Merck & Co., Inc v Arrow Pharmaceuticals Limited - [2002] APO 13

OFFICIAL NOTICE
DECISION OF A DEPUTY COMMISSIONER OF PATENTS
Patent: No. 535944 in the name of Merck & Co., Inc

Title: Hypocholestermic fermentation products from Aspergillus.

Action: Request under s.70 for an extension of the term of the patent, and opposition thereto under s.75 by Arrow Pharmaceuticals Limited

Decision: Issued . 17 April 2002
Abstract
Opposition successful, and the extension refused.
The test for whether a pharmaceutical substance in substance falls within the scope of a claim is intended to be substantially the same as the test under s.102(2)(a) for allowability of amendments. In assessing whether something in substance fell within the scope of a claim, the courts have had regard to the principles of infringement – but this is not a determinative test. Infringements arising from ‘the only reasonable use’ of a substance involve a temporal component arising from when the ‘only reasonable use’ is assessed – and if applied as the test for what in substance fell within the scope of a claim, would introduce a temporal element on the substance of a claim. The Distillers test is appropriate to assess the extent of the infringement required in order for a substance to in substance fall within the scope of a claim. In the present case, while human pharmaceutical use of Lovastatin may reasonably lead to infringement of the claims to its metabolite, an amendment to claim Lovastatin per se would not be allowable as such a claim would give rise to infringement with respect to Lovastatin however used.
The nature of the Australian Register of Therapeutic Goods (ARTG) discussed in detail.
The relationship between s.70(3)(a) and the ARTG discussed. In order for a request to comply with s.70(3)(a), the pharmaceutical substance forming the basis of the request for the extension of term must be included in that part of the ARTG that is publicly accessible – that is, available under reg. 46(2) of the TGA Act. Also, the pharmaceutical substance must also be included in the ARTG as an active ingredient – and, in particular, not merely as an impurity.

PATENTS ACT 1990
DECISION OF A DEPUTY COMMISSIONER OF PATENTS
Re: Patent No. 535944 in the name of Merck & Co., Inc, Request under s.70 for an extension of the term of the patent, and opposition thereto under s.75 by Arrow Pharmaceuticals Limited.
BACKGROUND

This matter concerns a request for an extension of the term of patent 535944 in the name of Merck & Co Inc (Merck), and an opposition thereto by Arrow Pharmaceuticals Limited (Arrow). The matter concerns a substance called Lovastatin, and its b–hydroxy metabolite. Lovastatin is entered on the Australian Register of Therapeutic Goods (ARTG) and is used for the control of cholesterol levels. Lovastatin is a pro-drug which, when administered, is metabolised into the b–hydroxy form that is the substance that directly interacts with the physiological system to produce the therapeutic effect. There is no dispute that such a pro-drug falls within the definition of a pharmaceutical substance.
The Sankyo patent
Normally the considerations under s.70 would not involve the consideration of any other application. However the facts of this case require a limited understanding of patent 532626 in the name of Sankyo Company Pty Ltd.
The Sankyo patent (532626) was applied for on 19 Feb 1980, with a priority date of 20 Feb 1979. The term of that patent expired on 19 Feb 2000. The patent claims a compound of a particular formula that Sankyo called Monacolin K – but which is now more generally referred to as Lovastatin. The patent describes obtaining the substance from micro-organisms of the genus Monascus. It does not disclose the b–hydroxy metabolite of Lovastatin. No extension of term has been sought for this patent.
The Merck patent
The current patent was applied for on 3 June 1980, with priority dates of 15 June 1979, 21 Sept 1979, and 23 Jan 1980. It describes substances of several formulae identified as compounds I to IV, which were obtained from the genus Aspergillus. [Both parties agreed that no issue arose from whether the compounds were obtained from Aspergillus or Monascus.] Compound I corresponds to Lovastatin; compound III is the b–hydroxy metabolite of Lovastatin. The patent indicates that compounds I and III are generated at the same time by the micro-organism, and are separated. There is no teaching that the compounds should be kept together. The specification as filed included 2 broad claims to substances. The first (claim 8) specified compounds I & II. Claim 9 specified compounds III & IV. It is noteworthy that compound I was identical to the compound claimed in the Sankyo patent (i.e. to Lovastatin.) The issues concerning the request for the extension of term revolve around compounds I and III.
In the first examination report of the Merck patent, the examiner objected that claim 8 (and 16) was prior claimed by claim 1 of the Sankyo patent. [The objection extended also to claims 20 and 28, which define a pharmaceutical composition based on the relevant compounds.] No other objection was taken. In response to the examination report, compound I was deleted from claim 8. Claim 16 (which was specific to compound I) was deleted. Similar amendments were made with respect to claims 20 and 28. The patent was subsequently granted on this basis – that is, with the chemical entity Lovastatin having been specifically excised from the claims of the Merck patent.
