# Oral argument — Abbott Labs v. Sandoz > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A3399 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** December 3, 2007 ## Text Next case is number 07-1300 Abbott Laboratories v. Sandoz, Inc. Mr. Feder Good morning, Your Honors. May it please the Court. We are here because the District Court improperly granted preliminary junction against Sandoz in this case. Following first principles of patent law, the record shows that the 718 claims are vulnerable to obvious mischallenge and that Sandoz's product is not infringed. I will address the claim construction first, followed by non-infringement and obviousness determination. The District Court reached a claim construction for pharmaceutically acceptable polymer, a claim term in the claims at issue, that is an error. The term pharmaceutically acceptable polymer is defined in the sentence spanning columns 3 to 4 of the 718 patent specification, and a scientist of ordinary skill in the art reading this patent would know what a pharmaceutically acceptable polymer is. At that passage, the inventors state, quote, The pharmaceutically acceptable polymer is a hydrophilic water-soluble polymer selected from the group consisting of, and then they enumerate seven or eight classes of polymer structures. Judge Kaur did not give meaning to the term pharmaceutically acceptable polymer as it is defined in the patent. First, Judge Kaur failed to give the word is in that sentence its due meaning. The sentence, a pharmaceutically acceptable polymer is a hydrophilic water-soluble polymer, et cetera, et cetera. He purported to follow the ruling in the Andrix case, and in the Andrix case the court said the word means signifies better than the word is when somebody is trying to be his own lexicographer. But the Andrix court did not say that the word is cannot signify a definition. So he purported to follow that, but what the court was saying is that a means b might signify a little better than a is b, but we would submit that a is b is perfectly understandable and clear to the person of ordinary skill in the art. Second, Judge Kaur ignored the intrinsic evidence of record showing that the inventors, when they wanted to give a limited group and define something, used the word is in the sentence of a pharmaceutically acceptable polymer that I just discussed. Versus when they wanted to give examples of compounds, they described pharmaceutically acceptable excipients in another paragraph that says they are compounds such as, and then it enumerates examples. We submit that the intrinsic evidence showing the juxtaposition between the pharmaceutically acceptable polymer definition and the exemplification of pharmaceutically acceptable excipients shows that the pharmaceutically acceptable polymer language defines that claim term. Third, the language selected from the group consisting of has plain meaning. The Andrix court said, oh, that's Marcouche language, and Marcouche language has no meaning in the specification. Sandoz submits, Your Honor, that the words selected from the group of a, b, c, d, and e clearly signify to a person of ordinary skill in the art that f, g, h, and i are not in the group and they cannot be selected from the group because they're not there. Fourth, judge court erred in relying on the Andrix holding that the meaning of pharmaceutically acceptable polymer is broader than that stated in the definition. That holding was based on an admission that is only of record in the Andrix case. In the Andrix case, Andrix's expert admitted incorrectly that there were certain eutrogyte polymers, methacrylic acid copolymers, that were water insoluble. That is not the record in our case. In our case, Dr. Chambliss, Sandoz's expert, says that eutrogytes have, in fact, water-soluble hydrophilic polymers. And also, the intrinsic evidence is consistent with this. The W0422 publication, which is of record in this case, it's cited on the face of the patent, has at page 25, lines 4 to 5, which is appendix, joint appendix, page A2247, it states that the acrylic polymers are water-soluble at pH greater than 5.0. This shows that those persons of ordinary skill in the art, when they're talking about water-soluble eutrogytes, or eutrogytes, which are the methacrylic acid copolymers, understand that term, water-soluble, and that it was used consistently in the 718 patent specification. One other way we know it was used consistently in the 718 patent specification is because claim 3, which contains the Marcouche group, depends from claim 2, which requires water-solubility. So, even Abbott's inventors concede that water-soluble polymers are, there are species of the eutrogyte methacrylic acid copolymers that are, in fact, water-soluble. Additional intrinsic evidence is in the written description, because the written description contains no description of water-insoluble polymers. Had somebody tried to put in a claim claiming, wherein the pharmaceutically acceptable polymer is a water-insoluble polymer, a new matter objection would have been appropriate, because there is no description in the specification of insoluble polymers. Fifth, Judge Core read into the claim a matrix requirement. The word matrix has no place in the interpretation of pharmaceutically acceptable polymer. The specification expressly states