Oral argument — Purdue Pharma L.P. v. Amneal Pharmaceuticals, LLC

Our second case this morning is No. 15-1654, Purdue Pharma v. Amneal Pharmaceuticals. Mr. Castanhas. Yes, good morning, Your Honors, and may it please the Court. The newspapers every morning seem to have reports of opioid and opiate abuse. This has been characterized as a scourge. It's been characterized as an epidemic. Now, extended-release OxyContin, which is the product at issue here. Extended-release OxyContin contains a large amount of Purdue's powerful pain reliever, which is enough for 12 or more hours of continuous pain relief. When the extended-relief version was put on the market for the first time, abusers quickly realized that they could crush or dissolve this extended-release formulation and get an immediate and powerful high. The result of this were addictions, overdoses, and deaths. It's not insignificant, is it, that that was on the market for a fair bit of time before—I'm not quite sure what the right language is—before the problem of that abuse became very, very well known. I'm remembering, I guess, and correct me if I'm wrong, because it's not an insignificant consideration in your so-called objective indicia feature, that not really before 2000 was there significant publicity about this. That was when, according to the record here, the extent of the OxyContin abuse had peaked. As even my friends in their brief on the other side say, though, opioid and opiate abuse has been a problem for centuries. But it was in response to that peak that this invention came about. Right, but your long-felt-need argument does depend on the word long. And my recollection of the record is that it really wasn't until 2000 that there began to be significant recognition, and then you got into the act in 2001, and poof, you had a solution. But the need became most urgent in 2001. But when we get to secondary considerations, which I hope to, by starting with obviousness and then concluding with indefiniteness, I hope that you'll see that there are other more powerful secondary considerations in this case. But on the prima facie case issue, it was well known to add gels to compounds to prevent abuse, right? And what you're suggesting is, well, that was for immediate release tablets rather than for controlled release, no? Well, so Judge Dyke, I think this takes me right to where I want to go with regard to the prima facie case, and that's the Baston reference. As you point out quite correctly, gelling to prevent abuse was known at least as far back as the 1975 Sean Hoffmeister references. The use of PEO to control release was known as well. The case that was made by the trial court and by our friends on the other side of the trial court was that, well, it was just an obvious invention waiting to happen. But if this was so well known for so long, other than the peaking of the market being the only reasonable explanation for it, why was it that people didn't come with it, come up with it? And the answer, I think, can be found in Baston. You're saying it's a teaching away in Baston, but that was a different medication in a different additive, right? It was a different medication in a different additive. But let's take a look at what the district court found in Baston was that it actually taught the combination. It actually implies was the language that Judge Stein used, and I think that's at page A77. And what he relied upon, and this is in the third volume of the appendix, the small one at page A79490. This is at page five of the Baston reference at the bottom, and it's in a paragraph that begins in contrast. He relies, as my friends on the other side do, on lines 29 through 35 of that reference. And I'll read that to you so that we get the context. Which page are you on? This is page five of the reference. It's A79490. And it is in the bottom paragraph. And this is the language relied upon by the trial court. In contrast, a combination of the active drug substance and gelling agent in the same layer has the disadvantage. First word, disadvantage, important to note. That the gelling action is likely to retard the release of the drug. And then here's the language that the district court found compelling. In a manner similar to some known sustained release products, which include water-swellable, high molecular weight polymers to retard drug release. So that's the end of the quote that's relied upon, and that's what the district court thought suggested that for an extended release this would be perfectly good. But he didn't look at the next sentence. Reduction of the gelling agent concentration to a level which would not inhibit release of the drug substance severely limits the abuse resistance potential of the table. Now what does that mean, severely limit? Now let's turn to another part of Bastin which we argued but the district court did not read. And this is important because this court's rules say that you look to the reference in its entirety. Look at Table 4, which is also in our reply brief at page 10. Table 4 tells us about that single-layered tablet and the sort of severely limited release. With the gelling agent in the tablet, 50% of the drug came out after two hours, and no further release up to 12 hours. No further release. That's not controlled release. That's no release. But he had in front of him, did he not, testimony that someone skilled in the art would have easily figured out how to accommodate the use of the gel