On Appeal, Abuse-Deterrent OxyContin Patents are Invalid

By Dennis Crouch

Purdue Pharma v. Epic Pharma (Fed. Cir. 2016)[1]

In an important Hatch-Waxman related decision, the Federal Circuit has affirmed the lower court’s holding that Purdue Pharma’s abuse deterrent OxyContin patent claims are invalid as anticipated and/or obvious over the prior art.[2]

Of interest, it appears that the patentee was harmed here by the Patent Act’s objective approach to obviousness that, according to the statute, must be considered without regard to the actual “manner in which the invention was made.”  In particular, during the process of creating its low-ABUK oxycodone, the innovators first discovered an 8α impurity isomer created during drug manufacture that led to a problematic 14-hydroxy compound.  The claimed invention was a product having a reduced 14-hydroxy  level.  In the appeal, the Federal Circuit noted that, while the 8α impurity may have been a new discovery, it was not necessary to achieve the claimed low-ABUK results.  In particular, the court noted that reduction of both the 8α and 8β impurities would also achieve the same result without having to distinguish between the two — affirming that “a skilled artisan would recognize that hydrogenation could be used to remove the remaining 14-hydroxy, regardless of the source of the 14-hydroxy.”

Of course, the claims did expressly require that the impurity being reduced be “derived from 8α.”  That limitation, however, was deemed an illusory product-by-process limitation.

We also conclude that, because “derived from 8α[]” is a process limitation, the district court did not err in disregarding the limitation in its obviousness analysis. We have clearly stated that “‘[i]n determining validity of a product-by-process claim, the focus is on the product and not the process of making it.’” Greenliant Sys., Inc. v. Xicor LLC, 692 F.3d 1261 (Fed. Cir. 2012) (quoting Amgen Inc. v. F. Hoffman-La Roche Ltd., 580 F.3d 1340 (Fed. Cir. 2009)). “That is because of the . . . longstanding rule that an old product is not patentable even if it is made by a new process.” Id.; see also SmithKline Beecham Corp. v. Apotex Corp., 439 F.3d 1312, 1317 (Fed. Cir. 2006) (“It has long been established that one cannot avoid anticipation by an earlier product disclosure by claiming . . . the product as produced by a particular process.”); In re Thorpe, 777 F.2d 695, 697 (Fed. Cir. 1985) (“If the product in a product-by-process claim is the same as or obvious from a product of the prior art, the claim is unpatentable even though the prior product was made by a different process.”).

In Amgen, the Federal Circuit did find that the particular claimed process will be relevant if it imparts distinguishing structural differences on the resulting product.  Here, however, the court found that the fact that the negative 14-hydroxy byproduct was derived from 8α “imparts no structural or functional differences” over the prior art that already included low-ABUK product achieved without focusing on the 8α isomer.

Generic OxyContin has been on the market for the past year – following the district court judgment that has now been affirmed – and so this decision should not have a major market impact.

This case here is one part of a complex web of battles that Purdue appears to be fighting to protect its exclusive rights to sell OxyContin – or at least delay competition. Although these asserted claims have been found invalid, Purdue is asserting a set of new patents, including U.S. Patent Nos. 8,309,060, 8,337,888, 8,808,741, 8,894,987, 8,894,988, 9,060,976, 9,034,376 and 9,073,933.

= = = = =

[1] App. No. 2014-1294 (Fed. Cir. 2016); on appeal from In re OxyContin Antitrust Litig., 994 F. Supp. 2d 367 (S.D.N.Y. 2014).

[2] Asserted patents include U.S. Patent No. 7,674,799 (“’799 patent”), U.S. Patent No. 7,674,800 (“’800 patent”), U.S. Patent No. 7,683,072 (“’072 patent”) (collectively, “the low-ABUK patents”), and U.S. Patent No. 8,114,383 patent (“’383 patent”).

About Dennis Crouch

Law Professor at the University of Missouri School of Law. Co-director of the Center for Intellectual Property and Entrepreneurship.

42 thoughts on “On Appeal, Abuse-Deterrent OxyContin Patents are Invalid

  1. I’m not sure how the case (which was a correct holding) relates to the subheader of this post (which is also a correct statement of law):

    Of interest, it appears that the patentee was harmed here by the Patent Act’s objective approach to obviousness that, according to the statute, must be considered without regard to the actual “manner in which the invention was made.”

    This simply doesn’t seem to be in play here. The question in the case is a claim construction issue about whether a limitation which has no structural or functional bearing is given patentable weight. The answer to that is unquestionably no.

