Teva v. Sandoz
Federal Circuit Oral Argument
Federal Circuit Oral Argument
Deanne Maynard argued in front of the U.S. Court of Appeals for the Federal Circuit where she obtained a ruling of patent invalidity for indefiniteness.
Unofficial transcript for users of mofo.com
Speaker 1 (00:51):
Ms. Maynard.
Deanne Maynard (00:53):
Mr. Chief Judge, it says I have eight minutes, but I believe the court granted me 10. Is it already counting out my rebuttal?
Speaker 1 (01:01):
That’s, yeah, I think that’s—
Deanne Maynard (01:07):
Okay. Thank you very much.
Speaker 1 (01:08):
[Laughing]
Deanne Maynard (01:09):
Mr. Chief Judge, and may it please the court. My name is Deanne Maynard, and I represent Appellants Sandoz and Momenta. And Mr. Chesler represents Mylan and is going to be presenting the arguments from their brief. I’d like to focus my time on our indefiniteness and prosecution history disclaimer arguments. First, as to indefiniteness. Teva concedes that a person of skill in the art would understand that one of three different average molecular weights are potentially at issue in the patents in suit: peak, number, and weight average molecular weight. This difference matters because the special molecular weight range is key to the less-toxic, yet still effective copolymer-1 claimed in these patents.
Speaker 1 (01:55):
The district judge spent a lot of—a marvelous opinion, a lot of effort on this. And they said a person of skill in the art would look at the graph, Figure 1, and be able to figure out where those peaks are. And you’re going to get to prosecution history in a moment here. But, Teva said that was just a mistake. The first statement was just a mistake. Why isn’t that enough to render this definite?
Deanne Maynard (02:25):
Well, let me address the expert first.
Speaker 1 (02:28):
Okay.
Deanne Maynard (02:29):
Because as an initial matter, this court has repeatedly and as recently as last month, held that, whether a claim is indefinite is a question of law. But of course, one can look at an expert to help explain the extrinsic—can use it as extrinsic evidence. But under Phillips, intrinsic evidence is still the key. And when you look at Dr. Grant’s declaration, even by his own map, Figure 1, which is key to the District Court’s error, the District Court says Figure 1 points to peak. Dr. Grant says Figure 1 points to peak. But the legend in Figure 1 says the average molecular weight of these curves is 7.7 and 12. Dr. Grant says the peak is not that, by his own measurements. Dr. Grant measures the peak at 6.8 and at 10, which is a significant margin of error from the information in the graph. By his own calculations, Dr. Grant calculated both number average molecular weight and weight average molecular weight from the data in [inaudible]. And significantly the number he came up with weight average molecular weight is closer to the numbers in the legend than his numbers for peak. Yet, he chooses peak. And in fact, his declaration says he assumes it’s peak because his numbers for peak are close to those in the legend. But his numbers for weight average molecular were closer to the numbers in the legend.
Speaker 1 (03:54):
But he goes on to explain where a very high level of skill in this art and that a person with ordinary skill is going to be able to adjust that and make it work.
Deanne Maynard (04:05):
But there’s no support in the intrinsic record, Chief Judge, for the fact that there’s any error in the graph. So that’s the kind of expert testimony that this court doesn’t accept, which is the ultimate question of what does the specification mean, is a question for the court. Regardless of even if you—whatever standard of deference you apply.
Speaker 4 (04:27):
Counsel, there are group patents directed to Group I claims and Group II claims. You follow me, know what I’m talking about right?
Deanne Maynard (04:34):
There are three different kinds of claims.
Speaker 4 (04:37):
Okay? But Group I claims and Group II claims, two different sets of claims. The Group II claims are the 75% claims. Now follow me?
Deanne Maynard (04:44):
Yes, I am with you.
Speaker 4 (04:45):
So the Group II claims have the limitation copolymer-1 having over 75% of its molecules within the molecular weight range from about 2 to about 20. That is not a claim that covers average molecular weight at all. It’s not asking for Mw, Mp, Mn, or anything else. I guess there are four types. I don’t remember number four—M-something else. So it’s not asking for any of those. It’s saying each—looking at every one of the individual pupils—75% of them have to be within this range. So it’s not—it’s irrelevant. There is no Mw or Mp there. So why is that claim term indefinite?
Deanne Maynard (05:23):
Because when you’re talking about this copolymer-1, all of the weights are necessarily averages. And even Dr. Grant considered—
Speaker 4 (05:32):
No, this limitation says, “Copolymer-1 having 75% of its molecules within the molecular weight range from 2 to 20 kilodaltons.” That’s each individual molecule. That’s not 75% of each of the individual molecules. Not average. That doesn’t make sense when you’re talking about each one.
Deanne Maynard (05:51):
What Dr. Grant, even some of the other claims that what I think you’re calling the Group I claim saying having a—
Speaker 4 (05:57):
Well, I’m calling these Group II claims. Group I claims would be the ones that actually turn on average molecular weight.
Deanne Maynard (06:02):
Yes. But even those claims say a copolymer having a molecular weight of 4 to 9. And Dr. Grant said that—
Speaker 4 (06:10):
No, that’s different because that’s the copolymer. So you’re talking about the average of all the molecules, but this one actually says 75% of each molecule—75% of all the molecules have to be within this. That’s not an average. I mean, it’s a simple mathematics. This isn’t complicated chemistry because otherwise I wouldn’t get it. So 75% of the individual molecules have to be in this range. So, molecular weight and any indefiniteness about Mw or Mp, it’s irrelevant to these claim terms. So I don’t see any indefiniteness argument for these.
