Takeda Chemical Industries v F Hoffmann-La Roche Aktiengesellschaft
[1996] APO 3
•18 January 1996
official notice
decision of a delegate of the commissioner of patents
Application : No. 592527 in the name of Takeda Chemical Industries
Title: Non-glycosylated Human Interleukin-2 Recombinant Production
Action: Opposition under s.59 of the Patents Act 1952 by F. HOFFMANN-LA ROCHE AKTIENGESELLSCHAFT
Decision: Issued .
Abstract: Opposition successful in part. Claims 1-16, 18 are not fairly based because they do not contain all of the essential features of the invention. Claims 1-7, 14-16, 18 do not define a manner of manufacture because they define a known goal by desiderata.
None of the documents filed in evidence prior claim the current claims nor deprive any of the claims of their novelty.
The Opponent has not established that any of the claims lack inventive step.
patents act 1990
decision of a delegate of the commissioner of patents
Re: Patent application no. 592527 by Takeda Chemical Industries, and opposition by F. Hoffmann-La Roche Aktiengesellschaft
background
Patent Application 592527 by Takeda Chemical Industries was lodged on 21 November 1984 as a convention application claiming priority from application JP 58-225079 filed on 28 November 1983. The application was advertised accepted on 18 January 1990 and a Notice of Opposition was lodged by F. Hoffmann-La Roche Aktiengesellschaft on 18 April 1990.
The serving of evidence was completed 14 September 1994. On 24 March 1992, the applicant filed a voluntary request to amend the complete specification. These amendments were subsequently allowed on 2 August 1993.
A hearing to determine the Opposition took place in Canberra on 20 July, 1995. The applicant was represented by Mr David Catterns (QC) of counsel instructed by Mrs Ann Kurts and Ms Volga N. Vegar, patent attorneys from Griffith Hack & Co., Sydney. The opponent was represented by Dr Annabelle Bennett of counsel instructed by Mr John O'Connor, patent attorney of Spruson and Ferguson, Sydney.
As the application was filed and accepted before the commencement of the 1990 Patents Act, the provisions of section 234(3) of the 1990 Act, and regulation 23.3 are applicable.
The grounds of opposition listed on the notice of opposition and relied on at the hearing are laid down in paragraphs 59(1)(c), (d), (e), (f), (g), (h) and (i) of the Patents Act 1952, namely prior claiming, prior publication, manner of manufacture, obviousness, novelty and non-compliance with section 40 of the Act.
specification
The patent specification as amended after acceptance is entitled "Non-glycosylated human Interleukin-2 Recombinant Production". Natural interleukin-2 (also called T cell growth factor) is a glycosylated protein produced by T cells after stimulation with, for example, a lectin. According to the specification, interleukin-2 is a lymphokine which affects other T cells (including natural killer and helper T cells) making it useful in long-period culture of T cells and potentially useful in the treatment of cancer and other diseases.
The admitted prior art acknowledges that recombinant non-glycosylated interleukin-2 was known and its potential clinical application had been identified. Consequently, it was a recognised need (or obvious goal) to develop highly purified, recombinant interleukin-2.
The specification asserts in the consistory statement (page 4, lines 16-21) that the invention resides in:
"...a non-glycosylated human recombinant interleukin-2 protein having a specific activity of not less than 104 U/mg and having a purity of not less than 99% and displaying a single band on SDS-polyacrylamide gel electrophoresis analysis under both reducing and non-reducing conditions..."
However, as this statement is clearly simply restating the recognised need (the known desiderata), it cannot possibly be the invention. Therefore I need to go beyond this simple assertion and deduce the invention from the specification as a whole.
The object of the invention was to provide a highly purified interleukin-2 in large amounts for clinical use. The specification suggests that this is achieved by:
"...a method of producing said human recombinant interleukin-2 protein which comprises cultivating a transformant carrying a DNA having a base sequence coding for human recombinant interleukin-2 and purifying said protein from the culture broth."
As cultivating transformants with DNA encoding human recombinant interleukin-2 is acknowledged prior art, the method of purification is the only possible advance in the art provided by the specification.