    Prof. Crouch seems to note as much, saying at 1.1:

    Of course, it was important here that the claim was directed to a product itself rather than being directed toward a method.

    But of course not really – while the discovery of a non-obvious problem + an obvious solution may result in a non-obvious invention, in this case the “discovery of a non-obvious problem” is an invalid act under Alice – it’s doing nothing more than reporting a fact of nature. Applying a conventional solution to a previously undiscovered natural fact doth not an invention make nowadays (see Ariosa v. Sequenom).

    This isn’t a case of poor claiming, this is a case where the subject matter could never have been reached. Patentability wasn’t negated by the manner in which the invention came about – there never was an invention because the art already knew how to solve a problem that was lying dormant in nature.

      1. No – the prior art was creating a non-obvious solution to a problem. The most that can be said for this patent was that it was finding a related non-obvious problem to which the solution was already known.

        You can generate a valid claim from finding a non-obvious problem in a man made system. You cannot have a valid claim from finding a non-obvious problem in nature, because that is exactly what the patentee in Ariosa did.

          1. I agree with anon on this one. The claim was directed to a chemical/product and such chemical/product was already known prior to filing the patent application.

          2. Yes anon, I said that. I then went on to say that Prof. Crouch suggested that claiming as a method (which would give weight to the 8a language) would lead to a different result. It would not.

            1. Perhaps – but importantly perhaps not.

              You do understand differences between method claims and article claims when it comes to the product by process doctrine, right?

          3. “The claim was to an object, already old in the art.” Not so — the “object” was oxycontin free (or substantially free) of a dangerous compound, which had never before been known or made. And there was no claim of anticipation as to these patents.

            “This is a case where the subject matter never could have been reached … because the art already knew how to solve a problem that was lying dormant in nature.” If the patent law won’t protect a new, safer drug (the subject matter), isn’t that indicative of a problem with the patent law itself? As for the “problem lying dormant,” that was the Eibel Process case distinguished away by the court.

      1. You and Random really need to try to stop h1-jacking everything to fit your drive-by monologues.

        What you want here is just not here.

        Move along.

          1. Do…?

            LOL – gee, we are on the internet, what would you like me to do?

            sniveling little cr e ep

            Oh Noes, what will the politeness police make of that?

            What a chump.

    1. Random: The question in the case is a claim construction issue about whether a limitation which has no structural or functional bearing is given patentable weight. The answer to that is unquestionably no.

      Wouldn’t it be nice if this were in the MPEP?

      1. It already is in the Claim as a Whole doctrine…

        (no mere aggregations – and one does not even need to have any claim elements in an aggregation to be of the sort of “mental step” for that doctrine to make Malcolm’s pet theory attempt mundane)

      2. Wouldn’t it be nice if this were in the MPEP?

        Indeed, the problem is not that non-functional descriptive material section is so minor.

        1. Random,

          What is the difference between “non-functional descriptive material“…

          functional descriptive material“…

          and

          functional material” (which said material is merely material that is in written form)?

          There is an easy Set Theory explication available.

  2. it was important here that the claim was directed to a product itself rather than being directed toward a method.

    Three thoughts on that:

    1) I don’t think the status (product, process, or product-by-process) would make a difference here given that that the recited hydrogenation step was obvious regardless of the choice of the source of the material to be dehydrogenated. The claims at issue were amended in an attempt to draft around a prior art process (“Chapman”) which described the claimed hydrogenation step but didn’t particularly identify one of the molecules (“8-alpha”) that was, in fact, hydrogenated.

    2) Whether they appear in system claims, process claims, or “composition” claims, computing devices that are described functionally are properly subjected to the same analysis that the court used here. The issue is: does the “source” limitation impart an objectively cognizable structural distinction to the claim vis a vis the prior art? If not, then the limitation is spurious (in the case of computing, it’s typically an abstract content descriptor). Note that the analysis is not determining whether it’s possible to imagine one structurally distinguishable species falling within the scope of the claim and then concluding “distinct”. It’s whether the claim scope provided by the source identifer includes one species that is structurally indistinct from the prior art, in which case the “limitation” ceases to be one.

    One way to look at the issue is to recognize whether any novel instructions must necessarily be written by the instruction writer when he/she is programming the computer. Does the computer care whether the number it’s going to cruch is “derived from Greenland ice melt data” or “derived from Super Bowl cheese fry consumption data”? Typically not.

    3) Buried in this opinion there is a brief discussion of a Hail Mary “limitation” which is utterly and transparently devoid of meaning, at least in the context of a chemical composition or formulation. Specifically I refer to Purdue’s recitation — and reliance upon! — the limitation in claim 1 of the ‘383 patent that recites “active ingredients with abuse potential (see footnote 3 of the decision). Thankfully Purdue’s gambit was swept aside here but … seriously?