Deanne Maynard (06:41):
Well, the indefinite argument is that when you separate species, you ultimately, no matter how thin you slice the species—and Dr. Grant testified to this—no matter how thin you slice it, you’re always going to have a mix of different size molecules in each slice, and therefore you’re necessarily talking about an average. And so even with respect to the 2 kilodaltons to 20 kilodaltons where you’re talking about the species—and I do agree with you—you’re still talking about an average and at the margins between 2 and 20, it will make a difference which type of average you’re talking about. And so it is indefinite. But even if the court thought there was a distinction that one could draw between these different claims, the ones that you’re calling the Group IV claims, Your Honor—
Speaker 4 (07:19):
Group II. Group II.
Deanne Maynard (07:21):
I’m, no—I’m now talking about the 4 to 9 and the 5 to 9 claims are indefinite because a person of skill in the art can’t discern from the patent.
Speaker 4 (07:29):
But I guess here’s my problem. If I don’t agree with you that the Group II claims are indefinite, because we’re talking about 75% of the total area under the curve, and it seems like it’s pretty definite to me, then it doesn’t matter whether the Group I claims are or not because the district court made a separate infringement holding on this basis, which I can look up and point to in the record—because I had my clerk look this up. So the entire thing’s in affirm it’s the bottom line because you’ve got an injunction here. That’s all that this is about is an injunction. It’s not about damages. So the injunction stands if any of the claims are infringed.
Deanne Maynard (08:00):
No, Your Honor, with respect, the ’808 patent has a longer expiration date. And the ’808 patent has one claim, and it’s what you’re calling the species one type of claim. And it’s invalid under our indefiniteness argument. And it does make a difference. In addition, a number—
Speaker 4 (08:16):
Okay. Wait. So what you’re telling me is, even if I am absolutely dug my heels in on these Group II claims, that I still need to look at indefiniteness for the Group I claims because the injunction—the length of the term of the injunction will be affected by the Group I claims, because that’s the longest life patent. Is that right?
Deanne Maynard (08:34):
Yes, Your Honor. And also some of the other claims—some of the other claims which are in Orange Book patents, which would also make a difference because the longest running patent’s not an Orange Book patent. And so the patents that have claims that only contain the 4 to 9 and 5 to 9 are indefinite under an argument. And even if you draw a distinction, those claims should be held invalid, and it will make a difference. So it does matter. And we do think that if there ever is an indefinite claim, these are they. It’s just like Honeywell. They’re trying to add in information that isn’t there using their expert, and they’re relying on the expert to trump the intrinsic evidence that is there. The evidence that is there says the average molecular weight in Figure 1 is 7.7, and their expert says, “No, it’s not, it’s 6.8.” And this Phillips is clear: the expert can’t be used, whatever is used for to trump. I would like to touch on the prosecution history disclaimer, because even if one draws a distinction, Judge Moore, for purposes of indefiniteness, the prosecution history disclaimer argument goes to all the claims because they made the statements in all types of patents. So they made the disclaiming statements in the molecular weight profile patents. They made the disclaiming statements in patents with the Group II-type claims that Judge Moore is referring to.
Speaker 5 (09:52):
Can’t they correct that with their second statement?
Deanne Maynard (09:56):
The prosecution of history disclaimers—Your Honor, I’ve switched arguments now.
Speaker 5 (10:00):
Yeah, I’m with you. Can’t they correct the mistake that they made that you say sets up an ambiguous dichotomy? Can’t they correct it by making a second statement?
Deanne Maynard (10:12):
No, if we’re—I was trying to move on to my second argument. But with respect to indefinite, it’s no. This court’s case law makes both of those prosecution history statements relevant, and they can’t correct it. And the answer, the reasoning given for the second answer, Your Honor—the peak answer in the prosecution history—is just as questionable as the answer that they gave, the explanation they gave with the respect to the first prosecution history because it points to Figure 1, which doesn’t show peak in the legend. But Your Honor, I was trying to make the point that they repeatedly and unequivocally disclaimed the weight range in the ’550 patent in order to get these new patents on their new copolymer. And whatever else these patents claim, none of them can cover a weight average molecular weight of 10 or more kilodaltons because they repeatedly gave up that to get this coverage. If I can save what little time I have left for rebuttal, I’d appreciate it.
Speaker 5 (11:08):
Okay, thank you.
Speaker 1 (11:17):
Mr. Chesler.
Evan Chesler (11:19):
Good[ASM(L1] morning, Your Honors. May it please the court. My name is Evan Chesler. I represent Mylan. I want to focus this morning on only one issue in the time—brief time I have—and that’s the non-infringement/infringement issue. Judge Moore, in a different context, just said something is simple mathematics here. That’s the only kind of mathematics I can understand, and that’s what I’m here to point out the court below erred with respect to.
Speaker 1 (11:41):
It’s pretty unusual where you—in fact, I can’t think of another case where you suggest that a bioequivalent doesn’t infringe.
Evan Chesler (11:50):
Well, Your Honor, it may be unusual that a bioequivalent doesn’t infringe. I agree with that. But there is, in this case, a very, very simple reason upon which the district court made a clear error—regardless of the review standard that’s applied, whether you look at this as a de novo standard, because it’s claim construction or a factual mistake—we get to the same place. And it has to do with one word, and that’s the word “approximately.” And the mistake the district court made is how it compared two ratios. And if I can, I’d like to use an analogy to try to make the point. I want you to assume, Your Honor, that there’s a model that predicts what the percentage of people in America is who will have each of four eye colors: brown, hazel, blue, and green. And the model tells you that the ratio of those people will be 6:5—6:2 to 5:1.
