Expectation of Success – How Much is Reasonably Needed for Obviousness

https://depositphotos.com/14517523/stock-photo-how-much-concept.htmlThis is an important question, but one that is not always satisfactorily answered.  There are times when it may not be clear what the hypothetical person of ordinary skill in the art would have taken away from the prior art at the time of the invention.  In such cases, it is not possible to divine with much confidence whether the ordinarily skilled person would have reasonably expected to arrive at the claimed invention.  UCB, Inc. v. Accord Healthcare (“UCB”), decided by the Federal Circuit in May 2018, fits this scenario.  In UCB, the majority found UCB, Inc.’s asserted claim nonobvious under non-statutory double patenting.  The dissent, however, found overwhelming evidence in support of obviousness.  The majority determined that prior art teaching was insufficient for a skilled artisan to have a reasonable expectation of success.  To the dissent, however, the majority’s analysis was flawed because it ignored the fact that the law required only a reasonable expectation of success, not a guarantee.  One might say the majority viewed the glass as half full and the dissent half empty.

The invention at issue in UCB was a small molecule pharmaceutical, and obviousness turned on the likelihood of success of placing two desired substituents at specific positions in the molecule relative to a substituent at a third position.  The case arose in the Hatch Waxman context.  UCB, Inc. and three other companies (collectively, “UCB”) own and/or license U.S. Patent No. RE38,551 (the ’551 patent), which covers lacosamide, an anti-epileptic drug marketed under the tradename Vimpat®.  The effective filing date of the ’551 patent is March 1996.  A group of generic drug manufacturers including Accord Healthcare, Inc., (collectively “Accord”) sought FDA approval to market generic versions of Vimpat®, certifying that the ’551 patent was invalid, unenforceable, or not infringed by the proposed generic products.  In response, UCB sued Accord for patent infringement in a United States District Court.  The court held the asserted claims not invalid.  Accord appealed, arguing that the district court had misapplied legal standards.  This post discusses only the issue of obviousness-type double patenting, one of the three patentability issues addressed by the court.


The ’551 patent discloses and claims lacosamide which belongs to a class of compounds known as functionalized amino acids (“FAAs”).  The development of FAAs as anticonvulsants began in the 1980s with the work of Dr. Kohn, the inventor named in the ’551 patent.  The work led to the issuance of U.S. Patent Nos. 5,378,729 (the ’729 patent) and 5,654,301 (the ’301 patent) to Dr. Kohn.  Unlike the ’729 patent, the ’301 patent is not prior art to the ’551 patent.  However, claims 44 and 45 of the ’301 patent were alleged by Accord to render asserted claim 9 of the ’551 patent obvious under obviousness-type double patenting.

FAAs have the general structure shown below (left), in which R, R1, and R3 are variable.  Claim 45 defines a genus in which R3 is fixed as methoxymethyl whereas R, and R1 positions can have different groups, including, respectively, unsubstituted benzyl and unsubstituted methyl (middle).  Lacosamide (at least 90% pure R enantiomer) is a species of this genus, having a methoxymethyl group at R3, unsubstituted benzyl at R, and unsubstituted methyl at R1.  Asserted claim 9 of the ’551 patent is directed to lacosamide (right).

Double patenting analysis requires determining the differences between the compounds claimed in the reference and asserted patents and then determining whether those differences render the claims patentably distinct.  The analysis requires identifying some reason that would have led a chemist to modify the earlier compound to make the later compound with a reasonable expectation of success.  In UCB, it required determining whether an ordinarily skilled artisan, starting with claim 45 of the ’301 patent (methoxymethyl at R3), would have been motivated to place an unsubstituted benzyl at R and an unsubstituted methyl at R1 with a reasonable expectation of success.  The district court determined that the artisan would not have had a reasonable expectation of success because of a lack of data showing the effect of placing these two substituents at their respective positions.