On page 8, the specification states that purification of interleukin-2 can be carried out by any "appropriate combination of per se known techniques". The specification supplies a list of well-known protein purification techniques, but does not teach the skilled worker how to select an "appropriate combination" of them. Without such teaching, the skilled worker would have an undue burden of experimentation to identify the complete range of "appropriate combinations".
Thus, although the specification acknowledges that there could potentially be numerous "appropriate combinations" which would purify non-glycosylated recombinant interleukin-2 to the level defined in the claims, it clearly does not disclose all the possible "appropriate combinations". In reality, the specification only discloses one particular combination of techniques which achieves the requisite level of purity and this is in the preferred embodiment. This combination (in any order) is (from pages 9 and 10 and the examples):
anion exchange column chromatography,
gel filtration column chromatography, and
reverse phase HPLC.
The skilled worker wanting to achieve the level of purity defined in the claims is taught to use this particular combination of purification techniques because no other successful combination is disclosed. Thus, each step of that combination is essential to the invention disclosed in the specification.
The applicant argued that reverse phase HPLC was the only essential element of the "appropriate combination" disclosed in the specification. I disagree. Although the specification noted that reverse phase HPLC is "very effective" (page 9), it did not suggest that any combination of techniques which included this step would achieve the object of the invention.
Further, reverse phase HPLC was never taught as having special or surprising qualities which made it essential to all "appropriate combinations". It is clearly a useful step in the appropriate combination disclosed, but it is only part of a successful combination. Thus, although I accept that reverse phase HPLC was an important part of the "appropriate combination", I cannot accept that it was the only essential part of the disclosed combination.
I conclude the particular combination of steps which I have summarised above is essential to the invention disclosed in the specification. I therefore construe the invention to reside in:
"a non-glycosylated human recombinant interleukin-2 protein having a specific activity of not less than 104 U/mg and having a purity of not less than 99% and displaying a single band on SDS-polyacrylamide gel electrophoresis analysis under both reducing and non-reducing conditions produced by a purification method including the following steps:
1.anion exchange column chromatography,
2.gel filtration column chromatography, and
3.reverse phase HPLC."
Having construed the invention, I can now refer to the claims. The specification ends with 21 claims, of which 8 (claims 1, 8, 14, 16, 17, 19, 20, 21) are independent. The two independent claims of particular importance in this decision are claims 1 and 8:
A non-glycosylated human recombinant interleukin-2 protein having a specific activity of not less than 104 U/mg, a purity of not less than 99% and displaying a single band on SDS-polyacrylamide gel electrophoresis analysis under both reducing and non-reducing conditions.
A method of producing a non-glycosylated human interleukin-2 protein having a specific activity of not less than 104 U/mg, a purity of not less than 99% and displaying a single band on SDS-polyacrylamide gel electrophoresis analysis under both reducing and non-reducing conditions which comprises growing a transformant carrying a DNA having a base sequence coding for human interleukin-2 to cause production and accumulation of human interleukin-2 in the culture broth and subjecting the thus obtained human interleukin-2 to a purification process which includes a high performance liquid chromatography using a reversed phase column.
essential features of the claims
Of the features in claim 1 as currently drafted, only two are in dispute regarding their "essentialness". These are whether the interleukin-2 molecule is:
(a)recombinant (or produced by a transformant carrying a DNA having a base sequence coding for human interleukin-2); and
(b) non-glycosylated
which the opponent submitted were inessential.
In my view, the feature of interleukin-2 being "recombinant" (or produced by a transformant carrying a DNA having a base sequence coding for human interleukin-2) is clearly essential. The object of the invention (page 3) is to produce a highly purified interleukin-2 in large amounts. Genetic engineering is crucial to achieving this object and therefore must be an essential feature of the invention and claims (Catnic Components v Hill and Smith Ltd, [1982] RPC 183).
From a construction of the specification as a whole, the feature of "non-glycosylated" is also essential. It is mentioned throughout the specification as an essential (not a preferred) feature and it is present in the consistory statement, all the examples and claims. I also accept the applicant's submissions (see Kato's declaration, page 4, paragraph 16) that the non-glycosylated human interleukin-2 is uniform in molecular weight and does not contain inter and intra batch variation. This is particularly advantageous in clinical applications as the composition of the administered interleukin-2 is precisely known and controllable.
decision
Section 40
I concluded above that both the protein and its potential clinical application were known and the purification of the protein was a known goal.
