Albany Molecular Research Inc v Alphapharm Pty Ltd
[2011] FCA 120
•18 February 2011
FEDERAL COURT OF AUSTRALIA
Albany Molecular Research Inc v Alphapharm Pty Ltd [2011] FCA 120
Citation: Albany Molecular Research Inc v Alphapharm Pty Ltd [2011] FCA 120 Parties: ALBANY MOLECULAR RESEARCH INC v ALPHAPHARM PTY LTD (ACN 002 359 739)
ALBANY MOLECULAR RESEARCH, INC v ARROW PHARMACEUTICALS PTY LTD (ACN 003 144 170), SIGMA PHARMACEUTICALS LIMITED (ACN 088 417 403), SIGMA PHARMACEUTICALS (AUSTRALIA) PTY LTD (ACN 004 118 594), SIGMA COMPANY LIMITED (ACN 004 132 923) and AVENTIS PHARMACEUTICALS, INC
File numbers: VID 219 of 2007
VID 883 of 2007Judge: JESSUP J Date of judgment: 18 February 2011 Catchwords: INTELLECTUAL PROPERTY – Patents – Whether patent invalid for want of novelty – Whether want of novelty established by disclosure of specific compound in terms in prior art base – Whether prior art needs to provide, for a skilled person working in the relevant area, information as to the means by which compound could practically and effectively be prepared – Whether prior art disclosed effective means – Whether further experiments and further information required in order to prepare compound – Whether further experiments and further information more than ordinary methods of trial and error
INTELLECTUAL PROPERTY – Patents – Whether patent invalid for lack of inventive step – Whether a skilled person faced with the problem of preparing the relevant compound in substantial purity would have been led, as a matter of routine, to try that which was invented under the patent in suit
INTELLECTUAL PROPERTY – Patents – Whether patent invalid due to inadequate description of invention or lack of clarity – Whether the expression “substantially pure” sufficiently clear
INTELLECTUAL PROPERTY – Patents – Amendment to specification – Whether amendment affected priority date – Whether matter claimed post amendment fell within scope of claims before amendment
INTELLECTUAL PROPERTY – Patents – Whether patent a manner of manufacture – Whether patent disclosed a new invention – Whether patent merely claimed a substance with desired attributes
INTELLECTUAL PROPERTY – Patents – Whether patent invalid for lack of sufficient description – Whether patent gave the best method known to the inventor of performing the invention
INTELLECTUAL PROPERTY – Patents – Whether patent invalid for lack of fair basis in the claims
INTELLECTUAL PROPERTY – Patents – Whether patent invalid due to false suggestion – Whether false suggestion made to commissioner – Whether false suggestion materially contributed to commissioner’s decision to grant patent, or was a material, inducing factor which led to grant
Legislation: Patents Act 1990 (Cth) ss 7, 18, 40, 114, 138
Patents Regulations 1991 (Cth) reg 3.14Cases cited: Abbott GMBH and Co KG v Apotex Pty Ltd (2010) 87 IPR 561
Acme Bedstead Co Ltd v Newlands Brothers Ltd (1937) 58 CLR 689
Aktiebolaget Hassle v Alphapharm Pty Ltd (2002) 212 CLR 411
Apotex Pty Ltd v Sanofi-Aventis (2009) 82 IPR 416
Eli Lilly and Co v Pfizer Overseas Pharmaceuticals (2005) 218 ALR 408
Hill v Evans (1862) 1A IPR 1, 6
H Lundbeck A/S v Alphapharm Pty Ltd (2009) 177 FCR 151
Inre O’Farrell (1988) 853 F 2d 894
Martin v Scribal Pty Ltd (1954) 92 CLR 17
Merck and Co Inc v Arrow Pharmaceuticals Ltd (2006) 154 FCR 31
NV Philips Gloelampenfabrieken v Mirabella International (1993) 44 FCR 239
Olin Corporation v Super Cartridge Co Pty Ltd (1977) 180 CLR 236
Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157
Pfizer Overseas Pharmaceuticals v Eli Lilly and Co (2005) 225 ALR 416
Ranbaxy Australia Pty Ltd v Warner-Lambert Co LLC (2008) 77 IPR 449
Smithkline Beecham plc’s (Paroxetine Methanesulfonate) Patent [2006] RPC 10
Van der Lely NV v Bamfords Ltd [1963] RPC 61
Welch Perrin and Co Pty Ltd v Worrel (1961) 106 CLR 588
Wellcome Foundation Limited v VR Laboratories (Aust) Pty Ltd (1981) 148 CLR 262Date of hearing: 29-30 November, 1-3, 6-10 December 2010 Place: Melbourne Division: GENERAL DIVISION Category: Catchwords Number of paragraphs: 223 Counsel for the Applicant: Mr A Archibald QC with Mr A Ryan SC and Ms K Beattie Solicitor for the Applicant: Phillips Ormonde Fitzpatrick Counsel for the Respondent
in VID219 of 2007:Mr D Catterns QC with Mr N Murray Solicitor for the Respondent in VID 219 of 2007: Mallesons Stephen Jaques Counsel for the Respondents in VID 883 of 2007: Mr D Shavin QC with Ms H Rofe Solicitor for the Respondents in VID 883 of 2007 Corrs Chambers Westgarth
IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY
GENERAL DIVISION
VID 219 of 2007
BETWEEN: ALBANY MOLECULAR RESEARCH, INC
Applicant/Cross-ClaimantAND: ALPHAPHARM PTY LTD (ACN 002 359 739)
Respondent/Cross Claimant
JUDGE:
JESSUP J
DATE OF ORDER:
18 FEBRUARY 2011
WHERE MADE:
MELBOURNE
THE COURT ORDERS THAT:
1.The parties file and serve memoranda setting out their submissions as to the orders which should be made to reflect the court’s reasons published this day, together with any submissions as to costs, in accordance with the following timetable:
(a)the respondent/cross-claimant within seven days;
(b)the applicant/cross-respondent within a further seven days;
(c)the respondent/cross-claimant in reply (if necessary), within a further seven days.
Note:Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.
The text of entered orders can be located using Federal Law Search on the Court’s website.
IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY
GENERAL DIVISION
VID 883 of 2007
BETWEEN: ALBANY MOLECULAR RESEARCH INC
Applicant/First Cross-RespondentAVENTIS PHARMACEUTICALS, INC
Second Cross-RespondentAND: ARROW PHARMACEUTICALS PTY LTD (ACN 003 144 170)
First Respondent/First Cross-ClaimantSIGMA PHARMACEUTICALS LIMITED (ACN 088 417 403)
Third Respondent/Third Cross-ClaimantSIGMA PHARMACEUTICALS (AUSTRALIA) PTY LTD (ACN 004 118 594)
Fourth Respondent/Fourth Cross-ClaimantSIGMA COMPANY LIMITED (ACN 004 132 923)
Fifth Respondent/Fifth Cross-Claimant
JUDGE:
JESSUP J
DATE OF ORDER:
18 February 2011
WHERE MADE:
MELBOURNE
THE COURT ORDERS THAT:
1.The parties file and serve memoranda setting out their submissions as to the orders which should be made to reflect the court’s reasons published this day, together with any submissions as to costs, in accordance with the following timetable:
(a)the respondents/cross-claimants within seven days;
(b)the applicant/cross-respondents within a further seven days;
(c)the respondents/cross-claimants in reply (if necessary), within a further seven days.
Note:Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.
The text of entered orders can be located using Federal Law Search on the Court’s website.
IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY
GENERAL DIVISION
VID 219 of 2007
BETWEEN: ALBANY MOLECULAR RESEARCH, INC
Applicant/Cross-ClaimantAND: ALPHAPHARM PTY LTD
Respondent/Cross ClaimantIN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY
GENERAL DIVISION
VID 883 of 2007
BETWEEN: ALBANY MOLECULAR RESEARCH INC
Applicant/First Cross-RespondentAVENTIS PHARMACEUTICALS, INC
Second Cross-RespondentAND: ARROW PHARMACEUTICALS PTY LTD (ACN 003 144 170)
First Respondent/First Cross-ClaimantSIGMA PHARMACEUTICALS LIMITED (ACN 088 417 403)
Third Respondent/Third Cross-ClaimantSIGMA PHARMACEUTICALS (AUSTRALIA) PTY LTD (ACN 004 118 594)
Fourth Respondent/Fourth Cross-ClaimantSIGMA COMPANY LIMITED (ACN 004 132 923)
Fifth Respondent/Fifth Cross-ClaimantJUDGE:
JESSUP J
DATE:
18 FEBRUARY 2011
PLACE:
MELBOURNE
REASONS FOR JUDGMENT
There are two proceedings presently before the court. Each relates to Australian Patent No 699799 titled “Piperidine derivatives and process for their production”, the priority date of which is 24 June 1993. The applicant in each proceeding is the patentee, Albany Molecular Research, Inc (“AMR”). In proceeding VID 219 of 2007, AMR alleges that the respondent, Alphapharm Pty Ltd (“Alphapharm”) has infringed the patent by making, selling and otherwise dealing in the pharmaceutical product known as “Xergic”, the active ingredient in which is fexofenadine hydrochloride, the hydrochloride salt of one of the compounds covered by the patent. In proceeding VID 883 of 2007, AMR alleges that the respondents, Arrow Pharmaceuticals Pty Ltd, Sigma Pharmaceuticals Ltd, Sigma Pharmaceuticals (Australia) Pty Ltd and Sigma Company Ltd (together, “the Sigma respondents”) infringed the same claims of the patent by making, selling and otherwise dealing in various pharmaceutical products – many of which carry a label which includes the term “fexo” – of which the active ingredient is fexofenadine hydrochloride.
The respondents in both proceedings contend that the patent in suit is, and has at all material times been, invalid for want of novelty, lack of an inventive step, inadequate description of the invention, failure to define the invention, lack of clarity and succinctness in the claims, lack of fair basis in the claims and false suggestion. They also allege that the invention is not a manner of manufacture within the meaning of s 6 of the Statute of Monopolies, and that, if the patent is otherwise valid, the priority date is 15 September 1998. Likewise, in both proceedings, there are cross-claims for revocation, in which the grounds correspond broadly to the points taken by way of defence. In VID 219 of 2007, Alphapharm is the cross-claimant and AMR is the cross-respondent. In VID 883 of 2007, the Sigma respondents are the cross-claimants, AMR is the first cross-respondent and Aventis Pharmaceuticals, Inc is the second cross-respondent, joined as exclusive licensee of the patent in suit.
According to the “background” section of the complete specification to the patent in suit, the compound 1-(p-tert-butylphenyl)-4-[4'-(α-hydroxydiphenylmethyl)-1'-piperidinyl]-butanol, known as “terfenadine”, is a non‑sedating anti‑histamine. It is said that terfenadine has been linked to potentially fatal abnormal heart rhythms in some patients with liver disease, or who also take the anti‑fungal drug ketoconazole or the antibiotic erythromycin. It is said that, in animal and human metabolic studies, terfenadine was shown to undergo high first‑pass effect, resulting in readily measurable plasma concentrations of the major metabolite 4-[4-[4-(hydroxydiphenylmethyl)-l-piperidinyl]-l-hydroxybutyl]-α,α-dimethylphenylacetic acid, also known as terfenadine carboxylic acid metabolite.
To the extent presently relevant, the invention under the patent in suit is said to relate to substantially pure piperidine derivative compounds of the formula –
wherein
Rl is hydrogen or hydroxy;
R2 is hydrogen;
or R1 and R2 taken together form a second bond between the carbon atoms bearing R1 and R2;
R3 is -COOH or -COOR4 ;
R4 is an alkyl with 1 to 6 carbon atoms;
A, B, and D are the substituents of their rings, each of which may be different or the same, and are selected from the group consisting of hydrogen, halogens, alkyl, hydroxy, or alkoxy, or other substituents.– or a salt thereof. It is said that “these compounds are useful in pharmaceutical compositions, particularly as antihistamines, antiallergy agents, and bronchodilators”. When all the substituents marked as A, B and D are hydrogen, when R1 is a hydroxyl group, when R2 is hydrogen and when R3 is –COOH, the compound is 4-[4-[4-(hydroxydiphenylmethyl)-l-piperidinyl]-l-hydroxybutyl]-α,α-dimethylphenylacetic acid, the synthetic form of terfenadine carboxylic acid metabolite. This compound is known by the non‑proprietary name “fexofenadine”.
