[00:00:25] Speaker ?:
Thank you.

[00:00:47] Speaker 01:
211729, AstraZeneca vs Mylan.

[00:01:19] Speaker 01:
Mr. Dufresne, is that it?

[00:01:22] Speaker 02:
That's right, Your Honor, thank you.

[00:01:25] Speaker 02:
May it please the Court.

[00:01:26] Speaker 02:
The District Court's claim instruction in this case was wrong.

[00:01:31] Speaker 02:
Throughout the specification, the key consistent notion expressed again and again was that formulations of 0.001% PVP were different.

[00:01:40] Speaker 02:
They were better, they were more stable than all other disclosed formulations.

[00:01:45] Speaker 02:
And before the District Court,

[00:01:47] Speaker 02:
AstraZeneca urged a broad construction that treated 0.001% PVP as one in the same with 0005%, which was disclosed as a different, less stable formulation.

[00:02:00] Speaker 01:
AstraZeneca's broad- Does anything in the specs say that the lower figures are unacceptably unstable?

[00:02:09] Speaker 02:
I don't know.

[00:02:10] Speaker 02:
There's no active disparate event like that in the specification.

[00:02:15] Speaker 02:
We don't need to show that here, because we're not arguing disclaimer.

[00:02:18] Speaker 02:
We're arguing that, in the first instance, the person of ordinary skill in the art looking at this specification would understand that 001% was something different than 0005%.

[00:02:30] Speaker 02:
And 001 couldn't be construed to be the same as 0005%.

[00:02:34] Speaker 01:
But it's not the same, right?

[00:02:36] Speaker 01:
A little rounding room bubble around 001 is different from a little rounding

[00:02:43] Speaker 01:
room bubble around 0005, even though there's a tiny overlap.

[00:02:49] Speaker 02:
Non-overlap does not mean the same.

[00:03:00] Speaker 02:
Well, overlap in this case does mean that 0001, when the claims say 0001 in this case under the district court's construction,

[00:03:08] Speaker 02:
That also means 0005.

[00:03:10] Speaker 02:
It includes 0005.

[00:03:12] Speaker 01:
It means exactly pinpoint from 0005 up to, down to 0005, but it doesn't mean what 0005 means, which is the bubble around, the rounding room bubble around 0005.

[00:03:31] Speaker 01:
If they had claimed 0005,

[00:03:33] Speaker 01:
then that would have actually been a different non-pinpoint bubble, a different range.

[00:03:42] Speaker 01:
The fact that one overlaps with the other doesn't mean that they are no longer distinguished, because they're not identical, even if they overlap.

[00:03:52] Speaker 01:
That's why I thought your argument really did kind of turn on your suggestion, at least in your opening brief, that 0005 was, let's just use the word, disparaged.

[00:04:05] Speaker 01:
Forget about whether it's teaching away disparaged.

[00:04:07] Speaker 01:
In some way disparaged.

[00:04:09] Speaker 01:
That if you go that far down, you're off into bad territory.

[00:04:14] Speaker 01:
And I don't see the spec saying that.

[00:04:17] Speaker 02:
Definitly, there's several questions in there, I think, Your Honor.

[00:04:21] Speaker 02:
The specification definitely tells skilled artisans that 0005 wasn't worse than 001.

[00:04:31] Speaker 02:
Throughout the specification, again and again, in column 2, over and over again in column 6, and all the data that are described in column 6, the specification describes 001 as the best, the most stable.

[00:04:46] Speaker 02:
That's comparative language.

[00:04:48] Speaker 02:
That's telling the skilled artisan that 001 is better than 0005.

[00:04:53] Speaker 02:
And it makes the problem here more jarring.

[00:04:58] Speaker 02:
It makes it clearer to the skilled artisan reading the specification that these two things are different.

[00:05:05] Speaker 02:
And it would be improper to read 001 as broadly as the district court did in a way that makes 001 extend all the way to 0005.

[00:05:14] Speaker 01:
In that last sentence, when you say 0005, you're no longer talking about bubble 0005.

[00:05:23] Speaker 01:
Rounding room meaning of 0005 in the world in which there is such a thing as significant digits.

[00:05:30] Speaker 01:
Now you're talking about mathematical 0005, a point on the line.

[00:05:35] Speaker 02:
I'm talking about the real world formulation.

[00:05:40] Speaker 02:
that's described in the specification as having 0005%.

[00:05:44] Speaker 02:
So these numbers, when they're using the specification, they aren't describing an abstract math problem.

[00:05:50] Speaker 02:
That's not what's going on here.

[00:05:52] Speaker 02:
What we have are these numbers used to express particular formulations that were described in the specification that the inventors prepared and tested

[00:06:02] Speaker 02:
And the 0005 performed much worse than the 001 in every one of those tests.

[00:06:07] Speaker 02:
That was a consistent theme throughout the specification that the 0005 was worse.

[00:06:12] Speaker 00:
All of the other formulations were worse.

[00:06:16] Speaker 00:
What about the fact that you yourself seem to suggest, or you do suggest, advocate a claim construction that relies on a significant digit?

[00:06:25] Speaker 00:
You go to the second digit for your model.

[00:06:30] Speaker 00:
In that situation, why is the second digit?

[00:06:35] Speaker 00:
Everybody seems to agree that there should be some rounding error because it's really hard to have a precise number, right?

[00:06:45] Speaker 00:
But why is yours correct?

[00:06:46] Speaker 00:
And why doesn't yours suggest that there's some agreement that when you look at a number here in this pharmaceutical composition, it's not just a target number, there's some bubble around it.