In my opinion, claims to the purified protein produced by any method are mere desiderata and are clearly speculative:
"A patentee.....is not entitled to claim a monopoly more extensive than is necessary to protect what he has himself said is his invention'. He cannot claim all solutions to a problem unless invention lies in the identification of the problem."
David Kahn Inc. v. Conway Stewart, (1974) RPC 279 (at pages 319 and 320)
The only advance over the prior art must be in a particular method of purification of the protein and (as I stated above) the particular method must contain the following steps (in any order):
(i) anion exchange column chromatography;
(ii) gel filtration column chromatography; and
(iii) reverse phase HPLC.
These features are not in claims 1-16, and 18. For this reason these claims are not fairly based.
The opponent submitted that there were further section 40 problems:
(a)that the phrase "very low in pyrogen" in claim 2 was unclear because it was indeterminate in scope; and
(b)claim 1 was not fairly based because, in defining an activity of "not less than 104 U/mg", it was inconsistent with the description (page 10) which describes the activity as "not less than 2 x 104 and up to 4 x 104".
On the first point, I accept that the phrase does not have an exact lower limit. However, the levels of pyrogens are clearly a recognized problem in this art and "low levels" of pyrogens are clearly desirable. It seems to me that a skilled worker would understand the term and be able to provide a general limit (perhaps non-toxic levels of pyrogen) to the phrase. I therefore find that the term is clear.
On the second point, I do not see a major inconsistency between the claims and the specification as other parts of the specification refer to the same level of activity as claim 1. Claim 1 may be inconsistent with page 10 but not with the specification as a whole. I therefore find that there is no fair basis problem with the level of activity.
priority dates
The opponent argued that the applicant was identifying reverse phase HPLC as a key step in an "amorphous mass" of other methods to try and suggest an inventive step in the specification. The opponent submitted that this was a "selection" which was not fairly based on either the basic specification or the specification as filed (Coopers Animal Health v Western Stock Distributors 11 IPR 20).
I accept that the basic specification and specification as filed did not identify reverse phase HPLC as a key or essential purification step but as I concluded above, neither does the specification as amended after acceptance. All three documents specify that it is the "appropriate combination" of purification techniques which achieves the object of the invention rather than the step of reverse phase HPLC per se.
I concluded above that only claims with a particular "appropriate combination" are fairly based on the specification. I will only consider such claims in my assessment of prior claiming.
However, as an aside, I note that claims 8-13 directed to any purification process of non-glycosylated interleukin-2 involving reverse-phase HPLC are not fairly based on the basic specification for the same reasons as they are not fairly based on the complete specification. As a result, these claims are only entitled to an earliest priority date of the date of filing of the complete specification and it seems likely that these claims are prior claimed by the claims of the opponent's patent (AU, B, 36995/84 (580276) filed 20 December 1984) which have an earliest priority date of 23 December 1983.
Prior claiming
The opponent relied on 3 Australian patent applications with earlier priority dates than the earliest possible priority date of the claims of the instant application to argue their case that the claims of the instant application were prior claimed:
(a)AU 21472/83 (patent number 556353) earliest priority date 15 December 1982; published in Australia 21 June 1984) by Ajinomoto Co. Inc. and the Japanese Foundation for Cancer Research;
(b)AU 19160/83 (patent number 559677) earliest priority date 16 September 1982; published in Australia 22 March 1984) by F Hoffman la-Roche;
(c)AU 71970/87 (patent number 598251) (divisional application of 19160/83 with the same earliest priority date as its parent and published in Australia 13 August 1987).
Document (a) discloses the cloning of interleukin-2 and its expression in both eucaryotes and prokaryotes - the recombinant product from the latter expression system being non-glycosylated. It does not disclose recombinant non-glycosylated interleukin-2 with the level of purity defined in the claims of the instant application.
