Oral argument — Takeda Chemical v. Alphapharm Pty

The Federal Circuit is now open and in session. John Sabin of the United States and his Honorable Court. Please be seated. Good afternoon ladies and gentlemen. We have a surprisingly large crowd. We had three cases on the calendar, and maybe more than that actually, last week. But through death in the family that others of you know more about than I do, we accommodated, out of respect for the situation, to reschedule. And in fact for the veterans case scheduled this afternoon, that was postponed from last week also because there was a death in the family, that couldn't be rescheduled for today. And we wanted to give you the chance to get out of here before you get iced in. So I hope no one really objects to the unexpected hearing this afternoon. In any event, we have two related cases. Takeda Chemical v. Alphapharm, 06-1329, Mr. Murphy. Thank you. And let me say, the Court's closed now. And we are grateful to have staff helping us out, and loyal and devoted law clerks. So please proceed. May it please the Court, good afternoon, Your Honors. This report made many errors within the scope of its opinion, but I'd like to focus my commentary today in the argument on three of the most serious of those errors. The District Court's error with respect to the scope and content of the prior art by carving out from the prior art a Takeda patent prosecution history where Takeda made very damaging admissions to its position with respect to the predictability that flows out of the 6-methyl and the motivation that flows out of the 6-methyl closest prior art compound. Number two, the District Court's decision to allow the patentee to buttress its case of unexpected results by allowing post-patent filing evidence and preventing the challenger, Alphapharm, from rebutting the case of unexpected results because it was post-filing. We view that as an abuse of discretion. And finally, I'd like to bring to Your Honor's attention the errors of law with respect to the analysis of obviousness requiring of Alphapharm evidence of absolute predictability in an unpredictable art and the claim that there are no reasonable expectations that any medicinal chemist can have, but they would sit and be paralyzed rather than undertake a research program, particularly one with a clear motivation shown in this art. First, the scope and content of the prior art. Could I understand what you're arguing? As I understand it, we have claims 1, 2, and 5 here. And if I understand correctly, claim 2 is to the 5-ethyl, but that claims 1 and 5 would also cover the 6-ethyl, correct? That's right. That's exactly right. Are you making a separate argument that claims 1 and 5 are obvious, let's say, over the earlier 200 patent? No. Everything is obvious over the 6-methyl compound. And the 6-methyl compound was disclosed in the 200, not particularly claimed in the 200. There are many millions of compounds. What we're essentially saying is that Takeda claimed a compound as a selection invention from the 779 patent that was earlier disclosed and explained in Soda 2. Just so that I can understand, are you then saying that in your view claims 1, 2, and 5, the obviousness of those claims, depends in each instance on the 5-ethyls being obvious? The position is that claims 1, 2, and 5 are all obvious and that the 6-ethyl is slightly more obvious than the 5-ethyl. Well, but that's the problem I'm having because 1 and 5 cover the 6-ethyl, right? That's right. Claim 2 is limited to the 5-ethyl. That's correct. So conceivably, we could determine that claims 1 and 5 are invalid but the 2 isn't. That's right. That's not the relief we're asking for here. What we're saying is that they're both obvious. Homiligation and ring walking were well known and well practiced in the prior work and there's every motivation to do what is being claimed here. That is taking the 6-ethyl and then moving it to the next available position of the ring to preserve potency, the potency of the 6-methyl while tweaking its side effects. But weren't the tweaking, I mean, tweaking is, this isn't a valve. Wasn't there a demotivation to start working from that closest prior art, 6-methyl? I mean, it had toxic effects. It was less desirable than pioglitazone. Only if you cut off the prior art analysis, Your Honor, of 1982. So what the district court had claimed was that there was a teaching away for all the prior art if you focus on 1982. But the district court, to find that, had to carve out of the prior art clear evidence of motivation within the 779 patent and the 779 patent prosecution history. So any demotivation was gone by 1984 when the 779 patent issued. The 779 patent specifically claimed the 6-methyl. The 779 patent taught that the class of substituents to use on the pyridyl ring on the left-hand side of the molecule were methyls, ethyls, etc. But in the final analysis, didn't pioglitazone have superior properties and unexpectedly superior