[00:06:57] Speaker 02:
As you suggested, in this case, it is clear that there has to be some kind of wiggle room, right?

[00:07:04] Speaker 02:
Because you can't make these formulations to like a level of atomic precision when you're putting them together.

[00:07:08] Speaker 02:
So there has to be some boundary.

[00:07:10] Speaker 02:
This isn't the case where it can be an exact number.

[00:07:13] Speaker 02:
And the question becomes how much wiggle room is appropriate.

[00:07:19] Speaker 02:
We submit that in this case, a person of ordinary skill and they are looking at this specification would conclude that

[00:07:26] Speaker 02:
that rounding from the third decimal place can't be correct here because that would mean that my double zero one would extend to triple zero five and so you would say all numbers need to be rounded to the triple zero five position yes exactly and that's and that's also that's also consistent with the way that the inventors

[00:07:45] Speaker 02:
describe the precision that we used to make these formulations when they talked about them in specification.

[00:07:50] Speaker 02:
So we have the 001 formulation, but we also have the 0001 and the 0005.

[00:07:57] Speaker 02:
And those also produce different data when they were tested.

[00:08:03] Speaker 02:
They didn't show exactly the same performance.

[00:08:05] Speaker 02:
They were different too.

[00:08:07] Speaker 02:
And so that also tells us, the person of our nearest skill in the art,

[00:08:10] Speaker 02:
When those were formulated and the formulator is putting in a certain amount of PVP to get to that triple zero one or triple zero five, they were able to do it precisely enough that those two tested differently and they were different.

[00:08:24] Speaker 02:
And so that level of precision was available when they're making up all of these formulations.

[00:08:28] Speaker 00:
What other evidence do you think there is thinking about the point of no dysfunction of claims and understanding claims?

[00:08:35] Speaker 00:
What other evidence is there to support

[00:08:38] Speaker 00:
You know, the scientific knowledge or knowledge of one and more nearest family art of how we're supposed to determine what the wiggle room is around this particular number without having the word about, for example, before it.

[00:08:53] Speaker 02:
Right, so we believe that the evidence I've talked about, the improper overlap between 001 and 005, the level of precision shown in the specification going out to the fourth digit and making the 001 and 005 indicate that our construction is correct.

[00:09:12] Speaker 02:
But I think your honor is asking about is there expert testimony.

[00:09:15] Speaker 00:
Yeah, is there anything else?

[00:09:17] Speaker 00:
I mean, I'm familiar with the specification, the prosecution history, the removal of ranges and the removal of the bow, but I'm not sure that

[00:09:23] Speaker 00:
Again, it's a question of public notice.

[00:09:28] Speaker 00:
What would Opposa understand from this language in light of that evidence?

[00:09:31] Speaker 02:
So in this case, neither side presented expert testimony at the time of claim construction.

[00:09:35] Speaker 02:
And I think that's because this was kind of a late-breaking claim construction.

[00:09:41] Speaker 02:
We talked about the history of how this progressed, but the parties had agreed at one point that they would both use the ordinary meaning.

[00:09:49] Speaker 02:
And then it came close to trial.

[00:09:50] Speaker 02:
I think this was about

[00:09:51] Speaker 02:
2 months before trial when this construction was done and it became clear that the parties didn't agree on on what the ordinary meeting was and so the district court said I want briefing on this and I'll decide it and so she she heard briefing or she took briefing and heard argument and decided that on that basis but I think there just wasn't time for the parties to put in expert testimony on the on the question you're talking about your honor and your your current position about mathematical exact 0.001 plus some I forget you had

[00:10:21] Speaker 01:
We use the term like legal room.

[00:10:24] Speaker 01:
I forget what it was.

[00:10:25] Speaker 01:
Small deviation or something like that.

[00:10:27] Speaker 01:
Minor deviation.

[00:10:28] Speaker 01:
Minor variation maybe.

[00:10:31] Speaker 01:
And when you put numbers on that, the numbers you put on that were precisely the rounding error if you added a second significant digit, namely made it 0010.

[00:10:41] Speaker 01:
Precisely.

[00:10:42] Speaker 01:
And that's what you put in your paragraph four letter as well, right?

[00:10:47] Speaker 01:
That was in the notice letter.

[00:10:51] Speaker 01:
And I've been curious until I read that what Judge Keeley meant when she said the parties agree that the significant digit understanding of these numbers is the right methodology of understanding and I didn't know until I saw that what the basis for that was but that sounds exactly right.

[00:11:10] Speaker 01:
In fact, your

[00:11:11] Speaker 01:
Now, minor variation from the mathematical exact figure is identical to the rounding version of a significant digit, if you add a significant digit, namely a zero at the end of 0001.

[00:11:27] Speaker 02:
Right, so we agree that rounding is, in general, the way that people can look at numbers and figure out what kind of

[00:11:33] Speaker 02:
but we submit that in the context of this particular patent with this specification and this file history that focused on lifting up the 001 formulation as being different and better than all the other ones that just blindly rounding to the third decimal place that's used in the claim would not be

[00:11:56] Speaker 03:
The specification recognizes that precision is important to hear out to the fourth digit because it tested at least a couple of formulations out to the fourth digit, so it must have understood those compositions could be different and have different results than just out to the third digit.

[00:12:27] Speaker 03:
Because otherwise you would have just tested out to the third and done rounding there.

[00:12:32] Speaker 02:
Yeah, I think that's an important part of it, Your Honor, because the fact that those formulations out to the fourth digit were there and were showing that the difference between those two, the 0001 and the 0005, had a meaningful effect in the performance of those formulations.

[00:12:46] Speaker 02:
That shows that the formulations were able to be made with precision out to that level.