Evan Chesler (12:39):
And when you do the percentages, you come out with exactly the percentages that the district court came out with, which are captured on page 31 of our main brief in a table that we’ve prepared. Now assume you just take a thousand people off the street, and you look at them as a sample and you determine whether their distribution of eye colors is approximately the same as what the model says. The percentages that the district court calculated, which are 42.9 and 14.3, etc., where the district court just added—it took 14, which is the total of those numbers in the ratio, and divided each one by 6, by 2, by 5.1. Now suppose, Your Honor, it turns out that, in a thousand people, your model says—just as the judge’s ratio did—that you’d have 429 brown-eyed people, which is the alanine ratio.
Evan Chesler (13:24):
And in fact, what happens in your sample is you have 427. It’s a tiny difference. Turns out it’s two-tenths of 1%. Suppose for the hazel-eyed people, model says 143 out of the thousand, which is 14.3%, have 144. A small difference, 0.1%. For the blue-eyed people, the model says you should have 357 people of a thousand. And in fact you have 336—21 different, 21 less than the model predicted. It is a difference of about 5.5%. And lastly, green-eyed people. The model says you’d have 71; in fact, you have 92. Now that’s a 30% difference. The mistake the district court made here was it added the difference in percentages as if they were absolute numbers and said, “Well, the difference for the blue-eyed people is 2.1 percentage points, even though it was a difference of 21 people off of a base of 357.”
Evan Chesler (14:20):
“And the difference for the green-eyed people is 2.1 percentage points, even though it was a difference of 21 people off of an expected number of 71.” Those are simply not the same, Your Honor. Those 21 people represent a huge difference off the small base and a very small difference off the big base. And yet what the district court did was a simple mathematical error. She added up all those percentages and said, “It’s only a 4.5 percentage difference among all of them.” And that’s not a big difference because she—
Speaker 7 (14:50):
I’m not sure that the analogy holds. Does there—does it have any functional difference?
Evan Chesler (14:54):
Yes, it does, your Honor. It has functional difference. And I want to make a couple of points for why it’s a functional difference.
Speaker 7 (15:00):
Okay. Tell me.
Evan Chesler (15:01):
First of all, the tyrosine—which is the one amino acid for which there’s this dramatic difference—it’s 30% difference. Tyrosine is in fact very difficult, relatively more difficult to dissolve in water. And, in this circumstance, if you vary that one amino acid by 30%, it could have an effect upon efficacy. Yet the court never got to that.
Speaker 7 (15:27):
So it’s not a bioequivalent?
Evan Chesler (15:28):
It can be a bioequivalent, Your Honor. But if it has too much of that material in it so that it has an effect upon the environment in which it sits, which is water, it could have an effect upon the efficacy, even though—even if it is bioequivalent in terms of registering it in the Orange Book. There, this question here is not whether it’s bioequivalent for Orange Book purposes. The question is whether it’s an infringement under the patent claim. And I would submit to, Your Honor, that the word “approximately” has to mean something. Now, what did the district court do when she erroneously added up all these percentages and said the total of the percentage differences is 4.5? She said, “Let me find the example that had the greatest difference when you add them all up.” And it’s a particular example in the Teitelbaum reference, which is incorporated in the patent.
Evan Chesler (16:11):
She said the biggest difference there is lysine is 12%, 4.5 is less than 12, hence it must be approximate. Well, if she had done it the correct way, if she had looked at each of the amino acids and found—which the numbers plainly show in eighth-grade mathematics—that there’s a 30% difference for tyrosine. If she had gone to exactly the same example in exactly the same publication, batch two in Teitelbaum, she’d have found that the difference there is only 16%. And I would submit, at a minimum, the district court has to reconsider, on remand, whether a 30% difference for tyrosine is “approximately” the same as that ratio when in fact the greatest difference in the record here is 16%.
Speaker 7 (16:52):
Remind me, did the district judge do that on her own? Or was that following a formula that was presented by your opponent?
Evan Chesler (16:59):
Following formulas presented by our opponent.
Speaker 7 (17:01):
Right, so it wasn’t something she just dreamed up?
Evan Chesler (17:02):
No, no, Your Honor. She—
Speaker 7 (17:04):
She was doing what judges do—that’s listening to both sides and trying to pick one.
Evan Chesler (17:08):
And if one—yes, Your Honor, I agree. And if one side proposes something that’s mathematically wrong and the district court follows it, it’s clear error.
Speaker 7 (17:15):
And that you think that’s a question of—that’s not a question in fact?
Evan Chesler (17:20):
I think, well, frankly, I could argue either one. But let’s for this purpose—
Speaker 4:You could go either way. I like you. I like a guy like that. Very good lawyering.
Evan Chesler:I could go either way, but let’s assume—thank you, Your Honor. I like you, too. Let’s assume for purposes of this argument’s question of fact, the alternative argument which we’ve made in the brief and I think is a credible argument, is the word “approximately” must be understood in order to interpret the claim. Because it says a ratio—it’s calls copolymer-1, and the spec defines it as approximately 6:2 to 5:1. So under—
Speaker 4 (17:48):
The district court know the prosecution history in reference to other references to figure out what approximately meant. Why is that erroneous? I mean, that’s exactly what one skilled in the art would do here.