together with the controlled release feature. And he made a finding that there was not a discouragement to doing that. And we don't run away with that. We have to review those findings based on a clearly erroneous standard, and so why isn't there support for his view that the Bastin reference and knowledge of someone skilled in the art could accommodate the use of the gel in a controlled release formulation? Well, first of all, the answer is that we're not shrinking away from the clear error standard here. I think the court should be left with a definite and firm conviction that he made a mistake by not reading the entirety of Bastin. And with regard to what next— We don't know. He didn't read. He just didn't mention. Well, he didn't, but I think this would have been very important language, especially in light of the fact that we pointed it out to him and in light of the fact that— Well, you did point it out to him. And we did, absolutely. So he was aware of it. He was. So the fact that he didn't conclude that it was as important as you say it is was a matter of judgment call on his part, which factors into the question of whether there was clear error. Well, I think, Judge Plager, with respect, I think that the failure to consider all of the aspects of the reference that we mentioned and instead to only consider the parts of them. Well, my point simply is you did call all of that to his attention. We did, absolutely. So he had to have considered it. At least he heard it. He heard it, but he did not read the reference for everything that it discloses. He didn't come out your way. And I think that particularly with regard to the fact that if you look at page A6199 of the record, you'll see that Judge Stein at trial was particularly concerned about Bastin. He says this. And our friends at Amnial on the other side point out when he says that he's concerned that they had not even addressed it in their testimony. The only testimony about the content of Bastin and whether or not it teaches away comes from our expert, Dr. Davies. The implication or the inference that Judge Stein drew, and that's his language, it says that it implies that, is strictly him giving it his own reading. Suppose we are not taken with the argument that because he didn't mention it he didn't consider it. Let's assume we conclude that he had it in front of him, he must have considered it. What is it that prevents him from reaching the conclusions that he did about the prima facie case of obviousness and the ability of someone skilled in the art to know how to combine the inhibiting gel together with a controllable release formulation? Judge Dyke, I would say look at the judge's opinion and look where he finds his motivation to combine. He finds it in the prior art only from this implication in Bastin. He doesn't find it in any of the so-called gelling references, and he can't find it in any of the so-called PEO references. The only place he finds it is this implication, and we say that it's exactly the opposite. And he only got to that implication by considering part of the reference. In fact, not just part, but only one sentence when there is data in here. Well, maybe I'm mistaken. I don't recall his saying that he wasn't relying on this testimony that was in front of him about what someone skilled in the art would know how to do. In fact, I thought he specifically mentioned that in his findings. Am I mistaken? No, no, no. He absolutely did consider that. But when you weigh the legal conclusion of obviousness against this erroneous finding and the Bastin reference considered as a whole, what you see is you see gelling for controlled release and gelling for abuse deterrence. And he says, Bastin says, they live together. We say, Bastin says, stop. Don't do that. Go a different direction. That is exactly what the clear error standard, whether considered or not judged plagiarism, whether considered or not by the judge. The simple fact is that this reference cannot be read reasonably as You mentioned a few minutes ago that the district court in its analysis, did not look back to the allegedly anticipatory references. Is that right? I don't think that I said that. Oh. What about the set of, it seemed to go by many different names, the 963 patent, the 591. The 963, 591. I mean, neither of those had the gelling to deter abuse. Neither of them had the viscosity limitations. The Joshi reference had no control. I don't remember, but do they talk about gelling and controlled release and whether those are like oil and water and they don't mix or maybe they might mix. Well, they talk about using the PEO for controlled release. But they don't talk anything about using it to deter abuse, to keep it from being sucked up into the hypodermic needle or to be snorted. So there is art on both sides of the equation here, on the abuse deterrent side on the one hand and on the controlled release side. Bastin says don't mix them together. They don't work together. They'll trap it. It'll keep the drug from ever getting released. That's the main point on our prima facie case. But the gel was used to control release, right? Absolutely. Well, so why would it be inconsistent to use gel in a controlled release formula? Because what Bastin tells you is that the amount of gelling that it takes to control release is not compatible with the amount of gel that it takes to deter abuse. That's exactly what that sentence at the end of the portion that was not