Home | Databases | WorldLII | Search | Feedback Federal Court of Australia

You are here: 

Cipla Australia Pty Ltd v Novo Nordisk A/S [2024] FCA 1414 (12 December 2024)

Last Updated: 12 December 2024

FEDERAL COURT OF AUSTRALIA

Cipla Australia Pty Ltd v Novo Nordisk A/S [2024] FCA 1414

File number:	NSD 516 of 2024
Judgment of:	PERRAM J
Date of judgment:	12 December 2024
Catchwords:	PATENTS – patent for formulations of liraglutide – validity of extension of term of patent – where application for extension must concern a patent which discloses a ‘pharmaceutical substance per se’ under s 70 of the Patents Act 1990 – whether ‘pharmaceutical substance’ includes formulations of active ingredients and excipients – whether excipients required to be for distinct therapeutic use
Legislation:	Acts Interpretation Act 1901 (Cth) s 11B Intellectual Property Laws Amendment Act 1998 (Cth) Intellectual Property Laws Amendment Act 2006 (Cth) Intellectual Property Laws Amendment Act 2015 (Cth) Patents (World Trade Organization Amendment) Act 1994 (Cth) Patents Act 1903 (Cth) s 84 Patents Act 1952 (Cth) ss 6, 90, 94 Patents Act 1990 (Cth) ss 70, 71, 77, 78, 119A, 192, 223(2), 230, Sch 1  Patents Amendment Act 1989  (Cth) Therapeutic Goods Act 1989 (Cth) s 3 Customs (Prohibited Imports) Regulations 1956 (Cth) reg 2 Patents Amendment Bill 1988 (Cth) Patents Amendment Bill 1989 (Cth)
Cases cited:	Alphapharm Pty Ltd v H Lundbeck A/S [2008] FCA 559 Alphapharm Pty Ltd v H Lundbeck A/S [2014] HCA 42; 254 CLR 247 Baini v R [2012] HCA 59; 246 CLR 469 Boehringer Ingelheim International GmbH v Commissioner of Patents [2001] FCA 647; 112 FCR 595 Boehringer Ingelheim International v Commissioner for Patents [2000] FCA 1918 Commissioner of Patents v AbbVie Biotechnology Ltd [2017] FCAFC 129; 253 FCR 436 Commissioner of Stamps v Telegraph Investment Co Pty Ltd [1995] HCA 44; (1995) 184 CLR 453 H Lundbeck A/S v Alphapharm Pty Ltd [2009] FCAFC 70; 177 FCR 151 Hill v Zuda Pty Ltd [2022] HCA 21; 275 CLR 24 Novartis AG v Pharmacor Pty Ltd (No 3) [2024] FCA 1307 Pharmacia Italia SpA v Mayne Pharma Pty Ltd [2006] FCA 305; 69 IPR 1 Prejay Holdings Ltd v Commissioner of Patents [2003] FCAFC 77; 57 IPR 424 Project Blue Sky Inc v Australian Broadcasting Corporation [1998] HCA 28; 194 CLR 355 Spirit Pharmaceuticals Pty Ltd v Mundipharma Pty Ltd [2013] FCA 658; 216 FCR 344
Division:	General Division
Registry:	New South Wales
National Practice Area:	Intellectual Property
Sub-area:	Patents and associated Statutes
Number of paragraphs:	219
Date of last submissions:	11 November 2024
Date of hearing:	30 October-1 November 2024
Counsel for the Applicant/Cross-Respondent:	Mr D Larish and Ms J McKenna
Solicitor for the Applicant/Cross-Respondent:	Maddocks
Counsel for the Respondent/Cross-Claimant:	Mr J Cooke SC and Ms M Evetts
Solicitor for the Respondent/Cross-Claimant:	Corrs Chambers Westgarth

ORDERS

THE COURT ORDERS THAT:

1. The parties bring in short minutes of order to give effect to these reasons within 7 days.

REASONS FOR JUDGMENT

PERRAM J:

INTRODUCTION

1. This case concerns Australian Patent No. 2004290862 (‘the 862 Patent’) which is a patent for formulations of liraglutide, a peptide used to treat type 2 diabetes. The question is whether an entry in the Register of Patents recording that the 862 Patent expires on 26 August 2025 should be rectified under s 192(1) of the Patents Act 1990 (Cth) (‘the Act’ or ‘the 1990 Act’) so that it records instead that the patent expires just over eight months earlier, on 18 November 2024. Section 192 provides:

192 Orders for rectification of Register
(1) A person aggrieved by:

(a) the omission of an entry from the Register; or
(b) an entry made in the Register without sufficient cause; or
(c) an entry wrongly existing in the Register; or
(d) an error or defect in an entry in the Register;

may apply to a prescribed court for an order to rectify the Register.
(2) On hearing an application, the court may:

(a) decide any question which it is necessary or expedient to decide in connection with the rectification of the Register; and

(b) make any order it thinks fit for the rectification of the Register.

(3) The Commissioner:

(a) must be given notice of an application; and
(b) may appear and be heard in the proceedings; and
(c) must appear if directed to do so by the court.

(4) An office copy of an order must be served on the Commissioner by the Registrar or other appropriate officer of the court.

(5) On receiving an office copy of an order, the Commissioner must rectify the Register accordingly.

2. The Register had formerly recorded that the 862 Patent expired on 18 November 2024. Following a successful patent term extension application made in February 2011 by the Respondent (‘Novo Nordisk’), however, the Register was amended to record, as it currently does, that the 862 Patent would expire on 26 August 2025. The Applicant (‘Cipla’) alleges that the entry of this extension of term was made without sufficient cause by the Commissioner for Patents (i.e. s 192(1)(b)), is wrongly existing (i.e. s 192(1)(c)) or is an error (s 192(1)(d)).
3. Cipla’s broad submission was that the extension was wrongly granted by the Commissioner. The extension regime is contained in Part 3 of Chapter 6 of the Act. Within Part 3, s 70 allows a patentee to apply to the Commissioner for an extension of the term of a patent but only if, inter alia, at least one ‘pharmaceutical substance per se’ is both disclosed in the patent’s complete specification and falls within its claims.
4. The 862 Patent is set out in more detail at [26]-[27] below but in broad outline it is a patent for two formulations involving liraglutide (an active pharmaceutical ingredient used to treat type 2 diabetes) together with excipients. The Macquarie Dictionary defines an excipient as ‘a more or less inert substance, as sugar, jelly, etc., used as the diluent or vehicle for the administration of a medicine, as in a tablet or capsule’. The first formulation consists of liraglutide and two excipients (disodium phosphate dihydrate and propylene glycol). The second formulation is the same but adds another excipient, phenol.
5. Cipla submitted that a formulation cannot be a ‘pharmaceutical substance’ for the purposes of s 70 because of the way in which that expression is defined in Schedule 1 to the Act. It did not dispute that if either formulation is a pharmaceutical substance then it is also a ‘pharmaceutical substance per se’ for the purposes of s 70. Although not directly relevant to this case, it is useful to know for some of the arguments to be considered that the words ‘per se’ have been held by the Full Court of this Court to entail that an extension cannot be granted for a patent which discloses claims for a method and is confined (subject to presently immaterial exceptions) to patents claiming products: Boehringer Ingelheim International GmbH v Commissioner of Patents [2001] FCA 647; 112 FCR 595 (‘Boehringer (FC)’) at [37] per Wilcox, Whitlam and Gyles JJ, affirming the decision of Heerey J in Boehringer Ingelheim International v Commissioner for Patents [2000] FCA 1918 (‘Boehringer’); Commissioner of Patents v AbbVie Biotechnology Ltd [2017] FCAFC 129; 253 FCR 436 at [49] per Besanko, Yates and Beach JJ. The definition of ‘pharmaceutical substance’ in Schedule 1 is in these terms:

pharmaceutical substance means a substance (including a mixture or compound of substances) for therapeutic use whose application (or one of whose applications) involves:

(a) a chemical interaction, or physico‑chemical interaction, with a human physiological system; or

(b) action on an infectious agent, or on a toxin or other poison, in a human body;

but does not include a substance that is solely for use in in vitro diagnosis or in vitro testing.

6. It is not in dispute that the formulations disclosed by the 862 Patent are, within the wording of the definition, ‘mixtures’ of liraglutide, disodium phosphate dihydrate, propylene glycol and phenol (as relevant), albeit it is in dispute whether the formulations are mixtures within the meaning of the definition above. The difference between the parties is whether formulations, that is to say, mixtures of active ingredients and excipients such as those disclosed by the 862 Patent, are excluded by the definition of pharmaceutical substance (as Cipla contends) or are encompassed by it (as Novo Nordisk contends).
7. If the latter construction is correct, then on Cipla’s ‘alternative case’ it becomes necessary to consider at a factual level whether each of the excipients satisfies the balance of the definition. If they all do, then the formulations will be pharmaceutical substances. If any of them does not, however, then the formulations will not be pharmaceutical substances.
8. Two issues therefore arise for determination. The first is whether, properly construed, the definition of ‘pharmaceutical substance’ in the Schedule to the Act includes formulations. The second issue only arises if the first is resolved in the affirmative and has two limbs: (a) whether the excipients in a formulation must have (individually or together) a therapeutic use distinct from that of the active ingredient; and (b) whether, assuming that to be so, the excipient components in the formulations claimed in the 862 Patent are all substances which each satisfy the balance of the definition of ‘pharmaceutical substance’ in Schedule 1.
9. In January 2025 Cipla proposes to launch a product that contains a formulation materially identical to those disclosed in the 862 Patent. If it does so before the 862 Patent expires, it is not in dispute that Cipla will infringe the claims of that patent. The currently recorded expiry date for the 862 Patent will therefore delay Cipla’s product launch by around 7 months. It is for that reason that Cipla now seeks to have the Register rectified to record a patent term expiry date of 18 November 2024 rather than 26 August 2025. For its part, Novo Nordisk has cross-claimed for injunctive relief to prevent Cipla from launching its product on the basis that its sale will infringe the 862 Patent.
10. The trial was conducted on the basis that the resolution of the two issues I have identified above would resolve the entitlement of both parties to the relief they seek. If either formulation disclosed in the 862 Patent is a pharmaceutical substance then Cipla’s rectification suit will fail and Novo Nordisk will be entitled to the injunctive relief it seeks up until 26 August 2025. On the other hand, if neither formulation is a pharmaceutical substance, then Cipla’s rectification suit will succeed and Novo Nordisk’s claims for injunctive relief will be dismissed.

Liraglutide

11. At the heart of the dispute, although not its direct subject matter, is a drug known as liraglutide. Liraglutide is used to treat type 2 diabetes, a chronic condition which is commonly associated with hyperglycaemia, that is to say, persistently high blood sugar levels. Glucagon-like peptide-1 is a peptide hormone which stimulates the release of insulin from pancreatic beta cells (a peptide is a short chain of amino acids that are linked together by peptide bonds). Glucagon-like peptide-1 is usually abbreviated to ‘GLP-1’. GLP-1 stimulates the secretion of insulin by binding on to the GLP-1 receptor in cells in the pancreas known as pancreatic beta cells (it also binds on to cells in other parts of the body such as the brain). In turn, insulin then binds on to receptors in cells throughout the body causing them to absorb glucose and lowering blood sugar levels. In the liver it has the additional effect of preventing the production of glucose.
12. Liraglutide is an analogue of GLP-1 produced by recombinant DNA technology. Its molecular composition is 97% the same as that of GLP-1 produced by the human body (‘native GLP-1’). The molecular formula for liraglutide is C₁₇₂H₂₆₅N₄₃O₅₁ and its molecule looks like this:

13. The are two differences between liraglutide and native GLP-1. The first is that in liraglutide an amino acid at position 34 (lysine) has been replaced with another amino acid, arginine (symbol in the diagram ‘R’). The second is that at position 26 a palmitic acid group has been attached via a glutamyl spacer to the lysine amino acid at that position.
14. Liraglutide is sufficiently similar to native GLP-1 that it will bind to the GLP-1 receptor stimulating the secretion of insulin and thereby causing blood sugar levels to drop. However, the differences at positions 26 and 34 give liraglutide a longer half-life than native GLP-1 and it is therefore able to be used as a daily administered drug to treat hyperglycaemia.

The Patent for Liraglutide

15. From 22 August 1997 Novo Nordisk was the registered owner of the now-expired Australian Patent No. 732957 which was entitled ‘GLP-1 derivatives’ (‘the 957 Patent’). It is not in dispute that claim 47 of the 957 Patent is a claim for the chemical compound liraglutide. Its standard term of 20 years was due to expire on 22 August 2017. However, because the 957 Patent was certainly a patent for a pharmaceutical substance it could, in certain circumstances, be extended beyond that 20 year period under s 70 of the Act. I will return to the provisions concerning such extensions in much more detail later in these reasons, but for present purposes it may be noted that broadly speaking such extensions are granted because of the time it can take to get regulatory approval to bring to market a drug containing the patented substance. The duration of any such extension is a function of the period between the date of the filing of the patent application and the date upon which regulatory approval is finally obtained. On 21 December 2010, Novo Nordisk applied for and was granted an extension of the 957 Patent to 22 August 2022. Since 22 August 2022, Novo Nordisk’s rights in relation to its liraglutide product (known as Victoza) have been confined to those conferred by the 862 Patent (which is for formulations containing liraglutide and not for liraglutide itself).

Formulations

16. It is not in dispute that liraglutide cannot be administered orally since, as a peptide, it will be broken down in the gastro-intestinal tract. The 957 Patent refers to the possibility of liraglutide being administered subcutaneously, that is to say, into the innermost tissue layer of the skin which lies closest to the muscle. It is not in dispute that for liraglutide to be administered subcutaneously by means of an injection it must be ‘formulated’. Formulation is the process of combining an active ingredient, such as liraglutide, with other substances to facilitate its administration. As already mentioned, the other substances are usually referred to as ‘excipients’. It is convenient, although as this case shows not altogether without difficulty, to refer to the drug substance that brings about the therapeutic effect (for example, liraglutide) as the active pharmaceutical ingredient (or ‘API’) and to say of the excipients that they are not the active ingredients. The factual question which arises in this case is whether the non-active ingredients in the formulations of liraglutide disclosed in the 862 Patent have therapeutic effects in their own right.

