[00:00:00] Speaker 00:
The next case for argument is 24-1094 Teva Pharmaceuticals versus Eva Lily.

[00:00:09] Speaker 00:
We're ready when you are, Ms.

[00:00:12] Speaker 03:
Wright.

[00:00:13] Speaker 03:
Good morning, Your Honors, and may it please the Court.

[00:00:15] Speaker 03:
Your Honors, the district court found that the jury could have credited testimony that the disclosed murine anti-CGRP antagonist antibodies could be routinely humanized, and that a post-suit would know they could treat headache, and that a jury could have credited testimony

[00:00:29] Speaker 03:
that all humanized anti-CGRP antagonist antibodies will treat headache, but the court thought as a legal matter that it could not consider those antibodies for purposes of its written description or enablement analyses because they had not actually been humanized.

[00:00:43] Speaker 03:
The court then compounded that legal error by taking into account expert testimony that was impeached and contradicted of the losing party's expert witness.

[00:00:52] Speaker 00:
OK, can I interrupt you just because you've read all the briefs.

[00:00:56] Speaker 00:
Trust me.

[00:00:57] Speaker 00:
It's a little in the weeds, but I need to understand it, which is, can you walk me through the steps here for making one of these antibodies?

[00:01:08] Speaker 00:
Here's how I understand it could have been well understood

[00:01:12] Speaker 00:
as of the priority date.

[00:01:14] Speaker 00:
One, you object mice with CGRP.

[00:01:17] Speaker 00:
Two, you gather the antibodies the mice produce.

[00:01:22] Speaker 00:
Three, you screen them for binding and antagonizing ability.

[00:01:26] Speaker 00:
And four, you humanize them.

[00:01:28] Speaker 00:
Is that right?

[00:01:30] Speaker 03:
So that is correct.

[00:01:31] Speaker 00:
OK.

[00:01:32] Speaker 00:
And the district court described a process a bit differently.

[00:01:36] Speaker 00:
So I want to, I guess, understand if you agree with her description or you think there's daylight between yours and hers.

[00:01:44] Speaker 00:
Because at appendix 18 and 19, she describes four steps.

[00:01:49] Speaker 00:
One, in vitro assay to identify candidates.

[00:01:52] Speaker 00:
Two, test the best candidate antibodies in animals.

[00:01:56] Speaker 00:
Three, get the antibodies from the animals, and four, humanize them.

[00:02:03] Speaker 00:
Do you have any criticism about that description of the steps?

[00:02:07] Speaker 03:
I mean, I would say when she says find the best candidates, she's essentially describing what you might do if you are trying to find a drug candidate, as opposed to merely practicing the invention.

[00:02:17] Speaker 03:
So I think I would take issue with that.

[00:02:19] Speaker 03:
But I want to step back, because she's also describing what you would do if you

[00:02:23] Speaker 03:
if you wanted to create the antibodies as if these were claims to the antibodies themselves, and they're not.

[00:02:29] Speaker 03:
These are claims to a method of treatment using an already well-known class of antibodies.

[00:02:34] Speaker 03:
All APOSA has to do to practice this method of treatment is take an antibody that's already been identified as an anti-CGRP antagonist antibody and has already been humanized.

[00:02:42] Speaker 00:
But she talks about the test the best candidate antibodies.

[00:02:46] Speaker 00:
Is animal testing required?

[00:02:49] Speaker 00:
Some of your own experts seem to say it was.

[00:02:53] Speaker 00:
So is there any evidence in terms of how complex this testing would have been?

[00:03:00] Speaker 03:
So the evidence from our experts and from the inventors, which the jury would have been able to credit, was that this is not actually incredibly complicated.

[00:03:09] Speaker 03:
They're all processes that you set up.

[00:03:12] Speaker 03:
They don't require teams of people to be acting.

[00:03:15] Speaker 00:
Did they describe it as routine?

[00:03:16] Speaker 03:
Yes.

[00:03:17] Speaker 03:
Do you have any sites who need that?

[00:03:22] Speaker 03:
I don't, Your Honor, because I honestly wasn't focused on this precise issue.

[00:03:25] Speaker 03:
We can get it to you.

[00:03:26] Speaker 03:
We can provide them.

[00:03:29] Speaker 03:
But this goes to one of the major problems in the case, though, which is that the district court treated this case as if it were like this court's cases or the Supreme Court's decision in Amgen.

[00:03:39] Speaker 03:
in which the invention is actually the class of antibodies, where the inventor is saying, I have come up with something new.

[00:03:44] Speaker 03:
There's a new class of antibodies, where there's uncertainty in the evidence about whether all members of that class will work for purposes of the invention.

[00:03:53] Speaker 03:
That is not this case, as the district court itself agreed.

[00:03:55] Speaker 03:
If you look at what the district court said, a jury could have found.

[00:03:59] Speaker 03:
A lot of those factual determinations actually support the plaintiff in this case.

[00:04:03] Speaker 03:
She agreed that, as I was starting to say, a jury could have found that.

[00:04:06] Speaker 03:
you can routinely humanize these antibodies.

[00:04:09] Speaker 03:
And once they're humanized, a person would understand that they all would work for purposes of the invention.

[00:04:14] Speaker 03:
That is a constructive.

[00:04:15] Speaker 04:
One of your argument, I think, hinges on kind of the Hirschler, Angino-Moto, kind of cases.

[00:04:20] Speaker 04:
Is that fair to say?

[00:04:22] Speaker 03:
Exactly right, Your Honor.

[00:04:23] Speaker 04:
And then how does this humanization step, though, fit into that sort of mix?

[00:04:29] Speaker 04:
Using that as your potential framework, I understand, for setting up your argument.