properties? And wasn't that a finding of the district court? The district court did find so, but the district court had to strike from evidence and had to claim as irrelevant evidence the evidence that showed that the claim by Takeda of unexpected results over the closest prior art was not so. What Takeda claimed to the examiner, and it may have been true at the time, was that pioglitazone was unexpectedly superior in three measures. One was heart weight, one was liver weight, and one was anemia. These were on the basis of preliminary screens, two-week studies. What AlphaForm found out in discovery was that Takeda did more robust studies a little later that said that heart weight and anemia were not actual toxic effects, that these are related, in fact, to the... So you're saying they developed and are selling as one of these highly successful compounds the wrong compound? No, not at all. Simply put, it's a very conservative position. It doesn't have to be any more than this, that Takeda went forward to the examiner and told the examiner, we have unexpected results here in three measures. Then they did more robust testing. They did testing at four weeks and 13 weeks and found out that actually it's a pharmacological effect, non-toxic. Heart weight and anemia were non-toxic. So two-thirds of the unexpected results case is basically thrown out. Now you're talking about mere differences of degree, not kind. You're not talking about patentability anymore, and it's based on Takeda's own admissions in its more robust testing. That's what the issues are here, and that's why it was so harmful to strike out those particular tests. It's almost as if... It's exactly akin to someone who... A doctor, for example, who sees patients over two weeks and he gives them high doses of acetaminophen. They have liver cancer. And he notices that there are tumors that shrink in those particular patients. Now he runs to the patent office because he thinks he has something patentable. He says, well, we have unexpected results here. Look at this. Acetaminophen. It's a new property. It can treat liver cancer. Then he's intrigued, and he does longer-term studies, and he finds out through the course of those studies that, no, there are confounding factors and the claims to unexpected results are not true. They're false. So what the district court did was say, no, no, no, no patentee can be subject to a challenge later on from post-patent filing evidence that the results were not unexpected. That's just wrong. It's a matter of law. It's wrong. It's a matter of evidence law. It's wrong. It's a matter of policy. It's wrong. The patentability cuts both ways, and unexpected results cuts both ways. Now, with respect to the district court's finding of obviousness... finding of non-obviousness, the district court took the position, based on Takeda's own... Takeda's expert, Dr. Daniszewski, that when you make a change to a particular portion of the left-hand side of this molecule, you have no idea what's going to happen in this medicinal chemistry. Absolutely not. So nothing can ever be obvious. Absolutely none? Well, any variability on the left-hand side of the molecule will... any varying of the substituent on the left-hand side of the molecule will be unpredictable. So there's no reason that you can have an expectation for any particular result. That's ultimately what he said, and that's ultimately what the district court found. Now, that obviously has a great deal of effect in the obviousness law on chemical obviousness. There can be no chemical obviousness if there are no reasonable expectations. If the skilled artisan has to know in advance that a particular modification will have a particular result. So in effect, it's requiring of the challenger to show absolute predictability, to be able to look through a crystal ball and know in advance what's going to happen. When in reality, what medicinal chemists do is they look at lead compounds and they see where the predictability is, and to Katie Heer... Katie Heer made admissions in the 777 patent application, and it made admissions in the 779. It said in the 779 that there's no reason to expect that this compound, the 6-methyl, would serve as a blood glucose-lowering, a triglyceride-lowering agent with low toxicity. It talked about prediction there. So there, to get that patent, to cover its own research program on substituted puritals from competition, it alleged predictability on the basis of the 6-methyl. The same can be found in the 777 patent, where they approached the examiner and they told the examiner, look at this data on the 5-ethyl. The 5-ethyl shows a superior toxicity result relative to the 6-methyl. And they claim, claims 1 and 5, claim the 3 position, the 4 position, and the 6 position. So they're essentially telling the skilled artisan that they have every reason to predict that the safety profile is going to be better there as well. But their trial position was contrary to