# Oral argument — Outsourcing v. FDA > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A104082 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** March 30, 2026 ## Text Good morning again, your honors. May it please the court. There are many similarities between these two cases, but I'd like to address some of the differences between them because I think that may be helpful. I think a fundamental difference, this time around in the facts, is that the FDA actually admitted in its briefing, and this is on page 48, that it simply accepted the manufacturer's data in the form that it was provided. In other words, rather than prescribe, here are the parameters that we think are required to make this determination, the agency simply eyeballed what the manufacturer sent it and said, well, that's good enough. It did not make any decision on its own as to how the data should be formatted and in turn how that would then drive the analysis. And I think that explains in this case how the FDA missed that the only measure of supply versus demand over time recited in its decision used a faulty measure of demand that was capped at the amount that Novo Dornis was able to supply. In other words, it could never show a shortage. And it gets worse than that. Novo Nordisk's own data shows that it was turning away nearly half of wholesaler orders for starter dosages in the months leading up to the delisting action. Novo told the agency, and this is quoted in the decision, that it did this to ensure that patients starting treatment would be able to continue with the larger follow-on doses. And that is an outright admission that Novo was manipulating future demand for the larger follow-on maintenance doses because otherwise it couldn't keep up. There's no indication that the FDA even understood the gravity or the meaning of that admission, and it certainly made no attempt to rationalize it. Instead, it simply accepted an alternative incompatible explanation. And FDA's deference to Novo led it to ignore or wave away once again all other evidence, which again all pointed to an ongoing shortage. That includes wholesaler stock reports, patient survey data, even admissions of demand outpacing supply by Novo's CEO and the CEO of one of the big three drug distributors. This is not reasoned decision-making. I'd like to begin by focusing on the demand manipulation point because I think this is very important to understand. So the way it works is that a patient begins with a smaller dose of one of these products and then titrates up over a period of time to the larger maintenance doses that then continue indefinitely. So the .25, the .5, and the one milligram starter doses are generally taken for about a month apiece. So it's a three-month period to ramp up to the larger 1.7 and 2.4 milligram maintenance doses. So this occurs over a period of months. As Novo admitted, it was not fulfilling orders for those initial starter doses because it had concerns about being able to satisfy the demand for the much larger maintenance doses that would be required for each person who made that journey titrating up. If you just play out the numbers, what that means is that Novo's not filling those orders would commensurately reduce demand for those larger maintenance doses three to four months in the future. As I said, the agency didn't seem to appreciate this. Oddly, it reprinted the admission in its decision and then just skipped right past it. But I think this calls into question basically all the figures contained in the decision because they all assume that demand was not being manipulated. In other words, that Novo was confronted with just ordinary market demand and then kind of took it as it came. Whereas in reality, what Novo admitted was that it was actually manipulating demand into the future in a way that affected the time period that was subject to the agency's analysis. Again, I noted that the FDA, and a particular metric or standard, but instead accepted Novo's data in the form that it was provided. In other words, the use of numbers like realized sales to measure demand, which made it impossible for there to be a shortage in the data, inventory and sales data that was unmatched by commensurate demand figures, and then the use of averaging in various ways. And this really does, I think, in a certain sense, apply to both cases, if not in specifics, then the general point, which is the manufacturer knows what the data are. And so if the data are presented, in other words, the form, the days we're going to take particular measures, whether we're going to use averages, how those averages are calculated, the manufacturer can effectively rig the outcome. In other words, the manufacturer can then decide this presentation supports our position better than some more fulsome presentation or some other way of putting the data, some other form of data. And the manufacturer has the ability to choose those forms that support its case while not providing the data in the forms that might be more revealing. And this is why it is so important that the agency has to be the one to make the determination. We think that's true in both cases with respect to the time period under analysis. I think it's notable that the time period in this case, again, the agency didn't make a finding on this, but if you infer from the agency's reasoning that it undertook a six-month analysis, I think it raises questions why it did a six-month analysis in this case versus a 15-month analysis in the prior case. Maybe there's some reason, I don't