Novo Nordisk A/S v Eli Lilly And Company
[2000] APO 67
•7 November 2000
OFFICIAL NOTICE
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Application : No. 682061 in the name of Novo Nordisk A/S
Title: Acylated insulin
Action: Opposition to the grant of a patent by Eli Lilly and Company
Decision: Issued
Abstract
The opposition is unsuccessful on all grounds.
Section 40 matters: One form of the invention is not exemplified, and there is no general method of preparation disclosed. A method of preparation would immediately suggest itself, so the claims are fairly based and there is a full description.
Novelty:The invention is novel. None of the citations disclose an insulin that is mono-substituted on the e-amino group of the LysB29 amino acid.
Inventive step: The invention is inventive. The evidence does not establish that it would have been routine to try acylating the LysB29 amino acid, nor that it would have been routine how to mono-acylate the e-amino of the LysB29.
Manner of manufacture: There is an invention on the face of the specification.
PATENTS ACT 1990
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Re:Patent Application No. 682061 by Novo Nordisk A/S, and an opposition to the grant of a patent by Eli Lilly and Company
BACKGROUND
Patent application number 76520/94 was filed as an international application designating Australia (number PCT/DK94/00347) on 16 September 1994 by Novo Nordisk A/S (hereafter referred to as Novo). The application claims priority from DK 1044/93 (which was filed on 17 September 1993) and US 190829 (which was filed on 2 February 1994). The application was examined by the Commissioner, and advertised accepted under the serial number 682061 on
18 September 1997.
Eli Lilly and Company (hereafter referred to as Lilly) filed a notice of opposition on 17 December 1997. A statement of grounds and particulars was served on 17 March 1998, and amended on 27 August 1998. The evidence stages were completed on 18 January 2000.
The matter was heard in Melbourne on 3 and 4 August 2000. Novo were represented by Mr Glenn McGowan of counsel, assisted by Mr Chris Schlicht and Dr Debbie Yin Foo of Phillips Ormonde & Fitzpatrick. Ms Birgitte Stephensen, a European patent attorney of Novo, was also present. Lilly were represented by Dr Annabelle Bennett S.C., instructed by Mr Mark O'Donnell and Dr Bill Pickering of F.B.Rice & Co. Dr Andrea Walsh and Dr Mark Brader from Lilly were also present.
GROUNDS OF OPPOSITION
The statement of grounds and particulars specifies the following grounds of opposition:
· section 40 matters
· novelty
· inventive step
· manner of manufacture
I note that in order to find that the grounds of opposition are made out, I must be clearly satisfied that the patent, if granted, would not be valid (see F.Hoffmann-La Roche AG v New England Biolabs Inc [2000] FCA 283 at [67]).
EVIDENCE
The evidence in support of the opposition consists of declarations by Mr O'Donnell, Dr Jeffrey and Professor Chisholm. The evidence in answer consists of declarations by Dr Markussen, Professor Colagiuri and Professor Hearn. The evidence in reply consists of declarations by Mr O'Donnell, Dr Widmer and Dr Jeffrey.
Further evidence was served by Novo. This evidence consists of a further declaration by Professor Hearn. Evidence in response to the further evidence consists of additional declarations by Dr Widmer and Dr Jeffrey.
Copies of a large number of documents are attached as exhibits to the declarations. I will refer to these declarations and the exhibits where relevant in my decision.
"FURTHER EVIDENCE"
At the hearing Dr Bennett presented me with a copy of United States patent number 3,528,960. This document is not particularised, and has not been served as evidence. Dr Bennett argued that the present application lacks novelty in the light of this patent, as well as in the light of the other documents in the statement of grounds and particulars.
Mr McGowan objected to this document. Dr Bennett stated that the document was part of the admitted prior art in the present application, so either it did not take Novo by surprise or it was part of the common general knowledge (based on the decision in Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 46 IPR 553). It is clear to me that if Lilly intend to rely on this document as a citation, it must be a part of their case and be introduced in evidence. There is presently no formal request to amend the statement of grounds and particulars or to serve the document as further evidence.
I informed both counsel that I would deal with this matter by taking basic submissions on the technical content of the document during the hearing. If the document is clearly of high relevance, then I will determine what should be done about introducing the document into evidence. (Detailed submissions on the document could be made once the document is in evidence.) If the document is clearly not relevant, then it is not necessary to deal further with this matter. I am extremely grateful that Mr McGowan was able to make submissions on this document at short notice.
My conclusion is that this document is clearly not sufficiently relevant. Consequently it is not necessary to consider any issues connected to this document. My reasons for finding that this document is not relevant are given under the heading of “NOVELTY”, later in this decision.
THE SPECIFICATION
The specification relates to derivatives of human insulin. The description occupies 56 pages, and includes 30 examples. Sequence listings occupy the next 31 pages. The specification ends with 22 claims and three pages of drawings.
The description
The description begins with a review of the literature on protracted insulin compositions (pages 1 to 3). There follows a summary of the problem addressed by the specification:
"However, there still is a need for protracted injectable insulin compositions which are solutions and contain insulins which stay in solution after injection and possess minimal inflammatory and immunogenic properties.
One object of the present invention is to provide human insulin derivatives, with a protracted profile of action, which are soluble at physiological pH values.
Another object of the present invention is to provide a pharmaceutical composition comprising the human insulin derivatives according to the invention.
It is a further object of the invention to provide a method of making the human insulin derivatives of the invention."
[page 3 line 23 to page 4 line 5]
The invention is then described in terms of the structure of the insulin derivatives. The broadest form of the invention is an insulin in which five amino acid positions are variable:
"Accordingly, in its broadest aspect, the present invention relates to an insulin derivative having the following sequence:
wherein
Xaa at position A21 and B3 are, independently, any amino acid residue which can be coded for by the genetic code except Lys, Arg and Cys;
Xaa at position B1 is Phe or is deleted;
Xaa at position B30 is (a) a non-codable, lipophilic amino acid having from 10 to 24 carbon atoms, in which case an acyl group of a carboxylic acid with up to 5 carbon atoms is bound to the e-amino group of LysB29, (b) any amino acid residue which can be coded for by the genetic code except Lys, Arg and Cys, in which case the e-amino group of LysB29 has a lipophilic substituent or (c) deleted, in which case the e-amino group of LysB29 has a lipophilic substituent; and any Zn2+ complexes thereof, provided that when Xaa at position B30 is Thr or Ala, Xaa at positions A21 and B3 are both Asn, and Xaa at position B1 is Phe, then the insulin derivative is a Zn2+ complex."
[page 4 line 11 to page 5 line 19]
Consequently, there are three forms of the invention, characterised by the B29 and B30 positions. The key aspects of these forms of the invention are:
Form a):the B30 position is replaced by a non-codable, lipophilic amino acid;
Form b):the B29 position has a lipophilic substituent on the e-amino group, and the insulin is a zinc complex if the insulin is porcine or human; and
Form c):the B29 position has a lipophilic substituent on the e-amino group, and the B30 position is deleted.
The common feature found in all three forms of the invention is a lipophilic substituent attached to the LysB29 position. The substituent may either be attached to the e-amino group, or attached as the B30 group. The relationship of these insulins to natural insulins can be summarised by considering the key amino acid positions in the insulin structure. Table 1 below identifies the differences. All other positions are identical.
Table 1: Comparison of bovine, porcine and human insulins with the insulins of the present invention
Position Bovine Porcine Human Present invention A8 Ala Thr Thr Thr A10 Val Ile Ile Ile A21 Asn Asn Asn Any amino acid except Lys, Arg, Cys B1 Phe Phe Phe Phe or deleted B3 Asn Asn Asn Any amino acid except Lys, Arg, Cys B30 Ala Ala Thr Three different options are defined
The exemplified insulins can be summarised by having regard to the variable parts of the molecule:
Table 2: Summary of the structures of insulin derivatives exemplified in the specification
Example No A21 B1 B3 e-B29 B30 Zn? Form § 8 Asn Phe Asp Benzoyl Thr Yes b) 9 Asn Phe Asp Lithocholoyl Thr Yes b) 10 Asn Phe Asp Decanoyl Thr Yes b) 11 Asn Phe Asp H Des No * 13 Asn Phe Asp Decanoyl Des No c) 14 Asn Phe Asp Dodecanoyl Des No c) 15 Asn Phe Asp Myristoyl-a-glutamyl Thr No # 16 Asn Phe Asp Undecanoyl Des No c) 17 Asn Phe Asp Tridecanoyl Des No c) 18 Asn Phe Asp Myristoyl Des No c) 19 Asn Phe Asp Palmitoyl Des No c) 20 Asn Phe Asp Suberoyl-D-thyroxine Thr No # 21 Asn Phe Asp 2-succinylamido-myristic acid Thr No # 22 Asn Phe Asp Octyloxycarbonyl Thr No # 23 Asn Phe Asp 2-succinylamido-palmitic acid Thr No # 24 Asn Phe Asp 2-succinylamido-ethyloxypalmitic acid Thr No # 25 Asn Phe Asp Lithocholoyl-a-glutamyl Des No c) 26 Asn Phe Asp 3,3',5,5'-tetraiodo-thyroacetyl Thr No # 27 Asn Phe Asp L-thyroxyl Thr No # § This column identifies whether the compound belongs to the a), b) or c) form of the invention as claimed
* Not a form c) compound because there is no e-substituent.