AMR sues on claims 1, 6, 7, 8, 9, 10 and 11 of the patent in suit. They are also the claims on which the respondents’ validity case is based. To the extent presently relevant, Claim 1 is expressed in broad terms corresponding to those set out in para 4 above. Claim 6 is based on Claim 1, and R1 must be a hydroxyl group. Claim 7 is based on Claim 6, and R3 must be –COOH. Claim 8 is for “[a] pharmaceutical composition comprising … a pharmaceutical carrier and the substantially pure piperidine derivative compound according to any one of claims 1 to 7”. Claim 9 is for “[a] unit dosage formulation comprising a pharmaceutical composition according to claim 8, wherein said substantially pure piperidine derivative compound is present in an effective antiallergic amount”. Claim 10 is for “[a] method of treating allergic reactions in a patient comprising administering to the patient said pharmaceutical composition according to claim 8 in an effective amount”. Claim 11 is for a compound, in accordance with Claim 1, with reference to certain preparatory examples contained in the patent.
THE CARR PATENTS
Central to an understanding of the patent in suit, and to the respondents’ novelty cases, is a series of previous USA patents of which the inventors were Carr and others. The two that are presently important are Nos 4,254,129 and 4,254,130. It is convenient to refer to them as “Carr 129” and “Carr 130” respectively. It is common ground that Carr 129 ostensibly disclosed piperidine derivative compounds of the classes covered by the patent in suit, including fexofenadine, but the inventor under the patent in suit asserted that, if the methods disclosed in Carr 129 were followed, substantially pure compounds would not be the result. Indeed, that circumstance lay at the core of the justification for – and the claimed inventiveness of – the present invention. These are aspects to which I shall have to give substantial attention in my consideration of the respondents’ novelty cases, but my present concern is merely to identify the compounds disclosed in the Carr patents (and the Australian equivalent of Carr 129, to which I shall turn) and how they relate to the compounds claimed in the patent in suit.
Carr 129 was publicly available in Australia on or about 3 March 1981. It relates to:
… novel substituted piperidine derivatives. More particularly, this invention relates to substituted phenyl 4-substituted piperidinoalkanol derivatives which are useful as antihistamines, anti allergy agents and bronchodilators and to methods of making and using the same.
The compounds disclosed in Carr 129 are of the following formula:
wherein R1 is hydrogen or hydroxy; R2 is hydrogen; or R1 and R2 taken together form a second bond between the carbon atoms bearing R1 and R2; n is an integer of from 1 to 5; R3 is -CH3, -CH2OH, -COOH or -COOalkyl wherein the alkyl moiety has from 1 to 6 carbon atoms and is straight or branched; and each of A and B is hydrogen or hydroxy; with the provisos that at least one of A or B is hydrogen and one of A or B is other than hydrogen when R3 is -CH3; and pharmaceutically acceptable salts thereof.
Among the many examples of the compounds covered by the invention in Carr 129 is 4-[4-[4-(hydroxydiphenylmethyl)-l-piperidinyl]-l-hydroxybutyl]-α,α-dimethylbenzeneacetic acid, that is to say, fexofenadine. That compound exists when, in the formula diagrammatically depicted above, R1, R2, A and B are hydrogen, n is 3 and R3 is -COOH. Claim 1 in Carr 129 is for a compound of the formula set out above, wherein –
… R1 represents hydrogen or hydroxy; R2 represents hydrogen; or R1 and R2 taken together form a second bond between the carbon atoms bearing R1 and R2; n is an integer of from 1 to 5; R3 is -COOH or -COOalkyl wherein the alkyl moiety has from 1 to 6 carbon atoms and is straight or branched; and each of A and B is hydrogen or hydroxy; with the proviso that at least one of A or B is hydrogen; and pharmaceutically acceptable salts and individual optical isomers thereof.
As explained above, fexofenadine is also disclosed by this claim.
There is an Australian equivalent of Carr 129: Patent No 531146 (“Carr 146”), published on 16 October 1980. Although the terminology is not precisely the same in all instances, it is sufficient to say that this patent also ostensibly discloses compounds of the same classes as does Carr 129, including fexofenadine. The one special aspect of Carr 146 which may be presently material is that Claim 2 therein, which otherwise follows the terms of Claim 1 in Carr 129, claims a “substantially pure” compound of the formula set out above, and has the following additional wording:
·R3 may also be –CH3 or –CH2OH;
·There is an extra proviso that, when R3 is -CH3, “one of A or B is other than hydrogen”;
·“pharmaceutically acceptable bioprecurors [sic]” are also included.
In Carr 129, a series of examples (to which I shall return in some detail below) ostensibly sets out how to prepare 4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1- hydroxybutyl]-α,α-dimethylbenzeneacetic acid, that is, fexofenadine. The first example describes a synthesis for the intermediate compound ethyl 4-(4-chloro-1-oxobutyl)-α,α-dimethylphenylacetate. That is a compound in which the chlorobutyl ketone is substituted on the dimethylated benzene ring at the para position (ie at 6 o’clock relative to the existing substituent). However, according to the patent in suit:
Applicant has discovered that the preparation of ethyl 4-(4-chloro-1-oxobutyl)-α,α-dimethylphenylacetate by reaction of 4-chlorobutyryl chloride, aluminum chloride, and ethyl α,α-dimethylphenylacetate in carbon disulfide, as described in Example 1 of U.S. Patent Nos. 4,254,130 and 4,285,958 provides an inseparable mixture of monosubstituted aromatic regioisomers of the formula:
wherein the chlorobutyryl substituent is attached at either of the three aromatic carbons which are meta or para to the dimethylacetate substituent. These regioisomers are not separable by standard techniques of thin layer chromatography, or column chromatography, and low field proton nuclear magnetic resonance spectroscopy is inconclusive in identifying the product of this reaction as a mixture.
It is stated that it was known in the art that a monoalkyl substituent on a benzene ring is ortho (ie at 2 and 10 o’clock relative to the existing substituent), para directing in electrophilic aromatic substitution reactions such as a Friedel-Crafts reaction. Thus, it would be expected that the Friedel-Crafts reaction of 4-chlorobutyryl chloride with ethyl α,α-dimethylphenylacetate would yield predominantly a para-substituted product, because of the electron donating, para-directing character of the dimethylalkyl substituent combined with the steric hindrance associated with substitution at the ortho positions. However, the inventor states that, in practice, “the inductive electronic withdrawing effect of the carboxylic ester of ethyl α,α-dimethylphenylacetate counteracts the expected alkyl electron donating effect, resulting in no significant directing effect for the aromatic substitution reaction.” It is said that “a statistical mixture of meta to para regioisomers results, with the two meta positions predominating.” (The meta positions are those which are situated at 4 and 8 o’clock relative to the existing substituent.)
In the patent in suit, the inventor claims that the mixture of regioisomers will endure if the product of this example in Carr 129, as achieved in practice, is further reacted in an attempt to derive any of the compounds claimed by that patent. Referring to the mixture –
– it is said:
Although the … mixture of regioisomers can be analyzed by HPLC experiments, a practical separation to obtain gram quantities of substantially pure regioisomers has not been achieved. Each mixture (including the first), would be expected to contain 33% of the para isomer and 67% of the meta isomer. Since these components are inseparable, it has not been possible to obtain either of the regioisomers in each mixture in substantially pure form.
Thus it is said that Carr 129 (and implicitly Carr 146) does not in fact give substantially pure para‑substituted compounds.
In the patent in suit, there is a claim for various methods to prepare these piperidine derivative compounds as such, ie to prepare compounds which are unembarrassed by the presence of meta regiosomers. However, it is not the method claim in the patent which is of principal concern in this proceeding. The inventor claims to have invented the compounds themselves, including, most relevantly for present purposes, fexofenadine. It is asserted in the specification that the relevant prior art, contrary to ostensible indications, did not disclose these compounds in substantial purity because any compound made in accordance with it would be “an inseparable mixture” of the para and meta regiosomers. By contrast, the inventor here makes a claim to “substantially pure” piperidine derivative compounds.
THE RESPONDENTS’ PRIMARY CASE ON NOVELTY: LUNDBECK AND APOTEX
The patent in suit will be susceptible to revocation if the invention to which it refers, so far as claimed in any claim, is not novel by comparison with the prior art base as it existed before the priority date (Patents Act 1990 (Cth) ss 138(3)(b) and 18(1)(b)(i)). The invention is to be taken to be novel in this sense unless it is not novel in the light of information of the kind referred to in s 7(1) of the Patents Act. The respondents say that Carr 129 and Carr 146 are part of the prior art base in the sense that they were documents that were publicly available at the priority date. That is undoubtedly so. The contentious question is whether the invention claimed in the patent in suit is not novel in the light of the information set out in Carr 129 and Carr 146.
The respondents say that the invention claimed in Claim 1 of the patent in suit is for a class of compounds that were in terms disclosed in the Carr patents, considered as documents. According to the respondents, in the case of a compound patent, once the specific compound is found to have been disclosed in terms in the prior art base, the conclusion follows that the invention is not novel and, therefore, not patentable. For this proposition they rely upon the recent judgments of the Full Court in H Lundbeck A/S v Alphapharm Pty Ltd (2009) 177 FCR 151 and Apotex Pty Ltd v Sanofi-Aventis (2009) 82 IPR 416. In Lundbeck, Bennett J said (with the assent of Middleton J) (177 FCR at 192 [180]): “Where the prior publication discloses exactly what is claimed, there is anticipation”. That was not the situation in Lundbeck itself, but it was in Apotex, in which case the court held that the disclosure of a particular enantiomer in the prior art anticipated a later patent which claimed that enantiomer, notwithstanding that a compound consisting only of that enantiomer had never, apparently, been made.
In the present case, it was accepted by AMR that the Carr patents disclosed the relevant compounds, including fexofenadine, in terms. It was also ultimately accepted by AMR that that disclosure was of compounds that were pure as such. In the case of fexofenadine, looking only at the Carr patents as documents, substantially pure fexofenadine was disclosed. However, it was submitted that the Carr patents were not anticipatory because they did not provide, for a skilled person working in the relevant area, information as to the means by which the compounds to which they referred could practically and effectively be brought into existence. In this respect AMR relied on Hill v Evans (1862) 1A IPR 1, 6, Acme Bedstead Co Ltd v Newlands Brothers Ltd (1937) 58 CLR 689, 707 and Olin Corporation v Super Cartridge Co Pty Ltd (1977) 180 CLR 236, 260-261. The respondents’ primary case on novelty is that, even if this be so, it does not matter: if the prior art discloses exactly what is claimed in the patent in suit, the invention will not be novel. This first point is substantially one of law and, if correct, would require me to uphold the respondents’ novelty cases without more.
AMR commences with the words of Lord Westbury in Hill v Evans (1A IPR at 6-7):
The question then is, what must be the nature of the antecedent statement? I apprehend that the principle is correctly thus expressed: the antecedent statement must be such that a person of ordinary knowledge of the subject would at once perceive, understand, and be able practically to apply the discovery without the necessity of making further experiments and gaining further information before the invention can be made useful. If something remains to be ascertained which is necessary for the useful application of the discovery, that affords sufficient room for another valid patent. By the words of the statute of James, it is necessary for the validity of a patent that the invention should not have been known or used at the time. These words are held to mean “not publicly known or publicly used”. What amounts to public knowledge or public user is still to be ascertained. One of the means of imparting knowledge to the public is the publication of a book, or the recording of a specification of a patent. If, therefore, in disproving that allegation which is involved in every patent, that the invention was not previously known, appeal be made to an antecedently-published book or specification, the question is, what is the nature and extent of the information thus acquired which is necessary to disprove the novelty of the subsequent patent? There is not, I think, any other general answer that can be given to this question than this: that the information as to the alleged invention given by the prior publication must, for the purposes of practical utility, be equal to that given by the subsequent patent. The invention must be shewn to have been before made known. Whatever, therefore, is essential to the invention must be read out of the prior publication. If specific details are necessary for the practical working and real utility of the alleged invention, they must be found substantially in the prior publication.
