Obviousness of the New Mixture

Apotex v. Wyeth[1] is a relatively simple and straightforward nonobviousness decision in the pharma space (albeit non-precedential).  The Wyeth patent at issue here claims the compound that includes tigecycline mixed together with an acid as well as lactose.[2]  Although not claimed, the specification indicates that the mixture is beneficial because it helps tigecycline avoid degradation.  At the time of the invention (2005) tigecycline was a known tetracycline known to act as a broad spectrum antibiotic that degrades rapidly at a neutral PH.

The key prior art in the obviousness case was a Chinese patent publication that discloses minocycline mixed with both lactose and an acid in order to avoid degradation.[3]  Minocycline is a closely related tetracycline that is both structurally similar and operates by similar action.  (Tigecycline is a derivative of minocycline).

In both the Inter Partes Review proceedings and the appeal, the courts found insufficient motivation to combine the cited references by substituting the tigecycline into the minocycline compound.  Importantly here, the courts made this determination as a factual finding that “a skilled artisan would [not] have had a reason to combine the references.”[4]

Apotex (the challenger) argued that the similarity between the two forms of tetracycline and the fact that tigecycline was “known to work where other antibiotics have failed” both pushed toward a finding of a motivation to combine.  The courts however rejected this argument – finding that the structural similarities did not create “a potential motivating factor for a skilled artisan to substitute tigecycline for minocycline.”

[W]hile tigecycline is closely related to minocycline structurally and in terms of benefit, the Board did not err in concluding that there was insufficient basis in the record to show that it would have been obvious to a skilled artisan to substitute tigecycline in the prior art minocycline composition.

Although not mentioned by the courts, I would think it would also be important motivation to consider that – at the time of the invention here – tigecycline was not just some random minocycline derivative – rather tigecycline was being considered as an important last-resort MRSA treatment and under consideration for fast-track FDA approval.

= = = = =

A first thought that comes to my head in this case is whether the outcome here stems from the fact that this is a pharmaceutical case rather than one focusing on electronics or mechanical engineering.  Perhaps not, the court does explain here the lack of assurances (pre-invention) that the claimed mixture would substantially reduce degredation.  That factual-setup can be distinguished from a case like KSR where a good engineer could give pre-assurances that they could get the combination to work.

= = = = =

One aspect of the decision involves underlying distinctions between the ultimate question of obviousness (an issue of law) and the factual underpinnings (such as  motivation to combine references).  Here, of course, once the facts were determined, the question of law was instantly answered.

= = = = =

Notes:

[1] Apotex Inc. v. Wyeth LLC, App. No. 15-1871 (Fed. Cir. August 16, 2016).

[2] U.S. Patent No. 7,879,828, claim 1 taken as typical.

[3] Chinese Patent Pub. No. 139055A.

[4] See In re Hyon, 679 F.3d 1363 (Fed. Cir. 2012).

52 thoughts on “Obviousness of the New Mixture

  1. 8

    Most of these comments characterize the decision as acknowledging that tigecycline and minocycline are similar, but somehow not recognizing that a composition that stabilizes minocycline would also stabilize tigecycline. However, a closer reading shows this is not how the decision reads. Look at what the decision actually says about the compounds.

    Minocycline degrades by oxidation at pH above 5. Therefore, it is compounded with a strong acid to prevent oxidation. At lower pH, the primary degradation pathway in solution is by epimerization, which can be slowed by adding solubilizers (note – lactose is not listed as a solubilizer). However, the dry salt of minocycline on a supporting agent is stable. The stability of minocycline hydrochloride does not depend on the presence of lactose; it is simply one of many inactive supporting agents that can be used to prepare stable formulations.

    What Wyeth found is that tigecycline degrades much faster through epimerization than minocycline and other tetracyclines. A person having ordinary skill in the art would not have used tigecycline at all. The acid is needed to prevent oxidative degradation, but in the presence of acid, tigecycline rapidly undergoes epimeric degradation. You would expect that the tigecycline-acid combination would be unstable regardless of the supporting material. But although certain solubilizers are known to reduce the epimeric degradation of tetracyclines, lactose is not. The surprising aspect of the discovery is that lactose, of all the choices for supporting materials or anything else in the compounding pharmacists cupboard (excipients, solubilizers, etc.) stabilizes tigecycline hydrochloride.

    So yes, it might have been obvious to try tigecycline hydrochloride as a substitute for minocycline hydrochloride in the CN ‘550 formulation, but there was no reason to expect that it would work. In fact, the expectation would be that it would be as unstable on lactose as it was on any other support. Without some reference showing that lactose is known to stabilize tetracyclines, there just isn’t any reason to expect the combination tigecycline-hydrochloric acid-lactose to be stable. Apotex didn’t provide such a reference and the Board didn’t find that teaching in any of the references that were provided.

    1. 8.1

      Great comment. This is the key information missing from the article. As i mentioned in my earlier comment, the determination of obviousness turns on whether its reasonable to expect success. From this it seems that the mixture does function significantly different than the prior art despite the similarities.

      1. 8.1.1

        edstirling the mixture does function significantly different than the prior art

        This isn’t a method claim. It’s a composition claim. There are no “functional” limitations.

