Hello and welcome to my Quimby presentation on how recent case law has reshaped the BPCIA. My name is Christopher Loh. I'm a pharmaceutical patent litigator at the law firm Venable LLP. And if you're watching this presentation, I'm going to presume that you have a basic understanding of what biologics are and what the BPCIA is. But because this material is quite complicated, I'm gonna do my best to go slow and give you the background you need to understand what I'm talking about. Just as a reminder, the opinions and advice I'll be providing in this presentation are solely my own, and don't reflect those at Venable LLP or its clients. And in any event, I'll generally tend to stick to the facts and keep my editorializing to a minimum. Now, let's start with the fundamental question underlying this presentation. What are biologics? Put simply, biologics are drugs whose active ingredients are large, complex molecules that are produced in living cells. These include antibodies, large proteins, and large nucleic acids. Well-known examples of biologics include Humira, or adalimumab, an antibody that's produced in Chinese hamster ovary, or CHO, cells, and Enbrel, or etenercept, a fusion protein that's also made in CHO cells. Because biologics are large, complex molecules that are produced in living cells, they are by nature more variable in terms of their structure and function than small molecule drugs. For example, antibodies produced in living cells are glycosylated, meaning that the cells which produce the antibodies attach carbohydrates called glycans to them. And glycosylation is a heterogeneous process that results in variations in the structure of the resulting antibodies. So unlike, say, the small molecule drug Tylenol, in which the structure of the acetaminophen drug molecule, which is shown on the right, is relatively small and simple and uniform from molecule to molecule, the structure of the Humira antibody, adalimumab, on the left, is complex and may vary from molecule to molecule. And that's where we get the term biosimilar. So everyone knows that we have generic small molecule drugs that are equivalent to branded small molecule drugs in terms of their therapeutic and pharmacokinetic properties. But due to the inherent variability of biologics, it's generally not feasible to make generic biologics that are equivalent to an earlier FDA approved biologic in terms of their therapeutic and pharmacokinetic properties. It is, however, feasible to make biologics that are similar to an earlier FDA approved reference biologic in terms of their safety, purity, and potency. And these later products are called "biosimilars." The earlier product to which they're similar to is referred to as the "reference product." Now, although biologic biosimilars offer the promise of delivering therapeutic benefits that are similar to their reference product counterparts at a lower cost, that promise has yet to be fully realized. And currently there are fewer than 30 biosimilar products on the market in the U.S. Now, why? Some of this is due to weakness on the demand side. For example, physicians and patients may be unfamiliar with biosimilars and their therapeutic or safety attributes, and some of this is due to weakness on the supply side. Drug makers may encounter significant technological, regulatory, and patent related issues that may dissuade them from developing biosimilars. And that explains why we have the BPCIA, or the Biologics Price Competition and Innovation Act. The BPCIA, which is codified at 42 U.S.C. Section 262, was passed by Congress in 2010 to spur the development of biosimilars by establishing an abbreviated FDA approval pathway for biosimilars. So just sort of smoothing the flow to get more biosimilars to the market. Now, similar to the Hatch-Waxman Act for generic small molecule drugs, the BPCIA links the FDA approval process for biosimilars to the patent mitigation process between the biosimilar maker and the reference product maker. Now, the purpose of this linkage is to provide a mechanism for the biosimilar maker and the reference product maker to raise and resolve patent disputes before the actual commercial launch of the biosimilar. So like in Hatch-Waxman litigation, the parties to the BPCIA litigation process get to go to a federal district court and hash out their patent plates before an infringing product actually shows up on the market. Now, for those of you who are already familiar with Hatch-Waxman litigation, I've prepared this handy slide highlighting three key differences between the BPCIA litigation process and the Hatch-Waxman litigation process. And of course these aren't all the differences, these are just some of the key ones that I've highlighted. First in BPCIA litigation, the reference biologic maker can assert its process patents. That is the patents that cover the manufacturing processes for biologics. And this is really important in the biologics context, because oftentimes the process dictates what the key attributes of the product will be. In Hatch-Waxman litigation, by contrast, the reference drug maker cannot assert process patents. Second, in BPCIA litigation, unlike in Hatch-Waxman litigation, filing suit doesn't result in an automatic 30-month stay of FDA approval of the biosimilar. So those of you who are familiar with Hatch-Waxman litigation know that if you bring suit within a certain deadline, you'll be able to obtain a 30-month stay that will prevent the FDA from finally approving a generic drug. That stay isn't present in the BPCIA litigation process. And third, as we'll see in a moment, the BPCIA sets forth a relatively complex pre-litigation information exchange process between the reference product maker and the biosimilar maker, whereas the Hatch-Waxman does not. Now, because the patent litigation process that's prescribed by the BPCIA is complicated and depends largely on the voluntary compliance of the biosimilar maker and the reference product maker, incentives