[00:04:32] Speaker 03:
Yeah, what I would say is that when you look at cases like Hershler and Agenomoto, what distinguishes those from the cases that the district court relied upon is that in those cases, you have a well-known genus in the background.

[00:04:44] Speaker 03:
And the question is, will all members of that genus work for purposes of the invention?

[00:04:49] Speaker 03:
And so as the district court agreed, the jury could have found that a POSA

[00:04:53] Speaker 03:
would know that you could take the anti-CGRP antagonists and antibodies, that well-known class, and routinely humanize them, and that the humanization would work, and that once you've done the humanization, they all will work for purposes of the invention.

[00:05:06] Speaker 03:
That makes this case just like Hirshler and Ajinomoto, which we did cite to the district court, and the district court did not address in its decision below.

[00:05:13] Speaker 01:
Well, Ajinomoto didn't add something like four identified promoters

[00:05:19] Speaker 01:
citing to articles that had another dozen promoters.

[00:05:23] Speaker 01:
So you had a lot of promoters.

[00:05:25] Speaker 01:
But in this case, don't you only have one identified humanized antibody?

[00:05:33] Speaker 03:
So Ginomoto did have more.

[00:05:35] Speaker 03:
But I think that's a difference.

[00:05:36] Speaker 03:
It's not a distinction.

[00:05:37] Speaker 03:
Hershler only had one, for example.

[00:05:39] Speaker 03:
There's only a single steroid.

[00:05:41] Speaker 03:
And what the CCPA said in that case was that whatever other differences may exist between the steroids for other purposes,

[00:05:49] Speaker 03:
For purposes of this invention, they'll all work.

[00:05:51] Speaker 03:
They'll all behave similarly.

[00:05:52] Speaker 03:
And that's our case.

[00:05:54] Speaker 03:
We identified antibody G1.

[00:05:56] Speaker 03:
That was enough to direct a POSA to the class of antibodies.

[00:06:00] Speaker 03:
But I also want to make another point, too, which is there was antibody 4901, which was disclosed on the face of the specification.

[00:06:07] Speaker 03:
This was an antibody that what we said

[00:06:09] Speaker 03:
And the specification was available for sale.

[00:06:12] Speaker 03:
We know that both Teva and Lilly obtained antibody 4901.

[00:06:15] Speaker 01:
I'm sorry, was that a humanized antibody?

[00:06:18] Speaker 03:
It was a mirroring antibody.

[00:06:19] Speaker 03:
But the specification says there's antibody 4901.

[00:06:22] Speaker 03:
Here's where you can buy it.

[00:06:24] Speaker 03:
The evidence was it was available for sale to a post-it for just $250.

[00:06:28] Speaker 03:
And the specification then goes on.

[00:06:29] Speaker 03:
It says in some embodiments of this invention, the antibodies are humanized.

[00:06:34] Speaker 03:
It explains why a POSA would want to humanize the antibodies so that they won't be rejected by the human patients.

[00:06:41] Speaker 00:
First, the humanization is also described in the specification, right?

[00:06:44] Speaker 03:
It is extensively described in the specification.

[00:06:46] Speaker 00:
And what about the 12-mirror mine?

[00:06:48] Speaker 00:
I mean, do we put them in as representative because humanization was routine?

[00:06:53] Speaker 00:
Is that your analysis, too, that there's not just one?

[00:06:57] Speaker 03:
There's not just one.

[00:06:58] Speaker 03:
There's at least 4901, which Opposa could have obtained from the catalog.

[00:07:02] Speaker 03:
And then there are other antibodies that are described in the literature.

[00:07:05] Speaker 03:
And I think, thinking about this from a jury trial perspective as well, this jury heard extensively that Lilly had successfully argued to the PTO that this class of murine antibodies was already well known to Opposa.

[00:07:19] Speaker 03:
They even got the evidence.

[00:07:21] Speaker 01:
Sorry, when they were arguing that this class was well known, were they talking about the murine antibodies, or were they talking about the humanized antibodies, or were they talking about both?

[00:07:31] Speaker 03:
So that would be the murine antibodies.

[00:07:32] Speaker 03:
The specification then says, you should humanize these, and here's how to go about humanizing them.

[00:07:37] Speaker 03:
And so we believe under this court's precedence,

[00:07:40] Speaker 03:
that we've cited to the district court in our brief, those all should have been taken into account.

[00:07:47] Speaker 03:
That is exactly the kind of information that a post-it would be looking at in deciding, does this inventor

[00:07:52] Speaker 01:
have possession of what they're claiming which is the class of a different way of saying this was It was obvious how you would get the humanized antibodies.

[00:08:02] Speaker 03:
It's not obvious in the sure honor I know they argue obviousness but this is not a case where there's some limitation missing from the specification and we're trying to argue that a person would have thought that limitation was obvious and

[00:08:13] Speaker 03:
Here, we had antibody G1.

[00:08:16] Speaker 03:
We actually took an antibody all the way through.

[00:08:18] Speaker 03:
And what this court has said in cases like Immunex versus Sandoz that we discussed in our brief is that a district court commits error if it does not consider what a post-it would have considered outside the four corners of the specification.

[00:08:33] Speaker 03:
But here, we don't even have to rely upon information outside the four corners because the specification does point to literature

[00:08:39] Speaker 03:
which describes anti-CGRP antagonist antibodies.

[00:08:42] Speaker 03:
It does identify 4901.

[00:08:43] Speaker 03:
It does point to and explain how to humanize them.

[00:08:48] Speaker 03:
And it tells APOSTA why they would want to humanize them for purposes of the invention.

[00:08:52] Speaker 01:
Does it make any difference to you that, or should make any difference to us, maybe, that there, I think, are three epitopes, if I'm using the right word, your antibody was one of the three

[00:09:04] Speaker 01:
Lilies was involved, a different one of the three.