# Not a form b) compound because there is no zinc complex
Clearly there is no exemplification of the a) form of the invention. There are several examples of the b) and c) forms of the invention.
The mechanism by which the compounds act is not explained in the specification. Dr Widmer seems to believe that the lipophilic groups increase the binding to albumins (second declaration at paragraph 8). Professor Hearn thought that the addition of a lipophilic group is such a substantial modification of the structure that it is impossible to predict its effect (second declaration at paragraph 23). Professor Colagiuri is unsure of the exact mechanism by which the compounds achieve a protracted action (first declaration at paragraph 21). The evidence as a whole suggests that it is likely that the lipophilic group enhances binding to albumins, and this may be important in protracting the action of the insulin. However, this is not a certain conclusion, and this is definitely not disclosed in the specification.
There was also disagreement about the meaning of certain terms used in the specification. My opinion is as follows.
Protraction
It is clear from the specification that it is concerned with producing insulins with a protracted profile of action. Various protracted insulin compositions are part of the admitted prior art. It seems that there is no simple definition of a protracted action. The specification refers to a prior art composition having an intermediate protraction. Thus it is clear that there are degrees of protraction. However, it seems that the invention lies in achieving an insulin that has a protracted action compared to normal insulin, no matter what the degree of protraction.
Soluble
The normal meaning of soluble is that the material dissolves in a specified solvent. However, solubility is seldom an all or nothing phenomenon. Most substances are soluble to some extent, and it is possible to determine the degree of solubility of a compound, for instance in grams per litre. Given the fact that solubility is not an absolute event, how would the term be understood by the person skilled in the art?
Given the context of the specification, I believe that the compounds must be completely soluble in the concentrations that they are likely to be used as pharmaceuticals, and must remain in solution upon injection, i.e. under physiological conditions. The solubility is a property of the insulin, rather than as a result of the use of solubilising additives.
Insulin
The compounds of the application are insulin derivatives. An insulin derivative is a compound derived from insulin by modifying the structure. The question is whether the derivatives must retain the natural activity of insulin. While a compound is technically a derivative if it is derived from insulin, as a practical matter it is not reasonable to regard a compound as an insulin derivative unless it retains the activity of insulin.
Effect of zinc on solubility
At the hearing there was a considerable discussion of the effect of zinc on the solubility of the insulin derivatives. Mr McGowan contended that the solubility of the compounds increased with increasing amounts of added zinc, which is the reverse of the situation with insulin itself. This is a most unexpected property, but it is not something that the specification indicates is a problem addressed by the specification. Mr McGowan agreed that this property is more properly characterised as a bonus effect.
The claims
The claims of the specification are directed to a number of aspects of the invention.
Claims 1 to 17 define an insulin derivative.
Claim 18 defines a method of preparing the insulin derivatives.
Claims 19 to 20 define a pharmaceutical composition of the insulin derivative.
Claim 21 to 22 define a method of treating diabetes using the insulin derivative.
The claims that are in dispute are claims 1 to 22, i.e. all of the claims. Claim 1 is the key claim, and this claim is reproduced in full in Annex 1 of this decision. It is apparent that claim 1 is the same as the general statement in the body of the specification. The claims encompass derivatives of human and porcine insulin. However, in these cases the insulin derivative must be in the form of a zinc complex. The three forms is a convenient way to consider the scope of the claim:
Form a):the B30 position is replaced by a non-codable, lipophilic amino acid
Form b):the B29 position has a lipophilic substituent on the e-amino group, and the insulin is a zinc complex if the insulin is porcine or human
Form c):the B29 position has a lipophilic substituent on the e-amino group, and the B30 position is deleted
SECTION 40 MATTERS
A number of section 40 matters were argued at the hearing. Many of these matters are interconnected, or are alternative ways of looking at a single issue. Not all of these matters were included in the particulars. As they were fully argued at the hearing I will deal with all of the issues. I will attempt to fully reflect the range of issues raised.
Fair basis
The test for fair basis is to ask whether there is a "real and reasonably clear disclosure" of the matter that is claimed (Rehm Pty Ltd v Websters Security Systems (International) Pty Ltd (1988) 81 ALR 79 at 95, Patent Gesellschaft AG v Saudi Livestock Transport and Trading Co (1997) 37 IPR 523 at 530). This test is explained in the Rehm case as:
"the question is whether there is a real and reasonably clear disclosure in the body of the specification of what is then claimed, so that the alleged invention as claimed is broadly, that is to say in a general sense, described in the body of the specification"
[page 95]
It was argued that the compounds that are claimed are a series of disconnect structures having nothing in common, and consequently it was speculative to claim them as a single invention. I do not agree with this argument. As previously stated, the common feature of the compounds is the attachment of a lipophilic group to the LysB29 position. It follows that the compounds of claim 1 are not a disconnected group of compounds, and there is a basis for extrapolating to the members of the group.
It was argued that the claims are speculative as there is no evidence that all of the class of insulin derivatives meet the promise of the invention. Another way of looking at this is that the claims extend into an unexplored field. While it is true that not all of the claimed compounds have been made and tested, there is no necessity for this. Fair basis is achieved if what is explicitly disclosed is a reasonable basis for extrapolating to the rest of the claimed subject matter. In general terms, it seems to me to be technically reasonable to extrapolate to the range of compounds claimed. However, there is a specific difficulty with the a) form of the invention, as discussed in the next point.
It was argued that the a) form of the invention is not fairly based, as there are no examples of a) form compounds. While there are no examples of such compounds, the description does refer to suitable B30 amino acids at pages 7 and 8 of the specification; for instance, a-amino decanoic acid, a-amino undecanoic acid, a-amino dodecanoic acid, a-amino tridecanoic acid, etc, are specifically named. I consider that there is a disclosure of the structures of these compounds. However, a disclosure of a structure without a means of preparing the compound (whether it be explicit or implicit) is not a real and reasonably clear disclosure of the compound (see my discussion in Biochem Pharma Inc v Emory University [1999] APO 50) - it is a prediction of a further compound that should possess the useful property, but the preparation of the compound is left as a puzzle for the reader. In truth, such a compound has not yet been disclosed.
The present specification does not include any general method of preparing a) form compounds, and no a) form compounds are actually prepared (as noted by Dr Jeffrey at paragraph 15 of his first declaration). Consequently, there is no explicit disclosure of a method for preparing a) form compounds. Is there an implicit method disclosed? Dr Jeffrey declared that he believed a) form compounds could be synthesised (paragraph 26 of his first declaration), but is silent on how, or whether this was implicit in the specification. I attach little weight to this assertion. Professor Hearn declared that he would have no difficulty preparing these compounds using the information in the specification. Professor Hearn gives detailed information to support his opinion:
"I would have no difficulty following the instructions set out in the passages and adopting [presumably this should be adapting] the procedures to produce an insulin derivative where the amino acid at position B30 is a non-codable lipophilic amino acid having 10-24 carbon atoms and the e-amino acid of B29 lysine is substituted with an acyl group of a carboxylic acid of up to 5 carbon atoms. I could make such insulin derivatives without undue experimentation. For example, starting with the (des B30) insulin as described on page 22 of 682061 or the e acyl Lys B29 des B30 insulins as described on pages 45-49, and modification of the a-amino groups using the di-tert-butyl dicarbonate reagent as described also on page 22, the introduction of the non-codable amino acid at position B30, as a 10-24 carbon atom derivative, could be achieved using solution phase enzymatic semi-synthesis, by procedures analogous to those described in publications by T Oka et al. And a paper by K Morihara et al or alternatively by procedures analogous to those employed to elongate (desB30)insulin to (X-B31)insulin, where X is a non-codable amino acid residue, as described in a publication by C Birr et al., a paper by C Niu et al. and a publication by T Weiland and M Bodanszky."