exist to test compliance with the BPCIA's provisions. And this has resulted in a body of case law that's reshaped, in sometimes surprising ways, the canonical patent litigation process that is set forth in the BPCIA. To better understand how that recent case law has reshaped the BPCIA, it's important that we first understand how the BPCIA's canonical patent litigation process is supposed to work. That process, as I mentioned, includes several information exchange steps between the biosimilar maker and the reference product maker, which are commonly referred to as the "patent dance." So who are the dancers? Well, there's two of them in the BPCIA patent litigation process, and I've been talking about them for the last couple of minutes. It's the reference product maker and the biosimilar maker. Now, for the remainder of this presentation, I'm going to refer to the reference product maker as the reference product sponsor, or RPS, and the biosimilar maker as the biosimilar applicant, or BA. Now, in the wild you'll likely hear many alternate terms for both of these dancers, like the BLA holder and the Section K applicant. But to keep things simple, I'm gonna stick to the terms RPS and BA here. Now, the RPS is the owner of the patents that are the subject of the BPCIA patent litigation process. The RPS files its biologics license application and obtains FDA approval for the reference product pursuant to 42 U.S.C. section 262a. To obtain FDA approval for the reference product, the RPS must prove to the FDA that its reference product is safe, pure, and potent. And upon FDA approval of the reference product, the RPS is entitled to four years of data exclusivity and 12 years of market exclusivity. This means that the FDA may not accept an application for a biosimilar until four years after the approval date for the reference product, and the FDA cannot approve that application until 12 years after the approval date of the reference product. So this timeframe gives the parties an eight year gap. That is the time between the expiration of data exclusivity in year four and the expiration of market exclusivity in year 12 to resolve their patent disputes. And the entire BPCIA litigation process is supposed to sit within that eight year gap, which is a pretty generous time to resolve patent disputes. But as we'll see in a moment, this process can be distorted in situations where the RPS' data and market exclusivity have already expired. So without that eight year gap, there's an incentive for the parties to rush, as we'll see. Now what about the BA? The BA is the party that's developed the biosimilar and wishes to obtain FDA approval for that biosimilar. To do so, the biosimilar files an abbreviated biologics license application, or aBLA, under 42 U.S.C. section 262k. And that's why you'll sometimes hear a BA referred to as a section K applicant. The BA need not undertake the same amount of effort as the RPS to show that its biosimilar is safe, pure, and potent, but it still must demonstrate that the biosimilar is highly similar to the reference product in terms of its safety, purity and potency. And unlike the RPS, the BA ordinarily is not entitled to any period of market exclusivity unless its biosimilar is the first, quote, "interchangeable" biosimilar for that reference product. And I'll discuss interchangeability in a moment. Now, one potentially confusing aspect about the participants in the BPCIA litigation process is that, unlike in the small molecule drug market, there's no clear dividing line between the companies that appear as RPS's and BAs in BPCIA litigation. So many biologics companies make both reference products and biosimilars, and the same company may show up as the RPS in one lawsuit and as the BA in another, as you'll see in some of the cases that I'll be citing. Just something to be bear in mind. Just to add some rigor to our discussion of biosimilars, here are the BPCIA's definitions of the terms biosimilar and interchangeable. The term "biosimilar" means that the BA's product is, quote, "highly similar" to the reference product and that there are no clinically meaningful differences between the BA's product and the reference product in terms of safety, purity, and potency. Now, what do the terms "highly similar" and "no clinically meaningful differences" mean? The BPCIA itself doesn't say. It leaves those terms to the FDA to define. And the FDA doesn't offer bright-line definitions of similarity, but instead has provide guidance to say that establishing similarity depends on a totality of the evidence, including the results of analytical studies, animal studies, and clinical studies on the biosimilar. The term "interchangeable" means that the BA's product not only meets the statutory definition of biosimilar, but also that the BA's product can be expected to produce the same clinical result as the reference product in a given patient, and that switching between the BA's product and the reference product poses no greater risk in terms of safety or diminished efficacy than the risk of using the reference product without any switch. Thus, interchangeability is a higher and more difficult standard for the BA to meet than mere biosimilarity. And for that reason, only four interchangeables have been approved by the FDA to date. There's some legislative action to lower the bar for interchangeability. We'll see where that goes, but it's on the table now. Now that we've identified the participants and the products, let's take a closer look at the BPCIA litigation process itself. Now a signature aspect of the BPCIA litigation process, unlike the Hatch-Waxman litigation process, and one that's quite important to this presentation, is that it's designed to occur in two phases. Phase one begins when the BA submits and the FDA accepts its aBLA. Under section 271e2C of the Patent Act, which is 35 U.S.C., the act of submitting the aBLA to the FDA constitutes a so-called artificial act of patent infringement that