[00:12:54] Speaker 02:
And if you do rounding out to the fourth digit, you get plus or minus.

[00:12:58] Speaker 03:
I mean, if the third digit is all that mattered, then presumably you wouldn't have bothered testing 0005 because it rounds up to 001 anyway.

[00:13:09] Speaker 02:
Right, and in fact it wouldn't have been possible if it was true that the only level of precision that matters is the third digit, you wouldn't have been able to reliably make one up to the fourth digit.

[00:13:20] Speaker 02:
You know, 0001 and 0005 wouldn't be reliably different ratios otherwise.

[00:13:26] Speaker 00:
I would like to go back to a question that Judge Toronto asked you and that was about the

[00:13:30] Speaker 00:
district court statement about how both parties agree that Opposite would interpret the specification to convey that 0.001% term is subject to rounding according to the number of significant digits.

[00:13:44] Speaker 00:
Is that true?

[00:13:46] Speaker 02:
Because I thought there was a difference between the parties.

[00:13:49] Speaker 02:
I think that there's a difference between the parties certainly as far as how they think rounding should apply to the claim term here.

[00:13:55] Speaker 02:
I think both parties

[00:13:56] Speaker 02:
have proposed constructions that use rounding to give, you know, to allow for the kind of imprecision that's required here in the context of this invention.

[00:14:06] Speaker 02:
But we certainly did not agree on exactly how that should be done or to what the decimal.

[00:14:10] Speaker 02:
Or they could use the third digit.

[00:14:12] Speaker 02:
Exactly.

[00:14:12] Speaker 02:
There's a difference on what kind of precision that should reflect.

[00:14:18] Speaker 03:
It seems to me like you potentially confused the district court when you were talking about agreeing on significant digits because

[00:14:25] Speaker 03:
At least to me, out of context, when you're looking at 001 in kind of general math terms, and God knows I haven't taken math since high school, so I don't really know what I'm talking about, but the significant digit is the 1, not the 10.

[00:14:42] Speaker 03:
So I get where she came from when you said that, but that seems like an acontextual view of what actually are significant for this patent by using that phrase.

[00:14:54] Speaker 02:
я думаю, что это точно так, господин, это возможно, что взяли 0.001, это, вы знаете, абстрактный номер на письме, кто-то может посмотреть на это и сказать, но это должно быть оборудовано, вы знаете, как district court в ее строительстве, но я думаю, в контексте этого патента, в котором district court пошло не так, она на самом деле оборудовала это так, как вы могли, я думал, что она оборудовала это до 0. whatever the five is и до 14

[00:15:24] Speaker 03:
но это действительно правильно, если раундин не раундовал даже далеко от 1.4, если бы вы не добавили много дизайнов?

[00:15:35] Speaker 02:
Я не утверждаю, что я математик или нет, господин, но я верю, что... Я буду не задавать вопросов на математику, потому что я думаю, что у нас тут математик.

[00:15:46] Speaker 02:
Я верю, что раунда будет от 0.0005, если вы раундаете до третьей дизайна, то будет от 0.0005 до 0.0014, потому что если вы попали до 0.0015, то вы раундовали до

[00:15:53] Speaker 03:
0.002.

[00:15:55] Speaker 03:
I guess what I'm saying is if you added even more digits, it'd be 1499 or whatever the number is.

[00:16:03] Speaker 03:
It's not 140, it's 14 something above zero.

[00:16:09] Speaker 02:
Yeah, you would end up having the plus or minus that you use at the end would get smaller and smaller by tenfold each time you go to a different digit to do your rounding.

[00:16:21] Speaker 02:
There are no further questions.

[00:16:22] Speaker 02:
I'll reserve the zero time I have left.

[00:16:34] Speaker 01:
So Mr. Burrow, talk about why the construction, why the claim should not be read in context to make clear, the whole aspect, make clear that what is significant, I don't mean significant digit, I mean significant in context is through four places.

[00:16:58] Speaker 04:
So the answer is, and I just like to make this clear because it was clear below, there was no confusion at all below about what the dispute was in this claim.

[00:17:07] Speaker 04:
I think they're trying to obscure it here.

[00:17:10] Speaker 04:
Everyone understood below that this dispute was, are there two significant digits or one?

[00:17:15] Speaker 04:
All of this business about the fourth decimal point is entirely new.

[00:17:19] Speaker 04:
Never mentioned before, never addressed before, and not the subject of construction below.

[00:17:24] Speaker 04:
They said below, and you can look, for example, at A7913, where their argument was that the POSA would understand the term to have a second significant figure, not four decimal points, a second significant figure.

[00:17:39] Speaker 04:
They said it two pages later.

[00:17:40] Speaker 00:
Mr. Burrell, I have a question for you.

[00:17:42] Speaker 00:
Are you suggesting that they have a different claim construction?

[00:17:45] Speaker 00:
и и

[00:18:04] Speaker 04:
0.0010.

[00:18:07] Speaker 04:
Two significant figures, I think that's the same construction.

[00:18:09] Speaker 04:
But as Judge Toronto noted, they obscured it in their brief.

[00:18:13] Speaker 04:
They used other words like minor deviations and other verbiage.

[00:18:20] Speaker 00:
I don't think they've changed the amount of deviation.

[00:18:23] Speaker 00:
I think they're still arguing the same amount of deviation.

[00:18:25] Speaker 00:
So assuming that's so, because they can't change the plane construction.

[00:18:27] Speaker 00:
What is your response to why, in light of this specification, their claim construction isn't correct, given that the level of precision in this specification, with the different numbers that were tested,

[00:18:42] Speaker 04:
ассооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооооо

[00:19:08] Speaker 04:
And that's simple math.

