Another update on $LQDA
Another update on $LQDA
It has once again been a minute since I’ve written, mostly due to the fact that I’m not that excited about most of the litigation/merger arb opportunities out there. A lot of folks asked me about MSFT/ATVI, which I stayed away from due to the international elements. Sometimes, it’s important to acknowledge what you don’t know, and I don’t know a damned thing about the inner machinations of the UK’s Competition and Markets Authority.
So that leaves the existing plays. For everyone’s reference, I still hold the same position, which has been a little painful this past month, post-earnings. I also hold , and considered adding when it dropped below $12 earlier in the month, though it was still above my cost basis and it’s already a pretty full position, so ultimately decided against it. I added to my position at various levels on the climb up to $12, and plan to hang on to that one, at least until we have some clarity on the relevant “control” date which dictates damages (the best discussion of the range of possible damages is from Andrew Walker’s YAVP podcast with Artem Fokin which you can find here starting at about 21:30). Jefferies just put out a $22.50 PT on the stock, which I think is a feasible valuation following damages and collection clarity.
It also leaves . Last month (May 3rd) we had oral arguments at the Federal Circuit for the appeal stemming from the District Court litigation. I barely touched on the District Court litigation in my prior writings because I had largely considered the appeal to be a write-off, at least in light of the PTAB appeal. For various reasons we’ll get into below, I put Liquidia’s odds of success in the District Court and PTAB appeals at about 50% and 90%, respectively (up from about 5% and 90% just a month ago). You may recall the landscape of the appeals looks something like this:
Today we’re going to focus on the left side of that chart—the District Court appeal. And buckle up because this is probably going to be the longest and driest post yet. There’s simply no way to do this in a fun way. Also, it’s not that the PTAB appeal isn’t important—indeed, it’s statistically more likely to go Liquidia’s way. We just don’t have any updates worth discussing in detail yet. The only real update at this point is that this past week Liquidia filed to expedite the briefing for the PTAB appeal, proposing the below schedule. We should know in the next week or so if the Federal Circuit will adhere to Liquidia’s proposed schedule, the original timeline, or some other abridged timeline of their own choosing.
Liquidia’s proposed timeline for the PTAB briefing (half the normal time allocated to the parties):
But back to the District Court. Why the huge increase in probability that the District Court appeal goes in Liquidia’s favor? 5% to 50% is no small jump.
The answer is that the oral arguments on May 3rd went pretty darn well for Liquidia. Really, you should listen to it all. It included Judge Lourie (one of three judges on the Federal Circuit panel hearing the case) calling UTHR’s arguments around the ’066 patent “absurd.”
Remember, this appeal is centered around both the ’066 patent, which Judge Andrews at the District of Del. found not infringed (with respect to claim 8) and invalid (with respect to claims 1, 2, 3, 6, and 9), and the ’793 patent, which Judge Andrews found valid and infringed, but which is also the subject of the PTAB appeal which found the patent invalid on anticipation (35 U.S.C. §102) grounds. It’s a headache, I know, but the chart above should serve as a guidepost.
Due to the Fed. Cir. Panel’s extreme incredulity of UTHR’s ’066 arguments, I’m not going to spend much time addressing them, although they are important (Liquidia losing the ’066 appeal would defeat their near-term commercialization plans just as the ’793 would). The relevant ’066 patent claims were found invalid in light of a 2004 paper by Robert Moriarty et. al which can be found here. The gist of UTHR’s argument is that Moriarty’s paper does not disclose treprostinil with the right impurities—and right level of such impurities—and therefore the compound claims are not invalid. However, the Fed. Cir. quickly shot down this argument, noting that the ’066 claims are “not a method of reducing impurities,” and that, given the ’066 claims are pharmaceutical product-by-process claims which do not expressly state impurity levels, the disclosure in Moriarty sufficiently discloses the treprostinil compound in the ’066 patent.
Assuming we have safely discarded discussion of the ’066 patent, that leaves us with only the ’793 arguments and, again referencing the chart above, the need for LQDA to only win its ’793 appeal on one side (i.e., either the PTAB appeal or District Court appeal will do).
While the ’793 patent is being appealed on both sides of the chart, the substantive grounds for each appeal are entirely different.