Novo Nordisk’s Formulation of Liraglutide

17. Prior to 2004, Novo Nordisk conducted Phase I and II trials on formulations containing liraglutide. The formulations used in these trials involved the use of mannitol, a sugar alcohol, as an excipient. The role of mannitol in the formulation was as a tonicity agent which means that it was intended to ensure that the formulation was isotonic such that the concentration of osmotically active solutes in the solution was equal to the concentration of solutes in the interstitial fluid (that is, the fluid in the tissue between the skin and the muscles). A difficulty with mannitol was encountered. During the production process crystals of mannitol sometimes formed and these could go on to block the injecting device. In 2004 Novo Nordisk overcame this problem by replacing mannitol with propylene glycol. In the same year, it obtained the 862 Patent to protect that not altogether astonishing discovery (the patent is discussed in the next section). The point for present purposes is that this patent is concerned with formulations of liraglutide but is not itself a patent for liraglutide. A person who made a different liraglutide formulation to Novo Nordisk’s patented formulation would infringe its liraglutide patent (the 957 Patent) whilst it remained in force, but would not infringe its formulation patent (the 862 Patent).
18. A formulation using propylene glycol known as ‘Formulation 4’ proceeded to Phase III trials between August 2006 and February 2008. This trial involved thousands of patients in multiple countries. On 15 July 2008 Novo Nordisk submitted Formulation 4 to the Therapeutic Goods Administration (‘the TGA’) for marketing approval and received approval on 18 August 2010. It will be observed that this was 13 years after Novo Nordisk had first been granted the 957 Patent which claimed the compound liraglutide.
19. Formulation 4 was registered on the Australian Register of Therapeutic Goods on 26 August 2010. It then entered the Australia market under the brand name Victoza. Victoza is supplied in the form of a pre-filled injecting pen. There are two formulations of Victoza approved by the TGA for marketing in Australia. These are the two formulations claimed by the 862 Patent which it will be recalled are, first, a formulation comprising liraglutide and two excipients, propylene glycol and disodium phosphate dihydrate; and, secondly, an identical formulation which also contains a third excipient, phenol.
20. As already noted, there is a factual dispute between the parties as to whether these three excipients contribute a therapeutic effect in their own right. At this stage it may be noted, however, that Novo Nordisk makes no such claims in the product information documents it has supplied to various regulators. At the trial, the experts agreed that the excipients had certain roles in the formulations but there was a difference between them in relation to other roles. As to the agreed roles, both experts agreed that phenol was a preservative, that disodium phosphate dihydrate was a buffering agent (to achieve a particular pH level) and that propylene glycol was a tonicity agent.
21. As to the roles which were not agreed, the dispute arose from the evidence given for Novo Nordisk by Professor Bernkop-Schnürch. His evidence was that after injection, the phenol and disodium phosphate dihydrate interact with liraglutide in the interstitial fluid to prevent the formation of large oligomers (of liraglutide) in that fluid. By way of explanation, depending on the conditions in which it finds itself, liraglutide has a tendency to ‘self-associate’ which means that it tends to bind to itself. In so doing, it forms chains of itself known as oligomers and, depending on how much self-associating it gets up to, these oligomers may be large or small. In the worst case, the large oligomers may bind together into even larger structures known as fibrils which can be visible to the human eye.
22. The point for present purposes is that large oligomers of liraglutide are too big to be transported into the bloodstream either via the blood capillaries or, alternatively, via the lymphatic capillaries. According to Professor Bernkop-Schnürch, this means that the phenol and disodium phosphate dihydrate increase the uptake of the liraglutide into the bloodstream and thereby contribute to the goal of getting the liraglutide to where it needs to be, namely, the GLP-1 receptor. This is a somewhat simplified version of Professor Bernkop-Schnürch’s account but will do for present purposes. Cipla called Associate Professor Keyser who did not agree with this account for various reasons.
23. Thus was the debate about phenol and disodium phosphate dihydride. In relation to the role of the third excipient, propylene glycol, the parties thought that they had reached agreement about what it did above and beyond its role as a tonicity agent. But on the second day of the trial, it emerged that their consensus was shallow and concealed a reef of disagreement. Fortunately, the debate about any role it had above and beyond achieving isotonicity emerged just before the luncheon adjournment on the second day of the trial which, as it happens, was midway through the concurrent evidence of Professor Bernkop-Schnürch and Associate Professor Keyser. After lunch, the concurrent evidence session was then rejigged to provide for the ventilation of this issue.
24. As the debate emerged, the question seemed to be whether by maintaining isotonicity the propylene glycol had the effect of avoiding pain at the injection site. If isotonicity was not maintained, then a phenomenon known as osmotic shock could ensue which, as I understood it, resulted from osmosis in the interstitial fluid leading to swelling and pain. On this view, propylene glycol had a therapeutic effect because of what it did not do. I will return to this topic in more detail later.
25. In any event, in light of its astonishing discovery that the formation of mannitol crystals could be avoided by not including mannitol in its formulation, Novo Nordisk now filed a second patent, the 862 Patent. It is this patent which forms the subject matter of the present litigation. It is not suggested in this proceeding that the use of propylene glycol rather than mannitol lacks an inventive step or is not novel.

A Patent for Formulations of Liraglutide

26. On 18 November 2004 Novo Nordisk filed an application for the 862 Patent, entitled ‘Propylene Glycol-containing peptide formulations which are optimal for production and for use in injection devices’. The patent was issued with effect from that date with a standard term of 20 years. Its expiry date was 18 November 2024. In February 2011, Novo Nordisk applied for and obtained an extension of this patent to 26 August 2025. Claim 1 of the 862 Patent is for:

A pharmaceutical formulation comprising at least one GLP-1 agonist and propylene glycol and a buffer which is disodium phosphate dihydrate, wherein said propylene glycol is present in said formulation in a final concentration of from about 1 mg/ml to about 100 mg/ml and wherein said formulation has a pH of from about 7.0 to about 10.0.

27. Claims 2 to 14 encompass a range of formulations of a GLP-1 agonist. The GLP-1 agonist does not have to be liraglutide in claim 1 but later claims include claims for formulations of liraglutide. For the balance of the reasons, I will refer only to the liraglutide formulation claims. Two sets of formulations of liraglutide are claimed:

(a) formulations of liraglutide with a ‘tonicity agent’ of propylene glycol and a ‘buffer’ of disodium phosphate dihydrate with various concentrations of the former and with fluctuating pH levels resulting from the application of the latter (claims 1-8); and

(b) formulations of the kind in (a) but with a preservative added of varying concentrations (claims 8-9). The specification states at p 14 that in a preferred embodiment of the invention the relevant preservative is phenol or m-cresol.

28. It is not in dispute that the formulations in Victoza fall within the claims of the 862 Patent. This is important because, as already mentioned, Cipla wishes to bring to market an identical formulation of liraglutide to Victoza which, it is agreed, will infringe the 862 Patent if its term has not expired.
29. The first issue to be resolved is the proper construction of the definition of ‘pharmaceutical substance’.

THE CONSTRUCTION ISSUE

Framework for the construction issue

30. The formal issue between the parties is whether the 862 Patent met the requirements for an extension and in particular whether it could be the subject of an application for an extension. Section 70 of the Act provides for such applications but it is s 70(2)(a) which is of central importance. The full text of s 70 is:

70 Applications for extension of patent

(1) The patentee of a standard patent may apply to the Commissioner for an extension of the term of the patent if the requirements set out in subsections (2), (3) and (4) are satisfied.

(2) Either or both of the following conditions must be satisfied:

(a) one or more pharmaceutical substances per se must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification;

(b) one or more pharmaceutical substances when produced by a process that involves the use of recombinant DNA technology, must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification.

(3) Both of the following conditions must be satisfied in relation to at least one of those pharmaceutical substances:

(a) goods containing, or consisting of, the substance must be included in the Australian Register of Therapeutic Goods;

(b) the period beginning on the date of the patent and ending on the first regulatory approval date for the substance must be at least 5 years.

Note: Section 65 sets out the date of a patent.
(4) The term of the patent must not have been previously extended under this Part.
Meaning of first regulatory approval date

(5) For the purposes of this section, the first regulatory approval date, in relation to a pharmaceutical substance, is:

(a) if no pre‑TGA marketing approval was given in relation to the substance—the date of commencement of the first inclusion in the Australian Register of Therapeutic Goods of goods that contain, or consist of, the substance; or

(b) if pre‑TGA marketing approval was given in relation to the substance—the date of the first approval.

(5A) For the purposes of paragraph (5)(a), disregard an inclusion in the Australian Register of Therapeutic Goods of goods that contain, or consist of, a pharmaceutical substance if the inclusion was sought for the sole purpose of exporting the goods from Australia to address a public health problem in an eligible importing country:

(a) in circumstances of national emergency or other circumstances of extreme urgency; or

(b) by the public non‑commercial use of the goods.

Note: This subsection also applies in relation to an application for an extension of the term of a standard patent (see paragraph 71(2)(b)).
Meaning of pre‑TGA marketing approval

(6) For the purposes of this section, pre‑TGA marketing approval, in relation to a pharmaceutical substance, is an approval (however described) by a Minister, or a Secretary of a Department, to:

(a) market the substance, or a product containing the substance, in Australia; or

(b) import into Australia, for general marketing, the substance or a product containing the substance.

31. The debate is whether the two formulations of liraglutide disclosed in the claims of the 862 Patent are claims for a ‘pharmaceutical substance’. Whilst it will be necessary to return to what a ‘pharmaceutical substance per se’ in s 70(2)(a) means, it is not in dispute in this case that if the 862 Patent discloses a claim for a pharmaceutical substance then it also discloses a claim for a pharmaceutical substance per se. The definition of ‘pharmaceutical substance’ is set out at [5] above. It must be read with the definition of ‘therapeutic use’ in the Schedule which provides:

therapeutic use means use for the purpose of:

(a) preventing, diagnosing, curing or alleviating a disease, ailment, defect or injury in persons; or

(b) influencing, inhibiting or modifying a physiological process in persons; or
(c) testing the susceptibility of persons to a disease or ailment.

32. As already noted, the dispute between the parties is whether the definition of ‘pharmaceutical substance’ includes formulations. The task therefore is one of statutory construction.
33. As Cipla developed the argument, its broad outline had five components. First, there were two appellate authorities which bound this Court to the conclusion that a formulation could not be a pharmaceutical substance. These were the Full Court of this Court’s decision in H Lundbeck A/S v Alphapharm Pty Ltd [2009] FCAFC 70; 177 FCR 151 (‘Alphapharm (FC)’) at [238], [242] and [244] per Bennett J (with whom Middleton J agreed) and the High Court’s decision in Alphapharm Pty Ltd v H Lundbeck A/S [2014] HCA 42; 254 CLR 247 (‘Alphapharm (HC)’) at [23] footnote 40 per Crennan, Bell and Gageler JJ.
34. Secondly, if the Court was not bound by those decisions, then the ordinary meaning of the definition of a pharmaceutical substance excluded formulations.
35. Thirdly, the legislative history of the extension provisions and the definition of ‘pharmaceutical substance’ showed that a formulation could not be a pharmaceutical substance, as did some of the secondary materials relating to that legislative history. There are many planks in this submission but one should be noted at the outset. It concerns the introduction into the Act of s 119A in 2006. That provision is not part of the extension provisions in Chapter 6 Part 3 but is instead found in Chapter 11 (entitled ‘Infringement’) and Part 1 thereof (entitled ‘Infringement and infringement proceedings’). Section 119A concerns ‘pharmaceutical patents’ and provides that a person does not infringe such a patent by exploiting an invention claimed in the patent if the exploitation is solely for the purpose of obtaining inclusion on the Register of Therapeutic Goods. I will return to s 119A in more detail later in these reasons but the idea underpinning it is to permit generic drug companies to take steps to obtain registration of a generic product on the Register of Therapeutic Goods whilst the patent of the originator company is still on foot so that the generic company may immediately enter the market upon the expiry of the patent. Colloquially, this is referred to as springboarding. Cipla submitted that the terms of s 119A are inconsistent with the proposition that a formulation may be a pharmaceutical substance. The details of this submission are quite complex.
36. Fourthly, the underlying policy of the extension provisions was said to support Cipla’s construction.
37. Finally, Cipla acknowledged that two first instance judgments of this Court had accepted that a formulation was a pharmaceutical substance: Pharmacia Italia SpA v Mayne Pharma Pty Ltd [2006] FCA 305; 69 IPR 1 (‘Pharmacia’) and Spirit Pharmaceuticals Pty Ltd v Mundipharma Pty Ltd [2013] FCA 658; 216 FCR 344 (‘Spirit’). In relation to these two decisions, Cipla submitted that both were decided before the High Court’s judgment in Alphapharm and considered versions of the Act which pre-dated the introduction of s 119A. They were thus to be distinguished. In the event this was not a sufficient basis to distinguish the two decisions, Cipla submitted that the decisions are plainly wrong and should not be followed.
38. For the reasons which follow, I conclude that a pharmaceutical substance includes a formulation and has done so since it first appeared in the Act in 1989. The structure of this section of the reasons is as follows:
    • Legislative History;
    • The Policy of the Extension Provisions;
    • The Ordinary Meaning of Pharmaceutical Substance;
    • Whether Meaning Affected by Other Provisions of the Act;
    • Case Law Concerning the Meaning of Pharmaceutical Substance; and
    • Conclusion.

Legislative History

39. The High Court outlined the legislative history of the extension provisions in the Act in Alphapharm at [42]-[60] but not in a way which was focussed on the meaning of ‘pharmaceutical substance’. As the Court there observed, ‘[t]he pre-existing law and legislative history should not deflect the Court from its duty to resolve an issue of statutory construction, which is a text-based activity’. As in Alphapharm, however, both parties in the present case sought to obtain benefit from this history. The Court accepted that resort to the legislative history was legitimate in that case (where the issue was different) and it also accepted that ‘questions of policy can inform the Court’s task of statutory construction’: [42].
40. The issue in Alphapharm (HC) was whether the Commissioner had the power to grant an extension of time for the bringing of an application to extend the term of a patent whereas the issue in this case is whether a formulation can be a pharmaceutical substance. Subject to what the legislative history shows, for the reasons given by the High Court in Alphapharm, I accept that a survey of the history of the expression ‘pharmaceutical substance’ should be embarked upon because it may throw light on whether the expression includes a formulation containing a pharmaceutical substance.

The Patents Act 1903 (Cth)

41. The starting point is the Patents Act 1903 (Cth) (‘the 1903 Act’). The 1903 Act contained as Part IV Division 5 an extension scheme covering all patents. A patentee was to petition the High Court or the Supreme Court of the State in which the Patent Office was situated at least six months before the patent lapsed: s 84(1). If the Court was satisfied, inter alia, that the patentee had been inadequately remunerated by the patent it could order the extension of the patent or grant a new patent which could be up to 7 years in duration and, in exceptional cases, 14 years: s 84. The extension regime did not distinguish between patents for pharmaceutical substances and other patents.

The Patents Act 1952 (Cth)

42. The Patents Act 1952 (Cth) (‘the 1952 Act’) largely retained the features of the extension scheme in the 1903 Act noted above save that the relevant provisions were now relocated to a Part IX, provided only for petitions to the High Court (not any Supreme Court) and changed the ordinary term of an extension to 5 years and the extraordinary term to 10: s 94. As with the 1903 Act, the provisions of Part IX did not draw any distinction between patents for pharmaceutical substances and other patents. Part IX was amended on 9 December 1976 to permit a ‘prescribed court’ to deal with such petitions. A ‘prescribed court’ was defined to be the Federal Court, the Supreme Court of a State, the Supreme Court of the Australian Capital Territory, the Supreme Court of the Northern Territory or the Supreme Court of Norfolk Island: s 6.

The Report of the Industrial Property Advisory Committee in 1984

43. The Industrial Property Advisory Committee reviewed the 1952 Act and reported to the Minister for Science and Technology on 29 August 1984. This report concluded at p 39 that the case for having extension provisions for patents was empirically unsupported and recommended the complete repeal of the extension provisions.

The Patents Amendment Bill 1988 (Cth)

44. The recommendations of the Industrial Advisory Committee were not acted upon. Instead, on 3 June 1988 the then Minister for Science, Mr Barry Jones, introduced a bill into the House of Representatives entitled the Patents Amendment Bill 1988 (Cth). The parties were agreed that the bill was never assented to and consequently did not become law. Had it become law, this bill would have made substantial amendments to the extension regime. There is a dispute about whether the terms of the bill and the explanatory memorandum which accompanied it may be used as aids to construction. As a general proposition I agree that the terms of a bill which is not enacted will not be a useful aid to construction and the same may be said of any explanatory memorandum which accompanied it. I would not exclude the perhaps obscure possibility that reference in later legislation to unpassed earlier legislation might, in an appropriate case, have some work to do and the same process of reference may make relevant something said in an explanatory memorandum (or second reading speech) relating to legislation which was not enacted. In this case, the only relevance of the Patents Amendment Bill 1988 is historical in that it shows that an earlier attempt at an extension regime for patents claiming pharmaceutical substances had been made a year before such a regime was introduced into the 1952 Act by the  Patents Amendment Act 1989  (Cth). I will therefore omit at this stage any reference to the Patents Amendment Bill 1988 and will mention it again only if it becomes necessary.