quoted by the district court says. Now, I'm already into my rebuttal time. I haven't even mentioned secondary considerations other than my colloquy with Judge Taranto or the indefiniteness issue, which I'm happy to leave to the briefs. But on secondary considerations, perhaps I should say the following. Two points. First of all, the epidemiological studies that Purdue submitted to the FDA showed that this was a game changer with regard to reducing abuse. It actually diverted abusers away from using OxyContin, and unfortunately to other drugs sometimes. But it was a huge success in that regard. These are the abusers who now cannot get this up their nose or into their arm. The FDA, I think, is even more important here because the FDA's recognition of this invention as a game changer, as one that changed the risk-benefit calculus, and they recognized that the old version of OxyContin was removed from the market for safety considerations. They would not approve generics of the old OxyContin to compete. So how do you fit that into the combined question? What secondary consideration, and sometimes that can be a list seemingly disconnected from the underlying inquiry, but more importantly, how do you connect that as a logical evidentiary matter to the question whether this would have been obvious or not? Absolutely. So first of all, KSR says any secondary consideration. This court has traditionally looked to commercial success in an ordinary competitive free market as the sort of thing. Well, and I assume that's because if there's great commercial success, it tells you something about the incentive that relevant skilled artisans would have had to discover it, and the fact that they didn't tells you maybe it wasn't all that obvious. Connect the FDA decision to that. Sure. So first of all, with regard to the invention here, the invention here is not a personal good. It's not something that either a legitimate user or an abuser wants necessarily. This is a public good. So it's the FDA that's looking out for the public good here. They're making this determination, and they're making a determination that now takes old OxyContin and its potential generics out of the picture. It's no longer on the market. It recognizes that old OxyContin was removed from the market because of these safety considerations, and it becomes such an invention that the FDA gives for the first time in its history labeling that recognizes the abuse deterrent effects of this drug. The purpose of the secondary consideration analysis is, as the Supreme Court in Graham and so many other cases from this court have made clear, it is to check judges, and I would dare say lawyers also, who may not be the experts in the area. But this is the way that objective people... Let me put my question this way. Assume for purposes of this question that what the FDA did here is confirmation that the reformulated OxyContin is so much better on the safety efficacy profile than original OxyContin that it took original OxyContin off the market. It wouldn't let generics in. You dropped your thing. So in terms of objective improvement to what was available in the market, this was a giant thing. Connect that to non-obviousness. Connect that to non-obviousness? Yeah. We built a better mousetrap. The FDA recognized it. That is exactly what the non-obviousness... But if the recognition for the need for it didn't come up until the day before you figured it out, why is that, as an evidentiary matter, connected to non-obviousness? Well, you're coming fairly close to saying to me that long-felt need is a necessity for a claim of non-obviousness. Whether this was a long-felt need or it was a short-felt need, it was a desperate need, and it was filled, and it was filled against the advice of the art, which was what Bastin told, and ultimately not just the approval and the withdrawal but the abuse deterrent labeling. This is something that other drug manufacturers tried to get and didn't get. We got it for the first time ever. That was a recognition that this truly was a game changer. It is objective evidence from disinterested parties that says this is... I'm sorry, I'm not getting the reference to other drug companies trying to get and didn't get. I'm sorry, I didn't hear the last part. You said something about other drug companies tried to get and didn't get. What are you referring to? Other drug companies tried to get, at least after. I know at least of, I believe it was Endo, with regard to its drug Opana, tried to get abuse deterrent labeling and did not achieve that. This was the first time the FDA ever did that. That is a remarkable achievement. I believe also the amicus brief on this. It's not the first time that people had used gels to control abuse. No, but it is the first time that people used gels to control abuse in a controlled release drug, and that was so important here because the controlled release was not just one hour of pain relief but 12 continuous hours of pain relief. That's the invention here. It's the combination. Controlled release, abuse deterrents. Again, I go back and I say Bastin said stop. No matter how clever it was, it's still a question as to exactly how that relates to the question of non-obviousness under the statute, and I'm having somewhat the same problem. It is a secondary consideration, is it not? We call them secondary for a reason. They are, except that you also have case law. Transocean is one case that comes immediately