Obviousness-type Double Patenting after Amgen v. F. Hoffmann-La Roche

    By Kevin E. Noonan --

Although Amgen's erythropoietin franchise has weathered its most recent challenge by F. Hoffmann-La Roche's pegylated EPO analog, Mircera® (see Amgen Inc. v. F. Hoffman-La Roche Ltd. (Fed. Cir. 2009)), the victory was not absolute.  The status of three of the patents-in-suit -- U.S. Patent Nos. 5,547,933 (the '933 patent), 5,955,422 (the '422 patent, claim 1) and 5,756,349 (the '349 patent, claim 7) -- were put in jeopardy by the Federal Circuit's reversal of summary judgment and remand to the District Court on the question of whether these patents are invalid for obviousness-type double patenting.  If the District Court were to make a determination that they are invalid, Amgen's EPO patent estate will expire almost two years earlier than expected, with loss of at least a portion of the company's significant revenues from its flagship product.  The likelihood of this outcome will depend on the parties' reaction to the latest installment of Federal Circuit jurisprudence on obviousness-type double patenting following its decision last year in Pfizer, Inc. v. Teva Pharmaceuticals USA, Inc.

The ultimate parent application, which matured into U.S. Patent No. 4,703,008, was subject to a restriction requirement identifying the following groups of claims:

I.  Claims 1–13, 16, 39–41, 47–54, and 59, drawn to polypeptide, classified in Class 260, subclass 112.
II.  Claims 14, 15, 17–36, 58, and 61–72, drawn to DNA, classified in Class 536, subclass 27.
III.  Claims 37–38, drawn to plasmid, classified in Class 435, subclass 317.
IV.  Claims 42–46, drawn to cells, classified in Class 435, subclass 240.
V.  Claims 55–57, drawn to pharmaceutical composition, classified in Class 435, subclass 177.
VI.  Claim 60, drawn to assay, classified in Class 435, subclass 6.

Pursuant to the restriction requirement, each of these claim groupings defined an independently-patentable invention, and as such would entitle Amgen to a separate patent unfettered by obviousness-type double patenting.  It is undisputed that Amgen prosecuted the claims of Group II in the '008 patent.  Thereafter, Amgen filed two applications claiming priority to the '008 patent, the '178 application and the '179 application; these applications are related to the patents-in-suit according to the following schematic diagram:

The Court characterized "in broad strokes" the claims of the three patents under consideration:

[T]he '933 patent claims recombinant EPO, a pharmaceutical composition comprising recombinant EPO, and methods of treating kidney dialysis patients by administering pharmaceutical compositions comprising recombinant EPO . . . .  The '422 patent claims a pharmaceutical composition comprising recombinant EPO . . . .  The '349 patent claims the process of producing recombinant EPO in vertebrate cells capable of producing EPO at a specific rate.

At first blush, the claims of the '933 patent would appear to correspond to Groups I and V from the '008 restriction requirement, and the claims of the '422 patent correspond to Group V.  Two other patents-in-suit -- U.S. Patent Nos. 5,441,868 (the '868 patent) and 5,618,698 (the '698 patent) -- recite claims for methods of making EPO.

Roche's position, unpersuasive at the District Court, was that the claims of the '933, '422, and '394 patents were invalid for obviousness-type double patenting over the claims of the '868 and '698 patents, and in turn the claims of the '868 and '698 patents were invalid for obviousness-type double patenting over the claims of the '008 patent.  This question depends on the meaning of 35 U.S.C. § 121, which sets forth a "safe harbor" provision protecting claims subject to a restriction requirement (and thus determined by the Patent Office to define separately-patentable inventions) from a determination of obviousness-type double patenting:

A patent issuing on an application with respect to which a requirement for restriction under this section has been made, or on an application filed as a result of such a requirement, shall not be used as a reference either in the Patent and Trademark Office or in the courts against a divisional application or against the original application or any patent issued on either of them, if the divisional application is filed before the issuance of the patent on the other application.

