Humira Patent Invalid for Obviousness Type Double Patenting

In AbbVie Inc. v. Kennedy Institute of Rheumatology Trust, the Federal Circuit affirmed the district court’s finding that a second patent covering  AbbVie’s Humira product is invalid under the doctrine of obviousness-type double patenting. The decision reaffirms the court’s stance in Gilead that the doctrine remains valid under the 20-year patent term system, and serves an important public policy goal of ensuring that once a patent protecting an invention has expired, the public is free to practice that invention and its obvious variants. The court may have gone too far, however, in relying on the disclosure of the reference patent to support its obviousness determination.

The Patent at Issue

The patent at issue was Kennedy’s U.S. 7,846,442, directed to “Methods of treating rheumatoid arthritis with … anti-TNF-alpha antibodies and methotrexate.” The Federal Circuit cited claim 1 as representative:

1. A method of treating an individual suffering from rheumatoid arthritis whose active disease is incompletely controlled despite already receiving methotrexate comprising adjunctively administering with methotrexate therapy a different composition comprising an anti-human tumor necrosis factor-α antibody or a human tumor necrosis factor-α binding fragment thereof to the individual, wherein the anti-human tumor necrosis factor-α antibody or fragment thereof (a) binds to an epitope on human tumor necrosis factor-α, (b) inhibits binding of human tumor necrosis factor-α to human tumor necrosis factor-α cell surface receptors and (c) is administered at a dosage of 0.01-100 mg/kg, and wherein such administration reduces or eliminates signs and symptoms associated with rheumatoid arthritis.

The ’442 patent was filed September 12, 2005, with a priority claim to August 1, 1996. It was granted in 2010, and was awarded 750 days of Patent Term Adjustment, leading to an expiration date of August 21, 2018.

The earlier patent was the grandparent of the ’442 patent, Kennedy’s U.S. 7,846,422, directed to “Anti-TNF antibodies and methotrexate in the treatment of arthritis and Crohn’s disease.” The Federal Circuit cited claim 8 as representative:

8. A method of treating rheumatoid arthritis in an individual in need thereof comprising co-administering methotrexate and an anti-tumor necrosis factor alpha antibody or an antigen-binding fragment thereof to the individual, in therapeutically effective amounts.

The ’442 patent is a continuation of the ’766 patent (filed August 1, 1996), but the ’766 patent included an earlier priority claim to October 8, 1992, that was not maintained in the ’442 patent. The ’766 patent was granted in 2001 and expired at the end of its 20-year term, on October 8, 2012.

The District Court Proceedings

As summarized in the Federal Circuit decision, AbbVie licensed the ’766  patent in 2002, and obtained FDA approval to market Humira (an anti-TNFα antibody) for use alone or with methotrexate in the treatment of rheumatoid arthritis. When the ’442 patent was issued, AbbVie did not want to “pay further royalties for the right to sell the same product.”

AbbVie brought a declaratory judgment action challenging the validity of the ’442 patent and the district court held the asserted claims invalid under the doctrine of obviousness-type double patenting. Kennedy acknowledged that the patents “encompass the same inventive subject matter,” but argued that the ’442 patent was patentable over the ’766 patent, because it was directed to a “narrower species” and associated with “unexpected results.”

The Federal Circuit Decision

The Federal Circuit decision was authored by Judge Dyk and joined by Judges Wallach and Chen.

Judge Dyk begins his analysis with a review of the doctrine of obviousness-type double patenting. Judge Dyk notes, “While often described as a court-created doctrine, obviousness-type double patenting is grounded in the text of the Patent Act. …. Thus, § 101 forbids an individual from obtaining more than one patent on the same invention, i.e., double patenting.”

Reviewing Supreme Court cases from the 1800′s, Judge Dyk concludes:

The ban on double patenting ensures that the public gets the benefit of the invention after the original period of monopoly expires.

Rejecting Kennedy’s arguments that the main purpose of the doctrine–curbing abuses of submarine patents–is no longer necessary under the current 20-year term system, Judge Dyk states:

[T]his argument ignores another crucial purpose of the doctrine: It is designed to prevent an inventor from securing a second, later expiring patent for the same invention. …. That problem still exists. Patents claiming overlapping subject matter that were filed at the same time still can have different patent terms due to examination delays at the PTO. See 35 U.S.C. § 154(b) (patent term adjustments) …. So too where, as here, the applicant chooses to file separate applications for overlapping subject matter and to claim different priority dates for the applications, the separate patents will have different expiration dates since the patent term is measured from the claimed priority date. When such situations arise, the doctrine of obviousness-type double patenting ensures that a particular invention (and obvious variants thereof) does not receive an undue patent term extension.