    1. I find your analysis on #2 to be very wrong. Ok, so the patent relates “A to B.” B is known, or obvious, then any patent on the structure of B fails. But, even if B is known or obvious, that doesn’t mean that the “to”, the process”, cannot be independently patentable.

      Europe knew how to get to India around Africa for a long time. A new, shorter way (North West Passage) would have been valuable in and of itself.

      1. J so the patent relates “A to B.” B is known, or obvious, then any patent on the structure of B fails. But, even if B is known or obvious, that doesn’t mean that the “to”, the process”, cannot be independently patentable.

        I never stated otherwise, J. I’m talking about a more specific set of facts and you should feel free to address those.

        Europe knew how to get to India around Africa for a long time. A new, shorter way (North West Passage) would have been valuable in and of itself.

        Setting aside the general question of the patentability of “new routes” on the earth’s surface, the “limitations” I’m referring to are not remotely like “new routes” but more like “use a ship to get from Europe to [insert name of new city in India].” I presume you can see the problem with that, and I presume you can recognize the parallels in the computing arts. If not, let me know. Then it’s just a baby step to understand that it’s not just “ships” that are in the prior art. Not even close.

        1. You stated that the issue is:

          “does the ‘source’ limitation impart an objectively cognizable structural distinction to the claim vis a vis the prior art?”

          The structural distinction only matters if it is a claim on the structure. That is what I am saying.

          I simply used the Northwest passage as a metaphor, setting aside patent eligibility and looking at novelty and obviousness. I noticed you inserted “new city in India.” I’m saying the city can be old, or known, but if you have a route that is non-obvious to other routes, that process will pass 102 and 103. (Metaphorically speaking.)

          1. J The structural distinction only matters if it is a claim on the structure.

            Nope. The structural distinction also matters if the process acting upon the “source limited” object/information also happens to be in the prior art (as it was in this case).

            I simply used the Northwest passage as a metaphor

            Right. But you missed the point I was making. That’s why I took your metaphor and modified it — to help you understand the point I was making.

            As for your proposition that if you have a route that is non-obvious to other routes, that process will pass 102 and 103

            it seems more than a bit circular to say that, setting aside eligibility issues, “a non-obvious process will pass 103.” As I’ve tried to make clear, I’m not arguing otherwise (how could I?).

    2. ” The issue is: does the “source” limitation impart an objectively cognizable structural distinction to the claim vis a vis the prior art? If not, then the limitation is spurious (in the case of computing, it’s typically an abstract content descriptor).”

      Fine. Computer plus new software claims pass this test, even when described functionally, as the functional descriptions encompass a range of cognizable structural distinctions (i.e., bit states) that configure the computer to do the functions.

      1. There is some ambiguity that is relevant that makes you both right:

        Claiming software by function describes a definite-but-unposited number of bit states. It is “cognizable” under 112(b) but not 112(a). Further, I think the Supreme Court would find it invalid under 112(b) as well, because their conception of 112(b) looks to the spec to determine if the claim accurately tracks it. Disclosure of a single algorithm (or even multiple algorithms) is not even remotely coextensive with all algorithms that achieve a result.

    3. “One way to look at the issue is to recognize whether any novel instructions must necessarily be written by the instruction writer when he/she is programming the computer. Does the computer care whether the number it’s going to cruch is “derived from Greenland ice melt data” or “derived from Super Bowl cheese fry consumption data”? Typically not. ”

      Huh? Are you in Colorado? What are you smoking? Can you say non-sequitur?

      This almost makes sense: “One way to look at the issue is to recognize whether any novel instructions must necessarily be written by the instruction writer when he/she is programming the computer. ”

      This, has nothing to do with that:

      “Does the computer care whether the number it’s going to cruch is “derived from Greenland ice melt data” or “derived from Super Bowl cheese fry consumption data”? Typically not. ”

      And, oh by the way, computers don’t have feelings (yet) and never “care” where numbers come from.

      When computers do care where numbers come from, the method that they perform to achieve that emotion will be patentable.

      1. Les When computers do care where numbers come from, the method that they perform to achieve that emotion will be patentable.

        And you’ll be right there filing emotional distress claims on the computer’s behalf, I suppose.

        computers don’t have feelings … and never “care” where numbers come from.