The request for an extension of term
On the first working day after the expiration of the Sankyo patent, Merck applied to Therapeutic Goods Administration (TGA) for listing a good called Mevacor on the ARTG. Listing occurred some 4 weeks later. The printout from the ARTG indicates one active ingredient – Lovastatin – which is the entity that was excised from the claims of the Merck patent. The printout from the ARTG does not explicitly identify the b–hydroxy metabolite of formula III, the claims to which form the basis of the request to extend the term of the patent.
Subsequent to ARTG listing, Merck applied for an extension of the term of the patent. The Commissioner’s consideration of the application under s.72 was favourable in the first instance, and the request was advertised for opposition purposes. Arrow opposed the request. Following the usual service of evidence, the opposition was heard in Sydney on 31 January 2002. Merck was represented by Mr D Catterns (senior counsel), Ms K Howard (counsel), instructed by Mr J O’Connor & Ms S Irahni (patent attorneys). Arrow was represent by Dr A Bennett (senior counsel), instructed by Dr T Davies (patent attorney).
In support of their opposition, Arrow argued that the application for an extension of the term of the patent should be refused because:
·     Lovastatin did not in substance fall within the scope of any of the claims of the Merck patent;
·      The Merck patent when filed included claims to Lovastatin, which were deleted during prosecution in response to an objection of prior claiming. Accordingly, Merck was estopped from asserting that Lovastatin did fall within the scope; and
·      the b–hydroxy metabolite was not included on the ARTG.
In justification of the extension of term of their patent, Merck makes two separate, discrete arguments:
· The substance Lovastatin is clearly entered on the ARTG. Despite the deletion of any express reference to compounds of formula I from the claims, Lovastatin falls within the scope of the claim to the b–hydroxy metabolite of Lovastatin (i.e. the claims to the compound of formula III), because the only reasonable use of Lovastatin produces the b–hydroxy metabolite which would be an infringement of their patent.
· The b–hydroxy metabolite form of Lovastatin is in fact included on the ARTG. A “Certificate of Analysis” allegedly accompanied the request for registration. That certificate indicates that the metabolite was present in a batch of Mevacor in the amount of 0.2%. And (Merck asserts) that certificate is part of the ARTG.
Some peripheral observations are as follows:
·      The original term of the patent expired on 19 Feb 2000. Entry onto the ARTG occurred more than 10 years after the filing date. Accordingly, if an extension is granted the extended term will expire on 19 Feb 2005. It is worth noting that already 40% of the extended term has expired without any extension being granted. If, ultimately, the extension is refused the patentee will nevertheless have had the benefit of a de facto extension of the term for the duration of these proceedings;
·      While the intended purpose of the extension of term provisions was to compensate for losses arising from regulatory delays, the time period between application for ARTG listing and its listing was a mere 4 weeks;
·     The TGA extract indicates that the manufacturers of the goods are Merck Sharp and Dohme in Puerto Rico, and Merck Sharp and Dohme in Sydney. At the hearing I enquired as to whether Merck were actually manufacturing the goods in Australia. A Company representative present at the hearing advised that it was not – although there was an intention to do so at some date in the future.
I make these observations in passing to particularly note that while these issues might not be completely aligned to the rationale for the scheme to extend the term of a pharmaceutical patent, they in fact have no direct bearing on what I need to decide.
EXTENSION BASED ON LOVASTATIN BEING INCLUDED ON THE ARTG
s.70(2) of the Patents Act provides:
(2) Either or both of the following conditions must be satisfied:
(a) One or more pharmaceutical substances per se must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification;
(b) One or more pharmaceutical substances when produced by a process that involves the use of recombinant DNA technology, must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification.
As the current invention does not involve the use of recombinant DNA technology, only subparagraph (a) is relevant.
Merck’s primary argument in support of their extension is that a good containing the substance Lovastatin is clearly entered on the ARTG, and that Lovastatin falls within the scope of the claims to the b–hydroxy metabolite of Lovastatin. There is no suggestion that b–hydroxy metabolite of Lovastatin is Lovastatin. Rather, Merck argues that the only reasonable use of Lovastatin is for a pharmaceutical use; that upon use Lovastatin is metabolised into the b–hydroxy form; that such use constitutes infringement of claims to the metabolite; and therefore Lovastatin must fall within the scope of claims to the metabolite. They particularly rely on Beecham Group Ltd v Bristol Laboratories Ltd 1B IPR 665 to support the proposition that the use of Lovastatin would be an infringement of the claims to the metabolite.