bootstraps subject matter jurisdiction. And that's what allows district courts to hear BPCIA litigation in advance of the actual commercial launch of an infringing product. So ordinarily in patent litigation, you have to wait until there's an infringing product on the market or about to come on the market in order to get into district court. But the BPCIA, like the Hatch-Waxman Act, provides that there is this artificial act of infringement that allows the parties to come into court before that time. And the aBLA submission also starts the patent dance. During the dance, the BA discloses information about its biosimilar to the RPS, the parties negotiate which of the RPS's patents should be litigated in phase one, and then the parties proceed to litigate those patents in phase one. Now, phase two begins when the BA provides the RPS with 180 days advance notice prior to the first commercial marketing of its biosimilar. This kicks off a period during which the RPS can sue the BA on any RPS patents that were identified but not litigated in phase one, and any newly issued or licensed RPS patents before the commercial launch of the biosimilar. And this phase two process is also a key difference between the BPCIA litigation process and Hatch-Waxman litigation. Oftentimes in Hatch-Waxman, the parties have to independently negotiate a notice provision, whereby the generic will provide advance notice of when it will actually commercial launch its generic drug. Well, the BPCIA builds that notice right into the statute itself. Note that both phases are triggered by the actions of the BA and not the RPS. This gives the BA considerable control over the start and timing of both phase one and phase two patent litigation, as we'll see in a moment. So let's zoom in on phase one. And as you can see from this slide, phase one is quite complicated, so I'm going to baby step you through it. And for reference, the phase one steps on this slide that are numbered 3 to 10 constitute the steps of the patent dance. So phase one begins when the BA submits aBLA to the FDA and the FDA accepts the aBLA for review. 20 days after the FDA accepts the aBLA for review, section l2A of the BPCIA provides that the BA shall provide its aBLA and the manufacturing information for its biosimilar to the RPS. 60 days thereafter, the RPS provides to the BA a list of the RPS's patents that the RPS believes could reasonably be asserted against the biosimilar described in the aBLA. And this list of patents is called the l3A list, logically. 60 days after the RPS provides its l3A list, the BA provides a list of additional patents that it believes could be asserted by the RPS against the biosimilar, as well as its non-infringement and invalidity contentions for the patents that are set forth on both the l3A and l3B lists. 60 days after the BA provides its l3B list of contentions, the RPS provides its responsive l3C infringement inability contentions for all the listed patents. And after the RPS provides its l3C contentions, the BPCIA requires that the RPS and BA engage in good faith negotiations over what patents to litigate in phase one. Note, however, that the BPCIA does not set forth any deadline as to when such negotiations must begin, which is why I have these three question marks for this particular part of this phase one process. If the RPS and BA reach agreement over what patents to litigate in phase one, then the RPS must, within 30 days of reaching agreement, file suit on those patents. If the RPS and BA don't reach agreement over what patents to litigate in phase one, the BPCIA sets forth the dispute resolution process that I've shown in the red boxes at the far right here. In this process, the BA must notify the RPS of the number of patents that should be litigated in phase one. Five days thereafter, the parties must exchange competing l5B lists of the patents they believe should be litigated in phase one. And 30 days thereafter, the RPS shall bring suit upon all of the patents included in the section l5B lists. So that's the way the process is supposed to work. But because phase one is so complicated and requires a significant amount of time and effort to complete, both the BA and the RPS have incentives to vary or circumvent steps in the patent dance. One key step that the BA has incentive to vary occurs at the very start of the patent dance. So a BA here may be particularly reluctant at the outset to provide its proprietary manufacturing information for its biosimilar to the RPAs, despite the fact that the BPCIA includes at section l1 a default protective order for that information. And the reason for that reluctance is because, as I mentioned earlier, manufacturing information can be particularly valuable in the context of biologics, because key attributes of the BA's biosimilar product may arise out of proprietary aspects of the BA's manufacturing process, such as the cell lines, the growth medium, and bioreactor parameters that are used to make the biosimilar. So put simply, there's a lot of secret sauce that goes into making biologics, including biosimilars, and a BA generally will not want to disclose that proprietary information to its competitor. Now turning to the l3A and l3B list, the BPCIA gives both the RPS and BA discretion over what patents to include on those lists, because the BPCIA leaves that up to the RPS and the BA to determine which of the RPS's patents, quote, "could reasonably be asserted" against the biosimilar. So this is another area of variation in the process. Note that if the RPS does not identify all of its potentially relevant patents in its l3 list to the BA, the BA will have to rely upon its own independent due diligence on the RPS's patent estate to find and include on its l3B list any additional patents that it believes should be included in phase two litigation. Now there's been a recent change to the BPCIA, which is that in June, 2021, the BPCIA was updated to require the publication of any l3A list that the RPS has