        The prior art teaches that you can make a stable lyophilized preparation of a structurally very similar drug using lactose and hydrochloric acid. The fact that Wyeth was able to show differences between the two drugs and “how the mixture functions” to achieve stability is beside the point. Nobody is disputing that the drugs have different properties or that tigecycline was less stable that minocycline. That was already known.

        And by the way, if these mixtures function soooo differently to stabilize the two drugs and everything is soooo unpredictable, it’s kinda odd that both the CN ‘550 and Wyeth’s spec teach that you can use different carbohydrates and acids (e.g., from Wyeth’s spec: “other suitable carbohydrates may include higher saccharides such as polysaccharides; complex carbohydrates such as hetastarch, dextran; and celluloses such as hydroxypropylmethyl cellulose and hydroxypropyl cellulose. It is further expected that combinations of carbohydrates, including monosaccharides and trisaccharides, will be suitable carbohydrates and work to reduce epimer formation according to the invention.”; CN ‘550 teaches dextran as a substitute for lactose).

        Oh, and this, too, from Wyeth’s spec: Any carbohydrate capable of reducing epimer formation in the invention is a suitable carbohydrate and this invention is not limited to compositions employing those carbohydrates specifically identified.

        Wow. So let’s get this straight. A teaching that lactose and low pH can be used to make a stable formulation of a very similar drug is hand-waved away, but a teaching that a handful of similar sugars work with tigecycline means that Wyeth is entitled to compositions with “any carbohydrate that works”. Sure, that makes sense.

    2. 8.2

      Mike, congrats on kicking up much of the same dust that was kicked up by the patentee here.

      You write: lactose … is simply one of many inactive supporting agents that can be used to prepare stable formulations.

      Right. And it was used to prepare a stable formulation of a structurally closely related drug (minocycline). Hence: obvious to try it.

      the dry salt of minocycline on a supporting agent is stable. The stability of minocycline hydrochloride does not depend on the presence of lactose

      That’s nice. The cited prior art taught that stabilization of minocycline — a structurally closely related drug — was stabilized with lactose as the supporting agent. Hence: obvious to try it as the supporting agent (it’s not an unusual supporting agent in any event).

      although certain solubilizers are known to reduce the epimeric degradation of tetracyclines, lactose is not

      The specification teaches that “Epimerization is a known degradation pathway in tetracyclines generally.” This statement is key because it shows exactly where Wyeth started playing games with its treatment ‘550 patent. They knew about the ‘550 patent when they wrote the spec. This is what they wrote:

      “Chinese patent application CN 1390550A discloses that minocycline could be combined with an acid to increase the stability toward the oxidative degradation. It further discloses the use of a caking agent, such as mannitol.”

      Note how they conveniently left out lactose from that description (the application expressly teaches lactose as an alternative to mannitol). But here comes the money quote:

      “This reference says nothing about tigecycline”

      But the reference does say that “[T]he invention belongs to broad-spectrum antibiotics”, and tetracyclines are certainly within that class. More importantly, the fact that tigecycline isn’t expressly called out doesn’t mean anything for the obvious issue.

      Wyeth’s spec continues: “nor does [the ‘550] suggest that carbohydrates could be used to reduce either oxidation or epimerization for minocycline in reduced pH environments.”

      Except that “epimerization is a known degradation pathway in tetracyclines generally” and the ‘550 patent teaches a stabilization formulation that works in “reduced pH environments” (“optimum pH between 2.0 and 3.5”, and hydrochloric acid is expressly called out as a “pH adjusting agent”).

      A person having ordinary skill in the art would not have used tigecycline at all.

      Except that people of ordinary skill were using the drug because … it worked. Wyeth admits that. It has to. The formulations just weren’t as stable as certain formulations of related drugs. The problem was : how to improve the stability of tigecycline. One obvious solution: look to the prior art and see what other people used to stabilize structurally very similar drugs and try that. Why on earth would anyone make it harder on themselves? Lactose and hydrochloric acid (along with other acids and other carbs like glucose, sucrose, mannitol, sorbitol, glycerol, trehalose etc) are sitting on Wyeth’s shelves, just like they are on the shelves at every drug company. They are what is known as “standard reagents” — especially if you are lyophilizing drugs. Making lyophilyzed compositions and testing them is what these companies do eight days a week.

      Wyeth’s problem — or, rather, Wyeth’s patent attorney’s problem — was that the solution Wyeth found was taught in the prior art. And Wyeth knew about that art when it prepared its patent specification. So what did Wyeth do? It used hindsight to concoct an elaborate explanation as to why one would never, never, never ever have expected that to work. But for some mysterious reason, Wyeth went ahead and tried it anyway! Maybe a lightning bolt struck the bench and some lactose puffed into the tigecycline solution. Total serendipity!

    3. 8.3

      Mike: The surprising aspect of the discovery is that lactose, of all the choices for supporting materials or anything else in the compounding pharmacists cupboard (excipients, solubilizers, etc.) stabilizes tigecycline hydrochloride.

      According to Wyeth’s’ spec, there are reams of different carbohydrates that would be expected to work. That’s based on Wyeth’s testing of a tiny handful of other small sugar molecules.