Documents (b) and (c) both disclose reverse phase HPLC methods to purify native (glycosylated) interleukin-2. However, neither document teaches that recombinant (non-glycosylated) interleukin-2 could be purified using the same technique. The opponent argued that the term "interleukin-2" in the claims of documents (b) and (c) would encompass both the recombinant and native forms leading to prior claiming. I disagree. Document (b) states that the invention relates to human interleukin-2 derived from induced human malignant cells (page 3) (i.e: native glycosylated human interleukin-2) and clearly distinguishes native from recombinant interleukin-2 on page 7 by referring to "recombinant interleukin-2".
I find that none of these earlier documents claim recombinant non-glycosylated interleukin-2 with the level of purity defined in the claims of the instant application and hence none of the documents prior claim the instant application.
novelty
The closest related art published before the earliest priority date and the citation relied on in evidence and submissions for assessing novelty was:
K. Welte et al: "Purification of human interleukin-2 to apparent homogeneity and its molecular heterogeneity" J. Exp. Med. vol 156 (August 1982) pages 454-464 (exhibit JOC-18) - hereafter referred to as "Welte".
This citation was published in Australia on or before 10 November 1982. It disclosed a procedure for purifying interleukin-2 from a lectin-activated culture of human peripheral blood lymphocytes (PBL). The procedure involved ammonium sulphate precipitation, anion-exchange chromatography, gel filtration, and dye-ligand chromatography and resulted in a highly purified product with a specific activity of over 1 x 104 U/mg (table 1).
The generally accepted test for anticipation is the "reverse infringement" test as set out in Meyers Taylor Pty Ltd v Vicarr Industries Ltd (1977) 137 CLR 228 at 235 where Aickin J stated:
"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."
Infringement of a claim occurs when each and every one of the essential integers of the claim have been taken (Rodi and Wienenberger AG v Henry Showell Ltd [1969] RPC 367).
Both sides agree that the product of Welte's method has the same purity and activity as the interleukin-2 defined in the independent claims and that this level of purity and activity are essential features. There are, however, 2 key differences between Welte and the independent claims. The human interleukin-2 claimed in the instant application is both:
(a)recombinant (or produced by a transformant carrying a DNA having a base sequence coding for human interleukin-2) and
(b)non-glycosylated.
I have previously concluded that an interleukin-2 which is both non-glycosylated and recombinant (or produced by a transformant carrying a DNA having a base sequence coding for human interleukin-2) is essential to the invention and therefore to all the independent claims in which it appears. Since the product of the Welte method has neither feature, Welte does not contain all the essential features of any of the independent claims and does not deprive any of the claims of their novelty.
None of the other documents tendered in evidence and published prior to the earliest priority date disclose a non-glycosylated human recombinant interleukin-2 dependent claims of a particular level of purity. Since all these features are essential, the claims are novel in light of the evidence before me.
For the same reasons as given under prior claiming, documents (a)-(c) mentioned there do not deprive any of the claims of their novelty even if the priority date of the claims was after the publication date of the documents.
obviousness
The opponent suggested that protein purification techniques (including those in the specification) were well-known at the priority date of the claims and that it would have been routine experimentation for the skilled addressee to select from commonly known purification techniques to purify a known protein such as recombinant non-glycosylated interleukin-2.
In support of this, Dr Bernard provided evidence of a range of well-known techniques available to the skilled worker. However, she did not suggest that any or all of these techniques would have been expected to successfully purify the protein nor that the skilled worker would have been directly led to try any particular techniques at the priority date which would purify the protein. In my opinion, it is not routine experimentation to try each possible purification technique and combination of techniques to devise a successful purification strategy. Since the opponent has not established that the skilled worker would have selected particular techniques from the myriad available to them, I am not convinced that the skilled worker could have successfully purified recombinant non-glycosylated human interleukin-2 without an inventive step.
Later in her evidence, Dr Bernard made specific statements about particular techniques which could be used to purify recombinant non-glycosylated interleukin-2, but these statements were based on documents which were either not published at the priority date or not part of the then common general knowledge in Australia (on which obviousness is assessed under the 1952 Patents Act (Minnesota Mining and Manufacturing Company and Another v. Beiersdorf (Australia) Limited, 144 CLR 253)). Therefore these statements are not relevant for assessing obviousness.