[00:19:14] Speaker 00:
The question is not how many decimal points it goes out to with respect to the level of precision and how much bubble room there is.

[00:19:21] Speaker 00:
Don't I have to look at that Phillips case in terms of how it was claimed?

[00:19:25] Speaker 00:
I mean, it was claimed in a different way, right?

[00:19:26] Speaker 00:
Yes.

[00:19:26] Speaker 00:
That was a different mathematical way of writing it.

[00:19:28] Speaker 04:
It was, but the court noted that it's the same thing.

[00:19:30] Speaker 00:
The claim there, I think, said 10 to the minus 6, and the court... The court also said that in this context of this patent,

[00:19:37] Speaker 00:
Expressing it logarithmically like that meant something different, right?

[00:19:42] Speaker 00:
That it wouldn't have wiggle room.

[00:19:44] Speaker 04:
Well, it had some wiggle room, but it meant something different than adding a significant digit.

[00:19:48] Speaker 04:
And the court understood that the bubble room, so to speak, depends on significant digits.

[00:19:53] Speaker 04:
And I think the Viscasi case stands for the exact same proposition, where the claim said 0.91.

[00:19:58] Speaker 04:
And the question in that case, as it is here, and that case was argued below and of course argued in our briefs here,

[00:20:05] Speaker 04:
was how much wiggle room is there around that.

[00:20:07] Speaker 04:
There, they said there's not much wiggle room, notwithstanding the usual meaning of .91, two significant digits, we need more because the patentee during prosecution defined the claim to be .910 and below.

[00:20:21] Speaker 04:
And in this specification, did the same thing, added a third significant digit.

[00:20:25] Speaker 04:
That evidence is completely absent from this intrinsic record, 23 times.

[00:20:31] Speaker 04:
This amount of PVP is addressed in the specification.

[00:20:35] Speaker 00:
Can I just say something?

[00:20:37] Speaker 00:
Let me tell you what my concern is a little bit with your argument.

[00:20:40] Speaker 00:
Because I want to make sure you hear it.

[00:20:41] Speaker 00:
I'm a little worried about the public notice function.

[00:20:43] Speaker 00:
And let me explain that a little bit more.

[00:20:45] Speaker 00:
I mean, you have in the prosecution history and in other related patents, I see you know how to use the word about.

[00:20:55] Speaker 00:
I see you know how to claim ranges.

[00:20:57] Speaker 00:
And I'm left looking at this and thinking,

[00:20:58] Speaker 00:
What a pose of looking at this really know that this is how I'm supposed to interpret this the way the court did here given that you don't have the word about you.

[00:21:10] Speaker 00:
This is it's it's there's some more precision you could have more

[00:21:15] Speaker 00:
I'm

[00:21:33] Speaker 04:
And that was used consistently throughout the specification.

[00:21:36] Speaker 04:
And what that means, there's no dispute between the parties.

[00:21:39] Speaker 04:
That it's not mathematical, as Judge Toronto called it, .001.

[00:21:43] Speaker 04:
It has a bubble.

[00:21:44] Speaker 04:
It has a range.

[00:21:45] Speaker 04:
The parties agreed below about that.

[00:21:46] Speaker 04:
The parties agree here.

[00:21:47] Speaker 03:
So let me understand this, because it was not apparent to me from the briefing that this is what you're arguing.

[00:21:55] Speaker 03:
But you originally claimed a broad range, got rejected, and you ultimately ended up, at least what we're dealing with here, as a claim for double zero y.

[00:22:04] Speaker 03:
your view is what that really means is a range from 0005 up to 014 one could call it a range one could call it bubble room around it what it really means is one significant digit 0.001 everyone agrees it's a range or has a bubble the question here is simple how much of a bubble is there but see this is my problem and it's very similar to judge stole so about the notices here you

[00:22:30] Speaker 03:
If you knew how to claim a range, if you really wanted that range from 0005 all the way up to 014, you could have claimed it.

[00:22:40] Speaker 03:
You're saying that that's inherent in your 001 argument, but this pattern is replete with other ranges that you go to further position.

[00:22:52] Speaker 03:
I think you heard me say to your friend that it seems to me that suggests that everybody, that you all understood that precision out to four digits could make a difference.

[00:23:05] Speaker 03:
So why shouldn't we read that as basically claiming

[00:23:09] Speaker 03:
a four digit number, even though you didn't put the 01 after the one.

[00:23:14] Speaker 03:
In the context of this, I mean, again, I understand why the district court got here, because it seems to me that if you're looking at this one number and not reading it in the context of the patent, but just looking at 001 and saying, what does that round up to and round down to, you are precisely correct.

[00:23:33] Speaker 03:
But that doesn't answer the question of what this patent meant when it was claiming

[00:23:38] Speaker 04:
0.001.

[00:24:01] Speaker 03:
to this specific recipient, it uses four digits when it wants to use a much lower percentage, right?

[00:24:09] Speaker 04:
One significant digit again, and the party's agreed below that the level of precision depends on the number of significant digits, not how many decimal points it goes out to.

[00:24:17] Speaker 04:
I think that's a different thing.

[00:24:18] Speaker 04:
In response to your honors submission that

[00:24:21] Speaker 04:
We could have just claimed 0005 to 0014.

[00:24:24] Speaker 04:
Had we said that, that would have been a claim of a different scope than this one, because 0005, as used in your honor's example, likewise, would have been accorded significant digits and would have gone down as would 00014.

[00:24:38] Speaker 04:
The point here, and everyone agreed below, is that these numbers depend on how many significant digits there are.