      So, to summarize:

      1) Lactose plus acid formulation stabilizes drug that is structurally very very similar to tigecycline
      2) The same formulation stabilizes tigecycline
      Wyeth: 2 is totally surprising in view of 1

      on the other hand

      1) Lactose stabilizes tigecycline
      2) Carbohydrates that are structurally very very different from lactose can be substituted for lactose
      Wyeth: 2 is totally expected in view of 1

      I think there’s a pretty clear disconnect in Wyeth’s line of reasoning.

      Let’s try an analogy. Prior art “in the field of seafood for cats” teaches that banana or apple juice plus Pacific salmon is highly appealing to cats. The article appeared in Pacific Salmon Monthly.

      I claim “an improved cat food composition comprising Atlantic salmon and banana juice”. I discovered that the added banana juice slows down the rotting of the fish, and Atlantic salmon — previously sold as cat food, by the way — tends to rot faster than Pacific salmon. Plus cats like the taste of the banana juice. Can I avoid a finding that my broad composition claim is obvious by arguing that banana juice was added in the prior art only to provide a flavor of interest to cats?

      1. 8.3.1

        The hole in Apotex’s argument and in your discussion of it, MM, is the statement the lactose was used to stabilize minocycline. You say “minocycline — a structurally closely related drug — was stabilized with lactose.” It wasn’t. It didn’t need to be. You later say, in your summary, as Apotex tried to say, “lactose plus acid formulation stabilizes drug that is structurally very very similar to tigecycline,” but as the PTAB and federal circuit understood it, this was an overstatement. The acid formulation stabilized minocycline, but there was never any stabilization imputed to the addition of lactose. Your statement that minocycline is “very very similar to tigecycline” ignores the crucial fact that tigecycline is much less stable than minocycline to epimerization. Two very similar compounds have very different properties. One of those little things that makes the chemical arts so much less predictable than the mechanical arts. Without that link, without some indication that a few select disaccharides would show a stabilization effect to one molecule that was unknown and unnecessary to the other, there is no reason to think that tigecycline would be any more stable combined with lactose than combined with any of the other
        10 million or so known chemical compounds, and therefore no reason to expect that the formulation of CN ‘550 would be an appropriate delivery form for tigecycline.

        1. 8.3.1.1

          You say “minocycline — a structurally closely related drug — was stabilized with lactose.” It wasn’t. It didn’t need to be.

          “It wasn’t” and “It didn’t need to be” are two very different propositions in this context.

          A formulation of lactose and low pH was indisputably taught in the art to stabilize minocycline. Given the very similar structures, that’s all the motivation you need to try the combination to see if it stabilizes tigecycline. And lo and behold: it works!

          Wyeth’s argument is really a teaching away argument but there is no teaching away in the record. There’s no teaching or evidence in the prior art that lactose is destabilizing minocyline (or any other tetracycline derivative) at low pH, nor is there any teaching that a lyophilization formulation comprising lactose at low pH would further destabilize tigecycline (note that the claims read on lyophilized formulations, in their lyophilized state).

          The acid formulation stabilized minocycline, but there was never any stabilization imputed to the addition of lactose. Your statement that minocycline is “very very similar to tigecycline” ignores the crucial fact that tigecycline is much less stable

          Sure, it’s less stable. But both drugs are subject to epimerization — that’s a property of tetracyclines, as admitted by Wyeth.

          without some indication that a few select disaccharides would show a stabilization effect

          But according to Wyeth, it’s not just a “few select disaccharides”. It’s vast multitudes of carbohydrates. Lactose is just one of them. It’s also one of the carbohydrates that was taught in the art as useful in a formulation for stabilizing a very stucturally similar drug (perhaps the most structurally similar drug?) that is — like all tetracyclines — subject to degradation by epimerization. Again, there is no teaching that using lactose or any other carbohydrate (e.g., polysorbate 20) in a lyophilization formulation destabilizes any tetracycline. But there is a very clear teaching that a low pH formulation including lactose and other carbohydrates stabilizes a structurally very similar drug.

          I’ll grant you, Mike, that your comments make me believe that this case is a shade closer than it appeared when I first read it. But I still don’t think it’s a close case. I think it was decided wrongly, and given the language in the case I’m glad it’s not precedential.

          1. 8.3.1.1.1

            MM, I think we’re a lot closer in thought on this argument than our posts would indicate. I don’t think the Wyeth patent should have been validated, but I blame Apotex argument rather than the PTAB or the court. They merely declined to fill in the gap in Apotex’s argument.

            Apotex still hasn’t pointed to any evidence that lactose stabilizes minocycline. I’m afraid I have to dispute your statement, “A formulation of lactose and low pH was indisputably taught in the art to stabilize minocycline.” The art can only be pushed as far as stating that a formulation of low pH was taught in the art to stabilize minocycline. I have the same argument with your statement “It’s also one of the carbohydrates that was taught in the art as useful in a formulation for stabilizing a very stucturally similar drug.” I have still seen no teaching that lactose had any effect on stabilizing minocycline.