that the compositions of the invention include suspensions, and suspensions are not matrices. Sixth, the claim differentiation presumption on which Judge Core relied is rebutted by all of the intrinsic evidence that I have just discussed. Turning to non-infringement, there is no dispute in this case that there's no literal infringement under Sandow's claim construction. Mr. Fetter, just going back to the issue of obviousness for a minute on the PK limitations. Yes, sir. Was it pretty well known in the art at the time about the PK limitations and what range of limitations to use within the particular scope of the claims? Yes, Your Honor. If you want to talk about the PK limitations, the person who organized... Is that the real issue here, whether or not the PK limitations were obvious at that point? I don't think so, Your Honor. I would point your attention to the 422 publication at page 7, which is the last paragraph in the summary of the invention. And this is where the inventors of the 422 publication are trying to explain why the azithromycin-extended release compositions are patentable. And in there, they say that it was known in the art to provide reduced dosing frequencies for short half-life compounds and to reduce fluctuations in plasma concentrations. Reducing fluctuations in plasma concentrations is one of the PK limitations that Judge Kaur said is not known in the prior art. Only Adam invented that. But this passage here reinforces and shows in the first instance that the person of ordinary skill in the art is a person of ordinary creativity. This person is imbued with the knowledge of extended release compositions, and this person knows that when you put a pharmaceutically acceptable polymer, such as HPMC, the most common polymer to be used for making controlled-release substances, that you're going to get a reduced fluctuation. Now, reduced fluctuation is defined mathematically. C-max minus C-minimum over the C-average. Well, the C-max is going to be lower. It inherently is lower when you put in the polymer that lower C-max. That's what people of ordinary skill in the art know. And the AUC is directly related to the AUC area under the curve is directly related to how much drug gets into the system. Well, under the CFR, under the regs, 37 CFR 325.20, I believe, if I've stated it correctly, says that when you're making an ER, an extended-release composition, you should match the area under the curve of your proposed ER formulation to the IR formulation. So here again, a person of ordinary skill in the art would be motivated and would reasonably expect to be able to create an extended-release formulation where an AUC of the ER matches the AUC of the IR. With respect to the C-minimum limitation, well, first of all, let me say this. With respect to Claim 1 of the 422 patent, it requires that the C-max, I'm sorry, it requires that the DFL, the fluctuation index, be lower, and that the AUC be equivalent to the AUC of the immediate release. Well, based on the 422 publication telling what the person of ordinary skill in the art knows, plus based on a true analysis of what a person of ordinary skill in the art formulation does know, he would understand that when you put in HPMC, when you put in a polymer with a composition such as And so Claim 1, I'm sorry, you're going to be motivated to get AUC equals to the AUC of the IR, and that's a reasonable expectation you'll get that. Claim 1 would have been obvious under a proper analysis, especially under KSR. With respect to Claim 4, it requires the PK limitation that the C-max is lower and that the C-minimum and the AUC are substantially equivalent. Well, I've already explained why the person of ordinary skill in the art would understand that the C-max would be lower. If the fluctuation is lower, the C-max is going to be lower. Also, the AUC, why it would be equivalent. The C-minimum, a person of ordinary skill in the art wants to keep the effect of the antibiotic across the entire 24-hour period. So the person of ordinary skill in the art is motivated to keep the C-min about equal to what the C-min of the IR is given twice a day. That allows the drug to have effective amounts. So these are all reasonable pharmacokinetic parameters. They'd be well understood to be the ones that are desirable. And Abbott's inventor, Linda Gustafson, at deposition admitted that lowering the C-max, keeping the AUC the same, and keeping the C-min about the same would all be reasonable expectations. I see I've only got 3 minutes and 15 seconds. Yes, you do. How about saving that for rebuttal? I will do so. Thank you, Your Honor. Thank you, Your Honor. Thank you, Mr. Fader. Mr. Dane. Thank you, Your Honor. May it please the Court. Since Mr. Fader focused on the issues of plain construction and obviousness, that's what I will focus on unless the Court has particular questions about the other issues in the case. There is a question that I would like to at least discuss regarding whether or not the landscape has changed for preliminary injunctions since eBay was issued by the Supreme Court. Thank you. Thank you, Your Honor. I'm glad you raised that because although so far this Court has not addressed the issue of whether the presumption of irreparable harm in the preliminary injunction context should survive eBay, in our view, clearly it should. The