[00:24:44] Speaker 04:
That defines the so-called bubble room.

[00:24:46] Speaker 04:
And the only question is, two significant digits or one?

[00:24:49] Speaker 04:
The intrinsic record is clear that the answer is one over and over and over again.

[00:24:54] Speaker 04:
And I'd like to address the prosecution history briefly, Your Honor.

[00:24:57] Speaker 04:
It's not correct that in the prosecution history the claim simply issued because it said 001 and somehow excluded other concentrations.

[00:25:06] Speaker 04:
That's not true.

[00:25:07] Speaker 04:
That's not what it says in the notice of allowance.

[00:25:09] Speaker 04:
The examiner gave us other claims that went up.

[00:25:12] Speaker 04:
The examiner gave us claims that didn't have any PVP concentration whatsoever.

[00:25:15] Speaker 04:
So that's sort of a revision of what actually happened.

[00:25:18] Speaker 04:
What happened was that there was this combination of elements and that all of those combinations of excipients and percentages together were deemed non-obvious and patentable.

[00:25:28] Speaker 04:
And one of those percentages is 001.

[00:25:30] Speaker 04:
And that was not defined more precisely than 001, one significant digit.

[00:25:36] Speaker 04:
And I couldn't have said it better than Judge Taranto.

[00:25:38] Speaker 03:
Can I ask you a different question?

[00:25:39] Speaker 03:
I understand your point.

[00:25:41] Speaker 03:
Okay.

[00:25:42] Speaker 03:
If we disagree and adopt your friend's construction, does that affect the validity argument at all?

[00:25:51] Speaker 04:
I don't think it affects the invalidity argument.

[00:25:56] Speaker 01:
If I remember right, the other side makes one point along those lines, namely in discussing the unexpected results component of the analysis that you had really, really good results only in part of the bubble range and not all of it.

[00:26:14] Speaker 01:
And if that bubble range shrinks, it changes that analysis.

[00:26:19] Speaker 04:
Right.

[00:26:20] Speaker 04:
let me answer

[00:26:30] Speaker 04:
make the validity case stronger on our behalf because the scope of unexpected properties would be narrower that we would have to show.

[00:26:37] Speaker 04:
I don't think any of that matters.

[00:26:39] Speaker 04:
The court found for at least four independent reasons separate and apart from unexpected properties that this claim is not obvious and those would all have to be overturned in any event.

[00:26:49] Speaker 04:
So I think the answer to your question is that the validity analysis wouldn't change.

[00:26:54] Speaker 01:
That's on the infringement question.

[00:27:00] Speaker 01:
Maybe it's just me, but I don't remember seeing in cases that I've sat on at least, this significant digit concept play a real role.

[00:27:21] Speaker 01:
Why should the app, first of all, have I just missed a lot of use of it?

[00:27:25] Speaker 01:
And second, if indeed it's fairly rare,

[00:27:28] Speaker 01:
What explains why this seems an unusual concept for claim construction in technology patents where numbers are around all of the time?

[00:27:42] Speaker 04:
Yeah, I'm not sure exactly how to answer that.

[00:27:44] Speaker 04:
I don't think it's altogether rare.

[00:27:46] Speaker 04:
I think the issue... You cited a few cases, but it was...

[00:27:49] Speaker 04:
alagan vs. teba case for example out of the eastern district of texas which judge bryson was sitting by designation we cited of course viscase which addresses this very question do we add an extra significant digit and i would submit that's the standard that the court applied there and should apply here does the intrinsic record mandate that based on statements and prosecutions i just wanted to follow up on that question

[00:28:11] Speaker 00:
I understand there's a number of cases, but one would think in pharmaceutical composition cases, where amounts of different materials seem to be so important, that there would be a ton of cases, or there would be some convention of how these numbers are going to be understood when there is a single member.

[00:28:31] Speaker 00:
And so, you know, again, going back to public notice and

[00:28:34] Speaker 00:
вы знаете, я имею в виду, есть ли какой-то понимательный способ, как мы всегда должны посмотреть на номер 1 и сказать, что это не просто 1, это 0.5 до 1.49, я имею в виду, как-то 1.4, я имею в виду, как-то мы должны смотреть на каждый номер в кlaimе?

[00:28:53] Speaker 04:
Ну...again, I think it depends on the context.

[00:28:56] Speaker 04:
I think they bring up the example of one apple, and obviously that's a different situation.

[00:29:00] Speaker 04:
One wouldn't come in with one and a half apples and say that's the same thing.

[00:29:03] Speaker 04:
That's something you count versus something you measure.

[00:29:06] Speaker 04:
But my understanding based on doing this for a few decades is that when there are concentrations in claims like 001, the person still looks at that with respect to significant digits.

[00:29:16] Speaker 04:
That's the basic way that this is understood.

[00:29:20] Speaker 00:
But there's no evidence to that effect.

[00:29:21] Speaker 00:
In this case, there is absolutely no evidence in this case of expert testimony or scientific true disease or anything saying that when a person who deals with pharmaceutical compositions looks at a number, they know that they are to look at .001 and understand that they wound up a little more down based on the significant digit.

[00:29:44] Speaker 04:
Both parties agreed about that below.

[00:29:46] Speaker 04:
It wasn't disputed that significant digits are the way to look at this.

[00:29:50] Speaker 00:
Mylon just added a second significant digit and said it's double... I understand, but I see what you're saying, but I have two different digits to pick from here.

[00:29:59] Speaker 04:
Correct.

[00:30:00] Speaker 04:
And I think the intrinsic record answers the question of how many significant digits there are.