            I will agree that there is no teaching away of the use of lactose. If Wyeth could have pointed to anything that showed that lactose destabilized tetracyclines, they would have had the argument in the bag. The teaching away is in the replacement of a stable tetracycline, minocycline, with an unstable tetracycline, tigecycline, in a pharmaceutical preparation. A PHOSITA would expect the tigecycline in that preparation would still be unstable. It simply would not be expected that a supposedly inert bulking agent like lactose would stabilize a tetracycline known to be unstable. It’s like packing gunpowder in a rolled up piece of cardboard and nitroglycerin in a rolled up piece of cardboard, and finding out that they are both now equally stable to shock. There is simply no “expectation of success” in the KSR “obvious to try” analysis.

            One other comment – polysorbate 20 is not a carbohydrate. If it was, Wyeth’s argument would have totally collapsed, because polysorbate 20 and other solubilizers were shown in the cited art to stabilize tetracycline salts to epimeric degradation. Apotex’s gap is that no references were provided to show whether lactose or any other carbohydrate stabilizes tetracycline salts. Without making this connection, there is no way to rebut Wyeth’s demonstration of unexpected results.

            1. 8.3.1.1.1.1

              Mike: The teaching away is in the replacement of a stable tetracycline, minocycline, with an unstable tetracycline, tigecycline, in a pharmaceutical preparation.

              As other commenters have noted, this is a strange way to look at what’s happening. Yes, minocyline was known to be more stable than tigecycline. But they are very very similar. The whole point of Wyeth’s work was to increase the stability of tigecycline. They clearly believed it was possible to achieve that.

              A PHOSITA would expect the tigecycline in that preparation would still be unstable

              But that assertion is not “teaching away” from trying the stabilizing formulation that was successfully used to stabilize the most similar drug. If you ask me if it’s going to rain tomorrow and I say “unlikely” that’s different than me saying “No chance because there are no rainclouds in the hemisphere.” Wyeth didn’t present any evidence of references “teaching away” from making a lyophilized composition of tigecycline comprising lactose and a low pH because there isn’t any.

              Mike: It’s like packing gunpowder in a rolled up piece of cardboard and nitroglycerin in a rolled up piece of cardboard, and finding out that they are both now equally stable to shock.

              That would be surprising. But this is key: Wyeth’s composition claims don’t recite any limitations requiring any level of stability. Would I be surprised to find that cardboard improves the shock-resistance of nitroglycerin compared to nothing at all? Of course not. Nobody would be surprised by that. Even if I discover that cardboard improves the shock-resistance of nitroglycerin so that it’s just as shock-resistant as dynamite (e.g, because of some special property of nitroglycerin), that still doesn’t entitle me to the broad claim of “nitroglycerin in cardboard.”

              One other comment – polysorbate 20 is not a carbohydrate.

              My bad. I’m not sure where I pulled “polysorbate 20” from…? I probably meant to say “dextran.”

        2. 8.3.1.2

          One additional comment regarding this:

          The hole in Apotex’s argument and in your discussion of it, MM, is the statement the lactose was used to stabilize minocycline.

          This is what I wrote: Lactose plus acid formulation stabilizes a drug that is structurally very similar to tigecycline. I’m trying to be as consistent and clear as possible about the CN ‘550 patent teaching. Note again, that it was not just lactose that was taught in the CN’ 550 patent — other very different carbs are taught as working in the low pH formulation. And Wyeth’s spec makes it clear that their “discovery” was that pretty much any carbohydrate will work in their low pH formulation as well.

          The take-home point from my side of the argument is that it doesn’t matter *why* the CN ‘550 added lactose to their formulation. The point is that the authors of the CN’550 knew that lactose (as well as other carbohydrates). and low pH worked to stabilize a structurally very similar drug Absent a bona fide teaching away, a later party discovering why that formulation works on certain very similar tetracyclines shouldn’t be entitled to repatent the lyophilization formulation.

          Look at it this way: if the person in charge of this project at Wyeth knew about the CN ‘550 reference (which is assumed to be the case) and failed to try a lypholization formulation comprising low pH and a carbohydrate, what would their supervisor say when another company’s patent on that formulation showed up? “I can’t believe they tried that”?! That seems very unlikely. More likely the supervisor is going to chew out the person in charge for not testing the effect of the CN’ 550 formulation.

          The CN’ 550 published in 2003. Wyeth’s spec was filed two years later. That means that the experiments with the lowpH+carb lyophilization formulations were performed and written up in the preceding year following the CN’ 550 publication. There is a strangely placed passage in the spec (looks like a last moment paste-in) referring to “In another experiment” that seems designed for the sole purpose of creating the issue regarding “epimerization”. Note that the passage doesn’t describe a very well executed “experiment” at all. It describes an observation (presumably Wyeth’s) that “lyophilized” minocycline was epimerized at pH 5.0 to a level of 2.65% after 20 days but the passage doesn’t tell us what else was present in the lyophilization formulation, nor does it tell us what effect lactose has on the rate of epimerization of minocycline at low pH. Curious, isn’t it?

  2. 7

    One of the considerations that is commonly overlooked in the inventive step argument is the entropy of innovation.