know. The agency didn't explain either determination, and frankly, we don't even think it made those determinations. But even if you were to infer it, there is no explanation of the time periods, and of course, the time periods can drive the outcome. Even worse here is that, again, in both of these cases, the agency didn't make the finding that it repeatedly concedes is required. The agency in its briefing, and it said it six times in its brief in this case, that all it did was decide whether one number was bigger than another number. Look at the brief very carefully. The agency actually doesn't tell you what those numbers are. It's not in the decision either. We don't know, and we've been litigating this for over a year. We still don't know what the agency thinks those numbers are, and this is really a fundamental problem of the decision. The decision lays out, and this is in the second paragraph of the decision, and then it's repeated elsewhere in the decision, that the agency was required to simply determine whether over a period of time, demand exceeded supply. That's what the agency recognizes, is the statutory definition of a shortage, and therefore, the inquiry that the agency was required to undertake. At no point, after laying out that inquiry and sort of defining for itself the path that it was required to take, does the agency actually go ahead and do that. We've said that that's the case with respect to the time period. Again, there's no finding with respect to what the proper time period of analysis is, but it's also the case with respect to the outputs of that decision, the supply and demand over the time period. Instead, in this case, once again, the agency deferred to Novo's choices. We think that's arbitrary in itself for the reasons that the Supreme Court explained in State Farm. Obviously, a corporation, a business that is participating in a regulatory process, has incentives that may differ from those of the public at large, and so the agency is required to make these determinations itself, rather than simply deferring without explanation to the agency's, I'm sorry, to a corporation's and a business's framing of things. And there is, in this instance, some indication of gerrymandering of the data. I already discussed figures one and two. These are the only cumulative supply and demand over time figures, the ones where, as I noted, the demand figures are based on realized sales, so it doesn't include unrealized sales, sales that could not be fulfilled, which means it could simply never show, it could never show any sort of shortage. This is not useful in determining whether or not there is a shortage. There's also, as we identified in our briefing, a mismatch between different types of numbers. If you're going to get all these sorts of incommensurate forms of data from a manufacturer, I think it behooves the agency, just as a matter of diligence, to kind of line them up when that's possible. I mean, that's basically the entire profession of accounting, as I understand. And in this instance, you have tables four and five that identify the quantities that Novo actually supplied, and then you have figures one and two that identify realized sales. In other words, what Novo described as demand, but it's realized sales. And in some measure, those two numbers should line up. But the problem is that when you do line up the data, it shows that demand exceeds supply. Now, maybe there's some explanation for this. Maybe there's some reason the numbers don't line up and you wind up with these negative results. But the agency didn't even notice that, and if it had noticed that, it should have inquired into it, because it suggests that the form the data was being presented in was not revealing the state of the world, but perhaps was obscuring it instead. Then you have the consistent statements with respect to shortage made by Novo Nordisk's CEO and the CEO of Cardinal Health. The decision, you know, I look at this and I think, well, if the CEO of a company that's claiming there is no shortage gives an interview with a major media outlet saying we can't keep up, that there's still unsated demand, that we're going to be selling into that unsated demand for months and months and months and months. You know, there's a reason a CEO would say that, but it strikes me as very probative of that company's business, because that's the reason the CEO is making that comment. Likewise, with respect to the CEO of Cardinal Health. Cardinal Health is an interested party in a certain respect, in that it wants to sell more of Novo's products. It's a wholesaler. It sells the products that are made by the major drug manufacturers. It doesn't sell the products that are made by compounders. So Cardinal Health would have every incentive in the world for the shortage to be cleared over at the earliest possible opportunity. And yet the CEO of Cardinal Health said, no, there's an ongoing shortage with respect to these GLP-1 medications, and that's something that's been affecting our sales and our revenue numbers. And then finally, I want to note that this idea of an ongoing shortage is also consistent with the on-the-ground evidence. Here, once again, you have what's referred to in the briefing as the screenshot evidence. These are reports from the inventory systems of the major wholesalers showing persistent shortages over a long period of time from a number of different pharmacies seeking to purchase these specific drugs. I mean, if you think about it from a 10,000-foot level, if the state