# Oral argument — In Re Columbia Laboratories, Inc > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A10484 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** December 11, 2014 ## Text Next case, In Re Columbia Labs, Inc. Securities Litigation. Good morning, Your Honor. It's Jeremy Lieberman on behalf of Appellants. I'd like to request four minutes for rebuttal. Your Honor, there's been an attempt both by defendants and characterization at the district court level that this case is about whether or not Columbian and Watson engaged in a process where they proceeded with an NDA that was doomed from the outset and they failed to disclose to investors that that NDA was doomed. That's how the claim has been portrayed. That's not a claim, Your Honors. The claim is that an NDA was brought to the FDA, brought by both Columbia and Watson, containing numerous flaws in the study and in the trial which were not disclosed by two investors. And as a result of the risk of those flaws, investors did not properly have an ability to weigh the value of Columbia securities. As you framed the argument and you framed it there, how does that description of your case show the answer? The description is that they have knowledge of the flaws in the study, yet they're not disclosing it to investors. Let's take first one in the statement, Your Honors. There was a requirement by the FDA for a single-study NDA, and it's critical here, that we have a single-study NDA which has much higher requirements. In a single-study NDA, there's very little room for any imperfections in the study as well as any flaws in the data. Do they really have to disclose each and every flaw, imperfection in the study? Is there a recognized duty that they have to disclose that? When there are statements made regarding the results of the study, and if there are flaws that impact into how those results were calculated, certainly, yes, we would say so, Your Honors. Were any of the statements that were made inaccurate? Well, Your Honor, we'll take one statement. There was a requirement by the FDA that a single study, if it's going to be successful to the FDA, that it be robust. Now, that's a requirement. Okay, so right there is a recognition that you could have a single study that would be sufficient if it was robust. Oh, certainly. If it's robust, there's no doubt. Okay, so the question in that particular single study was whether or not the findings were robust enough to show what we were trying to prove. That's correct, Your Honors. And the issue is that here, the FDA provided Columbia with a definition of robust. In 2004, at the study 300, they said this study needs to be robust on a single study MDA. Is that, in terms of your allegations, is that the sole basis for your allegation that they had knowledge of a requirement for the .01 P value, for what robust is defined to be, for this study? Not the study four years earlier, but for this study? The allegation is that they were given a definition of what robust meant. There's no inference that the definition of robust, which is a requirement by the FDA, would somehow change over the course of three years. There's nothing the FDA regulations indicate as such. There's simply nothing in the regulations that would show that somehow the definition of robust changed. Where in your allegations or the documents that are incorporated or referenced and incorporated, is there anything about an actual requirement? You use the language required, mandated a number of times. Where is that requirement that you allege they had knowledge of? The requirement that the results be robust? Is that the question? No, the requirement that it have a particular P value. No, no, no. The requirement is that the results be robust. The FDA provided this definition of robust to Columbia. Who knows what robust means? The FDA says, I'll tell you what it means. It means a P value of .01. Now the question is, was there any basis to believe that somehow that definition changed between study 300 and 302? They indicated that as to the earlier study. They indicated what the definition was. And the question is, is there any reason to believe that that definition of robust would somehow change over the course of two studies? There's simply no basis to believe so. And at least it provides a risk that that definition would remain the same. I'm trying to get to the false statement. Where was the false statement? When the company says that we have robust results. The study achieved robust results. They were given a definition as to what robust meant. And yet, they're not telling investors that these results actually did not meet the FDA's representation. There was a discussion between defendant Condella and investors at an analyst call in May of 2011 where analysts asked, is this going to meet the, do you have a lower P value because this is a single-study MDA? In 2007, the FDA informed Columbia that a new MDA would require at least one additional adequate and well-controlled study. They didn't go out and tell them again, look, we told you in 2004 you need a robust statistically significant reduction. They tried something different in 2007. No, I disagree with that, Your Honor. Well, they told them what they told them in a different way. Well, they told them what they told them. And they told them the same thing in a different way. What they told them then was that they said actually we would want to power this study to a .05 statistically significant requirement. The FDA said you do that at your own risk. And the question is, is that? They didn't say there you