You may recall from my prior posts and appearance on the YAVP podcast that Liquidia’s arguments around prior art invalidity of the ’793 patent never came up in the District Court trial. At the PTAB, Liquidia argued anticipation (35 U.S.C. §102) and non-obviousness (35 U.S.C. §103). This is of course no surprise. 102 and 103 grounds are the only invalidity grounds available to Petitioners at the PTAB. But these arguments didn’t show up at the District Court at all. Instead, Liquidia only argued invalidity based on lack of enablement and lack of sufficient written description (35 U.S.C. §112, which we’ll get into momentarily), despite all arguments—including §102 and §103 grounds—being available to it there. Remember Voswinckel JESC and Voswinckel JAHA? When it comes to the District Court appeal, those abstracts are irrelevant.
Liquidia dropped the arguments to which those references relate—the arguments based on 35 U.S.C. §102 and 35 U.S.C. §103—on the second day of trial at the District Court for strategic reasons, likely so as not to create a possible split decision between the PTAB and District Court on the same issues.
What we’re left with is a PTAB appeal that hinges on an abstract written by the ’793 inventor being counted as a “prior art reference,” and a District Court appeal that hinges on claim interpretation and the wording of the ’793 patent itself.
PATENT LAW EDUCATION PT. 1: OVERVIEW OF THE PATENT STATUTES
Let’s do a little exploration of the relevant patent statute before we fully dive in. I’ve probably touched on most of this in some form or another in prior posts, but it’s worth a refresher. Title 35 of the U.S. Code sets forth patent stuff, some of which I mentioned above. The individual sections are grouped according to subject matter. Sections 101-105 relate to deciding what should and should not be patentable. Sections 111-123 describe things you have to do in your application to get the patent. Sections 106-110 don’t exist :(
By way of example:
Section 101 of Title 35 says you can’t patent things like gravity or other laws of nature, nor can you patent things like the human genome itself (your mom might think you’re special, but your DNA isn’t patentable).
Section 102 says your invention must be new (you might have genuinely come up with the idea for a spork on your own after crushing a can of chili, but sadly it’s been done before whether you knew of the spork or not).
Section 103 says it must be “non-obvious” (attaching a piece of rope to a door handle is unlikely to get you a patent on a new type of doorknob; we already know ropes pull things . . . although it would probably have saved me a lot of ‘pushing’ on a ‘pull’ embarrassment).
Section 112, is a bit convoluted, and states:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
There are actually 3 separate requirements in that Section 112 paragraph:
the invention must be adequately described,
a person of ordinary skill in the art (“POSA” or “PHOSITA” or “POSITA” or whatever other permutation I’m missing) should be able to recreate the invention without “undue experimentation,” and
where the invention has multiple embodiments, the inventor must disclose the best mode of the invention. #3 is now a “paper tiger” and not really a requirement, but I digress…
Sections 102 and 103 are, as noted above, critical to the PTAB appeal. Section 112 is critical to the District Court appeal.
PATENT LAW EDUCATION PT. 2: WRITTEN DESCRIPTION
In order to sufficiently describe your invention, the patent must, well… describe it. If I tell you I’ve invented a vehicle but leave out any detail on an engine/motor/etc. (i.e., I haven’t told you how it moves anywhere in my patent) then I clearly have not provided enough of a description. Similarly, if I merely state “it has a propulsion method” then I *probably* haven’t sufficiently described it there either. From a public policy standpoint, this makes sense. We don’t want to give someone a monopoly on making and using this “invention” if they can’t even tell us what the invention is. How would we then know if they have invented anything useful at all? If I try to claim “a method of treating cancer with single dose of aspirin,” but don’t tell you how that dose would actually affect cancerous cells, it doesn’t really matter if later we discover aspirin really IS helpful in treating some forms of cancer. The quid pro quo of patent law is that you have to tell the public what you’ve invented in order to be granted that monopoly for 20 years, and you have to do it in a way that demonstrates you know what the hell you’re talking about.
The test to determine whether or not written description is satisfied therefore is whether or not a POSA/PHOSITA/POSITA can reasonably conclude that the inventor was in possession of the claimed invention.
