Eli Lilly and Company v Pfizer Overseas Pharmaceuticals [2005] FCA 67 (10 February 2005)

PATENTS – patent for treatment of impotence in man – application for revocation of claim 10 of patent – cross-claim for infringement of claim 10 of patent

PATENTS – construction – whether "cGMP PDE_V inhibitor" means an inhibitor selective for that PDE or an inhibitor affecting that PDE to any extent regardless of effect on other PDEs

PATENTS – novelty – whether the invention was novel when compared to prior art as it existed at priority date – application of reverse infringement test – whether Murray or Korenman teach the use of a cGMP PDE_V inhibitor to treat impotence – whether Korenman needed to disclose the PDE inhibitory activity of pentoxifylline for there to be an anticipation

PATENTS – obviousness – whether the invention compared to the prior art base as it existed before the priority date involved an inventive step – whether the notional team would include a medical practitioner experienced in sexual medicine – whether the team on reading Rajfer would be led as a matter of course to try oral cGMP PDE_V inhibitor to prevent impotence – whether fact that Rajfer’s experiments were in vitro rendered invention not obvious – whether Murray formed part of common general knowledge at time – whether the team on reading Murray would directly be led as a matter of course to try an oral cGMP PDE_V inhibitor to cure impotence – whether secondary indicia of commercial success and satisfaction of long-felt want established inventive step

PATENTS – fair basis – whether claim 10 is fairly based on the body of specification – claim 10 not confined to "compounds of the invention"

PATENTS – sufficiency – whether there is sufficient disclosure that will enable addressee to produce something within each claim without new inventions, additions or difficulty – whether there is sufficient disclosure if the skilled addressee could make a single embodiment within the method claimed – whether the best method is disclosed in complete specification – what is relevant date for assessing if the best method has been disclosed

PATENTS – whether best method disclosed – whether best method known to patentee must be identified as such – whether "complete specification" includes amendments

PATENTS – manner of manufacture – whether the invention is a manner of new manufacture – whether what is claimed in claim 10 is mere desideratum – whether known properties of compounds disclosed in Bell patents made them suitable for orally effective treatment for erectile dysfunction

PATENTS – false suggestion – whether there was false suggestion which materially contributed to the decision to grant the patent

PATENTS – infringement – whether Lilly has infringed claim 10 of patent by supplying and selling product Cialis, a cGMP PDE_V inhibitor – whether Cialis cures or prevents impotence within meaning of claim 10 – whether mere sale or importation can infringe a method claim

WORDS AND PHRASES – "complete specification"

Patents Act 1990 (Cth) ss 7, 13, 18, 40, 116, 117, 138(3), sch 1

Eli Lilly and Company v Pfizer Research and Development Company [2003] FCA 988 cited
Herbert Adams Pty Ltd v Federal Commissioner of Taxation [1932] HCA 27; (1932) 47 CLR 222 at 228 applied
Allsop Inc v Bintang Ltd (1989) 15 IPR 686 at 697 applied
Flexible Steel Lacing Co v Beltreco Ltd (2000) 49 IPR 331 at [81] applied
Bristol-Myers Squibb Company v F H Faulding & Co Limited [2000] FCA 316; (2000) 97 FCR 524 at [52], [61]-[67], [85] applied
Canadian General Electric Co Limited v Fada Radio Limited [1930] AC 97 at 104 applied
Hill v Evans (1862) 4 De G F & J 288, 45 ER 1195 at 301, 1199 cited
Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59; (2002) 212 CLR 411 applied
In re O’Farrell (1988) 853 F 2d 894 at 903 cited
Lockwood Security Products Pty Ltd v Doric Products Pty Ltd [2004] HCA 58; (2004) 212 ALR 1 at [49] applied
Rehm Pty Ltd v Websters Security Systems (International) Pty Ltd (1988) 81 ALR 79 at [95] applied
Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd [2001] HCA 8; (2001) 207 CLR 1 at [14], [16] applied
No-Fume Ltd v Frank Pitchford & Co Ltd (1935) 52 RPC 231 at 243 cited
Edison and Swan Electric Light Co v Holland (1889) 6 RPC 243 at 277
Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2004] UKHL 46 at [78] cited
Esber v The Commonwealth [1992] HCA 20; (1992) 174 CLR 430 at 448-449 cited
Frederikshavn Vaerft A/S v Stena Rederi Altiebolag [2002] FCA 1024; (2002) 124 FCR 243 at 249 cited
Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation [1981] HCA 26; (1981) 147 CLR 297 at 304, 321 applied
Illinois Tool Works Inc v Autobars Co (Services) Limited [1974] RPC 337 at 369 cited
C Van Der Lely NV v Rushton’s Engineering Co Ltd [1993] RPC 45 at 56 applied
Commissioner of Patents v Microcell Ltd [1958] HCA 58; (1959) 102 CLR 232 at 249 applied
Welcome Real-Time SA v Catuity Inc [2001] FCA 445; (2001) 51 IPR 327 at [106]- [107] cited
NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd (1993) 44 FCR 239 at 265 cited
NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd [1995] HCA 15; (1995) 183 CLR 655 at 663 cited
Prestige Group (Australia) Pty Ltd v Dart Industries Inc (1990) 19 IPR 275 at 279-280 applied

Blanco White, Patents for Invention, 5^th ed (1983) at 4-801

ELI LILLY AND COMPANY & ORS v PFIZER OVERSEAS PHARMACEUTICALS & ANOR
NO V604 OF 2002

HEEREY J
10 FEBRUARY 2005
MELBOURNE

IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY	V604 OF 2002

BETWEEN:	ELI LILLY AND COMPANY & ORS APPLICANTS/CROSS-RESPONDENTS PFIZER OVERSEAS PHARMACEUTICALS & ANOR RESPONDENTS/CROSS-CLAIMANTS
JUDGE:	HEEREY J
DATE OF ORDER:	10 FEBRUARY 2005
WHERE MADE:	MELBOURNE

Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.

IN THE FEDERAL COURT OF AUSTRALIA
VICTORIA DISTRICT REGISTRY		V604 OF 2002
BETWEEN:	ELI LILLY AND COMPANY & ORS APPLICANTS/CROSS-RESPONDENTS PFIZER OVERSEAS PHARMACEUTICALS & ANOR RESPONDENTS/CROSS-CLAIMANTS
JUDGE:	HEEREY J
DATE:	10 FEBRUARY 2005
PLACE:	MELBOURNE

1.0 INTRODUCTION [1]
2.0 THE PHYSIOLOGY OF PENILE ERECTION [9]
3.0 PREVIOUS METHODS OF IMPOTENCE TREATMENT [22]
4.0 THE PATENT [25]
5.0 PRIOR PUBLICATIONS [52]
5.1 Rajfer [53]
5.2 Murray [59]
5.3 Korenman [68]
6.0 CONSTRUCTION [71]
6.1 Q1 What is the meaning of "cGMP PDE_V inhibitor" in claim 10? [71]
7.0 PRIORITY DATE OF CLAIM 10 [88]
7.1 Q2 What is the priority date of claim 10? [88]
8.0 NOVELTY [89]
8.1 Q3 Does Murray teach the use of a cGMP PDE_V inhibitor to
treat impotence? [91]
8.2 Q4 Would the teaching of Murray be understood to be at least oral
administration? [92]
8.3 Q5 Was Korenman publicly available before the priority date of
claim 10? [93]
8.4 Q6 Does Korenman teach a method of orally treating impotence
with an effective amount of cGMP PDE_V inhibitor? [94]
8.5 Q7 Is pentoxifylline a cGMP PDE_V inhibitor within the meaning
of claim 10? [96]
8.6 Q8 Does Korenman need to disclose its PDE inhibitory activity to
be an anticipation? [98]
8.7 Q9 Is claim 10 liable to be revoked for want of novelty? [99]

9.0 OBVIOUSNESS [100]
9.1 Q10 Does the notional team constituting the skilled
addressee include a medical practitioner experienced
in sexual medicine? [105]
9.2 Q11 What facts, additional to those mentioned in sections
2.0 and 3.0 above, were part of common general knowledge
of the addressee in Australia at the priority date? [107]

useful alternative to existing methods of treatment? [110]
9.3.1 What Rajfer conveyed to witnesses [110]
9.3.2 What Rajfer conveyed to Pfizer [134]
9.3.3 Satisfying the test for obviousness [135]
9.3.4 Pfizer’s obviousness arguments [138]
9.3.4.1 The source of NO is not disclosed [139]
9.3.4.2 The addressee would only be led to try NO donors [140]
9.3.4.3 The degree of relaxation required for erection is not disclosed [142]
9.3.4.4 PDE_V was not known to be the predominant isoenzyme
in the corpus cavernosum [143]
9.3.4.5 Zaprinast is not a selective PDE_V inhibitor [148]
9.3.4.6 Oral administration is not taught [149]
9.4 Q13 Can Rajfer be relied on to establish a lack of inventive step? [150]
9.5 Q14 Did Murray form part of the common general knowledge
of the addressee in Australia at the priority date? [151]
9.6 Q15 Was Murray a s 7(3) document at the priority date? [153]
9.7 Q16 Would the team on reading Murray (with or without Rajfer)
directly be led as a matter of course to try an oral PDE_V
inhibitor to cure or prevent impotence, in the expectation
that it might well produce a useful alternative to existing
methods of treatment? [154]
9.8 Q17 Can Murray be relied on to establish a lack of inventive step? [158]
9.9 Q18 Do secondary indicia establish inventive step? [159]
10.0 FAIR BASIS [168]
10.1 Q19 Is claim 10 fairly based on the matter described in the
body of the specification? [170]
11.0 SUFFICIENCY [178]
11.1 Q20 Who is the skilled addressee for the purposes of sufficiency? [180]
11.2 Q21 What is the date for determining sufficiency? [182]
11.3 Q22 What common general knowledge and prior art can the
skilled addressee take into account? [183]

11.4 Q23 Is there sufficient disclosure if the skilled addressee could make a single embodiment within the method claimed? [184]

1 Australian patent No 676571 (the Patent) is concerned with methods of treatment of impotence. Although the Patent bears a priority date of 9 June 1993, as a consequence of various subsequent amendments the priority date for the claim in issue in the present proceeding is 10 May 1994. Under the Patent and corresponding patents in other countries the Pfizer Group has marketed Viagra, which has had huge commercial success worldwide with sales of US$1.78 billion in 2002 and increasing thereafter.

2 The applicants Eli Lilly and Company and Eli Lilly Australia Pty Ltd (collectively, Lilly) market a rival product called Cialis. Lilly seeks revocation of claim 10 of the Patent. Viagra itself comes within another claim of the Patent, namely claim 8, the validity of which is not in issue in the present proceeding. However Pfizer in its cross-claim alleges that Lilly’s dealings with Cialis infringe claim 10.