Evan Chesler (17:57):
Exactly, Your Honor. And let’s, so let’s take it as a question of fact in honor—in answer to Judge Benson’s question. Say it’s a question of fact, is it clearly erroneous? Because if you look at the very reference the district court went to, to determine whether on her erroneous methodology of comparison, the answer was approximately the same, she got approximately the same. She got 4.5 is less than 12%, it’s approximately the same. And what I’m saying to Your Honors is go to exactly the same reference, but instead of making the mathematical mistake she made, look, in fact, at the fact that when there’s a 21-unit difference on a base of 7.1%, it’s a 30% difference, not a 2.1% difference. And say is 30% close enough to be approximate? Go to exactly the same reference—batch two in Teitelbaum—and you’ll see that for tyrosine, the difference, which is the biggest difference of any reference that’s incorporated in the patent, is 16%.
Evan Chesler (18:49):
And I would submit to, Your Honor, that therefore under her own methodology of finding what is approximate, it would’ve been a completely opposite result and hence non-infringement had she not made that one mathematical mistake. Let me make one more analogy if I may, and then I’ll stop. I live in a small town in West Chester County in New York. If I ran for mayor of that town and I won by a thousand votes, it would be a landslide. If I ran for the governorship of New York and I won by a thousand votes, there’d be a recount. That’s the mistake the judge made—2.1% is not the same if they’re off of completely different bases.
Speaker 4:Doesn’t the difference in Mylan’s product really mean one amino acid?
Evan Chesler:No, Your Honor, because the size of these molecules, in fact, vary enormously. And some of them are huge, and some of them are smaller. They picked one as an example to make a very good rhetorical point. But, in fact, if you look in the real world at the actual drug, some of them are enormous numbers and the differences are much, much larger than that. So it’s not one molecule. I see that I’m within my—
Speaker 4:One amino acid.
Evan Chesler:It’s not one molecule of the amino acid. In fact, it depends upon which particular version of it you look at. And there are many, and some of them have huge numbers for that same difference of 30—29.6% could be an enormous number of difference in molecules.
Speaker 5 (20:01):
But you agree that the—under the statements made in the prosecution history, your product would infringe? If the statements made in the prosecution history are the way to properly consider what “approximately” means, then your product infringes.
Evan Chesler (20:14):
No, Your Honor, respectfully, I disagree. I disagree.
Speaker 5 (20:17):
Really? Because on pages 35 and 36, you agree. Thought that you basically agreed.
Evan Chesler (20:21):
No, Your Honor, if, in fact, the comparison were done correctly—By the way, in the prosecution—
Speaker 5 (20:28):
No, I’m not saying done according to your preferred methodology. I’m saying done according to the statements that they made in the prosecution history to further define or explain what this ratio should mean.
Evan Chesler (20:41):
Your Honor, even under that standard, they would not—until the trial, Teva had never in fact calculated the ratio other than by normalizing to the lowest populous amino acid, which in this case would be tyrosine [inaudible]. The first time they used the methodology that they proposed to calculate the ratio was at the trial. So under the prosecution history, it would not in fact have been in print. Thank you, Your Honors.
Speaker 4:Thank you Mr. Chesler. Thank you. Ms. Holland?
Elizabeth Holland (21:11):
May it please the court. I’m going to start with the point that Mr. Chesler just left off with. It’s Mylan’s position in this case that there is a 30% difference in tyrosine, and that’s the way you should assess infringement. And, you know, we agree with the district court—that’s not the right way to look at this issue. This is a composition. You have to look at all the components. But even assuming Mr. Chesler was right about the 30%, the district court had a specific factual finding after hearing all the evidence in the case that any difference in a single amino acid, even up to 40%, would be approximately the same to a person of ordinary skill in the art because that is the error associated with amino acid analysis at the timeframe, 1994. Dr. Kent, Mylan’s expert below, said the right way to look at the term “approximately” is to figure out what the error associated with amino analysis was.
Elizabeth Holland (22:07):
The district court didn’t agree with that, but said, “I’m going to look at the issue with that framework anyway,” and made a specific finding that anything in the range of 20 to 40% would be something that the person of skill would assume within the meaning of approximately. There’s an additional reason why Mr. Chesler’s argument can’t be right, Your Honor. It has to do with the process that’s given in the patent from making copolymer-1. Dr. Kent, again, Mylan’s expert below, conceded that if you follow the process in the patent using all high-quality reagents, high-quality HBR, you’re going to get something with Mylan’s amino acid molar ratio. It cannot be that you follow the process in the patent, you get exactly what Mylan gets, and then Mylan’s product is not approximately 6:2 to 5:1 and is not copolymer-1. Finally, on this point, Mr. Chesler mentioned—on page 31 in Mylan’s brief, Table 3—I think if you look at that chart, which shows the molar fractions of Mylan’s product versus the molar fraction in a sample that would be exactly 6:2 to 5:1. It’s striking how the two are almost exactly the same. And certainly it was not clearly erroneous for the district court to find that they were approximately the same.
Elizabeth Holland (23:34):
I’m going to turn now to the indefiniteness argument. A claim can’t be held invalid for indefiniteness unless it’s insolubly ambiguous. The district court here, in a very thorough opinion, found that the term “average molecular weight” would not be insolubly ambiguous to the person of ordinary skill in the art. And she approached the issue exactly as this court’s precedent told her to—
Speaker 5 (24:00):
The words “average molecular weight” in a vacuum would be ambiguous, right? Because it could be any of—I think it’s four different measures for molecular weight average. Is that correct?