eBay decision was limited to the specific point of whether in a permanent injunction context the four-factor equitable test should be applied, where this Court in the case under review had not applied the four-factor test and had issued a bright-line rule that said that you automatically get an injunction unless there are exceptional circumstances, to bring the patent law into line with other areas of law, including copyright, which is important. The Supreme Court said in all instances the four-factor test should be applied. There's no question about that in the preliminary injunction context. Courts have, in patent cases as in other cases, applied the four-factor test. That's what the Court did here. The irreparable harm presumption is a separate question. It's a question as to one of those four factors, whether there should be a presumption applied in the event that one of the other factors, the likelihood of success on the merits, has been demonstrated. Nothing in the Supreme Court decision indicates that that should be changed, and in fact the Supreme Court itself admonished this Court to act in accordance with other areas of law. In the copyright law context, the presumption of irreparable harm does apply, where there has been a substantial showing of likelihood of success on the merits, and this Court itself has recognized that in the case of Atari Games against Nintendo America, 975F2nd832. So our position is that clearly eBay did not disturb the presumption of irreparable harm work. Their case of infringement has been shown. But that presumption still can be overcome by evidence submitted. Absolutely, it is only a presumption, Your Honor, and it's not an irrebuttable presumption. It can be rebutted. That is correct. So I've answered the Court's question. I'll turn to the issue of claim construction. And I note that two of the members of this panel were members of the panel that decided an earlier appeal involving the same patents with a related party, which was the Teva case. I did not hear counsel in argument urge a point that they raised in their briefs, was that the Andrix and Teva decisions on claim construction were in conflict, and therefore the earlier Teva decision should govern on the issue of claim construction. Our view is that certainly is not correct. With two members of the panel here, the Court, I'm sure, is aware that in the Teva case, the issue of claim construction was not at issue. There was conceded infringement, interestingly conceded infringement, involving a polymer that is not among the exemplary polymers that are included in the specification. The only issue in the Teva case was obviousness. The obviousness prior art that was cited all involved the preferred pharmaceutically acceptable polymer of HPMC. The District Court noted in its preliminary injunction decision that there was no claim construction issue. This Court did comment in its appeal that it saw no reason to disturb the Court's preliminary claim construction, but that claim construction was not necessary to its decision, so for that reason alone it should not be binding in any sense on future cases. And second, this Court has recognized that in the preliminary injunction context, courts can issue rolling constructions that change as more evidence comes in. So we, and to be clear, we are not taking the position at all that this panel is necessarily bound in any legal sense by the Andrick's decision. It is not prohibited from revisiting the claim construction issue. However, based upon the evidence in the record, there simply is no reason to depart from the Andrick's decision. In the Andrick's case, it was very important to get the claim construction correct because that was a case that was based upon an infringement argument under the doctrine of equivalence. And of course the first step in determining equivalence is determining what is the proper claim construction of the literal scope of the claim term before you determine how long the scope, or how extensive the scope of equivalence may be. All of the arguments pretty much that have been made by Andrick's here were raised in front of the Andrick's panel. But here we're talking about literal infringement. We are, but claim construction is still claim construction. And the analysis, Your Honor, I think remains the same in trying to determine what pharmaceutically acceptable polymer means. The court would have to take the same analysis as a first stage of literal infringement as it would for equivalence. But is the real issue here in claim construction the PK parameters? Should we focus on the PK parameters at all? I think that is relevant in terms of the validity issues, Your Honor. In terms of the claim construction issues, the only term that was raised as being in dispute was pharmaceutically acceptable polymer. And so separate from that were all the elements of what the PK requirements were. There was no dispute as to what they meant in terms of the requirement of a statistically significantly lower EFL or a substantially equivalent Cmin or AUC. So I think that those are very important on the validity side, but I don't think they're implicated by the construction of pharmaceutically acceptable polymer. On the pharmaceutically acceptable polymer point, I just briefly would point out that Sandoz has focused very much on a single