I am prepared to assume, for the purpose of this decision, that human pharmaceutical use of Lovastatin generates the b–hydroxy metabolite, and that such use would constitute an infringement of claims to the b–hydroxy metabolite in the Merck patent. The question is whether this infringement necessarily means that Lovastatin in substance falls within the scope of those claims. In addressing this issue, I am particularly mindful that the Patents Act does not per se define what constitutes infringement. Rather, the rights given by a patent are specified in s.13(1), which provides:
Subject to this Act, a patent gives the patentee the exclusive rights, during the term of the patent, to exploit the invention and to authorise another person to exploit the invention.
and the dictionary defines exploit as:
"exploit", in relation to an invention, includes:
(a)where the invention is a product – make, hire, sell or otherwise dispose of the product, offer to make, sell, hire or otherwise dispose of it, use or import it, or keep it for the purpose of doing any of those things; or
(b)where the invention is a method or process—use the method or process or do any act mentioned in paragraph (a) in respect of a product resulting from such use;
This shows that infringement rights are broader than the express terms of a claim. Thus, for example, the rights in respect of a product include inter alia a right to use the product; and the rights in respect of a method or process include inter alia rights to hire, sell, and use the product of that method or process. Accordingly, the argument that the b–hydroxy metabolite of Lovastatin in substance falls within the scope of the claims to Lovastatin on the basis that the only reasonable use of the metabolite would give rise to infringement, requires careful consideration.
The requirement that the pharmaceutical substance must ‘in substance fall within the scope of the claims’ was first introduced into the Patents Act when the 1952 Act was amended in 1989 to introduce a scheme for extending the term of pharmaceutical patents. The phrase was included in amendment (4) in the amendments to the Patents Amendment Bill 1988 that were moved by the Government. In the Explanatory Memorandum accompanying those amendments, paragraph 8 states:
"… section 90(1)(a), which sets out the necessary relationship between the pharmaceutical substance and the patent concerned, is to be replaced by a criterion corresponding closely to the criteria employed elsewhere in the Patents Act in relation to the allowability of new or amended claims in a specification."
With the introduction of the 1990 Act, and the commencement of the current regime for extending the term of such patents, I can find no evident intent to depart from this intention. Accordingly, I consider it appropriate to assess compliance with the requirement for ‘in substance fall within the scope of the claims’ employing the criteria applied to the allowability of amendments. It follows that if, for the purpose of s.70(2), Lovastatin in substance falls within the scope of the claims to the b–hydroxy metabolite, it should be possible to make an amendment to the patent to insert a new claim directed to Lovastatin per se without offending against the requirements of s.102(2)(a).
When dealing with the issue of whether an amendment falls within the scope of the claims before amendment [i.e. the criterion of s.102(2)(a)], courts have had regard to the principles of infringement. In AMP Incorporated v Commissioner of Patents (1974) 3 ALR 283, the High Court stated at page 289:
"Without expressing a concluded view upon this question I turn to the second submission, namely, that the amended claim would in substance fall within the scope of the claims before amendment (s 78(2)). When the purpose of the claim is borne in mind it becomes clear that the limits imposed in s 78(2) were intended to have the effect that, as a result of the amendment, nothing should become an infringement of the patent which would not have been an infringement of the patent before the amendment. The exception to this is where the amendment is only for the purpose of correcting a clerical error or an obvious mistake (s 78(3)).
Although it is useful to have regard to the language of previous legislation and to the language of the United Kingdom legislation, one must primarily have regard to the language of s.78 itself. The amending claim must in substance fall within the scope of the claims before the amendment. The reference to substance imports the kind of test which is appropriate to a consideration of the question whether or not a particular act or particular article is an infringement of the patent. Although each claim is a definition, it is well recognised that an interpretation of the claim in the light of the state of the art may and probably will disclose that there are in it what may be described as essential as distinct from inessential integers. The substance of the claim must be discovered in order to determine whether a particular product is an infringement. However at the same time as one does this one must bear in mind that one is interpreting a claim in the nature of a definition and the ordinary rules of interpretation apply to it. The substance must appear from the language itself interpreted in accordance with ordinary rules of interpretation with a knowledge of the prior state of the art to which the claim relates.” (emphasis added)
The second paragraph quoted above provides the basis for Merck’s argument that because use of Lovastatin would infringe the claims to the b–hydroxy metabolite, it in substance falls within the scope of the claims to the metabolite.
The difficulty for Merck is that the infringement they rely upon comes about by the use (or intended use) of Lovastatin, not Lovastatin per se. The chemical entity Lovastatin is not embraced by the chemical formula in the claims – including allowing for routine variation of substituents that a person skilled in the art might have made. Infringement only comes about as a result of the use of Lovastatin – which use results in the formation of the b–hydroxy metabolite that is the subject of the claims. While Merck have argued that the only reasonable use of Lovastatin is a human pharmaceutical use, the fact remains that use of Lovastatin in any other manner, or possession for a purpose other than such use, would not constitute an infringement of the claims to the metabolite.
Accordingly, the first paragraph quoted above is particularly relevant. It is a restatement of the so-called Distillers test that was given in The Distillers Co. Ld.'s Application (1953) 70 RPC 221 at page 223 – viz:
"would the amendment make anything an infringement which would not have been an infringement before the amendment?"
While Merck has argued that 'since Lovastatin is an infringement of the claim to the metabolite, it must be in substance within the scope of that claim', the Distillers test puts the question the other way around, giving focus on the basis or extent of infringement required.