that. Their trial position was that anything, any change, is going to have unpredictable consequences. We believe that you should hold them to that position, hold them to the position that they took to secure the 779 and the 777 patents. Now, with respect to the harm, let me focus on now the scope and content of the prior article. The district court made a specific finding as to a Takeda prosecution history. That finding was as follows. The file record for the, and you'll find this on page 78 of the opinion at footnote 59, the file record for the 779 patent was simply not accessible to one of ordinary skilled in the art. Although it was available to the public, now I'm paraphrasing, although it was available to the public in 1984, more than a year, by the way, before the 777 patent was filed for, the file history would have had to be specially obtained at the patent office. So... Let's suppose she's wrong about that. Is it really harmful error? Certainly it is. Certainly it is. The file history for the 779 patent gave clear evidence of motivation to the skilled person, clear evidence that there was no teaching away in SOTA 2. What SOTA 2 says is that there are problems of brown fat and weight gain associated with this content. If that's unlikely to be productive of the results sought, then why is Takeda specifically claiming this six-month rule two years later? Why is Takeda specifically telling the examiner to that patent that this compound was especially important? Why is Takeda telling the examiner that you can predict as long as you use the narrow and well-defined range of constituents, the substituents to use on this left-hand side period of range? So there are a lot of admissions as to predictability that are inconsistent with the district court's ruling, and there are a lot of admissions as to motivation, and there's real harm here. There can be no finding of teaching away under these circumstances. I see that my time is up. I just want to make sure I've answered your question, Your Honor. Good deal. We'll save the rest of your time, Mr. Murphy. Thank you, Your Honor. Mr. Conlon. This was pretty close prior art, wasn't it? Structurally? Structurally, it was fairly close. Structurally with a similar activity, right? Hypoglycemic. It was glycemic. They found that to be the case. But in truth, however, it's eons away. You know, when you're trying to treat diabetes, toxicity is really, really important. People are going to be taking these medicines every day for the rest of their lives. So toxicity is king here. And the question is, really, is this an unexpected, is this an unobvious invention? This is a wonderful invention. And I think it's shown by many different aspects. And the district court's opinion, really the only question you have is, was there clear error shown in her holding that they didn't prove obviousness by clear and convincing evidence? And her holdings show intimate knowledge of the detail. That opinion was written by her. It was not an adoption of counsel's recommendation. She knew. And if you see how she ran the trial, you know. I don't see her addressing the 6-ethyl. Isn't it clear that claims 1 and 5 are in balance? Put aside 2 for the moment, and 1 and 5 may not be the focus. But claims 1 and 5 cover the 6-ethyl, don't they? They do indeed, Your Honor. And the court did realize that. Those claims are invalid if the 6-ethyl didn't have unexpected results. I think that the question is whether they've proved it to be invalid. But my statement's correct, right? That those claims are invalid unless the 6-ethyl has unexpected results, right? Well, I believe that they haven't been proved whether or not. I think your point is well taken, and it's much closer. But the point that the judge relied on. And she didn't even make a separate finding about the 6-ethyl, did she? She found at one point that there was no predictability of success of going from the methyl to the ethyl. But she doesn't find any unexpected results with respect to the 6-ethyl, does she? No, she did not find any unexpected results. And, you know, I think these claims don't rise or fall together. I think we're really interested primarily in claim 2. That's the one that covers the compounds on the market. There's no question about it. The appeal does apply to all three, correct? Of course. And I don't think that they've proven obviousness even of the 6-methyl. I think there's, you know, we don't know. They say that we took the position that these are only because of closeness, of structural closeness. Our inventors were available. They could have asked these questions to them. They had a lot more information in 1985 when they wrote the 7-7-7 case than they had in 1979 when they wrote the 7-7-9 application. Do you think for the generic claim to be invalid on grounds of obviousness, every species thereof must have unexpectedly beneficial properties over the closest priority? I believe that for them to be proven invalid, they