[paragraph 24 of the first Hearn declaration]
I find Professor Hearn's opinion to be very helpful. The conversion of c) form compounds to a) form compounds is a sensible method. Unfortunately, the attachment of a non-codable, lipophilic amino acid to the B chain of a c) form compound would not produce an a) form compound, because the B29 e-substituent would be a lipophilic group rather than an acyl of up to 5 carbon atoms. Taken at face value, Professor Hearn's evidence might suggest that there is in fact no implicit method of preparation. However, had Professor Hearn carefully considered the structures of the compounds I am sure he would have recognised that only a small alteration is needed - c) form compounds with an acyl group of up to 5 carbon atoms rather than a lipophilic substituent would need to be prepared. These compounds could probably be prepared by a simple alteration of the methods used to prepare the c) form compounds. Thus, while I do not consider it appropriate for me to rewrite Professor Hearn's declaration, I think it is clearly appropriate to extrapolate his comments to include the minor modification I have suggested. Since Professor Hearn has detailed the reasons for his opinion I can appreciate how he reached his conclusion, and I attach considerable weight to his evidence.
I consider that there is an implicit disclosure of a method of preparing the a) form of the invention. Consequently, the claims to this form of the invention are fairly based.
The compounds as claimed are not defined as having a protracted profile of action , and are not defined as being soluble at physiological pH. It is true that the compounds of claim 1 do not have this qualification. It seems to me that the claims are directed to a group of compounds that are alleged to have the desired properties of solubility and protracted action. In this sense, these are inherent features of the claims, so their absence is not a problem.
The term "lipophilic substituent" as used in several of the claims is said to be overly broad. Clearly the term is very broad, and encompasses a vast range of organic groups. However, broad terms are not automatically lacking in fair basis. The question is whether there is a disclosure that is sufficiently broad to justify such claiming. In the present case there is a disclosure of a range of lipophilic groups, and I believe that this represents a real and reasonable clear disclosure of lipophilic substituents generally.
Claim 9 is alleged to be speculative due to the use of the expression "acyl group derived from a carboxylic group having at least 6 carbon atoms". For the same reasons as given for the term "lipophilic substituent", this terminology is fairly based.
Claim 10 is alleged to be speculative due to the use of the expression "acyl group, which may be branched, comprises a main chain of carbon atoms 8-24 atoms long". For the same reasons as given for the term "lipophilic substituent", this terminology is fairly based.
Claim 11 is alleged to be speculative due to the use of the expression "acyl group of a fatty acid having at least 6 carbon atoms". For the same reasons as given for the term "lipophilic substituent", this terminology is fairly based.
Does not define the invention
It was argued that claim 17 does not define the invention. Claim 17 is drafted in the form of an omnibus claim:
"An insulin derivative, substantially as hereinbefore described with reference to any one of the Examples."
Some of the examples are directed to compounds that are outside the scope of the invention due to the proviso relating to zinc. If claim 17 includes these compounds, then the claim does not define the invention.
An omnibus claim in the form of claim 17 is construed as being limited to the essential features of the invention described in the specification. It is clear that the invention described in the specification excludes those compounds covered by the proviso. Thus claim 17 does not include compounds covered by the proviso.
It was argued that the claims do not define the invention because the claims include insulin derivatives that are poorly soluble. The evidence of the low solubility is in the declaration by Dr Markussen. Dr Markussen describes experiments to determine the solubility of various compounds. Dr Jeffrey took the view that the technique for measuring solubility was not appropriate (paragraphs 5 and 6 of the second Jeffrey declaration and paragraph 3 of the third Jeffrey declaration). I do not think that this is a matter that needs to be resolved, and I am prepared to accept the data at face value.
The interesting piece of information in paragraph 6 of Dr Markussen's declaration is the solubility of NeB29-tetradecanoyl des(B30) human insulin, zinc free. The solubility of this compound is quite low compared to the other tested compounds. Interestingly, the solubility is much less than that of NaB1-tetradecanoyl des(B30) human insulin, which is not a compound of the invention. The solubility is greatly increased by the addition of zinc, which seems to be the point of the table. However, it is clear that this compound is less soluble than other compounds of the invention, and much less soluble than a comparison compound. Dr Jeffrey notes the poor solubility of this compound in paragraph 9 of his second declaration.
Dr Markussen also provides data for the solubility of GlyA21,NeB29-tetradecanoyl des(B30) human insulin, zinc free. This compound has the lowest solubility of the compounds in the table (paragraph 9), including the equivalent compound where the B30 amino acid of human insulin is present.
Both of these compounds are c) form compounds. The compound NeB29-tetradecanoyl des(B30) human insulin, zinc free clearly has poor solubility compared to the other compounds, and the same is true of the compound GlyA21,NeB29-tetradecanoyl des(B30) human insulin, zinc free. The key question is whether these compounds can still be regarded as soluble. I note that both compounds were subjected to a number of tests by Dr Markussen. I can find no statement whether they formed solutions or suspensions. I am unable to answer the key question of whether the compounds are soluble for practical purposes. The evidence available does not satisfy me that these compounds are clearly outside the scope of the invention.
Full description
The requirement for full description is a requirement to disclose the invention to a technical person.
"The specification contains a full description if it makes the nature of the invention plain to persons having reasonably competent knowledge of the subject and also makes it plain, to persons having reasonable skill, how to perform the invention"
[Patent Gesellschaft case at 530]
Dr Jeffrey stated (in paragraph 12 of his first declaration) that he considered that there is no evidence in the specification that the derivatives in which the B30 is replaced by a non-codable amino acid have a protracted profile of action and are soluble at physiological pH. This is a consequence of the fact that these particular derivatives have not been exemplified (as noted above). As I noted previously, there is no explicit method of preparation of the a) form compounds, but there is an implicit method of preparation. It is made clear that the a) form compounds should possess the same properties, and it is possible to deduce the probable basis for this prediction (the presence of a lipophilic substituent in the form of the B30 amino acid). Consequently there is a full description in relation to the a) form compounds.
Dr Jeffrey also pointed out (paragraph 13 of his first declaration) that the zinc free derivatives in the description include compounds that are excluded by the proviso. This is not really a problem, because the zinc derivatives are prepared from the zinc free form by the addition of zinc. Consequently, the zinc free form is a necessary precursor to the preparation of the zinc complex.
It was argued that as some of the compounds within the scope of the invention do not work (in that they do not have all the properties desired), there is insufficient direction to the skilled addressee across the scope of the claims. As discussed above (under the heading of "Does not define the invention") there are two compounds that do not possess high solubility. The fact that there are two compounds that are subsequently shown to be beyond the promise of the invention does not of itself cause the description to fail to fully describe the invention. The specification provides sufficient information to enable the compounds to be prepared.
It was stated that the specification does not provide a definition of insulin derivative, protracted profile of action and soluble at physiological pH. I have previously discussed these terms, and decided that it is possible to understand these terms.
The particulars state that the specification is inconsistent because not all of the compounds have a lipophilic group bound to the e-amino of the LysB29. As already noted, this is not an issue since all the compounds have a lipophilic group attached to the LysB29, either on the e-amino or the acid.
The particulars state that the specification use the term "about" in connection with the pH and solution concentrations. Such terms are acceptable if their meaning would be understood by persons skilled in the art. It is not uncommon to use the term "about" in technical descriptions, and there is no evidence that it causes the nature of the invention to be uncertain in the present case.
Conclusion
All of the section 40 grounds are unsuccessful.