[00:09:08] Speaker 01:
And yet, because you have the one, you get to claim all three.

[00:09:15] Speaker 03:
So that does not make a difference, Your Honor, because what the district court agreed was that a posto would be aware from the prior art of antibodies binding to all three epitopes.

[00:09:26] Speaker 03:
This was at the court's decision at page 828.

[00:09:30] Speaker 03:
And there was evidence in the record of antibodies that bind to all three epitopes.

[00:09:35] Speaker 03:
And the district court agreed that a poster would understand that all members of the class would work for purposes of the invention.

[00:09:41] Speaker 03:
So the binding spot did not matter.

[00:09:45] Speaker 03:
And I can provide the court the references to the.

[00:09:48] Speaker 03:
to the three epitopes.

[00:09:49] Speaker 03:
But there's an important point here.

[00:09:51] Speaker 03:
And this goes against the jury trial issue, which is that one of the epitopes is the mid-region epitope.

[00:09:58] Speaker 03:
Lily made a very big deal at trial about the fact that supposedly the prior art did not disclose a mid-region epitope anti-CGRP antibody.

[00:10:07] Speaker 03:
Their experts said that several times, and that was disproven on cross-examination.

[00:10:12] Speaker 03:
That was one of many instances in which their expert's credibility was undermined.

[00:10:17] Speaker 03:
And once you discredit your expert on issues like that, on issues like how a post-it would go about trying to develop these antibodies,

[00:10:25] Speaker 03:
His theory was this random substitution theory.

[00:10:28] Speaker 03:
And there was extensive contrary testimony that you would not do random substitution.

[00:10:33] Speaker 03:
You would use this urine process.

[00:10:35] Speaker 03:
They really didn't have a clear and convincing case that the number of antibody candidates would be extraordinarily large or that the genus was very small.

[00:10:43] Speaker 04:
So how would you summarize your best record evidence to support the district court's conclusion that a jury could find?

[00:10:49] Speaker 04:
that the assurpends teach all humanized, anti-CGRP antagonists antibodies would treat the head.

[00:10:57] Speaker 03:
So this would be the testimony of one of our experts, Your Honor.

[00:11:00] Speaker 03:
If you look at record appendix 4174, our expert explained we'd expect that if one anti-CGRP antagonist antibody works in the closed cranial window, then you would expect them all to work.

[00:11:14] Speaker 03:
He put in similar testimony at 4272.

[00:11:16] Speaker 03:
This is the testimony that the district court cited when drawing that factual finding.

[00:11:21] Speaker 03:
And then there was additional testimony that the closed cranial window is a test that's predictive of the ability of a drug to reduce headaches or migraine in humans.

[00:11:33] Speaker 04:
And how would you address the statement?

[00:11:35] Speaker 04:
I think they rely on regents, some for some of their arguments.

[00:11:38] Speaker 04:
How would you respond or address

[00:11:41] Speaker 04:
That portion of regions that talks about describing a method of preparing a cDNA is not enough to satisfy written description.

[00:11:48] Speaker 04:
Is it just that cDNA would be a special case, or how do you respond?

[00:11:52] Speaker 03:
Yeah, I mean, that was a case where there was no ability to get from rat cDNA to human cDNA.

[00:11:58] Speaker 03:
It was certainly not routine to get from rat cDNA to human cDNA.

[00:12:01] Speaker 03:
Whereas here the evidence was that it was routine, again, including Lilly's own statements to the PTO, that it would be routine to get from a urine antibody to a humanized antibody that was essentially

[00:12:14] Speaker 03:
It was not, that was their theory at the PTO.

[00:12:19] Speaker 03:
They won on that theory.

[00:12:21] Speaker 03:
I want to focus on, I'll go ahead, Your Honor.

[00:12:24] Speaker 04:
I'm sorry.

[00:12:25] Speaker 04:
I was going to say, what do you think is the best case going your way on enablement?

[00:12:28] Speaker 04:
I feel like we talked a lot about re-description.

[00:12:30] Speaker 04:
I want to make sure before you sit down, you.

[00:12:31] Speaker 03:
Yeah, I think the key point for us on enablement is if you look at the Supreme Court's decision in Amgen, what it's really focused on is a novel body of genus of antibodies where there's uncertainty whether it will all work for purposes of the invention.

[00:12:45] Speaker 00:
And that's what they've claimed.

[00:12:47] Speaker 03:
And that's what they've claimed.

[00:12:47] Speaker 03:
That's the novel aspect of the invention.

[00:12:49] Speaker 03:
And so our best case would be all this court's cases in which it said that what you have to enable is the novel aspect of the invention.

[00:12:57] Speaker 03:
Here, the novel aspect is not the underlying class of mirroring antibodies.

[00:13:01] Speaker 03:
That was well-known.

[00:13:02] Speaker 03:
It was not humanization.

[00:13:03] Speaker 03:
That was routine.

[00:13:04] Speaker 03:
It was the use and headache.

[00:13:06] Speaker 03:
That is all the claims that these patents cover.

[00:13:10] Speaker 03:
That's exactly what we invented.

[00:13:12] Speaker 03:
That's exactly what we claimed.

[00:13:13] Speaker 03:
It's all that we claimed.

[00:13:15] Speaker 03:
And so we should survive the enablement test.

[00:13:16] Speaker 00:
So maybe this is related, but I'm not clear on why you're disputing whether it's large or not large.

[00:13:23] Speaker 00:
Because even if it's large, assuming I think it's very large, why does that matter your ability to satisfy enablement if we also accept as true that every one of those antibodies works for treating headaches?