[00:30:03] Speaker 04:
In Biscase, the specification said three significant digits.

[00:30:07] Speaker 04:
It repeatedly said 0.910.

[00:30:08] Speaker 04:
In prosecution, as the court noted at 1321 of that case, they said 0.910 as defined by the present claims.

[00:30:17] Speaker 04:
So they added a third significant digit, a trailing zero in that case, clearly to define their term.

[00:30:23] Speaker 04:
And so under the Phillips rubric, the court asked the question, is there enough here to change the typical meaning, the ordinary meaning, which everyone would agree to and see for 0.912 significant digits?

[00:30:34] Speaker 04:
In that case, they said yes.

[00:30:35] Speaker 04:
You have rejected anything above .910.

[00:30:39] Speaker 04:
You've said so in specification.

[00:30:41] Speaker 04:
You've said so in distinguishing claims.

[00:30:43] Speaker 04:
In distinguishing prior art that disclosed .910 to .940.

[00:30:47] Speaker 04:
And that's how you defined it in your intrinsic record.

[00:30:50] Speaker 04:
We haven't done that here.

[00:30:53] Speaker 04:
The court came to grips with it and looked at the intrinsic record and noted that there's only one significant digit every single time.

[00:30:59] Speaker 04:
No trailing zero.

[00:31:00] Speaker 04:
This is no more and no less than engrafting a number

[00:31:04] Speaker 04:
from nowhere, with no support in the intrinsic record, onto the claim where we didn't write it.

[00:31:10] Speaker 04:
They can't rewrite our claim.

[00:31:11] Speaker 04:
The law doesn't permit them to do that.

[00:31:12] Speaker 04:
And the Biscase case, I think, is probably the clearest example of how the court addresses the issue of significant digits.

[00:31:21] Speaker 03:
Understanding the claim meeting?

[00:31:24] Speaker 03:
Can I just ask you this kind of... This is more abstractly, although it's obviously related.

[00:31:28] Speaker 03:
If you were comparing the numbers 0005 to 001,

[00:31:32] Speaker 03:
would you consider them the same number or a different number?

[00:31:35] Speaker 04:
They're absolutely different numbers.

[00:31:38] Speaker 03:
They both have bubbles.

[00:31:39] Speaker 03:
They overlap.

[00:31:39] Speaker 03:
How do you know they're different numbers?

[00:31:41] Speaker 04:
Because they have different meanings.

[00:31:42] Speaker 04:
The 001 has a meaning of 0005 to 0014.

[00:31:47] Speaker 04:
The 0005 has a meaning of 00045 to 00054.

[00:31:52] Speaker 04:
They're different numbers.

[00:31:53] Speaker 04:
They were different embodiments in specification.

[00:31:55] Speaker 04:
We're not saying they're the same embodiment.

[00:31:57] Speaker 04:
If Your Honor held those two embodiments in their hand,

[00:31:59] Speaker 04:
In your hand, they would be two different embodiments.

[00:32:02] Speaker 04:
But one can't import a level of precision to how much PVP those embodiments had that this intrinsic record doesn't permit and assume that we know that's .0010 rather than .001.

[00:32:13] Speaker 01:
This specification doesn't say that.

[00:32:18] Speaker 01:
So the spec sometimes goes out to four decimal places.

[00:32:21] Speaker 01:
Now one possibility for why one would do that is one is just getting

[00:32:26] Speaker 01:
Smaller and smaller and smaller, getting closer to zero.

[00:32:30] Speaker 01:
And you've got no choice but to add a fourth decimal place or a fifth decimal place.

[00:32:36] Speaker 01:
That doesn't tell you anything about amounts of precision.

[00:32:41] Speaker 01:
Is there any use of four decimal places in the neighborhood of 001 or above it?

[00:32:50] Speaker 01:
No.

[00:32:50] Speaker 04:
Not with respect to PVP concentrations.

[00:32:53] Speaker 04:
Again, with respect to budesonide concentrations, they are more precise in specification, and it will say something like 75.0.

[00:33:01] Speaker 04:
That's telling the person of ordinary skill 74.8.

[00:33:03] Speaker 01:
I think this is correct, but let me just make sure I understand.

[00:33:07] Speaker 01:
It is just a...

[00:33:10] Speaker 01:
a mathematical corollary of significant digits that the size of the bubble shrinks as you get close to zero so there's more wiggle room the further north you go less and does that that's right right that's correct okay and is there anything about range of experimental error that makes it sensible to have

[00:33:37] Speaker 04:
keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene keene

[00:34:05] Speaker 04:
The wiggle room is smaller and the number is smaller.

[00:34:07] Speaker 04:
Where numbers get bigger, the wiggle room gets bigger.

[00:34:10] Speaker 01:
And that's just how... Well, that's the convention of significant digits.

[00:34:15] Speaker 01:
It doesn't have to be that way, but that's the convention.

[00:34:19] Speaker 01:
It's true that you have to go to more decimal points the further south you go.

[00:34:22] Speaker 04:
Correct, but again, at the risk of repeating, both sides argued below that significant digits were the issue, not the number of decimal points.

[00:34:28] Speaker 04:
If you looked at their argument, 7909 through 7915, that's entirely clear.

[00:34:34] Speaker 00:
Could you explain what you mean by that?

[00:34:38] Speaker 04:
Sure.

[00:34:38] Speaker 04:
So the number of decimal points, as I understand it, again, they never argued this below, is how many numbers follow the decimal point.

[00:34:47] Speaker 04:
So if you go out to .0001, that's four decimal points, and you just count that way.

[00:34:50] Speaker 04:
That may be relevant to something.