3 Until the second day of final addresses this proceeding had been tried together with another revocation claim, V111 of 2003, brought by Bayer Aktiengesellschaft (Bayer) with a corresponding infringement cross-claim by Pfizer. Bayer and Pfizer then resolved their differences and were by consent granted leave to withdraw application and cross-claim.

4 Lilly contends that the alleged invention lacks novelty and is obvious. Lilly also says that the Patent does not comply with the requirements of the Patents Act 1990 (Cth) (the Act) in that it

• does not disclose the best method known to the patentee for performing the invention (s 40(2)(a));

5 Pfizer disputes these grounds. It says the invention the subject of the Patent was a result of serendipitous discovery: in clinical testing of a compound intended as a treatment for angina unexpected erections in men were noted.

6 Lilly says that it has only imported and sold Cialis and that such conduct does not constitute infringement. Further it says that in any event Cialis does not infringe claim 10.

7 Resolution of the issues will involve, amongst other things, questions of the construction of the Patent and also questions as to the relevant date at which the various criteria are to be applied, having regard to amendments made to the Patent. Likewise Pfizer’s cross-claim for infringement turns in part on a point of construction.

8 I requested counsel to submit a schedule of agreed issues of fact and law. Counsel were able to formulate some issues, but could not agree on others. The issues which appear in this judgment, and which are identified by italicised subheadings, are in part based on those submitted by counsel.

9 Within the penis there are two symmetrical compartments, or corpora, each of which is called a corpus cavernosum. Each contains a network of very small blood vessels or passages. It is changes in the blood flow inside these compartments which lead to tumescence or detumescence.

10 When the penis is in flaccid state, the flow of blood into the corpora cavernosa is restricted by constriction of the smooth muscle which surrounds the vessels in the incoming arterial network. An erection occurs when the smooth muscles relax. The arteries open up, blood flows in, each corpus cavernosum swells and the penis stiffens. When the smooth muscle is contracted the influx of blood is restricted and the penis returns to its normal flaccid state. So, paradoxical as it might seem to the layman, it is relaxation in the penis which causes stiffness and erection.

11 The smooth muscle in the penis is made to relax or contract by the body’s autonomic (involuntary) nervous system. This system operates many other organs and muscles within the body which are not subject to conscious control, such as the vascular system surrounding blood vessels. Sexual stimulus in the brain as a result of something a man sees, feels or thinks is transmitted to the penis by non-adrenergic non-cholinergic (NANC) nerves, that is to say nerves distinct from other nerves in the body called adrenergic and cholinergic.

12 The chemical messenger which causes the cells in the smooth muscle to relax is nitric oxide (NO), a labile (unstable and short-lived) gas. Nitric oxide is produced or released from at least two sources.

13 First, endothelium cells lining the smooth muscle in the penis produce a "relaxing factor" which operates to relax the smooth muscle. This factor used to be known as ERDF (endothelium-derived relaxing factor). In the late 1980s ERDF was discovered to be nitric oxide. Secondly, the NANC nerves themselves also release nitric oxide.

14 In both cases nitric oxide is produced by a reaction from a chemical called L-arginine. Nitric oxide enters the cell and causes a chemical reaction which produces cyclic guanosine monophosphate (cGMP). This in turn activates certain other enzymes which cause or promote chemical reactions which relax the smooth muscle. Because it plays a part in the transmission process cGMP is often referred to as a second messenger.

15 But cGMP can itself be inactivated. A group of enzymes called phosphodiesterases (PDEs) cause cGMP to be turned into 5’GMP, a chemical which is ineffective to relax muscles. The series of biochemical events just discussed is known as the NANC (or sometimes NANC/L-arginine/NO/cGMP) pathway and is illustrated in the following diagram:

16 However penile smooth muscle is also relaxed by other agents, as well as cGMP, and in particular by a chemical called cyclic adenosine monophosphate (cAMP). It is produced in a different pathway, from a different starting material to cGMP, and in response to different messages. But again, there are PDEs which can make cAMP ineffective.

17 Before 1993 it was found that PDEs could be classified into five enzyme groups or families (subsequently more groups have been identified). These were distinguished by Roman numeral subscripts _Ito _V. These groups do not affect cGMP and cAMP uniformly. PDEs _I _IIand _IIIaffect both cGMP and cAMP, PDE_IVaffects cAMP only and PDE_V affects cGMP only. A PDE which affects only, say, cAMP is said to be cAMP specific. A PDE which affects

18 both cGMP and cAMP is not specific, but if it is more effective, or potent, in affecting one than another it is said to be selective to the former. PDE_III for example, is more effective in destroying cAMP than cGMP, so it is said to be cAMP selective.

19 Some chemicals inhibit (wholly or significantly block) the operation of PDEs. But different inhibitors have different effects on different PDEs. For PDE_II there are no known inhibitors. The remaining PDEs are affected by different inhibitors. In particular, it was known prior to 1993 that a compound called zaprinast inhibited PDE_V. It also inhibited PDE_I, but to a lesser extent. Zaprinast was produced by May & Baker and was also called M&B 22,948. The inhibition of a PDE is a reversible effect which is dependent on the concentration of the inhibitor used. One measure of the power, or potency, of an inhibitor is the figure which causes 50 per cent inhibition of the enzyme activity. This is called the IC₅₀ value. Thus the lower the IC₅₀figure, the more potent the inhibitor.

20 Different PDEs are present in different tissues and organs of the body. There is thus a basis for systemic therapeutic use of PDE inhibitors to relax smooth muscle in specific tissues and organs.

21 When a PDE inhibitor has more effect on one PDE than another, for example more effect on PDE_V than PDE_I,it is said to be selective to the former. However there is an issue in this case as to whether the term "cGMP PDE_V inhibitor" appearing in claim 10 of the Patent means an inhibitor selective for that PDE or an inhibitor which affects that PDE to any extent, regardless of its effect on other PDEs.

22 The foregoing may seem disarmingly straightforward. But we need to keep in mind that major features of the physiology of male erection, a process essential to the very survival of the species, were only discovered in the last decade of the twentieth century.

23 Impotence, referred to technically as erectile dysfunction, is defined in the Patent (page 1 lines 6-11) as a lack of power, in the male, to copulate and may involve an inability to achieve penile erection or ejaculation, or both. More specifically, it is said to be an inability to obtain or sustain an erection adequate for intercourse. As recently as the late 1970s it was thought that by far the predominant cause of impotence was psychological. Now the current consensus of international opinion is that from 75 to 90 per cent of men with chronic impotence are suffering from a physical or organic cause.

• Self-administered penile injection with various medications such as papaverine, a PDE inhibitor which acted as a vasodilator (an agent that causes or facilitates the dilation of blood vessels). This was an effective but unappealing form of therapy and had side effects including priapism and scarring of the corpus cavernosum with long term use;

• Amyl nitrate applied topically (by patches to the penis). Again there were side effects such as lowering of blood pressure, unpleasant sensation for the sexual partner and lack of effect over time;

25 There was no available oral treatment. A compound called yohimbine had been available for oral use since the 1970s in the United States and Europe but was registered as a poison in Australia. Clinicians, and no doubt patients, saw oral treatment as something highly desirable.

26 The Patent is entitled "Pyrazolopyrimidinones for the Treatment of Impotence". It begins by identifying its subject matter as relating to "the use of a series of pyrazolo [4,3-d] pyrimidin – 7 - ones for the treatment of impotence" (page 1 lines 3-5). Although reference is made to the use of particular compounds and groups of compounds for the treatment of impotence, the critical claim for the purposes of the present case, claim 10, is, as will be seen, framed differently.

27 The specification then discusses impotence and the prevalence of the condition (page 1 lines 6-18). There is a discussion of the prior art and methods of treatment (page 1 line 19 to page 2 line 9). It is then said (page 2 lines 10-33):

The earlier disclosures referred to are two European patent applications made by Pfizer. In
the present case they have been designated Bell I and Bell II. The structure of "a compound of formula (I)" is then set out and the specification continues (page 4 lines 20-24):

The words underlined (including the word "Unexpectedly" above) were added by amendment on 13 May 1994 (PCT application). By reason of the numerous permissible permutations and combinations, the number of compounds which might fall within formula (I) is of astronomic proportions.

28 After a discussion of various aspects of the formula (I) compounds there is a discussion at page 5 lines 7-16 of the salts of formula (I) in these terms (this passage was also added on 13 May 1994):

29 The specification then goes on to identify subsets of the compounds of formula (I). Each is smaller but more desirable than its predecessor. There is a "preferred group of compounds", a "more preferred group", a "particularly preferred group" and "especially preferred individual compounds". The last group is said to "include" nine compounds, one of which is a compound known as sildenafil. The citrate salt of that compound, sildenafil monocitrate, is the active ingredient of Viagra. However sildenafil monocitrate itself is not one of the nine "especially preferred" compounds identified.

30 The specification then states (page 7 lines 22-28) that the compounds of formula (I), and their pharmaceutically acceptable salts, processes for their preparation, in vitro test methods for determining their cGMP and cAMP inhibitory activities, pharmaceutical compositions of them and routes for their administration for human use are described in Bell I and Bell II.

31 Commencing on page 7 line 29 there is a discussion of a "preliminary investigation" carried out to isolate and characterise the cyclic nucleotide PDEs of the human corpus cavernosum. It is said at page 9 line 23 that the investigation identified three PDE isoenzymes. The cGMP-specific PDE_V is said to be predominant. However PDE_II and PDE_IIIare also present. In discussing the results of the investigations it is said (page 9 lines 7-8) that the first fraction "was found to be insensitive to stimulation by calcium/calmodulin". Pfizer says that the skilled reader would understand by this that PDE_Iwas not present.

The specification then gives an example of what is meant by this potent and selective inhibition, saying that "one of the especially preferred compounds of the invention" (although the compound is not identified) strongly inhibits PDE_V but "demonstrates only weak inhibitory activity" against PDE_II and PDE_III. It is then said (page 9 line 35):

33 At page 10 line 5 the specification discusses toxicity testing of some of the compounds of the invention in rat, dog and mouse. The compounds are not identified.

34 At page 10 line 13 there is reported oral administration to volunteers of "certain especially preferred compounds" in both single dose and multiple dose studies. It is stated that these studies have confirmed that "one of the especially preferred compounds induces penile erection in impotent males". Again, the compounds are not identified.

35 The matters discussed in [30]-[33] above were added to the Patent on 13 May 1994.

36 At page 10 line 25 it is said that oral administration of "the compounds of the invention is the preferred route being the most convenient and avoiding the disadvantages associated with i.c. administration" and that a preferred dosing regimen for a typical male is 5mg-75mg of compound three times daily.