[00:13:37] Speaker 03:
We think we can win two.

[00:13:39] Speaker 03:
I think our point there is that their argument, and they have the burden to prove by clear and convincing evidence, depends upon it being large, and as

[00:13:46] Speaker 03:
in part because of the deficiencies of their expert and the incredible theory that he relied upon, that even the jury thought was not credible at summary judgment.

[00:13:55] Speaker 03:
They simply did not meet their burden to prove here, especially to overcome J-MAL.

[00:14:00] Speaker 04:
So it sounds like you're raising two independent arguments on enablement.

[00:14:03] Speaker 04:
Is that fair to say?

[00:14:04] Speaker 03:
The true arguments would be that they didn't necessarily show that the number of candidates was very large, and they also did not show that the number of antibodies within the genus was especially large.

[00:14:13] Speaker 03:
As the district court said, the jury might have found that it wasn't necessarily large at all.

[00:14:18] Speaker 03:
Maybe it wasn't necessarily small, but it wasn't necessarily large.

[00:14:23] Speaker 00:
But even if you're wrong or we don't agree with you on that, we have a second argument.

[00:14:28] Speaker 00:
Yes.

[00:14:29] Speaker 04:
And that second argument, just to get it clear on the record, is what you told us about the inventive aspect and the focus needing to be there, correct?

[00:14:36] Speaker 03:
I mean, the inventive aspect is that all members of the genus will work for purposes of treating headache.

[00:14:41] Speaker 03:
And so whether that's a large genus or a small genus, we have established through the testing that's disclosed in the specification, as opposed to what I understand based upon our expert's testimony,

[00:14:51] Speaker 03:
that all of them will work.

[00:14:52] Speaker 03:
And that's what sets this case apart from the enablement cases that the district court and Lilly rely upon.

[00:14:58] Speaker 00:
Just to restate it, because the class itself was well known and creating members of the class was routine.

[00:15:06] Speaker 03:
Yes, Your Honor, and that makes this case like a genomoto, where the court said, look, it's a well-known class, and you can routinely generate these things, opposed to would know how to do it.

[00:15:16] Speaker 01:
And so just at the risk of redundancy, the representative species here includes not only what your inventors invented, but also the rat antibodies.

[00:15:32] Speaker 03:
It would include the rat antibodies because the specification says, take those antibodies, humanize them, and here's how to humanize them.

[00:15:38] Speaker 03:
So opposed to reading that, would understand that those are members of the genus.

[00:15:43] Speaker 01:
Is there a case where this court has said a representative species can include something that's not actually a species in what's claimed?

[00:15:53] Speaker 03:
I think probably the closest case that we cite in the briefing would be this court's decision

[00:16:02] Speaker 03:
in completely blank.

[00:16:05] Speaker 03:
Oh, sorry.

[00:16:06] Speaker 03:
It would be this course decision in Streck, which is the case involving reticulocytes, where as we read the case, there were reticulocyte analogs.

[00:16:17] Speaker 03:
There were true reticulocytes.

[00:16:19] Speaker 03:
The specification only had a reduction to practice with reticulocyte analogs.

[00:16:24] Speaker 03:
But what the court basically said in considering the written description question was that because true reticulous sites were well known to Ocosa, and Ocosa would understand that they also would work for purposes of the invention, then those would be taken into account for purposes of written description.

[00:16:39] Speaker 03:
But I also don't want to keep losing track of 4901, because 4901 was disclosed in the face of specification.

[00:16:45] Speaker 03:
Specification said you could humanize it, and it'll work.

[00:16:49] Speaker 03:
They had the burden approved by clear and convincing evidence.

[00:16:51] Speaker 03:
They put in nothing about 4901, even though it was discussed extensively by the inventors and by our experts.

[00:16:57] Speaker 03:
And so a jury could have taken that into account, concluding they failed to meet their burden approved.

[00:17:01] Speaker 01:
And as far as anybody's 4901 goes,

[00:17:05] Speaker 01:
You would say that is disclosed as a representative species because it's a rat antibody that's in the specification.

[00:17:15] Speaker 01:
And the specification describes how you would go from having that rat antibody to having a humanized antibody for the 4901.

[00:17:25] Speaker 03:
Agree, Your Honor.

[00:17:26] Speaker 03:
And also, the specification, if you look at it, also discusses extensively experiments that were run with 4901.

[00:17:34] Speaker 03:
For example, it shows that both 4901 and G1, the humanized antibody, would work in the saphenous nerve assay, which is one of the tests they ran to see what is an anti-CGRP antagonist antibody.

[00:17:47] Speaker 03:
So opposed to looking at the specification, we'd be very interested in 4901, which was available for public sale quite cheaply.

[00:17:56] Speaker 00:
Thank you, Your Honors.

[00:17:58] Speaker 00:
Thank you.

[00:18:04] Speaker 00:
Before the clock starts running, just let me get a clarification.

[00:18:07] Speaker 00:
You had marked down time for rebuttal responding to the cross appeal.

[00:18:13] Speaker 00:
Everything on 112, Your Honor.

[00:18:16] Speaker 00:
OK.

[00:18:16] Speaker 00:
Well, we may not see it that way.

[00:18:18] Speaker 00:
But you didn't plan to cover your cross appeal.

[00:18:22] Speaker 02:
We're going to waive the cross-appeal.

[00:18:24] Speaker 02:
We want to focus exclusively on the 112th issue.

[00:18:27] Speaker 00:
You're waiving the cross-appeal in terms of the challenge to the amount, or you're waiving your argument time?

[00:18:35] Speaker 02:
All of it?

[00:18:36] Speaker 02:
The whole thing.

[00:18:37] Speaker 02:
We forfeit our argument on the cross-appeal.