sentence in the specification, which does use the word is. This court has previously indicated in the Pfizer-Teva case, 429F1364, even when a specification includes a word such as is, and in that case involved IE, which the Latin is for that is, that that's not necessarily definitional. The court always must look at the intrinsic evidence as a whole. That is what the Andrix Court did. Here we have a patent that has a section that provides definitions. It includes 10 different terms in quotes, and it says this means that, this means that. It goes through 10 different terms, and then it gets into a description of exemplary ingredients and components of the patent, and it's from that portion of the specification that they're attempting to derive and characterize something as a definition. In addition to the fact that there is a definitional section in which this term is not included, the claims themselves, claims 1 through 3, are differently worded and clearly worded in a way in which the only difference among the claim terms is that pharmaceutically acceptable polymer is used in claim 1, in claim 2 there's a further limitation that the pharmaceutically acceptable polymer must be water-soluble and hydrophilic, and in claim 3 the Markus Group, which they point to in the specification, is a further limitation in dependent claim 3. We noted that when Sandoz said that the doctrine of claim differentiation can be disregarded, and we recognize it is not an absolute rule that is determinative or dispositive of claim construction, but we noted that they were unable to find a case in which the court had found that dependent claims that differ from an independent claim only in one particular respect, that the court would interpret all of those claims to be of exactly the same scope, which would essentially render the dependent claims in violation of the patent code, because a dependent claim must have an additional limitation over an independent claim. In their reply brief they cited one case, the fantasy football case. That case is distinguishable, as are all the other cases that they cited, because in those cases, the prosecution history, the patentee, was required, in order to get the claim issued, to give up the subject matter that it later tried to get in its independent claim. And when it said that the independent claim was much broader than the dependent claim, that was rejected, because the patentee had specifically given up that subject matter in the court's prosecution history. We do not have that here. This is not an instance where, as this Court has said, the intrinsic evidence mandates that the doctrine of claim differentiation should be disregarded. And it is also not an instance where they have overcome the normal presumption that where the claim term, pharmaceutically acceptable polymer, clearly has the broad meaning that we indicate for it. A polymer is just a compound, a high molecular weight compound consisting of a link of joined monomers. And pharmaceutically acceptable means is one that can be ingested into the body. They have not overcome the presumption that that is the scope of the claim that this Court should find. With my remaining time, I'd like to turn to the issue of obviousness. Could you cover the PK parameters at this point, please? As to whether or not those were well known in the art? I will do that now, Your Honor. Thank you. They have argued that, and I should be clear as to the PK parameters, they cite one of the inventories as indicating that oh, these were obvious, everybody knows this is the profile you want for an extended release composition. That is not what the testimony says, and I would urge the Court to go back and read the entirety of that testimony. Well, she said at one point that, speaking to the PK parameters, a few years' experience, I guess, where did you get these parameters by a few years' experience? I mean, certainly the CMX and the AUC are very basic. And the next question was, you say that these PK parameters are pretty much known in the art. Absolutely, yes. That's in Appendix 3841. That is correct, Your Honor. And the point I would like to make on that is that Dr. Gustafson is a pharmacokineticist. This is her specialty. What she, and I think a reasonable reading of what her testimony says there, is that these metrics, are known in the art. CMAX, it is the case in clinical studies, you typically would measure things such as CMAX, AUC, sometimes CMIN, sometimes DFL. They're basic in the sense that these are tools that pharmacokineticists use. These are metrics that people are aware of. She did not testify that developing a formulation that has the specific profile that is described in 718 is something that is obvious. She did not testify that it's something that you would do for most compositions. And our expert, and this is at A10240 of the appendix, testified that it is not standard for standard release compositions to have this type of profile. That in fact the ideal composition is usually one in which, because you don't want to wait for the person to start getting the medicine, that you actually have a spurt. You have medicine that's given right away. So you won't necessarily have a maximum level that's lower than that of the immediate release composition. And just in terms of the testimony of Dr. Gustafson, later she