product B. The method with which the innovator chose to make its application should not and cannot affect a generic company’s rights. Those rights are determined by the Directive and the jurisprudence of the Court, explaining that Directive.

Underlying MSD’s particular attempts to draw distinctions between the instant case and the previous jurisprudence lay its attempt to distinguish between the scope of the proviso and the period of protection afforded by Article 10. It seems to me impossible to draw any conceptual divide between the scope of Article 10 and the proviso and the right of a generic company to cross-refer to data relied upon in support of a previous authorisation, whether for a period of not less than six to ten years or a shorter period.

Article 10 makes no reference to the identity of the applicant. The right of a generic company to cross-refer to data is the same right exercisable by the competent authority to which the High Court and the European Court of Justice referred in Novartis in questions one and two. Indeed, it is noteworthy, in answering that question, that the Court happily elided the identification of “an applicant for marketing authorisation for product C” (at paragraphs 65 to 66) with “the competent authority of a Member State” (see paragraph 67). There is no other statutory provision which confers the right to cross-refer. Nowhere, other than in Article 10, will one find any reference to the right to cross-refer. Thus it is Article 10 which reconciles the objectives identified in the 2^nd, 3^rd, 9^th and 10^th Recitals.

The right to cross-refer stems from the notion of one medicinal product being “essentially similar” to another. Public health is safeguarded by ensuring that (with the exception of recognition in a detailed scientific bibliography) a medicinal product is only authorised without reliance on further data if it is essentially similar to a product which has already been authorised.

The absence of any need for further data, in circumstances where product C is essentially similar to A or B, also achieves the objective identified in Recital 10 of avoiding unnecessarily repetitive tests on humans or animals. That objective was referred to as one of the principal objectives at paragraph 71 of the Court’s judgment in Generics.

Thus the only justification for requiring generic companies to produce further data, would be to extend the period of protection for an innovator. But innovators are protected by the requirement that the product has been authorised for not less than six or ten years, in addition to other systems under domestic and Community law (see paragraph 73 of the Court’s judgment in Generics).

In my judgment, MSD’s submissions either impede or are inconsistent with the objectives of the Directive to which I referred. Firstly, the suggestion that the generic companies should be compelled to produce their own data, from their own tests, in relation to a product which is identical to Fosamax Once Weekly 70mg has nothing whatever to do with safety or efficacy. It has not been suggested to the contrary. Secondly, the purpose of MSD is to extend the protection for data it produced beyond the protection afforded by Article 10. It should be recalled that Article 10 is only a supporting means of protecting the development work of MSD. In the front line is protection afforded by patent law supported by the additional period introduced by EEC Regulation No 1768/92 to which the Advocate General referred in paragraph 53 of his opinion in Generics. MSD lost the protection it held when Teva and other generic companies successfully brought revocation proceedings in the United Kingdom Patents Court in respect of two patents for Alendronate in January 2003. The European patent was revoked on 19 August 2004.

There is no sensible basis for affording MSD any additional period of protection unless it can establish that the developed product B, Fosamax Once Weekly 70mg, falls outwith the proviso to Article 10(1)(a)(iii). For the reasons I have given MSD has failed in that attempt.

Thirdly, MSD’s contentions are wholly inconsistent with the objective of avoiding unnecessary testing. Since it is agreed that the product generic companies wish to develop is identical to Fosamax Once Weekly, any further data the generic companies were required to produce would involve unnecessary testing.

The essential submission, advanced in reply, as to why the submissions of MSD do not offend the objectives of the Directive, started with the somewhat unpromising confession that Mr Gordon had only recognised the force of the argument during the course of Mr Sales’ submissions. Perhaps appreciating that his attempts to exclude Fosamax Once Weekly from the provisions of the proviso might not succeed, Mr Gordon QC was compelled to contend that the Court in Novartis and APS was wrong. Such a submission is laudable in its intellectual honesty. Its honesty can be appreciated in the nature of the submission advanced. The submission was that generic companies had no right to cross-refer to data if product C was essentially similar to product B but not to product A as there was no provision within Article 10 permitting them to do so. The generic companies’ product C was not essentially similar to Fosamax 5mg and 10mg (product A) so they could not rely upon Article 10(1)(a)(iii).

Nor could the generic companies rely upon the essential similarity of product C to product B (Fosamax Once Weekly 70mg). Fosamax Once Weekly is not a medicinal product authorised within the Community for not less than six to ten years. It thus can find no place as a reference product in Article 10(1)(a)(iii). Nor is the proviso of any assistance because product C is not intended for a different therapeutic use or to be administered by different routes or in different doses from product B. product C must obtain a route for authorisation within Article 10(1)(a)(iii) or not at all.