[00:18:39] Speaker 02:
We withdraw it.

[00:18:40] Speaker 02:
Are you dismissing the cross-appeal?

[00:18:42] Speaker 02:
We are dismissing the cross-appeal, yes.

[00:18:44] Speaker 04:
Oh, so we don't need to decide the cross-appeal is what you just told us, correct?

[00:18:47] Speaker 02:
Correct.

[00:18:49] Speaker 02:
Please proceed.

[00:18:51] Speaker 02:
All right.

[00:18:52] Speaker 02:
Thank you, Your Honor.

[00:18:54] Speaker 02:
I think I want to start with the district court's finding that the jury could not have found that the inventors were in possession of an anti-CGRP antagonist antibody that could bind to all three regions of CGRP.

[00:19:10] Speaker 02:
an exceedingly well-supported statement, all of the antibodies in TEVA specification bound to the C-terminal region of CGRP, the C-terminal end.

[00:19:20] Speaker 02:
Antibody 4901 is tested in TEVA specification.

[00:19:25] Speaker 02:
It also binds to the C-terminal region of CGRP.

[00:19:30] Speaker 02:
TEVA attempted to make antibodies that bound to the mid-region of CGRP, which is what Lilly's antibody binds to.

[00:19:40] Speaker 02:
They failed.

[00:19:41] Speaker 02:
They said, objective not achieved.

[00:19:43] Speaker 02:
That's an appendix 24062.

[00:19:46] Speaker 02:
Eight years after filing their patent application, they attempted to make antibodies that bound to the N terminal region of CGRP because they had failed previously.

[00:19:59] Speaker 02:
And what did they write?

[00:20:00] Speaker 02:
They wrote, how to do this?

[00:20:02] Speaker 02:
Question mark.

[00:20:03] Speaker 02:
That's an appendix 5192.

[00:20:09] Speaker 02:
There's nothing in Teva's specification that points to possession of the genus of antibodies that binds to the mid-region or the N-terminal region of CGRP.

[00:20:25] Speaker 02:
Now, they made quite a bit of discussion in their time that the antibodies aren't the invention.

[00:20:33] Speaker 02:
This is a one-step method.

[00:20:35] Speaker 02:
This is a claim.

[00:20:37] Speaker 02:
directed to a method of using humanized anti-CGRP antagonist antibodies.

[00:20:44] Speaker 02:
And cases like Ariad and Rochester v. Searle tell us that regardless of whether a compound is claimed per se or a method is claimed that entails the use of the compound, the inventor cannot lay claim to that subject matter unless he can provide a description of the compound.

[00:21:02] Speaker 02:
And that's what they did and not do.

[00:21:05] Speaker 04:
And so what they're doing now... I mean, you just did a double... I felt like I heard a double negative.

[00:21:10] Speaker 04:
Can you just restate what you were saying?

[00:21:11] Speaker 04:
Because I want to make sure I understand what you're saying.

[00:21:14] Speaker 02:
They need to demonstrate possession of the compounds claimed in their method.

[00:21:18] Speaker 02:
That's the law.

[00:21:19] Speaker 04:
And then you said something they, I think you meant to say, did not do.

[00:21:21] Speaker 04:
And can you tell me what you meant to say that they did not do?

[00:21:24] Speaker 02:
Teva did not demonstrate possession of antibodies that bind to the mid-region or the end-terminal region, but they claimed them.

[00:21:33] Speaker 02:
They're encompassed by their claims, but there is no description of that invention.

[00:21:37] Speaker 01:
And I'm sorry, which two cases did you just say have that requirement?

[00:21:42] Speaker 02:
Areat, that's 598 F third at 1355.

[00:21:46] Speaker 00:
And Rochester- Areat was a composition.

[00:21:50] Speaker 02:
It was a method of using a composition.

[00:21:53] Speaker 02:
And so the point was that the particular class of molecules that were recited in that method were not adequately described.

[00:22:03] Speaker 00:
And that was sort of the core- But here wasn't the class well known?

[00:22:07] Speaker 02:
Yeah, we disagree that the class of molecule that they've claimed was well known because there were no humanized anti-CGRP antagonist antibodies that existed in the prior art.

[00:22:20] Speaker 04:
Do you agree that it was routine to humanize them from the mirroring antibodies?

[00:22:27] Speaker 02:
I think that if you knew the sequence of the myriad antibodies, none of which were known, you needed to have the sequence in order to do the humanization.

[00:22:38] Speaker 00:
Does the specification discuss the humanization?

[00:22:41] Speaker 02:
But it doesn't disclose the sequences of any of the myriad antibodies.

[00:22:46] Speaker 02:
Teva did not make that available to the public, and none of the prior art references disclosed the sequences of the myriad antibodies.

[00:22:56] Speaker 02:
That means that there's no structural information.

[00:23:01] Speaker 02:
So you can't visualize what antibodies are inside the scope of the claim and outside the scope of the claim.

[00:23:07] Speaker 00:
What about the IPR and the statements made in the IPR proceeding?

[00:23:13] Speaker 00:
The claim to the class was struck down.

[00:23:16] Speaker 00:
It was obvious, but you made the statement.

[00:23:19] Speaker 02:
Thank you, Your Honor.

[00:23:21] Speaker 02:
Those statements were directed to the Miriam class of antibodies.

[00:23:26] Speaker 02:
that there was a well-known class, but there was never a statement in the IPRs that the structures of the members of that class were known or that the sequences of the members of that class were known.

[00:23:41] Speaker 02:
And that's the information that a person of skill in the art would need both to make humanized antibodies and to visualize the antibodies

[00:23:50] Speaker 02:
that are within the scope of the claim or outside the scope of the claim.