In the present case, infringement of the claims to the b–hydroxy metabolite by Lovastatin will only occur when there is pharmaceutical use of Lovastatin such that it is metabolised. Use (or intended use) of Lovastatin for any other purpose will not lead to the metabolite, and therefore will not infringe the claims to the metabolite. Accordingly an amendment to insert a claim to Lovastatin per se would make something an infringement that would not otherwise have been an infringement of the patent (i.e. Lovastatin per se, not being used pharmaceutically), contrary to the requirements of the Distillers test. It follows that Lovastatin per se is not in substance within the scope of the claims to the b–hydroxy metabolite – and that therefore the requirements of s.70(2) have not been met.
There is a further difficulty with asserting that a substance falls within the scope of a claim based on the only reasonable use of a substance leading to infringement. For perhaps most infringement scenarios, an action that constitutes an infringement at one time during the term of a patent would be an infringement at any time during the term of a patent. Yet the type of test for infringement that arises in Beecham v Bristol, and in the contributory infringement provisions of s.119, involve a temporal element – to establish infringement, the test of ‘only reasonable use’ must necessarily be applied as at the date of the alleged infringement. In the case of a new substance, it is quite common for new uses to be subsequently discovered – as is recognised by that body of patent law dealing with ‘new use of a known substance’. Consequently, while at one date in the life of a patent there might be only one reasonable use of a substance, it cannot generally be said that there will only ever be one reasonable use. Assessing what in substance falls within the scope of a claim on the basis of an ‘only reasonable use’ infringement argument has the consequence of rendering this test time-dependant. The effect would be that matter in substance falls within the scope of the claim during the early life of the patent, but when a different use is discovered at a later date that matter would no longer fall within the scope of the claim. This consequence is completely at odds with the normal understanding of claim interpretation. It highlights the significance of the comments of the High Court in AMP v Commissioner as quoted above, and illustrates the fallacious basis of the ‘only reasonable use’ infringement argument vis-à-vis determining the issue by the application of the Distillers test.

The opponent also submitted that the patentee was estopped from arguing that Lovastatin is within the scope of the claims to the b–hydroxy metabolite, on the basis of file wrapper estoppel. Gummow J discussed this estoppel at some length in Prestige v Dart 19 IPR 275. The particular issue in the present case is that the patent when filed included claims to Lovastatin per se, unlimited to any particular use or environment. During examination, the examiner raised an objection of prior claim based on the Sankyo patent. In response, the claims were amended to remove any explicit reference to Lovastatin in the claims. That is, the express monopoly rights in respect of Lovastatin per se were given up. Given my findings above, I do not need to form a concluded view on the issue of file wrapper estoppel in this case. However in my view it would be appropriate to have regard to the deletion that occurred – see s.116 of the Patents Act. In particular, I think that it would be inappropriate to assess such overtly deleted monopoly rights as in fact continuing by reason of what is ‘in substance’ within the scope of other claims – absent clear intention. In the present case there is no such intention evident – only a present desire to construe the claims in this manner such that there is a basis to extend the term of the patent.
In conclusion, it is plain that Lovastatin is not expressly within the scope of any of the claims to the metabolite. The use of Lovastatin may constitute an infringement of the claims to the metabolite. However that fact does not itself establish that the substance in substance falls within the scope of those claims. Applying the Distillers test, it is clear that Lovastatin per se does not in substance fall within the scope of the claims, as the monopoly rights to Lovastatin per se would extend beyond the basis for Lovastatin being an infringement of the claims to the metabolite. Accordingly the requirements of s.70(2) have not been met as between the listing of goods containing Lovastatin on the ARTG, and the claims to the b–hydroxy metabolite.
EXTENSION BASED ON THE b–HYDROXY METABOLITE OF LOVASTATIN BEING ENTERED ON THE ARTG
The second argument of Merck is that the b–hydroxy metabolite of Lovastatin is in fact included on the ARTG. In their request they provide a copy of a Certificate of Analysis which they assert was filed with TGA when they sought listing on the ARTG. This certificate is an analysis of a batch made in 1999. It shows the presence of Lovastatin in the amount of 95 – 102%, with the b–hydroxy metabolite being present at 0.2%. Interestingly, it does not indicate the presence of any of the three other ‘impurities’ listed in the British Pharmacopoeia – even though the British Pharmacopoeia indicates that they should be present in amounts greater than the b–hydroxy metabolite. Merck argue that the ARTG includes all the documents submitted to TGA in support of the listing, that therefore this Certificate of Analysis is on the Register – and consequently the b–hydroxy metabolite is included on the Register.
The opponent argued that the b–hydroxy metabolite was not listed on the Register. They provided copies from the web-site of TGA that purportedly indicate that the Register was maintained electronically. They provided in evidence a copy of a letter from TGA stating that the b–hydroxy metabolite was not listed on the ARTG. Accordingly they argue that there is no basis for granting the extension of the term.