37 From page 11 line 4 to page 12 line 5 there is what is, arguably at least, a consistory clause. It is made up of six components.

39 Secondly, there is "further provided" a process for preparing such a composition.

This is the first mention in the Patent of a substance other than the "compounds of the invention" or "the compounds of formula (I)". The words underlined were added on 13 May 1994. The words double underlined were added on 4 April 1995 (PCT application amendments) and on the same date the words struck out were deleted.

These words in their present form were added on 3 October 1996 (Australian national phase amendments in response to the Examiner’s report). There follow chemical formulae which were added on 13 May 1994.

46 Claims 2, 3 and 4 narrow claim 1 by limiting various groups of formula (I). Claim 5 specifies nine particular compounds. These are the "especially preferred individual compounds" already referred to. The formulae in claims 2 to 5 were added on 13 May 1994.

47 Claims 6, 7 and 8 each specify one particular compound of formula (I). Claim 7 specifies the freebase known as sildenafil and claim 8 specifies sildenafil monocitrate. As already mentioned, it is the active ingredient of Viagra. Sildenafil has previously been claimed in the patent granted on the Bell I application. In each of claims 2 to 8 the claim is for "The method according to" the previous claim, the words "The method" having been added on 3 October 1996.

48 Claims 6, 7 and 8 were added on 6 January 1997 (Australian national phase amendments in response to the Examiner’s report).

49 Claim 10 is critical for the present case since the allegedly infringing product of Lilly (Cialis), does not contain any of the compounds of the earlier claims.

50 Claim 10 takes its present form from an amendment which I ordered, on the application of Pfizer, on 19 September 2003: Eli Lilly and Company v Pfizer Research and Development Company [2003] FCA 988.

There follow seven further claims but they are not relevant for the present case.

53 Many scientific publications were referred to in evidence. For present purposes the most important were the following.

54 In the issue of the New England Journal of Medicine for January 1992 there appeared an article "Nitric Oxide as a Mediator of Relaxation of the Corpus Cavernosum in Response to Nonadrenergic, Noncholinergic Neurotransmission" by Jacob Rajfer et al (Rajfer). Dr Rajfer and his fellow authors worked in the Division of Urology, Department of Surgery and the Department of Pharmacology at the University of California, Los Angeles. While there has been much debate in this case as to just what Rajfer would have conveyed to the skilled reader in 1992, there is no dispute as to its importance or scientific merit or that it was published in a pre-eminent journal of medical science. Witnesses variously described it as "elegant" and "captivating". One of the co-authors, Dr Louis J Ignarro, subsequently won the Nobel Prize. The publication of Rajfer was reported in the New York Times.

55 Rajfer records experiments with strips of corpus cavernosum tissue obtained from 21 impotent men who had undergone prosthetic implants. The stated purpose of the experiments was to determine whether NO is involved in the relaxation of the smooth muscle in the corpus cavernosum that allows penile erection.

56 Commencing the body of the paper, Rajfer notes that impotence is a major clinical problem in adult men with approximately 10 million sufferers in the US alone. Rajfer then summarises the existing state of knowledge. Corporal smooth muscle relaxes markedly in response to stimulation by NANC neurons (1988,1989). Various substances had been proposed as causes of this relaxation. Relaxation of rabbit corpus cavernosum mediated by NANC neurons was attributed to NO and cGMP (1990). The object of the study was to ascertain whether NO played a part in similarly mediated relaxation of the corpus cavernosum in humans, and therefore in penile erection.

57 The tissues were placed in an organ bath, in a solution designed to replicate the body’s environment, with additions which blocked adrenergic and cholinergic reactions, thus ensuring that the operation of the NANC system could be isolated. The tissues were attached to devices for measuring and recording relaxation. Electrical-field stimulation (EFS) was provided by electrical impulses at frequencies of 2, 4, 8, and 16 Hz to mimic sexual stimulation.

58 The experiments included adding S-nitroso-N-acetylpenicillamine (SNAP) to increase the production of NO and methylene blue to inhibit cGMP. Importantly for the present case, zaprinast was tested and it was found that it enhanced the effect the SNAP-added NO had on relaxation. The last-mentioned test was the subject of Figure 3:

The discs and circles represent readings taken respectively before and after the addition of zaprinast.

60 In the April 1993 issue of Drug News and Perspectives there appeared an article by Kenneth J Murray entitled "Phosphodiesterase V_A Inhibitors" (Murray). This was a review article. As such, it did not publish original research, but rather collected, summarised and analysed research already published and suggested possible future directions.

62 At the outset it is noted that cyclic nucleotide PDEs had long been regarded as potential targets for therapeutic agents. More recently, interest in this area had been renewed by the recognition that there were five distinct PDE families and that tissues have different complements of them. There was therefore a logical foundation for selective PDE inhibitors to be used to increase cyclic nucleotide levels in specific target tissues or organs. It is said that the article will discuss briefly the PDE families and describe one of these families, PDE V_A. (In the present case it is not suggested anything turned on Murray’s concentration on PDE V_Aas opposed to other members of the PDE_V family.)

63 After discussing PDE families and their nomenclature, the article summarises in tabular form the five known PDE families including their respective effects, if any, on cGMP and cAMP and their "selective inhibitors", if any. The article then discusses PDE V_A. It notes that, compared to other PDEs, PDE V_Aexhibits a rather limited tissue distribution. After pointing out the specificity for cGMP, the article suggests that the combination of limited tissue distribution and substrate specificity suggests that a specific PDE V_A inhibitor "could have a narrow range of physiological and pharmacological actions."

64 Under the heading "PDE V_A inhibitors" the article discusses zaprinast and concludes that there is no doubt that it is a "potent inhibitor of PDE V_A, but there is also evidence that it can inhibit PDE I". It mentions other inhibitors and gives the chemical structure for eleven of them, including zaprinast.

65 Under the heading "Physiological effects of PDE V_A inhibition" the article discusses the effect of various PDE V_A inhibitors (mainly zaprinast) on a number of in vitro smooth muscle preparations. Amongst the papers noted is Rajfer. There is also mention of various in vivo tests on animals, and in particular the effects of zaprinast on mean arterial blood pressure.

66 Under the heading "Potential therapeutic use of PDE V_A inhibitors" the article notes that the only such inhibitor which had been clinically evaluated was zaprinast. On adult asthmatics it was found zaprinast reduced bronchoconstriction induced by exercise, but not that provoked by histamine. In contrast, zaprinast had no effect on exercise-induced asthma in children. The article noted that these were small clinical trials using a compound that may have actions other than PDE V_A inhibition. At present therefore PDE V_A inhibitors, in the author’s view, "must be considered to be compounds with potential rather than established clinical use". Smooth muscle relaxation appeared to be "the most promising of the potential uses of PDE V_A inhibitors", and possible therapeutic utilities "could include vasodilation, bronchodilation, modulation of gastrointestinal motility and treatment of impotence". (It will be noted that the last-mentioned possible use was not included in the summary at the front of the article.)

67 The article goes on to note that such inhibitors will relax most smooth muscle types in vitro, but this did not necessarily mean that a nonselective action would be seen in vivo. Selective actions could be achieved by virtue of the fact that many effects of PDE V_A inhibitors are dependent on the level of guanylate cyclase activity. The author suggests that

68 The article concludes that the present therapeutic potential of PDE V_A inhibitors

69 In the April 1993 issue of the Journal of the American Geriatrics Society there appeared an article by Stanley G Korenman and Sharon P Viosca from the Department of Medicine, University of California, Los Angeles entitled "Treatment of Vasculogenic Sexual Dysfunction with Pentoxifylline" (Korenman).

70 Korenman reported the results of a double-blind, randomised clinical trial to evaluate the use of pentoxifylline to treat erectile dysfunction. Patients studied were diagnosed as impotent and had had no successful coital events or attempts at intercourse for an average of 5.7 years. Patients were orally administered 400 mg of pentoxifylline (Trental) three times a day for twelve weeks. Trental was available only in tablet form at that time. Effects of treatment were measured subjectively by the number of coital episodes per month and objectively by changes in penile-brachial pressure index (PBPI). Korenman reported a "significant improvement in erectile function" in half the patients completing the protocol:

71 Twenty four couples were selected. The men had had no successful coital events or attempts at intercourse for an average of 5.7 years. Five men did not comply, in that they either failed to take their medication or complete the daily log. One failed to complete the study because of an acute anxiety attack during the placebo stage. Of the remaining 18, nine improved from four unsuccessful attempts while on placebo to 90 successful attempts after treatment with pentoxifylline. Five of the 18 experienced erections while receiving pentoxifylline, but did not attempt intercourse. Only 3 of the 18 patients showed no improvement. However, 53 of the total 90 episodes were reported by the one patient. The other 37 episodes were spread among the remaining 8 patients, with at least one patient reporting only 2 successful episodes during the 12 week study. Moreover, pentoxifylline has a delayed onset of action of two weeks.

72 Pfizer contends that the term refers to a PDE_V inhibitor which is selective for PDE_V, that is to say a compound which has a lower IC₅₀for PDE_V than it has for PDE_I, PDE_II, PDE_III and PDE_IV.Lilly’s submission is that the term refers to any inhibitor that has any activity against PDE_V, regardless of whether it is more or less selective for that PDE than any other of the known PDEs.

73 As a matter of ordinary language, the term "cGMP PDE_V inhibitor" in claim 10 conveys to the reader the concept of a substance identified only by the presence of a particular function or capacity, namely that it inhibits cGMP PDE_V. Whatever else it might do, including but not limited to, the inhibiting of PDEs other than cGMP PDE_V, a substance which inhibits cGMP PDE_V will answer that description. Pfizer’s argument therefore necessitates reading a limitation or qualification into the ordinary language so that the reader will understand that the author of the document was intending to convey a narrower meaning.

74 In Herbert Adams Pty Ltd v Federal Commissioner of Taxation [1932] HCA 27; (1932) 47 CLR 222 the High Court was concerned with a claim for exemption from sales tax for a pastrycook’s product generally known as sponge. The relevant exemption provision excluded "cakes". The company relied on evidence of trade usage among pastrycooks in which "sponge" had a specialized meaning, namely an article in which some form of fat is a substantial or basic ingredient, as distinct from other things, such as sponge, which consumers might call cake. In rejecting this argument Dixon J (with whom Rich J agreed) said at 228:

75 Herbert Adams is perhaps not precisely on all fours since the court there had to choose between a wide ordinary meaning and a narrow technical meaning whereas the present case involves competing wide and narrow meanings of a technical term. Nevertheless, the linguistic logic of the passage cited is compelling, and particularly so once it appears that the suggested narrower meaning is not a matter of uniform expert opinion or usage and is not consistent with the language of the document as a whole.