[00:34:53] Speaker 04:
It wasn't argued to be relevant to the scope of this claim below or in any case that I'm aware of from this court that they've cited or that I've ever read.

[00:35:02] Speaker 04:
Rather the question is how many significant digits there are.

[00:35:04] Speaker 04:
And significant digits are defined by the first non-zero after the decimal point.

[00:35:09] Speaker 04:
And then trailing zeros thereafter.

[00:35:11] Speaker 04:
So, so-called leading zeros, the two zeros in 001%, are not significant digits under the basic conventions of mathematics.

[00:35:18] Speaker 04:
If you wanted to add a second significant digit, .0010, that trailing significant digit is a second significant digit.

[00:35:25] Speaker 04:
It's significant.

[00:35:26] Speaker 04:
It would narrow the bubble.

[00:35:28] Speaker 04:
That's what they want to do.

[00:35:30] Speaker 00:
That's not what the specification... Maybe their main argument, as I understand it, maybe I think it's their best argument, is that

[00:35:37] Speaker 00:
the wall .001 has a different digit than .0005.

[00:35:45] Speaker 00:
By having the patent specification test to that level of precision, that is going down to .0005, it shows that

[00:35:55] Speaker 04:
My response is that assumes that the claim can only cover one embodiment tested in the specification.

[00:36:06] Speaker 04:
They are conflating two questions.

[00:36:08] Speaker 04:
How much is in each of those embodiments?

[00:36:10] Speaker 04:
With the trenchant question here, which is what is the scope of this claim?

[00:36:14] Speaker 04:
And it's not correct that just because 001 was tested differently or was better than 0005, that doesn't mean 0005 is outside the scope of the claim.

[00:36:24] Speaker 04:
Repeatedly the patent conveys, Column 1, Column 5, Column 6, that 0005 is within the scope of the invention.

[00:36:30] Speaker 04:
And we cited the Continental versus Intel case, and I would submit that as instructive here, even an embodiment that was disparaged.

[00:36:38] Speaker 04:
We have no suggestion that 005 is outside the scope of the invention.

[00:36:50] Speaker 04:
The patent says otherwise.

[00:36:55] Speaker 04:
No suggestion that 0005 is in the prior art.

[00:36:58] Speaker 04:
Everyone agrees that it's not in the prior art.

[00:37:00] Speaker 04:
So the notion that that should somehow be excluded because it didn't perform quite as well as 001 is completely unsupportable, I would submit, under this course of planning construction jurisprudence.

[00:37:11] Speaker 04:
Did that answer your honor's question?

[00:37:15] Speaker 04:
I'm mindful that I haven't addressed obviousness at all.

[00:37:17] Speaker 01:
Neither did your opposite number, so I think we're going to leave that where it lies.

[00:37:37] Speaker 02:
Благодарю, генерал.

[00:37:38] Speaker 02:
Хочу поделить одну вещь, которую я слышал в Константе Эстерземка, об использовании дополнительных цифровых точек.

[00:37:47] Speaker 02:
И, вероятно, я слышал, что нет причины для того, сколько цифров было использовано в спецификации.

[00:37:53] Speaker 02:
И, я думаю, что это противопоставление к Вискасе, которую партия дискуссировала в briefingе.

[00:37:57] Speaker 02:
Если вы посмотрите на 261, 3 и 1322,

[00:38:02] Speaker 02:
In supporting its construction to add another significant digit to the claim term in that case said, in addition, I'm quoting, the first family specifications illustrate use of the third decimal place for the densities of other classes of polyepilenes, referring to that as an important piece of intrinsic evidence supporting the construction that was used in that case.

[00:38:28] Speaker 01:
What came before the third decimal point?

[00:38:32] Speaker 02:
Only zeros or were there other numbers?

[00:38:35] Speaker 02:
I believe it was 0.91 and then this court added a zero after the one.

[00:38:41] Speaker 00:
Okay.

[00:38:42] Speaker 00:
I think I have that right.

[00:38:42] Speaker 00:
Yes.

[00:38:42] Speaker 00:
There was some prosecution history there that made that, that's what made that important, right?

[00:38:46] Speaker 00:
There was a prosecution history where they said that they, I think, defined it as 0.910 during prosecution.

[00:38:54] Speaker 00:
Is that right?

[00:38:54] Speaker 02:
They referred to the limitation in that way in prosecution and that was a clear signal to this court and to One Skilled in the Art that, you know, that level of precision, that was where the rounding should take place in that case because that's where they were drawing the line.

[00:39:09] Speaker 02:
а а а

[00:39:27] Speaker 00:
0.005, which shows that in the different testing that they did, that number is the most significant digit.

[00:39:37] Speaker 00:
Is that what your point is, or am I missing that?

[00:39:41] Speaker 02:
I think that's right.

[00:39:41] Speaker 02:
Those data there demonstrate that those formulations were different, and the 0005 was worse than the 001.

[00:39:50] Speaker 02:
and the construction that would conflate those two would not be correct.

[00:39:58] Speaker 00:
Is there any other piece of intrinsic evidence or even extrinsic in the record that helps me to know that that is the digit that I should be looking at?

[00:40:09] Speaker 02:
Я думаю, что мы говорили об интенсивных доказаниях, о которых мы верим.

[00:40:16] Speaker 00:
Но специально о этом?

[00:40:19] Speaker 00:
Есть ли что-то другое, которое предпочитает, что я должен посмотреть на это место?

[00:40:23] Speaker 00:
Кроме слов об этом и так далее.

[00:40:24] Speaker 00:
Есть ли что-то другое, где есть какие-то предпочтения о том, что нужно идти к четвёртому месту?