was asked, in the same, it's on page 83842, she was asked, you just described these pharmacokinetic parameters basically for an extended release formulation. The C-max would have to be a little lower, your AEC would have to be substantially the same, and your Siemens about the same, and you said maybe it doesn't matter that much. So the question was assuming, I think your initial reading, Judge Garrison, the answer was it depends on the drug. And then the question again was, and the question is, isn't that the definition of extended release formulation? And her answer was, it depends on what definition of extended release you're using. If your definition of extended release is the ability to move from twice a day to once a day dosing, you don't need necessarily to have the lower C-max as shown by the alginate-based formulation, the MR, and the Siemens is not, in my opinion, substantially equal for the MR. So I guess the answer to the question is no. The answer to the question is no. Now, this is what is critical here in terms of the obvious analysis, because their position is that based on the 4-2-2 composition, which gives no in vivo data whatsoever, based on the 5-7-1 composition, which gives no in vivo data whatsoever, based on the 1-90 composition, which does provide in vivo data, and for the reasons we explained, it's not in vivo data that one can actually use to determine clarithromycin concentrations, because it combines clarithromycin with its active metabolite, and it doesn't give you enough information to figure out fluctuation. Based on those, someone would have been able to figure out how to develop an extended release composition that has the very precise profile that is required by the 7-18 claims, and there's just no support for that, because all that these other compositions are talking about, when they use the term sustained release, their argument is based on the notion that this brings into it all of these requirements of substantially equivalent semen, AUC, lower C-max, lower DFL. They don't. They don't. We know that from the 1-90, because the 1-90 is a controlled release patent, and the reason that Abbott developed the 7-18 patent is because the 1-90 didn't have this ideal profile. They wanted to improve upon it, and that is the motivation that's described in the background of the intervention, to do something that doesn't have the drawbacks of the 1-90, which is not able to achieve these PK characteristics, and instead come up with one that has the ideal profile. And it makes sense, and we have evidence in the record from our expert as to why it would not be automatic that you could develop this type of composition. You are playing with different dynamic factors. What you're trying to do is you're trying to make sure that you're not losing the bioavailability. So even though you're slowing down the release, you want to make sure that over a particular period, 24 hours, the person is going to have the same amount of drug in their system. You're doing this by giving them a tablet, which has twice as much drug in it as what they would take if they took it twice a day. We know that clarithromycin has nonlinear pharmacokinetics, meaning that normally when you give twice as much of it, you don't have just twice as much absorption, you have more than twice as much absorption, which complicates things. You're saying that this takes a long time to develop? Excuse me? Does this take a long time to develop as far as selecting the PK parameters are concerned? It takes a year, two years, to come up with the right combination? To determine the parameters that one would have as a goal or to achieve them. To achieve them. I think it all depends in different circumstances, Judge Garrison, and in different drugs it takes different amounts of time. In this case, there is evidence in the record that it took about one month to develop. Forty days. It took that long. I see that my time is running out. I'd just say on that point, what is also important here to understand that we think this is a hindsight analysis to say that this would have been obvious to do, is if one looks at the 422 composition and then the later 650 patent, we're not arguing the 650 patent is prior art. We're not arguing that someone of skill in the art would have known about it at the time, because obviously they wouldn't have. It was ten years later. But it's a reality check. Would someone look at a publication which discloses no PK information, which only discloses in vitro information, which doesn't tell you what the PK characteristics of the product may be, would they look at that, look at the dozens, scores of formulations that it describes, would they pluck out of it one and say, for a different drug, and we have evidence in the record that these are very different drugs, that you could take it for a different drug, plug in HPMC to clarithromycin, and you would expect that you're going to get this very specific profile. And as real-world evidence, that that would be unlikely that anybody would have that conclusion, consistent with what our expert testified to, Pfizer itself, ten years later, filed a new patent application, and they said, you know what, that previous application that we told you was going to be so good for sustained release, it flunked the most basic element of sustained release, which