Elizabeth Holland (24:10):
Yes, Your Honor, if you’re looking at average molecular weight with no context, without the specification, prosecution history could mean many different things. But we’re not doing that in this case. What the district court found is that a person of ordinary skill in the art here, high level of skill in the art—you know, Ph.D. in chemistry, a lot of experience in size exclusion chromatography—when they look at the specification and they see that the technique you’re supposed to use to determine molecular weight here is size exclusion chromatography, that tells them something very important. It tells them that they should assume that the average molecular weight means peak. It’s true that you can do manipulations and calculations and eventually get to weight average or number average, but the data that you get from an SEC chromatogram directly is just peak molecular weight. And the distr—
Speaker 4 (25:02):
How do you reconcile the two contradictory statements in the prosecution history? Are you asking us to just let the first one be a mulligan and only the second one counts?
Elizabeth Holland (25:15):
Your Honor, it—
Speaker 4 (25:16):
It doesn’t count in professional golf, does it?
Elizabeth Holland (25:18):
I guess not. I’ll take your word for it. No, what I would say about that, Your Honor, is that the person of ordinary skill, when he gets to look at the prosecution history comes in with the assumption based on the specification—we’re talking about peak average molecular weight here. He gets to the prosecution—
Speaker 5 (25:37):
What is the assumption based on the specification?
Elizabeth Holland (25:40):
Why is it, why is that—
Speaker 5 (25:41):
What is the assumption? You said starting with the—
Elizabeth Holland (25:43):
The average molecular weight means peak average molecular weight.
Speaker 5 (25:46):
No, why? Why does one of skill in the art reading this specification think that?
Elizabeth Holland (25:52):
Because the person of ordinary skill in the art, when they see the technique of size exclusion chromatography, that tells them that it’s peak average molecular weight unless they’re exclusively told otherwise.
Speaker 5 (26:02):
And where is your evidence of that?
Elizabeth Holland (26:04):
That was in Dr. Grant’s declarations, Your Honor.
Speaker 5 (26:08):
He says that because they say size exclusion chromatography in the specification that tells somebody it is peak molecular weight.
Elizabeth Holland (26:16):
Yes. Because we’re talking here not about a lay person looking at the specification, we’re talking about a person that’s skilled in the art, and the specific identification of the technique to use tells them what they should assume average molecular weight means. And that’s what Dr. Grant said in his declarations and explained quite clearly and—
Speaker 4 (26:35):
Is this contested that once it mentioned SEC, it was automatic, everyone knew what method you were using?
Elizabeth Holland (26:42):
It’s not contested that when a person of ordinary skill in the arts has an SEC chromatogram, the only thing they can get from that chromatogram without further mathematical manipulations is the peak. That’s not contested.
Speaker 4 (26:55):
Excuse me. Proceed. Well then, in the prosecution history, the first time you say a person of skill in the art would use weight average molecular weight. How does that square with what you just told us about the spec?
Elizabeth Holland (27:12):
Yes, Your Honor. So, when the person of ordinary skill in the art comes to the prosecution history, they already have the assumption in mind—based on the specification, this means peak. So when they get to prosecution history, they see two statements. They see a statement in the ’539 prosecution history, a definitional statement. It says, “average molecular weight means peak.” And that is completely consistent with and confirmed what they would’ve already understood based on the specification. They get to the statement of the ’847 prosecution history. First of all, it doesn’t make sense in the context of the rest of the intrinsic evidence which points to peak. But in addition to that, there is no dispute here, and Sandoz expert below conceded it is a simple scientific error. This statement is “a person of ordinary skill in the art could understand that kilodalton unit implies weight average molecular weight.” That’s just wrong. As a matter of science, kilodalton unit is nothing. It’s just a unit of measurement. Like the inch doesn’t tell you anything about the type of average molecular weight that’s being discussed. So in the context of the intrinsic evidence as a whole—the specification, which is telling you peak; ’539 prosecution history, which is telling you peak—the person of ordinary skill is not going to be confused by an erroneous statement in the ’847 prosecution history. I’d like to address Ms. Maynard’s comments about Figure 1 in the specification because I don’t believe Dr. Grant’s opinions there were stated clearly. What Dr. Grant said is the person of ordinary skill in the art, they’ve already seen the reference to SEC and the specification, which is telling them this is peak—
Speaker 5 (28:54):
But a person of skill in the art can calculate Mw with SEC also. So I don’t—I just don’t understand your argument that SEC means only peak. It doesn’t. A person of skill in the art, that’s how they calculate Mw also.
Elizabeth Holland (29:09):
Correct, Your Honor. It can be calculated, but the point is that a person of ordinary skill in the art, the only data that they get directly from the chromatogram is peak. So, unless someone is telling them that you’re talking about a different kind of average molecular weight, which of course, could happen—
Speaker 5 (29:23):
The only thing they get is peak. The problem is the figure that you have in the patent, the number on that figure doesn’t correspond to Mp. So, one of skill in the art looking at that is going to say, “Hmm.” In chemistry, especially when you’re talking about these sorts of things, it seems to me that precision is important, you know, and any—
Elizabeth Holland (29:47):
Your Honor, any technique you use in science is going to have some error associated, some lack of precision. When the person of ordinary skill gets to Figure 1, what Dr. Grant explains clearly is not exactly what Ms. Maynard said. Figure 1 is not itself an SEC chromatogram, it’s based on data that’s transferred from an SEC chromatogram. In the process of transferring that data and making a new graph, you would expect that the peaks to shift a little bit. And I can explain to you in a little more detail how the transfer takes place, but it’s in Dr. Grant’s declaration
Speaker 5 (30:22):
And the number is what, 6.6? Is that right? Six point—I don’t remember what the number was. What’s the number?