    For those unfamiliar with entropy, consider two bags, one of sand and one of salt. The bags are opened and mixed together. The result is a disordered structure, where it is very difficult to separate the individual grains of the sand and the salt from the mixture. Entropy is easy to increase but difficult to decrease.

    Similar considerations apply to prior art references. Before the invention is made, the references exist in the prior art at random, and the likelihood of their being brought together is remote. The references exist in the disordered structure of the prior art at a high level of entropy. Once an invention has been made and defined in a claim, then the search becomes structured. If some features are traced to a first reference, the search for the remaining features is not random but is structured. The result flows from an enormous and often unappreciated decrease in the search entropy resulting from knowledge of the invention.

    The question then becomes: Would the skilled person have the reference in a single field of view absent hindsight knowledge of the invention? On the facts in this case, the answer arguably is: Not very likely.

    1. 7.1

      PC Before the invention is made, the references exist in the prior art at random, and the likelihood of their being brought together is remote.

      In fact, in all cases the references are already “brought together” because they were both published on Earth. In many other cases, the references are “brought together” hundreds of times a day by creative or curious people doing keyword searches on a topic of interest that is touched upon by both references. You know, a topic like “stability of tetracycline and glycylglycine compounds”.

      Sometimes the references are even in the same state, in the same city, in the same building, in the same laboratory, sitting on the same shelf. Why? Because the inhabitants are intellectually curious and they like to know what’s going on.

      Would the skilled person have the reference in a single field of view absent hindsight knowledge of the invention? On the facts in this case, the answer arguably is: Not very likely.

      Really? Because a skilled person doesn’t know how to use the Internet to search for relevant teaching?

    2. 7.2

      “Entropy of Innovation” – interesting concept, but just does not fit into the legal realm.

      A better distinction between 102 Art and 103 Art is available.

      All art that is extant is 102 Art – for the sole purpose of novelty. But NOT all extant art is available for 103 Art. So while PHOSITA has “universal knowledge,” this does NOT mean that all extant art is available in the 103 Prior Art sense.

      Notwithstanding any “internet search” capability ;-)

    3. 7.3

      Paul, that is why both the Supreme Court and the Federal Circuit require there to be a problem known in the art before the invention for which the combination of reference provides a solution. It is easy to take the claims and use them as a roadmap. It is less easy to show a problem in the prior art other than a general desire to improve things.

      The PTAB routinely invalidates claims simply because the petitioner files prior art showing all the elements. But the Feds are pushing back at this skullduggery, and are actually enforcing KSR the way it was written.

  3. 6

    The way they combined the references was wonky. Making the chinese reference the primary and substituting in tigecycline was a mistake.

    I think they would have had a better shot if they reversed the references. Make tigecycline the primary prior art. State that it is known that tigecycline has stability problems. Then argue it would be obvious to add lactic acid of the Chinese reference to the tigecycline; the motivation being to increase stability. Finally, argue that increase stability would be expected because both tigecycline and minocycline were structurally similar and operate by a similar mechanism.

      1. 6.1.1

        Is it possible that the divergence of obviousness opinions between electromechanical and biochemical arts could be driven by the judges’ differing levels of skill in the relevant arts? In other words, are the judges finding obviousness where they are confident in understanding the art, and nonobviousness where they are not confident? I ask because you reference a good engineer being able to give pre-assurances, and this is a case where it seems like a good (Ph.D.) biochemist also could give pre-assurances based on structural similarities and binding actions between the chemicals; particularly so where software is available to analyze Gibbs free energy for various chemical structures.

        1. 6.1.1.1

          What appears to be happening is the “rewarding” of a lack of maturity. In other words, the “gosh, we don’t know why this is so hard, so we must think that it is not obvious” line.

          Absent (so far) from the comments has been Random’s views on “predictability” and how this (errant) line feeds into that (and the necessary logical next step of the either Flash of Genius or unpredictable serendipity as the onlypaths to earning a patent…).

  4. 5

    It is telling to see some of the most vehement opponents of IPRs on this blog rise in defense of the IPR decision in an extremely rare IPR claim invalidity decision that is being reversed on 103 [rather than procedural] grounds by a Fed. Cir. panel.

    1. 5.1

      Telling…?

      Not in the way that you are trying to portray Paul.

      I don’t see ANY “pro-IPR” comments here, but rather, many comments in the underlying law that appears to have been bent (and bent badly) in favor of Pharma.

      What is “telling” is your rather bad attempt at spin to turn the comments into something that they are not.

  5. 4

    Perhaps this is non-precedential for a reason. The knowledge that tigecycline is effective ‘where other antibiotics have failed’ would COMPEL a person to substitute it for other antibiotics, especially closely related compounds that have shown remarkably similar behavior in other aspects (like, they all kill bacteria through the same mechanism).
    On a theoretical note, if one is compelled to make the combination (it’s obvious to try), how does it become patentable just because one is not sure it will work? The combination is still obvious, no matter what the outcome.

    1. 4.1

      K.D.C., the people who know what they are talking about in this arena all seem to agree that both the PTAB and the Federal Circuit were making egregious mistakes in favor of Wyeth against Apotex in this end-run of Hatch Waxman. What that tells me is that whole system of IPR/Federal Circuit is enormously corrupt. It is simply filled with politics and does not provide due process to non stakeholders.