needed .01. They said you do so at your own risk. The question is, is that a risk disclosed to investors? Because investors had no idea at the time that the FDA had prior told them that the definition of robust means, this definition means one thing, .01. Well, the question goes back to my initial question. I mean, this is an interesting debate, you know, as to the FDA rules, how you proceed with an MDA. But the question goes back to CNTR. And CNTR means something under the Private Securities Litigation Reform Act. Sure. CNTR means that there's a knowledge by the company that it's giving material, that their material risks are not being disclosed. That's how we define CNTR. Okay. Your Honor, and here we have material risks. But the FDA is going to use the same exact definition for robust that applied to Study 300. It's a matter of, there's a requirement of robust. How does Columbia or Watson know that? They know the risks, certainly. They were told. The FDA guidelines list out five criteria. You know, having statistically very persuasive findings is just one of them. And even for the single trial study, the example that's given in the FDA guidelines has a .05 as part of their example for what's statistically significant and could still be considered effective. Your Honor, but in this situation, they're told it's required to be robust, and they're told it's a .01. So the question is, what basis would Columbia have to believe that somehow the definition would have changed? My information is the same as Judge Passer's. The significantly statistical number changed from .01 to .05 by 2005. How did it change? That's what the FDA said. To have an acceptable study, it has to reach a p-value of .05. We're no longer talking about the Study 300. We're talking about the Study 302. We don't have the FDA. We're not aware of what the FDA said. You need only .05. Your Honor, I'm not aware of that. I'd like to see that. We would like to see that. Sorry, it isn't the point that nothing precise about a p-value was communicated. He's trying to allege that there is a strong inference of knowledge of material risks that weren't disclosed, but where is there any evidence or allegation that you've made that that information was communicated, that supposed requirement was ever communicated to Columbia as to Study 302? The allegation, Your Honor, is that the requirements by the FDA between the time that they met on Study 300 in 2004 and the time they conducted the study did not change. If there's a suggestion of a change, we would like to see it, Your Honor. And that because it did not change, there was a requirement of being robust for Study 300. That definition does not change. It's a definition of a standard. That definition does not change. Turning also, there was a statement, Your Honors, regarding statistical significance at the primary end. Say that again. There was a requirement of statistical significance, even at the .05 level, that the company says they achieved that number. They only achieved that number, Your Honors, by including data from two sites outside of America, where in Belarus and South Africa, that only handled 7% of the whole entire trial. There's perfect results on the… How does that make it a false statement? It makes it a false statement because it's not just a matter of being factually correct, Your Honors. It's a matter of giving information a full record to understand the basis for a statement. If you say we achieved statistical significance… Because they omitted the Belarus and the South Africa studies. That's correct, which were, especially with respect to a single-study NDA, those were contradictory data. You take out that data from South Africa and Belarus, you have no statistical significance, even at 95%. The situation is you have in Belarus and South Africa, you have the company achieves 50% on placebo. In Belarus, achieves pre-term birth zero when they take the progesterone. For South Africa, 36% on placebo get pre-term birth zero. Pre-term birth zero, again, perfect when you take progesterone. So you have a clear situation where there's very curious data here. And if you compare that data to the other data sets, it doesn't even match. In the United States, 16% that took the placebo, only 16% actually had a pre-term birth. It doesn't seem to make your statement inaccurate, though. It's inaccurate because they included the countries of Belarus and South Africa. They should have excluded them, and then they would have given you a different number. No, it's inaccurate because when you give data to the public and you tell them we met the primary endpoints, and there are serious flaws in that data, you have to disclose to investors the full risks of the investment and the full risks of the statements you're making. It's not just a matter of factually. It would be offensive after your honors. Wasn't it the FDA that they went back to them and said, look, you have an issue with your studies? The FDA in the warning letter said there's an issue with the U.S. results in June 28th. Right away, once the NDA was accepted, the FDA hit them with a letter right away saying, in the United States, you don't have statistical significance. The key primary driver of the study, which is to cover the U.S. market, never achieved statistical significance, only achieved a 2.5 percent result, which was that statistical significance was .65, clearly falling short of anything near statistical significance. So you have in a single study NDA a requirement that there be no contradictory data, yet there's contradictory data all over the place. It's