[00:40:32] Speaker 02:
Так что, я думаю, что я бы хотел посоветовать две вещи.

[00:40:36] Speaker 02:
As you noted in your questions to my colleague, there were claims that were presented that had ranges.

[00:40:43] Speaker 02:
There was a claim at Appendix 16-321, Claim 23 that was presented at one point that recited triple zero one, triple zero five, or double zero one, indicating that, again, that those are different formulations that would be considered differently in specification.

[00:41:04] Speaker 02:
I mentioned to Your Honor before that there was no expert testimony presented during the claim construction.

[00:41:09] Speaker 02:
However, Miland's expert, Dr. Pritchard, did touch briefly on the district court's claim construction, or the issue of claim construction during his testimony at trial.

[00:41:18] Speaker 02:
And this is at appendix 9778 to 9779, where he said that himself looking at this as one skilled in the art would look at rounding based on the recorded, the precision of the reported formulations, which is similar to the approach we've taken in our construction.

[00:41:34] Speaker 03:
I understand that.

[00:41:37] Speaker 03:
I mean, that kind of makes sense to me, but your opponent makes the point that, you know, in this field, when you're using these numbers with the one digit, that that's the important digit, and everybody would understand that the rounding would occur from that, so that even if they use a formulation with 0005, 0001, and then 001, they actually met

[00:42:02] Speaker 03:
They did not mean the same rounding for each.

[00:42:05] Speaker 03:
They intended different scope for all of those numbers.

[00:42:09] Speaker 03:
And so, 0005 may round to a much more limited range than 001, but that's what somebody reading this specification would understand from these tests.

[00:42:20] Speaker 03:
What's your response to that?

[00:42:21] Speaker 02:
So that might be the case if this, again, was just an abstract math problem.

[00:42:26] Speaker 02:
Somebody looking at just bare numbers on the page might think that's the way each individual one is rounded.

[00:42:31] Speaker 02:
there might even be cases where that's appropriate.

[00:42:33] Speaker 02:
I think the problem here in this case is that taking that approach with the intrinsic record we have here means that the 001 in the claims would mean something different than it meant in the specification.

[00:42:46] Speaker 02:
And that, I think, is an untimable situation in terms of claim construction.

[00:42:52] Speaker 03:
Well, what's your support for that?

[00:42:54] Speaker 03:
I mean, I still, you say that, but what in the specification suggests that

[00:42:59] Speaker 02:
0.01 0.01 0.01 0.01 0.01

[00:43:29] Speaker 02:
you know, that broad reaching that far.

[00:43:32] Speaker 02:
In addition, I don't think it makes a lot of sense.

[00:43:34] Speaker 02:
If you take that approach, there would be, you know, different magnitude of wiggle room depending on which formulation you're looking at.

[00:43:41] Speaker 02:
So the 0005 and 0001 would have plus or minus half a 10,000th of a percent.

[00:43:48] Speaker 02:
That magnitude would get 10 times bigger at the 001 formulations and then 10 times bigger again.

[00:43:56] Speaker 03:
Did they test any at the like 0.0 something range?

[00:43:58] Speaker 02:
Yeah, there was 0.01, 0.03 and 0.05.

[00:44:02] Speaker 02:
Those were all tested and again showed different performance in all the tests.

[00:44:08] Speaker 01:
And some up in that area were what was distinguished in the prosecution history.

[00:44:16] Speaker 01:
There's two sources.

[00:44:19] Speaker 01:
One was like 0.025 and one was much higher than that.

[00:44:21] Speaker 01:
Both of them north of the bubble range that the district court adopted here.

[00:44:26] Speaker 02:
There was prior art before the examiner that had numbers like that.

[00:44:32] Speaker 01:
This is, yeah, Weir and Meade.

[00:44:36] Speaker 01:
Both of them were above the northern end of the bubble range that the district court adopted here.

[00:44:41] Speaker 02:
Well, Meade did not provide a number at all.

[00:44:42] Speaker 02:
And I think this is an important point about the prosecution.

[00:44:45] Speaker 02:
If Your Honor looks to appendix 16205, that's where the examiner is discussing the obviousness rejection that the examiner was making.

[00:44:57] Speaker 02:
And the examiner is not citing any particular number from either of the primary prior references that are part of the rejection.

[00:45:04] Speaker 02:
The examiner is relying on an optimization rationale to get to the 001 that's in the claims.

[00:45:10] Speaker 02:
So that the examiner was not citing some higher number or some lower number.

[00:45:15] Speaker 02:
The examiner was saying, these references give us the ingredients, and one of skill in the art would optimize to get there.

[00:45:23] Speaker 02:
And so the applicants in responding to that were

[00:45:26] Speaker 02:
we're trying to differentiate their 001 as being, you know, unexpectedly superior, better versus the prior to get around that kind of rejection.

[00:45:44] Speaker 02:
I think your clock has run down.

[00:45:48] Speaker 02:
May I make one more quick point, Your Honor?

[00:45:49] Speaker 02:
Quick.

[00:45:50] Speaker 02:
I just want to

[00:45:51] Speaker 02:
Let the court know that there is some real world urgency for Mylan here in this particular case, as reflected by the party's stipulation to an expedited briefing schedule here.

[00:46:02] Speaker 02:
The district courts, we submit erroneous judgment in this case, has blocked Mylan's first file to ANDA and continues to do so until it's out of the way.

[00:46:14] Speaker 02:
So we would request that the court reverse the claim construction or reverse the infringement judgment that was based on that.

[00:46:21] Speaker 01:
Alright.