    2. 4.2

      KDC: On a theoretical note, if one is compelled to make the combination (it’s obvious to try), how does it become patentable just because one is not sure it will work? The combination is still obvious, no matter what the outcome.

      One theory that has been applied without too much controversy is that an “obvious to try” combination can become patentable if the applicant can demonstrate results of the combination that are sufficiently unexpected and useful that the combination is deemed patent-worthy in spite of the fact it was “obvious to try”. In my opinion, enforcement of any patents granted in this situation should be limited to uses commensurate with that unexpected result and the doctrine shouldn’t apply when there’s evidence of near-simultaneous discovery. The other important aspect of the doctrine to make it as fair as possible for everyone involved is to require any evidence unexpected results to be presented at the time of filing (currently not the case!) and to require that the claim be narrowed to the species formulations that yield the unexpected result (almost always narrower than the obvious-to-try teachings).

      For example, let’s say Reference A teaches a chocolate and tumeric bar, and reference B teaches that rounded peanuts usually makes chocolate bars taste better. In light of that teaching, it’s obvious to combine the teaching to try the combination to see if it makes a better tasting bar.

      So let’s say you try the obvious-to-try combination and, lo and behold, it does make a better tasting bar! Should be you be entitled to claim “A composition comprising chocolate, tumeric and rounded peanut pieces”? That doesn’t seem correct. But what if you find that the combination unexpectedly results in a bar that stays fresh-tasting ten times longer than the existing bars? What sort of patent protection should you be entitled to then? It seems to me that you should be able to get a claim limited to something narrower than the broad “obvious to try combination”, i.e., something in line with the specific concentrations and peanut sizes that you showed yield the unexpected result. As I alluded to above, it also seems that you’d need to both show at least some of those unexpected results and recite the limiting ranges which yield those results in the spec as filed. Providing that information years later via, e.g., an expert declaration or, worse, citation to a later-filed report by a third party, shouldn’t suffice (but presently it would seem to suffice … unfortunately).

      1. 4.2.1

        you appear to have forgotten about inherency….

        (besides the numerous new law “creative” efforts of yours…)

      2. 4.2.2

        MM:
        You are more generous than I am. I think that if it’s obvious to try, it’s obvious. Unexpected results, lack of assurance, and synergy are irrelevant to the fundamental question: would it be obvious for an artisan to do that?
        In your example, providing a specific blend of chocolate, peanuts, and tumeric might be patentable even under my view, since it might not be obvious that any blend might enhance shelf-life. But a claim merely reciting a synergistic but obvious combination of ingredients seems too much. And synergy clams are always suspect.

        1. 4.2.2.1

          Obvious to try = obvious under 103? Hang on! What if there are 999 compounds that are all equally “obvious to try”, all at least plausible as cures for disease X, but you’ve selected, singled out and specifically claimed the only one in that List that turned out to have any significant effect against X? Still no contribution to the art? Still no patent for you?

          The whole area of “obvious to try” is very much disputed, in any patent jurisdiction. Anybody who tells you “it’s all very simple” to get it right is having you on. Ask the specialist patent judges in England.

          Innovative pharma, and the generics companies, deserve better than this, from the patent courts and litigators. The climate of our times is captured in that BREXIT theme “I think that the public’s of a mind that by now it’s heard quite enough from the experts, thank you”. But I think having expert specialist patent judges is exactly what is needed.

            1. 4.2.2.1.1.1

              (for example, “obvious to try” requires a limited number of ways to try – if you violate that premise then you are no longer talking about “obvious to try.”)

          1. 4.2.2.1.2

            “Obvious to try = obvious under 103?”

            Yup. Ever since the Supreme nonsense issued in KSR, which is summarized in part thusly:

            I.EXEMPLARY RATIONALES

            Exemplary rationales that may support a conclusion of obviousness include:

            (A) Combining prior art elements according to known methods to yield predictable results;
            (B) Simple substitution of one known element for another to obtain predictable results;
            (C) Use of known technique to improve similar devices (methods, or products) in the same way;
            (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results;

            (E) “Obvious to try” – choosing from a finite number of identified, predictable solutions, with a reasonable expectation of success;

            (F) Known work in one field of endeavor may prompt variations of it for use in either the same field or a different one based on design incentives or other market forces if the variations are predictable to one of ordinary skill in the art;
            (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention.

            MPEP 2143

            link to uspto.gov

            1. 4.2.2.1.2.1

              So I have a combination lock for my bike, with four discs each marked with numbers 0 to 9.

              The single number that opens the lock is not obvious, I hope.

              There are 10,000 permutations of the lock. Each one is indisputably and perfectly equally “obvious to try”.

              Agreed?

              But it’s different, when searching for a new and effective drug, because you don’t know whether any of the permutations to be tried will actually work. Agreed?

              So, obviousness. When when looking for a new and effective drug, when does any one permutation to be tried become “obvious”? Les, how about you have a go at writing the test.