not even close to statistical significance for the U.S. population. Time went by so fast. The light is on already. Okay, we'll get you back. Okay, thanks. Mr. Donovan. May it please the Court, Your Honor. John Donovan for Columbia Laboratories. Let me start with Mr. Lieberman started about the nature of the claim and whether it's about disclosure of something known and doomed to failure, as we said in our briefs below, and repeat it here, or something just a risk. I direct you to Paragraph 65 of the complaint, which says precisely that Columbia failed to meet a known requirement. Go on to Paragraph 66 and 67 and 68, where it said making the risk of approval not viable. That's precisely what the complaint says. Now, getting to your question, Judge Fischer, about CYENTR and where it is, there is nothing in the regs that requires a confidence level of .01. There is no requirement in the guidelines of .01. The guideline to the contrary is .05. There is nothing in the complaint that says that Columbia was told by the FDA or anyone else that with respect to this study a confidence level of .01 would be required. Why not tell the investors about that .01? Pardon me, Your Honor? Why not tell the investors about that .01 that the FDA said in 2004? Well, that's the deconstruct the plaintiff's argument there for a minute. What he's really saying is that in connection with a different study seven years ago involving a different population and a different indication, we told you about .01. Our guidelines require statistical significance at .05. We therefore are insisting, or Mr. Lieberman is insisting, that Columbia should have deduced that over seven years in a different population, in a different study, a .01 requirement would nevertheless be applied. But they did reiterate in 2007 if you want to go to .05, you're doing so in your own risk. Absolutely, Your Honor. Okay. Now, isn't that a known, specifically, isn't that a known risk that would have been important to the potential investors? Your Honor, absolutely, and it was disclosed. I mean, no number was disclosed because no number was required. The only thing that is required is .05, and in the context of a single study that there be some additional showing. It's precisely what Columbia told the world. They said all we can tell you about our results is that they're statistically significant. Mr. Lieberman just told you that Columbia said the results were robust. It never said that. There's no obligation to say that. You never said in 2004 with Study 300 we were told we had to have a P value of .01. We were told that that study failed. Because there was not a P value of .01. Precisely. You didn't disclose that. No, that was not disclosed. It was disclosed that the study failed and no NDA was ultimately submitted. So the FDA never got the chance to say .01 either applies or doesn't. I mean, wouldn't investors like to know you've been down this road before? You deviated a little bit. Again, Your Honor, different study. And what investors were told is we are required to meet statistical significance. That's what we meant. But the FDA could take a different point of view. The FDA could insist upon additional things. There is no assurance that this will be approved. And that's what investors were told every step of the way. What was the intent behind not disclosing the subgroups and how the subgroups affected the top-line results? Subgroups did not affect top-line results. This is a matter of statistics, Your Honor. Top-line results are the overall effect on the overall population. You do not disclose subgroups and do not disclose subgroups along the way, in part because of the risk of gamesmanship. But the subgroups had an effect on the NDA. Your Honor, we were content they did not. And, in fact, the statistical analysis plan that Columbia submitted and agreed to with the FDA required them to test for variation among subgroups to determine if there were spurious results that should be eliminated. And all of them fell within normal tolerances. The only thing that happened is the FDA, 11 months after the NDA was submitted, went outside of the SAP, the plan, looked at additional data, and it was spurious data. Those numbers change dramatically if you take the subgroups. Well, Your Honor, you could pick any two subgroups and say the same thing. If you took the two subgroups that Mr. Lieberman now focuses upon, which were 16 patients, and said those drove the results, you could take, you know, out another subgroup or the U.S. population. If you took the U.S. population out, it would have met 0.01. So the point is just that. You are not allowed to play games with stats after the facts to say that subgroup data should have been disclosed. What is the risk to investors? The risk to investors is whether they know that there is a opportunity, a risk, a possibility, that there will not be approval of the product. And that's exactly what investors knew all along. Now put it in the context, too, of knowledge. What Mr. Lieberman is saying, what the plaintiffs are saying here, is that knowledge of some data, some data point, some prior communication by the FDA, posed a risk of misleading investors because someone knew it. But what he claims is that they knew some fact. Not that it posed a risk of misleading investors. What the company always told investors was that our risk of approval of this product is at risk. The FDA might not approve. They can look at any fact. Our risk is what? That they might not approve. The disclosure in every 10-K, 10-Q, periodic disclosure, the disclosure of results and