Slipping on (olive) oil: key patents on olive-derived phenolic supplements declared invalid

Ever since studies suggested that the lower incidence of heart disease associated with the Mediterranean diet could be linked to the consumption of olive oil (for an overview of the health effects of other fats, see here), researchers have focused on the phenolic compounds contained in olives, which include oleuropein, tyrosol and hydroxytyrosol. Among the companies that first sought to commercialize olive extracts containing phenolic compounds, CreAgri was quick to seek patent protection, filing two patent applications in the US, in 2001 and 2003, subsequently granted as US Patents No. 6,416,808 and 8,216,599.

In 2011, CreAgri filed a lawsuit against Pinnaclife, a competitor selling olive-derived supplements, alleging infringement of both the '808 and '599 patent. The defendant sought declaratory judgments of non-infringement and invalidity of the asserted patents. In a judgment rendered on 18 December 2013, the District Court for the Northern District of California found both patents to be invalid. The decision examines several interesting issues (ranging from claim interpretation, overlaps between reexamination proceedings and litigation, prophetic claims, written description requirement, etc.), and deserves an in-depth look. Before reviewing it, let me add that a great write-up of the case by Virginia Brown Keyder provided inspiration for this post. Virginia also wrote a nice article on olive oil legislation in the US and EU here.

The court first turned its attention to the '808 patent, which essentially claims dietary supplements comprising an aqueous extract of olives containing hydroxytyrosol and oleuropein, or hydroxytyrosol and tyrosol, at certain weight ratios. The defendant alleged, inter alia, invalidity due to anticipation, under 35 USC § 102, in light of two prior art references, the US Patent No. 6,358,542 ('Cuomo') and an article ('Romani') that disclosed the polyphenolic content of five different cultivars of olive fruits from Tuscany. Pinnaclife also observed that the US Patent and Trademark Office, in a recent ex parte reexamination, had rejected all the claims of the '808 patent, as anticipated by Cuomo and Romani, or obvious in light of these references. The court, however, refused to take into account the findings of the USPTO, noting that (i) while a patent is presumed valid in litigation, ex 35 USC § 282, there is no such presumption in reexamination actions, to the effect that '[t]he difference in standard of proof significantly reduces the relevance of the PTO's conclusions', and (ii) the PTO's decision was still potentially subject to substantial review before the Patent Trial and Appeal Board and the Federal Circuit.

Judge Koh highlighted that the plaintiff had not disputed that the references disclosed the weight-ratio changes indicated in the asserted patent. Thus, the only issue concerned the disclosure, in prior art, of an 'aqueous extract of olive'. The claim construction order, previously issued by the court, had defined the aqueous extract as 'an aqueous solution containing a water-soluble preparation from an olive plant', with no restriction on the process by which the solution is obtained. In this perspective, the judge noted that CreAgri's late attempt to restrict the notion of 'aqueous extract' to pure water contrasted with the arguments raised during claim construction, and failed on the merits, as the specification of the '808 patent clearly refer to an 'aqueous or aqueous-alcoholic extract', when describing the invention 'as a whole'. The court held that 'the specification's use of the disjunctive is ambiguous at best', and is insufficient to exclude aqueous-alcoholic extracts from the scope of the invention. A different conclusion, added the court, would improperly import a purported limitation from the specification into the claims.

Analysing the references, the judge held that the 'aqueous extract' was clearly disclosed by Cuomo's Examples 4 and 11, which jointly described a method for obtaining an aqueous methanol solution, in which a composition, obtained from olive pulp, is dissolved. The same conclusion was reached with regard to the Romani reference, which disclosed the use of an 'aqueous extract', as an intermediate step of the extraction process. Thus, the court found all the claims of the '808 patent to be invalid due to anticipation.

The second part of Judge Koh's decision examined the validity of the '599 patent, which claims methods of treating a number of inflammatory conditions, through the use of various forms of hydroxytyrosol, solely or in conjunction with oleuropein. Pinnaclife maintained that the specification of the contested patent did not provide any relevant data to support the anti-inflammatory effects of the olive-derived preparations, thus failing to meet the requirements of written description, enablement, and utility.