….
Upon principle, therefore, I conclude that the prior knowledge of an invention to avoid a patent must be knowledge equal to that required to be given by a specification, namely, such knowledge as will enable the public to perceive the very discovery, and to carry the invention into practical use.AMR relies also on what Dixon J (in Acme Bedstead 58 CLR at 707) described as the “well-settled rule” –
… that a prior paper publication, giving information that does not become part of common knowledge, does not invalidate a subsequent patent unless it supplies enough information to enable a person of proper skill in the art to produce the mechanical device or appliance or carry out the process claimed in the later specification.
According to AMR, the principle to which Lord Westbury referred was endorsed by Stephen and Mason JJ (with the concurrence of Barwick CJ) in Olin. That case concerned a patent for a plastic shotgun cartridge, in which an earlier patent, described as “the Core patent”, had been cited as anticipatory. A plastic shotgun cartridge, and the general means by which it might be produced, had indeed been disclosed in the Core patent, but Stephen and Mason JJ held that the disclosure was not sufficiently developed to sustain the case of anticipation. Their Honours said (180 CLR at 260-261):
The Core patent did not disclose a process or means of producing the wanted article. Nor did it reveal any advance in knowledge or development on which the appellant's patent depended for its success. All that the Core patent did was to describe an article without describing the effective means by which it could be produced. To our mind this is not enough to justify the conclusion that the appellant's claim fails for want of novelty. So much at least emerges from the remarks of Lord Westbury L.C. in Hills v Evans:
“[in order to invalidate the subsequent patent] the antecedent statement must be such that a person of ordinary knowledge of the subject would at once perceive, understand, and be able practically to apply the discovery without the necessity of making further experiments and gaining further information before the invention can be made useful ... the information as to the alleged invention given by the prior publication must, for the purposes of practical utility, be equal to that given by the subsequent patent”.
In N. Guthridge Ltd. v Wilfley Ore Concentrator Syndicate Ltd., Griffith C.J. quoted Lord Westbury L.C. in Betts v Menzies, when his Lordship said that a prior patent "ought not to be held to be an anticipation of a subsequent discovery, unless you have ascertained that the antecedent specification discloses a practicable mode of producing the result which is the effect of the subsequent discovery". In the same case, Barton J, applied the observations of Lord Westbury L.C. in Hills v Evans and Betts v Menzies.
I should deal straight away with two submissions about Olin that were made on behalf of the Sigma respondents. They submitted first that the passage set out above was obiter, and that I am not, therefore, bound to follow it. Sitting as a single member of the Federal Court, I do not find this an attractive submission. It is true that, of a number of challenges to validity in Olin, their Honours in the High Court upheld the fair basis point, but they gave detailed consideration to the novelty aspect, explaining clearly the respects in which the Core patent would not have been anticipatory. If Olin stood alone and were directly applicable to the facts of the present case, clearly it would have to be followed. Secondly, the Sigma respondents submitted that the respects in which Olin had been followed or applied in this court related to aspects of validity other than novelty – obviousness and fair basis for instance. It was said that the principles set out by Mason and Stephen JJ on novelty had never been followed here. Whether or not this is so, the point is really no more than a development of the first one. For reasons to which I have referred, even in the absence of subsequent application in this court, I consider that the only course properly open to me would be to follow Olin, if it was not to be distinguished.
As it happens, the two recent Full Court judgments on which the respondents relied did deal with the principle for which Olin stands, and it is by reference to them, and to a proper understanding of them, that the legal dimension of the respondents’ novelty case must be approached.
The novelty point which arose in Lundbeck was whether a patent which claimed the (+) enantiomer of a compound had been anticipated by certain prior art information which disclosed a racemic mixture of (+) and (-) enantiomers (177 FCR 151 at [193]), or by other prior art information which disclosed the (-) enantiomer as such, but stopped short of characterising the (+) enantiomer, the preparation of which would, it was conceded, require the taking of another inventive step (at [205]). Because the prior art considered in Lundbeck stopped short of disclosing the very thing claimed under the patent in suit, the result of that case is not directly applicable to the facts of the present case.
In Lundbeck, however, Bennett J (with the assent of Middleton J) undertook a detailed examination of the law of novelty in respects which are, on any view, presently relevant. Her Honour said (177 FCR 151 at [173]):
The following general propositions emerge from the authorities:
•An invention is a piece of information (Merrell Dow Pharmaceuticals Inc v HN Norton & Company Ltd (1995) 33 IPR 1 at 8). It follows that a disclosure is the communication of information.
•Commonly the only question may be whether the prior publication describes the claimed invention with sufficient clarity (Bristol-Myers 97 FCR 524 at [67]).
•The disclosure is assessed by reference to the skilled addressee, a person of ordinary skill in the art.
•The question is whether the prior publication is sufficient to make the claimed invention apparent to the skilled addressee (Nicaro Holdings Pty Ltd v Martin Engineering Company (1990) 91 ALR 513 at 529).
•A prior publication does not invalidate a patent unless it supplies sufficient information to enable a person of ordinary skill to produce the product subsequently claimed (Acme Bedstead Company Ltd v Newlands Brothers Ltd (1937) 58 CLR 689 at 707). A specification is not to be read as in a vacuum but by the reader having at least the common knowledge of the art (Acme Bedstead 58 CLR at 701; Nicaro 91 ALR at 530).
•The requirement is that a person of ordinary knowledge of the relevant subject would be able practically to apply the prior published discovery without the necessity of making further experiments (Hill v Evans 1A IPR at 6-7).
•The further experiments do not include those that formed part of standard procedure or common general knowledge. They are experiments with a view to discovering something not disclosed (Van der Lely 1A IPR at 90).
•The further experiments do not mean ordinary methods of trial and error (Van der Lely 1A IPR at 90).
•If the alleged anticipation is to a process that produces the claimed product, it is not an anticipation if the process would not necessarily achieve the result claimed for it (Olin Corporation v Super Cartridge Company Pty Ltd (1977) 180 CLR 236 at 260-261).
•Something less than a full description of the invention allegedly anticipated may be sufficient to invalidate it for want of novelty (Nicaro 91 ALR at 529).
•Something less than a full description of an effective means by which the combination claimed in a patent may be produced may be sufficient to a reader having common general knowledge in the art (Nicaro 91 ALR at 531).
•A direction, recommendation or suggestion may be implicit in what is described (Bristol-Myers 97 FCR 524 at [67]).
•A disclosure that describes an effective means by which a claimed invention may be produced falls short of anticipation if it requires the exercise of inventive ingenuity or the taking of any inventive step (Nicaro 91 ALR at 531).
•Where the prior disclosure is to a broad chemical claim encompassing many compounds, there may not be anticipation in the absence of the skilled addressee understanding or perceiving a specific compound in the disclosure (Imperial Chemicals Industries Pty Ltd v Commissioner of Patents (2004) 213 ALR 399 at [64]-[65]). That is, there is no actual description of the particular compound to the skilled addressee; there is no relevant disclosure. There may be a distinction, albeit fine, between a “fleeting” or “paper” disclosure or the “intellectual content” of a disclosure on the one hand and a “disclosure for novelty purposes” or “enabling disclosure” on the other (Imperial Chemicals at [68]; University of Georgia Research Foundation v Biochem Pharma Inc (2000) 51 IPR 222, a decision of Dr Barker of the Patent Office described by Crennan J in Imperial Chemicals as a “sound account of the relevant distinctions between a ‘paper disclosure’ and an ‘enabling disclosure’ in the field of chemistry” (at 412)). It depends on what the skilled reader would understand.
Bennett J considered Hill v Evans and continued (at [178]-[183]):
178Care must be taken to distinguish between the tests for novelty and want of inventive step, in particular when looking to see what the prior art “teaches”. The concept of novelty in Australia involves a comparison between the invention as claimed in the claims of the patent and prior art information. Often, this must be determined by looking to prior publications which are to be read by the skilled addressee to determine what they disclose. Generally speaking, the consideration of what a prior publication “teaches”, especially when one talks of “teaching away” from the claimed invention, tends to be relevant to questions of obviousness and inventive step.
179As Lord Hoffmann said in Merrell Dow 33 IPR at 8, an invention is a piece of information and making matter available to the public therefore requires the communication of information. Whether or not such information has been communicated depends on the subject matter of the claim and the extent of the prior disclosure to the skilled addressee.
180Where the prior publication discloses exactly what is claimed, there is anticipation. This can be objectively determined and, apart from an understanding of terms of art, the evidence of the skilled addressee is not likely to be of much further assistance. However, this does not always occur and many of the authorities contain discussions of the extent to which a disclosure less than the entirety of the claim constitutes an anticipation of a product or a process to deprive the claimed invention of novelty.
181If the prior art discloses some but not all integers of a claimed patent to a product, such as a combination, there is anticipation if the skilled addressee would add the missing information as a matter of course and without the application of inventive ingenuity or undue experimentation (Nicaro 91 ALR at 530-531).
182It may be that the prior disclosure is of a method that produces the claimed product. If that method leads inexorably to the product, there is anticipation (General Tire 1A IPR at 138). If it may or may not result in the claimed product, there is no anticipation.
183It is these last two examples that, in Australia, could be said to be within a shorthand description of “enabling disclosure”. That is, the disclosure is not complete but it is sufficient to enable the skilled addressee, in the ordinary course and without invention, to add what is missing in the prior publication to obtain the claimed invention. The term “enabling disclosure” may also be apposite to disclosure to the skilled addressee of an asserted prior use: whether what the skilled addressee observes on inspection is sufficient to enable him or her to comprehend the complete invention (eg Insta Image Pty Ltd v KD Kanopy Australasia Pty Ltd (2008) 78 IPR 20; Jupiters [Jupiters Ltd v Neurizon Pty Ltd (2005) 222 ALR 155]), that is, whether it is sufficient to amount to a disclosure of the invention.
Bennett J then considered the speech of Lord Hoffman in Smithkline Beecham plc’s (Paroxetine Methanesulfonate) Patent [2006] RPC 10, and in particular the UK requirements of disclosure and “enablement” on a want of novelty case. Her Honour continued (at [189]-[190]):
For the purposes of disclosure, the prior art must disclose an invention which, if performed, would necessarily infringe the patent. Once the very subject matter of the invention has been disclosed, the person skilled in the art is assumed to be willing to make trial and error experiments to get it to work. For the purposes of disclosure, the disclosure is either of an invention which, if performed, would infringe the patent, or it is not. When Lord Hoffmann went on to say that, for the purposes of enablement, the question is no longer what the skilled person would think the disclosure meant but whether he was able to work the invention which the Court has held it to disclose, his Lordship was talking of what, in Australia, is covered by sufficiency of disclosure or description.
It follows that, where the prior publication is of the subsequently claimed invention, that is sufficient. Where the prior disclosure falls short of a complete disclosure, the question of the sufficiency of that disclosure arises. It is there that consideration must be given to the quality of a disclosure to the skilled addressee armed with common general knowledge. It is in that context that, in a limited fashion, questions of “enablement” can be said to arise. The use of that expression tends to cause confusion between anticipation and sufficiency. Rather, the Court, armed with the evidence of the skilled addressee as to terms of art and the nature and extent of the disclosure in the prior art document, must determine whether the prior disclosure is sufficient to enable the skilled addressee to perceive, understand and, where appropriate, apply the prior disclosure necessarily to obtain the invention.