everything else, always included that caveat. What about the analyst's question to Candela? Again, precisely making the point that the company does not have the information about what the FDA will require. The response was that we achieved statistically significant. Correct. Statistically significant. It could have misled investors more to say more. What Mr. Candela said was we have met the requirement of statistical significance. We don't know more until we go talk to the FDA about what they will specifically require in this context with all the data analysis done. In fact, the company had achieved a requirement substantially better than 95% confidence. It was over 98% confidence. It was .02. But he did not want to mislead investors into believing that because they had beaten the .05, that inevitably the FDA would approve. That's the way it comes across. I think it comes across the exact opposite way, Your Honor. I'm not going to get into this because I'm not going to mislead you in the wrong direction by overstating the results. Doesn't it suggest at least that there's an awareness that asking the FDA the question could provide more precise information? And there is a mechanism available for Phase III trials to get more precise information. So where you've been told in the past that robust for a single trial study means .1 or they want to see .1, why is it reckless not to ask as to the second study what P value the FDA wants to see? We don't know that did not happen, Your Honor. We don't have the full record of communications between the FDA and the company along the way. What we have is allegations of a complaint in which he says you should have deduced this result, elevated that deduction into a requirement, and robbed the company of the ability to advocate in front of the FDA. That, I think, is really a significant point, and that what the plaintiffs are complaining about here in many respects is about the day-to-day common interaction between the FDA and a regulated entity in which they take the comment of someone, a staff person, whatever, elevate it into both a requirement and gospel, and essentially preclude the company from advocating on its own behalf in front of the FDA. The company thought, and recognizes well, the rest of the context in which the FDA has on single studies approved many, many other drugs without a .01 requirement, even without a .05 requirement, including products that compete directly or that are in the same reproductive space as this product. If you rob the company of that ability to go in front of the FDA by saying that the securities law is trumpet and you are obliged to disclose whatever the FDA says to you and elevate it into gospel, then you are subverting the FDA and the ability of a regulated company. Was your problem, and this is a curious question, was your problem that you ran one test? I mean, you had one study instead of two studies. Precisely. Had you done a second and gotten the same numbers? The company spent essentially all of its resources on a single study because this is a... It's a matter of resources. I mean, I was just curious because you gave it one shot. You gave it one shot. You gave it one shot. What would happen if you had given it a second shot and you got the same numbers? You couldn't afford the $25 million it would take to do the study. I understand the cost factor, but I'm wondering how that would affect... Well, again, without knowing the results of the second study. If you got the same results of the second study. If you got the same results of the second study, you would have said that you easily met the test because if both studies produced the .02 result, you're well within the .05 tolerance. You probably wouldn't have had a panel convened at all. But that, again, goes to whether you're deliberately attempting to deceive investors. The science here is not just knowing something. It's knowing that what is undisclosed will actually deceive someone. There are different disclosures and different time periods. What about the June 28th letter? The June 28th letter is a filing letter required as a matter of law by the FDA, and it is not, as Mr. Lieberman just said, a warning letter of any sort. The FDA has an entire hierarchy of... I know. That June 28th letter says, look, the... Subgroup data is an issue, a review issue. It's an issue. It's a review issue. Precisely. And so that is... And that is as of at that time. Shouldn't you have disclosed that letter to the public? Because some people bought after June 28th. Sure, they bought after that. But I would say that's exactly what you should not be required to disclose, and the danger of misleading people if you did is much higher than if you didn't. If you tell the world, the FDA has said, this is something we want you to address during the course of our review process, and you thereby elevate that issue into something that is the FDA's point of view and bound to, you know, cause you a problem, indeed lead to inevitable rejection, as the complaint says, and that's not true, then the risk of misleading investors into believing that you are not going to be approved is much greater. You are, again, robbing the company of the chance to advocate for itself in front of the FDA, especially on a point of statistical significance on subgroups that, frankly, from the company's point of view, was absolutely wrong. I mean, the reason to do a study across multiple sites is precisely because individual sites can have a problem. In other words, you're saying disclosure at that time could be considered to be reckless in the