PGR decision provides much-needed clarity in fast-growing microbiome space

In a first-of-its-kind decision that finally provides some measure of the strength of broad US patents issued for microbiome therapeutics, the US Patent Trial and Appeal Board (PTAB) issued a ruling on 15th April in the post-grant review (PGR) of the University of Chicago’s US Patent No 9,855,302.

While all claims of the patent were ruled invalid, the basis of the decision provides some positive notes for those with issued microbiome patents, as well as some guidance for patent applicants in this space.

The claims of the patent were based on the discovery that the makeup of the gut microbiome influences patients’ responses to immune checkpoint inhibitor immunotherapy. Claim 1 of the ‘302 patent recited:

“A method of treating cancer in a human subject comprising co-administering to the subject an immune checkpoint inhibitor and a bacterial formulation comprising bacteria of the genus Bifidobacterium.”

Immune checkpoint inhibitors are a hot area in cancer therapy that act to stimulate the patient’s own immune system to attack tumors. The approach works remarkably well in some patients, but poorly in others, so the discovery that the presence of certain bacteria in the patient’s gut can increase the therapeutic efficacy of checkpoint inhibitors is a potential game changer.

The patent, which was exclusively licensed to Evelo Biosciences, based in Cambridge, Massachusetts, was challenged by petitioner Genome & Company, a Korean probiotics company engaged in a clinical trial collaboration with Merck KGaA and Pfizer, Inc. to test a single-strain bacterial formulation for efficacy in combination with the antibody checkpoint inhibitor avelumab in multiple cancer indications.

PGR permits validity challenge within nine months of patent issuance on a range of grounds including both prior art and sufficiency of disclosure. A panel of PTAB judges hears and rules on the challenges.

The procedure is not often used because, if unsuccessful, the petitioner can be estopped from raising any issue that was raised or reasonably could have been raised in the PGR as a defence in a later infringement suit. However, in this case, the petitioner argued that the claims of the ‘302 patent are invalid for lack of enablement over their full scope, and for obviousness over the prior art.

Regarding enablement, the petitioner argued that the claims are overbroad for encompassing the use of any checkpoint inhibitor, in combination with any bacterium of the genus Bifidobacterium, to treat any cancer. Petitioner argued that only one inhibitor, an antibody against PD-L1, was demonstrated to work in the patent, and that efficacy was only shown against two types of cancer in an animal model. Regarding the bacteria, petitioner argued that the genus Bifidobacterium encompasses a large number of different species and strains that may or may not enhance the efficacy of checkpoint inhibitor therapy.

While the board essentially agreed with the petitioner on enablement, the good news for microbiome innovators and investors is that the ruling hinged not on the breadth of the bacteria recited, but on overbreadth of the recited checkpoint inhibitors.

The board emphasized that the patent disclosure describes checkpoint inhibitors broadly as including antibodies as well as nucleic acid inhibitors, such as RNA interference molecules, but that only a single antibody inhibitor of a single checkpoint protein was shown to work. Only two of the narrowest dependent claims that recite specific antibody checkpoint inhibitors escaped invalidity for lack of enablement.

With regard to the cancers and bacteria, the board found that if sufficient guidance were provided regarding the checkpoint inhibitors, one of ordinary skill in the art could select an inhibitor appropriate for a given cancer, and, of greatest interest for other microbiome therapeutic patents, that the specification not only provided working examples of several species of Bifidobacteria, but also provided description to guide the skilled artisan in determining whether any given species or strain of Bifidobacterium would work.

Regarding obviousness, the petitioner presented a prior art reference that describes the use of an anti-PD-L1 antibody to treat colon cancer and a reference that describes Bifidobacterium longum as inhibiting the growth of colon cancer. They argued that it would be obvious to combine the treatments to improve the treatment of colon cancer. The board agreed. Perhaps the most instructive aspect of the obviousness ruling is that the board rejected the patent owner’s argument that it would not have been obvious to combine the references because the B. longum reference did not teach that the bacteria were immunostimulatory.

The board commented specifically that the claims did not require that the bacteria be immunostimulatory, and that motivation to combine was instead based on both references teaching an effect against the same kind of cancer. The value of functional claim language is often downplayed in the U.S., but where the activities of even known bacteria are only now being determined, patent applicants in the microbiome space may do well to consider including functional language that may help insulate the claims from an obviousness challenge.

The PTAB’s analysis delivers some good news for microbiome innovators and investors that are worth noting. The validity of each patent turns on its own facts, and the PTAB’s decision is particular to this patent and not binding on courts, but this decision, together with other important challenges, such as the European Opposition Division’s 2019 decision on Vedanta’s patent EP2575835, is bringing into focus what is and is not patentable, providing the microbiome community and investors with increasing certainty regarding what they can and cannot protect.

A number of clinical trials for microbiome-related therapeutics are underway, and the next patent challenges in this arena are likely to come in the form of defenses to infringement litigation as products gain regulatory approval and hit the market.