As taught by the Federal Circuit in Ariad Pharmaceutical et al. v Eli Lilly & Co., the written description test describes 'an objective inquiry into the four corners of the specification from the perspective of a person of ordinary skill in the art', to determine whether the specification shows that 'the inventor actually invented the invention claimed'. Judge Koh added that claims that have a chemical nature, or whose scope is broad, as in the case of the '599 patent, frequently require more supporting details to meet the written description requirement. She argued that the specification of the contested patent 'conclusively reveals that the inventor sought to claim a method of treating inflammation based on no more than a hope that olive-derived compositions would one day be used effectively to treat inflammation caused by a wide variety of factors'. The court also recited the testimony of the inventor:

(A) I learned early in my career that you can file patents, they are prophetic...
(Q) You anticipate that this would work?
(A) Right.
(Q) But you don’t actually have the data yet showing that it would work?
(A) Yeah...

Commenting these words, Judge Koh pointed again to Ariad, observing that 'the prophetic approach to patenting is not proper absent some indication that the named inventor has [...] performed the difficult work of invention', as a mere wish or plan is insufficient to meet the written description requirement. Similarly, while substantial prophetic examples can be sufficient to that aim, under Ariad, 'mere mentions of a desired outcome' are not. Evaluating the studies cited in the specification, the court found that they failed to provide any statistically significant results, realized or predicted, capable of supporting the claims of the '599 patent.

CreAgri argued that the anti-inflammatory properties of olive phenols were already known by others skilled in the art, pointing to an inherent disclosure, but the court dismissed this argument. Judge Koh noted that the written description requirement can only be satisfied by an inherent disclosure if 'the missing descriptive matter [is] present in the application's specification such that one skilled in the art would recognize such a disclosure' (PowerOasis v T-Mobile USA). Examining the specification of the contested patent, she concluded that the disclosure of a number of biomarkers, employed to determine whether the treatment has a therapeutic effect, is insufficient to create an inherent disclosure of the full scope of the invention: 'the ’599 Patent inherently discloses how to measure whether the treatment works as claimed, not that the treatment does work as claimed'. Further, none of the prior art references cited in the specification demonstrated the efficacy of the claimed olive-derived phenols for treating the inflammatory conditions described in the patent. Accordingly, the judge found all the claims of the '599 patent to be invalid for failure to meet the written description requirement of 35 USC § 112.

Assessing enablement and utility, the court concluded that, at the time of the patent filing, there was no scientific evidence demonstrating the link between olive-derived phenols and the claimed health benefits, to the effect that a person of ordinary skill in the art would have questioned the utility of the invention. The judge refused to consider post-filing articles submitted by CreAgri (but observed that, even when taken into account, said articles would not alter the court's findings on utility), discussing the Federal Circuit's holding in In re Brana:

[T]he Court cannot consider CreAgri’s post-filing test results as evidence of the utility of the claimed methods of treating inflammation. "Enablement is determined as of the effective filing date of the patent's application." [In re ’318 Patent Infringement Litigation]. Where results "were not available at the time of the application," they cannot be used to establish utility or enablement. [Id.]. The Federal Circuit has created a narrow exception to the rule that post-filing data cannot support utility. In In re Brana, the Federal Circuit allowed such evidence "to substantiate any doubts as to the asserted utility" where those test results "pertain [...] to the accuracy of a statement already in the specification." [Id.]. Read too broadly, however, the Brana exception would swallow the rule that "[e]nablement, or utility, is determined as of the application filing date." [Id.]. Where actual results, garnered post-filing, mirror or otherwise substantiate predicted results, it is plain that those results will pertain to the accuracy of a statement in the specification within the meaning of Brana. Here, however, the ’599 Patent makes no assertions whatsoever regarding the outcomes of the proposed studies [...], so the study designs provided in the specification are not sufficiently prophetic such that later-achieved results can support the utility of the claimed invention.

Finally, the court evaluated the existence of any data or reasoning capable of supporting the inventor's claims on the therapeutic efficacy of the olive-derived phenols, looking for a 'reasonable correlation' between the activity in question and the asserted utility. Judge Koh noted that, when 'the inventor did not know at the time of filing whether the invention was in fact operable and instead rests the invention’s asserted operability on [...] prophe[cy]', analytic reasoning may come to aid, but only if it is sufficient to demonstrate how and why the invention is expected to work. However, she could not identify, in the specification of the '599 patent, any valid argument or analytic reasoning establishing utility. Observing that the patent merely claims a research hypothesis, the court confirmed its invalidity, for failure to satisfy the enablement and utility requirements.