other direction. Absolutely, Your Honor. People are going to adopt their stock and sell. They're going to sell on the way down. They're going to get less. You're going to correct the problem, and people who bought at the low cost are going to profit. Exactly. And so what you tell investors is here is the single requirement we have to meet, statistical significance. We don't know what the FDA will require for robustness or whatever. That's the requirement we have to meet, and we make no promises. But if you highlight anything else, you run the risk of misleading people. Why is it highlighting or elevating it to anything? Why not just disclose the fact the FDA has identified a potential issue for review? If you said only that, again, I think you run the risk of telling people that this issue is serious. It's real. It's material. It could have an effect on approval when, from your point of view, it does not. It's just a review issue. Now, Mr. Lieberman has taken that, you know, review issue, elevated it into matter of law, and said it was a new requirement. Not true. But, again, that goes to the calculus. Ultimately, what you have to do in a securities case, in looking at scienter, is weigh competing inferences. It's the one place on a motion to dismiss, under 12b-6, where you get to actually look at the competing arguments of the parties. It's the only context in which that happens. So the question is, when faced with something like that, this is something we want you to address during the course of our review. We do not tell you it's dispositive. We may even be wrong in our wrong assumption. We're giving you the opportunity to advocate for yourself. Is it more likely that the company, by saying we will deal with this, we only have to meet statistical significance at 5% and we're not promising you anything? We have to get over to the state. Well, again, the one case I'd point you to on that very subject is your own decision, or this court's decision in June in Pfizer, where, again, a company raced ahead, took results of data, and moved to Phase 3, having said they have to be spectacular in order to get there. Now, this court said that was not even misleading. But the question in probability, in weighing the inferences, this court said, with the initiation of Phase 3, spending millions of dollars, required FDA approval, rendering it improbable, defendants would have continued if they did not believe their interpretation of the interim results or if they thought the drug a complete failure. That's this case all over. Sorry, one last question. Do I understand correctly from, you were saying before that if you had better statistical results for a second study that showed those results, that the advisory panel wouldn't have been convened. So can we take from that that you agree that the district court, in identifying the convening of a panel as raising a competing inference, was mistaken? Absolutely not. I was responding to Judge Fischer's question about if there had been a second study. If there had been a second study and both were .02, then you wouldn't need a panel. It would be obvious that you were within a .05. It would be easy. But the convening of a panel, the district court took that to mean that there's evidence for a positive outcome, that this was an inference the other way. From what you said, you seem to suggest that that's not the case. Absolutely. And the reason is this. If the requirement were .01 and the result was .02, why convene a panel? That's what Judge Hochberg realized, is that there would have been no reason to convene a panel and consider all kinds of things, including subgroup data or whatever, if it was so obvious. If Mr. Lieben was right that .01 is a requirement or subgroup data was dispositive, why convene a panel? That's what Judge Hochberg realized, is that all this indicated that all these issues were in play. And robbing the company of the ability to advocate for itself made no sense. And letting the securities laws trump the FDA's relationship with an issuer makes no sense. Thank you, Mr. Lieben. Thank you. Ms. Rudson. Good morning. Abby Rudson for the Watson defendants. This case against Watson is pretty unusual. We were Columbia's largest shareholder. So this is a suit brought by a class of Columbia's shareholders against Columbia and its executives. And then they've also added their fellow shareholder who, under their theory, was the biggest victim of this supposed fraud, because we suffered the single biggest loss when the FDA didn't approve the NDA. Up to this point, Watson can only be held responsible for the statements it actually made. I don't think the plaintiffs would dispute this. It's clear from the Supreme Court's decisions in Janus and the Stonebridge Capital case. There was a statement of a David DeVos. Is he an employee? He was a disruptor. Did he have any management role in Watson? He did not. He was just a sales. I mean, he might have been called a sales manager, but I don't consider that sort of corporate management. It seemed like early on they were making preparations to bail out. No, Your Honor, and that's an important point that I was going to make next, and that's on the timing issue. So Watson only made five allegedly misleading statements, five statements in the complaint. The last one was on April 26, 2011, which is only five months into the 14-month class period. They did not speak on this topic, or at least were not alleged to have made any misleading statements on this topic again. So that's in April 2011. The June 28 letter is on June 28. So more than two months