August 1, 2013

Copaxone® Confusion

Indefiniteness in light of Teva Pharma. v. Sandoz

Teva Pharmaceuticals markets Copaxone® (glatiramer acetate), a treatment for relapse-remitting multiple sclerosis. The Orange Book lists eight patents covering glatiramer acetate, all of which derive from the same priority documents and have substantially the same specification. Teva licenses these patents from Yeda Pharmaceuticals. Mylan and Sandoz filed abbreviated new drug applications with the FDA, both seeking authorization to market generic glatiramer acetate injections. Teva sued for infringement under 35 U.S.C. §271(e)(2)(A), asserting nine patents. Mylan and Sandoz argued the drugs described in their ANDAs did not infringe any of Teva’s claims, and further argued that all asserted claims were invalid for non-enablement, indefiniteness, and obviousness. The district court ruled in favor of Teva on all counts. Mylan and Sandoz appealed. The most interesting parts of the case concern the indefiniteness rejection, which is the subject of the present blog post.


          On appeal, Mylan and Sandoz argued that the claims in suit all recited limitations related to the “molecular weight” of the claimed drug, but the nine asserted patents all described three different measures of “molecular weight”: peak average molecular weight (Mp); number average molecular weight (Mn); and weight average molecular weight (Mw). It is not essential to the judgment to understand the distinctions among these three, except to know that the claim scope depended on which one was used.

The Federal Circuit divided the asserted claims into two groups: (1) Group 1 claims specified that the molecular weight of the drug lies within a claimed range (e.g., “molecular weight of about 5 to 9 kilodaltons”); (2) Group 2 claims specified that a percentage of the drug molecules have a weight within the claimed range (e.g., “over 75% of its molar fraction within the molecular weight range from about 2 kDa to about 20 kDa”). Mylan and Sandoz argued that because none of the claims specified which measure of molecular weight—Mp, Mn, or Mw—was intended, and because the claim scope changes according to which measure is used, the asserted claims are all insolubly ambiguous. Writing for a unanimous three-judge panel, Judge Moore agreed with Mylan and Sandoz that the Group 1 claims were insolubly ambiguous, but concluded that the Group 2 claims were clear.

To understand how the court arrived at this conclusion, one must know the prosecution history. The trial court had concluded that all asserted claims are definite enough to avoid invalidity, because during prosecution of U.S. 6,939,539 (‘539), Examiner Amy Lewis had made a rejection for indefiniteness, predicated on the ambiguity of the three measures of average molecular weight. In response, Yeda had argued to Examiner Lewis that an ordinary chemist reading the specification would understand from the graphs that “molecular weight” referred to Mp. The trial court considered that this assertion made it clear which molecular weight was intended, and the Markman order construed “molecular weight” to mean “Mp.”

At trial, however, Mylan and Sandoz had pointed out that this was not the only statement that Yeda had made during prosecution history. During prosecution of U.S. 6,620,847 (‘847), Examiner Frederick Krass had also made essentially the same indefiniteness rejection. In response to Examiner Krass’ rejection, Yeda had argued that “[o]ne of ordinary skill in the art could understand that [the use of] kilodalton units implies” that Mw is intended. In other words, Yeda had taken inconsistent positions during prosecution of the different patents in this family. At trial, Teva had countered this point by noting that the assertion made during the ‘847 prosecution was a scientific non sequitur—all three molecular weight measures are expressed in kilodaltons, so the kDa unit tells one nothing about which measure is intended—and therefore one skilled in the art would have disregarded the statement made in ‘847 in favor of the statement made in ‘539.

Although the trial court found this argument convincing, the Federal Circuit was not persuaded. Simply put, the Federal Circuit concluded that the inconsistent positions taken during the prosecution of the two different members of this patent family rendered the claim term “molecular weight” insolubly ambiguous in all members of the family. Although this ambiguity did not extend to the Group 2 claims, where the recitation of a percentage range combined with the molecular weight range was enough to make the claimed weight range clear, the inconsistency between the ‘539 prosecution and the ‘847 prosecution was enough to make all Group 1 claims indefinite. The take-home lesson here for patent applicants is that it is absolutely vital to make sure that positions taken during prosecution of one patent are consistent with those taken during prosecution of related patents.

In this instance, Yeda put Teva in a difficult position, because Yeda made the scientifically inaccurate statement in ‘847 before it was later hit with essentially the same rejection in ‘539. Yeda was stuck with the choice of reiterating a logically untenable position, or making a statement that contradicted its earlier position. This case shows that inconsistency is so damaging to the whole patent family, that it would have been better for Teva to have sought a re-issue in ‘847 before asserting any of the rest of the members of the family in a lawsuit.


          Another interesting aspect of this opinion concerns prosecution history estoppel. Recently, the Federal Circuit held in Biogen Idec v. GlaxoSmithKline, 713 F.3d 1090 (Fed. Cir. 2013) that when Biogen responded to a rejection for non-enablement by saying that “one of skill in the art could readily identify an antibody that binds to CD20 with similar affinity and specificity as does RITUXAN® using techniques that are well known in the art…”, this statement was sufficiently “clear and unmistakable” as to amount to a disclaimer of antibodies that bound to different epitopes of CD20 than the epitopes bound by Rituxan®. Id. at 1097.

By contrast, in the instant case Yeda distinguished its claims from U.S. 3,849,550 (Teitelbaum) by arguing that, “in contrast to Teitelbaum’s copolymer-1 with a ‘minimum molecular weight of 10 kilodaltons,’ the pending claims cover ‘copolymer-1 having a molecular weight of about 5 to 9 kilodaltons.’” Mylan and Sandoz argued that this constituted a disclaimer of drug with Mw greater than 10 kDa. The Federal Circuit disagreed, holding that Yeda’s statement was not “a clear and unmistakable disclaimer.” Rather, the Federal Circuit concluded that, because the term “molecular weight” was so ambiguous as to render the claims invalid, Yeda’s use of the term “molecular weight” during prosecution could not be “clear and unmistakable” enough to waive claim scope.

The above article was written by Greg DeLassus, an associate in the St. Louis, MO office of Harness Dickey.  Greg can be reached at 314.446.7687 or