76 Pfizer relied on expert evidence in support of the meaning for which it contended, including evidence obtained in the course of cross-examination of Lilly and Bayer witnesses. Evidence as to the meaning of technical terms is admissible although the construction of the specification is a matter of law and belongs to the court: Allsop Inc v Bintang Ltd (1989) 15 IPR 686 at 697, Flexible Steel Lacing Co v Beltreco Ltd (2000) 49 IPR 331 at [81].

77 There were Pfizer witnesses on this issue, Dr Fischmeister (France) and Professor Pittler (USA), who are or have been members of the international Cyclic Nucleotide Nomenclature Committee for the Classification of PDEs. But such evidence may be of limited value when construing a patent, which is not a document published as an exposition of abstract science, but is directed towards workers in a particular field at a particular point in time. If a question arose as to the meaning of a term in the argot of the Marseilles docks the opinion of the Academie française may not carry much weight.

78 There was contrary evidence from Lilly witnesses, in particular Dr Kruse and Dr Gristwood. The latter in particular was quite unmoved despite persistent cross-examination. I do not think this warranted criticism of him as partisan and I do not accept that he is somehow discredited as being a "serial" Lilly witness, if all that means is that he has given evidence for Lilly on more than one occasion in litigation over Viagra patents.

79 Importantly, there was evidence of usage which supported Lilly’s view. The words "selective" was commonly used to identify the PDE against which the inhibitor has the most activity. Thus in an article "Primary Sequenter of Cyclic Nucleotide Phyophodiesterase Isozines and the Design of Selective Inhibitors" by Beavo (the chairman of the Nomenclature Committee) and Reifonydo in TiPS April 1990 the authors provide a table (p154) headed "Inhibitors Selective for Phosphodiesterase Isozine Families" listing various inhibitors selective for PDEs _I to _Vand also a list headed "Common Non-selective Inhibitors". The article also contains a sub-heading "Selective Inhibitors Currently Available". Another example is Murray, who refers (p152) to zaprinast as "a reasonably potent and selective PDE_V inhibitor". His Table IV (p154) is headed "PDE V_A Inhibitors". In discussing the compounds appearing in that table Murray says (p152):

80 If Pfizer’s strict view is correct, the term "selective PDE_V inhibitor" would be tautologous. Yet, as the foregoing examples show, if the context requires it, the term is often used. Pfizer itself did so in its Canadian Viagra patent where claim 25 (corresponding to our claim 10) claims:

81 Likewise claim 24 of the Pfizer US patent speaks of "an effective amount of a selective cGMP PDE_V inhibitor".

82 So, as far as the evidence goes, especially evidence of usage, I conclude that if the context is appropriate, the qualifier "selective" may be used in conjunction with "cGMP PDE_V inhibitor".

83 There is a distinction to be made between a context where an identified substance is said to have a particular capacity or function ("zaprinast is a cGMP PDE_V inhibitor") and one where a particular activity or function identifies the substance or substances ("cGMP PDE_V inhibitor"). In the former case where those in scientific discourse are mutually aware that zaprinast is selective for PDE_V it is understandable that the particular quality of selectivity is sometimes not stated. In the latter case, neither the definite nor indefinite article is used. The only identifying criteria of the substance or substances is that it or they must do something in relation to cGMP PDE_V, namely, inhibit it. There is no logical reason why the further qualifier of selectivity must be implied.

84 The context of the Patent supports Lilly’s contention. The Act provides that in interpreting a complete specification the Court may refer to the specification without amendment: s116. Prior to the 2003 amendment, claim 9 spoke of "an orally effective amount of a cGMP PDE inhibitor" and claim 10 was for

85 Plainly claim 9 was not speaking of any selectivity; it extended to all inhibitors of any cGMP PDE. So in the absence of any express qualification as to selectivity, claim 10 would naturally be read as referring to a subset of claim 9. In other words, claim 10 focuses on those of the claim 9 group of cGMP PDE inhibitors (of necessity both selective and non-selective) which happen to have one particular characteristic, that of inhibiting cGMP PDE_V.

86 Further, claim 10 is not limited to "compounds of the invention" (or "compounds of formula (I)" if, contrary to my opinion, those terms are not synonymous). Therefore the reader who has seen at page 9 line 28 to page 10 line 4 that the "compounds of the invention" are "potent and selective inhibitors" of cGMP PDE_V would take it that compounds not "of the invention", but within claim 10, need not necessarily have that quality. The converse of the familiar aphorism is equally true: what is not unclaimed is claimed. Indeed, the reader would know that the specification stressed selectivity, not only in the passage just quoted but in page 2 lines 14-27 where it is said that "(t)his selective enzyme inhibition (scil for cGMP PDEs in contrast to cAMP PDEs) leads to elevated cGMP levels" providing the utilities already disclosed in the Bell patents. If selectivity is so important, the reader might think that where selectivity is meant, the author would say so.

87 This reading may be illustrated by a homely example, a cookbook containing a recipe for apple pie. In the discussion preceding the recipe the author says that the preferable variety of apple is Granny Smith and explains why – texture, water content, sugar content etc. But the actual recipe merely says "Take 1 kg apples". The natural meaning is that any apples will do. They may not produce as good a result as would Granny Smiths, but they will still produce an apple pie.

88 I conclude therefore that while sometimes "cGMP PDE_V inhibitor" may, without more, connote an inhibitor which is selective for PDE_V, in claim 10 in the context of the Patent as a whole it does not.

89 It is now accepted by Pfizer that the date is 13 May 1994, the date of filing of the PCT application.

90 It is a ground for revocation that the alleged invention is not a patentable invention: s 138(3)(b). One of the requirements for an invention to be patentable is that, so far as claimed in any claim, the invention, when compared with the prior art base as it existed before the priority date of that claim, is novel: s 18(1)(b)(i). The statutory definitions of the concepts of novelty and the prior art base are to be found in s 7(1) and the Dictionary in sch 1 of the Act.

91 The authorities on novelty are comprehensively discussed by Black CJ and Lehane J in Bristol-Myers Squibb Company v F H Faulding & Co Limited [2000] FCA 316; (2000) 97 FCR 524 at [61]- [67]. For present purposes it is sufficient to say that the question is whether the alleged anticipation would, if the patent were valid, constitute an infringement (the "reverse infringement test"). Where the alleged anticipation takes the form of a publication (a "paper anticipation"), the hypothetical infringement does not arise because of the publication per se; rather one asks whether the publication teaches, in the sense of directing, recommending or suggesting, the invention under consideration. As the Privy Council pointed out in Canadian General Electric Co Limited v Fada Radio Limited [1930] AC 97 at 104,

8.1 Q3 Does Murray teach the use of a cGMP PDE_V inhibitor to treat impotence?

92 The highest that Lilly could put their case was that such a use was a "promising potential therapeutic use". On its face, that falls well short of the required standard. The pervading tone of Murray is hypothetical, cautious, abstract and speculative, an approach encapsulated in the careful warning that such inhibitors

Amongst a number of potential uses of these inhibitors Murray gives no particular prominence to the treatment of impotence and does not mention that use in its statement of the theme of the article.

93 If, in the relevant sense, Murray taught the method of claim 10 (which I find not to be the case) that teaching would be taken to include oral administration; see particularly the reference to studies of oral asthma treatment. The answer to this question is: Yes.

94 Since the relevant priority date is 13 May 1994 (see Q2 above) Pfizer accepts that on this basis the answer to this question is: Yes.

95 Korenman states at the outset that Trental renders red blood cells more deformable thus allowing them to pass more readily through partially obstructed arterial channels. The study was therefore to test

Thus Korenman teaches away from the pith and substance of claim 10 of the Patent, which is the use of a substance identifiable by a particular function, namely the inhibition of cGMP PDE_V. There is no reference to pentoxifylline as a cGMP PDE_V inhibitor. Nor is there in the Trental product information leaflet. At best for Lilly, there is a passing comment towards the end of Korenman that pentoxifylline

There is a footnote referring to two 1989 papers, dealing respectively with the effect of pentoxifylline in rabbit pulmonary circulation (Kaapa et al) and canine arteries and veins (Hoeffner et al). These do not teach the method of claim 10. In particular, Hoeffner discusses cAMP, not cGMP. In any event, the tentative terms in which these papers are referred to, and the fact that they deal with animal, not human studies and conditions other than impotence, confirm the conclusion that Korenman, read as a whole, does not contain any directions, let alone clear and unmistakeable directions, to use the method of claim 10.

96 Moreover it seems that pentoxifylline does not act via a cGMP PDE_V inhibitory mechanism of action. It needs to be taken over a course of weeks. Korenmam explicitly recognises this, stating

Viagra and Cialis have effect shortly after the administration of a single dose.

8.5 Q7 Is pentoxifylline a cGMP PDE_V inhibitor within the meaning of claim 10?

97 Because I have rejected Pfizer’s construction argument that claim 10 is referring to a selective cGMP PDE_V inhibitor (see section 6.0 above), it is not necessary to discuss the experiment conducted by Lilly and designed to show that pentoxifylline was in fact a selective cGMP PDE_V inhibitor, with the consequence that Korenman was an anticipation of claim 10.

98 But in any event, I am not persuaded that at the relevant time pentoxifylline would have been identified as a selective cGMP PDE_V inhibitor. The burden of the evidence of Dr Ellis, Mr Burslem and Dr Cherry was that it was not so described in the literature. It would be referred to either as non-selective or a cAMP PDE inhibitor.

99 Yes. Otherwise there will not be disclosure of "the specific details necessary for the practical working and real utility of the alleged invention": Hill v Evans (1862) 4 De G F & J 288, 45 ER 1195 at 301, 1199. The contrast between the effectiveness of pentoxifylline on the one hand and Viagra and Cialis on the other has already been referred to ([95] above).

101 To be patentable an invention must, so far as claimed in any claim, when compared with the prior art base as it existed before the priority date of that claim, involve an inventive step: s 18(1)(b)(ii). Section 7(2) and (3) provide:

102 As is often the case, here the hypothetical skilled person will be constituted by a research team or group, the members of which contribute different skills to the notional task. The test is whether the group would, in all the circumstances, including a knowledge of all the relevant prior art, directly be led as a matter of course to try something in the expectation that it might well produce a useful result: Aktiebolaget Hassle v Alphapharm Pty Limited [2002] HCA 59; (2002) 212 CLR 411 at [52]. If it can be said that upon doing so the group would arrive at the alleged invention then obviousness is made out.

103 To become part of common general knowledge in the relevant field, a publication needs to be not just accessible; it must have been assimilated into the consciousness of the skilled worker: Alphapharm at [57].

104 In assessing whether an invention is obvious, the court must be aware of the dangers of hindsight, of starting with the alleged invention and looking back: Alphapharm at [21], [78].