35 U.S.C. § 121, third sentence; see also Applied Materials, Inc. v. Advanced Semiconductor Materials Am., Inc., 98 F.3d 1563 (Fed. Cir. 1996), cited in the Court's opinion.  The District Court held that the '933, '422, and '349 patents were immunized from obviousness-type double patenting, since they had issued from applications in the chain back to the '178 and '179 applications, which were filed pursuant to a Patent Office-imposed restriction requirement.  However, these applications when filed were not designated as "divisional applications," the language of the statute, but rather as "continuation applications," and Roche argued that this difference precluded the '933, '422, and '349 patents from the § 121 safe harbor.

The Federal Circuit agreed, to the extent that summary judgment and JMOL granted by the District Court was improper.  In its opinion, by Judge Schall and joined by Judges Mayer and Clevenger, the CAFC granted no deference to the District Court's decision, first applying its own standard of review regarding summary judgment, next using the 1st Circuit's standard of review for JMOL, and last for the substantive question of obviousness-type double patenting, which it interpreted as being analogous to claim construction, as a "matter of what is claimed" and thus subject to the Federal Circuit's de novo review precedent.  Using these standards, the Federal Circuit held that the '933, '422, and '349 patents were not entitled to the § 121 safe harbor because the '178 and '179 applications were filed as continuation applications and not divisional applications.  In interpreting the statute so literally, the CAFC noted that, unlike the claims in the Pfizer case that arose from a continuation-in-part application, each of the '933, '422, and '349 patents may have satisfied the substantive requirements for divisional application status -- being limited to the subject matter disclosed in the prior parent application, reciting claims determined to define a patentably-distinct invention, and being filed during the pendency of the earlier-filed parent application.  Nonetheless, the Court held that "[t]his distinction . . . does not justify departing from a strict application of the plain language of §121, which affords its [safe harbor] benefits to 'divisional applications.'"  In making this decision, the CAFC distinguished both its Applied Materials and Symbol Technologies, Inc. v. Opticon, Inc., 935 F.2d 1569 (Fed. Cir. 1991) precedent, because in those cases the "continuation" application was filed ultimately from a properly-designated divisional application.

The CAFC reached a different conclusion with regard to the '868 and '698 patents, because the parties litigated the obviousness-type double patenting question on the merits before the District Court.  As characterized in the Federal Circuit's opinion, the claims of the '008 patent recite a recombinant cell that expresses human EPO, and the claims of the '868 and '698 patents recite methods of producing recombinant EPO using such cells, either in vitro or in vivo.  The District Court distinguished the method claims of the '868 and '698 patents and the recombinant cell (product or composition) claims of the '008 patent because:

none of the '008 claims require:  (1) that the recited host cell actually express any EPO polypeptide; (2) that the recited host cell actually express a glycosylated EPO polypeptide; (3) that the host cell be capable of producing an isolatable amount of a glycosylated EPO polypeptide; and (4) that any glycosylated EPO isolated from cells grown in culture have the stated in vivo function.

The District Court relied on expert testimony for the proposition that, even if in possession of the recombinant cell disclosed in the '008 patent, at the priority date claims to methods for making EPO were novel, useful, and non-obvious.  (The lower court also noted that the Patent Office had found these claims to be patentably-distinct.)  The Federal Circuit agreed with this distinction, that while the recombinant cells claimed in the '008 patent might be "capable of producing" recombinant EPO having in vivo biological function, the claim did not recite a limitation that such cells actually produce functional EPO.  This requirement is affirmatively found in the '868 and '698 patent claims, rendering them patentably distinct.

