Eli Lilly & Company v Human Genome Sciences, Inc

Kitchin J held HGS’s EP (UK) 0,939,804 invalid, [2008] EWHC 1903 (Pat). His principal grounds were that none of the claims were “susceptible,” (Art. 57 EPC) - “capable” (s.1(1)(a) of the Patents Act 1977) - of industrial application and were obvious because the patent constituted mere speculation. The latter is the so-called AgrEvo (T-0939/92) ground. He also held some of the claims were insufficient, others not. He rejected other attacks on the patent.

HGS appealed all the findings adverse to it and Lilly cross-appealed some of those adverse to it. This Court unanimously upheld Kitchin J on the question of industrial applicability and did not go on to consider other points, [2010] EWCA Civ 33.

The Supreme Court reversed that decision and held certain claims sufficient and valid, [2011] UKSC 51. The combined effect of its decision and those below, i.e. the position to date, is that:

The Supreme Court decision left over for determination three issues. They concern claims 13, 18 and 19. I here use the current numbering which arose as a result of the decision of the TBA. Kitchin J used the original numbering.

HGS appeals Kitchin J’s finding that claims 18 and 19 (pharmaceutical and diagnostic compositions) are insufficient. Lilly appeals his finding that claim 13 is sufficient and that its amendment does not extend the protection it provides.

Commercially the really important claim is 13. At the time of the patent, neither the patentees nor anyone else, knew whether neutrokine-α or any of its antibodies or both would actually be valuable products. Claim 1 was to the isolated polynucleotide sequence for neutrokine-α (and variants of it). Claim 10 was to the polypeptide neutrokine-α and variants of it. None of these have proved valuable commercially.

But further research work by HGS and others, all of which was doubtless time-consuming, expensive and risky, has led to identification of specific antibodies to neutrokine-α which are promising. HGS in association with GSK have an antibody product for the treatment of lupus, Lilly an antibody product for the treatment of rheumatoid arthritis and multiple myeloma. We were told that Biogen and Genentech have potential antibody products too. These products, we were told, each have their own patents, as does HGS for its lupus product. Some may think that the evidence of this further substantial research work even in the face of HGS’s patent belies the notion that failure to grant patent protection far upstream from a practical product would have a “chilling effect” on research. That is not for us to say.

Commercially therefore what is at stake is whether the inventors and developers of these “downstream” products have to pay tribute to HGS because it has the master claim to all neutrokine-α antibodies. We are not of course concerned with the amount of such tribute which may be payable. I say no more than that if it were to be determined pursuant to an application for a compulsory licence pursuant to s. 47 of the Patents Act 1977 the question would be what is a “reasonable term” fixing the amount payable and that the assessment of that might well include the degree of risk, expense and work incurred by the respective parties.

As now amended this reads:

An isolated antibody or portion thereof that binds specifically to:

(a)

the full length Neutrokine-α polypeptide (amino acid sequence of residues 1 to 285 of SEQ ID NO: 2); or

(b)

the extracellular domain of the Neutrokine-α polypeptide (amino acid sequence of residues 73 to 285 of SEQ ID NO: 2)

Effectively this means any antibody that binds specifically to neutrokine-α. The sufficiency point was argued before us on that basis. Kitchin J held the claim sufficient.

The statutory language of the sufficiency objection is taken from Art. 83 of the EPC, which provides:

The European patent application shall disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

The Patents Act 1977 repeats these words as revocation ground s72(1)(c).

Kitchin J summarised at [239] the effect of this language as interpreted by case-law as follows:

The key elements of this requirement which bear on the present case are these:

i)

the first step is to identify the invention and that is to be done by reading and construing the claims;

ii)

in the case of a product claim that means making or otherwise obtaining the product;

iii)

in the case of a process claim, it means working the process;

iv)

sufficiency of the disclosure must be assessed on the basis of the specification as a whole including the description and the claims;

v)

the disclosure is aimed at the skilled person who may use his common general knowledge to supplement the information contained in the specification;

vi)

the specification must be sufficient to allow the invention to be performed over the whole scope of the claim;

vii)

the specification must be sufficient to allow the invention to be so performed without undue burden.

Mr Waugh did not challenge any of this. Nor did he challenge a crucial finding of fact Kitchin J summarised at [250]:

In the light of all this evidence I am satisfied that it did not require undue effort to make and identify specific antibodies to Neutrokine-α at the priority date.

The full implications of this need spelling out a little more. There will be many, probably millions, of antibodies to neutrokine-α. Some will bind well, others less so. But they can all be made and isolated.