of the world is that there was a shortage, this is where you would expect it to be manifest. You would expect to see it that pharmacies are unable to obtain the drugs in question, and that, in fact, is what was happening here. Finally, I would like to address the agency's treatment of demand as being a treatment of the demand that was being satisfied during the shortage by compounded drugs. The way the statute works is that it contemplates that if the manufacturer can't keep up, then some of that demand is allowed to be shifted to compounders so they can provide necessary drugs to patients. The agency, in its reasoning in this decision, assumed, for the sake of analysis, that there would be a one-to-one transfer. In other words, that when the shortage ended, then on a one-to-one basis, that demand was being satisfied by compounding would be transferred over to Novo Nordisk. But it failed to undertake any estimation of the amount that was being compounded based on what I think is the most probative record evidence. It didn't really even analyze that record evidence in its decision. And I want to note two pieces of evidence in particular. One is Novo's estimate that are provided to the agency that compounding was satiating 20% of the market for GLP-1 drugs. Novo thought that this was probative evidence. Novo attempted to use this to persuade the agency to act in the way that Novo wanted it to act. And now, in court, Novo turns around and says, well, it was just some random number, and we don't know where it comes from, and we don't stand by it. But that's not what Novo told the agency. That wasn't its representation. And the other is one of the comments submitted by the Outsourcing Facilities Association, one of the plaintiffs in this case, and my client, which presented numbers indicating that compounding was satisfying somewhere between 40% and 50% of the market. The agency, in its briefing, has said that it believes that these numbers are unreliable and that, therefore, it had no obligation to consider them. But this is provided by the Trade Association for Outsourcing Facilities, which are large-scale compounders who, again, are exactly the parties one would expect would be the ones manufacturing these drugs in large quantities. And so, at the end of the day, if the agency was going to assume, for the sake of analysis, that there was this one-to-one transfer, it actually had to conduct the analysis based on the evidence before it, rather than saying that it was not going to make any sort of predictive judgment with respect to the proportion of the market being satisfied. If the Court has no further questions, I reserve the remainder of my time for rebuttal. Thank you. Good morning, Your Honors. Since my friend on the other side focused primarily on the arbitrary and capricious argument in this section, that's where I'll start, unless you have further questions. Let me ask you something that I meant to ask last time around. It's not about arbitrary and capricious. It's about kind of what we are actually deciding with respect to adjudication versus rulemaking. In your view, given our precedent, the Supreme Court precedent, are we deciding, I guess, part one is, I mean, how much deference do we give to what the agency calls what it's doing? We put a label on it. There's part one. And second, are we asking whether the agency abused its discretion in proceeding along the track of adjudication or rulemaking? Maybe. Or are we asking, okay, let's look at what they did and say, is it adjudication, is it rulemaking? Well, what are we actually? I know it's a strange question. What are we doing here? What are we doing? I appreciate the question, Your Honor. I think in City of Arlington, this court took a sort of two-part approach evaluating what's the way we evaluated the label that the agency gave to the decision. And so that's part of the reason why we discuss it in our brief. But I do think the fundamental point is what is the nature of that decision? I think that's what this analysis more or less collapses down to and whether the agency, I mean, there is, in context when Congress hasn't specified a particular method, this court has said we look, we recognize that agencies therefore have discretion and we look at whether the agency's choice is reasonable. And so that's the framework that was provided in City of Arlington. But I think the arguments on both sides more or less come down to, again, what is the nature of this decision? Is it fact-based, et cetera? Counsel, can I ask one other question before we get to the ANC stuff? I'm surprised we haven't talked about Safari Club yet, so I'll break the seal on it. Can you help me understand whether we have to create a circuit split with the D.C. Circuit to rule with you on the adjudication rulemaking thing? Because this case looks a lot like, to my eye, Safari Club. I mean, it looks like the agency making a similar sort of, is this thing bigger or smaller? In that case, it's about the threatened species of elephants and they look at the importation from Zimbabwe of hunted elephants. So the inquiry looks similar, the non-parties look similar, the perspective only looks similar, and obviously the D.C. Circuit said that was a rule. So you would not have to make a circuit split, Your Honor. The difference is the finding in Safari Club incorporated a policy judgment, the determination of whether importing these kinds of trophies was better for the survival of the species as a whole. That's fundamentally a question about whether it's better