served on the BA within the FDA's publication called the "Purple Book," and the RPS has to provide those l3A lists to the FDA within 30 days after the list was provided to the BA. So this new provision should lessen the burden on the BA to have to conduct independent due diligence on the RPS's patent estate and figure out what sort of patents were previously asserted and potentially cover the reference product and the biosimilar product. But note that this will only be a benefit to BAs that sue after the first couple of BAs have been sued and l3A lists have been exchanged. So it's something that's gonna benefit later biosimilar applicants for a particular reference product, but not the early ones. One aspect of these June, 2021 amendments to the BPCIA that I thought would be important to note is that if you're the BA, then the publication of certain patents on an l3A list in the Purple Book may give the public general notice of, for example, the manufacturing process that you might be using for your particular biosimilar. Recall that biosimilars are similar to each other, they're not identical. So it's not always true that the RPS will assert the same group of patents against every single BA. So those differences could potentially, again, provide the public with some clue as to what processes you might be using. And again, because this is secret sauce, you might be potentially concerned about the public disclosure of that sort of information. This hasn't risen or shown up in litigation yet, but it's something that potentially might be a concern. Now, generally speaking, an RPS has an incentive to be as inclusive as possible in terms of including patents on its l3A list. And that's because if the RPS doesn't include a patent on its l3A list, then the RPS may later be precluded from asserting that patent, pursuant to section 35 U.S.C. 271e6C of the Patent Act. However, if an RPS includes a patent on its l3A list, that doesn't automatically mean that there's an actionable case or controversy over every patent on that list. As I've noted on this slide, in a 2020 BPCIA case involving Genentech and Amgen, the district court for the District of Delaware dismissed, for lack of subject matter jurisdiction, the BA's counterclaims against a patent that the RPS initially included on its l3A list, but on which the RPS ultimately did not assert in phase one litigation against the BA. Now one other thing to note about l3A lists is that the RPS is subject to an obligation under section l7 of the BPCIA to supplement its l3A list with any newly issued or newly exclusively licensed patents that it believes could reasonably be asserted against a biosimilar. Such supplementation may be made within 30 days after the issuance or licensing of any new patents. And if that supplementation is made, then the newly issued or licensed patents can be asserted in phase two litigation. So again, because this supplementation obligation is subject to that "could reasonably be asserted" language, there's some discretion here over what the RPS can choose to include in its l7 supplements. Now, another aspect of the patent dance that's subject to variation by the parties is the amount of detail that the parties need to provide in their l3B and l3C infringement and validity contentions. Now, the BPCIA says that the contentions must be detailed and provide on a claim by claim basis what those contentions are, but the BPCIA doesn't provide any further guidance. And in this connection, one concern that the BP, in this connection, one concern that the BA and RPS may both share is that the infringement and validity contentions exchanged during the patent dance potentially may be limiting or binding on their arguments once litigation begins. So both sides might be incentivized to provide less detail instead of more. The current case law, however, suggests that these contentions are not necessarily binding or limiting in litigation. So in the Genentech versus Amgen case that I just cited on the previous slide, the Delaware District Court there held that a BA's counterclaim asserted in phase one litigation are not limited to contentions that it disclosed to the RPS during the patent dance. However, I should note that the federal circuit has noted at least as dicta in another case, that contentions exchanged during the patent dance potentially may have probative value in phase one litigation. So not limiting based on current case law, but there is some intimations that they might potentially have an effect. Now the last aspect of the phase one process that can be susceptible to variation arises when the parties are unable to agree on what patents to include in phase one litigation. And under this circumstance, section l5 of the BPCIA gives the BA a significant amount of power to shape the overall scope of phase one litigation. First, if the parties can't agree on which patents to litigate, then the number of patents that the RPS can assert in phase one cannot be greater than the number of patents that the BA wishes to litigate in phase one. Second, if the BA refuses outright to list any or identify any patents for phase one, then the RPS will be limited to suing on just one patent in phase one, and that can cause some jam ups during phase two, as I'll discuss in a moment. Third, once the parties exchange their competing l5 lists of patents that they believe should be litigated in phase one, the statutory language says that the RPS has to sue on all the patents that are included on both lists. So as you can see, these steps give the BA significant power to narrow or potentially expand the scope of phase one litigation. Yeah, that was quite a bit and that's just phase one. Let's now zoom in on phase two, which is governed by section l8 of the BPCIA. As I mentioned, phase two is initiated when the BA provides the RPS with not later than 180 days notice of the first commercial marketing of its biosimilar. In other words, the BPCIA requires that the BA provides