Speaker 1 (30:26):
7, 7.7 and 12. But they’re shifted off, just —
Elizabeth Holland (30:29):
They’re shifted a bit.
Speaker 4 (30:32):
—6.8, aren’t they?
Speaker 5
And it’s actually closer to Mw than Mp. Correct?
Elizabeth Holland (30:37):
Actually, Dr. [inaudible], Sandoz expert below, said you can’t even calculate Mw from Figure 1. It’s not possible. That’s not what the person of ordinary skill in the art is doing when they look at Figure 1. What they’re doing when they get to Figure 1 is confirming whether or not their understanding of peak from the specifications use of size exclusion chromatography is correct. What they see are the peaks in Figure 1 about where they would expect them to be, based on the process of transferring the data from the SEC chromatogram to Figure 1. When you transfer the data, you do it from the chromatogram in slices. This is explained in Dr Grant’s declaration, the process of transferring the data results in a shift piece. That would be understood by the person of ordinary skill. And that is why it’s critical here to look at the specification from that perspective, not as a layperson. What would the person of ordinary skill in the art understand about Figure 1?
Speaker 1 (31:29):
But, Ms. Holland, my concern is that the examiner raises this issue and says they presumably have skill in the art. They see that it’s SEC, and they raised the issue that there’s some indefiniteness here. And your response to them is, “I’ll use average weight molecular weight, not use the peak.” Doesn’t that send everything off-kilter?
Elizabeth Holland (32:00):
No, I don’t believe it does, Your Honor. Again, I think the person of ordinary still in the art is approaching this, just like the court is supposed to approach it, looking at all the intrinsic evidence together. Person of ordinary skill in the art sees the ’539 patent, which teaches peak, which is what they would’ve understood from the specification anyway. They get to the ’847 prosecution history, and there’s no dispute. It’s a clear scientific error. There’s no reason for the person of ordinary skill in the art to say, “I’m going to take that clear scientific error and assume that that’s what was meant by us.”
Speaker 4 (32:31):
What is clear scientific error?
Elizabeth Holland (32:32):
I’m sorry?
Speaker 4 (32:33):
What’s a clear scientific error?
Elizabeth Holland (32:35):
The statement in the ’847 prosecution history says a person of ordinary skill could understand from the use of the kilodalton unit that average molecular weight is weight average molecular weight. That is not correct. Scientifically, the kilodalton unit doesn’t speak at all.
Speaker 4 (32:52):
No, kilodalton is the unit of measure for any of them.
Elizabeth Holland (32:55):
Correct.
Speaker 4 (32:55)
I get that. But SEC can be used to determine any of them, too. So, I just don’t get how you could have it both ways. You made the statement, the examiner said, “Ambiguous, because I can’t tell whether you want Mw, Mp, or some other M.” You guys come along and say it’s “Mw”. There’s no other way to interpret this sentence because this overcame that rejection. Immediately after this, those claims were allowed, and this is the only statement you made that was responsive to that rejection. So, you have to—one of skill in the art has to walk away from this prosecution with the understanding that you are referencing Mw. There is no other way to read it. So, they’re walking away with that understanding. So, then the question is, can you overcome that by coming along later and correcting it? Or, does this only add to the indefiniteness?
Elizabeth Holland (33:44):
Your Honor—
Speaker 4 (33:44):
That’s the question. I’m not going to—you’re never going to convince me that this statement doesn’t mean you have said it’s Mw. So, get to the next point, and try and convince me why I should say it’s not indefinite after you’ve said it’s one thing to get one patent and then said it’s a different thing to get a continuation later on down the road.
Elizabeth Holland (34:03):
Well, Your Honor, I mean, we start off with a premise of because this is an issued patent, it has a presumption of validity, and there has to be—it has to be insolubly ambiguous in order to get past that presumption of validity. In this case, you have a lot of evidence from the intrinsic evidence teaching you that it’s peak, regardless of whether you’d say that there was some piece of evidence that teaches away—
Speaker 4 (34:26):
But, you know, the same examiner was on both of these, which really cracks me up. Mr. Kraus, the same examiner, was on both the parent and the continuation. And both times, you are saying that one with skill in the art, based on this specification, would know it’s Mp. But both times, he rejected all of these claims, saying, “I can’t tell from this specification whether it’s Mp or Mw.” He was quite clear. So—
Elizabeth Holland (34:50):
Well, the fact that there was a second rejection, Your Honor, after the initial one that had the statement about weight average molecular weight—
Speaker 4 (34:58):
Trust me, it’s not lost on me. The fact that he rejected you again after you said Mw the first time, he obviously forgot. I don’t know. I don’t know what else to attribute that to.
Elizabeth Holland (35:08):
It could be attributed, Your Honor, to the fact that it—the statement that Mw could be understood from the kilodalton unit is just not right and doesn’t make sense.