PATENT LAW EDUCATION PT. 3: ENABLEMENT
In order to sufficiently enable your invention, someone who has skill in the field of the invention must be able to read and appreciate how to actually employ the invention after reviewing your patent. From a public policy standpoint, this makes sense as well. If you tell me you’ve invented a cure for cancer, but you can’t tell me how you made the cure, then all you’ve really done is come up with a theoretical cure for cancer. Inventions must be real, and again, they must be disclosed. You don’t get to hide some of the most critical information needed to practice the invention and still get the patent. And in an age of AI and computer-assisted design, it’s entirely possible the cure for cancer will be first rendered by software. But until that cure is reduced to practice in a way that tells the world you actually know it will cure cancer, it isn’t sufficiently enabled. Importantly, the full scope of a patent claim must be enabled. In other words, if there are a substantial number of embodiments of the claimed invention which are not enabled by the patent, then the claim fails.
We can break down enablement into 8 factors, below, though again the critical question is whether or not a person of skill in the field of the patent would understand and appreciate how to practice the patent without spending inordinate amounts of time conducting their own experiments. The specific factors include:
the quantity of experimentation necessary,
the amount of direction or guidance presented,
the presence or absence of working examples,
the nature of the invention,
the state of the prior art,
the relative skill of those in the art,
the predictability or unpredictability of the art, and
the breadth of the claims.
PATENT LAW EDUCATION PT. 4: WHAT THE HELL IS THE DIFFERENCE BETWEEN WRITTEN DESCRIPTION AND ENABLEMENT?
It might sound superficial, but there is an important distinction. Written description asks: did you describe your invention in a way that shows you know what you’re talking about? Enablement asks: can a POSA make it? Since that probably just sounds like I repeated the last two headers, let’s look at some additional examples.
An invention may be described without being enabling. I could describe a chemical compound that has some inventive properties (like oh, I don’t know, a pharmaceutical compound) but if there is no apparent method of making that compound, I haven’t enabled it. Conversely, a disclosure could be enabling without being properly described (for example a specification describing a method of making and using a paint composition made of specific ingredients within broad ranges would be enabling since you know how to combine the ingredients, but if I don’t actually describe any specific formulation then I haven’t sufficiently described the invention).
If this still sounds confusing, don’t worry. I’ve seen patent lawyers confuse the two in court.
LIQUIDIA’S ARGUMENTS
So now that we now what grounds on which Liquidia is challenging the ’793 in the District Court litigation, we can generally sum up Liquidia’s ’793 position with:
UTHR failed to describe its ’793 patent sufficiently because the patent doesn’t actually describe the treatment of any WHO Group 2 patients with treprostinil, and
UTHR did not enable its ’793 patent because a doctor would have to conduct undue experimentation to know whether or not isolated Group 2 patients could be treated with treprostinil.
On its face, it seems like a reasonable (and potentially winning) argument. If UTHR had no idea whether or not treprostinil could be used in isolated Group 2 patients, it would stand to reason that they weren’t actually in possession of everything they claim. Similarly, if your patent does not mention that your method of treatment of a disease could kill half the people with that disease, then it would seem further study is required to know how useful that patent really is. In fact, doctors did conduct such a study, and the results were not great:
It turns out that vasodilators like treprostinil don’t work for isolated Group 2 pulmonary hypertension patients (i.e., those with ONLY Group 2 and not some combined form of PH where patients have Group 2 + another Group) because the cause of the disease is fundamentally different with Group 2 patients compared to the other varieties of pulmonary hypertension. In Groups 1 & 3-5, the disease conditions affect the pulmonary arteries or “precapillary vessels.” Conversely, in Group 2, high pressure in the pulmonary arteries is actually caused by defects in the heart, which in turn affects the postcapillary veins. The term “postcapillary” arises because the left heart is downstream with respect to bloodflow from the pulmonary capillaries. Because these two categories—pre and postcapillary PH—are caused by different issues, the treatments are unsurprisingly very different. Precapillary PH is treated with vasodilators, while postcapillary PH is treated with diuretics. Use of vasodilators in patients with precapillary PH can cause death by way of excess fluid in the lungs and excess blood flowing to the left side of the heart, also known as pulmonary edema.