[00:23:55] Speaker 02:
So the critical thing for the purposes of description under Ariad and Juno and Abbvie is the structure of these antibodies.

[00:24:07] Speaker 02:
And that structure was not in the prior art.

[00:24:11] Speaker 02:
It was not well known.

[00:24:13] Speaker 02:
And Teva did not describe it in their patents.

[00:24:18] Speaker 02:
Let me talk a little bit about Ejinomoto, which was a case that was referenced in Hirschler.

[00:24:26] Speaker 02:
So in Ejinomoto, that's exactly right.

[00:24:29] Speaker 02:
There was a method disclosed, but there were four examples of these potent promoters disclosed in the specification.

[00:24:36] Speaker 02:
There were 14 more disclosed in the prior art.

[00:24:40] Speaker 02:
And there was a structure-function correlation.

[00:24:42] Speaker 02:
So there was a consensus sequence that had been established.

[00:24:46] Speaker 02:
And it was known that the closer the sequences were to that consensus sequence, the more active the promoter would be.

[00:24:55] Speaker 02:
So there was a known structure-function correlation.

[00:24:58] Speaker 02:
That's exactly what's missing here.

[00:25:01] Speaker 02:
Because there was no structural information whatsoever, there was no ability to determine, to visualize or recognize which antibodies were within the scope of the claim and which antibodies were not within the scope of the claim.

[00:25:18] Speaker 04:
Doesn't our case law support up that you don't need to disclose the structure of what's known in the art?

[00:25:26] Speaker 02:
I think in some instances, for example, if a technology is particularly predictable, that may be the case.

[00:25:36] Speaker 02:
But here, again, the underlying fact is that the structures of these antibodies, the sequences of these antibodies,

[00:25:45] Speaker 02:
were absolutely unknown.

[00:25:48] Speaker 02:
And so what Teb is, excuse me.

[00:25:50] Speaker 00:
I mean, just follow up to what Judge Cunha and your PEP, Judge Bryson said exactly that.

[00:25:55] Speaker 00:
If the data is well understood in the art and not itself the invention, is instead a component of the claim, background knowledge may provide necessary support.

[00:26:05] Speaker 02:
In EuroPEP, the specification disclosed the chemical structures of 10 different PDE5 inhibitors, the class of compounds.

[00:26:16] Speaker 02:
There were over 100 different structures reported in the prior art, including the accused product.

[00:26:24] Speaker 02:
So the accused product was actually in the prior art.

[00:26:27] Speaker 02:
This is a fundamentally different situation.

[00:26:30] Speaker 00:
Well, yours went because you can't humanize without the sequence.

[00:26:34] Speaker 00:
But weren't the sequences obtainable if you just procured the antibodies?

[00:26:39] Speaker 02:
Potentially, but that just the- Well, is that a yes?

[00:26:42] Speaker 00:
I mean, yes, right?

[00:26:44] Speaker 02:
If you assume that you could procure the antibodies, then you would get sequences.

[00:26:51] Speaker 02:
That's just a research project.

[00:26:52] Speaker 02:
That's the work that- That's routine, right?

[00:26:57] Speaker 02:
I think for doing it for an antibody, it's routine.

[00:27:02] Speaker 02:
But when you layer in sort of the incredible scope of what they're attempting to claim.

[00:27:07] Speaker 02:
And I guess the other point that I would make, when we talk about the well-known class of antibodies, the facts are undisputed that there are only six or seven such myriad antibodies that existed in the prior art.

[00:27:22] Speaker 02:
That's just a tiny corner, even if you did obtain those antibodies and even if you did sequence them, that's just a tiny corner of the incredible scope of what Teva is claiming.

[00:27:37] Speaker 02:
So I think that that's why because the numbers are so large.

[00:27:41] Speaker 02:
Because the numbers are so large.

[00:27:42] Speaker 02:
This was the situation, Your Honor, in Juneau, for example, where there was one of these particular antibody types in the specification.

[00:27:51] Speaker 02:
And there were, I think, four or five more known in the prior art.

[00:27:55] Speaker 02:
The court said that's not good enough.

[00:27:57] Speaker 02:
That's just a tiny piece because you're

[00:27:59] Speaker 02:
you're claiming functionally.

[00:28:01] Speaker 00:
What if we accept the other side's argument that they've got one disclosure of the human eyes, and we accept his argument about 4901 and about the 12 marine antibodies?

[00:28:17] Speaker 00:
You argue against that in your brief.

[00:28:20] Speaker 00:
What if we say yes to that?

[00:28:22] Speaker 00:
I didn't see your brief argue, even if that's the case.

[00:28:25] Speaker 00:
It's still insufficient.

[00:28:26] Speaker 02:
I mean, that certainly would still be insufficient for written description.

[00:28:30] Speaker 00:
I mean, did you argue that in your brief?

[00:28:34] Speaker 02:
I will.

[00:28:44] Speaker 00:
Sorry to take your time.

[00:28:46] Speaker 02:
So if you look at.

[00:28:48] Speaker 02:
Red brief, pages 49 to 50.

[00:28:51] Speaker 02:
The argument is they failed to rodent antibodies.

[00:28:55] Speaker 02:
Dr. McDonald confirmed that no sequence or structure was known for the rodent antibodies disclosed in Teva's specification of the prior art.

[00:29:03] Speaker 02:
Teva's experts similarly acknowledged that none of the sequences were known.

[00:29:07] Speaker 02:
And so this contention that they need not disclose

[00:29:12] Speaker 00:
But that's my question.

[00:29:14] Speaker 00:
I understood this to be with respect to the 12 Marines.

[00:29:19] Speaker 00:
But let's assume that we disagree with that.

[00:29:21] Speaker 00:
I didn't see another argument, which is even if they were disclosed, this disclosure was insufficient.