to mind that says they always have to be considered. You have a number of opinions pointing out that sometimes, and I would say here, it is at least as powerful evidence of innovation. Remember that obviousness is ultimately you're just looking for the presence of something we know it's not because it overcame 102. So is this under 103 something that should be granted? Is this the sort of thing that we want to innovate, that we want to encourage, that we want to reward with the grant of a limited exclusionary right? I would put forth that given all of the value of this invention, as recognized in the marketplace by the regulator, the answer to that question clearly has to be yes. Unless the court has further questions. Don't fail to read this morning's Washington Post as a front page article on the newest abusive drug that has nothing to do with your product. Your product may no longer be relevant. I'll get back to newspapers after this argument. Thank you. All right, we'll give you two minutes for remodeling. Thank you, Your Honor. Mr. Maloro. Thank you, Your Honor. May it please the court, the judgment of the district court should be affirmed. This is a judgment which was entered after a five-day trial, nine witnesses testified, and the court issued something like 63 pages of findings of facts. Can I ask you about Bastin? Did Bastin in describing its use of, is it the same gel as claimed or some gel? Bastin has a different gel. Did he structure the use of the gel in the pill in a particular way that wouldn't or would generalize to the problem of stopping release after two hours? There's no specific structuring in Bastin. There was a table, which counsel referred to in his argument, in which the gel was part of an immediate release tablet, and in that case the drug stopped going out. To the extent that you can explain this to me, and there's both a you and a me in that, what is it about the problem that Table 4 identified in Bastin that was overcome here? Is gel used in a different way, not layers with a kernel or something, or what? So the problem, so to speak, with respect to the table, was that you had an immediate release tablet. It wasn't a controlled release tablet. And so the object was to have immediate release of the drug. What Judge Stein found is that if you wanted to have a sustained release version of the drug, you would have the gel to modulate the properties, the release of the drug, and the prior art exactly taught that. There were other references, the Royce reference, for example, which Judge Stein discussed, specifically taught using PEO, the exact gelling agent here, as something that could modulate the release of the drug. I'm trying to get one layer down to some level of technical detail that I can imagine. What did your experts say about why Bastin didn't teach a way from trying to get controlled release using this gel? Essentially, it was the paragraph that counsel referred to at page 79490, and it was specifically the effects of the gel that they could retard the release of the drug, which would happen in a sustained release agent. But I assume at trial they saw that. They looked at Table 4. It retarded the release so much that after two hours there was no more release. They're thinking, that's a bad thing. For an immediate, sorry. It's fine for immediate release. It's a bad thing if you want controlled release. Did your experts say, oh, no, no, no, that's not the only way you can use gel. People would not have been deterred from trying to create a 12-hour OxyContin drug by reading Bastin. And if so, what did your experts say about why they wouldn't be deterred? Because of the teachings of the prior art in Royce and Hoffmeister and Shaw that specifically taught, and there are curves where you can look at the release of drug with PEO modulating it. So we have, for example, the Zhang dissertation that the court referenced in its opinion and the Zhang article which teach that PEO was a widely used agent for controlled release. And you can actually look at the curves that one skilled in the art can adjust the amount of PEO or the type of PEO to get the release of drug that you want. So it's quite apart from other substances which might control the release. The prior art taught that the use of gel alone could be used to control the release. That's exactly right, Your Honor. Royce is a good example of that. The Zhang article is a good example of that. Judge Stein carefully analyzed all those prior art references and found that one skilled in the art would have known, would have been motivated to use PEO, would have been motivated to use it in a sustained release dosage form, and would have been motivated to use it to solve the abuse problem. The Joshi reference, for example, which Judge Stein referred to in his opinion, that Joshi reference taught specifically the use of PEO as a gel to prevent abuse of pharmaceutical formulations. So here what we had was a known problem, motivation to solve that problem, known predictable techniques for solving the problem, and no unexpected results, not even really an allegation of unexpected results here, and secondary considerations that the court found didn't support a finding of non-obviousness here. Just to take one example, you referred to Royce several times. Can you point to me where in Royce something says if the end product you're trying to get is one that affirmatively provides controlled release, and not just control in the sense of stopping it after a short period, but having release of the drug for an extended period? Think about using a gel. Yes, Your Honor. Royce begins at A83985. Yes. And if you look to column three, around line 39, Yeah. you see that sustained release dosage forms may be prepared from the higher molecular weight polymers, i.e. molecular weight greater than about 300,000, and there's reference to polyox in the parenthetical there. And if you continue down, it goes on to talk about contemplating mixtures of the polymers, and the mixtures may comprise respective amounts of various polymers, as shall be within the skill of the worker. Thank you. The Zhang article at A45726. Yes. If we go to A45731, we can see there are two figures there. Figure four. 45731. 