have to be a showing that you would go there and that you would expect success in going there. I think that's missing now. Having said that, I believe that. What Judge Lurie is asking, I think, is if one of the claim compounds in claims 1 and 5 are obvious. If that's obvious, then the claim is invalid. Then the claim is invalid. Absolutely. Of course, a generic claim can have thousands, millions of compounds in it, and it's absurd to think that all of them have to be shown to be non-obvious. That's true. Here you have, what, four compounds. Four compounds. Four compounds. Why did the inventors believe that they were all patently distinct over the prior art? You know, if you look at what they did, they could have tried to claim much broader. There were some pretty unexpected results about the 5-methyl as well, but they were distinct. They only made a claim to the ethyl compound. They believed that the ethyls were different. We don't know why. They weren't asked. So you're saying if the generic claim is held to be invalid because the 6-ethyl has not been shown to be unexpectedly superior to the prior art 5-methyl, that's irrelevant to the validity of the species claim 5-ethyl? Absolutely. That's the law. Each claim is presumed to be independently valid, and claim 2 is independently valid beyond question over claim 6. But we have a determination here by the district court that claims 1 and 5 are infringed. That's a declaratory judgment. The judgment is incorrect, right? Is that true? I think that would be true, yes. But I think the court did not hold the claims invalid, and the reason is simply that the burden was on them to prove they were invalid. But I don't see that she really addressed the validity of claims 1 and 5 in the terms that we've been talking about them here. She did not specifically rule on whether they're shown unexpected results or anything like that, but she did hold that there was no prima facie case of obviousness. And that being the case, I think the claims are sustainable. But, again, I think we're primarily focusing on claim 2, which is the claim of importance in the case. So I think to get an impression, Mr. Murphy talked about one approach that he has taken is to say that the court precluded them from relying on later evidence as to whether there was obviousness or unobviousness. I think if you look at the portions of the record cited in that brief, all she did was rule on the admissibility of two documents that were not shown to be admissible. She really didn't take that position. And he said it may have been true at the time that this was unexpected, but in 1992 or 1999, it might have been expected. I think by saying that it may have been true at the time that it was unexpected, he said it's a valid patent because the obviousness is tested as of the time of the invention, under section 103. This is an incredible invention. I would like to show you one document that is from A4692. Takeda, as the judge pointed out, tested thousands of compounds. Thousands of compounds. And then in 1984, they chose the 50 best compounds to be tested because Takeda and Nupjohn were working on trying to determine where can we go forward? What kind of a species, what kind of a compound can we go forward with? And of the thousands of compounds, they chose the best 50. Is this what you're showing us? That's correct, Your Honor. And of the best 50, the only thing I would say is Dr. Fujita, who was at this meeting, and Dr. Morton, who was at the meeting, they testified in this case. And their testimony is particularly telling that Dr. Morton and Dr. Polka from Nupjohn came as volunteers and sat for a couple of weeks waiting to testify on their own dime with no connection to Takeda or anything else. They tell you what it was like in 1984 and what they were considering at that time. And this was incredibly important to them, and it was surprising to them. It was something that you cannot say this is not unexpected results. So in terms of whether or not the... What's the evidence of unexpected results for the 6-ethyl? The 6-ethyl, there is no direct evidence of unexpected results. And the 6-ethyl is covered by the 200 patent, right? The 200 patent covers generically the 6-ethyl as well. Yes, that's right, Your Honor. That's correct. It does, along with about another million compounds that are generically covered. Is it disclosed? 