The Federal Circuit agreed with the district court.  Out of all the work performed by Dr. Kohn and others, the district court had found that not a single reference disclosed any anticonvulsant data for any compound comprising a methoxymethyl group at R3.  Majority Op. at 20.  Further, most of the FAAs studied prior to the priority date experimented only with modifying the substituents at the R3 position while holding constant unsubstituted benzyl at R and unsubstituted methyl at R1.  Id.  As such, the Federal Circuit found the district court’s conclusion reasonable that the prior art data on FAAs did not provide sufficient insight into the effectiveness of placing benzyl and methyl at R and R1, respectively, relative to other substituents that could have been placed at those positions.  The Federal Circuit also took note of the finding that data disclosed in the ’729 patent showed that compounds having benzyl and methyl, respectively, at R and R1, but different groups at R3, had varying effectiveness, and of the expert testimony that prior art was silent as to what role benzyl and methyl played in the activity of the FAAs.  Id.

In UCB, prior art included the thesis of Philippe LeGall (“LeGall”).  The thesis disclosed 15 new FAAs, all having a benzyl group at R and a methyl group at R1.  These included compound 107e, the racemate of lacosamide.  No anticonvulsant efficacy data for compound 107e was disclosed, however.  Compound 107e, with methoxymethyl group at the R3 position, belongs to a class of compounds called “polar analogues”, all of which contain a nonaromatic group at the R3 position.  As a class, these compounds showed little to no potency.  However, LeGall speculated that because of structural similarities to another compound (having OCH2CH3 at R), which was more potent than the polar analogues, compound 107e may have good anticonvulsant activity.  LeGall concluded that the most active compounds studied had heteroaromatic groups in the R3 position (unlike107e, which had a nonaromatic group).  The district court acknowledged LeGall’s speculation regarding compound 107e, but considered it together with the expert testimony that an ordinarily skilled person looking to LeGall would have had no interest in pursuing compound 107e and would not have had a reasonable likelihood of success in pursuing an FAA with a methoxymethyl group at R3 as an anticonvulsant drug.

The dissent came to the opposite conclusion based on an extensive review of facts.  It found that not only many tests had been conducted on FAAs with benzyl at R and methyl at R1, but that 75% of Dr. Kohn’s experimental compounds contained benzyl at R and methyl at R1, most of which were unsubstituted.  Dissenting Op. at 5.  The first FAA compound (AAB) having anticonvulsant activity contained a benzyl at R, and a methyl at R1 (Dr. Kohn, 1985).  Id.  Sixteen structural AAB analogues disclosed two years later by Dr. Kohn used unsubstituted benzyl at R and unsubstituted methyl at R1 as a “reference point,” while testing analogues having modifications of these groups.  Id.  Among these, only one of the five R modifications showed activity comparable to unsubstituted benzyl at R; the others showed decreased activity.  As to R1, each of the three modifications decreased anticonvulsant activity relative to unsubstituted methyl group.  Id. at 9.  A 1990 paper by Dr. Kohn disclosed compounds in which R1 and R were kept constant as methyl and benzyl, respectively, and the most potent compound in the group, 2g (with an aromatic 2-furanyl structure at R3), was significantly more potent than APB (having phenyl as R3), at the time the most potent compound in the FAA family.  Id. at 6.  In a summary of his FAA work (published 1991), Dr. Kohn reported twenty-six compounds, all having unsubstituted benzyl at R and unsubstituted methyl at R1, with different groups at R3, and explained that one gets potent protection if there is a benzyl on one end (at R) and a methyl on the other (at R1).  Id.  The summary also identified a compound 3l which had the best activity till date for any FAA racemate and a methoxyamino at R3 (differing from lacosamide only by the replacement of one carbon by nitrogen in R3).  Id.  Also, a 1993 paper published by Dr. Kohn described compounds in which the 2-furanyl group at R3 was switched with other heteroaromatic groups, but all compounds had benzyl and methyl at R and R1 respectively.  Id.