[00:29:28] Speaker 00:
Did you make that argument?

[00:29:29] Speaker 02:
I don't have the site at my fingertips.

[00:29:32] Speaker 02:
My recollection is that we additionally argued that the limited number of species that would be available simply is not enough.

[00:29:40] Speaker 02:
In the AbbVie case where there were 300 antibodies disclosed, that wasn't enough because the

[00:29:47] Speaker 02:
antibody, essentially there was a quesive infringement there, was so functionally different, which is exactly the case here.

[00:29:55] Speaker 02:
In the Amgen case, there were 26 different species.

[00:29:59] Speaker 02:
That was not enough.

[00:30:01] Speaker 02:
Here you have fewer examples that were in the prior art.

[00:30:06] Speaker 02:
And moreover, you have this possession problem of them not being able to make antibodies that bound to the mid-region or to the N-terminal region, like Lilly's antibody.

[00:30:17] Speaker 02:
So there's this extra layer of complication.

[00:30:21] Speaker 02:
I'd like to turn to their statements about the credibility of Lilly's experts.

[00:30:28] Speaker 02:
So Lilly's expert offered the opinion that the potential scope of antibodies in the genus claimed by Teva was exceedingly large, 10 to the 78th power of antibodies.

[00:30:40] Speaker 02:
And this was consistent with a slide from Teva's inventor on the potential diversity of antibodies.

[00:30:48] Speaker 02:
This is at appendix 18.012.

[00:30:51] Speaker 02:
There is no contrary testimony on the number of candidate antibodies potentially within the scope of the genus.

[00:30:59] Speaker 02:
And so the district court concluded that the jury could only have found that the number of antibodies needed to be screened was large and that the size of the genus was unknowable.

[00:31:10] Speaker 04:
Do you think a novel aspect of the invention is using the class of antibodies to treat headache?

[00:31:15] Speaker 02:
I think it's the antibodies themselves.

[00:31:19] Speaker 02:
It's the entire scope of the claim.

[00:31:21] Speaker 04:
I thought that we've already established, though, that it's the method of treatment claims here, right?

[00:31:27] Speaker 04:
It's not just a claim to the antibodies.

[00:31:33] Speaker 02:
But the claim is a method of treatment.

[00:31:36] Speaker 02:
But under Ariad and under Rochester, if the method is claiming

[00:31:43] Speaker 02:
a genus of compounds, you need to adequately describe the compounds that are part of what you're claiming.

[00:31:53] Speaker 00:
Methods for treating headaches.

[00:31:55] Speaker 00:
Correct.

[00:31:55] Speaker 00:
That's what this invention is.

[00:31:57] Speaker 02:
No.

[00:31:57] Speaker 02:
It's a method of treating a headache of using a genus of compounds that was not adequately described or enabled.

[00:32:07] Speaker 04:
You threw a few extra in there, but they're not at a pre-described or enabled.

[00:32:13] Speaker 04:
Are you arguing the Pennison suit did not enable the method of treatment if the class of antibodies itself is enabled?

[00:32:19] Speaker 02:
Yes.

[00:32:20] Speaker 02:
That is also our argument.

[00:32:21] Speaker 02:
We do have the- What's your basis for that argument?

[00:32:27] Speaker 02:
The core of the argument is that you can't make predictions from antibody to antibody.

[00:32:32] Speaker 02:
That testimony is undisputed.

[00:32:34] Speaker 02:
And so consequently, you have this incredibly large genus of antibodies.

[00:32:39] Speaker 02:
And you have to make and test each antibody empirically to determine whether or not it's a member of the class.

[00:32:48] Speaker 02:
given the amount of time that's involved and the amount of work that's involved in doing that, where their witnesses testified that to do the mouse method is, quote, not very quick.

[00:33:02] Speaker 02:
It takes four months.

[00:33:04] Speaker 02:
That's Appendix 3391.

[00:33:05] Speaker 02:
Their expert testified that you do need to do testing in animals to determine what are

[00:33:14] Speaker 02:
are good candidates.

[00:33:15] Speaker 00:
Well, I guess the concern I have is that we have a jury verdict here.

[00:33:19] Speaker 00:
If we didn't have a jury verdict, we'd be in a different place.

[00:33:22] Speaker 00:
If the jury verdict were in your favor, we'd be in an entirely different place.

[00:33:25] Speaker 00:
But just because you have some testimony that supports what you're saying here today is not sufficient.

[00:33:32] Speaker 00:
Right.

[00:33:32] Speaker 00:
You get that.

[00:33:33] Speaker 02:
I do understand, but I think taking every assumption favorable to Teva, I think if you look at the whole of the evidence, the evidence is that you cannot predict which antibodies are going to work in the method and which antibodies aren't going to work in the method.

[00:33:48] Speaker 02:
That is fundamentally an enablement problem when you have a very broad genus, which is what Teva has here.

[00:33:55] Speaker 02:
And all of the testimony that we're citing for that is Teva's expert.

[00:33:59] Speaker 02:
Appendix 4257, Tevis Inventor, Appendix 1407, where they say that you have to do the experiments.

[00:34:07] Speaker 02:
You have to make and test each of these antibodies to determine whether or not they work.

[00:34:13] Speaker 00:
And then the question, taking everything you said as given, the question is, experimentation is not a problem unless it's undue experimentation, right?

[00:34:23] Speaker 00:
So is there testimony on undue experimentation?

[00:34:25] Speaker 02:
There is testimony on undue experimentation.

[00:34:29] Speaker 00:
There's been testimony on the other side by Tepes expert that undue experimentation was not required.