NOVELTY
Lilly asserted that the invention claimed in all claims is not novel in the light of the following documents:
JP 1-254699
WO 92/01476
Lindsay, D.G. and Shall, S., Biochem J 121: 737-745, 1971
Hashimoto, M. et al, “Synthesis of Palmitoyl Derivatives of Insulin and their biological activities”, Pharm Res 6(2): 171-176, 1989
Muranishi, S., et al, “Trials of lipid modification of peptide hormones for intestinal delivery”, J Controlled Release 19: 179-188 (1992)
Brange, J., “Galenics of Insulin” (Springer-Verlag), 1987 (particularly pages 18-29, 35-38, 69)
Brange, J. and Langkjaer, L., “Chemical Stability of Insulin 3. Influence of excipients, formulations, and pH” Acta Pharm Nord 4: 149-158, 1992 (particularly pages 152-153)
Blundell, T. et al, “Insulin: The structure in the crystal and its reflection in chemistry and biology” in “Advances in Protein Chemistry” 26: 279-402 (Academic Press), 1972 (particularly pages 296-297, 323-328, 337-345 and 365)
MIMS Annual 1991, Section 6d “Insulin preparations” (IMS Publishing) (pages 290-292)
MIMS Annual 1993, Section 6d “Insulin preparations” (IMS Publishing) (pages 349-351)
Prescription Products Guide 1992 (pages 942-944, 957-958, 1307-1308)
Schlichtkrull, J. in "Insulin Crystals" (Ejnar Munksgaard, Copenhagen), 1958
"Joslin's Diabetes Mellitus", 12th Edition, Marble, A. et al, eds (Lea & Febiger, Philadelphia) 1958 (see particularly Table 19-2, page 382)
Schade, D.S., et al, "Intensive Insulin Therapy" (Exerpta Medica Publishing, Amsterdam) 1983 (see particularly pages 7 and 304)
"Principles of Medicinal Chemistry", Foye, W.O., ed (Lea & Febiger, Philadelphia) 1974 (see particularly page 564)
"Wilson and Gisvold's Textbook of Organic Medicinal and Pharmaceutical Chemistry", 8th Edition, Doerge, R.F. ed (J.B.Lippincott, Philadelphia) 1982 (see particularly pages 774-776)
Montague, W., "Diabetes and the Endocrine Pancreas - A Biochemical Approach" (Oxford Press, New York) 1983
It is well established that the general test for anticipation is the reverse infringement test. The classic formulation of this test is that given by Aickin J in Meyers Taylor Pty Ltd v Vicarr Industries Ltd (1977) 137 CLR 228, at page 235:
"The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged anticipation would, if the patent were valid, constitute an infringement"
This test is satisfied if the alleged anticipation discloses all the essential features of the invention as claimed (see Nicaro Holdings Pty Ltd v Martin Engineering Co (1990) 91 ALR 513 at page 517; 16 IPR 545 at page 549). A citation is part of the prior art base for the purposes of novelty if it was published before the priority date of the claim (see definition of "prior art base" in Schedule 1 of the Act).
It is relevant to have regard to the recent decision of the Federal Court in Bristol-Myers Squibb Co v F H Faulding & Co Ltd (2000) 46 IPR 553. The Court noted that in the case of a paper anticipation, the reverse infringement test cannot be applied literally. This is because the infringement arises because someone hypothetically does what is suggested by the document. After reviewing the traditional authorities, the majority concluded (at page 576):
"What all those authorities contemplate, in our view, is that a prior publication, if it is to destroy novelty, must give a direction or make a recommendation or suggestion which will result, if the skilled reader follows it, in the claimed invention. A direction, recommendation or suggestion may often, of course, be implicit in what is described and commonly the only question may be whether the publication describes with sufficient clarity the claimed invention or, in the case of a combination, each integer of it."
The Court recognised that not everything published in a document is an anticipation. For instance, information that is a speculation or a proposal for a trial would not be an anticipation. The judges drew a distinction between a teaching of useful things in relation to a subject matter, and a teaching of the invention as claimed.
When a citation has a disclosure in generic form (as is common in chemical patents), the question arises as to which compounds within the scope of the generic disclosure are taught to the reader. I have considered this question in a number of earlier decisions. (For instance, see Application by American Home Products Corporation (1994) APO 58, Pharmacia Aktiebolag v Ueno Fine Chemicals Industry Ltd (1995) 34 IPR 445) It is not enough to find that a specific compound is within the scope of the broadest disclosure of the citation (which can be considered as the intellectual content of the disclosure). It is necessary to find that a reader would have understood that the specific compound was part of the technical information of the specification, and there must be an enabling disclosure of that compound (for instance, see my analysis of the doctrine of enabling disclosure in Biochem Pharma Inc v Emory University [1999] APO 50).
A further point is that an arbitrary limitation which is not related to the performance of the invention does not confer novelty (Re Application by National Research Development Corporation (1992) 24 IPR 123 at 128). Presumably, such an integer should not be treated as an essential feature.
It follows that there are three questions that need to be considered:
(i)does the citation embrace the relevant compounds in a generic sense (i.e. are they within the intellectual content);
(ii)is there a specific disclosure of the relevant compounds (i.e. are they within the technical content); and
(iii)is there an enabling disclosure of the relevant compounds.
In the present case the relevant compounds have a LysB29-lipophilic substituent.
Essential features
It is well established that the initial assumption is that all features of an independent claims are essential: Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183. A feature can be regarded as not essential if it does not appear to make any difference to the inventive concept (but noting the caution in Société Technique de Pulverisation Step v Emson Europe Ltd [1993] RPC 513 at 522). I have previously applied these principles to the situation of a compound defined by a structural formula in decision such as Application by American Home Products Corporation [1994] APO 58, and Nippon Kayaku Kabushiki Kaisha and Sankyo Co Ltd v Rohm and Haas Co [1997] APO 40. The conclusion that I reached in the Nippon decision was:
"If it is argued that a particular group in a structure is not essential to the activity of the compound, it would be necessary to demonstrate that the activity is present when that group is absent (i.e. replaced by hydrogen), and when the group is replaced by different groups, for example, groups that are neither isosteric nor isoelectronic."
[page 8]
This is consistent with the approach taken by the House of Lords in Beecham Group Ltd v Bristol Laboratories Ltd [1978] RPC 153. That case involved a derivative of a known compound, in which the derivatised portion is converted to the parent compound in normal use. The derivative carries out its normal activity as the parent compound, and possesses no advantage over the parent compound. The derivatised portion was effectively found to be a non-essential feature.
The present claim is in the generic form that is common in chemical patents. The generic formula defines a large number of compounds that are alternative forms of the invention. The initial assumption is that the structure of each of those compounds is an essential feature of that alternative.
It was argued that zinc is an optional feature, because it is not essential to all of the compounds. However, essential features are not determined in that way. It is inherent in the nature of a generic formula there are some groups that are not common to all of the compounds. It was suggested that the zinc is just doing its normal thing.
The way to determine whether any particular part of a structure is to use the approach that I recommended in the Nippon decision: does the particular moiety contribute to the activity of the compound. There is an argument that the zinc is not an essential feature of the compounds. This argument can be summarised by noting that most of the claimed compounds may be either zinc free or zinc complexes, which suggests that zinc makes no difference to activity. Clearly for those compounds the zinc is not essential. However, some of the b) form compounds are claimed as the zinc complexes only. It might be expected that these b) form compounds would behave in the same way as the other claimed compounds, but there is no clear data about whether these compounds have the same properties in the zinc free form. Thus there is a basis for predicting that zinc might not essential. However, it is not appropriate to classify a feature as non-essential unless I am satisfied that it does not contribute to the invention. I am not satisfied that it has been established that the zinc does not contribute to the activity of the compounds. I regard the zinc as an essential feature of those compounds that contain it.
There were suggestions that the other parts of the structure might not be important to the activity of the compounds. For similar reasons, I believe that a sufficient case has not been made out, so I regard all other parts of the structure as essential.
All structural features are essential features of the claim. The common element of the three forms of the invention is the lipophilic substituent on the LysB29 position, either on the e-amino group or in the B30 position. Thus the key feature in analysing the citations is the lipophilic substituent attached to the B29 position (and only the B29 position). The issue of the zinc complexes does need to be considered because the novelty of the primary structure is sufficient to resolve the matter.
JP 1-254699
My discussion of this document is based on the English language translation. Claim 1 of this document gives a feel for the nature of the disclosure:
"1. Insulin with a fatty acid bound to the amino group of the amino acid at the B1 or B29 position of the B chain of insulin."
These derivatives are lipid soluble without loss of insulin activity:
"As the result of our studies, we discovered that fatty acid bound to insulin yielded lipid-soluble insulin without loss of insulin activity, so that the insulin with fatty acids bound to it is effective hypoglycemic. The invention is the result of this discovery."
[page 5]"The insulins with bound fatty acids of this invention are prepared by bonding fatty acids to either one or both of the amino groups at the B1 and B29 position of the B chain of insulin, as shown in general formula (I).
[page 6]
The examples of the invention demonstrate the preparation of a palmitoyl derivative of bovine insulin. The material is separated by HPLC to give Lys-B29-palmitoyl-insulin as an isolated compound. This compound is designated pal-1. A graph of the changes induced in the blood glucose of male rats following administration of the pal-1 (and other compounds) is shown in Figure 2 of the document. It was stated that the shape of the graph demonstrated a protracted action.
Some doubt is cast on whether this document teaches a LysB29 substituted insulin by another document in evidence: an article in Pharmaceutical Research in 1989 by Hashimoto et al. Two of the authors of this article are listed as inventors of the Japanese application. This document seems to relate to the same work, and has a graph identical to Figure 2 of the patent application. In this article, the substance identified as pal-1 is called Palins-1, which is an abbreviation for B1-mono-palmitoyl-insulin. LysB29-mono-palmitoyl-insulin is apparently not produced according to this paper.