The  Patents Amendment Act 1989  (Cth)

45. Following the lapsing of the Patents Amendment Bill 1988 (Cth) there was then passed the  Patents Amendment Act 1989  (Cth). It substituted a new  Part IX  into the Patents Act 1952 which provided for extensions of patents for pharmaceutical substances. The central provision was s 90(1):

Where:

(a) a pharmaceutical substance is in substance disclosed in the complete specification of a standard patent and in substance falls within the scope of the claim or claims of that complete specification; and

(b) the patentee has requested the issue of a marketing approval certificate in respect of that substance;

the patentee may, by notice in writing in accordance with the prescribed form given to the Commissioner not later than 12 months before the end of the term of the patent, apply for an extension of the term of the patent in respect of that substance and any other pharmaceutical substance which is in substance disclosed in the specification and in substance falls within the scope of the claim or claims of the specification.

46. The definitions of ‘marketing approval certificate’ and ‘pharmaceutical substance’ inserted into s 6 were as follows:

‘marketing approval certificate’, in relation to a pharmaceutical substance, means a certificate given by the Secretary to the Department of Community Services and Health certifying that he or she has approved the marketing of the substance, or a product containing the substance, in Australia;
‘pharmaceutical substance’ means a substance (including a mixture or compound of substances) for therapeutic use the application, or one of the applications, of which involves a chemical interaction, or physico-chemical interaction, with a human physiological system, or involves action on an infectious agent, or on a toxin or other poison, within a human body, but does not include a substance that is solely for use in in vitro diagnosis or in vitro testing ... .

47. It will be seen that the definition of ‘marketing approval certificate’ observes a distinction between a pharmaceutical substance and a product containing a pharmaceutical substance. I do not consider that this throws any light on the meaning of pharmaceutical substance since the range of products which may contain pharmaceutical substances is capable of extending to items such as bottles and dispensers.
48. In relation to the definition of ‘pharmaceutical substance’ the following was said in the explanatory memorandum accompanying the Patents Amendment Bill 1989 (‘the 1989 Explanatory Memorandum’) at [8]:

The definitions of “pharmaceutical substance” and “therapeutic use” have the effect that the new patent extension scheme will apply only to pharmaceuticals for human use. The scheme will be available for “therapeutic substances” in the terms of the Customs (Prohibited Imports) Regulations, with added limitations that:
· only substances for use in relation to humans are included (those solely for veterinary use would not be included);
· only substances whose use involves a chemical or physico-chemical interaction with a human physiological system, or involves action on an infectious agent, or on a toxin or other poison, within a human body, are included (devices such as surgical ligatures are not included);
· substances whose sole use is in in vitro diagnosis or testing are not included.

49. This discloses a clear intention to adopt the meaning of ‘therapeutic substance’ contained in the Customs (Prohibited Imports) Regulations 1956 (Cth) with the three added limitations set out. It will be observed that in the explanatory memorandum the three limitations are expressed in terms of the use to which the substance would be put. By comparison, the first two limitations in the definition of ‘pharmaceutical substance’ are expressed in terms of the application of the substance whilst the third is expressed in terms of use. The likely reason for this is that the first two limitations appear immediately after the words ‘for therapeutic use’ in the definition so that the words ‘for therapeutic use whose use’ would have created uncertainty about whether the second use was the same as the first. It may be noted that in the third limitation, the word ‘use’ remains but that in that part of the definition it does not appear in proximity to the words ‘for therapeutic use’.
50. As I will explain later when I reject Cipla’s textual arguments about the meaning of the word ‘application’, I would interpret the word ‘application’ to mean ‘use’ both because of the oddity of Cipla’s proposed construction but also because the 1989 Explanatory Memorandum provides strong support for the proposition that ‘application’ means ‘use’.
51. As I have said, it seems clear, subject to the three use limitations, that it was intended that the definition of ‘pharmaceutical substance’ would operate in the same way as had the definition of ‘therapeutic substance’ in the Customs (Prohibited Imports) Regulations. At the time of the passage of the  Patents Amendment Act 1989  the Customs (Prohibited Imports) Regulations contained this definition of a therapeutic substance in reg 2(1):

therapeutic substance means a substance, including a mixture or compound of substances, that has a therapeutic use and includes a surgical ligature, suture or dressing, but does not include a vaccine prepared from microscopic organisms from the body of a person or animal for use in the treatment of that person or animal only.

52. Regulation 2(2) provided:

(2) For the purpose of regulations 5A to 5G (inclusive) of these Regulations—

(a) each form of a therapeutic substance shall be taken to be a separate and distinct therapeutic substance;

(b) if a therapeutic substance is manufactured according to two or more formulations—the substance manufactured according to a particular formulation shall be taken to be a different therapeutic substance from the substance manufactured according to the other or each other formulation; and

(c) a therapeutic substance having a particular strength shall be taken to be a different therapeutic substance from the substance having a different strength.

53. It will be observed that reg 2(2)(b) made clear that each formulation of a therapeutic substance was taken to be a different therapeutic substance.
54. Regulation 5A then prohibited the import into Australia of various kinds of therapeutic substances without the written permission of the Secretary of the Department of Community Services and Health. The effect of regs 2(1), 2(2)(b) and 5A was that separate written permission was required from the Secretary for the importation of each formulation of a therapeutic substance.
55. Regulation 2(2)(b) assumes that a pharmaceutical substance may be manufactured according to a formulation. But it only operates in circumstances where there are two or more such formulations and has no impact if there is only one such formulation. Its effect, where there is more than one formulation, is to require import permission for each.
56. This strongly suggests that a therapeutic substance must include a formulation of a therapeutic substance. If it did not, the importation into Australia of a single formulation would not be prohibited by reg 5A because it would not be a ‘therapeutic substance’; and yet if there was more than one formulation then the importation of each such formulation would be prohibited (since reg 2(2)(b) would deem each to be a therapeutic substance). This strikes me as a highly unlikely outcome.
57. As it happens, there is no difficulty in construing the definition of ‘therapeutic substance’ so that it includes a formulation. A formulation would be ‘a substance, including a mixture or compound of substances, that has a therapeutic use’.
58. It will be apparent from the definition of ‘pharmaceutical substance’ inserted by the  Patents Amendment Act 1989  that the deeming provision in reg 2(2)(b) was not carried across into it. However, there was no need to do so since no prohibition was imposed by the  Patents Amendment Act 1989  which utilised the definition in such a way as to make it necessary to deem multiple formulations of a pharmaceutical substance to be separate pharmaceutical substances. And, as I have explained, it is obvious that the definition of ‘therapeutic substance’ included a formulation. I detect nothing in the terms of the  Patents Amendment Act 1989  or the 1989 Explanatory Memorandum which provides any support for the idea that a pharmaceutical substance could not include a formulation. Indeed, and to the contrary, it appears to me to be nearly beyond argument that it did.
59. It is useful to observe at this juncture that the definition of ‘pharmaceutical substance’ inserted by the Patents Amendment Act 1989 is nearly the same as the current version that appears in the Schedule to the Patents Act 1990 save for two changes. The first is the replacement in the Patents Act 1990 of the words ‘the application, or one of the applications, of which’ with ‘whose application (or one of whose applications)’. The second is the reformatting of the definition so that it has two subparagraphs.
60. Cipla submitted that I should not conclude that the definition of ‘pharmaceutical substance’ included a formulation in the  Patents Amendment Act 1989  for five reasons. First, the 1989 Explanatory Memorandum paraphrased the definition of ‘therapeutic substance’ in the Customs (Prohibited Imports) Regulations and such a paraphrasing was not helpful in interpreting what the meaning of the definition of ‘pharmaceutical substance’ in the  Patents Amendment Act 1989  was. I do not agree with this submission. The statement at [8] of the 1989 Explanatory Memorandum does not paraphrase what the meaning of ‘therapeutic substance’ in the Customs (Prohibited Imports) Regulations was. It simply says that the extension regime inserted by the  Patents Amendment Act 1989  would be ‘available’ for substances which were therapeutic substances under that regulation subject to the three use limitations set out. Far from being a paraphrasing of different legislation, this is an explicit statement about what the terms of the definition of ‘pharmaceutical substance’ were intended to mean.
61. Thus, whilst I accept Cipla’s submission, based on the warning in Baini v R [2012] HCA 59; 246 CLR 469 (‘Baini v R’) at [14] per French CJ, Hayne, Crennan, Kiefel and Bell JJ, that paraphrasing of statutory language in extrinsic materials ‘may be apt to mislead if attention strays from the statutory text’, I do not consider that warning has any application to the present circumstance since no such paraphrasing is involved.
62. Secondly, Cipla submitted that the definition of ‘pharmaceutical substance’ in the current version of the Act and that in the  Patents Amendment Act 1989  were not in identical terms. In particular, Cipla submitted that the words ‘the application, or one of the applications, of which’ were found in the 1952 Act as amended by the Patents Amendment Act 1989 but not in the 1990 Act.
63. It is true as has already been noted that those words were replaced in the Patents Act 1990 with ‘whose application (or one of whose applications)’. But the articulation of the difference demonstrates sufficiently that it is immaterial and leads nowhere.
64. It may be that Cipla also had in mind that neither phrase is included in the definition of a therapeutic substance in reg 2 of the Customs (Prohibited Imports) Regulations. If this was the argument, then it would be relevant to Cipla’s submission about what meaning should be attributed to the words ‘whose application, or one of whose applications’. I deal with that submission below at [72] but as I have already explained, the history of the provision shows that the word ‘application’ was intended to mean ‘use’. For present purposes, however, it will suffice to say that I do accept that a difference did exist between the definition of ‘pharmaceutical substance’ in the Patents Act 1989 and the definition of ‘therapeutic substance’ in the Customs (Prohibited Imports) Regulations. I will explain later why I do not think that this difference was in any way material.
65. Cipla’s third submission on this topic was that the 1998 and 2006 Explanatory Memoranda did not make any cross-reference to the 1989 Explanatory Memorandum. I have not yet come to those memoranda and I will deal more fully with this submission when I get to those documents. By way of an early instalment, however, I will just say that neither Act to which those explanatory memoranda related (the Intellectual Property Laws Amendment Act 1998 or the Intellectual Property Laws Amendment Act 2006) made any change to the definition of ‘pharmaceutical substance’ and, indeed, the two explanatory memoranda in question exhibit an intention that that definition was not to be changed. A failure in those two memoranda to discuss [8] of the 1989 Explanatory Memorandum is, in that context, quite unsurprising. Indeed, if that omission does anything, it probably underscores that the two later explanatory memoranda exhibit a positive intention that the definition should not be amended.
66. Fourthly, Cipla submitted that if all of that were wrong so that it was legitimate to consult the terms of the 1989 Explanatory Memorandum then, in fact, it showed that a pharmaceutical substance could not include a formulation. The rhetorical force of this submission is perhaps dented by the fact that it was advanced only after much effort on Cipla’s part to show that the memorandum should not be consulted. The presentment that the submission might not have much in it is heightened when attention is turned to the passage on which Cipla now relies, the second bullet point in [8] of the 1989 Explanatory Memorandum quoted at [48] above. Cipla’s point is that a formulation cannot answer that description.
67. No doubt, this is true but it does not matter. As the astute reader will no doubt have already realised, the three bullet points were the limitations which the explanatory memorandum explained were to be applied to the definition of ‘therapeutic substance’ in the Customs (Prohibited Imports) Regulations. In the definition of ‘pharmaceutical substance’ these limitations appear after the words ‘for therapeutic use’. However, the fact that a therapeutic substance within the definition of the Customs (Prohibited Imports) Regulations may include a formulation does not spring from the words of limitation. This is because the words of limitation do not appear in the definition in the Customs (Prohibited Imports) Regulations at all. Indeed, it is their absence from that definition which provided explicit reason why they were included in the definition of ‘pharmaceutical substance’. As I have already explained, the fact that a therapeutic substance includes a formulation under the Customs (Prohibited Imports) Regulations springs from the words ‘including a mixture or compound of substances’. The problem for Cipla is that those words appear in the definition of ‘pharmaceutical substance’ as well.
68. Cipla’s fifth submission was based on the supplementary explanatory memorandum circulated in the lead up to the passage of the  Patents Amendment Act 1989 . At [1] of that document, this statement appears:

The main purpose of the amendments and new clause is to clarify and simplify the procedures contained in the Bill by which extensions of patent term for pharmaceutical substances will be obtained. In particular, the amendments will ensure that the patent extension scheme will cover all patented pharmaceuticals which receive marketing approval from the Department of Community Services and Health (DCSH) – whether or not they are imported; whether or not they had previously been approved in some other form or formulation; and including antibiotics, antivirals and antitoxins. The amendments will enable a patentee to obtain an extension for all pharmaceutical substances claimed in a particular patent – provided that marketing approval is obtained for one of those substances, and notwithstanding that the marketing approval granted may be restricted to a particular product or formulation containing the substance.

69. Cipla relied upon the last sentence which it submitted observes a distinction between a pharmaceutical substance and a formulation of that substance. I do not agree. Recourse to the whole passage shows the opposite. The second sentence shows that the extension scheme was to apply to all patented pharmaceuticals which had received marketing approval. The last sentence shows that those patented pharmaceuticals could include formulations. What [1] shows is an assumption that if a patented pharmaceutical substance could, in whatever form, be subject to an approval then it could be extended. This does not show that a pharmaceutical substance within the definition in the  Patents Amendment Act 1989  did not include a formulation. In any event, however, I do not think that anything that was said at [1] would be capable of dislodging the clear statements appearing at [8] in the 1989 Explanatory Memorandum as to what the definition was intended to mean.

The Patents Act 1990 (Cth)

70. There was then passed the Patents Act 1990 (Cth) which by s 230 repealed the 1952 Act. As originally enacted, Part 3 of Chapter 6 dealt with the extension of patents for pharmaceutical substances. The machinery of these provisions operated on a ‘pharmaceutical substance’ and, at this stage, there was no requirement that the patent claim a pharmaceutical substance per se. ‘Pharmaceutical substance’ was defined in Schedule 1 in these terms:

“pharmaceutical substance” means a substance (including a mixture or compound of substances) for therapeutic use whose application (or one of whose applications) involves:

(a) a chemical interaction, or physico-chemical interaction, with a human physiological system; or

(b) action on an infectious agent, or on a toxin or other poison, in a human body;
but does not include a substance that is solely for use in in vitro diagnosis or in vitro testing

71. This is identical to the earlier definition except that it was now reformatted with two subparagraphs and the foreshadowed substitution of ‘the application, or one of the applications, of which’ with ‘whose application (or one of whose applications)’. As already explained I do not consider that substitution to have had any effect on the definition. Moreover, the explanatory memorandum accompanying the Patents Bill 1990 contains nothing to indicate any intention to alter the meaning of pharmaceutical substance from that which had applied under the  Patents Amendment Act 1989  (Cth).
72. I conclude therefore that, as passed, the definition in the Patents Act 1990 of a pharmaceutical substance included a formulation of a pharmaceutical substance.

Patents (World Trade Organization Amendment) Act 1994

73. Neither party submitted that this Act has any impact on the issue. However, in the interests of completeness, it extended the term of a standard patent to 20 years and, at the same time, repealed the extension regime contained in Chapter 6 Part 3 Division 2. Even though it was no longer attached to any operative provisions, the definition of ‘pharmaceutical substance’ was not repealed.

The Industry Commission’s Report No 51 (The Pharmaceutical Industry)

74. On 9 June 1995 the Assistant Treasurer referred several matters concerning the pharmaceutical industry to the Industry Commission for report. The Commission produced its report, numbered 51 and entitled ‘The Pharmaceutical Industry’, on 3 May 1996. Chapter 16 dealt with intellectual property. Section 16.3 was headed ‘Effective Patent term and patent term restoration’ and section 16.4 was headed ‘Generic springboarding’. In Section 16.3 the Commission surveyed the extension regimes in the United States and Europe. Relevantly, this survey revealed that in Europe an extension could be obtained for one patent per chemical entity whilst in the United States it could be obtained for one patent per drug. As recorded in the report, the Government’s position was that it supported an extension regime for a ‘drug’ which would give a patent for that drug an effective life of 15 years. Whilst I think one can be clear that all parties including the Government, the Commission and industry bodies were clear that only one patent could be extended per substance, I do not extract from the report any clear view of whether that the substance had to be a drug or a chemical entity, still less whether either of those concepts included a formulation.