105 A court can look at what are termed secondary indicia of obviousness. If an invention is a commercial success, and/or fulfils a long-felt want, that may support a conclusion that it is not obvious because were it obvious somebody would have discovered it earlier. (Of course, if the invention had in fact been discovered earlier, it would be invalid for want of novelty.)

106 It is common ground that the notional team would include a pharmacologist and a chemist with experience in drug discovery and development (a medicinal chemist). I accept Lilly’s submission that the team would also include a medical practitioner experienced in the treatment of impotence.

107 Claim 10 is directed to a method of treatment, not to a composition. Thus those involved in treatment of impotence would be the best persons to explain the level of knowledge in the field (to fellow members of the team as well as to the court). Many of the medical practitioners who gave evidence had an interest in the underlying biochemical mechanism of impotence: Mr Pryor, Dr Cherry, Dr McMahon, Mr Cartmill and Mr Wisniewski. Their interest, directly related as it was to their clinical work, cannot be relegated, as Pfizer argued, to the level of an intellectual hobby. The notional team in this context is not necessarily the same as that to be postulated in considering issues of sufficiency and best method.

108 Because of the different ways in which counsel formulated this issue, it was not easy to isolate just what facts were disputed as being within common general knowledge or otherwise. This is partly due to the way in which Pfizer presented what senior counsel for Lilly, with some justification in my opinion, criticised as an unnecessarily complicated picture of the prior art. Moreover, it is an artificial exercise to make findings as to the common general knowledge sans Rajfer and Murray since they formed part of the common general knowledge (and in the case of the former, admittedly so).

109 Subject to those comments, I should record my rejection of Pfizer’s assertion that as at 13 May 1994 there was an "incomplete and inaccurate knowledge of the presence of PDE’s and their types in corpus cavernosum".

110 Also, in the light especially of Murray, it is an overstatement to say that as at the relevant priority date each of cGMP PDE_V and PDE_I "were known to be located in tissues situated throughout the body, including in the smooth muscle of the vasculature platelets and lung tissue".

111 Dr Christopher McMahon of Sydney has practised full time in the field of genito-urinary/sexual health since 1982. He read Rajfer when it was published.

112 He saw the paper as clarifying the role of NO as part of a "cascade of neurotransmitters in the development of penile erection". Beyond that "primary objective" the main conclusion he drew was that there may be a role for zaprinast and perhaps for PDE_V drugs as a treatment for men with erectile dysfunction.

113 Dr Denis Cherry is the Medical Director of the Perth Human Sexuality Centre which specialises in all areas of male sexuality with a special interest in impotence. He has been interested in the treatment of impotence since working as a general practitioner in Donnybrook WA between 1981 and 1989. At that time and subsequently he worked closely with the late Professor Keogh at the Reproductive Medicine Research Centre in Perth.

114 Dr Cherry read Rajfer late in 1992. The paper established to his mind that the NO/cGMP pathway was responsible for the human erectile response and that at an ex vivo level the pathway could be manipulated by either potentiating NO formation or by the use of a cGMP PDE inhibitor such as zaprinast.

115 Initially he focussed on NO but on close reading he realised that zaprinast was described as a drug. As is often the case in reading scientific papers, he said, you read it the first time noting the primary interest, in this case NO, but often you go back and find that there is something else which strikes a chord or an interest.

116 While he was aware that ex vivo data did not necessarily translate into in vivo data in human beings, the model used by Rajfer, penile tissue from impotent men, was the best ex vivo data that a researcher could obtain before moving into an animal model.

117 Mr John Pryor is a London urologist who has practised since 1975 specialising in uroandrology (problems relating to the male genital tract). Prior to 1993 he had professional contact with Australian urologists at international conferences and had visited Australia in a professional capacity. Whether or not he would qualify as a skilled addressee himself, his evidence is admissible. The whole point of the skilled addressee concept is that it involves an hypothetical person, not a real one.

118 He read Rajfer shortly after it was published. His "understanding of the neurological mechanisms fell into place" because the paper demonstrated that the cGMP PDE inhibitor zaprinast "was potentially useful in the treatment of erectile dysfunction because it elicited the needed physiological response, ie smooth muscle relaxation in penile tissue taken from impotent human patients".

119 Dr Robert Gristwood is the Research and Development Director of Arachnova Limited, a British pharmaceutical company specialising in identifying new therapeutic uses for existing drugs and providing consultancy services to the pharmaceutical industry. He holds a Ph D in Pharmacology from Oxford University (1982). He has had extensive experience in pharmacological research and development with a number of companies. He has had contact with Australian scientists working in his area.

120 He read Rajfer at the latest in February 1993 because on 27 February 1993 he gave a paper at a conference in Nice in which he referred to Rajfer. He said, citing Rajfer (Dr Gristwood’s paper itself was in Spanish):

121 In his evidence he said that on reading Rajfer he appreciated that impotence was a new therapeutic possibility for PDE_V inhibitors in light of the fact that zaprinast, a known selective PDE_V inhibitor, was shown to enhance smooth muscle relaxation of human corpus cavernosal tissue. If he were looking for a new treatment for impotence, Rajfer would have "strongly suggested" to him the use of PDE_V inhibitors.

122 Dr Lawrence Kruse of New Hampshire USA has since 1995 been a consultant to the pharmaceutical and biotechnology industry. Prior to that he was in charge of chemical research and development at Stirling Winthrop, with responsibility for over 200 scientists. He has had contact with Australian scientists and has lectured at universities in Australia.

123 He does not specifically recall reading Rajfer at the time of its publication, but believes, "given the significance of the scientific information" in it, a co-worker would have drawn it to his attention. In his view Rajfer "completely elucidate(s)" the NANC pathway leading to erection. It confirms that NO is the biochemical agent involved in the relaxation of the corpus cavernosum, which leads to penile erection. The article also confirms that the NO-induced relaxation is mediated through cyclic GMP. Smooth muscle relaxation induced by NO is mediated through an increase in the cGMP PDE inhibitor zaprinast. He knew by 1992 that zaprinast was a selective inhibitor of PDE_V.

124 Importantly for him, Rajfer used the most relevant experimental model possible, namely actual human penile muscle tissue from men suffering from erectile dysfunction. Because of this he would have had a "high level of confidence" that Rajfer’s results would have been directly transferable to the in vivo human situation.

125 Dr Peter Ellis of Canterbury UK is the Executive Director of Exploratory Development of that country’s member of the Pfizer Group. He has a Ph D in pharmacology from Edinburgh University (1983). He is one of the inventors of the invention described and claimed in the Patent.

126 He read Rajfer in late January 1992. In his affidavit he said that all Rajfer did, as far as he believed, was to confirm the early animal results and demonstrate that NO was a neurotransmitter in the NANC pathway in the penis in man. Rajfer, Dr Ellis said, deliberately excluded all pathways except the NANC system and used a "battery of standard tools" to establish that the NANC pathway, in the absence of other physiological regulation, could relax corpus cavernosal tissue and that the mediator appeared to be NO.

127 In cross-examination, however, Dr Ellis made a number of concessions. Amongst other things, he agreed that Rajfer showed that zaprinast, used as a tool, potentiated or augmented or enhanced the effect of electrical field stimulation and NO under the experimental conditions defined in the paper. Indeed, he agreed that the enhancement effect of zaprinast on the effects of NO was a well-known fact and that was why it was used in the study.

128 He adhered to an answer he had given in cross-examination in the UK proceeding and rephrased it as follows:

129 Dr Richard Palmer has a Ph D in pharmacology from University College, London. Since 1973 he has been involved as a research scientist involved in drug development. He is currently Chief Executive Officer of Alizyme plc, a drug development company based in Cambridge UK.

130 He had not read Rajfer before the present proceeding because it was "an area where NO was involved that was outside our interests". In his affidavit he discusses earlier publications and says that the "whole thrust" of Rajfer was "consistent with a focus on demonstrating/confirming the features of the NANC/L-arginine/NO/cGMP pathway in this preparation, thus bringing together previous work". He did not agree with the Lilly and Bayer witnesses that Rajfer disclosed a logical therapeutic use of PDE inhibitors for the treatment of impotence. He said that Rajfer "neither establishes nor seeks to establish" that the increase in relaxation shown in figure 3 in vivo "would enable a usable erection".

• that readers are able to take what they like from pieces of fundamental work like Rajfer in terms of deriving potential therapeutic possibilities;

• that Rajfer makes the clinical suggestion that a defect in the NANC pathway might be a cause of impotence;

• that the electrical field stimulation frequencies used in Rajfer are of the order of human physiological frequencies – otherwise the paper would not have been published; a peer review would have considered it physiologically irrelevant; and

• it is not known what percentage of relaxation is required in humans to achieve erection and that it has to be a matter of trying it in humans.

132 Dr Alan Robertson has a Ph D in synthetic organic chemistry from the University of Glasgow (1981) and has worked in Australia since 1992. As will be seen below, I found his evidence valuable on issues of sufficiency. However on the present issue he rather deferred to Dr Palmer.

133 Dr Robertson read Rajfer at the time of its publication as part of his general reading. He considered it "an excellent paper published in an important very high impact journal".

134 He is the inventor of a successful drug Zomig. In his work on that drug he extrapolated from in vitro tests on rabbit leg artery to a treatment of humans for migraine. In cross- examination he said:

135 Immediately after publication of Rajfer Dr Peter Ringrose, then Head of Discovery at Pfizer UK, circulated a copy to his management team with his comment written on it:

UK-92,480 was Pfizer’s internal designation of sildenafil monocitrate, then being studied as a potential angina treatment. What happened thereafter will call for further discussion. At the moment the observation can be made that Pfizer’s own reaction provides a neat symmetry to Dr Gristwood’s Nice paper. Both are contemporaneous, actual reactions by skilled readers to the Rajfer paper. They are by themselves strong indications that Rajfer would, when published, have taught a skilled reader that compounds that inhibit cGMP PDE_V were likely to be useful in the treatment of impotence.

9.3.3 Satisfying the test for obviousness

136 I accept the evidence of Lilly and Bayer witnesses as to what Rajfer would have conveyed. As the distinguished Australian scientist Sir Gustav Nossal said in a paper given at the Australian Legal Convention in 1997,

and scientific papers have a "cautious tone of voice" with a "liberal sprinkling of qualifications" or, as one of the witnesses in the present case put it, are couched in "science speak". In a discourse in that style and tone Rajfer conveys, not certainty, for that is not required, but a well-founded expectation that further testing of cGMP PDE_V inhibitors might well provide a treatment for impotence or, in Dr Robertson’s words, would be likely to have that effect.

137 It is not to the point that the "focus" or "thrust" of Rajfer might have been on NO, or that zaprinast was used as a laboratory tool, if the potential for cGMP PDE_V inhibitors is otherwise conveyed.