Speaker 4 (35:17):
Well, he allowed that patent, that patent issued, and you enforced it, so—
Elizabeth Holland (35:22):
Correct, Your Honor. But—
Speaker 4 (35:22):
It’s—so it’s not the fact that he looked at that statement and concluded that was inadequate. He allowed it, and you enforced it.
Elizabeth Holland (35:30):
The second statement in the ’539 prosecution history is the one that would comport with the understanding of the person of ordinary skill in the art based on the specification.
Speaker 4 (35:39):
Really? Because the second statement in the prosecution history strikes me as ambiguous and confusing as the first. Accordingly, the term average molecular weight is—oh, I’m sorry, I’m reading [inaudible] rejection. “One of ordinary skill in the art, given the specification would understand that average molecular weight refers to molecular weight at the peak of the molecular weight distribution curve shown in Figure 1, which is the percentage of total mass versus molecular weight. Figure one, the average molecular weight is 7.7.” The problem is that’s not right. Molecular peak weight is six point something, Mw is closer to 7.7. So, I think one with skill in the art would be as confused by that second sentence as they were by the first.
Elizabeth Holland (36:19):
The only evidence in the record before this court about what the person of ordinary skill in the art would be able to understand from the specification is Dr. Grant’s declaration—
Speaker 4 (36:28):
Numbers don’t lie. 7.7 isn’t Mp. So, if you have some expert that says, “Oh, one would understand 7.7 is Mp”, and everybody in the room knows 7.7 isn’t Mp. I mean, what am I deferring to him for?
Elizabeth Holland (36:41):
You were deferring to him, Your Honor, for an understanding of how—what Figure 1 is, what it’s meant to represent, and how it relates to the SEC chromatogram. That’s the original data that comes from the use of the size exclusion chromatography. There is no expert on Sandoz’s side disputing that you would have a shift in the peaks, Your Honor. There’s nobody on Sandoz’s side saying from an expert perspective, “I disagree with Dr. Grant, you wouldn’t expect that peak shift.” That is the evidence in the record of what the person of ordinary skill in the art would understand about Figure 1. There’s no basis in the record to dispute that on anybody’s part, the scientific matter based on a high level of skill in the art here.
Speaker 4 (37:22):
Well, you said there’s nothing in the record except the fact that the examiner rejected it despite that disclosure twice. And one time you said Mw, the next time you said Mp. So, guess what? Maybe the spec wasn’t ambiguous until you guys started answering all those rejections, and you made it so by going one way once and the next way the next time. I don’t know. I mean, it’s just hard for me to reconcile how one skilled in the art is supposed to walk away from this prosecution history with a clear understanding of what you all mean.
Elizabeth Holland (37:48):
As I said, Your Honor, the way to do that is to look at the intrinsic evidence as a whole, the person of ordinary skill in the art comes in with the assumption it’s peak. When they read the prosecution history, when they see the statement in the ’539 prosecution history, that confirms what they would understand based on the specification, Your Honor. Just to touch briefly on the prosecution history disclaimer argument that Ms. Maynard touched on briefly.
Elizabeth Holland (38:22):
The claims at issue here have a molecular weight range of 5 to 9 kilodaltons. The prosecution history statements that Sandoz points to are about 10 kilodalton copolymer-1. They don’t have anything to do with the scope, the literal scope of the claims as issued. But even if they did, there was nothing in the prosecution history that would tell the person of ordinary skill clearly and unambiguously that the references to 10 kilodalton copolymer-1 refer to weight average molecular weight copolymer-1. There’s just nothing in the prosecution history itself. What Sandoz does is constructs a very elaborate daisy chain going from the prosecution history to the ’550 patent, which itself does not talk about any particular kind of average molecular weight, to the ’71 Teitelbaum paper, which is not incorporated by reference in the ’550 patent. It’s just cited in the prosecution history of the ’550 patent.
Elizabeth Holland (39:21):
Go from there to the ’71 Teitelbaum papers’ reference to a Spinco E ultracentrifuge and make the argument that that reference in the ’71 Teitelbaum paper would somehow be a clear and unambiguous disavowal of claim scope of weight average molecular weight. It’s just too far from clear and unambiguous to be a disclaimer in this case. In addition, even if you did that whole analysis, it would ultimately fail, because the ultracentrifugation can give many different kinds of average molecular weight in addition to weight average molecular weight. That’s what the district court found as a matter of fact, and her conclusion should be affirmed. Any more questions? Or else I’ll stop here.
Speaker 5 (40:10)
Thank you, Ms. Holland. Ms. Maynard, you have 16 seconds.
Speaker 4 (40:18)
Speak quickly.
Deanne Maynard (40:20):
Both of our arguments are about notice, notice about what these patents cover.
Speaker 4 (40:23)
Talk about the SEC stuff. Because that, to me, is a very compelling argument on her point.
Deanne Maynard (40:29):
Well, as you correctly note, SEC can get you one of three different types.
Speaker 4 (40:33):
Only one can be read off the graph, right? The rest have to be calculated.
Deanne Maynard (40:35):
But there’s no basis to assume that one of these very highly skilled artisans going to take the easiest path. And what Ms. Holland ignores is the Teitelbaum article, which is incorporated by reference in these asserted patents, which everyone agrees shows a method that doesn’t get peak. The Spinco Model E ultracentrifuge, we say only gets weight, and that’s what the testimony shows. But no one says it gets peak, so it teaches away from you.
Speaker 5 (41:01)
I’m not sure that’s incorporated by reference. It is cited.