In making this determination, the Federal Circuit invoked a number of recent, post-KSR decisions on obviousness under 35 U.S.C. § 103, the standard used without using the '008 patent as prior art, including In re Kubin, In re O'Farrell (where the application, not the case, is recent), and Pharmastem Therapeutics, Inc. v. Viacell Inc.  The CAFC opined that the skilled artisan would not have reasonably expected to produce recombinant, glycosylated EPO having the properties -- biological activity, post-translational modifications and amount -- required by the claims of the '868 and '698 patents at the time of the invention.  In making this determination, the Federal Circuit rejected Roche's argument that the capacity for making recombinant EPO was inherent in the claimed cells (and did not address the issue of whether the claims of the expired '008 patent would have satisfied the utility or enablement requirements if the claimed cells did not inherently produce functional recombinant EPO).  The CAFC applied the Supreme Court's KSR warning against hindsight in refusing to consider the actuality that the cells claimed in the '008 patent produced biologically-active EPO in determining whether the skilled worker would have had an expectation that they would do so.

Finally, the Federal Circuit clarified the application of the principles it enunciated in Takeda Pharmaceutical Co. v.  Doll regarding the availability of later-disclosed (or discovered) methods to overcome obviousness-type double patenting invalidation of a process claim over an earlier-issued product claim.  In Takeda, the Federal Circuit held that art used by a patentee to support patentable distinctiveness of process claims filed after grant of product claims was limited to art-recognized processes known at the time the later (or "secondary" in the Takeda court's parlance) patent application was filed.  Roche argued that Amgen should not be permitted to introduce evidence relating to processes for producing recombinant EPO after the 1984 filing date of the '008 patent where Roche was precluded from asserting art available after the 1984 priority date.

Here, the Federal Circuit held that Roche's arguments "collide" with 35 U.S.C. § 120, wherein Amgen's later-filed applications are entitled to the benefit of priority to the '008 patent.  The CAFC found that interpreting its Takeda precedent as Roche advocated would "violate the plain language" of § 120, which entitles all of Amgen's patents-in-suit to the priority benefit of the '008 patent and accordingly prevents any of Amgen's patents-in-suit from being invalidated based on art available after the November 30, 1984 priority date of the '008 patent.  The Court's opinion acknowledged (and asserted that the Takeda court recognized) that this rule "could 'provide the patentee with the best of both worlds:  the applicant can use the filing date as a shield, enjoying the earlier priority date in order to avoid prior art, and rely on later-developed alternative processes as a sword to defeat double patenting challenges.'"  Nonetheless, the Federal Circuit characterized this seeming one-sidedness as being limited and hence not inequitable.  However, the CAFC noted that the '349 patent could not benefit from the application of the principles in the Takeda opinion, since it (unlike the '933 and '422 patents) recited claims to methods and not products.  Thus:

Takeda will permit Amgen, if it wishes to do so, to rely on alternative processes for making the products claimed in the '933 and '422 patents up to their filing dates to prove that the claims of those patents and the claims of the '868 and '698 patents are patentably distinct.  If Amgen pursues that course, Roche will be free to rely on subsequent developments in the art up to the filing dates of the '933 and '422 patents to prove that any alternative processes put forth by Amgen do not render the claims of the '933 and '422 patents and the claims of the '868 and '698 patents patentably distinct.

Unless the parties settle, on remand it can be expected that the District Court will be compelled to find the asserted claims of the '933, '422, and '349 patents invalid for obviousness-type double patenting, unless Amgen can establish patentable distinctiveness on the merits (i.e., outside the § 121 safe harbor).  Until such time (including any further appeals to the Federal Circuit or the Supreme Court), the significant holding in this decision is that whether a claim is entitled to the § 121 safe harbor depends, in the first instance, on whether the application reciting that claim is designated as a "divisional" and not a "continuation" application.  (Such claims must also satisfy the requirements for divisional applications.)  The Federal Circuit's opinion is stringent, harsh, and unyielding, and provides another "trap for the unwary" for which patent law is infamous.  It also raises the specter of malpractice liability in designating follow-on applications accurately, a practice tip that the patent bar ignores at its peril.