At the commencement of the hearing I raised an interpretation issue concerning s.70(3)(a). That provision states:
(3) Both of the following conditions must be satisfied in relation to at least one of those pharmaceutical substances:
(a) goods containing, or consisting of, the substance must be included in the Australian Register of Therapeutic Goods;
(b) ….
A simple interpretation of this phrase is that the requirement is met so long as a good entered on the ARTG contains the substance – irrespective of whether the particular substance is identified on the ARTG. Neither party agreed with this interpretation. Rather the debate at the hearing was predicated on requiring the substance being included somewhere on the ARTG, with unresolved differences as to what actually constituted the ARTG.
It was clear at the hearing that there was significant disagreement about what constituted the ARTG. Further, the arguments of the parties was based upon their respective understandings of what constituted the ARTG, with both parties expressing some difficulties in providing ‘definitive’ statements about the ARTG and what was on it. Indeed, even as late as at the hearing Merck tabled some correspondence from TGA about this issue – including a letter to Merck from a lawyer within TGA that was apparently written sometime on the day of the hearing. Given the significance of this issue, I indicated at the end of the hearing that I would be making my own enquiries as to what is the ARTG – and, if appropriate, I would provide material for the parties to further comment on. In the event, I concluded that there was nothing for the parties to further comment on, and both parties agreed to my proceeding with this decision without inviting further submissions.
The Australian Register of Therapeutic Goods (ARTG)
Under the Patents Act, Trade Marks Act, and Designs Act, the Register has historically been made up of a special set of books, with pages added when something is entered in the Register. In contrast, it would seem that the ARTG has never existed as a distinct entity.
The ARTG is provided for by s.17 of the Therapeutic Goods Act 1989. Subsections (1) to (3) provide:
(1)The secretary is to cause to be maintained a register, to be known as the Australian Register of Therapeutic Goods, for the purpose of compiling information in relation to, and providing for evaluation of, therapeutic goods for use in humans.
(2)Subject to subsection (3), the Register is to be kept in such form as the Secretary determines.
(3)The Register is to contain two parts, one relating to goods to be known as registered goods and another relating to goods to be known as listed goods.
I note that while (2) refers to a determination by the Secretary, I understand that as of the date of the hearing the Secretary had made no ‘formal’ determination. I therefore understand that the form of the ARTG is that form in which it exists by way of operational procedures of the Therapeutic Goods Administration. However the Register is not open to public inspection (see immediately below), which presumably creates some of the difficulties the parties have experienced when trying to establish the form of the ARTG.
Section 32(1), (2), (4) and (6) of the TGA Act relevantly provide:
(1)The Register is not open for public inspection, but a person in relation to whom therapeutic goods are registered or listed may make a written request to the Secretary for a copy of the entry in the Register in relation to the goods.
(2)If the person makes such a request, the Secretary must send to the person a copy of so much (if any) of that entry as is contained in any computer database maintained by the Department for purposes connected with administration of this Act (other than any part of that Entry that was supplied in confidence by another person).
(4)          Where:
(a)     the person in relation to whom therapeutic goods are registered or listed has asked the Secretary to vary product information included in the entry in the Register that relates to the goods; and
(b)     ….
the secretary must vary the entry in accordance with the request.
(6)In this section, product information, in relation to therapeutic goods, means information relating to the safe and effective use of the goods, including information regarding the usefulness and limitations of the goods.
From (1) it may be noted that the Register itself is not open to public inspection. From (2) it is apparent that whatever constitutes an entry in the Register, at least some of it is likely to be held on a computer database. And from (4), it is clear that an entry in the Register may contain product information as defined in (6).
Section 61(6) of the TGA Act provides for the release to a person of ‘therapeutic goods information’ of a kind identified in the regulations. Therapeutic goods information is defined in 61(1) as:
Therapeutic goods information means information in relation to therapeutic goods that came into the possession of the Department in connection with the performance of the Department’s functions (including functions relating to the Mutual Recognition Agreement).
Regulation 46(2)(e) relevantly provides for the release of therapeutic goods information:
(2)For the purposes of subsection 61(6) of the Act, the Secretary may release to a person, on application by the person, therapeutic goods information in respect of an entry in the Register, being therapeutic goods information of the following kinds:
(a)whether the goods are included in the Register, …
….
(e)if the goods are medicines, therapeutic devices that contain medicines or therapeutic devices that are presented in a pharmaceutical form:
(i)the quantity of goods to be in the primary pack; and
(ii)the entry relating to the goods in the Poisons Standard; and
(iii)the indications for the goods; and
(iv)the dosage form of the goods and their physical appearance; and
(v)the names and quantities of therapeutically active substances in the goods; and
(vi)the presence or absence of any specific excipient in the goods; and
(vii)the routes of administration of the goods; and
(viii)the type of container in which the goods are to be picked
The web site for TGA includes a page entitled “TGA General Information” It contains the statement
“The ARTG is a computer database of information about therapeutic goods for human use approved for supply in, or exported from, Australia.”