Getting back to the “FIRST study,” it sort of seems axiomatic that if you patent a method for treatment of a disease, and then doctors do a big study on that treatment method only to find out that there is a massive risk of death for a subset of the patients requiring termination of said study, that “undue experimentation” was, in fact, required to get to the bottom of your patented method. After all, I don’t expect my doctor to conduct a rigorous double-blind multi-month study for every drug that they may prescribe me.
However, UTHR counters Liquidia with a few arguments. First, the drug used in the “FIRST study” was Flolan® (the brand name of the vasodilator epoprostenol). On the Flolan label was a warning: treatment should be stopped if signs of pulmonary edema occur. So UTHR credibly argues that the risk was already known to healthcare providers, and that such risk would be considered in the prescribing of a vasodilator to any patient with any form of pulmonary hypertension.
UTHR also argues that it is always the case that in pharmaceutical compound method-of-treatment claims, efficacy is taken into consideration when prescribing, and that risk tolerance is a job for the FDA, not the USPTO.
That’s generally true. The USPTO doesn’t try to play referee when it comes to the nuances of the medical field. Some treatments which may be used for a variety of indications may have varying efficacy for each indication and various efficacy for different subpopulations. It’s easy to imagine how cumbersome patent claims might become if a patentee had to list out efficacy and/or preferred treatments in every claim. Judge Lourie on the Federal Circuit panel posed this question to Liquidia counsel in the May 3rd hearing:
UTHR was also quick to point out this ostensibly heavy lift in its briefing:
You can see from the sentence immediately following the highlight that a single inoperable enablement is probably unlikely to defeat a patent claim. Enablement becomes an issue if the number of embodiments that don’t work becomes too high. But what happens if the number of inoperable embodiment(s) is low, but the number of people affected by the inoperable embodiment(s) and aggregate effect is really high? What if only one of the possible patient pools that would fall under the scope of the patent is a really big patient pool? Like, really, really big.
It turns out isolated Group 2 PH patients comprise about 50% of the PH patient population.
Recall from our intro to enablement above that the full scope of a patent claim must be enabled. If a material amount of embodiments of the claim are not enabled or render the patent claim generally “inoperable,” such that finding the operative version is burdensome to a practitioner, the claim should fail. A single failed embodiment becomes more material where the claim only has a small set of embodiments. Atlas Powder Co. v. E.I. du Pont De Nemours & Co., 750 F.2d 1569 (Fed. Cir. 1984) (the Court found that it “has not been shown to be [a] case” where “the number of inoperative combinations becomes significant.”) Here, UTHR arguably lacks enablement for one of only two embodiments (pre vs. postcapillary disease), or at the very least one of five (Group 2 vs. Groups 1 & 3-5). Liquidia’s counsel answered Judge Dyk’s question on the burden of “multifurcation” of embodiments by highlighting the size of the inoperable patient population:
Looking at the claims themselves, we see the sole independent claim (Claim 1) begins with:
“A method of treating pulmonary hypertension comprising administering by inhalation to a human suffering from pulmonary hypertension. . .”
UTHR’s arguments that patent claims can get unwieldy, while true in theory, seems a little disingenuous here. It would not have been that hard to carve out Group 2 from the claims. The patentee easily could have claimed “a method of treating Groups 1, 3-5 Pulmonary Hypertension . . . ” or “a method of treating precapillary pulmonary hypertension . . .”
But they didn’t.
Instead, they sort of skirted the issue, made further evident based on the embodiments that the UTHR patentee did and did not describe. Patent claims usually contain several named embodiments. That is to say, patents usually describe several ways of how to actually practice the invention (e.g., you might patent a drug compound that can have several different inert molecular chains attached to it, so you may describe a few of these as exemplars, though you may not choose to describe every single embodiment due to efficiency or impracticality). The ’793 patent has a number of embodiments listed, but not a single one describes treatment of an isolated Group 2 patient. So while the claims describe treatment of “pulmonary hypertension” broadly, the embodiments conveniently leave out isolated Group 2 patients.
In its appellate briefing, UTHR oversimplifies Liquidia’s arguments, saying that Liquidia is arguing the claimed treatment must be the “preferred” method, and a “mainstay” of treatment.