6-ethyl is not disclosed, specifically disclosed, no. It's only generically within the scope of the claim. So in this case, there is... The other point that he tried to make was that we should be saddled to our position in the 7-7-9 patent and saddled to our position in the 7-7-7 patent with regard to generic coverage. I'd just like to make a couple of comments on that. In the 7-7-9 patent, if you look at what the attorney actually said, all he said was that there's no reason to doubt that these would be functional. There was a 112 enablement requirement rejection, and that's all you really need to expect. In the 7-7-9 patent, no question was raised and no statement was made by us, so there's no reason to tie or ticket any statements as to the scope of the claims. And in fact, if you look at the evidence that's actually in the 7-7-9, compare the methyl groups that are disclosed in there, three of which are not even in the prior art. We disclosed them for the interest of the examiner. That shows that there are significant differences between adjacent homologs and methyls and ethyl groups. So I think that we cannot be tied to any admission for any position in that case. The other point is that there is tremendous evidence of non-obviousness in this case. This drug has brought hope and equality of life to millions of people. It's the seventh fastest drug in the history of mankind, making a billion dollars a year in sales. So this thing is really, if there were even assuming that there were some evidence of obviousness, or that the court was wrong, that they did show that there was prima facie obviousness, that still would be way overweighed by the evidence of unexpected results and commercial success, long felt but unsolved need, and other evidence of non-obviousness that the court pointed out. Would you comment on the file record of the 7-7-9? Yes, I would. The court did have that footnote, which is a little bit confusing. She says at the beginning that it wasn't available, but she talks about it being available. But the court did not rule it was not prior art. In fact, she went on and considered it. Under the cases, what she has to do is look at the prior art as a whole. We sent a copy of the Ingray Lilly case from December this year to the court a week or so ago. It's clear that you have to look at the prior art as a whole, and that's what she did look at. But she made a mistake, did she not, by dismissing the prosecution history of the 7-7-9 patent, right? She didn't dismiss it, Your Honor. She did talk about it, and she said she considered it along with the teachings of the SOTA II article. She considered it in many places. I thought she said that someone skilled in the art wouldn't have looked at it, wouldn't have paid any attention to it. She said that looking at the article as a whole, they would be more led by the SOTA II article. That was the evidence, that was the testimony that was before her. She said it was not accessible to one of ordinary skill in the art, I think in the footnote. That's correct. Which certainly suggests that, in effect, it wasn't prior art. If you say it wasn't accessible, you're really saying it isn't something that's going to be considered as prior art and making a determination of whether that person, a hypothetical person of skill in the art, had material before him or her that would render it obvious. That's correct. That was an error, and it was a harmless error, because she went on and considered it as prior art despite that. If you look at page 880, she points out that one skill in the art would not have relied on the 779 reference to identify the compound B, given the more exhaustive and reliable scientific analysis presented by SOTA II. If you look at 889 of the opinion, she says, in any event, nothing in the 2779 patents or in any other publication from Takeda would have given an expectation of success in changing from the methyl to the ethyl. She did, in many cases throughout her opinion, go ahead and treat the 779 and treat it as part of the prior art as a whole, which is what she's required to do under the court's flexible standard of determining obviousness. So I think the other point that was argued in the brief is the question of whether or not it was a prime official case. And here, the prime official case never was made by the evidence. Once the evidence of unexpected results, once rebuttal testimony is given, the court quite correctly said, we have to begin all over again. We have to start fresh. That's the law of this court. And so doing that, she looked to see whether or not the evidence that they submitted at trial amounted to a prime official case of obviousness. She found that they failed in the proof to do that. Her findings are well supported by very substantial evidence in the record. And I believe that that is a good, solid basis on which this court should affirm the district court's opinion. This is really a classic case of a wonderful invention. It's the kind of invention that the patent system is designed to protect. And I think that the district court's decision is very meticulous and is detailed, very detailed. It's 120 pages long. It's hard to find anything that would have a minor mistake in it. But there's no harmful error that's been shown in this record. And the decision of the district court should be affirmed. Can I go back, if I could, just for a moment to the—I know neither party regards Claims 1 and 5 as the core issue in this case. But nonetheless, we have to deal with all of the claims that are in