        Thanks for the admission. I’m not sure why you put “care” in quotes, though. Are you confused about what the term means?

        functional descriptions encompass a range of cognizable structural distinctions

        That’s true of all functionalities and, as everyone knows, “new functionalities” are ineligible subject matter. Why? Because the “cognizable structures” you refer to must be set forth in the claim (or in the specification if you are lexicographing) in a manner which enables your new structure to be distinguished from all the zillions of structures in the prior art.

        Please alert me when someone describes their “do it on a computer” junk on the basis of the “bit states” that are “encompassed” by the desired functionality.

        1. I’m not sure why you put “care” in quotes, though. Are you confused about what the term means?

          Says the number one person of putting things in quotes….

          More C R P AccuseOthersOfThatWhichMalcolmDoes…

          Nice ecosystem Prof.

          (never mind the once again attempt to make an issue out of an optional claim format (as if it were the ONLY legal claim format – A G A I N)

      2. It’s not a non-sequitur. Concrete example:

        A invents a sorting algorithm which sorts numbers in decreasing value. A envisions said algorithm is used for sorting the test results of students in a class.

        B comes along and invents/claims the exact same algorithm A did, but B’s algorithm is used in the context of sorting the cost of automobiles.

        When B claims his algorithm there are going to be limitations about automobile costs in there. A obviously had no disclosure relating to automobiles.

        MM is simply stating that A anticipates B, not renders it obvious, but anticipates, because even though A has no disclosure related to automobile costs *from the standpoint of the computer* the algorithm is the same – both input values and sort the values. The human being operating the computer attaches a significance to the numbers which is irrelevant both structurally and functionally to the computer, and thus to the software engineer.

        Both algorithms are structurally the same, with identical Input (Set (x))’s. Therefore A’s disclosure anticipates B. B’s limitations directed to the source and meaning of the number values is, from the standpoint of the software coder, non-weighted intended use. It is a situation where drawing the proper art field is extremely relevant.

        1. RG: Both algorithms are structurally the same, with identical Input (Set (x))’s. Therefore A’s disclosure anticipates B. B’s limitations directed to the source and meaning of the number values is, from the standpoint of the software coder, non-weighted intended use.

          Yes. And this is a massive and increasingly transparent problem with the patent system as it’s presently set up. The USPTO (but not just the USPTO) has miserably failed to develop a robust system for finding, submitting and cataloguing the near infinite number of prior art algorithms and logical paradigms such that claims relying on “new” algorithms and “new” logic can be readily searched against them (contrast the PTO’s complete lack of effort in this regard with the manner in which claims to engineered nucleic acid and protein sequences were treated).

            1. “anon” That very same “wah” argument was made way back in 1968.

              Love it when the truth leaks out.

              Talk about the need to “grow up”…

              Indeed. When are you going to grow up?

                1. “anon” the same LAME arguments that did not work some nearly 50 years ago…

                  LOLOLOLOLOLOLOL

                  Meanwhile giant chunks of “do it on a computer” hide are falling by the curb every day.

                  You’re funny. Really, keep the laughs coming.

                  1. Meanwhile, Malcolm throws another non sequitur to this discussion and hopes no one is paying attention.

                    “Go figure Folks”

    4. MM: “The issue is: does the “source” limitation impart an objectively cognizable structural distinction…”

      That is it.

        1. Claim 1: Prior art machine, claimed in detail, where part A of the claimed machine is manufactured by X process.

          Now, if X process produces the same part A as any other part A, is the claimed machine new?

          1. Product by process is (still, and only): product.

            Malcolm’s spin here is to attempt to use his “objective physical structure” option as the only proper way to write a claim.

            It’s not.

            EVEN IF this claim was written in his preferred optional manner, the result would have been the same.

            ALL ELSE from him is just the empty clanging of his propaganda wagon.

  3. I was impressed by the CAFC’s acknowledgement that the “derived by” limitation failed to impart any relevant structural distinction to the claimed product, and likewise failed to make the proposed synthesis method any less obvious.

    Now all the CAFC and the PTO has to do is learn to ritually apply this basic reasoning to claims in the “do it on a computer” arts, where “data derived from [insert trendy source]” “limitations” are equally spurious but far, far more common.

    1. Malcolm,

      Your continued blight is truly v0mit worthy.

      You are absolutely wrong in your mantra here.

      Once again, Malcolm pretends that the optional claim format of “objective physical structure” is the per se ONLY legal structure allowed.

      That is NOT what this case is saying.

      Prof, your reply does not make that clear enough. And it’s not like just any old someone is “wrong on the Internet” – this is a known sAmeone trying to pass off the sAme known wrong version of law because that version matches his desired end.

      You do no one any favors with your “polite” answer.

Comments are closed.