later, that's when Columbia gets this letter raising potential review issue. It's not alleged whether Watson actually saw it at that time, but it doesn't matter, because even if Watson should have known when it got this letter on June 28, that wouldn't show scienter in statements it made praising the study months and months earlier. And the DeVos allegation, to your point, is that – after the June 28 letter, this thing wasn't going to get approved. Well, what he said was he learned that it probably wasn't going to get approved. They were making preparations in case it didn't get approved. But even assuming – and we have arguments in our brief about why one statement from one guy with no allegations about how he would know this, he was a sales manager. And yes, of course, Watson is making arrangements for the possibility that it wasn't going to be approved, because everyone understood that this is a risky thing, trying to get FDA approval. But even if you take it at absolute face value that by June 29, the day after the letter, Watson had figured out, we're screwed, we're not going to get approval, that does not show knowledge or scienter or anything about statements that were made months and months earlier. I agree with that. They're complaining about a statement we made in December 2010. But the question would be, what obligation do you have after that point to disclose? We have no obligation. This is an essential tenet of securities law, going back to Basic v. Robinson. Silence, absent a duty to disclose, is not misleading under Rule 10b-5. We don't owe our fellow shareholders any duty. We're not an insider. We're not a fiduciary. We're just another shareholder. So even if this Court determines that Columbia should have disclosed this warning letter, that doesn't mean Watson should have. In fact, there's no allegation Watson ever even saw the letter. And in fact, on that point, we actually did speak. We spoke to our own shareholders in an analyst call before the FDA briefing book was published. So before the first allegedly corrected disclosure, we told our own shareholders, yeah, we've got this product that we're buying from Columbia, and we told them the U.S. subgroup data is not statistically significant. We told them the results were better at foreign sites than at U.S. sites. And Columbia's stock didn't move at all. So for Mr. Lieberman to come in here today and say this was material, we totally should have known this, I don't see it. The shareholders didn't care, or at least the price didn't move, when we told our shareholders, hey, the U.S. data doesn't look so good. Foreign data is much better. When did you make that disclosure? That was January 5, 2012. It's in the record at A175758. And there was an analyst call where somebody from Goldman Sachs, an analyst from Goldman Sachs, was asking questions about the FDA panel that was coming up. And to Mr. Donovan's point, Mr. Bassaro, the CEO of Watson, said specifically, I'm going to be a little careful here because as we head toward the panel, I don't want to preempt the arguments that will take place. So as Mr. Donovan said, you advocate to the FDA. So to say that he should have told Watson shareholders they might point to all these flaws, that's not really a way to get a drug approved. And he was trying to caution shareholders, pros and cons here. And he said the study was successful. The study design was approved by the agency in advance, of course, of the study taking place. What we saw, what was seen as an anomaly, I guess, between the ex-U.S. population versus the U.S. population. And then he goes on to say if you pull out, both cohorts pass, okay, but they're not statistically significant. So to Mr. Lieberman's point that this data was anomalous, well, that was in the public domain weeks before the first corrected disclosure or alleged corrected disclosure, and there was no effect on that. All right. Thank you very much. Thank you very much. You save time for a rebuttal, Mr. Lieberman. I'd like to pose a fact scenario to your honors, a situation where a publicly held corporation, every quarter, says they have a billion dollars in cash on hand as of the end of the quarter. Now, it turns out, behind that billion dollars of cash, they have actually no cash that's their own. Every week, one week before the end of the quarter, they need to get a loan or they need to get money from an affiliate or they need to get money from a corporate friend or colleague. They'll give them that money. So the statement is factually correct. They've got a billion dollars of cash on hand, but the underlying facts of that disclosure are utterly false. And so that's really the question right before your honors. We have statements regarding the company we won't deny. Certain statements regarding fiscal significance at 95 percent, they're factually correct, but the underlying data, that's the question. We have subgroups here that are similar to that bank situation where there's a real suspicion of either inconsistencies, but really, let's look at it, your honors, it's really the issue is possible manipulation. You have such overly perfect results. In Belarus and South Africa, 50 percent of all people on progesterone, all of a sudden, for some reason in Belarus, they have preterm birth. Thirty-six percent in South Africa, it happens to all be in the same subgroup. So if you're in the same pool, you pool together, it's an outweighed result for the company. Zero of them, absolutely zero, which has not occurred in any other subgroup, zero of them get preterm birth on the progesterone. It's an