The Government’s Response to the Report

75. On 8 April 1997 the Government announced its response to the Report. The Minister issued a press release in these terms:

The Minister also announced the Patents Act 1990 will be amended to provide for extensions of term of up to five years for 20-year standard pharmaceutical patents and will also allow springboarding (preparation for generic copying of drugs) for eligible patents. This will bring Australia into line with international best practice.
The announcement forms part of the Government’s final response to the Industry Commission report on the pharmaceutical industry.
“The Government is determined Australia will maintain an internationally competitive operating environment for its pharmaceutical industry into the 21st century,” Mr Moore said.
“This is a knowledge-based, technology-intensive industry which is uniquely placed to develop and commercialise the results of long term investment in medical research.”
The Minister said a key element of the Government’s industry policy is ensuring Australia is an internationally competitive location for investment.
“Our decision supporting the future of the pharmaceutical industry is a reflection of this priority.
“The pharmaceutical industry is a major employer and the Government wishes to encourage these companies to plan their future investment with confidence. Our decision is designed to encourage the companies to invest, develop and create additional skilled jobs for Australians.”
Mr Moore said that two of the most important factors which shape investment in the pharmaceutical industry on an international basis are: the extent of price controls and the competitiveness of a country’s patents regime.
“The Government’s final response to the Commission’s report addresses these two key issues.”

76. Attached to the press release was this more detailed statement:

The Government will introduce a new extension of term scheme for pharmaceutical patents in recognition that the long development times and regulatory requirements for new pharmaceuticals significantly erode the time available under patent to exploit the invention. In line with the situation in the US, the EU and Japan, extensions of up to five years will be available for 20 year standard term patents for new pharmaceutical substances.
Where extensions are granted, third parties will be allowed to undertake activities solely for the purposes of meeting pre-marketing regulatory requirements without infringing the patents. This is in recognition of the legitimate interests of generic drug manufacturers in Australia.
Implementation details will be settled in consultation with industry and relevant Government departments.

Intellectual Property Laws Amendment Act 1998 (Cth)

77. The Patents Act 1990 was next amended by the Intellectual Property Laws Amendment Act 1998 (Cth) which was the legislation foreshadowed by the Minister in the press release above. In the explanatory memorandum which accompanied the bill (‘the 1998 Explanatory Memorandum’), under the heading ‘Background’, this passage appears:

In 1993, the former Government announced that Australia would introduce an extension of term scheme for pharmaceutical patents. A number of options were developed to provide extensions of up to five years to the standard 20 year patent term, however, a decision on which option should be implemented was not reached prior to the election in March 1996. Since then, further consultations have been undertaken on both the impact of options for providing, as well as the impact of not providing, an extension of term scheme.
The Minister for Industry, Science and Tourism, in presenting the Government’s final response to the Industry Commission’s Report No. 51, The Pharmaceutical Industry, announced in April 1997 that the Patents Act 1990 would be amended to provide for an extension of term of up to five years for pharmaceutical patents.
Extensions of up to five years on the standard 20 year term are available for pharmaceutical patents in the United States, the European Union and Japan in recognition of the exceptionally long development time and regulatory requirements involved in developing and commercialising a new drug. The aim is to provide an ‘effective patent life’, or period after marketing approval is obtained during which companies are earning a return on their investment, more in line with that available to inventions in other fields of technology.
The World Trade Organization (WTO) Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS) obliges countries to provide certain exclusive rights to all patents for a minimum of 20 years.

78. The amendments now reinserted a new Division 2 back into Part 3 of Chapter 6 (recalling that Division 2 had contained the former extension provisions but had been repealed by the Patents (World Trade Organization Amendment) Act 1994). This version now more closely resembled the current version. However, no change was made to the definition of ‘pharmaceutical substance’. The central provision was a new form of s 70 which now provided:
    70. Applications for extension of patent

(1) The patentee of a standard patent may apply to the Commissioner for an extension of the term of the patent if the requirements set out in subsections (2), (3) and (4) are satisfied.

(2) Either or both of the following conditions must be satisfied:

(a) one or more pharmaceutical substances per se must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification;

(b) one or more pharmaceutical substances when produced by a process that involves the use of recombinant DNA technology, must in substance be disclosed in the complete specification of the patent and in substance fall within the scope of the claim or claims of that specification.

(3) Both of the following conditions must be satisfied in relation to at least one of those pharmaceutical substances:

(a) goods containing, or consisting of, the substance must be included in the Australian Register of Therapeutic Goods;

(b) the period beginning on the date of the patent and ending on the first regulatory approval date for the substance must be at least 5 years.

Note: Section 65 sets out the date of a patent.

(4) The term of the patent must not have been previously extended under this Division.

(5) For the purposes of this section, the first regulatory approval date, in relation to a pharmaceutical substance, is:

(a) if no pre-TGA marketing approval was given in relation to the substance—the date of commencement of the first inclusion in the Australian Register of Therapeutic Goods of goods that contain, or consist of, the substance; or

(b) if pre-TGA marketing approval was given in relation to the substance—the date of the first approval.

(6) For the purposes of this section, pre-TGA marketing approval, in relation to a pharmaceutical substance, is an approval (however described) by a Minister, or a Secretary to a Department, to:

(a) market the substance, or a product containing the substance, in Australia; or

(b) import into Australia, for general marketing, the substance or a product containing the substance.

79. It will be observed that this contained for the first time a reference to a ‘pharmaceutical substance per se’. Paragraph 8 of the 1998 Explanatory Memorandum contains this statement (in a section headed ‘Notes on Clauses’):

A ‘pharmaceutical substance’ is defined in Schedule 1 of the [Act] and may comprise combinations of active ingredients or single active ingredients.

80. Paragraphs 9 and 10 of the 1998 Explanatory Memorandum also said this:
    9. The extension of term provisions will be available for patents that include claims to pharmaceutical substances per se (provided that the other criteria are met). These claims to pharmaceutical substances per se would usually be restricted to new and inventive substances. Patents that claim pharmaceutical substances when produced by a particular process (product by process claims) will not be eligible unless that process involves the use of recombinant DNA technology. Claims which limit the use of a known substance to a particular environment, for example claims to pharmaceutical substances when used in a new and inventive method of treatment, are not considered to be claims to pharmaceutical substances per se.
    10. An extension of term will not be available for claims to new processes of making pharmaceutical substances or new methods of using pharmaceutical substances where the substances themselves are already known.
81. It was this passage (and the specific reference to methods involving recombinant DNA technology in s 70(2)(b)) that led Heerey J and then the Full Court on appeal in Boehringer to conclude that the effect of the words ‘per se’ in s 70(2)(a) was to prevent an extension being granted in the case of a patent claiming a method. I return to these decisions later in these reasons.
82. The Act also introduced a new s 78:

78 Exclusive rights of patentee are limited if extension granted

(1) If the Commissioner grants an extension of the term of a standard patent, the exclusive rights of the patentee during the term of the extension are not infringed:

(a) by a person exploiting:

(i) pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(ii) pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification;

for a purpose other than therapeutic use; or

(b) by a person exploiting any form of the invention other than:

(i) a pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(ii) a pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification.

(2) If the Commissioner grants an extension of the term of a standard patent, the exclusive rights of the patentee after the grant of the extension are not infringed by a person exploiting:

(a) pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(b) pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification;

solely for purposes in connection with:

(c) having goods included in the Australian Register of Therapeutic Goods, where the goods are intended for therapeutic use; or

(d) obtaining similar regulatory approval under a law of a foreign country or of a part of a foreign country.

83. Section 78(2) is a springboarding provision. Its intent was to permit a generic pharmaceutical company, during any extended period of a patent for a pharmaceutical substance per se, to take steps to allow it to bring to market a generic version of the substance on the expiry of the patent. This is confirmed by [23] of the 1998 Explanatory Memorandum:

Subsection 78(2) sets out the so-called “spring-boarding” provisions whereby from the date of the extension being granted third parties will be able to undertake otherwise infringing activities solely for the purposes of meeting pre-marketing regulatory requirements for therapeutic use in humans. This has the effect of enabling a generic manufacturer to produce a generic pharmaceutical formulation containing the patented pharmaceutical substance solely for the purpose of obtaining regulatory approval while the patent is still in force. It therefore prevents a patentee from ending up with a further de facto extension of term which would occur if a generic producer could not commence any work on the patented pharmaceutical substance to meet these requirements until the extended term expired.

84. Cipla relied upon the passage extracted above at [79] and the second sentence of the passage quoted directly above from the 1998 Explanatory Memorandum to support a construction of the definition of pharmaceutical substance which does not include formulations. I am unpersuaded that the first passage shows that the author of the explanatory memorandum thought that a formulation was not a pharmaceutical substance. The use of the word ‘may’ makes such a conclusion too tenuous although I accept that it can be read the way Cipla suggests. I accept that the second passage does exhibit to some extent an assumption that a pharmaceutical substance might not include a formulation but I regard this as textually weak.
85. Even if these passages had more heft than I think that they do, however, I would not accept that the passage of the Intellectual Property Laws Amendment Act 1998 had any effect on the meaning of the definition of pharmaceutical substance. The Act did not purport to amend the definition. If the definition was amended it must be because the effect of the amendments introduced by the Act was impliedly to amend the definition. For that to have occurred, it would be necessary to identify something in the terms of the amendments which requires, as a matter of legal operation, that the definition was amended. The search for such a legal operation cannot be provided by the terms of an explanatory memorandum alone.
86. The terms of s 78(1) and (2) show only that the extension regime and accompanying springboarding provision were to apply to patents claiming pharmaceutical substances per se. There is nothing in the text of s 78(1) or any other provision inserted by this Act which is inconsistent with a pharmaceutical substance including a formulation. Indeed, s 78(1) takes as its point of departure a pharmaceutical substance which the explanatory memorandum shows was understood to be a defined term. The feature of s 78(1) which was novel was, for the first time, the insertion of a requirement that the patent had to claim not just a pharmaceutical substance, but a pharmaceutical substance per se. But it is beyond plain that whatever work was done by the words ‘per se’ was being done on the defined term, ‘pharmaceutical substance’.
87. I see no basis therefore upon which it can be concluded that the enactment of the Intellectual Property Laws Amendment Act 1998 had the effect of impliedly amending the definition. Once that conclusion is reached, the explanatory memorandum is irrelevant even if, contrary to my view, it provided any real support for the proposition that a pharmaceutical substance did not include a formulation.
88. It is not necessary to consider whether the passages relied upon by Cipla provide any support for the idea that a pharmaceutical substance per se may not include a formulation since Cipla expressly eschewed such a submission.

The Intellectual Property Laws Amendment Act 2006 (Cth)

89. There followed then the passage of the Intellectual Property Laws Amendment Act 2006 (Cth). It repealed the springboarding provision in s 78(2) and inserted a new springboarding provision, s 119A which it now relocated to the chapter of the Act dealing with infringement (Chapter 11 Division 1). It otherwise made no alteration to the extension regime. Whereas the former s 78(2) had only permitted springboarding during the extended term of a patent, the new s 119A now permitted springboarding at any time during the life of the patent (except in relation to the export of a substance for the purpose of obtaining foreign approval where the springboarding provision was limited by s 119A(2) to the period of any extension). Section 119A was (and is) in these terms:

119A Infringement exemptions: acts for obtaining regulatory approval of pharmaceuticals

(1) The rights of a patentee of a pharmaceutical patent are not infringed by a person exploiting an invention claimed in the patent if the exploitation is solely for:

(a) purposes connected with obtaining the inclusion in the Australian Register of Therapeutic Goods of goods that:

(i) are intended for therapeutic use; and

(ii) are not medical devices, or therapeutic goods, as defined in the Therapeutic Goods Act 1989; or

(b) purposes connected with obtaining similar regulatory approval under a law of a foreign country or of a part of a foreign country.

(2) Subsection (1) does not apply to the export from Australia of goods for purposes described in paragraph (1)(b) unless the term of the patent has been extended under Part 3 of Chapter 6 and the goods consist of or contain:

(a) a pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(b) a pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification.

Note: Part 3 of Chapter 6 provides for the extension of the term of standard patents claiming pharmaceutical substances.

(3) In this section:

pharmaceutical patent means a patent claiming:
(a) a pharmaceutical substance; or

(b) a method, use or product relating to a pharmaceutical substance, including any of the following:

(i) a method for producing a raw material needed to produce the substance;

(ii) a product that is a raw material needed to produce the substance;

(iii) a product that is a pro-drug, metabolite or derivative of the substance.

90. It will be seen that s 119A(3) defines the concept used in s 119A(1) of a ‘pharmaceutical patent’. There is a debate in this case about what s 119A(3) means. The explanatory memorandum accompanying the bill (‘the 2006 Explanatory Memorandum’) contains these relevant statements at p 19:

There are broadly four types of pharmaceutical patent: those on the active pharmaceutical ingredient (API); the formulation of the medication; the process for making the API; and methods of use of the medication. Only patents which claim a pharmaceutical substance (ie API) are currently eligible for patent extension in Australia. Pharmaceutical products are frequently the subject of multiple patents which cover different aspects of the product. These patents are potentially of different types, some of which may not be eligible for extension. In some cases the most important (or ‘blocking’) patent may not be extended and thus the most important springboarding work cannot be done until this patent expiries in Australia.
(emphasis added)

91. As Cipla correctly submitted, this passage in the 2006 Explanatory Memorandum exhibits an assumption that a pharmaceutical substance did not include a formulation.
92. At [158]-[159] of the 2006 Explanatory Memorandum this was said:
    158. The term ‘pharmaceutical substance’ is defined in Schedule 1 to the Patents Act.
    159. The definition of ‘pharmaceutical patent’ is intended to cover all patents that a generic pharmaceutical company would need to exploit in order to seek inclusion of a good other than a medical device or a therapeutic device on the ARTG. It is intended that patents claiming the following methods, products and uses relating to a pharmaceutical substance would be covered by the definition of ‘pharmaceutical patent’:

(a) a pharmaceutical substance per se; or

(b) a pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology; or

(c) a method of use of a pharmaceutical substance; or
(d) a method of administering a pharmaceutical substance; or
(e) a process for manufacturing a pharmaceutical substance; or

(f) a product or formulation incorporating a pharmaceutical substance or a mixture of pharmaceutical substances, including products such as layered or coated tablets; or

(g) other features of the pharmaceutical substance such as the colour or shape of a pill or packaging; or

(h) an apparatus or method specifically designed for manufacturing or testing a particular pharmaceutical substance.

(emphasis added)

93. Paragraph 158 shows that the author of the 2006 Explanatory Memorandum did not intend that there should be any alteration to the existing definition of ‘pharmaceutical substance’ in the Act. But that cannot be reconciled with the assumption, evident in [159(f)], that a formulation was not a pharmaceutical substance since, as defined in the Act, ‘pharmaceutical substance’ did include a formulation.
94. Paragraph 159 extends well beyond the actual text of s 119A(3). However, it is likely that each of (a)-(g) does fall within its terms. I am doubtful that (h) does so however there is no need to reach a conclusion about that.
95. The question which then arises is whether the insertion of s 119A into the Patents Act 1990 had the effect of amending the definition of ‘pharmaceutical substance’. Section 11B of the Acts Interpretation Act 1901 (Cth) provides:

11B Amending Act to be construed with Amended Act

(1) Every Act amending another Act must be construed with the other Act as part of the other Act.