The most reasonable interpretation to place on these facts is that the patent application is in error in the assignment of a structure to pal-1, and in fact the LysB29 derivative is not produced. I conclude that the Japanese application does not disclose a LysB29 derivative, or if it does then there is no enabling disclosure.
Thus the invention is novel in the light of this document.
WO 92/01476
This patent application is entitled "Novel drug delivery systems for proteins and peptides using albumin as a carrier molecule". The broad thrust of the application is given on page 4 as:
"A new drug delivery system for peptides and proteins wherein such molecules are modified with apolar substituents to promote noncovalent binding to albumin, which in turn serves as a macromolecular carrier for the drug within the circulation. In the most preferred embodiment of the invention the protein is modified at a single or limited number of sites with a long chain fatty acid derivative."
The application does not exemplify any substituted insulins, and does not mention LysB29 substitution or replacement of the B30 position. The citation does not disclose any compounds of the forms claimed by the present application. Thus the invention is novel in the light of this document.
Lindsay, D.G. and Shall, S., Biochem J 121: 737-745, 1971
This document is concerned with acetylation of insulin. The insulin that is used appears to be porcine insulin and ox insulin (page 737). The article discloses LysB29-mono-acetyl- bovine insulin at page 739. There is also a reference to LysB29-thiazolidine-bovine insulin in Table 2. Otherwise, I can find no reference to LysB29-monosubstituted insulins. There are no insulins with replacement of the B30 position. Thus the invention is novel in the light of this document.
Hashimoto, M. et al, “Synthesis of Palmitoyl Derivatives of Insulin and their biological activities”, Pharm Res 6(2): 171-176, 1989
This article appears to be a fuller report of the work in the Japanese patent, discusses above. Bovine insulin is derivatised with a palmitoyl moiety, but there is no reference to a LysB29-mono-substituted insulin or replacement of the B30 position. Thus the invention is novel in the light of this document.
Muranishi, S., et al, “Trials of lipid modification of peptide hormones for intestinal delivery”, J Controlled Release 19: 179-188 (1992)
This article relates to modified hormones. Derivatives of bovine insulin are specifically mentioned, and the B1-monopalmitoyl and B1,B29-dipalmitoyl derivatives are prepared. There is no reference to a LysB29-monosubstituted insulin or replacement of the B30 position. Thus the invention is novel in the light of this document.
Brange, J., “Galenics of Insulin” (Springer-Verlag), 1987 (particularly pages 18-29, 35-38, 69)
This book contains a wealth of information about insulin. The specific pages referred to relate to the pharmaceutical chemistry of insulin, protamine complexes and zinc complexes. On page 69 there is a short discussion of insulin derivatives. There is a discussion of a phenylcarbamoyl insulin, and desalanine(B30) porcine insulin. There is no discussion of LysB29-monosubstituted insulin or replacement of the B30 position. Thus the invention is novel in the light of this document.
The other 11 documents
The other documents describe natural insulins and commercial insulin compositions. They do not disclose any LysB29-monosubstituted insulin or replacement of the B30 position. Thus the invention is novel in the light of these documents.
Conclusion on novelty
It is clear that none of the documents cited in the opposition disclose insulin derivatives of the type covered by the present application. Consequently, all claims are novel in the light of all of the citations.
The further evidence document, US 3,528,960
This document relates to modified insulins. The invention is broadly described by reference to a formula that shows an insulin modified by the attachment of 1.5 to 2.5 carboxyaroyl groups. This formula could include compounds of the present application. The specific example that is relevant is Example IV, which reads:
"EXAMPLE IV
g-Carboxy-b-thiophenoyl insulin
Using the procedure of Example IV, 130 mg of 3,4-thiophendicarboxylic acid and 217 mg of 1-cyclohexyl-3-[2-morpholinyl-(4)-ethyl]carbodiimide metho-toluenesulfonate were reacted with 120 mg of insulin at pH 7 for 3 hours and 30 minutes giving 133 mg of g-carboxy-b-thiophenoyl insulin."
[column 4]
It is very uncertain what this example is teaching. First, it is unclear how many substituents are attached to the insulin by this procedure. The product is probably a mixture of the mono, bis and tris, but the large excess of acylating agent suggests that the tris is likely to predominate. Second, if there is any mono-acylation, it is unclear where the substituent is attached. Given the uncontrolled manner in which the reaction is carried out, it is likely that any mono substituted product is acylated at all available positions. Finally, it is uncertain whether a carboxythiophen can be regarded as a lipophilic substituent. I am satisfied that there is no clear teaching of an insulin derivative according to the present invention.
INVENTIVE STEP
It was alleged that the invention as claimed in all of the claims is lacking in inventive step in the light of the documents discussed above under the heading of novelty. Each of these documents will be discussed in turn below. It was also argued that the invention is obvious in the light of the common general knowledge alone. I will discuss the common general knowledge separately below. It was also argued that various admitted prior art documents in the specification are common general knowledge (based on the views expressed in Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 46 IPR 553).
The relevant law
Section 7(2) provides a definition of inventive step for the purposes of the Patents Act.
"(2) For the purposes of this Act, an invention is to be taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed in the patent area before the priority date of the relevant claim, whether that knowledge is considered separately or together with either of the kinds of information mentioned in subsection (3), each of which must be considered separately.
(3) For the purposes of subsection (2), the kinds of information are:(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 a person skilled in the relevant art in the patent area would treat them as a single source of that information;
being information that the skilled person mentioned in subsection (2) could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood and regarded as relevant to work in the relevant art in the patent area."
[The prior art base is defined in Schedule 1 of the Act.]
It is clear that inventive step is a matter that is presumed, unless it is demonstrated that the invention is obvious. The assessment of obviousness can be made against the common general knowledge alone, or the common general knowledge together with a document (or act) of the type covered by section 7(3). Section 7(3) documents must satisfy several requirements: the document must be publicly available inside or outside Australia (see the definition of "prior art base"); and the document would have reasonably been expected to have been ascertained, understood and regarded as relevant.
The normal approach to obviousness is the problem-solution approach: for instance, see Rhone-Poulenc Rorer S.A.'s Application [1995] APO 50. I note in passing that the problem-solution concept is evident in judicial decisions such as Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd (1980) 144 CLR 253 (e.g. at 298: "this solution to the known problem") and Winner v Ammar Holdings Pty Ltd (1993) 113 ALR 63 (e.g. at 67: "The problem and the solution were readily apparent"). Once the problem has been formulated, and the common general knowledge or prior art base has been determined, the question of whether the claimed solution is obvious must be addressed. The test for obviousness is whether it would have been a matter of routine to proceed to the claimed invention.
"It is still correct to say that a valid patent may be obtained for something stumbled upon by accident, remembered from a dream or imported from abroad, if it otherwise satisfies the requirements of the legislation. What is important is that the patent itself should involve an inventive step, whether or not it was consciously taken by the patentee and whether or not it appeared obvious to the patentee himself. The test is whether the hypothetical addressee faced with the same problem would have taken as a matter of routine whatever steps might have led from the prior art to the invention, whether they be the steps of the inventor or not."
[Wellcome Foundation Ltd v V.R. Laboratories (Aust.) Pty Ltd (1981) 148 CLR 262 286]
The question of what would be routine can be considered using the "obvious to try" approach. This approach is well explained in Beecham Group Ltd's (Amoxycillin) Application [1980] RPC 261 (which has been approved in Coopers Animal Health Australia Ltd v Western Stock Distributors Ltd (1986) 67 ALR 390 at 410 and W R Grace & Co v Asahi Kasei Kogyo Kabushiki Kaisha (1993) 25 IPR 481 at 492 - 494) at 290 - 291:
"It is clearly established that, for a particular step or process to be obvious for the purpose of either section, it is not necessary to establish that its success is clearly predictable. It will suffice if it is shown that it would appear to anyone skilled in the art but lacking in inventive capacity that to try the step or process would be worthwhile. Worthwhile to what end? It must, in my opinion, be shown to be worth trying in order to solve some recognised problem or meet some recognised need."
[citations omitted]
This approach suggests that an invention is obvious if it would have been considered well worth trying, with the likelihood of success being sufficient to warrant the actual trial. However, it must always be remembered that there is an unclear line between normal trial and error and an inventive step:
"However, the test of whether something was 'worth trying' involves questions of degree … If the expectation of success is sufficiently predictable, and the effort involved is not going to be very great, it may well be that there is no inventive step. On the other hand, if the expectation of ultimate success is doubtful and the effort involved is great, the person undertaking the work should be entitled to a monopoly. A patent monopoly is awarded, not to reward genius but to encourage the disclosure of information which is of value to the public in that it takes the store of knowledge ahead by the requisite 'inventive step'."