to sacrifice a few for the sake of the many, and that's a policy determination, which is different from the kind of numbers-based, is one number bigger than the other kind of determination here. But wasn't the inquiry that the Fish and Wildlife Service was doing in that case, wasn't it really looking at sort of, you know, the population statistics, the importation statistics, the increases over time of the number of elephants versus the decrease in time over the number of elephants. So it looks like sort of a supply-demand comparison, and you see what I'm getting at? Like the sort of the inquiry of it looks very similar, and it also has the same, you know, one of the key distinctions that you and your friends on the other side have in this case is, well, we're not actually making these policy determinations. The prohibition on compounding comes from the FDCA, to which, you know, if the D.C. Circuit was here to defend its position in Safari Club, they would say same thing, right? The Endangered Species Act and the attendant regulations of the CFR that talk to not just endangered species but also threatened species, those have the independent prohibitions on importation of the hunting and everything of the trophies. So like there's a bunch of similarities and analogous points to it. So I take the point that the agency in the D.C. Circuit case looked at data and making its ultimate policy judgment, but the finding at issue there was again, it's okay to sacrifice a few for the sake of the many. The finding was about whether permitting these trophies would promote conservation of the species, and I think that promote conservation is very critical in distinguishing the kind of finding at issue there and the kind of finding at issue here. Yonah, you also asked a question earlier about the nature of the listing decision. In that case, the FDA made a determination based on from the public and from manufacturers, and there is a statutory requirement that manufacturers notify FDA when there are disruptions in the supply chain. So I just wanted to provide some information about that. With respect to the arbitrary capricious argument, I'm happy to answer any specific questions that your honors may have. Otherwise, I think that the decision more or less speaks for itself in evaluating the thoroughness of the agency's consideration. It looked again at zoomed in, zoomed out evidence of the supply and demand market here. It looked at what was happening with respect to manufacturers, what was happening at the wholesale level. One theme of the presentation on the other side was the agency simply accepting the manufacturer's data without actually sort of critically analyzing it, and that data was manipulating. Of course, I can manipulating demand, and I can ask the order for Novo Nordisk whether they were doing that or not. But what is the agency's response to you just accepted it at face value, the data? So I think there are two responses to their notion that the agency was captured by Novo in this sense. The first is that the agency responded by looking at different kinds of data from multiple different angles. Again, looking at what was happening at every stage in the supply chain at the manufacturing level, the warehouse, wholesaler, and that answers the question that we just sort of handed the reins over to Novo with respect to this decision. I think the administrative record also supports the idea that FDA asked Novo, you know, what's the basis for these numbers? What's going on with wholesalers over here? I think the second argument is that the fact that we're here undermines the idea that the agency is captured by Novo. Sometimes the agency makes decisions that's in a manufacturer's favor, and sometimes we make decisions that are what the compounders want us to make. And we're only here because the agency determined in 2022 that there was a shortage of this particular drug, and that enabled the compounders to do what it is that they want to continue to do. And so I find it very hard to understand the basis for that argument, and I don't think it's borne out by the administrative record. Council brought up that what he said were the varying timelines of the time period, six months versus 15 months. What's your response to that? So the fact that the agency looked at different time parameters in this decision, and that I think only underscores why this is an adjudication that the agency is responding to fact-specific changes in, you know, regarding the health of the market on the ground. And the only reason, you know, the idea that we didn't announce a particular time period in advance is because what the agency was doing here is not a rule where the determination in one instance provides no guidance as to how the agency is going to make a determination in the other. The only relevant question is whether the FDA looked at enough information from which it can draw reasonable inferences about the state of that particular market. And in both cases, the answer is yes. The agency plainly looked at a sufficient amount of data from which it could say, we now think the shortage is over. And in this case, you know, the evidence supports that determination. NOVA was able to meet wholesaler orders, and it had a substantial surplus left over every time. Thank you. May it please the Court, Benjamin Block of Covington and Broom, LLP, on behalf of appellee intervener, Novo Nordisk. Consistent with its statutory mandate to ensure that a shortage list is up to date, when FDA declared that Ozempic and