45731. Thank you. And we can see the influence in figure four, the legend, is the influence of molecular weight of PEO on the release of CPM from tablets. So your actual experimentation where one skilled in the art can modify the PEO to get the release they want. This is well within the level of skill in the art. This is exactly the evidence that Judge Stein considered, carefully commented on. This was a ruling, a judgment that the rulings went both ways in the case. There were some defenses that were rejected by the court, other defenses that the court credited. There was some evidence from defense experts that was not credited by the court. There was other evidence that was. There was obviously evidence on both sides of many of these contentious issues, but there is no question that their product attained remarkable commercial success. In fact, their former product went off the market altogether, and they were able to market this new formulation widely. And those of us who have broken a leg in the last year or two may have had the benefit of it. That's pretty impressive evidence that they came up with something that nobody else had yet, perhaps not thought of, but perhaps been able to figure out how to do. Well, with respect to the sales of the product, there were high sales. Judge Stein recognized that. On the issue of commercial success, there was no connection. The evidence didn't support a connection between the patented features and the sales. And, in fact— I'm kind of baffled by this, and I kind of think I'm missing something. The reformulated OxyContin has so much better a safety efficacy profile that the FDA basically said, You can only do this. You can't even do the old OxyContin. How could the commercial success, even assuming we're going to isolate it from the actual 103 issue that it's about, not be attributable to that? It's so much better, and this is what's different about it. Because the sales of the product were the same with and without the safety feature. So if we're talking commercial success, there was no demand for those patented features. As a matter of fact, Purdue recognizes that had the generics been allowed to come on the market with the original OxyContin, they would have taken tremendous sales. I guess I keep thinking that we're getting caught up in thinking about secondary considerations as if it's a set of individual silos unrelated to the evidentiary inquiry. The fact is that this product had so much better a safety efficacy profile from what came before that the government regulators said, You can't even sell the old stuff anymore, and nobody else can either. It's that much better. Major contribution to the public good. What do we do with that fact in the obviousness analysis? Judge Stein recognized and credited that fact, and he put it into the, using the lists that the cases identify, he put it into the acclaimed category. He recognized that there was some acclaim associated with the invention that flows from the FDA's actions. This was not something where he couldn't fit it into a bucket and he ignored it. He recognized the evidence, but the evidence was counterweighed by the strong technical evidence of obviousness here, the predictable results that were achieved with the known techniques, and the fact that the other secondary considerations, when you looked at the commercial aspect, the sales aspect here, there was no nexus to the claimed invention. There was no skepticism. The inventor testified he couldn't identify any skepticism here. The FDA's actions were not deemed to be evidence of skepticism. It was just the FDA doing its job, waiting for data, and then approving the labeling for the abuse avoidance and taking the old product off the market. Is the weighing of the prima facie case and the secondary considerations, once they've been identified, a legal issue or a factual issue? I think there's a component of both there. The strength of the relative evidence is going to be highly factually dependent, and the findings of fact on the strength of the technical case, the findings of fact on the lack of secondary considerations as a whole, that's all highly fact-bound. At the end of the day, there clearly is a legal question on obviousness once you have all your facts pulled together. So if you say that even though there are some secondary considerations cutting against obviousness that's outweighed by a strong prima facie case, that's a legal determination? No, I believe that that's more factual than legal. Because the particular considerations, I don't think this is sort of a checkbox exercise where you determine is there a prima facie case, is there a secondary consideration or not, and then once