[ICI Chemicals & Polymers Ltd v Lubrizol Corp Inc (1999) 45 IPR 577 at 600-601, citations omitted]
It is worth noting that the presence of meaningless and arbitrary parameters is not sufficient to amount to an inventive step (see Nippon Steel Corporation v BHP Steel (JLA) Pty Ltd [1999] APO 70 and Raychem Corp.'s Patents [1998] RPC 31)
What was the problem
In the Rhone-Poulenc decision I decided that the problem can be formulated from a reading of the specification in the light of surrounding facts (see also my earlier decision in Biochem v Emory). In the present case, the specification provides a statement at page 3 that appears to encapsulate the problem:
"However, there still is a need for protracted injectable insulin compositions which are solutions and contain insulins which stay in solution after injection and possess minimal inflammatory and immunogenic properties.”
Mr McGowan suggested that the problem was to find insulins having protracted action by slowed diffusion across membranes. I am not sure it is necessary to insert the mechanism of action into the problem, so long as it is recognised that solubility and reduced inflammation at the site of administration is part of the problem.
I conclude that a fair formulation of the problem is that found in the specification:
To find a compound having protracted insulin activity that is soluble and remains in solution after injection, and has minimal inflammatory and immunogenic properties.
The common general knowledge
Common general knowledge was defined by Aickin J in Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd (1980) 144 CLR 253 at 292:
"The notion of common general knowledge itself involves the use of that which is known or used by those in the relevant trade. It forms the background knowledge and experience which is available to all in the trade in considering the making of new products, or the making of improvements in old, and it must be treated as being used by an individual as a general body of knowledge"
Dr Bennett alleged that the following matters were part of the common general knowledge:
1.the amino acid sequence and three dimensional structure of insulin
2.insulin forms dimeric and hexameric complexes upon addition of zinc
3.addition of excess zinc results in formation of crystallisation
4.zinc insulin crystals can be used in suspension formulations with a protracted duration of action
5.addition of zinc to insulin increases stability
6.the receptor binding surfaces of insulin comprise the amino acids
7.the amino acid LysB29 has a surface location in both monomeric and hexameric insulin forms
8.substantial modifications can be made to the C-terminal of the B chain without reducing biological activity
9.LysB29 provides a convenient site for attachment of various substituents
10.amino acids at positions A21, B1 and B3 can be varied without loss of biological activity
11.the amino acid at position B30 could be removed without loss of biological activity
12.reversible binding to albumin can decrease the clearance rate of lipophilic drugs and act as a reservoir of bound drug
13.adding a lipophilic substituent increases the lipophilicity of the insulin
It was also argued that various documents are part of the common general knowledge because they are part of the admitted prior art, or because they are routinely consulted in the art (paragraph 50 of the first Jeffrey declaration).
It is not necessary at this point to determine whether these matters were in fact common general knowledge.
Matters of routine
The critical question in the present opposition is whether it would have been a matter of routine to attach a lipophilic substituent to the e-amino acid of LysB29. As this issue is central to the obviousness of the invention, I will deal with it separately.
The argument is that it would have been obvious that a lipophilic substituent would have been an obvious way to produce a protracted insulin, since this would increase the binding of the insulin derivative to albumin. The e-position of the Lys B29 would have seemed a suitable position to attach a substituent (for instance, see paragraph 27 of the second Jeffrey declaration and paragraph 13 of the first Widmer declaration; although Professor Hearn did not think it was commonly known that this would produce a lipophilic insulin: paragraph 27 of his first declaration). One of the key points of disagreement is that binding to albumin can be used to produce a protracted drug. Dr Widmer declared that a LysB29 substitution was the obvious thing to do:
"I believe that had I been asked prior to 17 September 1993 to develop a human insulin with a prolonged profile of action, I would have considered the possibility of increasing binding of insulin to circulating albumin. To achieve that, I believe that I would have recognised that the simplest and most practical approach would be to attach to insulin a group which binds to albumin, such as a long chain group as in fatty acids, alkyl amines and alkyl alcohols. Further, in view of what was known of the structure of insulin (see, in this regard, the statements of Dr Jeffrey at paragraph 11 of his Statutory Declaration dated 18 January 2000), I would have also recognised that the e-amino group of LysB29 was a good, convenient candidate site for attachment of a group which binds to albumin. Thus, prior to 17 September 1993, it would have occurred to me that attaching a lipophilic group to the e-amino group of LysB29 would have been a clear initial approach to take in obtaining an insulin derivative with increased binding to albumin and a consequent prolonged profile of action."
[paragraph 9 of the second Widmer declaration]
Professor Hearn declared that decreasing the metabolism and/or excretion of insulin was unlikely to produce a clinically useful protraction (second Hearn declaration, paragraph 9). This seems to be saying that protraction might be expected, but it was unlikely that it would be clinically useful. This seems to be an argument about the degree of protraction that was possible, a matter that could only be determined by actually testing the particular compound. It was also argued that it was not common general knowledge that a LysB29 derivative would be lipophilic. At paragraph 25 of his first declaration, Professor Hearn goes through a number of documents that are supposed to support the allegation, and states that they do not support the conclusion of Dr Jeffrey. It is at best an arguable point.
Another issue is whether it would have been expected that attaching a lipophilic substituent would retain the solubility of the compound in water or a physiological solvent. I note that Professor Hearn declared that attaching a lipophilic substituent should make a compound less water soluble (paragraph 54 of the first Hearn declaration):
"Normally if you modify a molecule to make it less polar it is made less soluble in water. Thus modifying the molecule to make it more lipophilic will normally render the molecule more non-polar and thus less soluble in water or physiological buffers. For this reason, if the skilled person wanted to modify insulin to improve its solubility, they would not be motivated to increase its lipophilicity. Indeed, in the Hashimoto paper it is apparent that pal-2, which is more lipophilic than pal-1, was less soluble than pal-1. It is therefore surprising that the inventors of 682051 were able to obtain an insulin derivative that is soluble at physiological pH, yet is modified by the addition of a lipophilic moieties to the e amino group of lysine at position B29 or by exchange of the native amino acid at position B30 with a non-codable lipophilic amino acid."
Given that the problem includes the solubility of the insulin, it is doubtful whether an insulin substituted by a lipophilic substituent would have been expected to be soluble. The evidence does not clearly establish that it would have been routine to want to acylate the LysB29.
However, ignoring all these issues and assuming that it would have been obvious to wish to prepare a LysB29 derivative as a means to solve the problem, I am left with a more significant problem: would it have been a matter of routine to prepare a LysB29 derivative. This is quite different to the question of whether it would have been implicit from the specification how to prepare the a) form of the invention. The thirteen matters referred to by Dr Bennett do not refer to the routine preparation of LysB29 derivatives. Dr Jeffrey states (at paragraph 27 of his first declaration) that this would have been possible:
"I would have expected that such insulin derivatives could be synthesised and would retain biological activity"
Dr Jeffrey does not specify that the preparation would have been a matter of routine, or how he would have expected to prepare the derivative. Dr Widmer also declared along similar lines:
"There is no reason why there would be an inherent difficulty in producing such a modified insulin."
[paragraph 13 of the first Widmer declaration]
However, Dr Widmer is also silent on how to prepare a LysB29 derivative.
Professor Hearn's first declaration states (at paragraph 26) that substitution on the LysB29 amino acid is difficult. The evidence indicates that LysB29 substituted insulins are not well known compounds. The document JP 1-254699 refers to a B29-mono-palmitoyl-insulin. However, as discussed earlier, it is more likely that a LysB29 derivative is not disclosed. The document US 3528960 also refers to acylation, which could include LysB29-acylation. However, as discussed earlier, it is unlikely that B29-mono-acylation is disclosed. The document by Blundell et al refers to known derivatives of insulin. Included in Table IV (pages 343-344) are five B29-mono-substituted insulins: e-acetyl, e-DFTC (i.e. 2,2'-dimethyl-3-formyl-thiazolidine-4-carboxyl insulin), e-acetoacetyl, e-hemisuccinyl, [GlyB1]-acetylB29. All of this suggests to me that B29-mono-substitution is very uncommon. In contrast, the same Table lists 19 compounds where the LysB29 is substituted as well as one or two other positions. This suggests that selective derivatisation of the LysB29 position was probably not a matter of routine. I note that Professor Hearn refers to lipophilic insulins as "an exotic group of molecules" (paragraph 75 of the first Hearn declaration).