Turning to the merits, the court compared the claims of the patents, and determined that the main difference might relate to the recited patient populations. In particular, the Federal Circuit accepted for the sake of its analysis Kennedy’s arguments that individuals suffering from “active disease” as recited in the ’442 patent were a specific subset of patients in need of treatment of rheumatoid arthritis.

Thus, assuming Kennedy’s construction of the term “active disease” is correct, we must decide whether a patent that claims to treat a subset of patients with more severe rheumatoid arthritis (the ’442 patent) is an obvious variant of a patent that claims treatment of rheumatoid arthritis patients generally (the ’766 patent).

After summarizing the law of genus-species obviousness, Judge Dyk states:

Thus, species are unpatentable when prior art disclosures describe the genus containing those species such that a person of ordinary skill in the art would be able to envision every member of the class. Here, we think it is clear that a reader of the ’766 patent could have easily envisioned a species limited to sicker patients.

The Federal Circuit considered but rejected Kennedy’s arguments that the claimed methods achieved unexpected results:

To determine whether the ’442 patent is directed to a species that yielded unexpected results, we must necessarily look to the ’766 patent’s disclosures to assess what results were expected at the time the ’766 patent application was filed. The  demonstration of utility of the ’766 patent relies on the T-14 study, the very study that Kennedy now relies on to show that the ’442 patent led to unexpected results and merits a separate patent.

Kennedy argued that the study reported in the ’766 patent could not be used in the obviousness-type double patenting context, but the Federal Circuit disagreed:

It is true that a reference patent’s specification cannot be used as prior art in an obviousness-type double patenting analysis. …. But, it is also well settled that we may look to a reference patent’s disclosures of utility to determine the question of obviousness. …. [I]n In re Basell Poliolefine Italia S.P.A., we clarified that “[w]hile . . . it is impermissible to treat a ‘patent disclosure as though it were prior art’ in a double patenting inquiry, . . . the disclosure may be used . . . to answer the question whether claims merely define an obvious variation of what is earlier disclosed and claimed.” ….

There is no meaningful distinction between examining the disclosed utility of an earlier patent to determine the overall question of obviousness and looking at the disclosed utility of an earlier patent to determine whether the utility of the later patent was unexpected at the time of the earlier patent. Neither involves improper use of the reference patent’s specification as prior art.

Thus, the Federal Circuit affirmed the district court’s finding that the ’442 patent is invalid for obviousness-type double patenting.

Should the Court Have Relied on the ’766 Patent’s Specification?

The extent to which the Federal Circuit relied on the specification of the ’766 patent in reaching its decision raises concerns. While it may have had to review the specification in order to resolve the claim construction issues, the court’s own analysis seems to show that it went too far in relying on the specification to resolve the obviousness issue.

First, the court answered the threshold genus-species statement with reference to what “a reader of the ’766 patent” would have “easily envisioned.” Would it have reached the same conclusion based only on the claims of the ’766 patent?

Then, the court went even further with its “unexpected results” analysis:

To determine whether the ’442 patent is directed to a species that yielded unexpected results, we must necessarily look to the ’766 patent’s disclosures to assess what results were expected at the time the ’766 patent application was filed.

If the Federal Circuit were going to be true to the principle that “a reference patent’s specification cannot be used as prior art in an obviousness-type double patenting analysis,” shouldn’t it have looked to evidence other than the ’766 patent to determine “what results were expected”?

In this regard, I question whether the “utility” cases cited by the Federal Circuit justify its resort to the specification here. In those cases, the earlier granted claims were product claims. Because the grant of product claims confers a right to exclude others from using the product, the patent specifications had to be consulted to determine what uses of the product were disclosed–and thus already covered by the granted product claims.

Here, the ’766 patent claimed a method of treating a more general patient population than the ’442 patent. Thus, I think the question should have been whether the ability to treat the more specific patient population was obvious in view of the ability to treat the more general population. For example, I would have been more comfortable with the decision if the court cited evidence other than the ’766 patent showing that once a treatment has been shown to be effective in treating “patients suffering from rheumatoid arthritis and requiring treatment” (as the ’766 patent claims were construed), the treatment was expected to be effective for patients with “active disease.” Then, the Federal Circuit could have concluded that the ’422 patent claims were an “obvious variant” of the ’766 patent claims without relying on the ’766 patent specification as prior art.