(2) If:

(a) an Act (the amending Act) amends another Act (the principal Act); and

(b) a provision (the non-amending provision) of the amending Act does not amend the principal Act, but relates to an amendment of the principal Act made by another provision of the amending Act; and

(c) a term is used in the non-amending provision that has a particular meaning in the principal Act or in a provision of the principal Act amended or included by the amending Act;

then the term has that meaning in the non-amending provision.
Note: Subsection (2) covers, for example, application, transitional and saving items in a Schedule to an amending Act that relate to amendments of a principal Act made by the other times in the Schedule.

(3) Subsection (2) does not limit subsection (1).

96. The effect of this provision is, as Cipla correctly submitted, that the Intellectual Property Laws Amendment Act 2006 and the Patents Act 1990 must be read together as a combined statement of the will of the Parliament: Commissioner of Stamps v Telegraph Investment Co Pty Ltd [1995] HCA 44; (1995) 184 CLR 453 at 463 per Brennan CJ, Dawson and Toohey JJ.
97. The interpretative challenge arises from s 119A(3), which defines the concept of a ‘pharmaceutical patent’. The effect of s 119A(1) is that a pharmaceutical patent is not infringed by a generic company taking steps to have goods included on the Register of Therapeutic Substances. The definition of ‘pharmaceutical patent’ in s 119A(3) therefore operates to define the class of patents in respect of which springboarding protection is to be given to generics. Cipla’s point is that s 119A(3) observes a distinction between, on the one hand, a patent claiming a pharmaceutical substance referred to in s 119A(3)(a) and, on the other, a patent claiming a product relating to a pharmaceutical substance in s 119A(3)(b). Cipla submitted, and I agree, that that a patent which claims a product relating to a pharmaceutical substance in s 119A(3)(b) would encompass a patent claiming a formulation.
98. The question which arises is whether the word ‘or’ at the end of s 119A(3)(a) is disjunctive. If it is disjunctive, then two corollaries will follow. First, that a patent claiming a pharmaceutical substance cannot be a patent claiming any of the matters in s 119A(3)(b). Secondly, that the latter kind of patent must include a patent claiming a formulation (since a formulation is a product relating to a pharmaceutical substance). On the other hand, if ‘or’ is not disjunctive, then there is no reason why a patent claiming a formulation cannot also be a patent claiming a pharmaceutical substance in s 119A(3).
99. Thus, if the word ‘or’ is read disjunctively, it then will be necessary to read ‘pharmaceutical substance’ in s 119A(3) as not including a formulation. This may be achieved in two different ways. First, through s 11B it may be necessary to read the definition of ‘pharmaceutical substance’ in the Schedule to the Act as having been impliedly amended by the passage of the Intellectual Property Laws Amendment Act 2006. Secondly, it may instead only be necessary to read ‘pharmaceutical substance’ in s 119A(3) and not elsewhere in the Act as having this meaning because the definitions in the Schedule apply unless a contrary intention appears.
100. The interpretative choices are therefore these:

(a) To read ‘or’ as not being disjunctive so that no difficulty arises from the fact that a patent claiming a formulation falls within both s 119A(3)(a) and (b). In this case, a patent claiming a pharmaceutical substance will be a pharmaceutical substance via both s 119A(3)(a) and s 119A(3)(b). The extension provisions will continue to operate as they previously had and an extension will therefore be available for a patent claiming a formulation.

(b) To read ‘or’ as disjunctive and conclude that the definition of ‘pharmaceutical substance’ in the Schedule was impliedly amended to exclude anything appearing in s 119A(3)(b) including, relevantly, a formulation. This will mean that a patent claiming a formulation will still be a ‘pharmaceutical patent’ but only via s 119A(3)(b). It will also entail a substantive amendment to the operation of the extension provisions in that whilst formerly an extension could be obtained for a patent claiming a formulation, on the introduction of s 119A, this will no longer be so.

(c) To read ‘or’ as disjunctive and conclude that the definition of ‘pharmaceutical substance’ in the Schedule was not amended but that in s 119A(3) that definition does not apply with the consequence that in s 119A(3)(a) nothing which falls within the ambit of s 119A(3)(b) can ever be a pharmaceutical substance within the meaning of s 119A(3)(a). This will mean that a patent claiming a formulation will be a ‘pharmaceutical patent’ for the purposes of s 119A(3) but only via s 119A(3)(b). However, the extension provisions will continue to operate as they had formerly and an extension will be available in the case of a patent claiming a formulation.

101. Which of these choices is to be selected involves construing the two Acts together to give effect to the intention of Parliament as expressed in both. The relevant matters are:

(a) The definition of ‘pharmaceutical substance’ did include a formulation as at the date of the Intellectual Property Laws Amendment Act 2006.

(b) The Intellectual Property Laws Amendment Act 2006 does not amend the definition of ‘pharmaceutical substance’.

(c) The 2006 Explanatory Memorandum at [158] shows that the drafter of the memorandum intended that s 119A would operate by reference to the definition of pharmaceutical substance contained in the Schedule.

(d) The 2006 Explanatory Memorandum also shows at [159] that the drafter assumed that a patent claiming a formulation was not one to which the extension regime in s 70 applied.

(e) The assumption in (d) proceeded on an erroneous understanding of what the definition of ‘pharmaceutical substance’ in the Schedule meant.

102. In addition to these contextual matters, I consider it important to focus on what s 119A does. Section 119A is not concerned with the distinction between a pharmaceutical substance and a formulation of a pharmaceutical substance. It is concerned with giving springboarding protection to persons who would otherwise infringe a pharmaceutical patent as defined in s 119A(3). As constructional choices (b) and (c) above show, the operation of the provision is agnostic as to whether a pharmaceutical substance includes a formulation. If a pharmaceutical substance includes a formulation then springboarding protection will be available in the case of a patent claiming that formulation because it will be a pharmaceutical patent through both s 119A(3)(a) and s 119A(3)(b). But if a pharmaceutical substance does not include a formulation, then springboarding protection will still be available for a patent claiming a formulation because it will be a patent claiming a product relating to a pharmaceutical substance through s 119A(3)(b).
103. Considerations (a) to (d) and this textual consideration persuade me that the distinction between a pharmaceutical substance and a formulation is of no relevance to the operation of s 119A. That being the case, I do not think that Parliament can be taken to have intended to have amended the definition of ‘pharmaceutical substance’ when passing s 119A. That reading of s 119A comports with the contextual matters set out above at (a)-(c).
104. It does not comport with the assumption referred to in (d), that is to say, with the erroneous assumption made by the author of the 2006 Explanatory Memorandum that a pharmaceutical substance did not include a formulation. I do not think I should give such an erroneous assumption any weight. I have indicated above that I accept Cipla’s submission based on Baini v R that one should not rely on a paraphrasing of legislation in an explanatory memorandum but should instead focus on the text of the legislation itself. I am not sure that an assumption about what a provision means rises even as high as a paraphrasing of that provision, but assuming that it does, Baini v R requires me to prefer the text of s 119A as a surer guide to interpretation. This is particularly so where I am abundantly satisfied that the assumption is wrong in law.
105. It follows that constructional choice (b) must be rejected. This is enough to answer the present question contrary to Cipla’s interests for under either construction (a) or construction (c), a patent claiming a formulation is eligible to be extended and Cipla’s submission that it is not fails. Since the choice between construction (a) or construction (c) is not an issue in this proceeding, there is no formal need to determine which is correct. If it were necessary to express a view about it, I would prefer construction (a) since construction (c) appears pointless for the reasons I have given and would create a confusing situation where the meaning of pharmaceutical substance would depend on whether it appeared in s 119A. However, I have not heard argument on this issue and there may be answers to this if the issue ever arises.
106. In construing the Patents Act 1990 with the Intellectual Property Laws Amendment Act 2006, I therefore conclude that the definition of pharmaceutical substance in the Schedule was not amended and that a patent claiming a formulation continued to be eligible for an extension.

The Intellectual Property Laws Amendment Act 2015 (Cth)

107. The final legislative step was the passage of the Intellectual Property Laws Amendment Act 2015 (Cth) which inserted a new s 70(5). That provision has no relevance to the present debate.

Conclusions on legislative history

108. The legislative history shows that a pharmaceutical substance has always included a formulation. It is then necessary to turn to the policy of the extension provisions.

The Policy of the Extension Provisions

109. Cipla submitted, and I accept, that the general purpose and policy of a provision can be a guide to its meaning. Further, I accept that purpose and policy may be discerned from secondary materials such as explanatory memoranda. Cipla points to passages in the 1998 and 2006 Explanatory Memoranda to make good its submissions on purpose and policy.
110. As I have explained, I accept that these explanatory memoranda do contain some statements which are consistent with Cipla’s submission. However, I have explained above why I do not accept that either of the Acts to which those memoranda relate, namely the Intellectual Property Laws Amendment Act 1998 and the Intellectual Property Laws Amendment Act 2006, had any impact on the definition of ‘pharmaceutical substance’. Nor do I accept that anything said by the Industry Commission in its Report No 51 usefully bears upon the issue. What emerges clearly from the report is that there should be only one extension per drug but the Commission did not consider what that meant. It is true that its survey of the European position included a statement that there would be only one extension for each ‘chemical entity’ but the word the Commission went on immediately to use was ‘drug’. This material is far too ambiguous to be much use simply because the legislation mentioned above was the government’s response to Report No 51. Far more useful are the statements which actually appear in the relevant explanatory memoranda but, as I have explained, the statements in the 2006 Explanatory Memorandum plainly proceeded on a legally incorrect understanding of what a pharmaceutical substance was.

Ordinary Meaning of Pharmaceutical Substance

111. Notwithstanding the impression which the preceding 81 paragraphs may give, the task of statutory interpretation begins with the text. The text of the current definition of ‘pharmaceutical substance’ is set out above. Cipla made three relevant textual arguments (it conceded a fourth to be inconsequential).
112. First, Cipla submitted that the requirement that a pharmaceutical substance be ‘for therapeutic use’ confined the definition only to active ingredients (which necessarily have a therapeutic purpose). I do not see that this advances matters given there is a question, to be ventilated on Cipla’s alternative case, about whether the therapeutic use referred to in the definition must be of each constituent part of a pharmaceutical substance or may be of a mixture in globo.
113. Secondly, Cipla submitted that the inclusion of mixtures and compounds within the definition of pharmaceutical substance reflected an intention only to include combinations of active ingredients each of which were themselves pharmaceutical substances. This Cipla said arose from [8] of the 1998 Explanatory Memorandum, which stated that a pharmaceutical substance ‘may comprise combinations of active ingredients’. As I have already explained, however, I do not regard [8] as a firm basis to divine an intention to exclude formulations.
114. Thirdly, Cipla submitted that the word ‘application’ means the application of the pharmaceutical substance to the target in the human physiological system (in this case, the GLP-1 receptor). If this is correct, it entails that only the active pharmaceutical ingredient in a formulation may be a pharmaceutical substance. Why? Because a formulation is not ‘applied’ in this sense at the relevant target; only its active pharmaceutical ingredient is. For example, in this case, subject to a minor point, there is no debate that the excipients in the formulations disclosed by the 862 Patent do not ever bind on to the GLP-1 receptor. Thus, if the word ‘application’ entails application to the GLP-1 receptor, the formulation cannot be said to be so applied. On the other hand, Novo Nordisk submitted that ‘application’ means, in effect, ‘use’.
115. I accept that if Cipla is correct about the meaning of ‘application’ then this provides a reason to construe the definition of pharmaceutical substance as not applying to formulations.
116. The word ‘application’ has meanings which encompass both parties’ contentions. The Macquarie Dictionary’s definition of ‘application’ in definitions 1 and 5 is:
     1. the act of putting to a special use or purpose: the application of common sense to a problem.

...
5. the act of applying: the application of salve to a wound.

117. I do not think that the definition of pharmaceutical substances uses the word ‘application’ in the sense of definition 5 above. The application of a pharmaceutical substance in this sense would be satisfied by, for example, the application of an ointment to a wound (as the example tied to the dictionary definition makes clear). But I do not think that it would be correct usage to say of an antibiotic, for example, that it is applied in this sense to an infectious agent of the kind referred to in subclause (b) of the definition of pharmaceutical substance. So too, I do not think it a natural use of language to say that a blood thinner is ‘applied’ in this sense where the chemical or physico-chemical interactions mentioned in subclause (a) occur.
118. More generally, the use of the word ‘application’ in this sense begs the question of what the subject of the act of applying is. Where an ointment is applied to a wound, the answer is obvious: it is the wound. But the definition does not in terms identify anything to which the act of application is itself to be applied. It is, of course, into this convenient lacuna that Cipla injects the word ‘target’ and its accompanying and perhaps seductive suggestion that the pharmaceutical substance is to be applied at that target. But, when all is said and done, the word ‘target’ does not appear in the definition. And, whilst the absent word ‘target’ might well be consistent with the actions referred to in subclause (b) which are directed at unpleasant things such as infectious agents and toxins, it is difficult to make it work comfortably with the chemical interactions referred to in subclause (a).
119. It follows that I do not think that a pharmaceutical substance whose therapeutic use brings about a chemical interaction of the kind referred to in subclause (a) can be said to be ‘applied’ in Cipla’s sense of the word wherever that interaction occurs. This alone shows that Cipla’s deployment of the word ‘target’ is not apposite for the subclause (a) interactions unless, as Cipla appeared to submit, ‘target’ is broadened to mean the whole body. For example, ibuprofen reduces inflammation throughout the body and, no doubt, involves chemical interactions in bringing about that result. But I doubt that anyone would say a person who took a dose of ibuprofen had applied it throughout their body in the sense that an ointment is applied to a wound.
120. Thus, I think the word ‘target’ is not only absent from the definition of pharmaceutical substance but problematic in its relationship with the interactions in subclause (a). Rather than focussing on the word ‘target’ which the definition itself does not use, I think it of more utility to examine the word to which ‘application’ is undoubtedly attached in the definition: ‘involves’. The definition requires that the ‘application’ is to ‘involve’ the interactions in subclause (a) or the actions in subclause (b). I do not think that it would be an ordinary usage of the word ‘application’, in the sense that Cipla wishes to use it, to say that an application involves an interaction unless one had stated what the subject of the application was. Thus, whilst I might say that ‘The application of the ointment to the skin involves a chemical interaction’ I doubt that anyone would say in that sense ‘The application involves a chemical interaction’.
121. This leads to another problem. The definition clearly contemplates that there might be applications of the substance which do not involve the interactions in subclause (a) or the actions in subclause (b) (‘whose application (or one of whose applications)’). Why would the definition refer to applications in Cipla’s sense of the word which involve no interactions or actions? For example, in this case, why would anyone apply a peptide to their hair?
122. I would accept that none of this is completely decisive against Cipla’s construction. But viewed overall, reading ‘application’ in the manner suggested by Cipla lies somewhere between awkward and highly strained.
123. On the other hand, definition 1 of ‘application’ does not suffer from any of these problems. If ‘application’ means the act of putting to a special use or purpose then the word ‘involves’ is unproblematic. So construed, the lacuna to which Cipla’s submission gives rise by leaving unstated the subject of the application vanishes because ‘use’ assumes no subject. So too, there are no difficulties in saying that the use of the substance must involve the actions and interactions in subclauses (a) and (b). Finally, on this reading of ‘application’ as meaning ‘use’ the fact that the substance may have uses which do not involve the interactions and actions in the subclauses is unremarkable. Many pharmaceutical substances may have uses in animals, for example, which uses would not involve the relevant interactions or actions because there would be no ‘human physiological system’ or ‘human body’.
124. As a matter of ordinary language, I therefore conclude that ‘application’ in the definition of pharmaceutical substance refers to a use to which the substance is put. As it happens, this interpretation is consistent with the legislative history. I have explained above in relation to the  Patents Amendment Act 1989  why the explanatory memorandum accompanying that Act shows that ‘application’ there meant ‘use’.
125. On the basis that ‘application’ means ‘use’, it is then necessary to return to the question of whether a formulation can fall within the definition of a pharmaceutical substance. As a matter of ordinary language, I do not agree with Cipla’s construction. If the words in the first pair of parentheses were omitted from the definition then the word ‘substance’ would be construed to mean an element, a compound, or a mixture of elements or compounds. Whilst it is possible to imagine other substances, such as dark matter and ectoplasm, I do not think that the word ‘substance’ in a definition concerned with pharmaceutical substances would embrace these. It follows that if the words in parentheses were omitted from the definition then a formulation as a mixture of compounds would be a pharmaceutical substance so long as it satisfied the remaining requirements of the definition (namely that it be for therapeutic use and that its ‘application’ (scil. ‘use’) involve a chemical interaction with a human physiological system). Whether it is the mixture as a whole or each of its constituent parts which must be for a therapeutic use is the subject of Cipla’s alternative case discussed below.
126. The words in parentheses in the definition clarify that a substance includes a mixture or a compound of substances. Whilst I do not think this clarification is at all necessary, it is scarcely plausible that they were inserted to narrow the range of substances that could qualify as pharmaceutical substances. Yet this is the effect of Cipla’s submission. If narrowing in that fashion was intended, there were plenty of other, much clearer ways of indicating that such a limitation was intended. Another way of putting this is to observe that Cipla’s construction means that the word ‘including’ where it appears in the parenthetic phrase operates not to ‘include’ certain substances but, in fact, to exclude them. I am reluctant to interpret ‘include’ in this fashion.
127. I therefore do not accept Cipla’s submission about the ordinary meaning of the definition.