138 The present case is not one of trying each of a number of possible choices where the prior art gave no direction as to which of many possible choices is likely to be successful, or exploring a promising field of experimentation where the prior art gave only general guidance as to the particular form of the claimed invention or how to achieve it: In re O’Farrell (1988) 853 F 2d 894 at 903 per Judge Rich, cited with approval in Alphapharm at [76].

139 It is not disputed that Rajfer formed part of common general knowledge in Australia at the relevant time. However, Pfizer advanced a number of arguments against Lilly’s reliance on Rajfer to support a finding of obviousness.

140 This does not matter. Rajfer teaches that, wherever NO comes from, it is the NO/ cGMP pathway that causes smooth muscle relaxation, that it is defects in this pathway that can cause impotence and that relaxation can be significantly potentiated by the use of a PDE_V inhibitor.

141 There was evidence from Dr Palmer that he was focused at Wellcome on the "front end" of the pathway, that is production of NO. However this seems to have been due to commercial reasons: their expertise lay in that area and other companies were already working on PDE inhibitors.

142 Further, there were known problems with NO. It was an unstable free radical with a short half life and NO donors were unstable, highly polar compounds with problematic chemical lability which made them unsuitable for therapeutic application.

143 The Patent itself does not contain any specification of the percentage relaxation required to generate an erection sufficient for intercourse. Nor is it known what percentage of relaxation is required to achieve erection in humans, as Dr Palmer admitted.

144 Pfizer asserted that because the identity of PDEs in the corpus cavernosum and the tissue distribution of PDE_V throughout the body was not known at the priority date there was potential for any orally administered PDE inhibitor to have potentially dangerous effects elsewhere in the body.

145 However, there was evidence from Dr Palmer and Mr Wisnieski that the human body possesses mechanisms for controlling its response to agents and maintaining blood pressure at the right level.

146 Moreover, zaprinast was known to be a selective PDE_V inhibitor (see below). Since Rajfer taught that zaprinast potentiated the relaxation of corpus cavernosal tissue, that in itself showed that PDE_V was present.

147 At the priority date Dr Ellis was aware from earlier publications by Taher that PDE_III, PDE_IV and PDE_V were present in the corpus cavernosum. So also were other witnesses such as Dr Pryor and Dr McMahon. Determination of the isoenzyme profile within the corpus cavernosum had become a routine task.

148 It is convenient to note here that subsequently Murray stressed that systemic administration could be used for PDE_V inhibitors in view of their selective action. For example, he states:

149 Pre-priority date publications by Beavo (1990), Nicholson (1991) (as well as Murray) identified zaprinast as a selective PDE_V inhibitor. As already mentioned, a number of witnesses were aware at the time of this characteristic of zaprinast.

150 Oral administration was an obvious preferable route for drug administration, and particularly so in the case of impotence treatment. It hardly needs evidence to conclude that a man would prefer taking a tablet to injecting his penis.

151 For the foregoing reasons, Yes. The usual course in pharmacological discovery is from laboratory tests on animals to clinical testing on humans – first on healthy humans to test dosage levels for effectiveness, side effects and the like and then on patients suffering from the relevant condition. But a significant feature of the present case, as a number of witnesses stressed, is that the laboratory testing published by Rajfer was on human tissue from men suffering from the condition in question. Rajfer thus provided an unusually powerful indication that further examination of the PDE_V inhibitor possibility was worth trying and that the expectation that such examination might produce a worthwhile result was a very reasonable one.

152 Murray appeared in the April 1993 issue of a journal called Drug News and Perspectives, published in the UK. There is abundant evidence that it was read by people in the pharmaceutical industry. Pfizer seems to accept that it was read in Australia. At least two companies for which its witness Dr Robertson has worked in Australia subscribed to it and he himself read it as part of his "literature diet".

153 Pfizer attacked Drug News and Perspectives on the ground that it is an "industry focussed publication", is not one of the "higher flying journals", and there is doubt whether it is peer-reviewed. These grounds, even if made out, disclose no basis from excluding the contents of articles published in that journal from common general knowledge. In any case, in the practical world of patents, for a journal to be "industry focussed" seems to be a qualification for inclusion within the field of common general knowledge rather than the reverse.

154 For the foregoing reasons Murray formed part of common general knowledge (see Bristol-Myers at [52]) but also qualified as a s 7(3) document.

155 Murray specifically recommends the designing of PDE_V inhibitors for, inter alia, impotence. It is probably unnecessary and artificial to consider Murray without Rajfer because the former specifically refers to the latter (footnote 31).

156 The effect of Murray was recognised by one of the Pfizer inventors, Dr Terrett. He had written a memo on 17 May 1993 noting that his team had considered filing a broad "field of use" claim for all cGMP PDE inhibitors in impotence and raising the question whether they could assume that all such inhibitors, irrespective of potency, would have efficacy in impotence. He then read Murray and wrote a week later (26 May 1993):

157 Nevertheless Pfizer filed its UK application, notwithstanding the written advice of Mr I T Barmish from its patent department on 30 November 1993 that

158 In the present case Pfizer’s main answer to Murray is that it does not state that cGMP PDE_V inhibitors should be administered orally. However, as already stated, oral administration would self-evidently be the preferred route. Murray expressly deals with the question of limited tissue distribution of PDE_V. Moreover he states that the only clinical studies thus far of the selective PDE_V inhibitor zaprinast were in asthma studies where the administration was oral (footnotes 51 and 52). One of these was a study in children where clinical studies will not usually be undertaken unless the drug is extremely safe.

160 As already mentioned, Viagra has enjoyed enormous commercial success. In addition to the worldwide figure given above, gross sales in Australia in the year of launch (1999) were A$15.9 million (332,084 units) and the forecast for 2004 is A$26.3 million (515,070 units). The long-felt want for an oral treatment for impotence instead of the existing modes can be readily accepted. However, an examination of the chronology of the development of Viagra rather lessens the force of these facts and circumstances for present purposes.

161 In brief, Pfizer’s change of direction with UK-92,480, a known PDE_V inhibitor, from angina to impotence was due not to serendipity (the faculty of making desirable but unsought-for discoveries) but to the impact of Rajfer.

162 A response to Dr Ringrose’s request for suggestions as to the use of UK-92,480 ([134] above) came from Dr David Brown, a manager in Chemistry, in the form of a post-it sticker attached to a copy of Rajfer. It read:

On 19 November 1993 Dr Terrett sent a memo to Mr Barmish following what was described as the "recent evidence of efficacy of UK-92,480 in preliminary phase II impotence trials". The memo stated that the writer, together with Dr Ellis, was "currently trying to identify any documents prior to January 1992 that clearly suggest that impotence may be treated by the PDE inhibition mechanism". The purpose of this request appears from the following question in it:

163 On 31 January 1992 Pfizer’s Hypertension Task Force released a report entitled "Appraisal of New Approaches to the Treatment of Hypertension" written by Dr Ellis and Dr Terrett which stated, under the sub-heading "Impotence", the following:

The passages underlined are word for word from a report in the New York Times of 9 January 1992, datelined Los Angeles the previous day, which quotes Dr Rajfer and notes that his research is described in that day’s issue of the New England Journal of Medicine.

164 Pfizer commenced its Study 207 on UK-92,480 for angina in March 1992. Dr Ellis had drawn up the protocol for the study in the previous month. This was a study using healthy volunteers and designed to test dosage levels. Substantial concern among those conducting the study was then aroused by reports of myalgia and dyspepsia and this was the focus of attention. However, there were also reports of spontaneous erections at higher dose levels. According to Dr Ellis, this was the first occasion upon which spontaneous erections had been reported to him as an adverse effect during clinical trials of UK-92,480.

165 The results of Study 207 were formally reported within Pfizer on 26 June 1992. It was said that UK-92,480 would be "considered for the treatment of impotence on the basis of the spontaneous erections seen in Phase I multiple dose study".

166 On 18 January 1993 Dr Ellis and Dr Terrett made a Request for Patent Application within Pfizer

"Recent literature disclosures" is an obvious reference to Rajfer, notwithstanding the reluctance of Dr Ellis to frankly concede as much.

167 On 22 June 1993 Dr Ellis and Dr Terrett signed a Record of Inventorship as follows:

It will be observed that the "earliest record of the innovation" is the Appraisal of 31 January 1992, that is to say before Study 207 and its report of spontaneous erections in March 1992.

168 The foregoing demonstrates how intimately the ideas conveyed by Rajfer were bound up in Pfizer’s work which led to the alleged invention the subject of the Patent. No doubt Pfizer was lucky in having its existing angina project which could be steered in the new direction of impotence treatment. This gave it a head start, which helps to explain its commercial success. But the story of stumbling by chance on a solution which fulfilled a long-felt want does not fit the historical facts. Moreover, that story does not take account of the rapid developments in scientific knowledge in this area.

169 The claim of a patent must be "fairly based on the matter described in the specification": s 40(3).

170 In assessing whether there is fair basing within the meaning of s 40(3), it is necessary to split the patent into the claims and the body of the specification, in order to see whether the claim is fairly based on the matter described in the latter: Lockwood Security Products Pty Ltd v Doric Products Pty Ltd [2004] HCA 58; (2004) 212 ALR 1 at [49]. The inquiry is into what the body of the specification as a whole discloses as the invention: Lockwood at [99]. The rationale is that the inventor is not entitled to claim a monopoly more extensive than is necessary to protect that which he himself (in his specification) said is his invention: Blanco White, Patents for Invention, 5^th ed (1983) at 4-801, cited in Lockwood at [47]. This is a matter arising essentially on the contents of the complete specification: Blanco White, ibid. Fair basing is an issue quite separate from other grounds of invalidity such as obviousness and want of novelty. A patent may be found invalid for want of fair basing even though every other ground of challenge fails: Lockwood at [48].

171 Much of the discussion on the construction point (section 6.0 above) is relevant on this issue. The more one re-reads the Patent, the more apparent it becomes that claim 10 (and the abandoned claim 9) are an awkward fit. The Patent, both in its body and in claims 1 to 8, deals with the use of compounds identified by their chemical structure. As the opening words state (page 1 lines 3-5), the invention is said to relate to the use of a series of formulaically defined substances for the treatment of impotence. The detailed hierarchy of compounds has already been referred to ([28] above). Again, they are identified by chemical structure.

172 The terms "compounds of the invention" and "compounds of formula (I)" are used interchangeably. For example, in the Patent the "compounds of the invention" are said to be disclosed (page 3 line 17), and the "compounds of formula (I)" said to be described (page 7 line 28), in Bell I and II. Thus the same compounds are being referred to. There is no formula (II). It would appear from some other patents that were in evidence that the expression "formula (I)" is a conventional term often used by patent drafters even where there is only one formula.