Given that, Mr Waugh needed to and did advance a more sophisticated attack than simply saying the products of the claim cannot be made. I summarise the main lines of the argument as follows:

He reinforced his submissions in three ways:

I would reject all these submissions. I start with the main one. There are I think two, related, answers to it. Firstly, at the level of generality of this patent, all the antibodies have a “use.” Mr Thorley pointed out that “useful” could encompass uses other than pharmaceutical or diagnostic, for instance as a purifying agent if the binding is strong. He submitted that each member of the very large class of antibodies has, in a general sense, potential utility. Each member is susceptible of industrial application because the Supreme Court has said so. And that is enough. All the claim calls for is an antibody which specifically binds to neutrokine-α: that in itself is its potential utility. True it is that discerning any particular specific utility would or might involve a research project – undue effort – but that does not matter. Nor does it matter that one product may well be useless or much less so than other members of the class.

The second answer is one of construction. Mr Waugh’s argument involves reading in the further limitation that the antibody should be “useful” (assuming that this has a sufficiently precise meaning). Such a construction would divide antibodies to neutrokine-α into two classes, those which are “useful” and those which are not. And it would involve undue effort to find out whether a particular antibody was useful or not; to separate the wheat from the chaff.

The trouble with that submission is that the claim does not contain any limitation to “useful.” One does not read words into patent claims (or other documents for that matter) unless the context compellingly so requires. The context here does not. The skilled reader would know perfectly well that the patentees had discovered neutrokine-α, that it had some biological function similar to other members of the TNF ligand superfamily and that it or its antibodies might be useful. The antibody claim, just like the neutrokine-α claims, are all to things which could be valuable. He would not see from the patent any intention to limit the monopolies claimed to that which was “useful” for he would well know that the patentee had not limited himself to any particular utility – he was saying no more “I know of no particular utility yet, but all these products have potential utility.” So there is no reason why the skilled reader would read the claim as having a limitation to “useful” antibodies.

Moreover as I have already pointed out “useful” does not have any precise meaning. On any view, the reasonable skilled man reading the claim to find out what the patentee meant would not import a meaning of imprecise scope, which is what Mr Waugh’s argument necessarily involves.

I turn to the claims 18 and 19 subsidiary argument. Originally Mr Waugh had put this at the forefront of his case. It ended as a sort of “jury” point. It runs like this: Claims 18 and 19 are respectively for pharmaceutical and diagnostic compositions. They read as amended:

18.

A pharmaceutical composition comprising the polypeptide of any one of claims 10 to 12 or the antibody or portion thereof of any one of claims 13 to 17 and optionally, a pharmaceutically acceptable carrier.

19.

A diagnostic composition comprising the nucleic acid molecule of any one of claims 1 to 4, the polypeptide of any one of claims 10 to 12 or the antibody or portion thereof of any one of claims 13 to 17.

Mr Waugh submitted that these were limited to compositions with identifiable (from the teaching of the patent) pharmaceutical or diagnostic uses. Since such uses cannot be identified from the patent and could only be discerned after research amounting to undue effort, the claims were insufficient. Later I reject that submission when considering his specific sufficiency attacks on these claims. So the argument as the basis of an attack on claim 13 fails at this point.

But even id he were right, the argument would fail. It runs thus: within the class of antibodies of claim 13 there is a sub-class of antibodies to be used for the products of claims 18 and 19. But these cannot be identified without undue effort. So claim 13 includes within it antibodies which cannot be identified without undue effort. So it is not enabling across its breadth.

It fails because of the fundamental fact that all the antibodies of claim 13 can be made. Even if the antibodies to be used for the products of claims 18 and 19 do form a narrower class which cannot be ascertained without undue effort, that does not mean that the whole class is not enabled.

The same was true in Chiron v Murex, [1996] RPC 535. Chiron had discovered the hepatitis C virus. Claim 1 of its patent was to a class of polypeptides in substantially isolated form, namely those encoded by the genome of the virus or defined variants of it. I need not add more detail, the point is that it was to a large class of polypeptides, just as is claim 13 here. Claim 15 was to a vaccine composition according to claim 14. Claim 14 was to a composition of the polypeptide of claim 1 mixed with a pharmaceutically acceptable excipient. It was proved that a vaccine composition could not be made without undue effort at the very least (possibly at all). So the vaccine claim was insufficient.

Following the finding of insufficiency of claim 15, Chiron applied to amend to delete it. This was opposed on the grounds, inter alia:

4.

The proposed amendments are cosmetic only and do not cure, and are not intended to cure, the vices in the patent which were identified at the trial in this action, namely: (i) it monopolises vaccines yet fails to teach the skilled reader how to make any; (ii) it monopolises methods of growing HCV (being recombinant HCV) in tissue culture, yet it fails to tell the reader how to perform any of them. The patentee should expressly disclaim vaccines and methods of growing HCV in tissue culture.