the RPS with at least 180 days advance notice of the commercial launch date of its biosimilar. Now upon receipt of that notice, the RPS may, at any time prior to commercial launch of the biosimilar, sue to enjoin commercial launch by asserting against the BA those patents that were previously identified on the party's l3A or l3B list in phase one, as well as any patents that were newly issued or exclusively licensed by the RPS and included on those l7 supplements. Now superficially, phase two looks much simpler than phase one, because there are only two steps to it and not much in the way of information exchanges, but it turns out that there's still a fair amount of complexity to phase two. And focusing on just step one in phase two, the statutory language in the BPCIA requires that the BA provide at least 180 days advance notice of the commercial launch of the biosimilar. But the statutory language is silent as to whether a BA can provide notice more than 180 days prior to launch. And as I'll explain in a moment, this isn't necessarily a good thing if you're the RPS, because the statutory silence allows BAs to significantly reshape the canonical BPCIA patent litigation process. And those reshapings can affect both phase two and phase one. Now, one question that's arisen in connection with step one in phase two is whether the BPCIA requires only one commercial marketing notice per aBLA or whether supplements to a previously filed aBLA can trigger an obligation by the BA to provide new commercial marketing notices, and to wait 180 days after each such notice to launch its biosimilar. You can imagine if that was a requirement, the BA would be waiting a while to actually commercially launch. Well, the Delaware District Court in 2019 answered that question in the negative, holding that a BA who has already provided a commercial marketing notice to the RPS need not provide additional notices for supplements to its aBLA, at least where those supplements are for the purpose of informing the FDA of new manufacturing sites and labeling changes for the biosimilar. Now turning into the second step of phase two, here we have what appears to be a transfer of power from the BA to the RPS. So the second step of phase two allows the RPS to sue on any patent that was previously included on an l3A or an l3B list, but that was left out of phase one litigation, perhaps because the BA limited phase one litigation just to one patent. The RPS can also sue on any newly issued or newly exclusively licensed patents that were left out of phase one. But as I mentioned earlier, to sue the BA on those newly issued or newly licensed patents in phase two, the RPS must satisfy its obligation under section l7 of the BPCIA to supplement its l3A list with any such newly issued or newly licensed patents. Moreover, because phase two is closer in time than phase one to the actual commercial launch date of the biosimilar and may involve a significantly larger number of patents than were involved in phase one, phase two can impose significant time and resource burdens on the RPS. For example, if the BA limits phase one litigation to just one of many l3A listed RPS patents, then the RPS will have no choice but to wait until phase two to bring suit on all those other patents. And as you can imagine, the RPS may have to undertake a significant amount of work to move for injunctive relief on all those other patents within the relatively short 180 day window prior to the BA's commercial launch date. So the take-home message here, at least for RPS's, is to prepare well in advance for any potentially burdensome phase two litigation. Get your ducks in a row so that you can take advantage of that 180 day period and ensure that any patents that were left out of phase one are ready to go in phase two. Now that we've zoomed into the steps of phase one and two, let's zoom back out and talk about how recent case law has reshaped the overall process. As I've summarized on this slide, this is the canonical BPCIA litigation process, and this is how it's supposed to proceed, basically, in an orderly, sequential fashion. In phase one, as I mentioned, the BA provides its aBLA and its manufacturing information to the RPS. The parties engage in the patent dance, which takes up to a year, as you can see here, and then the parties litigate the agreed upon list of patents, or if there's no agreement upon the parties' combined l5 lists. In phase two, the BA provides at least 180 days notice of its commercial launch date, thereby giving the RPS an opportunity to sue the BA in advance of commercial launch on any previously identified l3 patents and any newly issued or licensed patents that were left out of phase one litigation. Now, as you can see, even when everything goes according to plan, this canonical process can take months or years, and that's on purpose. The canonical BPCIA litigation process was designed to fit, as I mentioned earlier, within that eight year gap between the expiration of the RPS's four year data exclusivity and the RPS's 12 year marketing exclusivity. But much of the recent BPCIA case law has arisen from litigation over reference products for which the data and market exclusivity already expired. So in those cases, the only thing holding the BAs off the market was the RPS's patents. And it's those types of situations that have generated further incentives for BAs to try to modify or shorten the BPCIA's canonical litigation process, and for the RPSs to fight back and ensure that there's sufficient time to litigate their patent rights. But in any event, it's those incentives that led one BA, Sandoz in this case, to reshape the BPCIA litigation process in a case that went all the way to the Supreme Court. And this is the only one that went to the Supreme Court so far. Now in this case, Sandoz decided that it would not provide to the RPS, Amgen, Sandoz's aBLA or the manufacturing information for Sandoz's filgrastim biosimilar, which goes under the brand