In Apotex, the d-enantiomer of a compound was claimed. The Full Court held that the relevant prior art disclosed both that enantiomer and the l-enantiomer (82 IPR 416 at [77]). However, the prior art did not contain a “description of a process to obtain the enantiomers” (at [78]). This led the patentee to advance a submission identified by Bennett and Middleton JJ as follows (at [79]):
Sanofi points out that no process of resolution of the racemate into the constituent enantiomers, or of preparation of the individual enantiomers of the derivatives, is described and that successful separation of the enantiomers was by no means assured, even if the prior art patents contained a direction to do so. Sanofi also says that, until the enantiomers are prepared, it is not known which is the d-enantiomer and which is the l-enantiomer, nor their levels of activity and toxicity. In short, Sanofi submits that further discovery is needed before the invention claimed in the patent is obtained.
The patentee submitted (at [95]) that “to anticipate a claimed invention, a prior paper disclosure that does not itself clearly show the production of the claimed invention must inevitably result in the skilled addressee arriving at the claimed invention”. Bennett and Middleton JJ noted that the patentee also made the following submission (82 IPR 416 at [96]):
Sanofi submits that the requirement is for clear disclosure and that, in relation to the prior art patents, where:
(a) the patentee had not made the d-enantiomer of PCR 4099;
(b) the patentee did not show a method of producing it by resolving it; and(c)but it is possible that the skilled addressee, using known methods, might be able to make it.
there is no clear disclosure of the d-enantiomer in the prior art patents.
The position adopted by the party challenging the patent on novelty grounds, as recorded by their Honours, was as follows (at [103]):
Apotex accepts that a disclosure will not be sufficiently clear if it is necessary to supplement the disclosure by means of experiments or other sources of information in order to perceive the disclosure. However, that is not the same as accepting that, where there has been disclosure of a product, the method of producing that product must also be disclosed.
Bennett and Middleton JJ referred to the judgment of the Full Court in Lundbeck, and specifically to the statement by Bennett J that, where the prior publication disclosed exactly what was claimed, there was anticipation. Their Honours continued (at [106]):
In this case, there was a disclosure of the enantiomers in the prior art patents. Those enantiomers were not only described, they were also claimed. Thus there was a clear direction to the skilled reader to prepare the enantiomers and in addition, but not necessarily, it was made clear that such enantiomers were, or were likely to be, useful for the desired purpose. The primary judge was not in error in concluding that the prior art patents disclosed the enantiomers as part of the invention and as compounds predicted to have the beneficial qualities of the compounds exemplified. If his Honour were correct in his conclusion that the skilled reader would understand to separate the enantiomers and would know the methods to apply, and that the preparation of the d-enantiomer was routine and involved no inventive step, it is hard to see how his Honour erred in concluding that there had been disclosure of the d-enantiomer to the skilled addressee and that a claim to the d-enantiomer had been anticipated: Ranbaxy.
Their Honours rejected the proposition that there could be no anticipation unless the compound disclosed in the prior art had actually been made and the making of it there described. Their Honours continued (at [108]):
That is, Sanofi’s submissions are to the effect that it would be necessary for a prior publication, in every case, to set out the method of preparation, no matter how routine, and presumably the detailed methodology of each step taken in the preparative process. That cannot be correct.
In the present case, the respondents say that the compounds of interest are disclosed, as such, in Carr 129 and Carr 146 and that, in accordance with Lundbeck and Apotex, that circumstance is sufficient to constitute anticipation. They point out that, in Apotex, the prior art disclosed no process for the preparation of the d-enantiomer and that there was no evidence that the d‑enantiomer had ever been prepared. If the patentee could not, in that setting, resist a challenge to novelty, AMR could be in no better a position in the present case, whether or not the Carr patents contained a viable description of how substantially pure compounds could be made.
As I have indicated earlier, AMR’s case on novelty is based on Hill v Evans, Acme Bedstead and Olin. It says that only if the prior art discloses both the compound in question and an effective means for preparing it will that art be anticipatory. The only exception to the requirement for the disclosure of effective means is where the skilled worker would, from the information about the compound given in the prior art, readily appreciate, as a matter of routine, how to go about preparing it. If that condition is satisfied, or if the effective means of preparation are disclosed as such, it does not matter that the compound may never have been prepared in fact. It is said that Apotex was a case in which, from the identification of the compound in the prior art, it was readily apparent to the skilled worker how it might have been prepared, and that that could have been done as a matter of routine. It was, according to AMR, neither here nor there that no-one had ever in fact prepared the compound.
AMR submits that the Full Court in Lundbeck could not have intended to set aside the principles long-established since Hill v Evans, and adopted in Australia in Acme Bedstead and Olin. Those principles required the disclosure not only of the compound claimed to have been invented under the patent in suit, but also an effective means of preparing that compound. AMR points out that Bennett J discussed Hill v Evans at some length, without any suggestion that it was not good law in Australia: and neither would such a suggestion have been easily sustained, in the light of the acceptance of the speech of Lord Westbury in Olin and elsewhere. As to Olin itself, it was pointed out that Bennett J referred to this important authority only in relation to a process patent (the 9th bullet point in para 173), and it is unlikely that her Honour intended to depart, sub silentio as it were, from the principles enunciated by Mason and Stephen JJ with respect to product patents.
The respondents – particularly in this regard the Sigma respondents – submitted that AMR had overstated the proposition for which Olin stood. Properly understood, to the extent that it relates to a patent for a product, Olin was to be seen as a case of a “missing integer”. The critical claim in the patent in suit in that case defined a product with (180 CLR at 259):
… [a] relatively thin walled tubular structure having a base at one end and formed of highly crystalline polyolefin polymer of substantially regularly ordered molecular structure, said structure being characterized by the fact that the wall of the structure has been formed in the solid state at an operative temperature below the crystalline melt temperature of the polymer to provide an increase in tensile strength of the polymer in said wall at a point remote from the base of at least twice the tensile strength of the polymer in said base.
Having dealt with the respects in which Mason and Stephen JJ held that the Core patent did not disclose an effective means of producing the article in question, the Sigma respondents drew attention to the following passage in their Honours’ reasons (180 CLR at 260):
Moreover, the Core patent made no mention of highly crystalline polyolefin polymer. Even if such materials were comprehended by the language of the patent it does not arithmetically express the increase of strength which will be obtained, whereas claim 10 asserts a twofold increase in strength.
Olin was, according to the Sigma respondents, more readily to be explained as a case of missing integers, rather than as a case in which the effective means of producing the product had not been disclosed.
There are indications in the judgment of Bennett J in Lundbeck that her Honour did not intend to endorse, or to apply to the facts before the Full Court, each of the “general propositions” set out in para 173 of her reasons. If read literally, some of those propositions would seem, with respect, to be difficult to reconcile with others of them, and, indeed, with the conclusions which her Honour expressed at para 180 thereof. I have in mind the proposition, drawn from Acme Bedstead, that –
… [a] prior publication does not invalidate a patent unless it supplies sufficient information to enable a person of ordinary skill to produce the product subsequently claimed;
and the proposition that –
… [t]he requirement is that a person of ordinary knowledge of the relevant subject would be able practically to apply the prior published discovery without the necessity of making further experiments;
and the proposition that –
… [a] disclosure that describes an effective means by which a claimed invention may be produced falls short of anticipation if it requires the exercise of inventive ingenuity or the taking of any inventive step.
I consider that Bennett J’s purpose in para 173 of her reasons was to go no further than to lay out the general propositions that emerged from the authorities. Not all of them were to be assumed to be applicable to the questions which her Honour subsequently determined.
Notwithstanding that qualification, although it is true that Bennett J made only passing reference to Olin, her Honour considered Hill v Evans in some detail. She cited the passage on which AMR relies here (177 FCR 151 at 191 [174]), and noted that Lord Westbury had explained what that meant “in a context where a prior publication does or does not actually disclose the subsequent invention”. Her Honour continued (at [174]):
He said that the information as to the alleged invention given by the prior publication must, for the purposes of practical utility, be equal to that given by the subsequent patent. It is apparent that in Hill v Evans, the disclosures in the prior publications were not of the integers of the subsequently claimed invention. His Lordship did not say that the invention had to have been previously made but that it had to have been previously made known.
Bennett J later said (at [177]):
Lord Westbury endorsed the conclusions in Househill Company v Neilson 1 Webst Pat Ca 718. In that case, Lord Lyndhurst held that where a prior publication included a distinct and clear description of a machine, there was anticipation if the description corresponded with that in the patent, even though the machine as described in the prior publication had never been worked. In Hill v Evans, there was information missing in the prior publications and there was no anticipation.
It was against the background of this discussion that Bennett J expressed her “conclusion” on the matter of anticipation, which I have set out at paras 23 and 24 above. There can, in my view, be little doubt but that her Honour intended the statement that “where the prior publication discloses exactly what is claimed, there is anticipation” to be understood in the categorical terms in which it was expressed. As her Honour made clear, some form of qualification was to be introduced only where there was “a disclosure less than the entirety of the claim”.
That understanding of Lundbeck was the very basis upon which the later case of Apotex was decided. It was submitted on behalf of AMR that Apotex could be explained by some less categorical proposition than that which I have identified as established by Lundbeck. Attention was drawn to the passage in the joint reasons in which their Honours observed that “the skilled reader would understand to separate the enantiomers and would know the methods to apply” (82 IPR 416 at [106]). AMR also stressed that its case here is in no sense inconsistent with what Bennett and Middleton JJ said at para 108, which I have set out at para 26 above. It was not AMR’s case that, in every case, the method of preparation “no matter how routine” had to be set out in the prior art before there could be anticipation. If the method of preparation was to be readily discerned by the skilled reader, and could be applied as a matter of routine, the later invention would not be novel. AMR stressed that its case is concerned wholly with the situation in which the method of preparation did not satisfy this requirement and had not been disclosed as such in the prior art: then, according to AMR, mere disclosure of the compound by name would not amount to anticipation.
The extent to which Bennett and Middleton JJ, in Apotex, considered that the reasons of Bennett J in Lundbeck were particularly apposite to the case before them is to be seen in the following paragraph in their Honours’ reasons (82 IPR at 433 [104]):
Some of the same issues concerning novelty arose in Lundbeck, a case which also involved a claim to the (+)-enantiomer of a racemate. From the consideration in Lundbeck, the following is apposite to the consideration of anticipation by the prior art patents in this case:
• Where the prior publication discloses exactly what is claimed, there is anticipation: Lundbeck at [180].
• There is anticipation if the skilled addressee would add missing information to what is disclosed in the prior art as a matter of course and without the application of inventive ingenuity or undue experimentation: at [181]. A disclosure is sufficient if it enables the skilled addressee, in the ordinary course and without invention, to add what is missing in the prior publication to obtain the claimed invention: at [183].
• If the prior art discloses the very subject matter of the invention, the person skilled in the art is assumed to be willing to make trial and error experiments to get it to work: at [189]. If the disclosure is of an invention which, if performed, would infringe the patent, there is anticipation.
• The question is whether the disclosure is sufficient to enable the skilled addressee to perceive, understand and, where appropriate, apply the prior disclosure necessarily, but within the ordinary limits of trial and error, to obtain the invention: at [190].
Of these four points, the first is the simple proposition of law on which the respondents rely here. The second is concerned with the “missing integer” situation. The third complements the first in a situation in which the prior art discloses something which, without “trial and error experiments”, does not “work”. The fourth, perhaps, provides the most obvious support for the position of AMR in the present case, but when para 190 in Lundbeck is examined, it will be seen that Bennett J was there concerned with the situation in which the prior art did not give “complete disclosure” (ie of the compound in question). In my view, the only unstrained reading of her Honour’s reasons in Lundbeck is that the disclosure of “exactly what is claimed” must be regarded as a “complete disclosure”, whether or not it is accompanied by a statement of the effective means of preparation.