I consider that there is insufficient evidence to establish that the mono-substitution of the LysB29 amino acid would have been a matter of routine. Rather, it seems to have been one of the achievements of the present application to find a method of mono-substitution.
My conclusion is that the derivatisation of the LysB29 position with a lipophilic substituent would not have been an obvious solution to the problem, and that it would not have been a routine matter to selectively derivatise the e-amino group of the LysB29 position. Because of this, all of the attacks on obviousness will fail. I will explain this in more detail below.
The obviousness question
Dr Bennett suggested that this is a case like that in Monsanto Co v Zeneca Ltd (1996) 36 IPR 120, where I found that the problem automatically suggested an answer (at page 132). In the Monsanto decision the problem was "the need to produce a dry glyphosate composition that could take advantage of the desirable properties of liquid surfactants" (page 131). The solution that was automatically suggested by the problem was "incorporate a liquid surfactant into an otherwise dry glyphosate composition" (page 132). It is clear that this solution arises from the very nature of the problem, without regard to the common general knowledge. In the present case I do not believe that the problem as formulated suggests a solution. It is necessary to have regard to the common general knowledge or the prior art.
I will turn now to consider whether the invention is obvious in the light of the common general knowledge. The key problem is that even if I accept that the common general knowledge indicates the attachment of a lipophilic substituent to the B29 amino acid would have been worth trying (and this is by no means certain), there is insufficient evidence that it would have been a matter of routine to prepare such a compound. Consequently the attack on obviousness fails.
JP 1-254699
This patent application is entitled "Novel drug delivery systems for proteins and peptides using albumin as a carrier molecule". The broad thrust of the application is given on page 4 as:
"A new drug delivery system for peptides and proteins wherein such molecules are modified with apolar substituents to promote noncovalent binding to albumin, which in turn serves as a macromolecular carrier for the drug within the circulation. In the most preferred embodiment of the invention the protein is modified at a single or limited number of sites with a long chain fatty acid derivative."
This document is relevant to work in the art. The specification goes on to refer specifically to insulin at page 12:
"The present invention also affords a novel class of long acting insulin derivatives which bind to albumin by means of an attached apolar group. Competition with endogenous free fatty acids gives rise to particularly useful pharmacokinetic properties for these derivatives. Normally there is an excess of available binding sites on albumin so that the modified insulin is almost 100% in the bound form. In this state, the insulin is inactive. Regular insulin is given intermittently to maintain normal metabolic control. If the levels of insulin administered are inadequate, the concentration of free fatty acids in the blood increases. This displaces some of the insulin bound to albumin and ameliorates the insulin deficiency. Such apolar modified insulin derivatives are particularly useful in preventing the complications of diabetic ketoacidosis, in which case the free fatty acid concentration can be as high as 3 to 4 mM."
There are no examples of attaching a substituent to an insulin. However, it is clear that this document is suggesting that the attachment of a lipophilic substituent should increase the protraction through binding to albumin.
However, as noted previously, it is questionable whether such a derivative would have been expected to be soluble in physiological solvent. This is reinforced by the fact that the compounds are dissolved in 1N hydrochloric acid for testing in rats. The pH of 1N hydrochloric acid is 0.1, which is not a physiological pH. However, the pH of the solution after dissolving the insulin is not specified so it is not clear whether it is physiological, but the fact that it is used for injection into rats suggests that it is still physiologically acceptable. Dr Jeffrey declared that:
"it is common for insulin compounds to be dissolved at low pH. Further, as is acknowledged by Dr Hearn, the pH can subsequently be adjusted back to neutrality"
[paragraph 12 of his third declaration]
However, I note that some of the test compounds are only suspended in 1N hydrochloric acid, so they are not sufficiently soluble even under these conditions. Also, it has not been established that it would have been a matter of routine to prepare a mono-B29 substituted insulin.
Consequently, the b) and c) forms of the invention are not obvious in the light of this document. There is nothing in the document that suggests the a) form of the invention. The invention as claimed is not obvious in the light of this document.
WO 92/01476
This patent application is entitled "Novel drug delivery systems for proteins and peptides using albumin as a carrier molecule". The broad thrust of the application is given on page 4 as:
"A new drug delivery system for peptides and proteins wherein such molecules are modified with apolar substituents to promote noncovalent binding to albumin, which in turn serves as a macromolecular carrier for the drug within the circulation. In the most preferred embodiment of the invention the protein is modified at a single or limited number of sites with a long chain fatty acid derivative."
The application mentions that it is suitable for insulin, but does not exemplify any substituted insulins, and does not mention LysB29 substitution. The citation does not disclose any compounds of the type claimed by the present application. This document is relevant to work in the art.
It is clear that this document suggests a general method of drug delivery which would be expected to provide the drug over a longer period than normal. It is clearly stated that it is suitable for insulin. While it is uncertain whether the insulin derivative would be soluble in physiological solvent, there is probably sufficient incentive to investigate the compounds. What is less clear is whether a LysB29 derivative would be obvious. As mentioned earlier, the preparation of a LysB29 derivative is not a matter of routine. Consequently the b) and c) forms of the invention are not obvious in the light of this document. There is nothing in the document that suggests the a) form of the invention.
The invention as claimed is not obvious in the light of this document.
Lindsay, D.G. and Shall, S., Biochem J 121: 737-745, 1971
This document teaches that LysB29 acetylation of insulin did not affect the biological activity of insulin. I consider that this document might have been relevant to work in the art. However, it is not at all clear that it suggests a solution to the problem of protracted action. It is only with the benefit of hindsight that it is apparent that LysB29 acylation has any relevance to the present problem. Consequently the b) and c) forms of the invention are not obvious in the light of this document. There is nothing in the document that suggests the a) form of the invention.
The invention as claimed is not obvious in the light of this document.
Hashimoto, M. et al, “Synthesis of Palmitoyl Derivatives of Insulin and their biological activities”, Pharm Res 6(2): 171-176, 1989
This document teaches that the lipophilicity of insulin can be increased by palmitoylation. Increased lipophilicity is expected to improve the transport of the insulin across membrane barriers. The test samples (which do not include a mono-B29-palmitoyl insulin) were dissolved in 1N hydrochloric acid for biological testing by injection into rats.
I consider that this document is of a type that would have been considered relevant to the work in the art. There is reason to regard this document as suggesting a solution to the problem of protracted action.
I note that the example involves dissolving the material in 1N hydrochloric acid. It was suggested that the derivative is either not soluble, or not soluble in a physiologically acceptable solvent. The pH of 1N hydrochloric acid is 0.1. As previously noted, it is not apparent what pH the resulting solution would have, but the fact that the solution was used for injection into rats suggests that it is a physiologically acceptable composition.
It is arguable that this document discloses soluble insulin derivatives. There is some disagreement as to whether the derivatives have a protracted action. The graph at Fig 5 shows an profile that can be regarded as protracted, even if it is not a good protraction. The compounds that demonstrate this activity are two palmitoyl derivatives of insulin: the B1-monopalmitoyl and the B1, B29-dipalmitoyl. There is nothing that directly suggests the B29-monopalmitoyl compound. It could be that it is a reasonable extrapolation from the two exemplified compounds to predict that the B29-monopalmitoyl would be worth testing. If I accept this proposition, it has not been established that it would have been a matter of routine to prepare the B29-monopalmitoyl compound. Consequently, it is not obvious to prepare this LysB29 derivative. There is no suggestion that other LysB29 derivatives should be investigated. Consequently the b) and c) forms of the invention are not obvious in the light of this document. There is nothing in the document that suggests the a) form of the invention.
The invention as claimed is not obvious in the light of this document.
Muranishi, S., et al, “Trials of lipid modification of peptide hormones for intestinal delivery”, J Controlled Release 19: 179-188 (1992)
This document teaches that the lipophilicity of insulin can be increased by acylation. It is speculated that this will improve intestinal delivery characteristics. It is noted on page 183-184 that:
"HCO-60 (nonionic surfactant, polyoxygenated hydrogenated caster oil) was used for solubilizing insulin analogues which were poorly soluble in water"
This clearly states that the type of derivatives disclosed are not in themselves soluble in water. This document is relevant to work in the art. However, the indication that the compounds require a surfactant to solubilise them is a concern. In my view, you would need to test the compounds to be sure about their solubility, but there are reasons for low optimism.