173 It is common ground that claim 10 is not confined to "compounds of the invention" or "compounds of formula (I)". But when, in the body of the specification, there is an account of the "preliminary investigation" and the recording of a finding of "potent and selective" inhibition of the cGMP-specific PDE_V (page 9 line 28), it is "compounds of the invention" that is referred to. The immediately following IC₅₀figure is for one of the "especially preferred" compounds of the invention.

174 In the six components of the (arguable) consistory clause commencing at page 11 line 4 ([36] et seq above), the only reference to compounds defined not by chemical structure but by inhibitory function (page 11 lines 19-27) is to inhibition of cGMP PDE, not cGMP PDE_V. This was obviously written to support the now abandoned claim 9.

175 In final submissions, senior counsel for Pfizer said that the words of claim 10 define the invention for the purposes both of obviousness and fair basing. But nowhere else in the Patent is there matter describing what is claimed in claim 10. Mere coincidence of language between a claim and part of the body of a specification does not establish fair basing if that part of the language of the specification does not reflect the description of the invention in the light of the specification as a whole: Lockwood at [87]. In the present case however there is not even coincidence of language.

176 Pfizer argued that the invention claimed in claim 10 related to a "principle of general application", namely the use of a cGMP PDE_V inhibitor as an orally effective treatment for the cure and prevention of male erectile dysfunction. The relevant disclosure, so the argument went, was (at page 9 lines 23-27) that PDE_V was the predominant PDE of the three PDEs present in the human corpus cavernosum and the "remarkable finding" that there was no PDE_I. It was said that the evidence of Dr Robertson disclosed that a person skilled in the art obtains sufficient information from the specification to enable the identification, design and synthesis of cGMP PDE_V inhibitors outside formula (I) which fall within claim 10 and to utilise that compound as an orally effective treatment to cure or prevent male erectile dysfunction.

177 I must say that the supposed importance of the absence of PDE_I has been conveyed in a somewhat oblique fashion. But more importantly, the issue of fair basing requires focus on the text of the patent to determine whether, in the words of Gummow J in Rehm Pty Ltd v Websters Security Systems (International) Pty Ltd (1988) 81 ALR 70 at 95, cited in Lockwood at [69],

What s 40(3) requires is that the claim be fairly based "on the matter described in the specification". Whether or not a reader might be able to achieve the result suggested by Dr Robertson, the Patent itself contains no such disclosure, and certainly no reasonably clear one. The Patent’s thrust directs the reader away from further investigation and testing because what it does disclose are the specified "compounds of the invention".

180 The specification, including the claims, must be read as a whole: Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd [2001] HCA 8; (2001) 207 CLR 1 at [14], [16]. In that case the High Court (at [25]) cited the statement in No-Fume Ltd v Frank Pitchford & Co Ltd (1935) 52 RPC 231 at 243 that

As Cotton LJ said in Edison and Swan Electric Light Co v Holland (1889) 6 RPC 243 at 277, it is not necessary that the addressee

181 It may be that in assessing the workability of the Patent, as distinct from the obviousness of the alleged invention, the medical practitioner experienced in sexual medicine might not be a member of the team.

182 This is a convenient point to note the observation of Lord Hoffmann in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2004] UKHL 46 at [78] that the hypothetical addressee should be taken to know the basic principles of patentability (including, one might add, some familiarity with the way patents are written).

183 Consistently with the conclusion I have reached on the similar issue arising in relation to fair basing (section 12.1 below), the date is, at the earliest, the date of grant. This means that claims 6, 7 and 8 and the other amendments made on 6 January 1997 must be taken into consideration.

185 The High Court said in Lockwood (at [60]) that all alternative means need not be disclosed; it is sufficient if the addressee can produce "something within each claim without new inventions or additions or prolonged study of matters presenting additional difficulties". This necessarily means that one embodiment is sufficient.

186 Whether or not this statement is, as Lilly contends, obiter, it is a considered statement by a unanimous bench of the High Court and, it hardly needs saying, should be followed by a judge at first instance. Nor do I see any logical ground for the distinction urged by Lilly between mechanical patents and chemical patents; certainly the Act provides no basis for such a distinction. In any case, there seems to be a practical dilemma inherent in Lilly’s argument. If one embodiment is not enough, are two, ten or twenty required, and who is to say how many are needed?

187 Pfizer’s witness Dr Robertson was in practice as a medicinal chemist in Australia at the relevant times. He has had substantial, and successful, experience in the making of medicinal compounds, including the reading and working of patent specifications. His experience included both new drug research and development and generic drug design and manufacture. He was subjected to lengthy and searching cross-examination. However, he seemed to me to deal adequately with matters put to him and I was left with a firm impression that he was a competent and practical expert in his field.

188 As to the identification of the compound subjected to testing and for which results are claimed in the Patent, I accept the evidence of Dr Robertson that the skilled addressee reading the Patent as a whole, and noting the progressive narrowing of claims, would be likely to focus on claim 8. He deposed:

189 Dr Scammells, Professor Charman and Dr Kruse all were prepared to accept that the addressee would narrow the area of interest down to the nine "especially preferred individual compounds" specified at page 6 line 32 to page 7 line 21 and in claim 5, but steadfastly refused to go any further. I did not find their evidence in this regard convincing. Professor Charman for example said there was no "data" to base any further narrowing. Apart from the fact that, as Dr Robertson said, one might reasonably take it that the data at page 9 line 30 was likely to refer to the compound which the structure of the text otherwise indicated was of particular interest, the issue here is not whether the Patent on its face provides a complete scientific rationale for a particular embodiment but rather whether the Patent identifies among a theoretically vast number of alternatives (apparently a common feature of chemical patents) one, or a reasonable number, which the addressee can select and work.

190 If the embodiment selected by the addressee for working comes within claim 10, it does not matter that it also might fall within claim 8. Put another way, there is no justification in my view for requiring the one embodiment of claim 10 relied on to be outside the compounds of formula (I).

191 The Bell patents are incorporated by reference into the Patent and they contain sufficient instructions to enable a medicinal chemist to synthesize the compounds described in them.

192 It is not disputed that there is sufficient information in the Patent (incorporating the information in the Bell patents) to enable a skilled medicinal chemist to:

193 To a medicinal chemist, the freebase compound claimed in claim 7 of the Patent is a sulphonamide. Dr Robertson identifies it as having the same chemical structure as compound 12 in Bell I. The latter patent sets out the steps involved in synthesising the sulphonamide compounds in formula (I) by general description and precise example. Dr Robertson’s evidence is that the steps set out in the Bell patents are all standard chemistry procedures.

194 It would be necessary to test for oral bioavailabilty, toxicity and effectiveness, but the evidence shows that while these steps call for skill, they are essentially routine for those skilled in this area. The term routine here (and in other contexts in this case) is not used as a synonym for simple and easy. In the present case the hypothetical skilled workers at the hypothetical workbench are persons holding academic qualifications at the Ph D level together with practical experience. It would not be necessary to employ such persons unless the task they had to perform was a difficult one. Yet this does not of itself mean that the Patent could not be worked without further invention.

195 There is no evidence that anyone attempted to make something falling within claim 10 but failed. Indeed Lilly succeeded; its product Cialis is within claim 10, hence the cross-claim for infringement. Lilly’s case on sufficiency would be stronger if it had revealed to the court the process by which it came to produce Cialis and shown that in the course of doing so it had to take steps which were truly inventive.

197 As already mentioned, s 40(2)(a) requires the complete specification, in the description of the invention, to include the best method known to the applicant of performing the invention.

198 The Act has a Dictionary in sch 1. It provides definitions of terms used in the Act "unless the contrary intention appears". One such definition is:

(Section 116 provides that the Commissioner or a court may, in interpreting a complete specification as amended, refer to the specification without amendment.)

199 The definition of "complete specification" in its present form with the reference to amendment of the specification was inserted by the Patents Amendment (Innovation Patents) Act 2000 (Cth). As its title suggests, this Act was primarily concerned with the introduction of a system of less formal patents called innovation patents. However, counsel were unable to suggest any connection between innovation patents and the amendment of the definition of "complete specification".

200 Lilly contends that the date is 13 May 1994, the date of the filing of the PCT application. Claim 8 which, as already mentioned, identifies sildenafil monocitrate, the active ingredient of Viagra, was not added by amendment (along with claims 6 and 7) until 6 January 1997. Pfizer submits that the relevant date is the date of grant (10 July 1997) or the date of commencement of this proceeding (17 September 2002).

201 Counsel on both sides addressed detailed arguments which included reference to English cases going back to 1778, together with a historical review of the United Kingdom patent legislation and a number of United States authorities. But without intending any disrespect, I think the question is a straightforward matter of statutory interpretation.

202 The term "complete specification" is used elsewhere in the Act in a context where it is clear the Act’s dictionary meaning, including amendments to the specification, is intended. Examples are s 105 (amendments directed by the court), s 112 (complete specification not to be amended, except by the court, where proceedings pending), s 218 (apportionment of costs where court finds some claims invalid but others not) and s 222(2) (Commissioner to arrange for selling copies of complete specifications). Using the term "complete specification" to mean the specification as amended is therefore a natural usage, and not really an artificial extension of what the term would convey even if the dictionary definition were not present. Therefore we might expect the Act to make it clear where, in a particular context, that meaning is not intended, as indeed it does in a number of instances: see ss 41(1), 42(1)(b), 79B(1)(a), 102(2A).

203 Section 138(3) sets out the grounds for revocation. Some are expressed in the past tense ((c) patentee has contravened a condition, (d) patent obtained by fraud etc). The ground presently under consideration is in the present tense ((f) that the specification does not comply with s 40(2) or (3)). This rather suggests that the court looks at the specification as at the time the proceeding is commenced; it would be odd if, for example, in dealing with a fair basing ground (s 40(3)) the court could not consider claims added or modified by amendment subsequent to the filing of the patent application. The general presumption is that judicial proceedings to enforce alleged rights are determined according to the law existing at the time when the proceedings are instituted, unless statute otherwise provides: Esber v The Commonwealth [1992] HCA 20; (1992) 174 CLR 430 at 448-449 per Brennan J. His Honour was in dissent as to the result in that case but the majority judgment is consistent with the principle stated: Frederikshavn Vaerft A/S v Stena Rederi Altiebolag [2002] FCA 1024; (2002) 124 FCR 243 at 249 per Sundberg J. A patent is a public document given legal force by the Act. It can be amended, as the Act provides, both before and after grant. There is therefore force in Pfizer’s primary argument that the court must consider the validity of the Patent at the time of commencement of the proceeding.