But this isn’t really an issue of “preference” or "the treatment being a “mainstay.” We now know that no isolated Group 2 patient should ever be treated with vasodilators since it doesn’t work on the source of the disease and creates an unreasonably high risk of death. Liquidia says that even a POSITA would not have appreciated that this method should not be used for Group 2 based on the patent because treprostinil had never been inhaled before. At the time of the invention, a healthcare provider might have assumed the novelty of inhalation would have caused treprostinil to work differently on isolation Group 2 patients, and therefore leaving this information out of the patent causes enablement to fail.
And though it isn’t strictly impossible to prescribe a vasodilator to an isolated Group 2 patient, the issue isn’t what is possible, the issue is whether or not undue experimentation is required, as discussed above. Trustees of Boston University v. Everlight Electronics Co., 896 F.3d 1357 (Fed. Cir. 2018) (“[t]he inquiry is not whether it was, or is, possible to make the full scope of the claimed device … the inquiry is whether the patent’s specification taught one of skill in the art how to make such a device without undue experimentation[.]”)
LEXICOGRAPHY AND OTHER LAWYERLY WORDS
Another key tenet of patent law is that the patentee is his own lexicographer. Which is a fancy way of saying you get to define your own terms. If I wanted to patent say, “a really large lever to move the earth,” I could elsewhere in my patent application say “wherein ‘move’ means to move emotionally, not physically.” I’m not saying this would get you a patent, I’m just saying patent interpretation defaults to the patentee’s own definitions.
But not all terms are clearly defined in every patent. After all, defining every single term of art would make already-lengthy patents brutally long and impossible to read. You do however want to define the really important terms. Because if you don’t, the Court may later interpret your patent differently than you had intended via a Claim Construction hearing (aka a “Markman hearing”). These hearings usually occur after the initial stages of written discovery and before depositions, and are sometimes day-long (or even multi-day) affairs where attorneys argue over technical terms by quoting decade-old IEEE dictionaries and whatnot—truly riveting affairs. But what they lack in excitement they make up for in importance. Markman hearings are one of the most critical stages of a patent litigation, and often result in settlements once the Court (note: this is a legal determination, so even where plaintiff has elected a jury trial, no jury is involved) decides what the claims of a patent mean for purposes of the litigation.
Here’s an example of a very brief claim construction order from the Liquidia litigation. You can see that sometimes a judge will simply say “plain and ordinary meaning” (aka “we don’t need a special definition for that”). Other times they swap out words or impose more narrow terms based on things like industry standards and technical definitions.
To make matters even more complicated, if you’re looking for a separate claim construction order for the ’793 patent, you won’t find it. That’s because Judge Andrews heard proposals for the ’793 claim construction during trial, and ruled on the ’793 claim construction in his post-trial opinion. Remember, this was a bench trial—Judge Andrews was the fact finder, law adjudicator, and ruler of the Klingons. If this had been a jury trial, we would have needed the claim construction for the jury prior to the jury hearing the facts at issue. But the ’793 was added to the case via an amended complaint due to the fact that the ’793 patent was issued to UTHR after it had filed its original case against Liquidia. The end result was a bit of a scheduling mess—but nothing improper—and a claim construction ruling that made an end-run to the finish line.
Why are we talking about all of this? Well, Liquidia got a really shit claim interpretation out of Judge Andrews on the '793. I’m definitely not saying that Judge Andrews did a shit job. Judge Andrews is a very smart man. Judge Andrews is much, much smarter than I am. I’m just saying that the construction was shit for Liquidia. You see, Judge Andrews looked at the patent claims and acknowledged that the claimed method had to be “therapeutically effective.”
Seems straightforward enough.
But what does that actually mean? Does it mean it will cure someone? Relieve their symptoms? Improve certain patient vitals, regardless of improvement in patient experience? Do psychosomatic improvements count? You can start to go down a rabbit hole fairly quickly if you’re really persnickety.
So Judge Andrews did what any judge should do in that situation. He looked first within the four corners of the patent, at the patent specification itself, hoping to find an answer. And an answer he did find.