In the view I take of the law as expounded in Lundbeck and Apotex, the disclosure of a compound by exact naming in the prior art is sufficient, of itself, to constitute anticipation. As a single Judge of the court, I do not believe I am in any position to consider the consistency of these Full Court judgments with earlier authority, however high. That authority was within the jurisprudence to which Bennett J gave extensive consideration in Lundbeck. I am bound by her Honour’s conclusions, to the extent that they relate directly to the facts of the present case. For reasons set out above, I take the view that they do so relate, and that the disclosure of the relevant compounds in Carr 129 and Carr 146 in terms anticipated the invention in Claim 1 of the patent in suit. It follows that it likewise anticipated Claims 6, 7, 8, 9 and 10. As indicated earlier in these reasons, Claim 11 is for a class of compounds made by reference to particular processes. It was no part of the respondents’ cases that the mere disclosure of those compounds in the prior art anticipated the invention so far as referred to in Claim 11. My conclusion in this part of my reasons does not extend to that claim.
Before leaving this area of controversy, I should say something about the claim in the patent in suit to “substantially pure” compounds. Ultimately, it was not submitted on behalf of AMR that the Carr patents did not disclose the relevant compounds as such. AMR’s point, rather, was that, if the skilled addressee set about to prepare any one of these compounds in accordance with the instructions in Carr 129, he or she would not obtain para‑substituted purity. But it was not submitted that, if the view I take of the law as expounded in Lundbeck and Apotex be the correct one, the patent in suit stood apart from Carr 129 only because of the claim to substantial purity. Thus the point raised particularly by the Sigma respondents that the invention in suit was on any view anticipated by Claim 2 in Carr 146, which claimed “substantially pure” compounds, does not arise. It would, however, have been a complete answer to any contention by AMR that the very compound claimed under the patent in suit had not, in terms, been disclosed in the prior art.
THE RESPONDENTS’ ALTERNATIVE CASE ON NOVELTY: EFFECTIVE MEANS OF PREPARATION
In the alternative, the respondents submitted that, if it were necessary (to establish want of novelty) that the prior art base disclosed not only the compounds claimed in the patent in suit but also an effective means for their preparation, Carr 129 and Carr 146 did so. What would be perceived and understood, and what could be practically applied (see Hill v Evans) are, of course, questions to be approached from the perspective of the skilled addressee. Conformably with this requirement, the parties called organic chemists of some standing in their profession to opine on these questions. AMR called Prof Christopher Easton, Distinguished Professor at the Research School of Chemistry, Institute of Advanced Studies, Australian National University; Alphapharm called Prof David Black, Professor of Organic Chemistry at the University of New South Wales, and Prof Bruce Wild, Emeritus Professor in the Research School of Chemistry at the Australian National University; and the Sigma Respondents called Dr Alan Robertson, Chief Executive Officer and Managing Director of Pharmaxis Ltd.
Although both Carr 129 and the patent in suit disclose a series of compounds, the question of effective means of preparation was addressed by the parties with reference to one only of them, 4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1- hydroxybutyl]-α,α-dimethylbenzeneacetic acid, that is, fexofenadine. The parties’ cases were based on the implicit assumption that the means of preparation of this compound disclosed in Carr 129 might stand, in effect, as a proxy for the like means so disclosed with respect to the other compounds. I shall proceed by reference to that assumption.
According to Carr 129, fexofenadine could be synthesised by the sequential combination of the methods described in Examples 5(A), 5(B), 2 and 3 therein, the product of each of which provided the starting material for the next. Part (A) of Example 5 is as follows:
To 700 ml of carbon disulfide containing 36.5 g (0.254 mole) of 4-chlorobutyryl chloride is added 74.5 g (0.56 mole) of aluminum chloride with stirring at -10° C. Stirring is continued for about 15 minutes at about 25° C. then the mixture is recooled to 5° C. and 48.4 g (0.294 mole) of ethyl α,α-dimethylphenylacetate in 100 ml of carbon disulfide is added. The reaction mixture is stirred on an ice bath for 3½ hours then stirred to 15½ hours at 25° C. then poured into HCl-ice water and extracted with chloroform. The extract is washed with dilute sodium carbonate solution, water and saturated sodium chloride solution, dried over magnesium sulfate, and evaporated giving as a solid ethyl 4-(4-chloro-l-oxobutyl)-α,α-dimethylphenylacetate.
Prof Wild described this as involving “a typical Friedel-Crafts acylation using aluminium chloride in carbon disulfide solvent”. He identified the reaction scheme as follows:
All of the experts who expressed a view on the subject gave effectively the same evidence as Prof Wild, namely, that, when they first looked at example 5(A), they anticipated a relatively uncomplicated Friedel-Crafts acylation.
Part (B) of Example 5 is as follows:
A mixture of 4.5 g (0.0163 mole) of α,α-diphenyl-4-piperidinemethanol, 6.1 g (0.0205 mole) of ethyl 4-(4-chloro-l-oxobutyl)-α,α-dimethylphenylacetate, 5 g (0.05 mole) of potassium bicarbonate and 0.05 g of potassium iodide in 50 ml of toluene is stirred and refluxed for 72 hours then filtered. Ether then ethereal hydrogen chloride is added to the filtrate, and the resulting precipitate collected and recrystallized several times from methanol-butanone and butanone to give ethyl 4-[4-[4-(hydroxydiphenylmethyl)-l-piperidinyl]-l-oxobutyl]-α,α- dimethylbenzeneacetate hydrochloride. M.P. 205.5°-208° C.
Prof Wild said that this described “a reaction carried out in toluene over 72 hours under typical coupling conditions with use of potassium bicarbonate as base and potassium iodide as catalyst.” He identified the reaction scheme as follows:
Example 2 of Carr 129 is as follows:
Ethyl 4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1-hydroxybutyl]-α,α-dimethylbenzeneacetate hydrochloride
A solution of 5.64 g (0.01 mole) of ethyl 4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1-oxobutyl]-α,α-dimethylbenzeneacetate hydrochloride in 200 ml of absolute ethanol and 50 ml of methanol and 0.5 g of platinum oxide is hydrogenated at about 50 psi for about 1 hour until the infrared showed no evidence of a ketone carbonyl function. The solution is filtered and the filtrate concentrated leaving a residue which is recrystallized from butanone and methanol-butanone to give ethyl
4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1-hydroxybutyl]-α,α-dimethylbenzeneacetate HCl, M.P. 185°-187° C.Prof Wild described this as “a reduction of the carbonyl group of the keto-carbonyl group of [the product obtained under Example 5(B)] with hydrogen at 50 psi for 1 hour in ethanol-methanol over a platinum oxide catalyst”. He added that Carr 129 indicated that sodium borohydride could also be used for this reduction step. He identified the reaction scheme as follows:
Example 3 of Carr 129 is as follows:
4-[4-[4-(Hydroxydiphenylmethyl)-1-piperidinyI]-1-hydroxybutyl]-
α,α-dimethybenzeneacetic acidTo a solution of 0.6 g of ethyl 4-[4-[4-(hydroxydiphenylmethyl)-l-piperidinyl]-l-hydroxybutyl]-α,α-dimethylbenzeneacetate in 20 ml of absolute ethanol is added 10 ml of a 50% solution of sodium hydroxide. The mixture is refluxed for 3½ hours and concentrated to a solid after which a minimum amount of methanol to dissolve the residue is added. 10% Aqueous HCl is added until pH 7 is reached, the methanol removed by evaporation and water (25 ml) is added. The resulting precipitate is recrystallized from methanol-butanone to give 4-[4-[4-(hydroxydiphenylmethyl)-1-piperidinyl]-1- hydroxybutyl]-α,α-dimethylbenzeneacetic acid, M.P. 195°-197° C.
Prof Wild described this as “the hydrolysis of the ethyl ester to the desired acid under typical conditions”. He identified the reaction scheme as follows:
Making reference to the steps described in these examples in Carr 129, Prof Black said that he would, as at 24 June 1993, have expected to be able to reproduce those steps and to prepare 4-[4-[(4-hydroxydiphenylmethyl)-1-piperidinyl]-l-hydroxybutyl]-α,α-dimethylbenzeneacetic acid as well as a pharmaceutically acceptable salt thereof. In his affidavit of 28 August 2009, which addressed Prof Black’s affidavit in great detail, Prof Easton did not venture a comment on this aspect. Indeed, Prof Easton himself said that, if in 1993 he had wished to obtain fexofenadine of a purity which would be suitable for use as a component in a pharmaceutical substance, and had been provided with the synthesis described in Carr 129, he would have expected that, if he followed that synthesis, he would have obtained the desired product. It seems to have been common ground as between these two experts, then, that Carr 129 disclosed an apparently efficacious means of preparing fexofenadine. Nothing said by the other experts called by the parties contradicted that position.
However, beyond that basic level of consensus, Profs Black and Easton parted company. Fundamental to their disagreement on the subject of what might have been expected to occur when Example 5(A) was carried out was the effect of the ester attached to the alkyl group substituted on the benzene ring in the starting material (ethyl α,α-dimethylphenylacetate). As I understand it, the following propositions would be regarded as uncontroversial:
·an existing alkyl group substituted on the benzene ring is activating (ie electron donating), and leads predominantly to secondary substitution at the para and ortho positions;
·an existing ester group substituted on the benzene ring is de-activating (ie electron withdrawing), and leads predominantly to secondary substitution at the meta positions;
·depending on the size and structure of an existing alkyl substituent, steric hindrance may discourage secondary substitution at the ortho positions, with the result that substitution at the para position might be expected to predominate.
In the case of the starting material under Example 5(A) in Carr 129, there was a benzene ring with an alkyl substituent, but that group in turn had an ester attached to it. In his affidavit of 6 February 2009, Prof Black explained what he might have expected in these circumstances as follows:
In my experience, this type of Friedel-Crafts acylation reactions commonly produce a mixture of regioisomeric products. I would predict that the para regioisomer would be the predominant product as the electron donating effect of the alkyl group induces the para arrangement. The electron withdrawing effect of the ester group also induces a meta arrangement. The effect of the ester is much weaker than the effect of the alkyl group however I would not be surprised to observe some meta isomer in the product. I would also predict that at least some ortho regioisomer could be formed, although steric hindrance of the bulky dimethylacetate functional group present in the ethyl α,α-dimethylphenylacetate reactant would limit the amount of ortho isomer formed.
That is to say, although not directly substituted on the benzene ring, the existence of the ester group would result in the production of some meta-regioisomeric forms. I note that Prof Black expressed these views before he was given access to the patent in suit, and (at least so far as disclosed in the evidence) without having been informed of the outcome of attempts made on behalf of Alphapharm to synthesise the product said to be the product of Example 5(A).
Prof Easton did not agree. In his affidavit of 28 August 2009, he said:
The Friedel-Crafts acylation described in example 5(A) of Carr '129 is said in that document to produce a para substituted regioisomer. Having regard to the fact that the mono-substituted starting material of example 5(A) of Carr '129 has a substituted alkyl substituent, which I would expect to be electron donating and therefore ortho/para directing, I would have expected that the Friedel-Crafts acylation of example 5(A) would have resulted in the production of the para regioisomer as described. While I would have expected that some amount of the ortho substituted regioisomer might form, I would have expected that there would be significant steric hindrance, such the para regioisomer would predominate. I would not have expected the formation of the meta regioisomer in any significant amount. In a mono-substituted benzene having an ester group directly attached to the aromatic ring, that ester group will have the effect of withdrawing electron density from the ring, thereby favouring substitution at the meta position. However, in the starting material used in Carr '129, the ester group is not directly attached to the aromatic ring, but rather is separated from the ring by an alkyl group. I would have expected that this separation would result in the ester group itself having no electron withdrawing effect on the ring.
Prof Black responded to this in his affidavit of 27 November 2009. He referred to “the first principles of electrophilic substitution … on a singly substituted benzene ring” as reported in standard teaching textbooks. He said:
A simple (unsubstituted) alkyl group is activating and consequently ortho, para directing. Accordingly, if a simple (unsubstituted) alkyl group is present on a benzene ring it will be an ortho, para director. However, when the alkyl group is itself substituted, the activating nature of the alkyl group will be directly affected by the properties of the substituent. In the case of a strongly deactivating carboxylic ester substituent (like that in Example 5(A) of [Carr] 129), this will result in a predictable deactivating effect that will encourage meta substitution to occur. In my opinion, these First Principles were well known both before and after the Priority Date. They were familiar to me and, to my knowledge, to my colleagues from before this time.