204 I accept that if the matter were unconstrained by statute, there are rational reasons which could be advanced for Lilly’s conclusion. However, there could also be grounds for adopting the Pfizer approach. But there is a statutory provision applicable. The question is not whether one side or the other has the preferable solution, but whether the plain meaning of the statute should govern or whether that meaning leads to a result which is irrational, absurd, extraordinary, capricious or obscure: Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation [1981] HCA 26; (1981) 147 CLR 297 at 304, 321. Pfizer’s construction hardly merits these latter characterisations. As mentioned, the date of commencement of the proceeding has a logical basis. Alternatively, there is much to be said for the view that it is the date of grant when the rest of the world, including potential infringers, should be able to assess the merits of the invention for which monopoly is granted and how it might be performed: Illinois Tool Works Inc v Autobars Co (Services) Limited [1974] RPC 337 at 369.

205 I conclude therefore that the relevant date is, at the earliest, the date of grant and not the date asserted by Lilly.

206 No: see C Van Der Lely NV v Rushton’s Engineering Co Ltd [1993] RPC 45 at 56.

207 I do not understand it to be disputed that the best method known to the Pfizer was (and is) the use of sildenafil monocitrate.

208 Dr Robertson’s evidence in this regard, which I accept, has already been mentioned in relation to the sufficiency issue (section 11.0 above). At the relevant time Dr Robertson would have identified the compounds of claims 6 and 7 as compounds of special interest. He would have noted that the salt claimed in claim 8 was the only salt specifically and individually identified. He would have understood the specification as teaching that these were the compounds of greatest interest to the patentee and he would have started at this point.

209 The Patent lists at page 5 lines 7-16 a number of salts, including organo-carboxylic acids, one of which is citrate. The evidence is that citrate salts are standard pharmaceutically acceptable salts. They are commonly used in pharmaceutical development, and have no toxicological consequences. It is a matter of standard routine chemistry to make the monocitrate salt form of a freebase. In any case, the identification of the salt is not essential. The salt breaks down in the stomach and is simply a means of facilitating administration. Sildenafil itself is identified as the third of the "especially preferred individual compounds" (page 7 line 7), the third compound of claim 5 and the single compound of claim 7.

210 I conclude therefore that to a skilled reader the Patent does disclose the freebase and salt of the especially preferred compound. Lilly’s argument to the contrary pays insufficient attention to the requirement that the specification and claims must be read as a whole. It is this requirement which distinguishes the best method ground from the fair basing ground which I have upheld. As already mentioned, there it is necessary to separate claim from the body of the patent.

212 Section 18 (1)(a) of the Act relevantly provides that an invention is a patentable invention if the invention (which includes an alleged invention – see Dictionary in sch 1), so far as claimed in any claim, is a manner of new manufacture within the meaning of section 6 of the Statute of Monopolies.

213 It is not an invention in this sense to take a substance and make out of it a new article for which the known properties of the substance make it suitable, even if the substance has not been used to make that article before: Commissioner of Patents v Microcell Ltd [1958] HCA 58; (1959) 102 CLR 232 at 249. A mere method or scheme is not an invention: Welcome Real-Time SA v Catuity Inc [2001] FCA 445; (2001) 51 IPR 327 at [106]- [107]. Nor is a "mere desideratum", that is to say something which is disclosed as no more than a wished for result: NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd (1993) 44 FCR 239 at 265. These criteria are to be applied to the specification on its face: NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd [1995] HCA 15; (1995) 183 CLR 655 at 663.

214 There can be no doubt that oral treatment of impotence was seen as desirable and much preferable to the existing alternatives of intracavernosal injection, surgical implantation, vacuum construction devices and reconstructive surgery. However, the alleged invention disclosed in the Patent goes further. The Patent specification tells the reader the manner in which the desired object is to be achieved, namely by the use of an orally effective cGMP PDE_V inhibitor. This cannot be characterised, as Lilly says, as a "bare idea".

215 In the Bell patents Pfizer disclosed a number of compounds including the formula (I) compounds which were suggested to be useful for a number of indications including the treatment of angina within a particular dosage range. The compounds disclosed in those two patents were identified as potent and selective inhibitors of both PDE_I and PDE_V.

217 The test is whether there was a false suggestion which materially contributed to the decision to grant the Patent: Prestige Group (Australia) Pty Ltd v Dart Industries Inc (1990) 19 IPR 275 at 279-280.

Lilly contends that this gives the false impression that one of the nine "especially preferred individual compounds" identified at page 6 line 30 to page 7 line 28 have been tested and worked and that the patentee was disclosing, as one of the nine compounds, its best method. Another ground alleging false suggestion was abandoned in final addresses.

219 The nine especially preferred individual compounds previously described at page 6 line 30 to page 7 line 28.

221 True it is that the compound which achieved the experimental results referred to on page 10 was not a freebase compound, as were the "especially preferred individual compounds". But, as Dr Robertson’s evidence established, a skilled addressee would not in this context draw a distinction between a compound in its freebase form and the same compound in its salt form. Sildenafil monocitrate on administration to man gets converted by the stomach from the citrate into the hydrochloride and then absorbed as the freebase, so it is the freebase or, to use Dr Robertson’s expression, the unfettered molecule, that is absorbed and causes the response, not the citrate itself.

222 There was no false suggestion, let alone one which materially contributed to the grant of the Patent.

223 A patent gives the patentee the exclusive rights, during the term of the patent, to exploit the invention: s 13(1). "Exploit", in relation to an invention, includes make, sell, hire or otherwise dispose of the product or, where the invention is a method or process, use the method or process or make, sell etc a product resulting from such use: Dictionary sch 1.

224 Generally speaking, infringement is the doing of an act, without the authority of the patentee, which the patentee has the exclusive right to do: Bristol-Myers at [85].

225 If the use of a product by a person would infringe a patent, the supply of that product by one person to another is an infringement of the patent by the supplier: s 117(1). "Use" of a product by a person is a reference to, inter alia, the use of a product in accordance with any instructions for the use of the product, or any inducement to use the product, given to the person by the supplier or contained in an advertisement published by or with the authority of the supplier: s 117(2)(c).

226 Lilly’s allegedly infringing product is marketed under the name Cialis. Its active ingredient is a substance known as tadalfil. Lilly has made certain formal admissions for the purposes of this case as follows:

227 Lilly has admitted the supply of Cialis, to divers doctors in Australia for the purposes of conducting confidential clinical trials in Australia in the periods September 2000 – February 2001 and March 2001 – September 2001. Lilly further admits that Eli Lilly and Company (the first cross-respondent) requested Eli Lilly Australia Pty Ltd (the second cross-respondent) to so distribute those Cialis tablets.

229 By the consumer medicine information included in the packaging for Cialis tablets, Lilly provides explicit instructions for use, including:

15.2 Q34 Does Cialis "cure or prevent" impotence within the meaning of claim 10?

230 Lilly contended that Cialis does not "cure or prevent erectile dysfunction in man in need of such treatment" within the meaning of claim 10 because it treats such dysfunction "on an ad hoc basis for a limited period of up to about 36 hours". This is not to "cure", so the argument goes, because the person is not treated so that he no longer has the disease and is restored to health. Likewise it is said that to "prevent" in this context means to forestall the onset of the disease, so as to keep it from occurring.

This would accord with the understanding of the term "impotence" (or its more scientific sounding synonym "erectile dysfunction") in ordinary speech. Accordingly claim 10 is not referring to some underlying physiological or psychological condition. In the vernacular, it is concerned with a performance problem. Cialis will treat that problem and thus "cure" or "prevent" it in the sense that a man taking Cialis will not have the problem for 36 hours. This question must be answered in the affirmative.

232 Lilly’s argument is that "mere sale or importation" does not infringe a method claim.

233 In Bristol-Myers at [82]-[96] Black CJ and Lehane J give detailed consideration to the construction of s 117. The case involved a use claim for a cancer treatment medication and the supply by a manufacturer of a product to medical practitioners with instructions consistent with the use claim. In that context, their Honours held there had been infringement by the supplier. Lilly did not seek to distinguish that decision and I therefore follow it.

234 I conclude that if claim 10 of the Patent were valid, Lilly’s conduct would have constituted an infringement of it.

235 The Patent contains a number of claims which define the scope of an alleged invention for the treatment of impotence. Claim 8 claims the use of a compound, defined in terms of its chemical structure, which is the active ingredient of Viagra. This claim is not alleged to be invalid. Claim 10 claims the use of all compounds which have a certain physiological effect. It is this claim which Lilly says is invalid, and which Pfizer says is infringed by Lilly’s product Cialis.

236 I find that claim 10 is invalid because it is obvious, that is to say it does not disclose an inventive step in the light of common general knowledge in this area, particularly having regard to two scientific publications which became part of that knowledge before the relevant date.

237 I also find that claim 10 is not fairly based on matter disclosed in the specification in the Patent.

238 On other issues Pfizer succeeds. Claim 10 is novel and is not invalid for insufficiency, failure to disclose the best method, not being a manner of manufacture and being obtained by a false suggestion.

239 I further find that were claim 10 valid, Lilly’s sales of Cialis would have constituted an infringement.

240 The proceeding will be adjourned to a date to be fixed. Counsel are directed to submit minutes of proposed orders.

I certify that the preceding two hundred and thirty-nine (239) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Heerey.

Counsel for the Applicants:	D K Catterns QC and K J Howard (with them D Katz QC of the New Zealand Bar)

Solicitors for the Applicants:	Blake Dawson Waldron

Counsel for the Respondent:	D Shavin QC, J Baird and H Rofe

Solicitors for the Respondent:	Corrs Chambers Westgarth

Dates of Hearing:	8-12, 15-19, 22-26, 27-30 November, 1, 11-15 December 2004

Date of Judgment:	10 February 2005

(a)	prior art information made publicly available in a single document or through doing a single act; and
(b)	prior art information made publicly available in 2 or more related documents, or through doing 2 or more related acts, if the relationship between the documents or acts is such that the person skilled in the relevant art in the patent area would treat them as a single source of that information;

(a)	Cialis is an inhibitor of cGMP PDE_V;
(b)	Cialis tablets are intended to be prescribed by doctors in Australia as a treatment to be taken orally for the treatment of erectile dysfunction in men;
(c)	Cialis (in 10mg and 20mg tablet form), has been approved by the Australian Therapeutic Goods Administration (TGA);
(d)	Lilly imports, sells, offers for sale or otherwise disposes of Cialis tablets in Australia; and
(e)	Cialis:
(i)	inhibits PDE_V;
(ii)	is >10,000 – fold more potent for PDE_V than for PDE_I, PDE_II, PDE_III and PDE_IV;
(iii)	is approximately 700-fold more potent for PDE_V than for PDE_VI.

•	synthesize each of the nine "especially preferred compounds" (claimed in claim 5); and
•	test and screen those compounds for PDE_V inhibitory activity.

Federal Court of Australia

Eli Lilly and Company v Pfizer Overseas Pharmaceuticals [2005] FCA 67 (10 February 2005)