The submissions of the parties about the content of the ARTG appeared to be based on the assumption (and perhaps an inference from this statement on the web page) that the printout obtained from TGA was a printout of the entire ARTG entry (at least in so far as it was held on a computer database), rather than the limited therapeutics goods information as provided for in s.61(6) / Reg. 46(2). However, it is my understanding that the content of such printouts is in fact dependent upon whether the request is made under s.32 (which can only be made by the person in relation to whom the goods have been registered or listed) or under s.61(6) / Reg. 46(2).
Applications for an extension of term are usually accompanied by a printout from a computer database from TGA. The printout supplied by Merck in support of the application contains more information than that on the printout supplied by Arrow. Specifically, the following information is only provided on the Merck printout:
·      Identity of the principal manufacturer, and other manufacturers;
·      List of excipient ingredients (6 components, with quantities specified); and
·      List of Proprietary Ingredients (none listed).
with the summary at the end of the report indicating, on the Merck copy, 7 ingredients; and on the Arrow copy, 1 ingredient. Although the reports were generated 10 months apart, the differences are indicative of the different bases for obtaining a printout from the ARTG. To the extent that a person in relation to whom therapeutic goods are registered or listed can obtain a printout under s.32 that contains more information than that available under s.61(6) / Reg. 46(2), that extra information is clearly not publicly available from TGA.
My understanding is that an application to TGA for Registration or Listing is typically accompanied by much information. For Registration, the information can be quite voluminous. This information is generally held on a paper file. When a decision is made to register or list a good, I understand the actions taken are:
·     Certain information concerning the good is entered into a computer database;
·     A letter of approval is issued. This letter will inter alia indicate any conditions associated with the Registration or listing. Non-standard conditions may refer to particular documents on the file. A copy of that letter is placed on the file; and
·     A certificate of Registration (or Listing) is issued, with a copy being placed on the file.
My understanding is that TGA considers that the Register entry consists of at least:
·     the information in the database;
·     the letter of approval;
·     the certificate of Registration or Listing; and
·     any document in the file that is referenced by the conditions in the letter of approval.
It is also my understanding that TGA considers that not all the documents on the file form part of the entry in the Register. The primary role of the ARTG is to regulate the supply of pharmaceuticals. It is an offence to market goods that are not listed in the ARTG. In the course of considering an application for Registration (in particular) much information is supplied to TGA. That information may well include information about goods that are not otherwise on the Register. To treat that information as being on the ARTG potentially has the effect of negating the principal purpose of the Register. The implication of this is that mere supply of documents to TGA in the course of obtaining Registration or Listing is not sufficient for that information to be considered as being included on the ARTG.
I also understand that when special conditions are applied, those conditions will not be noted in the printout supplied under s.61(6). In particular, while a special condition may reference documents on the file, the documents that relate to those conditions are not publicly available. In general, they are treated as documents supplied on a commercial-in-confidence basis and would not normally be released under the FOI Act.
Finally, Merck assert that they filed a Certificate of Analysis with their application to TGA. I understand that whether a Certificate of Analysis was filed with TGA is information that is not publicly available. Accordingly, the opponent or any other third party is unable to verify from TGA whether or not that Certificate was in fact filed.
Is the b–hydroxy metabolite on the ARTG?
As indicated above, there was much argument about whether the b–hydroxy metabolite appeared on the ARTG. Merck relied upon the certificate of analysis to assert that it was included. Arrow relied on the absence of any reference to it on the computer printout, and a letter from TGA asserting non-inclusion on the ARTG, to assert that it was not included.
In fact the issue is determined on a different basis. The TGA Act relies on standards to establish the chemical composition of goods on the ARTG – see s.25(1)(f), and 26(1)(f) of the TGA Act. That is, when a good is entered on the Register the composition of the good is taken to be that specified by the standard. That Act defines that 'standard' as:
"standard", in relation to therapeutic goods, means a standard that:
(a)is specified in an order under section 10 that is applicable to the goods; or
(b)if no such order is applicable to the goods but the goods are the subject of a monograph in:
(i)in the case of goods for use in humans - the British Pharmacopoeia; or
(ii)in the case of goods for use in animals - the British Pharmacopoeia (Veterinary);
is constituted by the statements in that monograph;
As may be noted, the default standard is the British Pharmacopoeia. While s.10 allows the Minister, by order published in the Gazette, to:
'determine that matters specified in the order constitute a standard for therapeutic goods or a class of therapeutic goods identified in the order (whether or not those goods are the subject of a monograph in the British Pharmacopoeia or the British Pharmacopoeia (Veterinary))',
I am not aware of any such relevant determination that applies to Lovastatin or the metabolite, and there is nothing else to suggest that the composition of Lovastatin is intended to be anything different to that given in the Pharmacopoeia.
The entry in the Pharmacopoeia for Lovastatin identifies the primary component, and four ‘impurities’ – one of which is the b–hydroxy metabolite of Lovastatin. It follows that a reference to the b–hydroxy metabolite does exist on the ARTG, as part of the entry for Lovastatin. However, it is not present as a good in its own right, or as an active ingredient in a good. It exists on the ARTG expressly as an impurity present in a good on the register.