Thus, to the dissent, Dr. Kohn’s extensive study of FAAs provided copious amounts of information from which an ordinarily skilled person would have formed a reasonable expectation that the selection of an unsubstituted benzyl and methyl for R and R1, respectively, would lead to the successful creation of an FAA with anticonvulsant activity.  The dissent acknowledged that the prior art also described a fluoro-substituted benzyl at R that yielded a comparable anticonvulsant effect but provided a far superior protective index (i.e. was safer), indicating that fluoro-substituted benzyl may be preferred at R.  This result notwithstanding, the dissent emphasized that the tests overall provided strong evidence that an ordinarily skilled person would have had a reasonable expectation of success in selecting unsubstituted benzyl and methyl for R and R1 positions, respectively, to create an FAA having an anticonvulsant effect.  Only a reasonable expectation of success, was needed, not a guarantee, the dissent observed citing Pfizer, Inc. v. Apotex, Inc., 480 F.3d (Fed. Cir. 2007) at 1364.

The dissent made another interesting point.  If, as the district court had found, all testing had focused on R3 and a person of ordinary skill would have attributed anticonvulsant behavior to R3, once R3 was fixed in the ’301 reference patent genus, plugging in unsubstituted benzyl at R and unsubstituted methyl at R1 (which had remained largely constant throughout the prior art testing) would have been a trivial selection.  Dissenting Op. at 8.

Finally, the dissent pointed to the importance of the disclosure in the LeGall thesis of compound 107e, a racemate, which has the same substituents at all three positions as lacosamide, and the hypothesis (which it found reasonable) that, based on structural similarities to another compound, 107e may have good anticonvulsant activity.  Only a reasonable expectation of success was needed, not a guarantee, the dissent reminded, again citing Pfizer, 480 F.3d at 1364.

In UCB, the district court recognized that many tests had been conducted on FAAs with benzyl at R and methyl at R1, but dismissed the resulting data because most of these tests kept the structures at R and R1 constant in order to be able to attribute the observed changes in anticonvulsant behavior and/or neurotoxicity to the structure at R3 rather than to the benzyl at R or the methyl at R1.  This dismissal of data was clear error, the dissent observed.  Dissenting Op. at 7.  “Where the prior art teaches that the selected substituent will work, even when it is selected from thousands of compounds, an inability to predict how any one substituent will work in the composition and a need for testing will not render that selection nonobvious,” the dissent explained, citing In re Corkill, 771 F.2d 1496, 1500 (Fed. Cir. 1985) (“Although [the inventor] declared that it cannot be predicted how any candidate will work in a detergent composition, but that it must be tested, this does not overcome [the prior art’s] teaching that hydrated zeolites will work.”).  Id. at 8.

As UCB shows, an expectation of success analysis can be highly fact-dependent and result unpredictable because of each side’s distinct view of what the facts teach.  Regardless of the outcome in UCB, testing to determine how any candidate will work in the makeup of a claimed compound or composition is not necessary where there is data from other sources that the candidate will work (In re Corkill, 1985), since the law only requires a reasonable expectation of success, not a guarantee (Pfizer, Inc. v. Apotex, Inc., 2007).


Image Source: Deposit Photos


Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Join the Discussion

2 comments so far.

  • [Avatar for Anon]
    July 26, 2018 12:09 pm

    Flash of Genius was expressly disavowed by Congress when they constrained the Supreme Court in the Act of 1952.

    Thus, any argument that starts with some type of dependence on that concept is suspect at best.

  • [Avatar for Peter Kramer]
    Peter Kramer
    July 26, 2018 08:27 am

    When the flash of genius occurs and an invention is conceived, there is ALWAYS an expectation of success. Otherwise the inventor would never shout, “EUREKA!” Of course, the question is whether PHOSITA would have had an expectation of success. The decision whether the invention was obvious or not therefore revolves around the horribly subjective question of whether the inventive insight was a product of PHOSITA or some higher talent.