you get done, you just weigh in a legal fashion the checkboxes. This is one where the evidence supporting all of those aspects of the obviousness analysis is highly factual. And so the ultimate legal conclusion, even with respect to the balancing, has to take into account the fact findings and the determinations that are made as to not just the existence of that evidence but the relative strength of it. Okay, anything further? Any further questions? Okay, if there's nothing further, I'll pass the vote. Thank you, Mr. Malone. Mr. Costanze, you have two minutes. Thank you, Your Honor. I think it's possible that either I misspoke or miscommunicated in my opening argument because one of my colleagues pointed out to me that Bastin does concern analgesics. I think the question to me might have been whether Bastin concerned other drugs. That's clear at A79484 in the abstract, and of course it's responding to Hoffmeister, which is one of those 1975 references, which is directed to analgesics. Justice Rothfeld? Yes. If this gel is used to control the release, why would it be startling that one would use the gel in a controlled release formulation, which is what you're suggesting? That's not what's startling. What's startling is the ability to use that gel in order to not only control the release but also to deter abuse. That goes back to our argument about Bastin. I'm sorry. I'm a little confused. Because on the one hand, it was used to control the release. On the other hand, it was used to deter abuse. So why is combining the two of them inconsistent? For exactly the reason that we pointed out with regard to Bastin. Bastin suggested, and again, I'll bring you back to the particular page, but it's that sentence that comes at the end of what was relied upon by the district court, that reduction of the gelling agent to a concentration to a level which would not inhibit release, in other words, which would avoid that holding it up for the rest of the 12-hour period that Judge Taranto was talking to my friend about, severely limits the abuse resistance potential of the tablet. That's part of the halt. Don't combine these two things that's in this reference. The data that's on page 28 of the reference, it's A79513, is another example of where the prior art says halt. When my friend talks about Royce and Zhang, all he's telling you is that there are prior art references that say, yes, you can use PEO to control release. What it doesn't say is that you can use PEO to control release and at the same time also get the remarkable, game-changing abuse deterrent. If I remember right, correct me if I'm wrong, but the basic logic of Judge Stein's analysis of this was a skilled artisan would have recognized that the gel has these two possible features, two kind of documented features. It controls release and it deters abuse. Then the second piece of the analysis was, and that would have given skilled artisans a reason to see if they can get both to work at the same time, and that was just a matter of, I'm going to call it routine experimentation, to get to the right balance. If you read Bastin as simply suggesting that that's what you do, then that's how he was able to do it because he thought Bastin was a teaching in favor of that holding. We say, and I think the document actually bears this out, Bastin says, no, it won't work, because if you reduce it to a level that will allow the drug to release, it won't have the abuse deterrent effects. That's what that sentence says. That's what the data says. If I remember right, Judge Stein did not use the language of obvious to try, right? He didn't use it, but it certainly seems like he was applying it. If I'm going to use that label, the first step seems to me that he said there was enough in the prior art, including Bastin, that made it obvious to try, to see if you could get both of these benefits at the same time, and then the rest was really routine experimentation. Why is there not enough evidence to support that? Because Bastin says that it's not going to work. But you had other testimony from skilled artisans that it would be routine to achieve that proper balance. Not about Bastin. Not about Bastin at all. Those skilled artisans did not consider Bastin. And again, A6199, and I'll quote Mr. Werner, Amnial's counsel, I was going to say we never talked about Bastin on direct for what it's worth. Judge Taranto, you asked my friend about what his expert said about Bastin. The answer is they didn't say anything about Bastin. The last two points I have very quickly are that the CPDD paper, which was written two years after the invention here, still didn't come up with this. It's not a very technical paper, right? It's not a very technical paper, but it is written by very expert people, including people from... My recollection is that it was more sociological than it was about technical choices one might make to solve the problem. It was about documenting the problem. Whether or not that's a fair conception of the CPDD paper, it also did have two specific suggestions for deterring abuse. One of which was the addition of antagonists, that the liver would filter out if taken orally, but that when crushed would counteract the opioid. And the other, of course, was controlled release itself, in and of itself, as an abuse deterrent. Okay, I think we're out of time. Okay, thank you very much for your time. We appreciate it. Thank you, both counsel, and the case is submitted.