The opponent also suggested that using reverse phase HPLC to purify non-glycosylated human recombinant interleukin-2 would not have involved an inventive step since the same technique had been used to purify other cytokines (including native gibbon interleukin-2) before the priority date.
Dr Sparrow provided 9 documents to support this allegation but provided no evidence that:
(a)any of the documents he tendered were part of the common general knowledge; nor that
(b)any of the cytokines purified in the documents (except, perhaps by implication, native gibbon interleukin-2) have similar physical properties to recombinant non-glycosylated human interleukin-2.
Further, each of the documents Dr Sparrow tendered showed the purification of native (glycosylated) proteins from eucaryotic sources. I do not accept that the skilled worker would have necessarily concluded at the priority date of the claims that any purification technique suitable for purification of glycosylated human interleukin-2 would also be suitable for the non-glycosylated form since the two may differ in size, charge and hydrophobicity.
Dr Sparrow conceded that carbohydrate side chains can modify the properties of a protein (see paragraph 8) and that this could, in turn, affect purification. Although, he suggests that the carbohydrate would not significantly affect the chromatographic properties of interleukin-2, his conclusion is based on an ex post facto analysis of both the Welte and the Robb and Smith (exhibit KK2) documents. Thus, Dr Sparrow has not established that the person skilled in the art would have arrived at the same conclusion at the priority date without the benefit of hindsight.
The only indication of what the skilled worker might have actually considered in 1983 comes from the opponent's patent specification AU, B, 36995/84 (580276) filed 20 December 1984 (approximately one month after the filing date of the opposed application) (exhibit JOC-20). This specification states on page 2:
"although these references describe the expression of interleukin-2 in transformed host cells, especially in E. coli, no purification procedures are disclosed which would permit one skilled in the art to obtain mature recombinant interleukin-2 in the homogeneous or substantially pure form.
The present invention relates to a process for the purification of recombinant interleukin-2 especially of recombinant mature human interleukin-2."
This suggests to me that there may well have been invention in determining a particular purification strategy for interleukin-2 at the priority date of the claims.
In any case, I do not need to determine where there is an inventive step, the onus is on the opponent to prove there was not one. I do not believe that they have done so as they have not established that any particular method (or range of methods) from the commonly known techniques would have been obvious to use in purifying non-glycosylated human recombinant HPLC at the priority date of the claims. Therefore they have not shown that any of the claims lack inventive step.
Manner of new Manufacture
I note the recent High Court decision in NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd (1995) AIPC ¶91-196 which concluded that:
"if it apparent on the face of the specification that the quality of inventiveness necessary for there to be a proper subject matter for letters patent under the statute of Monopolies is absent, one need go no further".
I previously concluded that purified non-glycosylated interleukin-2 produced by any method was a known goal. It follows that claims 1-7, 14-16, 18 which define this known goal by desiderata are not a manner of manufacture.
However, nothing in the specification leads me to believe that a skilled worker could have reasonably predicted that a particular technique or combination of purification techniques would have successfully purified recombinant, non-glycosylated interleukin-2. As a result, there is no evidence before me that a particular technique or combination of techniques lacked patentable subject matter. As a result, claims 8-13 all define a manner of new manufacture.
CONCLUSION
None of the documents filed in evidence prior claim the current claims nor deprive any of the claims of their novelty.
The Opponent has not established that any of the claims lack inventive step.
Claims 1-7, 14-16, 18 define a known goal by desiderata and do not define a manner of new manufacture.
Claims 1-16,18 are not fairly based because they do not contain all of the essential features of the invention.
I believe that the deficiencies I have noted above can be solved by amendment and allow the applicant 60 days in which to propose amendments.
costs
Costs normally follow the event and I see no reason to depart from this in this case. Therefore I award costs against the applicant.
Karen Ayers
Delegate of the Commissioner of Patents
Patent attorneys for the applicant : Griffith Hack and Co, Sydney.
Patent attorneys for the opponent : Spruson and Ferguson, Sydney
0
1
0