In summary, claim 1 would be infringed by a vaccine. But a vaccine is not enabled. So claim 1 is not enabled to that extent and should be amended by cutting out vaccines.

Aldous J struck that plea out, saying at p.267:

Paragraph 4 of the statement of objections alleged that the amendments cure vices identified at the trial, that the claims, for example claim 1, monopolise vaccines and in vitro growth. It concludes that the patentee should expressly disclaim vaccines and in vitro growth with the result that claim 1 and the other claims should be limited so as to permit anybody to produce a vaccine whether or not they use the invention of claim 1.

Paragraph 4 of the statement of reasons is based upon a misconception as neither claim 1 nor the other claims referred to in the pleading claim an invention of a vaccine or a method of growing HCV in tissue culture. Claim 1 is a valid claim and therefore it does not fall foul of any vice identified by me. The fact that the defendants may need to use the invention of claim 1 to produce a vaccine or to grow the virus in tissue culture is irrelevant. I conclude that paragraph 4 should also be struck out as it contains allegations which cannot succeed.

The position is exactly the same here. Claim 13 is enabled because (as found by the judge) the antibodies can all be made. The fact that one may need to use the invention of claim 13 to produce a pharmaceutical or diagnostic for an identified disease and the patent does not enable one to identify any such disease is likewise irrelevant.

The point arose also on appeal from Aldous J’s main judgment but in a different way. Between the first instance judgment and that appeal, Biogen v Medeva [1995] RPC 25 was decided by the Court of Appeal. This led to a different attack of insufficiency in the Chiron appeal. Morritt LJ (giving the judgment of the Court) said at p.615:

Accordingly we are bound by Biogen to consider whether in this case on the facts the invention can be performed across its whole width. ….

In our judgment the contention of the appellants is misconceived. As we have already held in relation to the point of construction claim 1 does not require the mapping or identification of all the antigenic determinants or the best of them. It may well be that subsequent developments will justify a claim to an invention which enables that to be done. But that is not what this patent claims. This claim is to a polypeptide comprising an antigenic determinant encoded by the HCV genome as defined; such a polypeptide can be identified in the sequence by exposure to an antibody and can be made by routine methods of molecular biology. The effect of the issue as formulated by the appellants is to widen the claim from what it is and then to contend that so widened it is not enabled across its width. In our judgment the issue is not as the appellants have formulated it. The consequence is that the invention as actually claimed is capable of being performed by a person skilled in the art because it is a claim to a polypeptide which can be both made and identified by routine means. Accordingly in our judgment claim 1 is not invalid on the ground of insufficiency.

Mr Waugh sought to escape the Chiron case by submitting that there, once the vaccine claim was deleted, the vaccine would no longer infringe claim 1. That was simply wrong. If anyone could find a vaccine it would almost certainly have employed a polypeptide falling with claim 1. It is quite clear that both Aldous J and Morritt LJ proceeded on that basis.

I turn to the argument based on Pharmacia v Merck. Claim 1 was to a Markush formula consisting of a large number of compounds. There was no limitation in the claim as to purpose or use. Read alone it covered all the compounds of the formula. The body of the specification disclosed that certain members of the class were Cox II inhibitors. Aldous LJ said:

[18] Mr Kitchin is correct that claims 1 to 12 do not include any limitation as to use. Thus when construed without recourse to the rest of the specification, the invention claimed is to the chemical compounds set out. But that construction makes the invention inconsistent with, amongst other passages, the description of the invention in the specification. It states that “A class of compounds useful in treating inflammation-related disorders is defined by Formula 1”. I will deal with this submission and the other submissions on construction later in this judgment in the context in which they arise.

And:

[20] I agree with the judge. Nobody reading the specification could believe that the “invention” was the compounds claimed in claim 1. The specification makes clear that the patentees had found a class of compounds that could be made which at least had anti-inflammatory action. It was that contribution that merited a 20 year monopoly. In my view the only question capable of argument is whether the compounds in the class were chosen merely for their anti-inflammatory action or because in addition they had reduced side-effects due to them being Cox II selective.

On that basis Aldous LJ held that the wide class claim was insufficient because it included compounds which did not have the promised activity – it was not enabled across its scope even though all the compounds could be made. Mr Waugh says the same applies here. Even though all the antibodies can be made, only those which are “useful” are the real invention and the claim is insufficient in the same way it was in Pharmacia.