Wegovy were no longer in shortage in February of 2025, the record before the agency showed that Novo was fulfilling all customer orders. On top of that, Novo had millions of packages of product in its inventory. On top of that, Novo had millions of packages of semi-finished product, which could quickly be converted into finished product. And on top of that, Novo was projecting that it was going to double the supply of these drugs, of these medicines, from its numbers in 2024, it was FDA considered all of that data, it looked up and down the supply chain, and it reasonably determined that the shortage was over, because the supply of these drugs exceeded the demand or projected demand. And when FDA did that in its declaratory order, it expressly advised that the agency, and I'm quoting from a record on appeal 1197 here, will closely monitor supply and demand over the following months. And I'm paraphrasing a bit or skipping a bit, but in the future, the agency said if nationwide demand exceeds supply, FDA will return semaglutide injection products to the shortage list, because that's the agency's job, to keep the shortage list up to date. We see that in the record at 1225, 1236, and record on appeal 1486, similar to comments from counsel in the prior hearing, the agency had Novo Nordisk commit to continuing to provide updates about its supply, which the company has done. Ozempic and Wigovi are not in shortage, the shortage has long been over, and subject to any questions that the panel has, we would ask that you refer. I'm sure you heard counsel opposite talk about the idea that Novo is manipulating its demand data. I think I'm getting that right. I heard the term realized sales versus unrealized sales. Do you, to the extent that that's a question, what are your responses? Your Honor, the response more generally to the suggestion that the agency just accepted everything Novo said, the record shows the exact opposite of that. Look no further than the first half of the excerpted records on appeal that we submitted with our brief, all the questions. Page limitations, we could only put part of the questions that the agency was asking. What does this mean? What does that say? On demand specifically, the agency said that we, that the agency disagreed, or said we can't agree that just what prescriptions are getting filled that Novo reports is necessarily the full picture of demand. And this is a case, I think there are cases, you know, it's probably an imperfect measurement of what the full demand is. The agency explained in detail how it assessed that demand, including transitional demand. I heard some comments about CEO statements with respect to demand, and that the agency didn't take those into account or didn't understand the significance of them. Yes, so the record confirms that the agency was aware of the statement and asked Novo about it. That's a record on appeal at 148, I believe. I might have missed a digit there. And the decision more generally confirms that FDA was focused on the FDA's decision, confirms that the agency was focused on the actual data. They asked a lot of questions about supply of orders in October and November of 2024, as well as December. And they looked at the supply at the wholesalers, and they looked at what the supply was, whether it was increasing or decreasing. It was increasing, the suppliers were retaining inventory of all doses. And it's also clear that the agency was most focused on the most recent data. We see that record on appeal 1170 and note 20, and in record on appeal 1175, where the agency says importantly, Novo fulfilled all orders in January 2025. So apologies, Your Honor, if that was a long-winded way of getting around to the question. We think the agency did address it. If the question is, why is there a specific reference to a statement from early November 2024, we would refer the court to, for example, the deep against bar case or the pension benefit guarantee case. Agency is not required to respond to every single piece that's in the record before it. So, Your Honor, can you discern the path the agency was following that the very detailed FDA's decision, you can discern the path that the agency was following, and it was the correct path. Why did the CEO say that? So, Your Honor, what it was, the CEO was talking about two different things that the article took out of context. There was a quote about the demand for drugs to treat diabetes and weight loss more generally. And there's certainly that there remains a demand for that, but it was not for new products and new innovations. And by the way, today, for example, now there is also, in addition to the injectable semaglutide, there's now FDA approved for Wigovi, the pill version of that drug. But so that was part of what he was talking about. The other part of the quote was actually about the worldwide market. It wasn't about the U.S. market specifically. Novo has invested heavily in its efforts and activities in the United States to end the shortage, and Novo has ended the shortage. We'd ask that Your Honor's affirm. Thank you. A few points regarding the arbitrary and capricious arguments. And if there's time, I'll address a couple of points regarding that as a comment. I'd like to begin with the point regarding demand manipulation. Judge Duncan asked my friend about that and actually didn't receive an answer. The quote disappears in the decision on page 1174 of the record, and I'll just read it. It says that Novo, quote, adjusted the quantities provided to wholesalers to ensure