is it didn't provide that equivalent bioavailability, because the problem is, when you put the brakes on the release of the drug, there's a chance that it's going to be released so far down the GI tract that it's going to be excreted, before you can get as much of it into the body as you would with an immediate release. So it's not a simple thing to develop these compositions, and we submit that there has not been a showing of obviousness in this case, and there's certainly not any error that is reversible by the district court, and the injunction should be affirmed. Any more questions? Thank you, Mr. Dane. Mr. Fader. Your Honors, let me address the claim construction issue a little bit more. I heard something about very precise in vivo profile. There is no precise in vivo profile here. C-max is lower than the C-max of the INR. That's something that a person of ordinary skill in the art would expect. Going from twice a day dosing to once a day dosing, that's something that the person of ordinary skill in the art would reasonably expect. AUC for the ER equivalent to AUC for the IR, a no-brainer. The CFR tells you that's what you want to do, so you don't have to go through more human testing if not necessary. Now, the Andrix and Teva decisions are in conflict in one area. In Teva, the court took a look at the claim construction from Judge Kaur and gave effect to the language selected from the group consisting of. In Andrix, the Andrix court said, MarCouche language has no meaning whatsoever in the specification. And we submit, Your Honor, that that is a conflict between the panels on a purely legal issue. And for that reason, Teva controls. That was our point. Not that the claim construction, but that MarCouche determination controls. Pharmaceutically acceptable polymer, in my opening argument, I made it clear why that's not exemplary language. I have all the intrinsic evidence that I listed, and I won't go back and list it again. The fantasy sports case, that had three claims just like we have three claims. It had an independent claim and two independent claims. Abbott challenged us and said, Oh, we couldn't find a case where there was claim differentiation doctrine was rebutted when you had a dependent and an independent claim. The fantasy sports case doesn't. Does it do it using some statements in the prosecution history? Yes. But is the specification intrinsic evidence? You bet. And it's very, very potent intrinsic evidence here. The pharmaceutically acceptable polymers that are defined and the pharmaceutically acceptable excipients that are exemplified both contain polymers. The excipients contain polyethylene glycol and starches, both of which are polymers. So there's a reason to give meaning to the pharmaceutically acceptable polymer is language because it distinguishes out of the universe of all known polymers, those that are the release-extended polymers, contemplated by the inventors for this invention. 30 days to develop this formulation. I think that speaks volumes, and I won't have to say anything more. I understand that there's been no argument about Section 103 on the issue of how you arrive at a patent. This is not a flash of genius. This was just applying what those of ordinary skill in the art would know, just like KSR said. In KSR, they say if a technique has been used to improve one device, a person of ordinary skill in the art recognizing that it could improve similar devices in the same way using the technique is obvious unless its actual application is beyond his or her skill. And finally, Your Honor, granting patent protection that would occur in the ordinary course without real innovation retards progress. And, Your Honors, I would suggest that filing extended release formulation patents just at the time that the immediate release composition is about to go off patent and not any earlier, not years earlier, shows that this is an invention that could have been, that this is a formulation that could have been made before and simply wasn't for commercial reasons, for commercial and patent evergreen reasons. With respect to the fluctuation index, I think I pointed out where actually there is a suggestion in the 422 patent itself. I would also submit that the EP571 patent, which shows the use of extended release technology using HPMC and a second polymer, also would render obvious claims 1 and 4 based on the teaching in the 571 patent and the fact that the person who organized the art is imbued with knowledge of extended release formulations and would be motivated to have these PK parameters. With respect to the fact that while the 571 patent requires a second polymer, an acrylic polymer, while the acrylic polymer is one of the pharmaceutically acceptable polymers expressly defined in the 718 patent, and in that definition, after they enumerate the marcoosh group... Mr. Feder, we need to wrap it up. I think you're going beyond rebuttal. Okay, I'm sorry, Your Honor. I'll just wrap it up with this one. The marcoosh group ends with the words and mixtures and derivatives thereof. So it's clear that it's one or more of those pharmaceutically acceptable polymers. Your Honors, thank you very much. Thank you, Mr. Feder and Mr. James. The case is taken under submission. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A3399. Public record. Not legal advice.