I do not accept that for two reasons. First I do not think the invention of this patent at its level of generality is limited in the same way as it was in Pharmacia to a specific activity. The skilled reader is not led to expect any specific activity, only a host of possibilities which may or may not in fact be. The fact of binding to neutrokine-α is all the invention of claim 13 is. Secondly however I am not clear that Pharmacia would be decided for the same reasons today. The objection is of an AgrEvo type – that the claim included within it many compounds as to which there was no technical contribution at all. This case is different because HGS have made a technical contribution – claim 13 is for products susceptible of industrial application.

Mr Waugh’s last reinforcement point, the technical contribution point, was supported by reference to Salk, T0609/02 and Exxon, T0409/91. But for the reasons already apparent, it fails: the technical contribution here is the limited one of a product which binds to neutrokine-α.

Art. 123(3) of the EPC provides that:

The European patent may not be amended in such a way as to extend the protection it confers

The UK legislation provides, by s.72(1)(e) for a ground of revocation on that basis.

By Art. 69(1) the EPC as amended in 2000 provides:

The extent of the protection conferred by a European patent or a European patent application shall be determined by the claims. Nevertheless, the description and drawings shall be used to interpret the claims. 

It is unnecessary here to consider the effect of the Protocol to Art. 69.

So to decide whether there has been an extension of protection by amendment the court must construe the claim before and after amendment. If, following that comparison, the court discerns that the amended claim provides more protection than the unamended claim, then the objection is made out. [I here do not consider the case where the claim concerned is a subsidiary claim whose protection is extended but not beyond that of a wider claim which remains unamended. There it may be that the protection of the patent as a whole is not extended.]

In construing the amended claim it must be legitimate to take into account the unamended claim, for the whole exercise depends on a comparison of the amended and unamended patent. I am not confident that it is also legitimate to take into account what passed between the patentee and the patent office. The subject of the use of the patent office file as an aid to construe a patent (or for other purposes, such as admissions of fact) is well-known to be controversial, with the German courts firmly saying no (as least as an aid to construction) and the Dutch Courts firmly saying yes. The Americans have an elaborate doctrine of prosecution history estoppel. The UK courts have yet to decide conclusively the matter one way or the other. It is unnecessary to go into it here.

The Judge mentioned the reason apparent from the prosecution file as to why the amendment was made, see [139]. However, I am not convinced he based his reasoning on that, for he said, correctly, “the answer lies in the wording of claim 1 in its original form”. But even if he did, it is unnecessary to decide whether he was right, for the reason is in any event obvious on a comparison of the unamended and amended specifications. I turn to consider the point in more detail.

I set out the claim so that the amendments can be seen:

Mr Waugh contends this was widening the claim: that it no longer includes the requirement that the antibody binds to the neutrokine-α portion of a Neutrokine-α polypeptide. It now is enough that it binds to (a) or (b), the full length polypeptide or its extra-cellular domain.

Whether that is so depends on what is meant by the neutrokine-α portion of a neutrokine-α polypeptide in the unamended claim. Did it mean something less than the neutrokine-α polypeptide itself? Only a part of the full length sequence of amino acids of neutrokine-α? If it did then there is widening: the claim would now include antibodies which bind to any part of the neutrokine-α polypeptide as opposed to those which bind to a lesser part.

When one goes to the unamended patent one can see what was meant by neutrokine-α portion there. The original claim required that the antibody should bind to the neutrokine-α portion of a polypeptide which could include an amino acid sequence different from that of neutrokine-α, either as an addition to it or within it. For the unamended claim 1 included not only (a) and (b) as set out in the amended claim but also polynucleotide sequences encoding for other polypeptides which included rather than consisted of neutrokine-α amino acid sequences (alternatives (c)-(f) which it is unnecessary to set out here). In that context the neutrokine-α portion of such a polypeptide is manifestly that bit of the whole corresponding to the neutrokine-α sequence. Binding to any part of that latter sequence was enough. Now that claim 1 has been limited to the nucleic acid sequences encoding for neutrokine-α or its extra-cellular domain, there is no need to refer to a portion of a longer or different sequence. So the deletion of “portion” involves no change of meaning.

Mr Waugh sought to persuade us otherwise. He said there were three reasons why the patentee must have meant only some smaller part of the neutrokine-α amino acid sequence when he used the phrase neutrokine-α portion of a neutrokine-α polypeptide in the unamended claim.

First he submitted that the reader would understand “portion” to be “a smaller part of a bigger bit.” Acontextually that is of course true. And so also in the context of the unamended patent – the “bigger bit” is a polypeptide which includes but does not consist solely of neutrokine-α sequence.

