Center for Advanced Study & Research on Intellectual Property


CASRIP Newsletter - Summer 1996, Volume 3, Issue 2

1995 EPO Decisions in Biotechnology

Karen Canady

The following is an overview of decisions in the area of biotechnology made in 1995 by the European Patent Office Boards of Appeal. The decisions will be referred to by case number, and a list of the decisions, including name and date, appear at the end of this article.

Patentable Subject Matter

T 356/93 & G 3/95: Claims to plants (or animals?) which encompass plant (or animal) varieties are not patentable.

After the 1990 decision confirming the patentability of the Harvard onco-mouse, the prevailing view was that transgenic plants and animals were patentable under the European Patent Convention (EPC) if the claims were directed to classifications higher tha n the variety level. The February 1995 decision, T 356/93 by the recently established Technical Board for Biotechnology, contradicts that prevailing view.

Opponent had appealed from the Opposition Division's decision to maintain a patent having the following claim 1:

Process for controlling the action in plant cells and plants comprising such cells of a glutamine synthetase inhibitor when the former are contacted with the latter, which comprises causing the stable integration in the genomic DNA of said plant cells of a heterologous DNA including a promoter recognised by polymerases of said plant cells and a foreign nucleotide sequence capable of being expressed in the form of a protein in said plant cells and plants, under the control of said promoter, and wherein sai d protein has an enzymatic activity capable of causing inactivation or neutralisation of said glutamine synthetase inhibitor.

The patent also contained claims to a process for producing the plant, to transformed plant cells, and to transformed plants.

The Board first held that the claims were in compliance with the requirements of Article 53(a) EPC, which prohibits the grant of patents in respect of inventions contrary to ordre public or morality. This aspect of the decision was discussed in th e Spring 1995 issue of the CASRIP Newsletter.

The Board then addressed whether the claims were patentable under Article 53(b) EPC, which excludes from patentability "plant or animal varieties or essentially biological processes for the production of plants or animals; this provision does not apply to microbiological processes or the products thereof."

The Board defined and clarified key terms: A plant variety is "any plant grouping within a single botanical taxon of the lowest-known rank which, irrespective of whether it would be eligible for protection under the UPOV Convention, is characterised by a t least one single transmissible characteristic distinguishing it from other plant groupings and which is sufficiently homogeneous and stable in its relevant characteristics."

A plant cell is neither a plant nor a plant variety, but is considered to be a microbiological product. Product claims encompassing plant varieties are not patentable.

An essentially biological process does not include "a process for the production of plants comprising at least one essential technical step which cannot be carried out without human intervention and which has a decisive impact on the final result."

A plant variety is patentable if it is the product of a microbiological process within the meaning of Article 53(b). A microbiological process includes any activity in which "an integrated use is made of biochemical and microbiological techniques, includ ing genetic and chemical engineering techniques, in order to exploit the capacities of microbes and cultured cells."

The product of a microbiological process is a product "made or modified by microorganisms as well as new microorganisms as such." The term microorganism includes bacteria, yeasts, fungi, algae, protozoa and human animal and plant cells, or "all generally unicellular organisms with dimensions beneath the limits of vision which c an be propagated and manipulated in a laboratory," as well as plasmids and viruses.

A technical process comprising a series of steps which include a microbiological step, is not a microbiological process, nor are the products of such technical processes the products of a microbiological process within the meaning of Article 53(b) EPC.

Regarding the claims at issue, the Board held the following:
1) The process for producing the transformed plant is not an essentially biological process, and therefore is not excluded from patentability. The transformation of cells is an essential technical step which has a decisive impact on the final result.
2) The claim that relates to plant cells cannot be said to encompass plant varieties.
3) The plant claim in the patent is not directed to a plant variety. This claim does, however, encompass plant varieties. The claimed subject matter includes "essentially derived varieties," because the plants are distinguishable, uniform and stable in their characteristics.

Therefore, the plants could only be patentable if they represent the product of a microbiological process. This is not the case, because of the series of agrotechnical and biological steps which follow the initial microbiological step of transfor mation.

This decision should be compared with T 19/90, the "Onco-mouse/HARVARD" decision, which dealt with the patentability of genetically-engineered animals. After holding that the patenting of animals per se was not prohibited by the EPC, the Board remitted the case to the Examining Division to determine whether the claims covere d animal varieties. The Examining Division in "Onco-mouse/HARVARD" held that the terms "mammal" and "rodent" read on taxonomic classifications much higher than the "animal variety" term used in Article 53(b) EPC, and therefore the subject matter of claim s directed to genetically altered mammals and rodents was not barred from patentability. Unfortunately, the more recent decision by the Technical Board of Appeal in T 356/93 implicitly overrules the "Onco-mouse/HARVARD" decision as it suggests tha t claims to stably transformed animals encompass animal varieties.