In his oral evidence in the concurrent session, Prof Black expanded on this by providing a simple explanation of the reason why the existence of a carboxylic ester substituent on the alkyl group would reduce the extent to which that group was activating. It related to the energy barrier of the transition state in the reaction by which the incoming group comes to be substituted on the ring. Prof Black said:
Once you change the group on the benzene ring from just methyls and you have C‑methyl‑methyl and then an electron‑withdrawing group, being the carboxylic ester, you have reduced the capacity of that group to be activating. What that means is that the energy barrier, the transition state for the formation of para, is raised and so it becomes closer to the same energy barrier for meta, and once you do that and you have the two trajectories relatively close together, then it is very difficult to get one of the isomers without getting the other one.
I should also set out a question and answer from the cross-examination of Prof Black:
As to the potentiality for the meta isomer, your expectation was that there might be some, but not in a significant amount. Is that correct? – My expectation was that there would be some. Whether it would be small or – my expectation was that it would be less than the para. Essentially, the group, if I can just explain my reasoning, the group that is attached to the benzene ring, the atom directly attached, is the main atom of directing influence and that is normally electron donating favouring para and ortho. But once you associate with that an electron withdrawing group, it will weaken the electron‑donating activation of the directly attached atom. Now, whether it is sufficient to override that directing influence or not is something that is a matter of guesswork, or was on my part at this stage. But it was clear that there would be an electron‑withdrawing effect that would at least weaken the para drive of that group as a whole, and whether it completely wipes it out – Uggeri, I think, says it would wipe it out. That is something you would really have to do the experiment in order to find out, but it would decrease that advantage of getting the para direction and if you decrease that, you then increase the possibility of getting meta.
In its written submissions, AMR referred to all but the first two sentences of this answer. I consider that Prof Black’s statement that his expectation was that the meta regioisomer would be less than the para regioisomer is significant in the present context.
In those submissions, AMR also attributed to Prof Black (amongst others) an expectation that “[a]ny meta regioisomer would be present in ‘small quantities’ only”. The quoted reference to “small quantities” appears to have come from Prof Black’s affidavit of 6 February 2009, in which he said (apropos the product of Example 5(A)):
No melting point is indicated for the solid product and the yield is also not reported at this first step so it is difficult to know precisely what the level of purity is. The impurities in the sample are likely to be small quantities of structural isomers. The only other impurity which may be present is unreacted starting material.
This statement came immediately after the paragraph of Prof Black’s affidavit which I have set out in para 47 above. Under cross-examination, Prof Black confirmed that the structural isomers referred to were the meta and the ortho regioisomers, and that, notwithstanding the presence of some of the meta regioisomer, his expectation would have been that the para regioisomer would predominate.
Prof Easton did not add anything to his earlier evidence in direct response to these explanations by Prof Black, but he made an observation in response to something said by Prof Black about the work of Prof Wild’s technical assistant, Mr Paul Gugger, under Example 5(A) (to which I refer below) which, it seemed to me, encapsulated his (Prof Black’s) position on the question whether some meta regioisomer might be expected from the reaction in Example 5(A). To set the context, what Prof Black said was:
… there is very little evidence given in the patent, there are no spectra, there is no real data, but that is typical of most of these things and equally typical of the Australian patent. When writing up the description of a compound, it would be normal to say, "This compound was prepared in a certain way and delivered as the experimental procedure is written down for 5(A), 5(B) and so on." It is very rare and extremely unlikely that there would be any comment about how difficult it was to actually get this and, "We had to separate it from all sorts of other things," people just don't usually write this. They might mention it if it is in a learned journal publication, something like that could be mentioned in the discussion process, part of that paper, but it is almost never written in the experimental section.
In describing how you would prepare a certain compound, you get the procedure for getting that compound without any discussion of a general nature. So it's not surprising, to me, that the patent says, where it describes the formation of 5(A), that it is the para isomer because it is the para isomer but with some other stuff there as well, but it goes on then to 5(B) because they are focusing on the synthesis of 5(B) and beyond.Prof Easton’s comment was:
… I would have to say that I would be very surprised to see a compound specifically named when it was only produced in 30 per cent purity or as a 1:1, approximately 1:1 mixture, of regioisomers. So I disagree with his comments that a compound specifically named as being the product of a protocol would not be produced essentially as a pure material.
In this passage, “30 per cent purity” was a reference to the largest single component in the product which Mr Gugger obtained under Example 5(A): inferentially the para regioisomer (see para 98 below).
Prof Wild also expected that a mixture of the para and meta regiosomers could result from the reaction described in Example 5(A) of Carr 129. His theoretical justification for that expectation was:
Due to the steric hindrance of the bulky dimethylacetate functional group present in the ethyl α,α-dimethylphenylacetate reactant, substitution at the ortho position of the aromatic ring would not be preferred. On the other hand, the para position of the ring is the least hindered and substitution in that position would lead to the major regioisomer of the product.
Prof Wild supported his view with a reference to a publication by E.D. Morgan entitled “Synthesis of p-Alkylphenylacetic Acids” (Tetrahedron, 1967, 23, 1735-1738) in which it had been reported that, following similar Friedel-Crafts acylations of phenylacetic esters with several acyl halides, mixtures of alkyl esters of acylphenylacetic acids in ratios of ortho:meta:para (4-6):(38-49):(44-58) were obtained. The conclusion in that reference was that the meta and para positions were preferred, with ratios of 38-49 and 44-58, respectively. Prof Wild continued:
Despite the para position being the least hindered and so preferred in the reaction, there is a similar proportion of meta regioisomer produced. This is due to there being two available meta positions, but only one para position. Despite there being two ortho positions, the proportion of ortho regioisomer formed is almost insignificant because of the severe steric hindrance of the dimethylacetate group that effectively blocks this position.
To employ the words of Lord Westbury in Hill v Evans, all of Profs Easton, Black and Wild perceived and understood how to prepare fexofenadine from the examples given in Carr 129. However, Prof Easton’s understanding was based upon his acceptance of the uncomplicated working of Example 5(A) in accordance with its terms. The understanding of Profs Black and Wild was based upon a perception that Example 5(A) would indeed give a mixture of regioisomers, which would then have to be resolved in the workings of the latter examples. What steps would be required in that regard, and whether the necessary departures from the text of Carr 129 would be within the principle in Hill v Evans, were matters of serious contention in the proceeding. The debate was played out in the context of attempts which scientists engaged by AMR and Alphapharm made to synthesise fexofenadine following these examples in Carr 129. I shall turn to such matters presently, but I first desire to say something about the legal setting in which the relevant issues arise.
The major focus of AMR’s energies on the novelty point in the present proceeding was on the second dimension of Lord Westbury’s injunction in Hill v Evans, namely, that, to be anticipatory, a prior art citation had to be such as would give the skilled addressee the ability to prepare the compound of interest, or, in the words of Stephen and Mason JJ in Olin, the “effective means” to do so. Here the onus of proof may be of some importance. AMR submitted that, since the respondents carried the legal onus of establishing want of novelty, the disclosure of effective means was a necessary part of their case, and they had to prove it. However, as pointed out above, a skilled addressee armed only with Carr 129 would have understood how fexofenadine could be prepared. That circumstance, in my view, effectively cast upon AMR the evidentiary onus of demonstrating that all was not as it seemed under Carr 129, and that Examples 5(A), 5(B), 2 and 3 did not “work”. As it happens, and as will appear presently, I have not found it necessary to decide this aspect of the case by reference to the locus of the evidentiary onus.
As will become apparent, Lord Westbury’s qualifier that it must be possible to apply the discovery “without the necessity of making further experiments and gaining further information” (1A IPR at 6) has become crucial in the present case. This qualifier, as I have called it, was the subject of elaboration by Lord Reid in Van der Lely NV v Bamfords Ltd [1963] RPC 61. According to his Lordship, the “further experiments” referred to in Hill v Evans did not include “the ordinary methods of trial and error which involve no inventive step and are generally necessary in applying any discovery to produce a practical result” ([1963] RPC at 71). In the submission of AMR, the only input that might be expected of the skilled addressee, in order to make the prior art work, are steps of an utterly routine or commonplace nature. It did not endorse the idea that there was a clear dichotomy between “ordinary methods of trial and error”, on the one hand, and processes which involved an “inventive step”, on the other hand. Departures from the prior art beyond the routine ought not be permitted, notwithstanding that they may not involve the taking of any further inventive step.
I was not referred to any decided case in which this distinction had been considered, or in which the actual content of Lord Reid’s elaboration had given rise to controversy in a practical situation. However, approaching the matter at the level of principle, and informed by the observations of Bennett J in Lundbeck, I think that the following things can be said. First, the question is not whether the synthesis of the compound of interest would involve an inventive step in the light of the relevant prior art citation. Rather, the question is whether “the prior publication is sufficient to make the claimed invention apparent to the skilled addressee” (Lundbeck 177 FCR at [173]). Secondly, the question is not whether a series of instructions disclosed in the prior art would inevitably lead to the preparation of the compound of interest. That question arises in a case, such as Abbott GMBH and Co KG v Apotex Pty Ltd (2010) 87 IPR 561, in which the compound as such is not disclosed. That is not the present case. Thirdly, the issue is concerned with the sufficiency of the information in the prior art to make the compound of interest It is assumed that the scientist will bring all of his or her skill and experience to bear on the problem, and will not feel obliged to approach that information as though it were a recipe. In the present case, the information on which the respondents rely consists of a series of recipe-like examples, but there will be cases where less prescriptive modes of expression of the information will nonetheless enable the scientist to understand how the compound can be produced. Fourthly, the very idea of “trial and error” implies that the making of the compound in accordance with the prior art information may be neither straightforward nor uncomplicated. Where the information consists of a series of instructions, it implies that those instructions may be incomplete or unhelpful. The notional scientist is assumed to be equipped with the intellectual and practical wherewithal to perceive in the instructions how the compound may be produced, notwithstanding such shortcomings. And fifthly, the idea of “trial and error” also implies a certain degree of perseverance. Here, much will depend on the scientist’s understanding of the general principles conveyed by the prior art information: if it seems to him or her, from that information, that preparation of the compound of interest should be possible, it might be expected that he or she would make every reasonable effort to achieve the intended result.
In Merck and Co Inc v Arrow Pharmaceuticals Ltd (2006) 154 FCR 31, 52 [63], the Full Court held that the following propositions were established on the authorities with respect to these provisions:
1.The opening words of s 18(1) (“a patentable invention is an invention that”) impose a threshold requirement that the “patentable invention” be an “invention”, that is to say an “alleged” “manner of new manufacture” within s 6 of the Statute of Monopolies (Philips at 663).
2.That requirement will not be met if, on the face of the specification, the subject matter:
(a)lacks the necessary quality of inventiveness under the Statute of Monopolies (Philips at 664);
(b)is not new ([National Research Development Corp v Commissioner of Patents (1959) 102 CLR 252] NRDC at 262; Philips at 664).
3.A new use of an old substance is not an invention if its known properties make it suitable for that use — in such a case the new purpose is "no more than analogous to the purposes for which the utility of the substance is already known" (NRDC at 262).
4.But there will be an invention if the new use consists in taking advantage of a hitherto unknown or unsuspected property of the substance (NRDC at 262).
The respondents’ first submission in this area of discourse was based upon the statement in the specification, contained within the paragraph which I have set out at para 12 above, that “a practical separation to obtain gram quantities of substantially pure regioisomers has not been achieved”. It was said that this amounted to an admission that less than gram quantities of substantially pure regioisomers had been achieved. Since the claims in contention were not limited by reference to gram quantities of the compounds claimed, it was said that the specification showed, on its face, that the subject matter if the invention was not new. AMR’s response to this submission was as follows:
The respondents contend that the Patent admits, at page 6, line 2 to 3, that separation of less than gram quantities of substantially pure regioisomer has been achieved. That is not a proper characterisation of the statement, “a practical separation to obtain gram quantities of substantially pure regioisomers has not been achieved.” There is no admission that a separation to obtain lesser quantities has been achieved. Further, the statement is qualified by the last sentence of the same paragraph, “it has not been possible to obtain either of the regioisomers in each mixture in substantially pure form.” Of course, separation of less than gram quantities by the inventor in the course of developing the invention would not be an admission that the invention is not “new”.