named Zarxio. In this case, Sandoz simply gave Amgen the commercial marketing notice at the time that it filed its aBLA, far in advance of the FDA approval and its projected commercial launch date. Faced with that situation, Amgen sued Sandoz to compel Sandoz, first, to provide its aBLA and manufacturing information in accordance with the patent dance, and second, to require that Sandoz wait until the actual FDA approval of its aBLA before providing its commercial marketing notice. So that suit made its way to the Supreme Court, which in 2017 ruled against Amgen on both grounds. The Supreme Court here held that there is no federal injunctive relief available to compel a BA to provide its aBLA or its manufacturing information to the RPS. And on remand, the federal circuit similarly held that there was no state injunctive relief available either to compel Sandoz to hand over its aBLA or manufacturing information. Now the Supreme Court ruled that the only remedy available to address a failure by a BA to provide its aBLA or its manufacturing information is the remedy that's expressly set forth at section l9C of the BPCIA. And section l9C provides that in the event of such failure by the BA, the RPS, but not the BA, may immediately bring action on any patent that claims the biological product or a use of the biological product. Note, however, that the language of section l9C is limited to product and method of use patents. It doesn't allow the RPS to bring immediate action on process patents. Additionally, and perhaps more importantly for purposes of this presentation, the Supreme Court in the 2017 Sandoz case held that the BA need not wait until it obtains FDA approval of its aBLA to give the RPS notice of commercial marketing. Instead, the Supreme Court held that under section l8 of the BPCIA, the BA can give the RPS its commercial marketing notice either before or after receiving FDA approval. And it's that ruling in particular that gives the BA the power to reshape the canonical BPCIA litigation process significantly simply by varying the timing of its commercial launch notice. Now, here we see the same scenario that was presented in the Sandoz case, in which the BA doesn't provide its aBLA or its manufacturing information, and instead immediately provides its notice of commercial marketing. In this scenario, the BA's actions enable it to bypass phase one entirely, including all steps of the patent dance. The parties then proceed directly to phase two, in which the RPS may seek a preliminary injunction to prevent the launch of the biosimilar. Now, there isn't much of a deterrent to the BA for attempting this type of scenario. The only remedy available to the RPS in this scenario under the Supreme Court's ruling is to sue the BA immediately on patents that claim the biosimilar or uses thereof. But if you're the BA and you figure that you're gonna get sued sooner or later on those patents in phase two, you may just decide that provoking and defending an earlier suit on those patents will require less effort than having to go through all the steps of the patent dance and the resulting phase one litigation. So in other words, just short circuit phase one and get onto phase two right away. And it's not surprising that another BA recently chose to do just that. Just a couple of weeks ago, Amgen, which is now the BA, gave early notice to an RPS, Biogen, of Amgen's intent to commercially market a biosimilar, a Biogen Stelara reference product, without first informing Biogen of when it filed its aBLA or whether it intended to participate in the patent dance. This scenario poses some procedural open questions, two of which I have highlighted on this slide. First, phase two litigation is limited by the BPCIA's statutory language to the patents identified on the party's l3A and l3B lists. But if phase one is bypassed and the parties never exchange l3A and l3B lists, what patents can be litigated in phase two? As I explained earlier, if the BA fails to provide its aBLA and manufacturing information to the RPS, then section l9, as I mentioned, kicks in and allows the RPS to sue on patents that cover the biosimilar product and its uses. But what about the RPS's process patents? Will they be left out of the scenario? Seems somewhat unfair that the substantive scope of the RPS's action will be limited in this way due to to the BA's non-compliance with the patent dance, but that's the outcome that this part of the statute seems to contemplate. Second, as I mentioned earlier, phase two is supposed to occur after phase one and closer in time to the actual commercial launch of the biosimilar. And it seems like a key purpose of the phase two litigation is to ensure that the RPS has an opportunity to obtain a preliminary injunction to prevent the imminent commercial launch of a biosimilar close in time to the actual launch. But obtaining a preliminary injunction typically requires proof of non-speculative, irreparable, and sometimes imminent harm. So if the BA provides its commercial marketing notice far in advance of any actual FDA approval or commercial launch, would that potentially impair the ability of the RPS to prove a non-speculative, irreparable, and potentially imminent harm? Again, these are some open questions. We don't have answers, but I expect that future litigation may come by and address them. Now here's a different scenario. What if the BPCIA, what if the BA provides less than 180 days advance notice of its intent to launch? And in this scenario, by providing late notice, the BA will compress the time period in which the RPS must file its phase two lawsuits, thereby increasing the time and resource crunch on the RPS in phase two. And recall that the BA can unilaterally limit phase one to just one patent, thereby forcing the RPS to wait until phase two to sue on all of the remaining patents. If that happens and the 180 day period for filing suit is also cut short, then the RPS will have to do a lot of work in a relatively