The President of the EPO sought clarification of this contradiction by referring the following question to the Enlarged Board of Appeal: Does a claim which relates to plants and animals but wherein specific plant or animal varieties are not individually claimed contravene the prohibition on patenting in Article 53(b) if it embraces plant or animal varieties? In G 3/95, the Enlarged Board rejected the referral based on its finding of no contradiction between T 19/90 and T 356/93. Un fortunately, the Enlarged Board did not comment on the cited decisions.


T 923/92: A claim which refers to a specific amino acid sequence cannot have priority based on a document disclosing a different amino acid & nucleotide sequence.

In this appeal from the opposition to Genentech's t-PA patent, the patentee needed to rely on either of two priority documents to avoid a novelty-destroying reference regarding the amino acid and nucleic acid sequences of t-PA. The two earliest priority documents, however, disclosed an amino acid sequence which differed by three residues as well as a correspondingly different nucleotide sequence.

Crucial to the outcome was the determination of whether the differing sequences could be considered the same invention. Article 87 EPC, which governs the right to priority, requires that the European application and the priority application relate to the same invention.

The Board concluded that "the primary amino acid sequence of a protein (or the nucleotide sequence of a DNA) constitutes a true technical feature" and that the particular sequence relied on for the definition of the subject matter of an invention is criti cal. The Board held that the relevant claim was not entitled to the earlier priority dates because it did not relate to the same invention as was disclosed in the priority documents.

Patentee had to instead define the DNA in its claims by reference to the source of the mRNA, the restriction map and the probes with which it hybridizes.

Prior Art

EP -B1 901 184: A research grant proposal is prior art once awarded and available to the public.

The Opposition Division, in deciding on the opposition to Hoffman-La Roche's PCR patent, held that a research grant application had become prior art for the purpose of assessing patentability once the grant had been awarded and was available to the public . This determination was made by interpreting the applicable U.S. law, the Freedom of Information Act (FOIA).

Factors supporting this decision include: the availability, through a prior publication, of the reference numbers with which one skilled in the art could identify the grant application; the availability of a search system for the identification of the gr ant application; the fact that a scientist had obtained a copy of the grant application upon request; and the public disclosure under FOIA of all but privileged information, such as salaries and budget items.


T 645/90: Claim language such as "ropy" and "slimy" is meaningful to one skilled in the art and can suffice to support novelty.

The opponents had attacked the novelty of claim 1 of a patent relating to fermented milk products. The opponents challenged as meaningless the terms "ropy" and "slimy", which were essential to distinguish the claimed product from the prior art. The Boar d disagreed, and held that the terms "ropy"and "slimy" are meaningful to one skilled in the art and sufficient to support novelty.

Claim 1 reads as follows:

A thick bodied liquid, fermented milk product which comprises: milk with a pH between about 4.2 and 4.7 which has been fermented with a mixed culture containing [various ingredients] to produce the thick bodied milk product without resulting in a slimy o r ropy product, wherein the product [has additional features].

T 63/94: A disclaimer may be introduced into a claim to establish novelty.

In this appeal from an opposition, the patentee was permitted to incorporate a disclaimer of the prior art in order to establish novelty. New claim 1 was allowed as follows, with the added disclaimer underlined:

Method for determining antigens by means of an immunological reaction, whereby the antigen must enter into a bond with at least two antibody molecules at least one of which is labelled, characterized in that hereby two or more different sorts of monoclona l antibodies are used directed against the same antigen; with the exception of the use of two monoclonal antibodies one of which is bound to a solid surface and rendered insoluble and the other of which is labelled, in sandwich assays.

The EPO ordinarily requires that the claimed subject matter be defined through the use of positive language. In this case, however, the Technical Board recognized that a positive limitation would restrict the claim to the embodiments presented in the exa mples, and would therefore unduly deprive the patentee of fair protection.

Inventive Step

T 63/94, T 906/91 & T 355/92: The replacement of polyclonal antibodies with monoclonal antibodies is not inventive in the absence of surprising advantageous properties or an unexpected solution to the problem.

Consistent with previous decisions, the Board in T 63/94 held that the replacement of polyclonal antibodies with monoclonal antibodies does not involve an inventive step. Those skilled in the art had the incentive to make such a replacement, expec ting it to improve the relevant assays. The Board noted that, if the broad claims of the patent were in fact adequately supported by the accompanying limited disclosure, then one skilled in the art would have considered the effort required to isolate the monoclonal antibodies to be reasonable.