Whether Meaning Affected by Other Provisions of the Act

128. It would of course be quite wrong to stop the inquiry at this point. It is one thing to ascertain the ordinary meaning of an expression but, as Cipla correctly submitted, the definition must be afforded a meaning consistent with the language and purpose of the Act: Project Blue Sky Inc v Australian Broadcasting Corporation [1998] HCA 28; 194 CLR 355 at [69] per McHugh, Gummow, Kirby and Hayne JJ. Cipla identifies the following provisions as relevant to this task, on the basis that each uses the term ‘pharmaceutical substance’: ss 70(2)(a), 70(2)(b), 70(3)(a), 70(6), 71, 77, 78 and 119A. I have already explained why I do not accept Cipla’s submissions about s 119A and there is no need to repeat those reasons. Section 70 is set out above. Sections 71, 77 and 78 are as follows (noting that the version of s 78 set out earlier in these reasons has since been amended to the version extracted below):

71 Form and timing of an application
Form of application
(1) An application for an extension of the term of a standard patent must:

(a) be in the approved form; and

(b) be accompanied by such documents (if any) as are ascertained in accordance with the regulations; and

(c) be accompanied by such information (if any) as is ascertained in accordance with the regulations.

For this purpose, document includes a copy of a document.

Timing of application

(2) An application for an extension of the term of a standard patent must be made during the term of the patent and within 6 months after the latest of the following dates:

(a) the date the patent was granted;

(b) the date of commencement of the first inclusion in the Australian Register of Therapeutic Goods of goods that contain, or consist of, any of the pharmaceutical substances referred to in subsection 70(3), as worked out under subsection 70(5A) (if applicable);

(c) the date of commencement of this section.

...
77 Calculation of term of extension

(1) If the Commissioner grants an extension of the term of a standard patent, the term of the extension is equal to:

(a) the period beginning on the date of the patent and ending on the earliest first regulatory approval date (as defined by section 70) in relation to any of the pharmaceutical substances referred to in subsection 70(2);

reduced (but not below zero) by:
(b) 5 years.
Note: Section 65 sets out the date of a patent.

(2) However, the term of the extension cannot be longer than 5 years.
78 Exclusive rights of patentee are limited if extension granted

If the Commissioner grants an extension of the term of a standard patent, the exclusive rights of the patentee during the term of the extension are not infringed:
(a) by a person exploiting:

(i) a pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(ii) a pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification;

for a purpose other than therapeutic use; or

(b) by a person exploiting any form of the invention other than:

(i) a pharmaceutical substance per se that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification; or

(ii) a pharmaceutical substance when produced by a process that involves the use of recombinant DNA technology, that is in substance disclosed in the complete specification of the patent and in substance falls within the scope of the claim or claims of that specification.

129. Cipla made the following points about these provisions:

Section 70(2)(a)

130. Cipla submitted that the words ‘per se’ in s 70(2)(a) support its construction because they have been held ‘to exclude additional matter and combinations of elements or matter concerning pharmaceutical substances’. Respectfully, I do not agree. The effect of the words ‘per se’, as has been held by the Full Court in cases already noted above, is to exclude from the extension provisions patents which are for methods. It may be that in some cases it is possible to argue that a patent for a formulation is a patent for a method of delivery, although it is difficult to think of an example and I express no view. It is also true that a patent for a method of making a formulation could not be a patent for a ‘pharmaceutical substance per se’ within the meaning of s 70(2)(a). But that throws no light on whether a pharmaceutical substance can be a formulation.

Section 70(2)(b)

131. Cipla drew attention to the fact that the reference to pharmaceutical substance in s 70(2)(b) assumes that the substance has been produced by the use of recombinant DNA technology and submitted that a substance so produced could not include excipients. Section 70(2)(b) has the effect of permitting an extension to a method patent which results in a pharmaceutical substance so long as the method involves recombinant DNA technology. It stands in contrast to s 70(2)(a) which applies only to product claims. I do not agree with Cipla’s submission. Section 70(2)(b) only requires the process to involve the use of recombinant DNA technology. The addition of excipients to that substance as part of a method would still be a method ‘involving’ the use of that technology.

Section 70(3)(a)

132. Cipla’s submitted that there are therapeutic drugs which consist solely of an active ingredient without excipients. I accept this submission. Cipla then argued that s 70(3)(a) observes a distinction between goods on the Register of Therapeutic Goods which contain the pharmaceutical substance and those which consist of that substance. There would be no point in drawing this distinction, so the argument went, if a pharmaceutical substance could be a formulation which included excipients.
133. One needs to be precise about the source of this problem. It does not lie in the words ‘consisting of’ because there is nothing problematic about a good consisting of a formulation. The difficulty arises from the word ‘containing’. Here the thinking is that a formulation is unlikely to be contained in the good because the good will consist of the formulation. Put another way, as a formulation it will already have excipients in it, and it is difficult to envisage circumstances in which more excipients would be added to such a formulation to produce the good on the Register of Therapeutic Goods.
134. There are, I think, two answers to this. First, it is possible for goods on the Register of Therapeutic Goods to have components which are neither an active ingredient nor an excipient. A capsule containing a formulation is an example. Section 3 of the Therapeutic Goods Act 1989 (Cth) defines a therapeutic good to include, inter alia, goods which are for use as a container for goods which are therapeutic goods. Thus, there is no problem arising from the word ‘contain’ in s 70(3)(a) which would operate coherently on formulations contained in such a container.
135. Secondly, Cipla’s submission conceals an assumption that formulations may not themselves be further formulated. The correctness of this assumption is not self-evident. For example, in this case there are two formulations disclosed by the 862 Patent. One is a formulation of liraglutide with disodium phosphate dihydrate and propylene glycol whilst the other is a formulation consisting of that formulation with added phenol. This second formulation contains the first. Cipla’s assumption is therefore shown to be false on the facts of this case.
136. It follows that do not think that s 70(3)(a) has the effect that Cipla submitted.

Section 70(6)

137. Cipla fastened on the words ‘market the substance, or a product containing the substance’ in s 70(6) to suggest that a pharmaceutical substance cannot be a formulation. I do not agree. The issue is essentially the same as that which arises under s 70(3)(a). There are many ways that a product may contain a pharmaceutical substance beyond the mere addition of excipients. For example, a pharmaceutical substance may be contained in capsules.

Sections 71, 77, 78

138. In its written submissions, Cipla contended that ss 71, 77 and 78 also advanced its position but it did not go on to explain why. Each provision is set out at [128] above.
139. In relation to s 71, the argument may be that the reference in s 71(2)(b) to goods ‘that contain, or consist of, any of the pharmaceutical substances referred to in subsection 70(3)’ delineates between goods which contain and goods which consist of a pharmaceutical substance. However, this is the same point Cipla makes in relation to s 70(3) and I would reject it for the reasons I have already given.
140. I do not see that s 77 advances Cipla’s argument.
141. In relation to s 78, I have already explained why I do not accept that the language of ‘pharmaceutical substance per se’ assists Cipla. This provision does not add anything to the debate beyond that issue.

Case Law Concerning the Meaning of Pharmaceutical Substance

142. There are eight cases to consider (using the abbreviated names of those already mentioned in these reasons):
     • Boehringer (Heerey J);
     • Boehringer (FC) (Wilcox, Whitlam and Gyles JJ);
     • Prejay Holdings Ltd v Commissioner of Patents [2003] FCAFC 77; 57 IPR 424 (‘Prejay’) (Wilcox, Cooper and Allsop JJ);
     • Pharmacia (Weinberg J);
     • Alphapharm (FC) (Emmett, Bennett and Middleton JJ);
     • Spirit (Rares J)
     • Alphapharm (HC); and
     • Novartis AG v Pharmacor Pty Ltd (No 3) [2024] FCA 1307 (‘Novartis’) (Yates J).
143. Dealing with each in turn:

Boehringer

144. This case concerned a patent claim 1 of which claimed:

A container comprising an aerosol or spray composition for nasal administration which composition comprises as active ingredient a quaternary tropane alkaloid derivative with atropine-like activity, the container being provided with a nozzle adapted for nasal administration of the composition.

145. It will be noted that this was a patent claiming a container which had two elements. First, a composition with a particular active ingredient. Secondly, a particular nozzle suitable for nasal administration. Whether a container can genuinely be said to comprise the substance it contains is an interesting question not necessary to assay. An application for extension was made to the Commissioner of Patents and refused. Judicial review proceedings were then commenced in this Court. The issue for determination was whether the patent claimed a ‘pharmaceutical substance per se’ within the meaning of s 70(2)(a). Heerey J concluded that a patent claiming a ‘pharmaceutical substance per se’ was limited to product claims, that what was claimed was a method claim, and hence that this was not a claim for a pharmaceutical substance per se: [13]-[16], [19]. There was no debate in the case as to whether the composition with a particular active ingredient was a formulation and hence not a pharmaceutical substance. The case therefore does not assist.

Boehringer (FC)

146. This was the appeal from the judgment of Heerey J. The Full Court (Wilcox, Whitlam and Gyles JJ) affirmed Heerey J’s judgment: [37]. Special leave to appeal to the High Court was refused. As with the trial reasons, no consideration was given to whether a formulation could be a pharmaceutical substance. Like its predecessor, therefore, it does not add anything to the issue.

Prejay

147. Claim 14 of the patent in Prejay was in these terms:

A method of hormonally treating menopausal or post-menopausal disorders in a woman, comprising administering to said woman continuously and uninterruptedly both progestogen and estrogen in daily dosage units of progestogen equivalent to laevo-norgestrel dosages of from about 0.025 mg to 0.05 mg and of estrogen equivalent to estradiol dosages of about 0.5 mg to 1.0 mg.

148. It was not disputed that progestogen and estrogen were both pharmaceutical substances nor that claim 14 was a claim for a method. A direct application of the Boehringer decisions would entail that claim 14 was not a claim for a pharmaceutical substance per se. The appellants made a formal submission that Boehringer was wrongly decided but primarily sought to distinguish the facts from Boehringer on the basis that the patent in Boehringer involved a physical apparatus. The Court declined to depart from Boehringer and, whilst accepting that a physical apparatus was lacking in claim 14, did not think that this was a relevant distinction. It concluded that it is not enough to constitute a claim for a pharmaceutical substance that a pharmaceutical substance ‘appears in a claim in combination with other integers or as part of the description of a method (or process) that is the subject of claim’: Prejay at [24] per Wilcox and Cooper JJ, [35]-[36] per Allsop J. As with Boehringer, I do not read the case to say anything about whether the definition of ‘pharmaceutical substance’ includes formulations.

Pharmacia

149. Pharmacia involved two proceedings, both related to a grant by the Commissioner of Patents of an extension to a patent. The patent was entitled ‘Injectable ready-to-use solutions containing an antitumor anthracycline glycoside’. Claim 1 was:

A sterile, pyrogen-free, anthracycline glycoside solution which comprises a physiologically acceptable salt of an anthracycline glycoside dissolved in a physiologically acceptable aqueous solvent therefor at an anthracycline glycoside concentration of from 0.1 to 50 mg/ml, which has not been reconstituted from a lyophilizate and the pH of which has been adjusted from 2.5 to 5.0 solely with a physiologically acceptable acid.

150. It will be apparent that claim 1 was, at least, a claim for a formulation of an acceptable salt of an anthracycline glycoside made in a particular way. The party opposing the patent extension in the first proceeding, Mayne, made the submission (recorded by Weinberg J at [37]-[38]) that claim 1 was a ‘product by process’ claim which was incapable of being a claim for a pharmaceutical substance per se because of the Full Court’s reasons in Boehringer. The second respondent, Interpharma, echoed Mayne’s submission but contended in addition that Pharmacia’s product was not a pharmaceutical substance: [62], [71]. It called an expert witness, Dr Elliott, who opined that ‘pharmaceutical substance’ within the meaning of the Act was limited to an active ingredient on its own so that claim 1 was not a claim to a pharmaceutical substance: [68]. Thus, both the question of whether the substance claimed in claim 1 was a pharmaceutical substance and the additional question of whether it was a pharmaceutical substance per se arose.
151. On the question of whether the substance claimed in claim 1 was a pharmaceutical substance, Weinberg J concluded that neither Heerey J nor the Full Court in Boehringer had distinguished between the active ingredients in a compound and the other ingredients in that compound: [106]. At [107], Weinberg J thought that Heerey J and the Full Court had both approved an analysis of the concept of what constituted a pharmaceutical substance in the Commissioner’s Patents Manual: [107]. The manual stated that: ‘In the case of such mixtures or compounds, the test of whether or not a substance is a pharmaceutical substance applies to the mixture or compound as a whole, not to an individual component of the mixture or compound’: [90]. In light of these matters, his Honour concluded that the claim 1 stated a claim to a pharmaceutical substance: [108].
152. On the question of whether claim 1 was for a pharmaceutical substance per se, Weinberg J concluded that when read fairly claim 1 was not a claim to a process but a claim to a new and inventive substance: [99]-[100].
153. About Pharmacia, Cipla made two submissions. First, it was distinguishable because at the time it was decided, s 119A had not been introduced into the Act. For the reasons I have already given, s 119A had no impact on the definition of pharmaceutical substance. It provides no reason to distinguish Pharmacia. Secondly, if it was not distinguishable, Cipla submitted that it was plainly wrong. There are aspects of the reasoning applied by Weinberg J which may be susceptible to criticism. It was never suggested in the Boehringer litigation that the claimed substance was not a pharmaceutical substance and I do not think his Honour was correct to derive support for his view that a formulation could be a pharmaceutical substance from those decisions. With respect, I also do not think that the Patents Manual is a legitimate interpretative source by which to interpret the Act. However, despite those matters it will be apparent from the legislative history traced earlier in these reasons that the conclusion reached by Weinberg J was correct. I therefore see no basis on which I could say it was plainly wrong.