As will be apparent, this passage contains three points in response to the respondents’ submissions. I shall return to the first presently, after I have dealt with the second and third.
The statement in the last sentence in the relevant paragraph – “it has not been possible to obtain either of the regioisomers in each mixture in substantially pure form” – refers to the previous sentence which, in my view, is both theoretical and tendentious, stating what the mixture “would be expected to contain”. In the final sentence in the passage, the inventor is, in my view, stating no more than that he was unable to separate out the para and meta regioisomers in the proportions of 33% and 67%. I do not consider that this statement would be effective to neutralise the implication upon which the respondents rely (that less than gram quantities of the para regioisomer had been obtained) if otherwise such an implication fairly arose.
As to the submission that AMR’s manner of manufacture case could not be defeated by pointing only to the experimental work done in the inventor’s own laboratory in the course of devising the invention itself, if the work referred to were part of the inventive process, I would be inclined to agree. However, the passage upon which the respondents rely appears in the “background” section of the specification, and constitutes a statement by the inventor of the shortcomings in the prior art. Properly understood, this passage described not the process of inventing something new, but the outcome which one would achieve (or rather, in the inventor’s words, would be unable to achieve) by following the prior art. If we were to find an effective admission that less than gram quantities of the compound claimed to have been invented were obtainable in 1993 by routine HPLC, it would be no answer that the inventor came to that position as part of his background work in connection with the invention in suit.
That brings me back to the first of the three points made by AMR. Is the statement that a practical separation to obtain gram quantities of substantially pure regioisomers had not been achieved the equivalent of an admission on the face of the specification that less than gram quantities of the substantially pure para regioisomer had been separated out? How the jurisprudence referred to in Merck has application to a circumstance in which the relevant admission is said to be implicit from some words, not expressly admitting anything, on the face of a specification is not a question upon which I was addressed by any party. I was not referred to any authority on the point. At the level of principle, I consider that a court should be cautious before treating a negative express statement in a specification as the equivalent of an implied admission in terms which would be conveyed by the corresponding positive statement. In the context of the present case, I am not prepared to hold that the specification, on its face, admits that the subject matter of the invention was not, at the priority date, new. I reach that conclusion for the following reasons.
First, the question of what is admitted on the face of the specification must be addressed by reference to the terms of the specification as a whole. No narrow, fragmentary or nicely grammatical approach is called for. I accept AMR’s submission that it is as clear as may be, from the terms of the specification in the present case, that the inventor is asserting that the relevant piperidine derivative compounds could not, consistent with the prior art, be synthesised at the level of substantial purity. I should be slow to treat that assertion as wholly negated by what is said to be an implication arising from an apparently uncontentious proposition that the inventor was unable to achieve gram quantities of substantially pure regioisomers. For all the reader of the specification knows, the inventor may never have attempted to achieve separation at less than gram quantities; or he may have attempted to do so and failed, but regarded that circumstance as inconsequential in a setting in which his objective was to achieve “a practical separation to achieve gram quantities”.
Secondly, what the respondents essentially seek to undertake here, in my view, is not the reading of an implication from the terms of the specification as such, but the drawing of a logical conclusion from the statement expressly made that the separation of gram quantities has not been achieved. There is an important distinction between the two. The first would involve a perception of something implicitly stated by the inventor, and would satisfy the requirement that the admission appear on the face of the specification. The second would involve , as it were, a joining of the dots between things stated in the specification to reason, as a matter of logic rather than terminology, that the inventor most probably had been able to synthesise the subject matter of the invention from the disclosures in the prior art. While this second approach may have some currency in the normal world of curial fact‑finding, it is, in my view, inappropriate as a means of identifying what appears on the face of the specification.
Thirdly, it was an important aspect of the respondents’ submission that, as at the priority date, HPLC could be used, and was routinely used, as a means to separate components from a mixture of regioisomers. Factually, that appears to be uncontroversial. According to the respondents, therefore, when the specification states that the mixture of regioisomers resulting from the examples in Carr 129 “can be analysed by HPLC experiments”, this should likewise be taken as an admission that the regioisomers could be separated by HPLC. But, in applying this branch of the law, I believe I am restricted to the terms of the specification itself. I am not permitted to bring to those terms some appreciation, obtained elsewhere, of what the techniques employed by the inventor would have permitted him to do. It does not appear on the face of the specification that the inventor in fact resorted to what has been called “preparative HPLC”, and I should not supply that deficiency in order to assist the respondents to make good their point.
Thus, I am not satisfied that the invention was not a manner of manufacture by reason of the “gram quantities” point.
It was also submitted on behalf of Alphapharm that the patent went no further than to claim a known compound with desirable attributes, that is to say, substantial purity. It was said that this was no more than a claim for “a mere desideratum”, in which respect reliance was placed upon NV Philips Gloelampenfabrieken v Mirabella International (1993) 44 FCR 239, 264-265. As it happens, the judgment of Lockhart J in that case says nothing about “mere desideratum” and, for my own part, I have difficulty finding any support in it for the point now advanced by Alphapharm. That was a case in which, relevantly to the present point, it was held that the combination of known, unremarkable, characteristics in an existing product (thereby giving the product an advantage over like products which lacked those characteristics) was not a “manner of new manufacture” or a “manner of manufacture”.
In Eli Lilly and Co v Pfizer Overseas Pharmaceuticals (2005) 218 ALR 408, Heerey J referred to the Full Court judgment in NV Philips as authority for the proposition that a “mere desideratum”, which his Honour paraphrased as “something which is disclosed as no more than a wished for result”, was not an invention in the sense of involving a manner of new manufacture within the meaning of s 6 of the Statute of Monopolies (218 ALR at 444-445 [212]). Like his Honour in that case, however, I consider that the specification in the present case, on its face, goes further than to disclose “a wished for result”. It asserts that the prior art does not permit the synthesis of the compounds of interest in substantially pure form and, therefore, that those substantially pure compounds did not previously exist. It “tells the reader the manner in which the desired object is to be achieved” (Eli Lilly 218 ALR at 445 [213]). In my view, there is no substance in this ground of objection advanced by Alphapharm.
It follows from what I have said above that I reject the respondents’ manner of manufacture cases.
SUFFICIENCY
The respondents submitted – perfunctorily if I may so observe without disrespect – that the patent in suit did not describe the invention fully, giving the best method known to the inventor of performing the invention, as required by s 40(2)(a) of the Patents Act. However, the patent sets out in detail the means by which compounds of the relevant class as such might be prepared, and none of scientists called by the respondents proposed that these methods would be insufficient for the purpose. Rather, the respondents’ case would require me to pass judgment upon the sufficiency of the specification, as perceived by a worker skilled in the art, without the assistance of the opinion of any such worker.
The respondents submitted that the specification was silent as to the identification, and the means of elimination, of impurities other than regioisomeric ones. As explained elsewhere, however, I take the view that the essence of the invention was a compound of substantial regioisomeric purity. I do not consider that the inventor was concerned with purity in other senses, from which it follows that he had no need, under s 40(2)(a), to describe the means by which it might be achieved.
Alphapharm also advanced a submission which was, in effect, in the alternative to its case on novelty. In its written outline, Alphapharm’s point was expressed as follows:
First, as submitted above, the Patent relies extensively on the skill of the skilled addressee in adjusting ordinary laboratory techniques in order to perform the alleged invention. If these skills are not available to the skilled addressee in construing or following the prior art, they cannot be relied upon by AMR and the Patent will be insufficient.
The limited basis upon which I have found, in my reasons above, that Carr 129 did not provide an effective means of preparing fexofenadine as such does not amount to a holding that “the skill of the skilled addressee in adjusting ordinary laboratory techniques” was not available to him or her in construing or following the prior art. Indeed, I would understand Alphapharm to have made common cause with the inventor in proposing that the crystallisation and other procedures referred to were within the synthetic stock‑in‑trade of the organic chemist at the priority date. That the inventor assumed an appreciation of such techniques, rather than describing them, does not warrant the conclusion that the specification in suit did not describe the best method known to the inventor of performing the invention.
I reject the respondents’ case under s 40(2)(a) of the Patents Act.
FAIR BASIS
The respondents submitted that the claims presently in contention were not fairly based on the matter described in the specification, as required by s 40(3) of the Patents Act. Their point was that the claims (with the exception of Claim 11) are for compounds, whereas the specification is confined to processes for synthesising those compounds. Where the specification describes processes only, claims for the compounds as such (however made) were not, it was said, fairly based.
This submission requires me to undertake a brief survey of the structure and content of the complete specification. Such a survey reveals the following:
·Pages 1A-6 are concerned with the “background of the invention” and deal with the limitations of the prior art (substantially the Carr patents) in respects to which I have referred at some length above.
·Pages 6-9 contain a “summary of the invention”, identifying the family of compounds covered thereby, and setting out, in very broad outline, how the compounds may be prepared.
·Pages 9-19 provide a “detailed description of the invention”, identifying the specific compounds that are covered by the various combinations referred to diagrammatically in the formula to which I have referred to in para 4 above. Certain compounds which are “particularly preferred” are identified. Various pharmaceutically acceptable salts, which are also included within the invention, are identified. The means by which the compounds covered by the invention might be administered therapeutically are referred to. Then the specification identifies a means by which the compounds may be prepared, moving on to the processes next mentioned.
·Pages 20-31 identify a series of synthetic processes for producing compounds covered by the specification.
·Pages 31-38 set out 13 examples of the preparation of various compounds covered by the invention.
·Pages 39-43 set out the claims.
So far as I can see, this appears to be a fairly conventional compound patent, in which the compound is identified and, conformably with s 40(2)(a) of the Patents Act, the known methods of preparing the compound are set out. With respect to the respondents, I have difficulty appreciating how it might be said that claims for the compounds are not fairly based in a specification such as this. The compounds for which claims are made are unambiguously referred to in the specification. The respondents’ point seems to be that the devising of a means to prepare compounds of substantial purity is the only advance which the specification made over the prior art. That is, of course, a separate point, but even then it must be remembered that the actual compounds to which the claims relate are substantially pure piperidine derivatives, it being asserted that the prior art contained no instance of such compounds. Assuming for present purposes, as I must, that the inventor was justified in making that assertion, I do not accept that the claims are not fairly based on the ground proposed by the respondents. If the claims are for a new invention, the compounds to which they relate are, in my opinion, correspondingly dealt with in the specification.
FALSE SUGGESTION
The respondents’ final point is that the patent was obtained by false suggestion, and should be revoked under s 138(3)(d) of the Patents Act. This requires the court to consider first whether a false suggestion was made, and secondly whether that suggestion “materially contributed to the commissioner’s decision to grant the patent or was a material, inducing factor, which led to the grant”: Ranbaxy Australia Pty Ltd v Warner-Lambert Co LLC (2008) 77 IPR 449, 468 [82].