In T 906/91, an appeal from an opposition, the patentee unsuccessfully defended the inventive step of replacing polyclonal immunoassays for theophylline with a monoclonal antibody defined in the claims by its low cross-reactivity (less than 5%) wit h caffeine. Patentee argued that, at the priority date of the patent, those skilled in the art were optimistic but also cautious about the replacement of polyclonal antibodies with monoclonals. This argument was based on the content of review articles w hich were concerned almost exclusively with the use of macromolecules as immunogens. Predictions based on macromolecules, Patentee argued, do not necessarily apply to small molecules (haptens) which require conjugation to a carrier before they can function as immunogens. This is especially true for theophylline, one of the smallest known haptens.

The Board first identified the closest prior art as a polyclonal antiserum obtained by the use of a hapten/carrier conjugate designed to minimize cross-reactivity. The best polyclonal antibodies of the prior art had a cross-reactivity with caffeine of 1% .

The Board then concluded that, as of the priority date, the advantageous properties of monoclonal antibodies over polyclonal antisera were known and that it was obvious to replace polyclonals with monoclonals with the reasonable expectation of improved sp ecificity. The Board continued its analysis by examining six criteria which can provide a basis for inventive step:

(1) whether the person skilled in the art would have concluded from the prior art that haptens can be successfully used for the production of monoclonal antibodies;
(2) whether the selection of the antigen used by Patentee for immunization was inventive;
(3) whether the degree of improvement in cross-reactivity achieved by Patentee was surprising;
(4) whether it was surprising that theophylline is a sufficiently strong antigen in mice;
(5) whether there were procedural difficulties to be overcome by Patentee; and
(6) whether the person skilled in the art would have tried to solve the technical problem by a route other than that taken by Patentee.

The Board found that support for inventive step of the claimed subject matter could not be found on the basis of any of these criteria and revoked the patent.

In T 355/92, an appeal from the rejection of an application by the Examining Division for lack of inventive step, the Technical Board found in favor of the applicant. The main claim of the application related to a monoclonal antibody to follicle s timulating hormone (FSH) having a cross-reactivity with other glycohormones of less than 3%.

Because FSH was known to be an immunogen, and because it was known that antibodies to glycohormones often crossreact with other glycohormones, the Examining Division considered the problem of looking for antibodies with low crossreactivities to be obvious . The solution was also considered obvious in light of the conventional K`hler/Milstein method.

The Technical Board of Appeal was persuaded by the applicant's arguments defending the patentability of the anti-FSH monoclonal antibodies:

(1) unlike T 906/91, a polyclonal antiserum with loss crossreactivity was not known for glycohormones;
(2) due to the high homology between glycohormones, all known monoclonal antibodies displayed a level of crossreactivity much higher than that of the invention;
(3) development of the monoclonal antibodies required application of a special immunization protocol involving an unconventional combination of adjuvants; and
(4) the prior art contained no suggestion that monoclonal antibodies with much lower crossreactivity could be obtained by using the requisite combination of adjuvants.

T 478/92 was a parallel decision concerning anti-LH antibodies, with the same result.

T 495/92: The provision of a gamma interferon variant, in the absence of unexpected advantageous properties, does not involve an inventive step.

The opposed patent claimed a DNA sequence encoding interferon-gamma with glutamine at position 9 of the amino acid sequence. The Opposition Division had maintained the patent and the opponent appealed. The Technical Board held that the subject matter la cked inventive step.

The closest prior art was defined as the wild type interferon-gamma which has lysine at position 9 and a variant which has glutamine instead of arginine at position 140. The technical problem addressed by the opposed patent was then to provide an alterna tive interferon-gamma. The patentee had submitted data from two experiments showing, by averaging three measurements, greater biological activity for the claimed variant. In both experiments, the three measurements showed a high degree of variance.

The Board based its finding of a lack of inventive step on the following:

(1) the provision of a different interferon-gamma was not unexpected;
(2) there was no prejudice in the art against the position and type of variation obtained; and
(3) there was insufficient evidence that the variant demonstrates any significant unexpected properties, as the data offered by the patentee was inconclusive regarding the biological activity of the claimed variant.

T 698/93: A process which combines two published steps is inventive where the authors of the second publication, approaching the problem 17 years after the first publication, chose not to incorporate the earlier published step into their method.

This case arose from an opponent's appeal from the Opposition Division's decision to maintain a patent with the following main claim:

A process for reducing the free amine content of a foodstuff with an amine oxidase which process comprises contacting the foodstuff with an amine oxidase enzyme derived from the microorganism Aspergillus niger at a temperature of between 20 and 50EC and i n a medium at pH of between 7 to 9, in the presence of molecular oxygen.

The opponent argued that the claim lacked inventive step on the basis of a document disclosing the oxidation of free amines to aldehydes in the presence of oxygen in combination with a document, published 17 years later, disclosing the removal of free ami nes from mackerel by treatment with diamine oxidase. Although the earlier publication made no mention of the treatment of food, the opponent argued that those skilled in food technology at the priority date of the patent knew that the amines referred to the earlier publication were usually present in food, making their removal from foodstuffs by the claimed method obvious.