short period of time to protect its interest. Now in this situation, can the RPS obtain specific enforcement of section l8 and force the commercial launch of the biosimilar until a full 180 days have elapsed from the time that notice of commercial launch was given? Well, this question actually reached the Federal Circuit Court of Appeals in 2016, and the federal circuit answered "Yes." According to the federal circuit's 2016 Amgen versus Apotex decision, a BA must provide the RPS with 180 days advance notice of commercial marketing under Section l8. And in contrast to the Supreme Court in the Sandoz case, the federal circuit here held that the remedies for the BA's violation of section l8 aren't limited to the remedies that are expressly set forth in the BPCIA itself. Now, the Supreme Court's denied certiorari of this Amgen versus Apotex case, and the Supreme Court's Sandoz decision doesn't address the range of remedies that are available to the RPS if the BA provides inadequate notice of commercial marketing under section l8. So the short of it is that Amgen versus Apotex is still good law, but one could consider whether or not the Supreme Court's decision and its reasoning that there's no federal relief available beyond that expressly provided in the BPCIA might potentially affect the future viability of Amgen versus Apotex. Now here's a hypothetical scenario that potentially could arise if a BA simply refuses to comply with the 180 day notice requirement of section l8, and no relief is available to compel the BA to do so. Now, in this situation, the BA simply launches this product at the time that it obtains FDA approval without providing any advance notice, thereby bypassing phase two entirely. And this situation is analogous to what we call an at-risk launch in the Hatch-Waxman world. Now, at present, this scenario is purely hypothetical. I don't wanna scare any RPS's. There's been no reported decision in which a BPCIA, in which a BA simply has launched its biosimilar without providing any advance notice under section l8 of the BPCIA. And in view of the fact that the Federal Circuit's 2016 Apotex decision is still good law, I don't know if this scenario is actually going to transpire in the real world, but I provide this for illustrative purposes. And in this situation, now that an infringing product is on the market, the parties effectively have left the BPCIA pre-launch litigation process behind and have entered into a phase of standard patent litigation. So accordingly, the RPS is now free to seek monetary damages, but the RPS is also put in the difficult situation of having to seek emergency injunctive relief in the form of trying to obtain a temporary restraining order and a preliminary injunction to stop infringing sales of the biosimilar. So this is a situation in which the RPS basically has to try to put the genie back in the bottle. Now, is there a way that the RPS can potentially get in front of such a scenario? Well, this case offers a potential solution. So this is the Federal Circuit's 2016 Amgen versus Hospira decision. And just as background to understanding this case, section 271e1 of the Patent Act provides a legal safe harbor that shields a BA from patent infringement liabilities for activities that are reasonably related to seeking FDA approval. So clinical development activities that have to be submitted by the BA to the FDA would be an example of things that are shielded by the safe harbor. By contrast, commercial activity that the BA undertakes in preparation for commercial launch aren't shielded from liability under the 271e1 safe harbor. And in this case, the BA, Hospira, had manufactured and stockpiled several lots of its erythropoietin active ingredient in advance of the commercial launch of its erythropoietin biosimilar. So Amgen sued Hospira for infringement of one of its process patents, and then in the resulting litigation, which, by the way, is not BPCIA litigation, this is just standard patent litigation, Hospira argued that all of its erythropoietin lots fell within the safe harbor because according to Hospira, all of the lots had been tested for reasons that it alleged reasonably related to FDA approval, such as biosimilarity, stability and/or continued process verification testing. The jury in that case, however, rejected that argument and found that 14 of the 21 lots that Hospira had manufactured fell outside of the safe harbor. And that was primarily because Hospira itself had characterized those lots as commercial inventory. And so in that case, the jury awarded Amgen $70 million in reasonable royalty damages, and on appeal the federal circuit affirmed the jury's verdict and its damages award. So the Hospira fact pattern prevents a potential option, again, outside of the BPCIA litigation process, for an RPS to assert its patents in advance of a commercial launch by the BA. Just following on from the facts of Hospira, if the RPS discovers, either as part of phase one litigation in the BPCIA process or otherwise, that the BA is stockpiling an infringing product or a component thereof in anticipation of commercial launch, the RPS potentially could sue the BA ahead of commercial launch and obtain monetary damages that flow from the BA stockpiling or other pre-launch commercial activities, much as Amgen did in the case I was talking about. So in this instance, the 271e1 safe harbor won't protect the BA, and moreover, the fact that the BA is amassing commercial quantities of the components it needs for its biosimilar may also give the RPS with evidence it may need to demonstrate that the commercial launch of the BA's product is imminent and that immediate injunctive relief is also justified. So this might be a way to fight back against a potential at-launch risk scenario in the biosimilar/reference biologics world. Now as I mentioned earlier, I'm gonna switch gears to talk about something that came up earlier. The canonical BPCIA litigation process can take months or