Mr Waugh submitted that the skilled reader would know that there are bits of the neutrokine-α polypeptide which matter and that so the patentee could only have had in mind those bits when he spoke of “a portion.” He sought to derive support from this from Professor Saklatvala’s evidence:

However at no point does the Patent explain what the “Neutrokine-α portion” of the polypeptide is. It is an expression only to be found in the claims and it is unclear what this [is] intended to convey to the reader. In other words, the reader may deduce from reading the Patent that there is an important part of the polypeptide to which antibodies must specifically bind but, on the other hand, the Patent does not identify where the important part is or how to assess when an antibody has bound to it.

I do not accept that this supports Mr Waugh’s construction at all. On the contrary, if one is to read the patent as it would be read by a rational skilled man, it is against it. The patent does not define the phrase, as the Professor points out. Nor does it say how the important bits of sequence are to be identified. So it is wholly unlikely that patentee was, by the word portion, limiting his claim to bits which he had not identified or told the reader how to identify.

Next, and it is really the same point, Mr Waugh took us to a number of passages in the Patent which say that there is or are a bit or bits of the neutrokine-α sequence which matter – the bit(s) that provide activity. The difficulty with that is that whilst the skilled reader would learn that (actually he would almost certainly know it already) from the patent, the patent at its high level of generality is far from concerned with identifying those bits. If the patentee had meant by “portion” to mean “portion which provides activity” he would surely not only have said so but told the reader how to identify them.

Finally Mr Waugh drew our attention to original claim 7 (a point not raised before the Judge). The point ran thus:

This argument is hopeless. It overlooks the important fact that “portion” does have a sensible meaning in relation to polypeptides which include but are not limited to neutrokine-α sequence amino acids. True it is that “portion” would not make much sense if the original claim had been limited to neutrokine-α sequence amino acids, but it is not. So “portion” has a perfectly intelligible meaning in the context of the claim as whole. This is a lawyer’s quibble and no more.

The extended matter appeal fails.

The Judge held these (which it was agreed before us stood or fell together) insufficient. He said:

[257] Overall I was left with the clear impression from the expert evidence that anyone seeking to develop a candidate antibody to Neutrokine-a for any therapeutic or diagnostic application based upon the teaching of the Patent would have been faced with a substantial research programme with an uncertain outcome. Depending upon his particular interests, the skilled person might choose to follow up an aspect of the teaching which would involve years of research or, even worse, lead him into a blind alley. Even today, the full range of activities of Neutrokine-a is still the subject of study.

And

[259 In the light of all these matters I have reached the conclusion that it would have required a research programme and been far from routine for the skilled person to produce a candidate pharmaceutical or diagnostic composition comprising an antibody to Neutrokine-a, that is to say the pharmaceutical or diagnostic equivalent of a workable prototype, on the basis of the information contained in the Patent and the common general knowledge. Indeed, such a project may have failed altogether depending on the route the skilled person chose to take. In my judgment proposed claims 20 and 21 are therefore insufficient.

[230] I would add that although both claims are directed to a composition they are extremely broad because the application of the composition is not specified. They extend to compositions for treating or diagnosing any of the many different conditions discussed in the specification. It follows from my findings that the specification is wholly insufficient to allow the inventions of these claims to be performed over their whole scope.

The Judge was not invited to construe the claims and did not expressly do so. He simply proceeded on the basis that they were limited to the pharmaceutical or diagnostic equivalent of a “workable prototype” (a phrase taken from Mentor v Hollister, [1993] RPC 7). If that were right, then the claims would indeed be insufficient on his findings of fact. But before us Mr Thorley contended that the Judge had proceeded on an erroneous construction. He submitted that read in the context of the specification as a whole the skilled reader would not expect the patentee to have intended these claims to be directed to compositions with immediate practical use as a pharmaceutical or diagnostic. On the contrary he would know that no such compositions had been disclosed and that what the patentee had discovered and disclosed is neutrokine-a and its antibodies with a practical use for these purposes yet to discovered So there is no reason to suppose that in these claims the patentee intended any specific application for the claimed compositions. They are not tied to any particular application. It follows that all he must have meant is compositions which could be formulated as suitable for administration as a pharmaceutical or suitable for use as a diagnostic. That could be done and so the claims are sufficient.

I accept that submission. It is in accordance with the principles of claim construction laid down in Kirin-Amgen [2005] RPC 9. The contrary view is not, involving as it does the skilled reader in ignoring the very general high level nature of this invention.

I would accordingly dismiss Lilly’s appeal concerning the alleged insufficiency or extended protection of claim 13 and allow HGS’ appeal concerning the sufficiency of claims 18 and 19. That result is in accordance with the decision of the EPO Board of Appeal to which it is unnecessary to refer. I would add that I also specifically agree with the judgment of Lewison LJ.