Deanne Maynard (41:05):
it’s not incorporated by reference in the ’550 patent, Your Honor, but it is indeed incorporated by reference in the very first column of the specification of all the asserted patents here. And so, when one’s trying to figure out what the asserted patents cover, one would look to the Teitelbaum article. And the Teitelbaum article says that the weight in that article—and the articles written by three of the named inventors of the asserted patents here, describes the ’550’s invention as being measured by ultracentrifugation. All the experts agree that doesn’t get you peak. So, it points away from peak. So, with the evidence—
Speaker 1 (41:41):
This is so insolubly ambiguous, why did a district court, after hearing five weeks of testimony, find it to be otherwise?
Deanne Maynard (41:51):
No, Your Honor. She found it to be not indefinite before the trial, after a claim construction hearing—
Speaker 1 (41:57):
She went all the way through the trial, heard everything. If she thought she’d made a mistake, she would’ve corrected it.
Deanne Maynard (42:02):
This wasn’t an issue at the trial, Your Honor. The trial—we preserved our objection. The trial proceeded on the assumption that it meant peak and, and this issue was not revisited. So, this was decided on the papers only on argument. And what she essentially did is use the expert to trump the intrinsic record, which points nowhere in terms of peak—in terms of what the average molecular weight means.
Speaker 4 (42:26):
Are you saying the statement in the spec, “All references cited herein are incorporated by reference in their entirety,” gets you Teitelbaum and gets you molecular weight Mw? Because we have a lot of case law—I mean, I don’t see another incorporation reference. Am I missing it? Maybe Column 1 contains an express incorporation of Teitelbaum, or a particular thing for which it stands? Because I’m looking at Column 1, and the only place where I see the words “incorporation by reference” are in a sentence that says, “all references cited here and are incorporated by reference.” And, Ms. Maynard, under our case law, if I’m—I mean if I’m mistaken and it appears somewhere else, please direct me to it. But under our case law, we’ve held that statements like this, I believe, are not enough to incorporate specifics in references into this one because this has too much generality. It’s like, why wouldn’t every patent say that all the time, then you have to incorporate my reference with specificity, with the particular component.
Deanne Maynard (43:23):
The line preceding where you’re reading, Your Honor, Line 14. That is the Teitelbaum article, the 1971 article, in two lines above where you are.
Speaker 4 (43:32):
Okay, I’m on A-847. So, am I in the right patent?
Deanne Maynard (43:34):
Oh, I’m sorry. If you look at A-344, I’m in the ’808, the first issued patent, but all the specifications are shared. I would—
Speaker 4 (43:41):
’808.
Deanne Maynard (43:42):
Okay, I’m on the ’808 patent. I’m looking at A-344.
Speaker 4 (43:45):
Okay, I’m on the ’808 patent. What line? 14?
Deanne Maynard (43:48):
I’m at line like 14. See the “Eur. J. Immunol 1971”?
Speaker 4 (43:53):
Yeah, but that doesn’t say anything about incorporating by reference or anything.
Deanne Maynard (43:55):
Then the next, very next sentence, “all references cited herein are hereby incorporated by reference.”
Speaker 4 (43:59):
Yeah, there’s no reference to that Teitelbaum reference or worse, a particular form of specificity within Teitelbaum. How long—isn’t that a three-volume treatise? Am I mistaken? Is that really—is that the long one? Hello? Who’s answering me? Not even a nod. Teitelbaum. What? What? You just going to get everything in automatically?
Deanne Maynard (44:23):
No, Teitelbaum is reprinted in—the 1971 Teitelbaum, Your Honor, is reprinted at A-49040. And it says on A-49042—
Speaker 4 (44:37):
No, but is my—am I understanding wrong of incorporation by reference law, Ms. Maynard? I’m sorry, we’re way over. But just yes or no, don’t you have to under incorporate by reference with specificity?
Deanne Maynard (44:47):
My understanding of your incorporation by reference law, which is cited in our briefing, is that it in things that are items that are incorporated by reference are incorporated in toto as if written in the patent itself. But even beyond that, Your Honor, if the person who’s skilled the art sees SEC, knows that can give you at least three different weights. There’s no reason to assume this person who’s skilled in the art is going to do the easiest thing. And they’re trying to figure it out. They’re going to look. But this article, written by three of the named inventors of the asserted patents, see how they were measuring it and it points away from P.
Evan Chesler (45:17):
Okay, thank you Ms. Maynard.
Deanne Maynard (45:18):
Thank you. Thank you for your indulgence.
Speaker 1 (45:24):
First time anybody said I’m indulgent.
Evan Chesler (45:30):
So I have 16 seconds. I’ll have—
Speaker 1 (45:32):
You have 16. No, just second. You have—you better start.
Evan Chesler (45:37):
I better start. But Your Honor, counsel with respect was incorrect when she said that Mylan uses the patented process and therefore must get a bioequivalent result. Neither Mylan nor Teva uses that process. It—they both use the process of a separate patent Delinsky, which appears in the record at A-32668 and following. And indeed, Teva went and got a patent on a process that fixed the problem by adding phenol. That fixed the fact that you lose 30% of the tyrosine, which creates the discrepancy which the district court missed by doing the math—wrong mathematical calculation. So, in fact, counsel is wrong when she said that we both practiced the process that’s in the patent.
Speaker 1 (46:14):
Okay. Thank you, Mr. Chesler.
Evan Chesler (46:15):
Thank you, Your Honor.
Practices