He noted that the specification was describing measurements of certain “hemodynamic” indicators like PVR (pulmonary vascular resistance), PAP (mean pulmonary arterial pressure) SAP (mean systemic arterial pressure), and SVR (systemic vascular resistance). And the patent describes several studies for groups of pulmonary hypertension patients (specifically WHO Groups 1, 3, and 4—the “group 1,” “group 2",” and “group 3” in the patent are merely enumerative and not indicative of WHO Groups) wherein these hemodynamic metrics—particularly PAP and PVR—improve.
So Judge Andrews concluded that “therapeutically effective” must mean “improvement in a patient's hemodynamics (reduced PAP or PVR).” If this is the claim construction, UTHR argues, then it does not matter that a healthcare provider would not prescribe a vasodilator to a postcapillary (isolated Group 2) patient. The point becomes whether or not the claimed method would have the effect of improving hemodynamics.
It probably does. There is some dispute as to whether the hemodynamics of an isolated Group 2 patient—a patient who should have “normal” hemodynamics—would change at all; but on the whole it appears that treprostinil would have some effect on hemodynamics nonetheless.
Think of it another way: salt increases blood pressure. If someone had chronically low blood pressure leading to fainting spells, maybe they would be prescribed a constant uptake of salt (I have no idea if this is true, but it seems logical to me). The mere fact that you don’t have fainting spells, or perhaps have fainting spells caused by something else would not mean that salt wouldn’t have the same “therapeutic effect” of raising your blood pressure. It might not be a therapeutic effect you need, but under this interpretation, the logic holds.
I generally find this to be a convincing argument. It’s even more convincing when you realize that the claim construction for “therapeutically effective” was not actually appealed. On appeal, claim construction—being a matter of law for the judge to decide—is reviewed de novo (meaning the Federal Circuit takes a fresh look at construction); however the underlying factual findings are reviewed for clear error.
The distinction can be highlighted between Judge Andrews’ discussion of the patent’s cited study on hemodynamics vs. the expert witnesses’ testimony on propensity to prescribe various therapies and the considerations that would go into each. Had the construction been appealed, the evaluation of the patent specification should be reviewed de novo, while the factual findings that providers would appreciate pulmonary edema risk for Group 2 patients would be evaluated under the clear error standard.
It would have seemed preferrable to attack the claim construction. De novo is a better standard of review for Liquidia since they would get a fresh bite at the apple, whereas “clear error” binds them to the prior ruling and imposes a fairly high hurdle. But instead of actually appealing the claim construction itself, Liquidia takes a different approach. Liquidia cleverly tries to create a construction argument by attacking the reliance on the FIRST study and witness testimony, essentially saying that it is impossible to interpret the “therapeutic[] effect[]” in the patent to mean effect on hemodynamics since no rational provider would use treprostinil on a postcapillary patient.
It’s certainly clever, though I’m still not 100% certain as to why Liquidia did not appeal the claim construction ruling. On one hand, I get it. Liquidia doesn’t want to betray their argument that there is no claim interpretation which can render the patent enabled and described due to safety concerns and the complete lack of Group 2 patient population references in the patent. Further, the intrinsic evidence in the patent is fairly clear, discussing on hemodynamic markers as the measure of therapeutic benefit. Importing safety terms by way of claim construction rather than enablement alone counts on extrinsic, weaker evidence. But on the other hand, the claim construction was critical to Judge Andrews’ decision, and a claim construction which requires safety and efficacy would almost certainly lead to defeat of the ’793 claims on both enablement and written description grounds. It’s a tough call.
Still, the Federal Circuit allowed Liquidia to get a foothold, focusing with intense scrutiny on the substantial amount of inoperability as to real-world treatment of patients with pulmonary hypertension, as we discussed above. The commentary by the Federal Circuit panel on this UTHR argument was slightly kinder than the commentary regarding the '066 arguments, but Judge Dyk still called UTHR’s ’793 inoperability rebuttal “kind of odd.”
Liquidia seems to have gotten away with conflating the testimony from experts on both sides admitting that treprostinil would have no beneficial effect on the disease with the claim construction requirement that the administered treprostinil have some effect on hemodynamics.