Alphapharm (FC)

154. Cipla submitted that I am bound by Alphapharm (FC) to conclude that a pharmaceutical substance cannot include a formulation. I do not accept this submission. The case was concerned with a patent for an antidepressant drug called citalopram. Citalopram is what is known as a ‘racemic mixture’ or racemate. Racemates arise in the context of chiral molecules. A chiral molecule is a molecule that can exist in two isomeric forms which are the mirror image of the other but are not superimposable on each other. The two non-superimposable mirror images are referred to as enantiomers. A racemate is a mixture comprising two such enantiomers in equal measure. The two enantiomers are designated either (+) or (-) according to their effect on the rotation of plane-polarised light. Being a racemate, citalopram is therefore a mixture in equal parts of the (+) and (-) enantiomers.
155. There were two patents involved in the Full Court litigation. One was a patent for the racemate, that is, a mixture in equal parts of the (+) enantiomer and the (-) enantiomer. It was not in dispute that the active ingredient in the racemate was the (+) enantiomer. The other patent was later in time and was for the (+) enantiomer alone. It reflected the fact that a superior way of isolating the (+) enantiomer had by then been found.
156. By way of context, the (+) enantiomer is more effective for treating depression than the racemate and much more effective than the (-) enantiomer. Whilst there were many issues in the Full Court litigation, only one was concerned with the extension of a patent. Lundbeck had obtained an extension to the term of its patent for the racemate and the Full Court was asked to determine whether this extension was valid (bearing in mind that its active ingredient was the (+) enantiomer). The extension issue in the Full Court did not therefore concern the (+) enantiomer patent. However, that patent is relevant to the High Court’s decision in Alphapharm next considered and should be kept in mind.
157. By s 71(2)(b), an extension of a patent for a pharmaceutical substance cannot be obtained, relevantly, unless the application is made within 6 months of the date of the inclusion on the Register of Therapeutic Goods of goods that ‘contain, or consist of’ that pharmaceutical substance.
158. Lundbeck had secured registration on the Register of Therapeutic Goods of two products. One contained the racemate (‘Cipramil’) and the other the more efficacious (+) enantiomer (‘Lexapro’).
159. The problem which Lundbeck encountered was that its racemate product, Cipramil, necessarily included the (+) enantiomer precisely because it was a racemate. The Cipramil product had been entered on the Register of Therapeutic Goods on 9 December 1997. The Full Court’s reasons include a variety of dates for this entry. Emmett J said at [106] that it was 29 December 1997 but at [108] that it was 9 December 1997. Bennett J said at [247] that it was 9 December 2012, which is I think a typo. Recourse to the primary judge’s reasons in Alphapharm Pty Ltd v H Lundbeck A/S [2008] FCA 559 at [476] shows that the correct date is 9 December 1997. It is also apparent from the primary judge’s reasons that the application for the extension was filed on 22 December 2003: [29]. This was well within 6 months of the date that the (+) enantiomer product, Lexapro, had been entered on the Register of Therapeutic Goods on 16 September 2003: [29]. But it was several years out of time if the relevant entry was that of the racemate product, Cipramil, on 9 December 1997.
160. Lundbeck sought to persuade the Full Court that the relevant registration for timing purposes was that of Lexapro (in September 2003) and not Cipramil (in December 1997). Largely these arguments turned on seeking to demonstrate that the active ingredient in the racemate was really the (+) enantiomer and that the correct product was therefore Lexapro and not Cipramil. The argument is succinctly summarised at [230]. As Bennett J recorded at [231] the parties agreed that the relevant pharmaceutical substance was the (+) enantiomer.
161. However, the statutory question in s 71(2)(b) was whether the racemate product (containing in equal measure the (+) and (-) enantiomers) contained the (+) enantiomer. This question tends to answer itself when framed this way, as the both the trial judge and the Full Court concluded. The application for the extension of the patent was therefore out of time and could not be granted.
162. There was no issue before the Full Court as to whether a formulation could be a pharmaceutical substance. The only question was whether Cipramil ‘contained’ the (+) enantiomer. Indeed, it is unclear from the report whether Cipramil was in fact formulated. There was evidence before me that there were some therapeutic goods which consisted entirely of an active ingredient. I will therefore not speculate on whether Cipramil was, in fact, formulated.
163. Cipla relied upon the reasons for judgment of Bennett J (with whom Middleton J agreed). Her Honour’s ultimate conclusion was that the racemate product Cipramil did contain the (+) enantiomer so that the extension could not be granted. Cipla relied upon several statements made by her Honour in the course of reaching that conclusion. The first is [231]:

There is no dispute that the pharmaceutical substance for the purposes of s 70(2)(a) is (+)-citalopram. However, the remaining approach suggested by Lundbeck finds no support in the language of s 70, particularly once s 70(3) is considered in context.

164. I accept that this provides some assistance to Cipla but I do not think that her Honour was specifically adverting her mind to the question of whether a formulation of the (+) enantiomer would be a pharmaceutical substance. However, Cipla also relied upon what her Honour said at [244]. Before setting that out it is necessary to understand a little more about the broader dispute in Alphapharm (FC). In the following passage her Honour refers to ‘the Patent’. That patent was not the racemate patent but rather the (+) enantiomer patent. At [244] her Honour observed:

I have concluded, as a matter of construction, that the subject matter of claim 1 of the Patent is the separated or purified or isolated (+)-enantiomer, (+)-citalopram. I have also concluded that the claim is not anticipated by the racemate. Lundbeck submits that such a conclusion is inconsistent with a finding that the racemate “contains” the (+)-enantiomer. However, the pharmaceutical substance per se is the molecule, the (+)-enantiomer. The racemic mixture is a solution that contains both (+) and (–) enantiomers in equal proportions. That (+)-enantiomer molecule per se fulfils the requirements of s 70(2). It is the molecule that “works” as a pharmaceutical substance alone, or together with other substances, in goods listed on the ARTG. These other substances may be components of a formulation or may otherwise be described as impurities, such as the (–)-enantiomer.

165. I would accept that this is consistent with a pharmaceutical substance per se being the active ingredient. However, it does not appear to throw light on whether a pharmaceutical substance can be a mixture of active and non-active ingredients. Indeed, on one view, the passage may be consistent with the proposition that a pharmaceutical substance may include a formulation given her Honour’s focus on the words ‘per se’, although this is a weak inference.
166. From these observations I would draw six conclusions. First, it is doubtful that her Honour was intending to express a view on whether a formulation could be a pharmaceutical substance since there was no issue about this in the case. Secondly, [231] does provide some support, limited perhaps, for Cipla but correspondingly the emphasis in [244] on ‘per se’ cuts to some extent in the opposite direction. Thirdly, the fact that the two passages may possibly point in opposite directions serves only to underscore that her Honour was likely not intending to address this issue. Fourthly, I do not therefore extract anything on the present issue from these passages. Fifthly, in any event, to the extent that something is being said about formulations in these passages the statements are obiter dicta. Sixthly, the fact that the statements may be contradictory underscores that the statements are not considered obiter dicta.
167. I therefore reject Cipla’s submission that I am bound by the Full Court’s judgment to conclude that a formulation cannot be a pharmaceutical substance. Indeed, I do not think that read in context it was intended to say, or indeed does say, anything about the present issue.

Spirit

168. Like Pharmacia, Spirit related to events before s 119A had come into force. However, for the reasons I have given, I do not accept Cipla’s submission that this provides any basis for distinguishing the case. Spirit involved a patent claiming a slow release formulation of the pain relief drug oxycodone. Claim 5 was in these terms:

A solid controlled release oral dosage form [of oxycodone], comprising:
(a) oxycodone or a salt thereof in an amount from 10 to 160 mg;

(b) an effective amount of a controlled release matrix selected from the group consisting of hydrophilic polymers, hydrophobic polymers, digestible substituted or unsubstituted hydrocarbons having from 8 to 50 carbon atoms, polyalkylene glycols, and mixtures of any of the foregoing; and

(c) a suitable amount of a suitable pharmaceutical diluent, wherein said composition provides a mean maximum plasma concentration of oxycodone from 6 to 240 ng/ml from a mean of 10 to 14 hours after repeated administration every 12 hours through steady-state conditions.

169. The issue in the case seems to have been whether the slow release formulation constituted a pharmaceutical substance per se and not whether it was a pharmaceutical substance. Nevertheless, Rares J expressed an opinion at [49] which, without setting out, appears to accept that a formulation is a pharmaceutical substance. His Honour did this as part of his analysis of the expression pharmaceutical substance per se. As such, and whilst the issue was not actually argued, it does appear to form part of the ratio decidendi. Interestingly, Rares J noted a submission that a claim for a pharmaceutical substance per se could only be a claim for the active ingredient and rejected it: [53]. Of course, Cipla does not advance such a submission in this case.
170. Since I agree that a formulation is a pharmaceutical substance, there is no need to assess Cipla’s submission that Spirit is plainly wrong.

Alphapharm (HC)

171. This was not an appeal from the the Full Court’s judgment in Alphapharm discussed above but is closely related to that litigation. After the Full Court had held that the extension of the racemate patent was invalid, Lundbeck then applied for an extension of the (+) enantiomer patent. This was based on the entry on the Register of Therapeutic Goods of the racemate product, Cipramil, on 9 December 1997. Lundbeck applied for the extension on 12 June 2009 which was, of course, far outside the six month period specified in s 71(2)(b). However, Lundbeck now applied for an extension of that time limit under a general provision providing for extensions of time towards the back end of the statute, s 223(2). A delegate of the Commissioner concluded that there was power to extend the time for Lundbeck to bring its extension application and granted this procedural extension of time (but did not deal with the substantive application for an extension of the term of the patent). This conclusion was affirmed by the Administrative Appeals Tribunal and then by the Full Court of this Court. The question before the High Court was whether this was correct. The High Court held that it was.
172. It will be seen that no part of the controversy before the High Court concerned the question of whether a formulation could be a pharmaceutical substance. Recourse to the summary of argument in the report of the decision likewise confirms that this was not an issue. What was in dispute before the High Court was the proper construction of s 223 and a regulation made under the Act.
173. The majority comprised Crennan, Bell and Gageler JJ (Kiefel and Keane JJ dissented). Having laid out the statutory provisions and introduction, the majority provided some context for the (+) enantiomer patent. This was under the heading ‘The background facts’ and appears at [23]-[24] in their Honours’ reasons. Paragraph 23 is in these terms:

A little more needs to be said about the Escitalopram Patent. Lundbeck, a Danish pharmaceutical company, applied for the Escitalopram Patent on 13 June 1989 (the expiry date of which became 13 June 2009), for an invention entitled “(+)-Enantiomer of citalopram and process for the preparation thereof”. There are six claims – claims 1 to 5 are product claims and claim 6 is a method claim, which, for present purposes, can be put to one side. Claim 1 claims a compound (an enantiomer) known as “(+)-citalopram” and its non-toxic acid addition salts, and claims 3 and 5 claim a pharmaceutical composition comprising, as an active ingredient, that compound. The pharmaceutical substance disclosed in the complete specification, (+)-citalopram, is used to treat depression.
(Footnotes omitted.)

174. The underlined sentence is accompanied by a footnote 40 which is in these terms:

Relevantly, the extension of term scheme under the Act covers standard patents for pharmaceutical substances per se pursuant to s 70(2)(a), hence patents for pharmaceutical methods or tablets do not fall within the scheme. It can be noted that pharmaceutical substances produced by a process that involves the use of recombinant DNA technology, the subject matter of s 70(2)(b), are not relevant to this case.

175. The statement that patents for pharmaceutical methods cannot be a patent for a pharmaceutical substance per se accords with the manner in which the words ‘per se’ has been interpreted in this Court in the authorities noted at [5] above. However, Cipla fastens on the statement that tablets do not fall within the extension scheme to advance the proposition that a formulation is not a pharmaceutical substance.
176. Cipla misreads footnote 40. The statement is that tablets cannot be a pharmaceutical substance per se. Since this appears immediately after an explanation of the uncontroversial proposition that a patent for a pharmaceutical method cannot be a pharmaceutical substance per se and since there is no reference to tablets in s 70(2) at all, I read this footnote as exhibiting a conclusion that a patent for a tablet is a patent for a pharmaceutical method of delivery. If not read that way, the reference to ‘tablet’ seems to come from nowhere although an alternative view may be that it derives from the passage in explanatory memorandum for the Intellectual Property Laws Amendment Act 2006 set out above which also refers to tablets. But wherever the reference to a tablet comes from, it is clear that the conclusion in footnote 40 is about the proposition that a method patent will not disclose a pharmaceutical substance per se within the meaning of s 70(2).
177. That may raise an interesting question in this case as to why Cipla did not submit that the 862 Patent was not a patent for a pharmaceutical substance per se. It is not necessary for me to speculate about this, however, and instead it suffices to observe again that Cipla’s submission is confined to the contention that the formulation disclosed by the 862 Patent is not a pharmaceutical substance. As I have noted above, it accepts that if it is a pharmaceutical substance, it is also a pharmaceutical substance per se for the purposes of s 70(2).
178. Returning then to the majority’s statement in footnote 40, my conclusion is that is says nothing about whether a formulation is a pharmaceutical substance. It is not necessary to consider whether footnote 40 is a seriously considered dicta of the High Court which I am bound to follow (see Hill v Zuda Pty Ltd [2022] HCA 21; 275 CLR 24 at [26] per the Court) or just an obiter dictum made in passing. Given its presence in a footnote in a section headed ‘The background facts’ in a case having nothing to do with this issue, I would tend to favour the latter characterisation. In any event, it does not matter because footnote 40 does not say that a formulation cannot be a pharmaceutical substance.

Novartis

179. The parties put on supplementary submissions about Novartis which was handed down following the conclusion of this hearing. Cipla contended that the decision supports the proposition that a mixture can only be a pharmaceutical substance if the components of the mixture are all pharmaceutical substances. In particular, at [286] Yates J stated that ‘[t]he definition makes clear that the “pharmaceutical substance” can be a mixture or compound of such substances’.
180. As Novo Nordisk submitted, however, whether the definition of ‘pharmaceutical substance’ extends to formulations was not in issue before his Honour. In addition, I would not read the statement that a pharmaceutical substance ‘can’ be a mixture of other pharmaceutical substances to exclude the possibility of formulations falling within the definition of pharmaceutical substance. I therefore do not take Novartis to be inconsistent with the views I have already expressed.

Conclusions on case law

181. The proposition that a formulation is not a pharmaceutical substance does not form part of the ratio decidendi of Alphapharm (FC) or Alphapharm (HC). Both are obiter dicta but I do not consider that either is considered. Obviously, anything said by the High Court, even if a footnote, must be taken into account. I do not consider that either of the Boehringer decisions assists since there was no issue about whether a pharmaceutical substance included a formulation in either. The same may be said of Prejay. Both Pharmacia and Spirit have as part of their ratio decidendi that a pharmaceutical substance may include a formulation. The reasoning in Pharmacia is susceptible to criticism and the point appears not to have been argued in Spirit. If it appeared reasonably clearly that a pharmaceutical substance did not include a formulation I would be prepared to depart from these decisions.
182. Viewing the course of authority as a whole, I am not persuaded that it would require me, or if not require then persuade me, to disregard my own conclusions following an exhaustive examination of the legislative history and the current provisions.

Conclusions on Interpretation of ‘Pharmaceutical Substance’

183. I reject Cipla’s construction of ‘pharmaceutical substance’. It is contrary to its ordinary meaning, is not required by other provisions in the Act, is contrary to the legislative history and is not required by the case law. Further, the suggested legislative policy has not been shown sufficiently to exist.
184. That said, I do accept that the reference to a tablet in footnote 40 in the High Court’s decision in Alphapharm, if not read as an illustration of a method, is inconsistent with the conclusion I have reached. My preferred reading of it is as a statement about method claims which would be consistent with the case law about what a pharmaceutical substance per se is. However, I acknowledge that I may be wrong about that and that if I am, then footnote 40 contradicts the conclusion I would otherwise reach.
185. It is not at all clear where the High Court obtained the reference to a tablet but it seems to me that the most likely explanation is that it comes from the explanatory memorandum which accompanied the 2006 Act. However, as I have explained, the statement in that explanatory memorandum is legally incorrect. Although the High Court in Baini v R counselled against utilising paraphrases of legislation which appear not only in explanatory memoranda but also in judgments of courts, I do not apprehend the Court to have been suggesting the courts below it should depart from its paraphrasing of legislation. I do not therefore regard Baini v R as a licence to ignore footnote 40.
186. I hesitate to depart from footnote 40, but it is not a considered obiter dictum and, as such, I am not legally required to follow it. Giving it the significant weight that it must nevertheless necessarily be accorded, I have come to the view that if it is not merely an illustration of a method claim, then I should respectfully not follow it. I do so hesitantly.