Comments

[EG]

"Takeda will permit Amgen, if it wishes to do so, to rely on alternative processes for making the products claimed in the '933 and '422 patents up to their filing dates to prove that the claims of those patents and the claims of the '868 and '698 patents are patentably distinct. If Amgen pursues that course, Roche will be free to rely on subsequent developments in the art up to the filing dates of the '933 and '422 patents to prove that any alternative processes put forth by Amgen do not render the claims of the '933 and '422 patents and the claims of the '868 and '698 patents patentably distinct."

Kevin,

I've puzzled over this statement and whether it's consistent with earlier statements in the Amgen opinion on what Takeda means. Is what's being said only that if Amgen puts in evidence of patentable distinctness for alternative processes up to the filing dates of the '933 and '422 patents, Roche is then (and only then) entitled to put in contervailing evidence to show that these alternative processes aren't patentably distinct?

[Kevin E. Noonan]

Dear EG:

I understand that the opinion stands for the proposition that Roche will be precluded from using after-filing art (i.e., after November 30, 1984) to attack the validity of the '933 or '422 patent claims. I think that Roche will be able to rebut any evidence Amgen presents regarding patentable distinctiveness based on new method art between the original filing date and the filing dates of the '933 and '422 patent, which is why the court says that Takeda is a "two-way street" within its confines.

Hope this helps. Thanks for the comment.

[EG]

Kevin,

It does. What you said is consistent with what this passage on what Takeda means says to me. I just remember reading something earlier in Amgen on the double patenting issue that might appear to be in conflict with this later passge on what Takeda means. The Federal Circuit isn't always as clear (or internally consistent) as it might be as it makes transitions between related issues.

[Dan Feigelson]

Kevin, I don't disagree with your conclusion about practitioners needing to be careful - this decision certainly grabbed my attention. Nevertheless, I'm not sure the situation is quite so dire. The Amgen patents in question are pre-1995 applications, thus entitled to 17 years from grant. Amgen could obviate any double-patenting issue by filing terminal disclaimers in the later-issued patents. But at four billion plus dollars a year, it's clear why Amgen doesn't want to do that.

In contrast, in the post-1995 world that we now inhabit (save for a handful of pre-1995 apps that are still languishing on an examiner's desk or before the BPAI), US patents get 20-years from earliest claimed US priority, irrespective of whether an application is filed as a divisional or continuation. Which means that often, there won't be a reason to NOT file a terminal disclaimer to avoid an obviousness-type double patenting rejection.

The one situation where a terminal disclaimer is clearly to be avoided is where the later-filed application is entitled to a patent term adjustment under 35 USC 154, since a TD forfeits the adjustment period. (A TD doesn't affect Waxman-Hatch patent term extensions under 35 USC 156.) Even in those cases, though, often the additional time at the end of the patent term won't matter. It's primarily (only?) in pharmaceuticals where each day of extra patent term is critical.

So I agree that following this decision, patent practitioners need to be careful and ensure that they file divisionals as divisionals and not as continuations. (They should probably also be careful in filing claims to ensure that they will invoke a restriction requirement, thus enabling them to avail themselves of safe harbor of 35 USC 121.) In the cases where this distinction is crucial, the sums of money involved can be quite high. But in many technologies, it won't matter, and even in pharma, not all later-filed applications will be entitled an extension under 154; or the later-filed applications may be directed to less-easily enforceable aspects of the invention, such as synthetic methods.

[Kevin E. Noonan]

Dear Dan:

There is one more consideration, which could be quite troublesome. Filing a terminal disclaimer requires the patents to be co-owned, and thus incapable of being independently assigned. There may be situations where that creates a problem.

But in general I agree with you. Thanks for the comment.

[BioPatAtty]

Can we please get rid of OTDP now? With patent term being calculated as 20 years from first non-provisional priority date, this doctrine is a fossil. And the argument that infringers shouldn't have to worry about more than one party coming after them is garbage. What other area of law do we worry about a liable party damaging more than one plaintiff? Join all rights holders as necessary parties, and that's that.

This is a dead horse. Let's stop beating it. If it's not a proper 103, move on.