On this latter point, the patentee suggested at the hearing that the presence of 0.2% b–hydroxy metabolite was in fact a significant component. They referred to a number of research papers that indicated that the absorption of Lovastatin was fairly poor, whereas the absorption of the b–hydroxy metabolite was good – and suggested that therefore the 0.2% of b–hydroxy metabolite was not really an impurity but a significant active ingredient. I find this suggestion ingenuous. In evidence in the opposition are a number of documents/articles that refer to Mevacor. The product information sheet for Mevacor, produced by Merck, makes no reference to the presence of the b–hydroxy metabolite other than in the context of this being the metabolite formed after ingestion of Mevacor. Several articles discuss Mevacor and the b–hydroxy metabolite, but none suggest the presence of the b–hydroxy metabolite in the Mevacor. The only document (other than the Certificate of Analysis) that refers to the presence of the b–hydroxy metabolite in Mevacor, is the British Pharmacopoeia entry where it is identified as an impurity. There is also no evidence whatsoever suggesting an intention to ensure the presence of the b–hydroxy metabolite in the manufacture of Lovastatin – indeed the specification teaches separation of the two compounds when extracting them from Aspergillus. If Merck really thought the b–hydroxy metabolite was an active component, they had an obligation to disclose it as part of the application for listing. The absence of any specific reference to the b–hydroxy metabolite in the Listing is consistent only with the b–hydroxy metabolite being treated at all relevant times as an ‘impurity’ as per the British Pharmacopoeia.
Relationship between s70(3)(a) [Patents] and the ARTG.
Turning now to the interpretation of s70(3)(a) of the Patents Act, the difficulty that I need to resolve is the interaction with the TGA Act. I understand that the exact constitution of the ARTG has not been a significant issue in the administration of the ARTG. It arises in the present proceedings in the context of the interface between the TGA Act and the Patents Act. It would seem extremely unlikely that the amendment to the Patents Act, to introduce the scheme for extending the term of patents by way of reference to what is included in the ARTG, was intended to alter the operation of the TGA Act. This leads to the preferred interpretation of s70(3)(a) of the Patents Act being one that is consistent with the actual operation of the ARTG, and otherwise achieving the objects of the legislation vis-à-vis the grant of an extension of the term of a patent.
It seems to me that there are four possible interpretations of s.70(3)(a) of the Patents Act. Those possibilities, with associated ‘issues’ are as follows:
(1)To obtain an extension, identify a good on the ARTG. Separately analyse the good to identify the chemical entities it contains, having no regard to what is on the ARTG.
This possibility involves an interpretation of s70(3)(a) where the concept of ‘containing’ has no relationship to any information on the ARTG. If this interpretation were correct, a number of difficulties would arise:
·      Neither the patentee, nor any third party, would have any ready way of ascertaining whether there was another good on the ARTG that contained the relevant chemical entity. A patentee would seek an extension with uncertainty about what impurities might be present in other goods already on the ARTG. A potential opponent would have an impossible task of conducting a chemical analysis of an unknown number of goods entered on the ARTG in the off-chance that such a good might containing the relevant chemical entity as an impurity – with the time-frame for filing a notice of opposition being totally incompatible with such a task;
·      While chemical entities not listed on the ARTG may exist in goods listed on the ARTG, their existence may be batch-dependant, or time-dependant.
·      Any invention that had as its basis the identification of the active chemical entity in a homeopathic or similar good entered on the ARTG, would be excluded from the extension of term mechanism.
I raised this interpretation at the start of the hearing. Both parties positively rejected this interpretation. They argued that, whatever the correct interpretation of the provision, it was necessary to find a reference to the relevant chemical entity somewhere on the ARTG. I take the bipartisan rejection of this interpretation as being supportive of this interpretation is not the ‘plain meaning’ of s.70(3)(a).
(2)To obtain an extension, identify a good on the ARTG. Ascertain whether the chemical entity is mentioned somewhere in the information filed with TGA for ARTG Registration or Listing.
This interpretation formed the basis of the arguments of Merck – in the sense that they argued that the b–hydroxy metabolite was mentioned in the Certificate of Analysis which was filed with the application. Arrow supported this argument so long as the metabolite was mentioned on the ARTG – with there being disagreement about what constituted the ARTG.
This possibility involves an interpretation of s70(3)(a) where the concept of ‘containing’ has some relationship to information on the ARTG. If this interpretation were correct, a number of difficulties would arise:
·      Whether or not the entity is mentioned would appear to be at the whim of the patentee. For example, in the present case the British Pharmacopoeia identifies three ‘impurities’ other than the b–hydroxy metabolite – all of which are present in greater quantities than the b–hydroxy metabolite. Yet the Certificate of Analysis only mentions the b–hydroxy metabolite.