[00:34:34] Speaker 02:
Yes, I don't think Teva's expert conceded it, but they do concede that in order to know which antibodies work, that you need to do the experimentation.

[00:34:46] Speaker 02:
And when you have to do experimentation for a class of antibodies that's this large, where there's so many antibodies, potentially within the scope of the claim, by law under WIATH and IDENIX and Amgen, that constitutes undue experimentation.

[00:35:14] Speaker 03:
Thank you, Your Honors.

[00:35:16] Speaker 03:
So to respond to some of the points made there, my brother started off by arguing about whether we had possession of the antibodies to all three epitopes.

[00:35:25] Speaker 03:
With the evidence, two points here.

[00:35:27] Speaker 03:
There's an evidentiary point, which is that the inventors testified

[00:35:30] Speaker 03:
that at a certain point they stopped attempting to make additional antibodies because they already had a good candidate.

[00:35:35] Speaker 03:
This is not pure science.

[00:35:36] Speaker 03:
These are companies.

[00:35:37] Speaker 03:
They don't do research.

[00:35:38] Speaker 03:
They don't need to.

[00:35:39] Speaker 03:
In addition, that entire argument ignores constructive possession.

[00:35:44] Speaker 03:
And again, the district court found that a poster would have understood that there were antibodies disclosed in the prior art that bound to all three epitopes, and that once they were humanized, they all would work for purposes of the invention.

[00:35:55] Speaker 03:
There was a discussion.

[00:35:57] Speaker 01:
He said something about there only being six or seven rat antibodies in the prior order that were known.

[00:36:05] Speaker 01:
Do you agree with that?

[00:36:07] Speaker 03:
I don't.

[00:36:08] Speaker 03:
His expert pointed to a few different pieces of literature and said in these pieces of literature, in these four pieces of literature, there are six disclosed.

[00:36:17] Speaker 03:
But for example, there was a discussion, extensive discussion,

[00:36:20] Speaker 03:
across multiple witnesses of the TAN prior art literature, which disclosed, I think, over 11 antibodies.

[00:36:28] Speaker 03:
Is that the graduate student?

[00:36:29] Speaker 03:
That was the graduate student.

[00:36:31] Speaker 03:
That's one of the aspects of this case, too, right?

[00:36:33] Speaker 03:
These are not antibodies like in Amgen, where you have multi-billion dollar corporations pouring resources into finding the one antibody that will hit the sweet spot.

[00:36:41] Speaker 03:
These are antibodies that a graduate student can make using a mouse.

[00:36:45] Speaker 03:
And they were freely sharing them before the priority date.

[00:36:48] Speaker 03:
They were just giving them away because no one actually really cared about the antibodies themselves.

[00:36:53] Speaker 03:
What is important about this case is the novel method of treatment that we came up with.

[00:36:58] Speaker 03:
That's what's the drug.

[00:37:01] Speaker 03:
That's why we're having the fight here today.

[00:37:03] Speaker 03:
And that's exactly what these particular patents claim.

[00:37:06] Speaker 03:
There was an argument made about knowing the sequences.

[00:37:09] Speaker 03:
And if you don't know the sequences, what can you do?

[00:37:12] Speaker 03:
Well, I would point this cord, as we do in our briefs, to Enzo Biocam versus GenProbe.

[00:37:17] Speaker 03:
If the material's out there and a poster has access to it, then they can sequence it.

[00:37:22] Speaker 03:
And that was the case, for example, with 4901, which they just ignored in front of the jury.

[00:37:27] Speaker 03:
And the jury could have taken that into account.

[00:37:31] Speaker 03:
Ajinomoto, there was a discussion about the number of species and whether the species there had a common structure.

[00:37:38] Speaker 03:
As we point out in our brief, Ajinomoto actually had two different

[00:37:43] Speaker 03:
pieces of its analysis.

[00:37:44] Speaker 03:
One was representative number of species.

[00:37:46] Speaker 03:
One was common structural feature.

[00:37:49] Speaker 03:
When it came to representative number of species, there was not a discussion about any kind of common structure.

[00:37:54] Speaker 03:
There, the court simply looked at what was disclosed in the specification about numbers, what was known to oppose it, and decided that enough was disclosed.

[00:38:01] Speaker 03:
And you're only going with representative species, right?

[00:38:04] Speaker 03:
We're only going with representative species, Your Honor.

[00:38:07] Speaker 03:
In Hirschler, there, again, only one was disclosed.

[00:38:13] Speaker 03:
And the important point in Hirschler was that despite the fact that only one was disclosed, a postal looking at that would understand if that one works, then whatever the variety of the genus is for other purposes, they'll all work for purposes of this invention.

[00:38:25] Speaker 03:
And that's exactly what the district court found the jury could have concluded here.

[00:38:30] Speaker 03:
The size of the Kennedy pool, every time.

[00:38:31] Speaker 03:
Oh, sorry, Your Honor.

[00:38:34] Speaker 03:
One sentence more.

[00:38:34] Speaker 03:
OK.

[00:38:34] Speaker 03:
So every time my brother referenced the large size of the antibody pool with a genus, he was relying upon Dr. McDonald's testimony, which, as we point out in our brief, was, of course, contested by our people.

[00:38:46] Speaker 00:
Thank you.

[00:38:47] Speaker 00:
Thank you.

[00:38:48] Speaker 00:
We thank both sides.

[00:38:49] Speaker 00:
The case is submitted.

[00:39:00] Speaker 00:
Is our district court, gentlemen, could I have your attention?

[00:39:04] Speaker 00:
Just do you want to file, just for our record-keeping purposes, a motion to dismiss the process?

[00:39:10] Speaker 01:
We'll do so.

[00:39:11] Speaker 00:
Thank you.

[00:39:12] Speaker 00:
Thank you.