obvious question. It's an obvious issue of manipulation, of the transparency of the data, of the integrity of the data. That's what the situation is. Possibly a factually correct statement, yes, but is it really giving investors a true understanding as to how that data came about? That would allow any corporation, your honor, make factually correct statements, and if there's fraud all around those statements, leading up to those statements, you don't have to disclose those. Man, that's fine. Mr. Lieberman, can I confirm that you are not challenging on appeal the district court's rejection of the legal significance of DeVos' statements? No, we do challenge that. We wasn't focused on the brief, but we do challenge that. It's not in your briefs. We haven't heard from you an argument. How is it that you have raised that on appeal, and what would you say about it? We have limited space to cover allegations, but you had someone who obviously understood the import of the warning letter, and both Watson and Columbia understood the import of the June 28th letter, and obviously that import was not disclosed to investors. As far as the Watson's disclosure at January 2012, all they discussed was the U.S. versus non-U.S. issue, and they discussed that in response to an analyst's question. They don't discuss the Belarus and South Africa issue, and they don't discuss the 99% issue. These are issues that have to be looked at holistically. There's an overwhelming problem here that they're not meeting the single-study NDA requirements. What case law or guidance can you point to as authority that disclosure of only top-line results can give rise to a 10b-5 claim? It's not about disclosure. You have Shapiro versus UJB, Third Circuit. When you make a disclosure, you don't just have to be factually correct. Once you speak, you need to speak fully and truthfully. We can find cases throughout the United States, appellate and Supreme Court. When you make a statement, it has to be a fully transparent statement. It can't be just factually correct by some technicality. It has to be fulsome. It has to give investors a full understanding as to what the material facts are regarding those statements. When you speak, you must speak fully and truthfully. But the focus that you want us to take to suggest that there was some knowledge of factual inaccuracy as to the U.S. statistics being primary, isn't it clear from what's stated in the panel meeting that that was a post hoc change in the analysis by the FDA? Oh, absolutely not. In the SAP plan in 2011, that was before the company submitted to the NDA, they agreed they would do a sensitivity analysis for the U.S. population. So absolutely not, that was post hoc. That was when part of the SAP agreed upon was you've got to take a look at the U.S., especially you're doing foreign studies, half the site is foreign. You want to know how is this working in the United States in a regulatory regime that is tight, that has a history of good regulations on FDA practices, and how is the study working out? Here, it didn't work out at all. It failed miserably. You had a 2.4% moderate success level, but only at a P of .659. It means there's a 65% chance that the U.S. results were basically by chance and pure happenstance. There were FDA criticisms of the 302 study, but wasn't the essence of those criticisms a difference between the FDA and Columbia over how to analyze the data? Oh, absolutely. We disagree. First of all, with respect to the T statistic, 99%, the issue is whether or not the FDA obviously was under the position that they clearly told Columbia what the regulation is. Columbia can say they disagree. They didn't understand that. Our question is whether or not Columbia believes that they told them or Columbia believed it. It was a risk to investors when Condell is making a statement, Columbia is making a statement, that the FDA gave the guidance. It may have been a risk to investors, but can Center be inferred from the failure to disclose that? Sure, absolutely. If the FDA is telling you how they interpret guidance, the FDA tells you how they interpret guidance, and investors are asking you here, does the FDA take a certain interpretation when it comes to single study? Doesn't the investor have any risk? Doesn't the investor have to assume some risk here? You invest in these companies based on various drug trials. You have to assume some risk that this NDA isn't going to be approved. You certainly assume the risk of non-approval. You don't assume the risk of not being told full information, not being transparent in violation of the securities laws. Mr. Donovan's argument regarding the conflict between FDA scenarios and the securities laws don't pass. This is a litigation, a securities litigation, a securities class action. The question is, did Columbia's statements and Watson's violate the securities laws? We could care less about the FDA, how it could hurt the FDA's processes. It's really not an issue here. If Mr. Donovan wants to bring a separate suit for preemption, let him do so. But we're bringing securities for a claim, and there's no weight given to the FDA processes. So any argument resting on that is completely flawed. This is about whether or not under the securities laws, the federal securities laws, was there a failure to disclose, period. FDA issues aside, it's really not an issue. Mr. Lieberman, thank you very much. Thank you. Thanks to all counsel for excellent arguments. The court really appreciates it. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A10484. Public record. Not legal advice.