·      Third parties have no right of access to such documentation, and therefore no way of readily ascertaining the presence of such chemical entities in goods (other than for entities listed in the British Pharmacopoeia). Consequently, a potential opponent is disadvantaged through not being able to independently assess the basis of the extension of term; and
·      As for the first interpretation, neither party has certainty about what documentation may exist with respect to other goods on the ARTG that might reference the chemical entity.
While Merck argued strongly for this interpretation (arguing that the certificate of analysis was ‘on’ the ARTG), it seems to me that the level of certainty and transparency is little different to (1) above.
(3)To obtain an extension, identify a good on the ARTG. Ascertain whether the chemical entity is mentioned in the publicly available information about goods on the ARTG [i.e. under reg 46(2)], but in any capacity.
This interpretation has the advantages over (1) and (2) that the basis of the extension of term is a matter of public record. Consequently, the patentee can seek an extension of term with some confidence that the chemical entity is not contained in any other good on the ARTG. And third parties can reasonably assess the validity of the extension of term without having to resort to any regime of chemically testing goods.
The principle difficulty is whether mere mention of the chemical entity on the ARTG is sufficient. The purpose of the scheme for extending the term of patents for pharmaceutical substances, as noted in the Explanatory Memorandum and other secondary material, is to encourage the development of new pharmaceuticals. While it may be well known that chemical impurities are sometimes found to be very valuable chemical entities in their own right, one would expect that value to be recognised as ARTG inclusion based on that chemical entity as an active ingredient. It would seem quite contrary to the intention of the legislation to justify the extension of the patent on the basis of the entity being present on the ARTG as an impurity in a good.
(4)To obtain an extension, identify a good on the ARTG. Ascertain whether the chemical entity is mentioned in the publicly available information [i.e. available under reg 46(2)] about goods on the ARTG as an active ingredient.
I think that this interpretation is the only one consistent with the intention of the legislation, and functional in the sense of providing certainty for the patentee, and for third parties wishing to ascertain the basis of the extension.

Accordingly, I conclude that the correct interpretation of s.70(3)(a) is that the pharmaceutical substance forming the basis of the request for the extension of term must be included in that part of the ARTG that is publicly accessible – that is available under reg. 46(2) of the TGA Act. Also, it must be included in the ARTG as an active ingredient – and, in particular, not merely as an impurity. And the issue of whether or not the substance is mentioned on a Certificate of Analysis or any other documents filed with TGA by the applicant is not a relevant consideration.
In coming to this conclusion, I should note that I have some reservations about referring to ‘active ingredient’. For example, there are entities called ‘pro-drugs’ that are converted within the human body to the ‘active’ chemical entity. Indeed, Lovastatin is apparently a pro-drug. In my mind, such pro-drugs are clearly ‘active’ ingredients. On the other hand, I think chemical entities identified as impurities are clearly not listed as active ingredients. There are other entities likely to be mentioned in the publicly available information, such as excipients. I have doubts about whether entities like excipients would constitute active ingredients. But I’m also concerned about whether the criteria for identifying substances as active or excipient in the ARTG are fully compatible with patent law concepts of combinations and pharmaceutical substances. However the issue has not been debated before me and it is unnecessary to decide this issue at this time.
In the present case, the b–hydroxy metabolite is included on the publicly available part of the ARTG by way of the British Pharmacopoeia. Although Merck suggested that it was in fact an active ingredient, all documentary material is consistent with it being present only as an impurity, and there is no evidence whatsoever to show any intention to ensure its presence in Lovastatin. Consequently (and despite its presence in Lovastatin) I conclude that goods containing, or consisting of, the b–hydroxy metabolite are not included in the ARTG in the manner required by s.70(3)(a) – which means there is no basis to extend the term of the patent based on the presence of the b–hydroxy metabolite in Lovastatin.
CONCLUSION
Merck argued in support of their request for an extension of term on the basis of two discrete arguments. On the first argument, I have concluded that while the use of Lovastatin might constitute an infringement of the claims to the b–hydroxy metabolite, Lovastatin per se does not in substance fall within the scope of the claims to the metabolite. Therefore the requirements of s.70(2) have not been met with respect to Lovastatin. On the second argument, I have concluded that the b–hydroxy metabolite is present on the ARTG, but only as an impurity; that the requirements of s.70(3)(a) are not met if the relevant substance is only included in goods on the ARTG as an impurity. Therefore the requirements of s.70(3)(a) have not been met with respect to the b–hydroxy metabolite.
It follows that the criteria for granting an extension of term of this patent have not been met. I am also of the view that the deficiencies are not capable of rectification. Accordingly I refuse the request for an extension of term of the patent.
COSTS
In opposition matters before the Commissioner, costs would normally follow the event. I see no reason in the present matter to depart from this, and accordingly award costs against the patentee – Merck & Co Inc.
D Herald
Deputy Commissioner of Patents
17 April 2002
Patent attorneys for the patentee : Spruson & Ferguson, Sydney
Patent attorneys for the opponent : Allens Arthur Robinson, Sydney