I agree with the judgment of Sir Robin Jacob. I add a few observations of my own in deference to the arguments we have heard.

The principal criteria for a valid patent are that it is granted for an invention that satisfies the following conditions:

Whether an invention is capable of industrial application involves asking the questions: what is it for; and what are the chances that it will work? In the case of an invention consisting of a product, whether the specification discloses it clearly and completely involves asking the question: can you make it? The difference between the two questions also emerges from the jurisprudence of the Technical Board of Appeal. In Case T 0939/92 (AgrEvo) the claim was a claim to chemical compounds. The description alleged that the claimed compounds had herbicidal activity; but this biological function was not part of the claim to the compounds themselves. As the TBA put it: “the present independent claim covers certain chemical compounds per se, and not just those having a particular biological activity”. Having thus construed the claim, the TBA went on to consider whether the specification was sufficient. They said:

“In the present case, it follows from the considerations contained in point 5 of the decision under appeal that the examining division had no doubts as to the possibility of preparing the claimed compounds. Furthermore, the examining division did not find that the description mentioned technical features as being essential features of the claimed invention which were not part of the definition of the present claim 1, nor could the board find any such feature. Instead, the examining division relied upon the fact that a skilled person upon reading the application documents would not have believed that all claimed compounds would or could be likely to possess the alleged herbicidal activity, which feature is, as already stated, not part of the definition of the subject-matter for which claim 1 seeks protection. Therefore, the facts of the present case differ from those underlying decision T 409/91, so that an objection of lack of support by the description cannot, in the board's judgment, be validly raised in the present case.”

However, the patent was refused because of a lack of inventive step; or as we usually call it “obviousness”.

When this case was last before this court we unanimously decided (in agreement with Kitchin J) that the claims were not capable (or susceptible) of industrial application: [2010] EWCA Civ 33, [2010] RPC 14. However, HGS appealed to the Supreme Court which allowed the appeal: [2011] UKSC 51 [2012] RPC 6. The Supreme Court held that the claims were capable of industrial application; and that claim 1 was sufficient. Remaining questions were remitted to this court. We must apply the law as the Supreme Court has laid it down.

Thus in our case the Supreme Court has answered the question: what is it for and what are the chances that it will work? They have said that there is a good enough chance that it will work. In my judgment, therefore, the only remaining question for us is: can you make it? (see H Lundbeck A/S v Generics (UK) Ltd [2008] EWCA Civ 311 [2008] RPC 19 § 52 per Jacob LJ). Obviousness is not one of the pleaded attacks on the patent pursued on this appeal. Obviousness in the AgrEvo sense was raised, but abandoned in the Supreme Court. So that is not an argument that is now open to Lilly.

Of course, you have to decide what “it” is that the patent must teach you how to make. In some cases the court has held that the specification must teach the skilled person how to make apparatus that will work (Valensi v British Radio Corporation [1973] RPC 337 per Buckley LJ) or a “workable prototype” (Mentor Corporation v Hollister Inc [1993] RPC 7 per Lloyd LJ). These glosses are fine as far as they go, and may be entirely apposite where the invention consists of a device suitable for a defined function; but they are no substitute for the statutory test.

As Lord Hoffmann pointed out in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2004] UKHL 46 [2005] RPC 9 § 104 an argument about sufficiency is often an argument about what the invention really is. We find out what the invention is by looking at the claim in question, and by interpreting it in the context of the description: Patents Act 1977 s. 125 (1). Thus as in Case T 0939/92 (AgrEvo) we must first construe the claim.

Claim 13 claims an isolated antibody that binds specifically to Neutrokine-α. The judge held that it did not require undue effort to make and identify specific antibodies that bound specifically to Neutrokine-α (§ 250). There is no appeal against that finding of fact. However, Lilly argue that, read in the context of the whole of the specification, this is a claim to useful antibodies. Lilly’s alternative way of putting this point was that claim 13 covered both useful antibodies and useless ones; and that there was no way of telling them apart without undue effort. The judge rejected Lilly’s argument. He said that all that was claimed was antibodies that bind specifically to Neutrokine-α. Whether they had therapeutic or diagnostic properties was not part of this claim. So it did not matter that you could not distinguish the useful antibodies from the useless ones. This is a question of construction of the claim. Mr Waugh accepted that he was asking us to read the word “useful” into the claims. But in the Supreme Court Lord Neuberger firmly set his face against reading words into claims (§ 137). Moreover, despite Mr Waugh’s valiant efforts, I was not able to see why claim 13 (which claimed the antibodies) was conceptually different to claim 1 (which claimed the polynucleotides encoding the protein) which the Supreme Court upheld both on industrial applicability and sufficiency. Mr Waugh said that claim 1 claimed a “thing” whereas claim 13 claimed a huge genus of diverse antibodies with no unifying principle. He said that if you claim a broad class which includes a functional characteristic the invention is only sufficiently described if a skilled person can reasonably expect that substantially all members of the class can be made and put into practice on the basis of what the patent teaches, combined with common general knowledge. The judge found as a fact that all members of the class could be made. But I cannot see that there is an additional requirement that, once made, all or substantially all members of the class can be “put into practice”. That would only be the case if the claim specified some use to which the claimed product had to be put. In this case the judge held that, as a matter of construction, claim 13 did not contain a use limitation.