play. Is there any—if we were to invalidate Claims 1 and 5, hypothetically, would it have any practical impact on the parties or on the subject of this litigation? No, Your Honor. In fact, you'll see at one point in the trial, Mr. Murphy was arguing about Claim 1 and the 6th Ethel. And the judge said, well, what difference does that make? What does that get you? And Mr. Murphy in Canada said, well, not much, Your Honor. But wouldn't it make a difference on attorney's feet? I don't believe it would on attorney's feet. Didn't the judge find that the position of the defendants was frivolous? Yes, I think—and the reason— So if Claims 1 and 5 are invalid, the claims weren't frivolous, were they? I don't think that they've been shown to be valid. I think the case— The position—the hypothetical is that we assume that they are found to be invalid. And the question then is, does it have any effect? And the question more precisely, as Judge Dyke correctly says, how about attorney's feet? I don't believe it should affect attorney's feet. Would the prevailing party be the same? Pardon me? Would the prevailing party be the same? The prevailing party would still be the same. And the question is whether they went ahead. Well, you wouldn't be the prevailing party on Claims 1 and 5, would you? If we found that they were invalid, as obvious, you wouldn't be the prevailing party on Claims 1 and 5. I think a patent would be valid and would be infringed, and we would go forward. Well, patents aren't valid or invalid. Claims are valid or invalid. Yes, sir. You're quite right. But I believe we still would be the prevailing party, but the patents would remain valid. I think the issue before the House in that case is different than what is the details of Claims 1 and 6. I think the issue as far as the attorney's fees, which is a case for a different day, I think the attorney's fees is whether they acted in good faith in proceeding the way they did. But what we did here was going to affect the attorney's fees. The attorney's fees judgment was based on her view of this case. Was it? It was. Thank you, Mr. Conlon. Thank you. And thank you again for making the arrangements so that we could be heard today. We appreciate that. Mr. Murphy, if you have a generic claim, does every species within the generic claim have to be non-obvious for the prior art to be valid? It certainly has to be enabled, and there has to be some predictability with respect to enabling that generic claim. But you certainly can have compounds within a generic claim, particularly really very, very broad generic claims, where you can have non-obvious compounds, certainly. It's not what we have here, of course. What we have here is a 779 patent prosecution history saying that this particular compound, the 6-methyl, is especially important. We have a problem in identifying the prior art with taking an unsubstituted period, getting activity when you stick a substituent out of the 6, and then having some associated side effects, which tells the skilled person to move that substituent up and perhaps lengthen it. But if we find the 6-methyl is not motivating prior art, then not only would the 5-ethyl compound, p-oglitazone, be non-obvious, but so would the generic claim. Is that right? I believe if you find the 6-methyl non-motivational, that's correct. Certainly all the evidence in the case, however, if it's in our view, points to the 6-methyl, including the district court's error in our view that soda 2 tore away. It did not. It actually motivated toward the 6-methyl. With respect to the determination of clearly erroneous factual findings, the bottom line there is that obviousness is a de novo determination of law, and there are underlying factual findings underneath it. So you can find that the court misapplied the law, and as a result, reached the conclusion that the district court did. That's Alford Farms' view. With respect to commercial success, for example, with respect to long felt but unmet need, neither of those were shown. We've talked about that in our briefs. Commercial success simply is no validity where there's a prior art dominating patent. That's the 7-7-9 patent. No one could get into this market, so there's no way, when I say market, that is on substituted periods. There's simply no way that persons could try and fail within the prior art time frame. Nor was there any long felt but unmet need, because Resolin was marketed, and then Rosaglitazone was marketed, and immediately filled the need before Actos came onto the market. Finally, with respect to the 5-ethyl. The 5-ethyl was the next natural, logical step that you take when you have a compound as efficacious as this one, where there are minimal problems that you want to avoid. It's the next available step on the way, and we've shown that. We showed a primary case as a result of that, and we think that the patent is in doubt. Thank you, Mr. Murphy. We'll take the case under advisement. Thank you.