In OSI v. Apotex, Appeal no. 2018-1925 (Fed. Cir., October 4, 2019), the panel reversed the PTAB and found that the method of treatment claims in U.S. Pat. No. 6,900,221 were not obvious over a primary reference taken with each of two secondary references. The claims were to a method of treating non-small cell lung cancer by administering to a mammal a therapeutically effective amount of erlotinib (Tarceva®). The panel found that the cited combinations of references, Schnur in view of Gibbs or OSI’s Form 10-K as it was presented in its 10-K submission, would not provide a reasonable expectation of success in treating NSCLC to the POSA.
The panel empathized that NSLC has, historically, been very difficult to treat, and that drug candidates have suffered a high failure rate. In view of these facts, and in the absence of any in vitro data, animal data or clinical (human) data, the panel found that the PTAB’s obviousness holding was not supported by substantial evidence.
Schnur discloses a laundry list of compounds, including erlotinib – a preferred compound – asserted to be useful to treat a laundry list of cancers, including “lung cancer” but did not mention NSCLC. Gibbs is a review article that discloses that erlotinib and another compound are “currently in clinical trials….However these [two] compounds appear to have good anticancer activities in preclinical models, with an acceptable therapeutic index, particularly in patients with [NSCLC].” Without considering anticipation, the panel drilled down into the art cited in this paragraph and found no data disclosing the use of erlotinib to treat NSCLC. OSI’s 10-K was also dismissed due to its lack of a specific disclosure of any test data showing that erlotinib had utility against NSCLC. But let’s look at the 10-K:
“[Erlotinib, which targets a variety of cancers including ovarian, pancreatic, non-small cell lung cancer and head and neck, achieved a significant milestone with the completion of Phase I safety trials and the initiation of Phase II clinic trials in the [U.S.] in cancer patients. [Erlotinib] is a potent, selective and orally active inhibitor of the [EDGFR], a key oncogene in these cancers.”
The panel concluded: “These references provide no more than hope—and hope that a potentially promising drug will treat a particular cancer is not enough to create a reasonable expectation of success in a highly unpredictable art as this….To be clear, we do not hold today that efficacy data is always required for a reasonable expectation of success. Nor are we requiring absolute predictability of success.”
This is certainly the case in an obviousness analysis, but what if the rejections had been based on the 10-K Form and the statement in the Gibbs review as anticipatory prior art. This thought reminded me of a Fed. Cir. case, Rasmusson v. SmithKline Beecham Corp., Appeals nos. 04-1191, 04-1192 (Fed. Cir., June 27, 2005). This appeal arose in an interference context in which Rasmusson et al. lost their priority date for a claim directed to a method of treating prostatic carcinoma with finasteride. The panel found that the application necessary to establish priority under s. 112 was non-enabling in not teaching the art how to use the claimed method. In other words, as of the filing date in question, the panel affirmed the Board’s finding that the POSA would simply not have believed that finasteride or any other selective 5aR inhibitor was effective in treating prostate cancer. The specification was devoid of any experimental data. As the panel summarized its affirmance:
“If mere plausibility [of efficacy] were the test from enablement under s. 112, applicants could obtain patent rights to ‘inventions’ consisting of little more than respectable guesses as to the likelihood of their success. When one of the guesses later proved true, the ‘inventor’ would be rewarded the spoils instead of the party who demonstrated that the method actually worked.”
This actually sounds something like an abstract idea analysis, but let’s get back to anticipation. Rasmusson et al. had lost their priority date, but they had an arrow left in their quiver. A counterpart EU application had been published more than one year prior to SmithKline’s filing date, and Rasmusson et al. argued that it anticipated the SmithKline patent in question. You may think that SmithKline had a slam dunk on this question, but you would be wrong. In a sentence, it is because there is no requirement that a piece of art teach “how to use” the claimed invention to be anticipatory. Here is how the panel summarized the law:
“In re Hafner, 410 F.2d 1403 (CCPA 1969), the court stated that ‘a disclosure lacking a teaching of how to use a fully disclosed compound for a specific, substantial utility…is, under the present state of the law, entirely adequate to anticipate a claim to either the product or the process and, at the same time, entirely inadequate to support the allowance of such a claim….The reason is that s. 112 provides that the specification must enable [the POSA] to ‘use’ the invention whereas 102 makes no such requirement as to an anticipatory disclosure.’”
It’s getting late in a dreary day, but have I just written that a patent can be anticipated by a prior art reference but still be found unobvious in view of that reference and a secondary reference. From a practical standpoint, if the PTAB had found the claims to be anticipated, they would not have considered obviousness, and the reasonable expectation of success question would be moot.