The Board noted that, although the document disclosing the use of molecular oxygen to oxidize free amines had been published 17 years earlier, the second group of authors did not choose to use molecular oxygen when treating mackerel with an amine oxidase from Aspergillus niger. The Board cited this as confirmation that those skilled in the art did not consider the amines referred to in the earlier publication to be those usually present in food and therefore removable from foodstuffs by the previously disclosed process.

T 923/92: An RNA fraction containing t-PA mRNA did not, as of 1982, create a reasonable expectation of success for cloning t-PA cDNA sequence.

The closest prior art was identified as the disclosure of a 19S RNA fraction from Bowes melanoma cells and containing t-PA mRNA. The Board then regarded the technical problem to be providing qualitatively and quantitatively adequate amounts of DNA for re combinant production of human t-PA.

The Board considered the state of the art of cDNA synthesis and cloning as of the priority date, and noted various factors which would affect the degree of confidence a skilled person would have had in addressing the technical problem. Because of these f actors (the risk of reagent contamination, the limited experience with relevant methods, vectors and screening procedures), the Board felt there was not a reasonable expectation of success in cloning a t-PA cDNA sequence using the disclosed mRNA fraction. The Board therefore acknowledged inventive step.


T 923/92: Enabling disclosure does not require that a specific example be exactly repeatable.

The appellants (opponents) challenged the sufficiency of the disclosure on the basis of later evidence that the human t-PA molecule did not express well in E. coli. The Board considered the evidence of poor expression of t-PA in E. coli and accumulation of most of the protein in the cytoplasm in an inactive form to actually confirm that a DNA sequence encoding human t-PA can indeed be expressed in an E. coli recombinant system.

The Board pointed out the importance of looking at the wider disclosure of the entire patent which included the full nucleotide and amino acid sequences. The Board also noted that the issue is whether the disclosure taught how to achieve expression of hu man t-PA in E. coli at any level.

The Board held that a skilled person, equipped with the patent's disclosure, could achieve expression in an E. coli system without applying inventive skill or undue experimentation.

T 923/92: Biological material necessary to practice the invention need not be deposited if it is already part of the state of the art.

The Opponents had also raised the issue that the Bowes melanoma cell line needed for the production of a t-PA cDNA sequence was not deposited and the invention was therefore not enabled because unlimited access to the cell line had not been provided.

The Board noted the large body of evidence that the cells were generally available and freely exchanged in the scientific community and that no restrictions had been placed on the dissemination of the cells. The Board considered the cells to form part of the state of the art at the priority date, and held that the patentee could not have been obliged to ensure their availability through a deposit.

T 918/94: Experimental verification of a reaction scheme is required where one skilled in the art would question feasibility of a reaction essential to the claimed subject matter.

This appeal arose from the rejection of an application claiming:

A vector including a structural gene encoding a polypeptide comprising the amino acid sequence of human calcitonin, wherein said polypeptide is enzymatically processable to produce human calcitonin using the C-terminal modification activity of the yeast e nzyme carboxypeptidase Y [CPD-Y].

The application did not contain a protocol for converting a calcitonin predecessor to calcitonin. Instead, the application referred to two publications by a group working on CPD-Y. Unfortunately for the applicant, later publications by the same group su ggest that CPD-Y would not work for the purpose suggested in the patent application. This same group had also stated that a given reaction cannot be considered feasible without experimental verification.

The Board found the claim unallowable because sufficient disclosure requires more than a mere reaction scheme. The skilled person must also be taught by way of experimental verification whether the reaction actually works, and which measures should be ta ken in the event of failure.

Decisions cited: T 645/90 "Fermented milk products/MICROLIFE" (February 1, 1995)
T 906/91 "Monoclonal antibody to theophylline/DuPONT" (April 25, 1995)
T 355/92 "Anti-FSH antibody/BOEHRINGER MANNHEIM" (October 24, 1995)
T 478/92 "Anti-LH antibody/BOEHRINGER MANNHEIM" (October 24, 1995)
T 495/92 "Gln(9)-interferon variant/CANCER INSTITUTE" (September 19, 1995)
T 923/92 "t-PA/GENENTECH"
T 356/93 "Plant cells/PLANT GENETIC SYSTEMS" (February 21, 1995; published in OJ EPO 1995 at 545)
T 698/93 "Amine removal/CPC" (October 16, 1995)
T 63/94 "Determination of antigens/AKZO NOBEL" (January 18, 1995)
T 918/94 "Calcitonin/CELLTECH" (July 6, 1995)
EP-B1 201 184 Opposition Division "PCR/Hoffman-La Roche" (December 14, 1995)

Last updated 4/27/2012