THE ALTERNATIVE CASE

187. It follows from the above that the two relevant formulations of liraglutide claimed in the 862 Patent are not precluded from being ‘pharmaceutical substances’ by virtue of their being formulations. It becomes necessary in light of this conclusion to consider Cipla’s alternative case that, even if ‘pharmaceutical substance’ is capable of encompassing formulations, it does not encompass either of the relevant formulations claimed by the 862 Patent.
188. The pith of Cipla’s alternative case is that here, unlike in Spirit, the excipients do not individually or together have a therapeutic use in the formulations separate from the liraglutide itself. As already noted, there are two limbs to this point: first, somewhat obscurely, whether it is necessary that the excipients in a formulation themselves be for a therapeutic use in order that the formulation be a ‘pharmaceutical substance’; and secondly, if so, whether the excipients in the formulations claimed by the 862 Patent are for a therapeutic use.

Limb 1: Must Excipients be for a Therapeutic Use?

189. The reader may be forgiven for struggling to understand this submission. As I understood it, the submission emerged as an illustration derived from the judgment of Rares J in Spirit and went as follows: in Spirit, the active ingredient was oxycodone but it was formulated as a controlled release tablet. This outcome was achieved by including the oxycodone in a substance (a deliberately neutral word) which resisted the process of digestion in the gastro-intestinal tract and slowed the release of the oxycodone. Rares J held that the formulation was a pharmaceutical substance. Cipla’s point, which was very briefly, perhaps vanishingly, expressed, seemed to be that if, contrary to its primary submission, a formulation including excipients could be a pharmaceutical substance then, as a matter of law, such a situation was confined to the circumstances obtaining in Spirit, that is to say, the situation where the excipient was for therapeutic use.
190. Unpicked in this way, this submission is revealed as a repetition of Cipla’s primary argument that a formulation can never be a pharmaceutical substance. The effect of Rares J’s conclusion that the controlled release excipient was for a therapeutic use inevitably entails that the controlled release excipient satisfied the definition of a pharmaceutical substance. Thus, whilst Cipla glancingly posits this as an alternative legal case to its broad contention that every element in a formulation must be for a therapeutic use, the facts of Spirit show that it was a case where every element of the formulation did have that quality.
191. Thus, when the fog lifts on this legal contention, it turns out to be the same as Cipla’s primary submission and is to be rejected for the same reasons. If the submission means anything else, then I do not understand it and reject it for that reason. Having cleared this out of the way, the remaining issues are the factual ones raised by Novo Nordisk on the assumption, contrary to my conclusion, that each excipient must be for therapeutic use in its own right.

Limb 2: Are the Excipients for a Therapeutic Use?

192. Under this heading, too, is concealed a legal question. Granted now that each excipient must be for a therapeutic purpose, how is that purpose to be ascertained and on the basis of what material? In this section, I approach the matter as follows: first, I examine the 862 Patent to see the purposes it discloses the three excipients were to serve. Secondly, I consider what other, if any, matters are relevant to the discernment of a purpose for therapeutic use. Thirdly, I assess the expert evidence about the uses of the excipients. Finally, I draw some conclusions.
193. As has already been noted, the two relevant formulations claimed by the 862 Patent are:

(a) a formulation of liraglutide with a ‘tonicity agent’ of propylene glycol and a ‘buffer’ of disodium phosphate dihydrate; and

(b) the same formulation to which is added a preservative (such as phenol).

The Tonicity Agent: Propylene Glycol

194. It is clear from the specification that the purpose of the propylene glycol is to ensure that the formulation is ‘isotonic’. This refers to a state where the concentration of osmotically active solutes in the formulation is equal to the concentration of osmotically active solutes in the interstitial fluid. If a formulation is not isotonic, then osmosis will occur which will result in the movement of water into or out of the cells in the region of administration. Where this occurs a phenomenon known as osmotic shock may develop which may result in pain, swelling and cell damage.

The Buffer: Disodium Phosphate Dihydrate

195. Both the terms of the specification and the claims themselves demonstrate that the purpose of the disodium phosphate dihydrate is to act as a buffer. The purpose of a buffer is to maintain a given pH level. Claims 5 to 7 set out varying pH levels for the formulations and it is clear from the body of the specification (p 1a lines 15-17) that the buffer is what achieves this in the formulation. In his first affidavit at §§66-67, Associate Professor Kayser explained that the selection of an appropriate pH level for a peptide formulation involves two concepts. First, because physiological pH is about 7.4 it is desirable for injectable solutions to have a pH range between 5 and 8. Outside this range pain may occur in the patient because the formulation is too acidic or too basic. Secondly, a peptide has a point known as the isoelectric point which is the pH at which the net charge on the surface of the peptide is 0. At this point, the peptide will precipitate out of solution. Thus, a peptide with an isoelectric point of 8 might be formulated with a buffer the pH of which is 6 or 7.

The Preservative: Phenol

196. Claims 8-9 refer to a preservative as part of the formulation at varying concentrations. The claims do not make a claim for any particular preservative, but the specification frequently refers to the preservative being phenol. In his second affidavit at §8 Associate Professor Kayser explained that a preservative prevents microbial growth during storage.

The Statutory Question

197. The definitions of ‘pharmaceutical substance’ and ‘therapeutic use’ are set out above but bear repetition at this point:

pharmaceutical substance means a substance (including a mixture or compound of substances) for therapeutic use whose application (or one of whose applications) involves:

(a) a chemical interaction, or physico‑chemical interaction, with a human physiological system; or

(b) action on an infectious agent, or on a toxin or other poison, in a human body;

but does not include a substance that is solely for use in in vitro diagnosis or in vitro testing.
therapeutic use means use for the purpose of:

(a) preventing, diagnosing, curing or alleviating a disease, ailment, defect or injury in persons; or

(b) influencing, inhibiting or modifying a physiological process in persons; or
(c) testing the susceptibility of persons to a disease or ailment.

198. In my view, whether a substance is for therapeutic use is to be determined by reference to the terms of the patent in which the substance appears. The word ‘for’ in the definition of ‘pharmaceutical substance’ and the explicit use of the word ‘purpose’ in the definition of a ‘therapeutic use’ show that the inquiry posited by both definitions focuses on the purpose for which the substance is used. The inquiry posited by the two definitions is objective and does not turn on the subjective state of mind of the patentee or, for that matter, anyone else.
199. The relevance of the two definitions is that they provide a gateway into the extension provisions in Part 3 and the springboarding provision in Part 11, s 119A. The former turns on a patent claiming a pharmaceutical substance per se, the latter on the presence of a pharmaceutical patent, i.e., a patent claiming a pharmaceutical substance; or, a method, use or product relating to a pharmaceutical substance.
200. What both have in common is a pharmaceutical substance. The concept of a pharmaceutical substance lies therefore at the core of Part 3 and s 119A. It must be known at the time that a patent is granted whether these provisions apply to the patent. The operation of the extension and springboarding provisions cannot be contingent on what may, at some later date, be shown about the therapeutic effects of a substance claimed in it. Were it otherwise, a patent claiming a substance for which it made no claim for therapeutic use could afterwards become a patent claiming a pharmaceutical substance on the adduction of evidence that the substance had a therapeutic use, even though no such claim was made in the patent. This would lead to a chaotic operation of the extension regime which can scarcely have been contemplated.
201. Thus, not only is the purpose inquiry posited by the two definitions objective in nature but the circumstances from which that objectively determined purpose are to be ascertained are limited to the terms of the patent construed in the orthodox way.
202. On Cipla’s rectification application, Novo Nordisk led evidence from Professor Bernkop-Schnürch about additional therapeutic effects that propylene glycol, disodium phosphate dihydrate and phenol had. For the reasons I have just given, I do not consider that evidence relevant to the inquiry at hand.
203. It is then necessary to consider what purpose is disclosed by the terms of the 862 Patent for the three excipients.

Therapeutic Use of Disodium Phosphate Dihydrate?

204. In terms of the definition, the buffer does not prevent, diagnose, cure or alleviate a disease, ailment or injury in persons; hence, it does not fall within subclause (a) of the definition of ‘therapeutic use’. It plainly does not fall within (c) either. The only substantive question is whether it can be described, in terms of subclause (b), as ‘influencing, inhibiting or modifying a physiological process in persons’.
205. The first use of the buffer mentioned by Associate Professor Kayser – the avoidance of precipitation of the peptide out of solution at the isoelectric point – does not fall within this definition. The question then becomes whether the second use of the buffer mentioned by Associate Professor Kayser – the avoidance of pain resulting from having a pH outside the range 5 to 8 – falls within (b). I am prepared to infer that pain is caused by a physiological process. However, I do not think that the buffer influences, inhibits or modifies whatever that process is. Rather, the significance of this second use of the buffer lies not in what it is but, rather, in what it is not. A formulation with a pH outside the range 5 to 8 causes pain. The buffer, by keeping the solution in the range 5 to 8, avoids that range. The buffer does not act upon the physiological process by which the pain is generated but rather simply avoids the circumstances which would provoke the pain response in the first place.

Therapeutic Use of Propylene Glycol?

206. As already noted, the patent contemplates the use of propylene glycol to maintain isotonicity. Where isotonicity is not maintained osmosis will occur which will result in the movement of water into or out of the cells in the region of administration. Where this occurs osmotic shock may develop which may result in pain, swelling and cell damage.
207. The use of a tonicity agent in this way does not fall within either subclauses (a) or (c) of the definition of ‘therapeutic use’. Again, the question is whether it falls within (b). As with the buffer, the significance of the tonicity agent lies in what it is not rather than what it is. It is not something which causes osmosis. For the reasons I have given in relation to the buffer, I conclude that this use of propylene glycol is not a therapeutic use.

Therapeutic Use of Phenol?

208. The only use of phenol disclosed in the patent is as a preservative whose purpose, as Associate Professor Kayser explained, is to avoid microbial growth in the formulation during storage. Plainly, this use is not a therapeutic use.

Primary Conclusion

209. On the terms of the patent, the tonicity agent, the buffer and the preservative are not used for therapeutic purposes.

Novo Nordisk’s Factual Contentions

210. On the view I take, Novo Nordisk’s attempts to prove via the evidence of Professor Bernkop-Schnürch that the three excipients had therapeutic purposes is misconceived. At a high level of generality Professor Bernkop-Schnürch’s evidence was that:

(a) the preservative, phenol, also prevents the formation of larger oligomers of liraglutide once the formulation is injected. This affects the pharmacokinetic profile of the liraglutide because smaller molecules can permeate the capillary wall to a higher extent than large oligomers;

(b) the tonicity agent, propylene glycol, prevents osmotic shock and avoids pain, swelling and cell damage; and

(c) the buffer, disodium phosphate dihydrate, keeps the pH of the formulation above 8.1 after injection for a period of time which prevents the formation of large oligomers of liraglutide.

211. The evidence in (b) is consistent with the terms of the patent and Associate Professor Kayser’s evidence about what the purpose of a tonicity agent generally is. The use of any tonicity agent (including propylene glycol) has this purpose for that is what tonicity agents are for. However, as I have explained above, the use of a tonicity agent to prevent osmosis is not a therapeutic use because it does not satisfy any of the subclauses of the definition of ‘therapeutic use’. In particular, it is not a use for the purpose of influencing, inhibiting or modifying a physiological process in persons. The use of a tonicity agent simply prevents the circumstances arising which would cause a physiological process (osmosis) to arise.
212. In relation to the preservative phenol in (a) and the buffer disodium phosphate dihydrate in (c), the difficulty with Novo Nordisk’s evidence is that, if accepted, it shows that the two chemicals have a pharmacokinetic effect which could, in principle, perhaps be of therapeutic use. However, the evidence does not descend into the level of detail which would be necessary to assess whether the pharmacokinetic effect actually constitutes a therapeutic effect.
213. To take the use of phenol first, it is one thing to say that phenol prevents the formation of larger oligomers so that the liraglutide will remain as smaller molecules capable of permeating the capillary wall. I can see in principle that this may well lead to an increased uptake of liraglutide. But there the trail runs cold. How does one know that this effect has any therapeutic consequences? There is no evidence that explains:

(a) what rate larger oligomers of liraglutide form using the formulation without phenol;

(b) how much phenol is necessary to achieve this outcome with a given amount of liraglutide;

(c) how long the effect continues at the injecting site; or

(d) what impact the use of the amount of phenol contained in the formulation has on the overall rate at which liraglutide acts at the GLP-1 receptor.

214. I am prepared to accept Professor Bernkop-Schnürch’s evidence that phenol retards the formation of larger oligomers of liraglutide but I am unable to fathom how, armed with that information, I am to infer that the use of the phenol has an actual therapeutic use. I can see that it might, but there is a gulf between the effect that Professor Bernkop-Schnürch describes and the utilisation of that effect for therapeutic benefit.
215. The same may be said of his evidence about the effect that the buffer, by keeping the pH at 8.1 at the injection site, also prevents the formation of large oligomers. I am prepared to accept that large oligomers are less likely to form where the pH is at 8.1. But as with the use of phenol, there is a chasm between accepting that the existence of that pharmacokinetic effect and concluding that the presence of the disodium phosphate dihydrate in these formulations actually has a therapeutic effect in terms of the action of liraglutide on the GLP-1 receptor.
216. Thus, if I am wrong that this evidence is completely irrelevant, I conclude in the alternative that I can make no findings based on the evidence that the excipients in these formulations have therapeutic uses. The question would then arise as to who bore the burden of proving that the excipients had a therapeutic use. It is Cipla which seeks rectification of the Register on the basis that the 862 Patent does not claim a pharmaceutical substance. Cipla therefore bore the burden of proving that the excipients do not have a therapeutic use. If one got to this somewhat upside-down part of the case, I would conclude that Cipla had not discharged that burden.
217. Lest I be wrong about this too, I should record that I found both Professor Bernkop-Schnürch and Associate Professor Kayser to be highly qualified to give their opinions and to be reliable witnesses. An attack was made on Professor Bernkop-Schnürch’s credit on the basis that he was somewhat dogmatic. I would agree that he was much more dogmatic that Associate Professor Kayser and perhaps a little less willing to make concessions. However, I do not think this in anyway affected the value of his testimony. A spirited debate about a paper, ‘the Marunaka paper’, also took place in the concurrent evidence session. For what it is worth, I do not accept that the Marunaka paper provides reliable evidence that the pH for the interstitial fluid of persons with diabetes can be as low as 6.6. It is a review paper and its reference to a pH of 6.6 does not appear to be referenced to any source.
218. I should say in fairness to both Professor Bernkop-Schnürch and Associate Professor Kayser that their difference of opinion emerged from the misdirected nature of the questions which Novo Nordisk had posed for Professor Bernkop-Schnürch’s opinion. Those questions, which lay outside the statute, created a formless field for debate in which both experts, understandably, struggled for structure.

CONCLUSIONS

219. Since a formulation is a pharmaceutical substance, it follows that Cipla’s rectification suit must fail. Novo Nordisk did not submit that Cipla’s rectification suit should also be dismissed because of its delay. There is no evidence before me as to why Cipla waited until 2024 to commence this suit at the very end of the patent’s life. But that delay required an expedited hearing to the disadvantage of other cases pending in the Court. Rectification is a discretionary remedy and, in a proper case, unexplained delay may provide a basis for refusing relief to which a party may otherwise be entitled. Cipla’s application will be dismissed with costs and Novo Nordisk’s cross-claim upheld with costs. The parties should bring in short minutes of order to give effect to these reasons within 7 days.

Associate:

Dated: 12 December 2024