the patients who initiate their treatment can continue with their treatment regimen. And then there's a footnote that then explains the titration process and how that leads to larger doses over a period of time. As I said, it's an outright admission that the Novo, the manufacturer, was manipulating demand with respect to future months, the very months that were under consideration by the agency. The agency simply reprinted that quotation and then ignored it. And my friend has no explanation as to what that could possibly mean other than what I just told you it means and what it really just says in the black and white of the decision it means. Second, there were some comments by the government's counsel regarding the thoroughness of the decision. I think this is a little bit of a red herring. The decisions are, in both cases, a number of pages. Our complaint isn't with the number of pages. We actually think the agency, if it took a more focused approach that used like regular inventory management, I'm sorry, regular inventory accounting protocols, could probably have much shorter decisions with not all that much additional reasoning to them. The problem is that the agency has to say, here's how we're going to do it. And that's something that we just haven't seen here. So we're not asking for more. We're not trying to nitpick. What we're saying is that the agency has to just decide on an approach and then apply that approach. And we don't think the approaches that would be appropriate are necessarily a complicated approach. Third, you know, as I mentioned, the agency said over and over and over again, and I mentioned this in my remarks, that all it did was decide that one number is bigger than another number. I didn't hear either of my friends identify what either of those numbers might be or where they might be found in the decision in this case or, frankly, in the last case as well. Again, I think that speaks volumes. Turning to the notice and comment issues, Judge Oldham, you asked about what degree of deferences do. We think the place to look for this is the professionals and patients for customized care case from 1995. It's cited in the W&T offshore case. And it says that the agency's labeling of a determination is entitled to minimal deference. So far as I'm aware, that reflects the earliest view of this circuit on that particular question. And so we think that's the one that's controlling. And then you also asked how to frame the determination. We, again, think that W&T offshore as well as Shell offshore, you know, are really correct in saying that the agency or, I'm sorry, that the court, rather, looks past the label and determines did the agency do what it said that it did as opposed to determining whether the agency was arbitrary and capricious in its choice of vehicle. That said, the court could look at it that way as the court did in City of Arlington. At the end of the day, I'm not sure that it makes all that much of a difference. Judge Oldham, you also mentioned the Safari Club case. And my friend representing the government said that, you know, what was at issue there was a policy determination. But if you actually look at the way the court described the determination in that case, it was really just all about the evidence. In fact, that was the reasoning for the determination that the agency issued in that case is that it felt it didn't have enough evidence to make the sort of judgment that was required so that it could issue a finding that would allow the importation of trophies. This is just a fact-based determination. It certainly was in that case. It won't surprise you to hear that we think the Safari Club is on all fours. And I think the court would have a very difficult time distinguishing it. I would also note that the proposed grounds of distinction is one that, as far as I'm aware, no court has ever applied. In other words, the court—and it's entirely unworkable—the idea that this court should draw a line between facts and policy and then somehow it would administer that in deciding whether something is a rulemaking or an adjudication. If you actually look at the way the cases work, like, for example, Florida East Coast Railway, that is not the way they took this. That is not the approach that they followed. The approach that they followed is the one that we've described and the one that's described in Safari Club. There is simply no way to draw a line between facts and determination. You know, whether a particular type of air emission is a pollutant, like, is that a fact question or a policy question? You know, it beats me. If you look at the standard, it seems pretty fact-based, but I guess you could argue it either way. I think there's a reason that that's not the way courts have looked at that question. And then finally, with respect to the timing that might be available for notice and comment, I think the court shouldn't be too focused on this idea that these are snap judgments that are made to list or delist certain substances. Because if you look—and this is Section 356C, sub B, and sub F of the statute—they both contemplate months-long periods leading up to a delisting—I'm sorry, to a listing determination. And nobody's ever claimed that there's any snap judgment that's required, there's any imperative to have an immediate determination with respect to a delisting decision. If the court has no further questions, we would ask that it reverse. Thank you. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A104082. Public record. Not legal advice.