Alphapharm’s false suggestion case was based on two letters sent by the then applicant for a patent to the Examiner on 20 June and 14 September 1997. Those letters came to be sent in the following circumstances. On 20 February 1997, the Examiner cited Carr 129 as anticipatory with respect to Claims 1-11 and 13 of the patent as applied for. In its letter of 20 June 1997, the applicant responded to this objection at length, in terms which, to a considerable extent, later found expression in the patent as granted. Relevantly to the present point, the letter covered the ground which I have traversed at paras 10-12 above, and continued:
The presence of meta isomer is not disclosed in US Patent No. 4,254,129. However, attempts to repeat the synthesis disclosed in these patents has demonstrated the existence of a sizeable quantity of the meta isomer. …
US Patent No. 4,254,129 does not mention the meta impurity problem, which the HPLC data … shows to be a major contaminant. The synthesis described in Example 5(A) of US Patent No. 4,254,129 calls for recrystallization; there is no indication of why this is needed or how it is carried out. However, as set forth in the attached August 25, 1993, letter from Louis J Wille … Corporate Patent Counsel, Marion Merrell Dow, Inc., then assignee of these patents, analysis of the actual sample corresponding to this example was found to contain 96.3% of the para isomer and 3.7% of the meta isomer. A second scientist at Marion Merrell Dow, Inc. followed this procedure and found that the product contained 95.9% para isomer and 4.1% meta isomer. Applicant's study of this chemistry has demonstrated that the Friedel-Crafts acylation of the ethyl ester of α,α-dimethylphenylacetic acid with 4-chlorobutyric acid reproducibly and consistently provides about a 67:33 mixture of the para to meta isomers. The purity levels achieved by Marion Merrell Dow, as described in the August 25, 1993, letter from Mr Wille, were apparently obtained by using a chromatography or crystallisation method. However, even such mixtures with higher levels of the para isomer do not constitute a “substantially pure” product, as required by the claims of the [p]resent application.Notwithstanding the ostensible indication in this extract, Mr Wille’s letter of 25 August 1993 made it clear that the samples of 96.3% and 95.9% para-purity referred to were in fact derived at the completion of Example 5(B), not Example 5(A), of Carr 129. Relevant extracts from Mr Wille’s letter were on the Commissioner’s file, and I doubt that the Examiner would have been under any misunderstanding on the point.
The letter of 20 June 1997 next referred to the importance of purity in pharmaceutical settings, and to relevant passages in the US Pharmacopeia. It concluded on this aspect:
From the above passage of USP, it is apparent that a pharmaceutical compound is considered “substantially pure”, in accordance with the present invention, where the level of impurities is less than 2%. As noted above, the level of meta impurities present when the subject piperidine derivative compound of the present invention is prepared in accordance with the process of US Patent No. 4,254,129 is significantly higher than 2%, even when that product is subjected to chromatography or crystallization steps which are at most poorly described in this reference. Accordingly, the compounds prepared by the prior art process are not “substantially pure”, as claimed by applicant, and therefore, it is respectfully submitted that the rejection based on this citation should be withdrawn.
Alphapharm’s first point about the letter of 20 June 1997 is that it contained a representation that “the product described in Carr US 129 was an ‘inseparable mixture’ of the meta and para regioisomers”. In terms, that was not strictly so. The letter stated that the product of Example 5(A) was “an inseparable mixture of monosubstituted aromatic regioisomers”. Speaking of the end product (ie that derived after Example 3), the letter stated: “Since these components cannot be completely separated, it has not been possible to obtain either of the regioisomers in each mixture in substantially pure form”. What was meant by “cannot be completely separated” is to be derived from the remaining passages to which I have referred. The applicant, while asserting that Example 5(A) of Carr 129 yielded an inseparable mixture of regioisomers, recognised that, with appropriate laboratory techniques, something of the order of 96% para-substituted purity was obtainable. However, according to the then applicant, that was not good enough. For a pharmaceutical compound, purity of at least 98% was required. It is as clear as may be that the applicant was asserting that such purity could not be obtained under Carr 129, even if 96% purity, or thereabouts, was achieved at the end of Example 5(B).
Although the “cannot be completely separated” statement in the letter of 20 June 1997 referred to the product of Example 3, that letter wholly ignored the contribution which Examples 2 and 3 may have made to purifying the product of Example 5(B). As Alphapharm submitted, Mr Gugger’s work in the present case demonstrated that a 96.39% pure compound at the 5(B) stage could, by ordinary methods applied under Examples 2 and 3, be converted into fexofenadine of 100% purity. The letter of 20 June 1997 contained the clearest of suggestions that this was not possible. That suggestion was, in my view, false, thereby satisfying the first requirement of s 138(3)(d), a deliberate intent to deceive not being a necessary ingredient of the provision: Pfizer Overseas Pharmaceuticals v Eli Lilly and Co (2005) 225 ALR 416, 495 [394].
In a letter dated 15 July 1997, the Examiner maintained the objection based on anticipation by Carr 129, but elaborated on it by reference to what seems to have been a different point from any that were dealt with in the applicant’s letter of 20 June 1997. There was no suggestion that the proposition that the 96% purity level achieved at the end of Example 5(B) under Carr 129 could not be improved on was not accepted. I infer that it was accepted. After some subsequent correspondence of no presently material importance, the patent proceeded to grant.
Given the stress which the letter of 20 June 1997 placed on the importance of at least 98% purity in the pharmaceutical context, I am in no doubt but that the suggestion in that letter which I have held to be false materially contributed to the Commissioner’s decision to grant the patent in suit. The present case is an instance of one in which “it may … be inferred that a representation in fact contributed to the decision to grant a patent, … [when] the representation was objectively likely to contribute to such a decision and the patent was in fact granted”: Ranbaxy 77 IPR at 468 [83]. It follows that the ground referred to in s 138(3)(d) of the Patents Act has been made out.
Alphapharm had another point about the letter of 20 June 1997. It related to the statement therein that “low field proton nuclear magnetic resonance spectroscopy is inconclusive in identifying the product of [the reaction in Example 5(A) of Carr 129] as a mixture”. It will be noted that this statement found its way into the corresponding part of the specification in the patent as granted. Alphapharm relied upon a declaration made on 30 January 1995 by the inventor under the patent in suit, and filed in the United States Patent and Trademark Office, which contained the following paragraph:
My experimental work started on or about May 9, 1992. However, I was unsuccessful in utilizing the synthesis disclosed in the Carr patents, because, according to nuclear magnetic resonance testing, there was a significant quantity of impurity in the product I prepared …. As a result, I concluded that if I continued with the approach of the Carr patents, I would only produce a mixture of para and meta aromatic regioisomers.
This was said to be directly contradictory of the statement in the letter of 20 June 1997, and to demonstrate that the latter was false.
I should deal first with a possible source of confusion on this point, to which the parties did not refer. The relevant passage in the letter of 20 June 1997 related to the product of Example 5(A) in Carr 129. The passage in the inventor’s declaration of 30 January 1995 was not specifically tied to any particular aspect of the Carr process. However, there was no suggestion by AMR that the inventor and the writer of the letter were not effectively speaking of the same stage in the synthesis – that resulting from the Friedel‑Crafts acylation. Indeed, as I shall show below, counsel for AMR responded to Alphapharm’s point on the unstated assumption that they were. I shall proceed in accordance with that assumption.
In response to this point, AMR first submitted that the inventor, having come to the end of Example 5(A) (or equivalent) with a regioisomeric impurity, “made the educated assumption that, if he continued, he would only produce a mixture of para and meta isomers”. Undoubtedly that is correct so far as it goes, but it does not tell the full story conveyed by the inventor’s declaration. That story included also the representation that he had used nuclear magnetic resonance testing to arrive at the conclusion that he had regioisomeric impurity. The present question is not whether such testing was sufficient to provide a quantitative regioisomeric characterisation of the material: it is whether such testing was conclusive in identifying the product as a mixture. The inventor’s declaration seems clearly to suggest that it was.
AMR next submitted that “Mr Gugger had the same problem … [t]hat is, the 1H NMR was insufficient to identify the three compounds present” in the material obtained under Example 5(A). However, Alphapharm submitted that “[i]t was not suggested to any of Alphapharm’s witnesses that the NMR data for the product of Example 5(A) of Carr did not reveal a mixture of regioisomers”. By the use of double negatives and slightly different concepts (“identify the … components” – v – “reveal a mixture”), these parties are here attempting to have essentially the same evidence perform service to opposite ends. What is the truth of the matter?
In his affidavit setting out the results achieved by Mr Gugger under Example 5(A), Prof Wild used the GC-MS data as the basis for his conclusion that the ortho:meta:para regioisomers were present in the proportions 1:46:53. He said that the 1H NMR spectrum was consistent with that. In his affidavit in response, Prof Easton said that, although the NMR analysis demonstrated the presence of more than one compound in the 5(A) material, and that the desired reaction product was probably present, it was not possible to say that the impurities were the other regioisomers of that product. In an affidavit in reply, Prof Wild did not accept Prof Easton’s reservations, but repeated his evidence that “[t]he patterns and chemical shifts of the NMR peaks for these impurities are, in my opinion, consistent with there being the expected ortho and meta regioisomeric by-products of the Friedel-Crafts acylation reaction”.
One of the topics covered in a concurrent evidence session (which did not involve Prof Wild) related to the analytical techniques used to test the purity of a substance as at the priority date. By reference to the 1H NMR analysis of the material obtained by Dr Simpson under his second run of Example 5(A), Prof Black explained what could be inferred from the various peaks there shown. He concluded that “the only clear thing that you can see is that there is some para isomer plus some other things”. Dr Robertson agreed, adding: “NMR indeed is seldom used as a pure analytical technique to look for very low levels of impurities, … because of the complications inherent in proton NMR spectroscopy”. In the course of a later concurrent session (which did involve Prof Wild), Prof Easton closely examined the 1H NMR analysis of Mr Gugger’s 5(A) product and reiterated, quite emphatically as I heard him, his view that that analysis did not permit one to reach any conclusion about the presence or absence of particular regioisomers. Prof Wild’s response was:
I agree with Prof Easton’s comments. If you look at the spectrum that we have been shown, it's an overlapping spectrum and we have relied completely on the gas chromatography measurements and did not attempt to integrate or deconvolute these peaks in this NMR spectrum. We had the answer already, and the peaks for those three isomers are in here because it is the same material that is being used to determine the two different results.
Prof Black also seemed to accept what Prof Easton had said, adding in relation to the 1H NMR spectrum “I think all you can say is that there is an indication”.
On the basis – which I take to be uncontroversial – that the “mixture” referred to was a mixture of regioisomers, the evidence to which I have referred would justify the statement that 1H NMR was inconclusive in identifying the product of Example 5(A) as a mixture. “Conclusive” is, of course, a strong term which implies an absolute. Anything short of that standard is to be regarded as “inconclusive”. That is, in my view, as accurate a description as one might get of the opinions of the scientists called in this case about the utility of 1H NMR in the analysis of the 5(A) material. I am satisfied, therefore, that the statement in the letter of 20 June 1997 was correct. I consider that it was the inventor’s declaration of 30 January 1995, rather than the letter of 20 June 1997, which may have conveyed a false impression in this respect.
While I have dealt above with Alphapharm’s submissions on false suggestion in terms, counsel for the Sigma respondents sought to have their own submissions similarly understood. I do not need to deal separately with the latter.
INFRINGEMENT
It was common ground as between AMR and each of the respondents that the latter produced and sold into the Australian market products that were amply within the 2% substantial purity tolerance of the patent in suit as I have construed it. The respondents accepted that, if their clarity point were not upheld, then, subject always to their other challenges to the validity of the patent, they had infringed claims 1, 6, 7, 8, 9 and 10 of the patent in suit. However, because of the findings I have made on the respondents’ validity cases, AMR’s infringement case under these claims cannot be upheld. The method claim – claim 11 – is in a different category. However, since there was no evidence of the methods used by the respondents to produce fexofenadine, this aspect of AMR’s infringement case must also be rejected.
DISPOSITION OF THE PROCEEDING
It follows from my reasons above that, so far as it relates to claims 1, 6, 7, 8, 9 and 10, the patent in suit should be revoked under s 138(3)(b) of the Patents Act for lack of novelty. It also follows that AMR’s infringement proceeding should be dismissed. There is a question whether the patent as a whole should be revoked for false suggestion under s 138(3)(d). I received no submissions specifically on this aspect, but I assume that the parties will address it in the written submissions which I shall require them to file as to the terms of the orders required to reflect these reasons, and as to costs.
I certify that the preceding two hundred and twenty-three (223) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Jessup. Associate:
Dated: 18 February 2011
15
19
2