It does pose an interesting question as to whether an embodiment is inherently inoperable if professional norms or custom dictate it should (almost?) never occur, but is still 100% possible and will have the claimed effect under the claim construction. Sure, professional norms could change, or it’s easy to imagine how a pharmaceutical compound may only be not prescribed due to certain side effects which are later mitigated by a joint therapy—in that situation a patent certainly should not be invalidated and the practical scope of that embodiment is better left to the FDA. But where the inoperability in practice is expressly due to the operability of the effects under the claim construction . . . I genuinely don’t know. Maybe I’m just trying to shoehorn in “[safely and effectively] treating pulmonary hypertension” in the same way Judge Andrews accused Liquidia of trying to inject the limitations. Under Biogen Int 'l GMBH v. Mylan Pharms. Inc., I suppose I shouldn’t be trying to find specific efficacy where none was claimed. But if another patent lawyer wants to show me a case I’ve missed, or at least point me in the right direction, I welcome your input.
I suspect the Judges Dyk, Lourie, and Stoll, upon returning to their chambers, will appreciate the claim construction that they have been dealt and adjudicate accordingly, but they certainly gave Liquidia’s arguments more credit than I originally thought due. It didn’t help that UTHR’s counsel was simply incapable of steering the conversation back to the unappealed claim construction.
OTHER THOUGHTS—PATENT MODIFICATION
A last remaining question you might have is: let’s say Liquidia is right, and that the ’793 patent is not enabled or lacks sufficient written description. Is there risk to Liquidia that the patent is simply abridged in favor of UTHR? Can the Federal Circuit chop up the patent and give UTHR the benefit of the doubt on the application of the claims to WHO Groups 1 & 3-5?
The answer there is a resounding “no.” We know from a prior Federal Circuit case that “Courts do not rewrite the claims to narrow them for the patentee to cover only the valid portion.” Alcon Rsch., Ltd. v. Apotex, Inc., 687 F.3d 1362, 1368 (Fed. Cir. 2012). If the patent is not enabled or sufficiently described as to Group 2, Liquidia will win.
SO IN CONCLUSION . . .
If I were writing the opinion, I’d probably rule in favor of UTHR on the ’793 and in favor of Liquidia on the ’066 (i.e., I’d just affirm the District Court ruling), but it’s an extremely close call, particularly in light of the Federal Circuit’s incredulity of UTHR’s position on all fronts. I’m just not sure that it’s “clear error” for Judge Andrews to have relied on the (albeit single, failed) FIRST study where the acute risk was one which was present on the Flolan® label. And though Flolan® was injected, not inhaled, it’s probably not clear error to think that a practitioner would have appreciated that both Flolan® and inhaled treprostinil are both vasodilators. In the end, it really comes down to the claim construction for me. If it isn’t clear error to rely on the FIRST study, and if it isn’t clear error to think that a practitioner would have appreciated pulmonary edema risk, then it’s reasonable to conclude that an isolated Group 2 patient could be treated with inhaled treprostinil for purposes of improving hemodynamics, and that a practitioner would at least proceed with caution without undue experimentation. It’s also easy to imagine how the Federal Circuit can uphold the Delaware ruling and punt the issue to the FDA. I can’t stress enough how close of a call this is though following the lines of questioning during the Federal Circuit hearing—UTHR has the claim construction, but Liquidia has common sense on its side.
I realize my 50% success prediction is a patently (hah!) unhelpful conclusion for most. Remember though, the original thesis had nothing to do with this appeal. I’ve both upped my equity and bought some July puts, assuming that if Liquidia doesn’t win outright here that it will erase the ~30% monthly gain since the hearing and trade back down to the $6.40-7.20ish level, where it will probably bounce around as it did prior to the hearing while we wait for an outcome from the PTAB appeal. And please don’t mistake put buying for a lack of faith in Liquidia; I still believe they will be commercializing by mid-2024. I’m just trading around these interstitial events based on the information we’ve got and trying to manage risk. The 30% run in May was a monster and you have to reconsider how to manage things given the still-substantial probability we don’t see a Liquidia victory out of the District Court. Anyway, if you’ve gotten nothing else out of this post, at least appreciate that:
Claim construction matters, and
Anyone who tells you they know the outcomes of a litigation immediately after a complaint has been filed is full of shit*
*unless that case is Twitter v. Musk—that was clear as day from the start.
HAVE A GREAT WEEKEND