              1. 4.2.2.1.2.1.1

                1. Don’t shoot the messenger. I’m just reporting the news. I agree with you, its nuts.

                2. In any event, I think you have an argument that if you don’t know if any of the 10,000 trial candidates will work, then none of them is obvious to try. I think obvious to try covers solutions one of ordinary skill in the art would expect to work.

                If dinner needs a little help taste wise, its obvious to try a little salt and pepper, but not drain cleaner…. even though they both might be found in a kitchen cupboard.

                1. Les, you get it. It’s all about whether one would “expect” it to “work”. With the bike lock, you know with 100% certainty, that one of the ten thousand permutations, a “finite” bounded and “limited” number of possibilities, must open the lock. So all of them are “obvious to try”. But in the unpredictable world of chem/bio, when does the level of expectation of success rise to a level high enough to make it “obvious”?

                2. MD in the unpredictable world of chem/bio, when does the level of expectation of success rise to a level high enough to make it “obvious”?

                  The short answer is “when it’s actually totally predictable.” The long answer is the same as the short answer “plus other situations.”

                  Example: Prior art teaches that 0.1M EDTA chelates trace amounts of magnesium in a zillion solutions comprising each of a zillion different nucleic acid polymers.

                  Now you’ve got a newly discovered nucleic acid whose sequence hasn’t been disclosed before. Will 0.1M EDTA chelate trace amounts of magnesium in a solution comprising this nucleic acid? Yes, it will.

                1. (the better example is that you have a single digit-wheeled “lock” with 0-9, and yes, then your “secret code” would in fact be both obvious to try and obvious)

                  As to any other examples and numbers of items, well, such becomes a bit more fact specific (do you remember that one Canadian patent case, the one that the applicant himself painted the viable pool of items numbering more seconds than the estimated age of the universe….?)

          2. 4.2.2.1.3

            Yep, no patent for you. I would expect a high school student to sort through all 999 things that are obvious to try.
            I know that the case law goes the other way, but I see no rational basis for it.
            Innovative pharma can get a patent if they combine something with no clear impetus in the art, and it turns out to do something good. If the art compels them to make the combination, then they are compelled to make the combination, and there is nothing inventive about that.

        2. 4.2.2.2

          KDC:
          Your statement that “unexpected results, lack of assurance, and synergy are irrelevant” if it’s “obvious to try” might be correct if it didn’t contradict Graham v. John Deere. Nice try though.

          1. 4.2.2.2.1

            You are correct, John Deere. Errant case law makes unexpected results, etc. legally relevant. I just think the obvious to try rule is wrong. Unexpected results, etc. are logically irrelevant.
            Is there any extant theory that explains that if it is obvious to combine a few things (MM’s chocolate tumeric and peanuts), but not obvious that they will taste good, then that negates the initial proposition that it was obvious to combine them? The rule hardly survives its own utterance.
            The law is full of legally relevant junk that turns out to be logically or factually irrelevant.

        3. 4.2.2.3

          K David: a claim merely reciting a synergistic but obvious combination of ingredients seems too much. And synergy clams are always suspect.

          Generally speaking, I agree with this sentiment. As I noted in my comment, the analysis is very fact-dependent (e.g., any “synergy” is not enough), the claims granted in such cases should be narrow, and enforceability should be limited. For instance, in my example, the “inventor” shouldn’t be able to obtain or enforce a claim that would keep everyone else from making a candy bar comprising chocolate, turmeric and rounded peanut pieces.

          1. 4.2.2.3.1

            Pray tell, in your candy bar example, exactly what claims would be allowed and how would enforcement of those claims be enacted?

    3. 4.3

      I havent read the disclosure, but from the explanation above the motivation to combine is the need to preserve the tigecycline. Its use as an antibiotic is only relevant inasmuch as its effectiveness is preserved. You’d have to ask the experts about the expectation that that the mixture would work for other tetracyclines. Sounds to me like there is absolutely sufficient motivation in the prior art (the need to preserve) and a reasonable expectation of success, depending on how fair it is to expect properties of tetracyclines to be transitive. The court’s statement of nonobviousness sounds lazily dismissive, but that could arise from counsel’s lack of clarity in stating their obviousness theory.

      1. 4.3.1

        As a matter of law (and so evident even here without much reading), I don’t see how “counsel’s lack of clarity in stating their obviousness theory” could get in the way of the judges’ actually following the law.

        What the judges have written shows almost an intent to not follow the law.

  6. 3

    100% agree with MM. I, too, am flabbergasted by this passage:

    “Dr. Nelson does not explain . . . why the knowledge that tigecycline is effective ‘where other antibiotics have failed’ would lead a person having ordinary skill” to substitute tigecycline for minocycline in CN ’550.

    The statement almost refutes itself. It is beyond obvious that the knowledge that substance X is effective where other substances have failed, would lead a POSA to substitute one of the failed substances with X.

    Lourie’s jurisprudence on obviousness is a huge mess.

    1. 3.1

      I don’t think this automatically follows by necessity – in a vast enough field, the fact that something is known to work “sometimes” may not make it obvious to try in every case.