Mr Waugh relied on the decision of this court in Pharmacia Corp v Merck & Co Inc [2001] EWCA Civ 1610 [2002] RPC 41. The claim in that case was a claim to a class of compounds. Read literally the claim did not require the class of compounds to have any particular characteristic. But the court held that in the context of the specification read as a whole no one reading the specification could believe that the invention was simply the compounds claimed in claim 1. The specification made it clear that the patentees had invented a class of compounds that could be made which at least had anti-inflammatory action. It was in the light of that construction of the specification in that case that Aldous LJ went on to say (§ 56):

“Where the claimed invention is to a class of compounds … the disclosure in the specification must enable the invention to be performed to the full extent of the monopoly claimed. Thus if the invention is a selection of certain compounds, in order to secure an advantage or avoid some disadvantage, not only must the specification contain sufficient information on how to make the compounds, it must also describe the advantage or how to avoid the disadvantage. Further the compounds monopolised by the claim must all have that advantage or avoid the disadvantage. The same principle applies where the claim is to a class of compounds. To be sufficient, the specification must identify the characteristics of the class and a method of manufacture. Further all the claimed compounds must in substance have the characteristics of the class.”

The characteristic of the class in that case, as a matter of construction of the specification, was that they had to have at least anti-inflammatory action. The characteristic of the class in our case is that the antibodies must bind specifically to Neutrokine-α. I do not consider that Pharmacia carries Mr Waugh’s argument any further.

I therefore agree with the judge’s interpretation of the claim. The judge decided that you could make what the claim claimed without undue effort and without invention. That was a finding of fact which cannot be challenged. In my judgment, therefore, the judge was right to hold that claim 13 was not insufficient.

Lilly have a separate point about claim 13. It was amended (as was claim 1) during the course of prosecution. They say that the amendment to claim 13 has impermissibly widened the scope of the monopoly it claims. The judge rejected this argument (§§ 138–140). He held, in effect, that the amendment to claim 13 was consequential on an amendment to claim 1 during the course of prosecution. The amendment to claim 1, far from widening the claimed monopoly, in fact narrowed it. So the consequential amendment did not widen the claimed monopoly. This is a one-off point confined to the particular claims in this case. I need say no more than that I agree with Sir Robin Jacob and the judge that there is nothing in this point.

Claim 18 claims a “pharmaceutical composition” containing Neutrokine-α or an antibody to Neutrokine-α and “optionally a pharmaceutically acceptable carrier”. In the case of a protein a pharmaceutically acceptable carrier is likely to be an inert fluid that can be injected into a human. Claim 19 claims a “diagnostic composition” containing the same. The issue again is: what is “it” that the patent must teach you how to make? If the question is: can you put Neutrokine-α (or an antibody to Neutrokine-α) into a pharmaceutically acceptable carrier, so that you could inject it into a human or an animal, the answer must be “yes”. But if the question is: do you know what it would do to a human or animal if you did inject it, then the answer is: “no”. Likewise in the case of a diagnostic composition.

Mr Waugh submitted that these claims should be limited to useful products either:

So far as the policy questions are concerned, they were the subject of the debate in the Supreme Court. Having lost the policy argument there, I do not consider that Lilly is entitled to resurrect it in dealing with sufficiency. Accordingly, in my judgment once again it is a question of identifying the invention as a matter of construction of the claims.

So as I see it, the real question is: what significance, if any, is to be attributed to the adjectives “pharmaceutical” and “diagnostic”? Lilly argues that the claims are not merely claims to compositions but to “pharmaceutical” and “diagnostic” compositions. One cannot describe a composition as a “pharmaceutical composition” unless it has some utility as such. In this context “pharmaceutical” means suitable for use as a medicine for the treatment of a disorder. A similar approach applies to a “diagnostic composition”. These adjectives require the compositions to “be of use as such”. You cannot call something “diagnostic” unless you know what it is that you want to diagnose. They point to the decision of the Technical Board of Appeal in upholding the validity of this very patent in which the TBA said:

“… the use of antibodies in the preparation of pharmaceutical or diagnostic compositions may imply the identification of a condition or disease where that protein is involved (which is also true for the compositions themselves).”