This paper relates to the same compounds as the Hashimoto document. There is nothing in the document that directly suggests the B29-monopalmitoyl derivative. It could be that it is a reasonable extrapolation from the two exemplified compounds to predict that the B29-monopalmitoyl would be worth testing. If I accept this proposition, it has not been established that it would have been a matter of routine to prepare the B29-monopalmitoyl compound. Consequently, it is not obvious to prepare this LysB29 derivative. There is no suggestion that other LysB29 derivatives should be investigated. Consequently the b) and c) forms of the invention are not obvious in the light of this document. There is nothing in the document that suggests the a) form of the invention.
The invention as claimed is not obvious in the light of this document.
Brange, J., “Galenics of Insulin” (Springer-Verlag), 1987 (particularly pages 18-29, 35-38, 69)
This document refers to the solubility of insulin at pages 18-20:
"In water insulin is practically insoluble at its isoelectric point of pH 5.4, but it is easily soluble at pH lower than 4. At neutral and alkaline reaction the solubility is strongly dependent on the concentration of zinc ions and on the species of insulin. Jeffrey et al found the solubility of zinc-free bovine insulin at pH 7.0 to be 4.5-5 g/l, corresponding to approximately 125 IU/ml. Milthorpe et al reported solubility of bovine zinc insulin crystals (2 Zn/hexamer) at pH 7.0 to be 0.9 g/l, corresponding to 24 IU/ml." [page 18]
At page 23 it is stated that:
"The porcine zinc-insulin complexes are soluble at neutral reaction when the zinc/insulin (hexameric) ratio is below 4, but at higher concentrations of zinc the solubility decreases and at a ratio of 6 Zn2+/hexamer complete precipitation of the insulin is observed."
Prolongation of insulin is discussed at pages 27-29. Insulin derivatives discussed in this context are the phenylisocyanate (also mentioned at page 69). The document teaches that insulin has a low solubility, and that the zinc complex is less soluble. There is a statement that some derivatives have a prolonged action. Clearly this document is relevant to work in the art. However, there is no evidence that the leap to the specific derivatives in the present application is a matter of routine. There is nothing in the document itself that suggests derivatives even close to the three forms of the invention in the present application. The application is inventive in the light of this document.
The other 11 documents
The other documents describe natural insulins and commercial insulin compositions. While these documents may be relevant to work in the art, they do not suggest anything about derivatives (with the exception of Blundell, which is discussed below). The application is inventive in the light of these documents.
The Blundell article lists the properties of a number of insulin derivatives in Table IV, including LysB29 derivatives. Solubility is not one of the listed properties. Consequently this document does not suggest anything that might be a solution to the problem.
Conclusion
The invention as claimed is not obvious in the light of any of the citations.
MANNER OF MANUFACTURE
The law on manner of manufacture has been examined in a number of recent decisions by Australian courts: the High Court in NV PhilipsGloelampenfabriken v Mirabella International Pty Ltd (1995) 183 CLR 655 and Advanced Building Systems Pty Ltd v Ramset Fasteners (Aust) Pty Ltd (1998) 194 CLR 171, the Federal Court in Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 46 IPR 553.
In the Bristol-Myers case, the majority summarised the effect of the Philips case as:
"Philips stands for the proposition (as a matter of construction of the 1990 Act) that if, on the basis of what was known, as revealed on the face of the specification, the invention claimed was obvious or did not involve an inventive step - that is, would be obvious to the hypothetical non-inventive and unimaginative skilled worker in the field (Minnesota at CLR 260 per Barwick CJ) - then the threshold requirement of inventiveness is not met."
[page 564]
It is clear that I must consider whether there is an inventive step in the light of what the specification states is known. What is known is not everything that is stated in the specification:
"Some elaboration, however, is required in relation to what the specification reveals as 'known'. If a patent application, lodged in Australia, refers to information derived from a number of prior publications referred to in the specification or, generally, to matters which are known, in our view the court - or the commissioner - would ordinarily proceed upon the basis that the knowledge thus described is, in the language of s 7(2) of the 1990 Act, part of 'the common general knowledge as it existed in the patent area'."
[Bristol-Myers at page 564]
The assessment is based on the common general knowledge as it is presented on the face of the specification. If information is stated to be common general knowledge or can reasonably be inferred to be common general knowledge, then the Commissioner can accept that conclusion without any further evidence. It is not correct to say that every piece of prior art referred to in the specification is part of the common general knowledge.
The relevant background art referred to by the specification are:
a) "Protracted insulin compositions are well known in the art" (page 1, line 14). I consider that this statement is equivalent to a statement that this is common general knowledge.
b) There are drawbacks with known protracted insulin compositions (page 1, line 20 and following paragraphs). There is no explicit statement as to whether this was generally known. The general way in which this is described seems to suggest that it is widely known and does not require detailed explanation. It seems likely that this is common general knowledge.
c) WO 91/12817 (page 2, line 23). The specification suggests that this document is known, but there is nothing that indicates that it is well known.
d) Substitution of one or more of the amino groups of insulin (page 2, line 27). The specification indicates that this is known from several sources. It is possible that this implies that this is common general knowledge. The specific documents that disclose these derivatives (US 3,528,960, GB 1,492,997, JP 1-254699 (page 2, line 27) are not indicated as well known in their own right.
It is apparent from the face of the specification that a protracted insulin is needed to overcome the known drawbacks. The fact that substituted insulins are common general knowledge does not lead to a conclusion that the specific LysB29 substituted insulins would be an obvious answer. Additionally, it is not obvious how to prepare such LysB29 substituted insulins. I believe that there is an invention on the face of the specification.
Another argument presented was that application is merely a collection of ideas for steps to modify the insulin structure. It is a well established principle that a mere idea is not patentable subject matter:
"No doubt you cannot patent an idea, which you have simply conceived, and have suggested no way of carrying it out, but the invention consists in the thinking of or conceiving something and suggesting a way of doing it … I think you can have a Patent for an idea, which is new and original and very meritorious, if you suggest a way of carrying it out. If you do not so suggest, you cannot no doubt have a patent"
[Hickton's Patent Syndicate v Patents and Machine Improvements Co Ltd (1909) 26 RPC 339 at 348]
In the present case there is not only a suggestion of how to carry out the invention, but there are several worked examples. Clearly the present case is not an example of a mere idea.
Another suggestion is that the application is merely a collocation of known chemical moieties, that do not achieve a new or improved result. The notion of a collocation is not normally applied to a single chemical, but I see no reason in principle why it cannot be. In the present case the compound can be viewed as a collocation of insulin and a lipophilic substituent and possibly zinc. One reason why chemicals are not normally viewed as collocations is that the overall activity of a compound may not simply be the sum of the activities of the parts that make it up. In the present case it seems that the compound is an insulin that has increased lipophilicity, i.e. a compound have a combination of the properties of the moieties that make it up. However, the specification also states that the compounds are soluble at physiological pH. This is a property that does not arise in a simple way from one of the moieties, i.e. there is no "water solubility moiety". Consequently, there is apparently a new or improved result achieved, so the compound represents a patentable combination.
CONCLUSION
The opposition fails on all grounds.
Unless the Commissioner is served with a notice of appeal within 28 days of the date of this decision, I direct that the application proceed to sealing.
COSTS
The power of the Commissioner to award costs is based on section 210 and regulation 22.8. This power is discretionary, so I must take into account all relevant considerations.
I have found that the opposition fails on all grounds. In this circumstance it is appropriate for costs to follow the event.
I award costs in accordance with Schedule 8 against Eli Lilly and Company.
Dr S.D.Barker
Delegate of the Commissioner of Patents
Patent attorneys for the applicant : Phillips Ormonde & Fitzpatrick, Melbourne
Patent attorneys for the opponent : F B Rice & Co, Melbourne
ANNEX 1: CLAIM 1 OF 682061
An insulin derivative having the following sequence:
wherein
Xaa at positions A21 and B3 are, independently, any amino acid residue which can be coded for by the genetic code except Lys, Arg and Cys;
Xaa at position B1 is Phe or is deleted;
Xaa at position B30 is (a) a non-codable, lipophilic amino acid having from 10 to 24 carbon atoms, in which case an acyl group of a carboxylic acid with up to 5 carbon atoms is bound to the e-amino group of LysB29, (b) any amino acid residue which can be coded for by the genetic code except Lys, Arg and Cys, in which case the e-amino group of LysB29 has a lipophilic substituent or (c) deleted, in which case the e-amino group of LysB29 has a lipophilic substituent; and
any Zn2+ complexes thereof,
provided that when Xaa at position B30 is Thr or Ala, Xaa at positions A21 and B3 are both Asn, and Xaa at position B1 is Phe, then the insulin derivative is a Zn2+ complex.
2
0
0