    2. 3.2

      I don’t think the analysis is whether it is obvious to substitute one antibiotic for another in the context of efficacy of the antibiotics. I believe the proper analysis should be based upon what the prior art teaches – that lactose added in a formulation helps stabilize minocycline from degradation. Thus, the question becomes whether it would be obvious to a PHOSITA to substitute tigecycline for minocycline in the prior art with a likelihood of successfully stabilizing tigecycline from degradation because of the lactose.

  7. 2

    Dennis: A first thought that comes to my head in this case is whether the outcome here stems from the fact that this is a pharmaceutical case rather than one focusing on electronics or mechanical engineering.

    That distinction shouldn’t matter too much for the motivation prong of the obviousness analysis. I think the CAFC just got this one wrong. Way wrong.

    Check out this passage from page 7:

    From the Board’s opinion, it is clear that it fully considered that potential motivation to combine and found it wanting. In its first paragraph of analysis, the Board acknowledged Apotex’s assertion that one of skill in the art “would find reason to substitute tigecycline for minocycline” in CN ’550 because it “was known to work where other antibiotics have failed,” and because “minocycline and tigecycline are tetracycline antibiotics” with “identical A and B rings.” … In
    addressing that assertion, the Board rejected Apotex’s proffered expert testimony on this point as unpersuasive.

    Okay, so let’s see why the expert testimony was deemed “unpersuasive”:

    “Dr. Nelson does not explain . . . why the knowledge that tigecycline is effective ‘where other antibiotics have failed’ would lead a person having ordinary skill” to substitute tigecycline for minocycline in CN ’550

    Excuse me? Drug X works in certain situations where other drugs don’t. That’s why you want to look for way to stabilize drug X. Where do you look? At methods for stabilizing the structurally most similar thing to drug X, which is minocycline. This has to be “explained”? C’mon. Regardless of that silliness, the fact is that this motivation was explained. That explanation was “acknowledged” by the Board (and that acknowledgment was acknowledged by the CAFC).

    And Apotex did not establish that minocycline was known to “have failed.”

    So what? Why on earth would that matter? It’s already “established” that different drugs work better under some circumstances. Tigecycline is non-identical to minocycline, i.e., it’s different. End of analysis. I don’t have to “establish” that “round wheels failed” to prove there is a motivation to use round wheels on my new vehicle. The PTO and the CAFC are being very silly here. But wait! There’s even more silliness:

    Moreover, the Board found that no evidence suggested that tigecycline would be as stable in the CN ’550 composition, as the notion of “identical A and B rings” alone was insufficient to show that.

    Two points: (1) the claims do not require that tigecycline be “as stable as minicycline is in the CN’550 composition” so this is a red herring; and (2) even if the claims did recite such a limitation, the lack of such evidence hardly equates with a lack of motivation. The motivation is to improve upon the art in some way. That’s all the motivation you need. In this case, a substantial improvement in stability was sought, so the ordinary skilled artisan would look for a prior formulation where a drug with a similar structure was substantially stabilized.

    Imagine a situation where you want to improve the stability of drug X, which is very unstable. The prior art a teaches that a very similar drug Y was 100% stabilized (wow!) by adding lactose and acid. There is no teaching away from using that formulation on other drugs. Somehow you obtain a claim to “Drug X with lactose and acid.” To demonstrate a motivation to combine your drug with lactose and acid, do I need to provide “evidence” that your drug X would be expected to be 100% stabilized? Of course not. That would be absurd.

    The CAFC’s errors here look like egregious legal errors, not just misunderstanding of facts. I’d take this en banc if I was Apotex and this was important to my business.

    1. 2.1

      Let me improve the analogy in my comment. I wrote:

      I don’t have to “establish” that “round wheels failed” to prove there is a motivation to use round wheels on my new vehicle.

      but I should have written

      I don’t have to “establish” that “prior art jeep with round wheels failed” to prove there is a motivation to use round wheels on my new vehicle.

    2. 2.2

      MM, here the PTAB and the Federal Circuit seem to have bent over backwards to help a well known patent holder against a generic who seems to have end-run Hatch Waxman by filing an IPR. You seem flabbergasted that they all seemed to go to sleep and stop thinking all of a sudden without considering the politics of the situation.

      Now that Wyeth has won, and Apotex has lost, just how does Hatcha Waxman work when the generic is required to file suit to challenge the validity of a patent?

    3. 2.3

      There used to be the notion of the “inherent uncertainty of chemistry” mentioned in past decisions. If that is what Dennis Crouch is alluding to then it makes sense. That is also why chemical compositions, especially in Biotech, are often held to a much higher enablement stardard – you can’t generalize that all proposed composition will work as expected. So in chemistry, you get an easier time rebutting obviousness and higher enablement requirement. Makes sense to me.

  8. 1

    Re: “A first thought that comes to my head in this case is whether the outcome here stems from the fact that this is a pharmaceutical case rather than one focusing on electronics or mechanical engineering. ”

    Or to put it another way, yes, assuming this still fits under the standard rubric and rational difference in 103 treatment for “the unpredictable arts?”

    Speaking of that, the best selling book “The Disappearing Spoon” by Sam Kean [about the Periodic Table and its contents] is full of fascinating stories about the chemical unpredictability of even the most closely related elements.

Comments are closed.