In fact, despite this passage in their decision, the TBA went on to decide that neither claim was insufficient, because of the high level of generality at which these claims were pitched. Although the judge did not in terms address this question of construction of these claims, in essence he accepted Lilly’s argument (although without reference to the TBA which had not then reached its decision). His reason for holding that these claims were insufficient was that it would have required a research programme and been far from routine “to produce a candidate pharmaceutical or diagnostic composition comprising an antibody to Neutrokine-α, that is to say the workable pharmaceutical or diagnostic equivalent of a workable prototype” on the basis of what was disclosed in the patent supplemented by common general knowledge (§ 259). He added that although both claims were directed to a composition, the “application of the composition is not specified” (§ 260). In so far as these are findings of fact they are unimpeachable. However, underlying these findings is the tacit conclusion that, as a matter of construction of the claim, the pharmaceutical or diagnostic composition had to be “workable”. I infer that the question whether the pharmaceutical or diagnostic composition had to be workable was an echo of Valensi v British Radio Corporation and Mentor Corporation v Hollister Inc to both of which the judge had referred earlier in his judgment in setting out the law.

HGS argues that the judge’s findings are based on the assumption that the claims did claim a medical use. In fact they do not. They claim products. Since claim 1 (to a compound) has been upheld by the Supreme Court both on industrial applicability and sufficiency, and claim 13 (to an antibody) must also be upheld, it must follow that the patentee is entitled to claim a product (manufactured without undue effort and without invention) containing the compounds covered by claims 1 and 13. They, too, point to the decision of the TBA which held that claims 18 and 19 were sufficient. As long as the product is suitable for pharmaceutical or diagnostic use, in the sense that it can be put into a suitable carrier, the claim is sufficient. It does not matter that the claim does not specify a particular condition or disease that such a product could treat or diagnose. It is open to pharmaceutical companies downstream to find specific medical uses for such compositions (in which case a “Swiss-type” patent for a second medical use could be granted); or to identify specific antibodies which have particular properties.

I prefer the submissions of HGS. In my judgment claims 18 and 19 claim products, not uses. There is no difficulty in making the products. The difficulty is in knowing which of the products would be worthwhile introducing into a human or animal body; and in what circumstances. But that, as I see it, is part of the question: is there a good enough chance that it will work? That question has been answered affirmatively by the Supreme Court. In my judgment the phrase “a pharmaceutical composition” is, as HGS submits, the common way of claiming ingredient X in packaged form. It does not promise any particular effect. I think that the same is true for “a diagnostic composition” although (despite the fact that neither party sought to differentiate between these two claims) there is, perhaps, more room for argument. This construction gains added force from a reading of the specification as a whole. It is clear from the specification that the patentee had no real idea what Neutrokine-α or its antibodies would do if introduced into a living creature. It is highly improbable that by using the words “pharmaceutical” and “diagnostic” in that context the patentee would have been understood to mean that the products claimed by these claims had an identifiable therapeutic use.

In addition the Supreme Court dealt in terms with the judge’s conclusion as expressed in paragraph 259 of his judgment. The judge was dealing with claims 18 and 19 at this point in his judgment; but the Supreme Court considered this passage in considering the sufficiency of claim 1. Lord Neuberger said (§ 133):

“Although the Court of Appeal did not consider this point, Jacob LJ did say at the end of his judgment, that he “rather suspect[ed]” that the insufficiency argument “would go hand-in-hand with Article 57” – [2010] RPC 29, para 159. Subject to one point, which turns on the meaning of Claim 1 (as well as some of the other claims), it seems to me that that must be correct. If Claim 1 is simply to the encoding gene of Neutrokine-α, then, subject to any other points which have yet to be decided by the Court of Appeal, the reason why I consider the Judge and the Court of Appeal were wrong to hold that Article 57 is not satisfied is the same reason for holding the claim to be sufficient.”

Lord Neuberger went on to say that there is a close connection “indeed overlap” between susceptibility of industrial application and sufficiency (§ 134). As I read it, what he was saying was that a conclusion on susceptibility of industrial application entailed a similar conclusion on the question of sufficiency. That conclusion was in favour of upholding the patent. Since the Supreme Court have held that the judge’s conclusion in paragraph 259 of his judgment did not render claim 1 insufficient it would, in my judgment, be extraordinary if the very same conclusion made claims 18 and 19 insufficient.

For these reasons in addition to the reasons given by Sir Robin Jacob I, too, would allow HGS’ appeal and dismiss Lilly’s appeal on the issues that have been remitted to us.

I agree with both judgments.