Accordingly, submits Mr Gordon, the fatal gap he has identified within the legislation merely confirms his original submissions. Whilst the generic company can rely upon previous data if product C is essentially similar to product A it cannot do so where it lacks that essential similarity to product A. Hence, Novartis and APS were wrongly decided because in neither of those cases was product C essentially similar to product A. The point is apparent in the incorrect analysis by the Advocate General and the Court in APS as to the procedural route by which generic companies can seek authorisation. The wording of Article 10(1)(a)(iii) and the proviso is inconsistent with their solution.

I reject these submissions. If Mr Gordon was correct then the only circumstance in which a generic company could make an application for authorisation under the abridged procedure, without relying upon its own fresh data, would be where its product C was essentially similar to product A. This is exactly the same submission that Novartis advanced, in the alternative (see Advocate General’s opinion at paragraph 44 of Novartis). The submission in the instant case is open to the same objection. Firstly, it would deprive the proviso of effect, save in relation to its reference to a different therapeutic use since a difference in administration or in dose will inevitably deprive product C of essential similarity to product A. Further, it does nothing to achieve the primary purpose of the Directive. Indeed, it is wholly inconsistent with it. If product C is essentially similar to product B, which has already been authorised, there is no further safety or efficacy objective to be achieved. Product C is no less safe or efficacious because it is essentially similar to B rather than A. The only purpose to be achieved is to give a further period of protection to innovators. But there is no greater reason why such protection should be afforded to MSD, which has increased the strength of Fosamax and reduced the frequency of its intake, than to Bristol-Myers Squibb which developed different uses for Captopril.

It seems to me that these arguments also offend the essential objectives of the Directive. The answer was given by both the Advocate General and the Court in APS. In APS the Advocate General referred to one or other form of the reference product:-

“An applicant is still required to show both essential similarity to one or another form of the reference product and that the product in question has been authorised within the Community for not less than 6 or 10 years.”

(Paragraph 84 of the Opinion).

It is entirely consistent with the objectives of the Directive to construe the reference to a medicinal product which has been authorised within the Community for not less than six or ten years as being a reference to not only the original reference product A but also its development product B provided only that that development falls within the proviso. The fact that neither the Advocate General nor the Court devoted substantial space to the problem is not evidence of any je m’en fichiste attitude but rather because, applying a purposive approach, it does not matter whether the generic companies’ application is regarded as being an application under Article 10(1)(a)(iii) or under the proviso, provided it is clear that the generic company does not have to adduce further tests. It is noteworthy that the language of the Advocate General “there are good reasons against requiring …” (paragraph 83) and of the Court “an application … may be made” (paragraph 30) does not suggest any compulsion to proceed under Article 10(1)(a)(iii). If the application was made pursuant to the proviso the competent authority could indicate that no further tests are appropriate. It really does not matter by what route the application is made so long as it is plain that the generic companies do not, in violation of the purpose of the 10^th recital, have to undertake further tests when their product is essentially similar to a product which has already received authorisation.

Mr Gordon also suggested that if the route which the generic companies must adopt was pursuant to Article 10(1)(a)(iii) there was no adequate definition of what was meant by a development of product A. But the Court in Novartis made plain what was meant by a development in paragraph 60. A development of product A is that which differs only in one or more respects identified in the proviso, as interpreted by the Court in Novartis and APS (see paragraphs 55 and 60 of the judgment in Novartis). The products in Novartis and in APS were the development of the products which had been authorised for six or ten years because they had features either expressly or impliedly identified in the proviso.

For those reasons I reject the submission of Mr Gordon that a reference is necessary because the Court overlooked the difficulties exposed by the wording of Article 10(1)(a)(iii). In my view it did not overlook such difficulties; it interpreted those provisions in a manner designed to achieve the objectives of the Directive.

In my view the principles enunciated by the Court in Generics, Novartis and APS are clearly applicable in this case and admit of no uncertainty. Fosamax Once Weekly 70mg is a development of Fosamax 5mg and 10mg. It falls within the proviso to Article 10(1)(a)(iii) no less because it is taken once weekly in the light of its increase in strength than if its therapeutic indication or its bioavailability or its pharmaceutical form had changed. It would be absurd if, in distinction to the innovators in the previous cases, MSD could acquire a greater period of protection merely by increasing the strength of the active ingredient and thus removing the need to take the product daily.

True it is that the Court has not specifically considered a change in posology before. But it is for the European Court of Justice to interpret the Treaty and the relevant principles by which it is to be interpreted. It is then for the domestic Court to apply those principles to the particular case. In reality, Mr Gordon has identified no new issue of principle; his arguments have merely served to underline how the application of those principles achieves the objectives of the Directive and precludes any need for yet a further reference. On three occasions, now, the Court has enunciated the basis upon which a generic company may seek authorisation in reliance on data used to support the previous authorisation of a product authorised for a period of six or ten years and its development. This is no occasion for a fourth consideration of those principles.