years to complete, but the median time for FDA approval of a biosimilar aBLA application hovers around 12 to 18 months, depending on who you ask. So this means that the FDA approval of an aBLA can occur while the parties are still doing the phase one patent dance. And if there's no marketing exclusivity left to keep the biosimilar off the market, then that particular fact pattern can give rise to the scenario that we see here. And this is the classic race to the courthouse scenario that we see in a lot of standard non-BPCIA patent case, only, here it's happening in the middle of the BPCIA litigation process. So in this scenario, the FDA approves the biosimilar while the parties are still engaged in good faith negotiations over what patents to litigate in phase one. And once it has its FDA approval in hand, the BA, of course, without any sort of regulatory exclusivity holding it back, would want to go through the BPCIA litigation process as quickly as possible. So in this scenario, what it does is it provides its commercial marketing notice on the very same day that it gets FDA approval, and then races to the courthouse in its favored district, here we call it district one, seeking a declaratory judgment on l3 listed patents. And that rapid action by the BA then spurs the RPS in turn to respond by filing a mirror declaratory judgment on the same patents in its favored district, which here we're calling district two. So the result is that we now have two substantively identical phase two declaratory judgment suits that are pending in two different districts. Maybe they were filed, you know, within hours of each other. So in this scenario, whose choice of venue prevails, the RPS's or the BA's? And it turns out, this scenario that I've just presented isn't hypothetical, it was actually litigated twice in 2018. So in this 2018 decision from the central district of California, the district court held that the RPS's choice of venue prevails. Here, the California court, exercising its statutory discretion not to hear declaratory judgment actions, dismissed the BA's first filed California phase two suit in favor of the RPS's second filed Delaware suit. The California court based its decision on the fact that the BPCIA intended the RPS to have the opportunity to sue first on its phase one patents before the parties proceeded to phase two, even though the Supreme Court's Sandoz decision technically allowed the BA to provide its commercial marketing notice at the time it obtained its FDA approval in the middle of phase one. Moreover, the California court dismissed the BA's California suit in favor of the RPS's Delaware suit, even though the cases weren't entirely identical. As the court noted, a few more patents were at issue in California, but the court nevertheless reasoned that the cases were significantly duplicative of one another. So this case comes from the Central District of California. This second case that I'm talking about, Celltrion versus Genentech, comes from the central, comes from the Northern District of California. And in this case, the court held that the RPS's choice of venue in Delaware prevailed. And in this case, a relatively easier fact pattern was presented for the court to resolve, because unlike the BA in the previous case, the BA in this case left the phase one negotiating table as soon as it received its phase one approval, and never came back to actually complete the patent dance pursuant to section l5. So presented with this fact pattern, the Northern District of California simply applied the statutory remedy prescribed by the BPCIA for this situation, so no discretionary denial of a declaratory judgment action necessary. The court basically followed the statute, and the statute says that the BA is deprived of its ability to file a phase two action for declaratory judgment if it doesn't comply with the steps the patent dance. So, a relatively easy outcome here. And I think that's a nice place for us to wrap up in terms of our discussion of the recent case law, because these two California cases nicely encapsulate the current state of affairs concerning BPCIA litigation and my key takeaways for this presentation. Now as we've seen, the BPCIA litigation process is very complicated, and those complications create opportunities and incentives on both sides to vary the process. For example, we've seen that the canonical phase one and phase two process may not leave sufficient time for the parties to resolve all of their patent disputes before the FDA approves the biosimilar at issue. And again, that's critical, especially in instances where the regulatory exclusivities, that marketing exclusivity that I was talking about earlier, have already expired for the reference product. So in those instances, there are particular incentives for the BA to shorten and bypass steps in the process. Now, we've also seen how the statutory language in the BPCIA and recent case law have given the BA significant control over the timing of phases one and two, as well as their scope. But as illustrated by the two California cases I just mentioned, courts do recognize, and I think they're increasingly starting to recognize, that the canonical phase one and phase two process were intended to follow an orderly sequence, and to give the RPS a full and fair opportunity to litigate the patents it wishes to assert. So that was setting all of the new dance moves that the recent case law, are limits on how much the canonical BPCIA litigation process can be bent to accommodate those competing incentives of the RPS and the BA. Now, in the meantime, I expect the parties will continue to test those limits and innovate and introduce new variations to the BPCIA litigation process. So I'll end with the caution that the new dance moves I've covered in this presentation aren't exhausted, aren't exhaustive, and I'm sure that clever and creative BAs and RPS's will continue to explore new ones. That's my presentation. Thanks for watching.