Daratumumab - USA

On Jan 28, 2019, Delaware court granted summary judgment as to noninfringement of the human antibody claims; granted-in-part and denied-in-part summary judgment of invalidity for lack of written description; granted summary judgment of invalidity for lack of enablement and denied summary judgment of invalidity for indefiniteness.

MorphoSys alleged infringement of U.S. Patent Nos. 8,263,746, 9,200,061, and 9,758,590. The asserted patents describe antibodies that can be used to treat blood cancer. The patents describe antibodies that bind to CD38, thus causing the destruction of the cancerous cells. Janssen produces an antibody drug; Darzalex® (“daratumumab”) that MorphoSys contends infringes the asserted patents.

Noninfringement:

Some of the asserted claims are directed to solely “human” antibodies, while other asserted claims are directed to “human or humanized” antibodies. Based on the claim constructions, in order for an antibody to be “human,” it must not be any of the following: (i) chimeric, (ii) humanized, and (iii) derived even in part from a nonhuman species.  Janssen moved for summary judgment of non-infringement of the “human” antibody claims, contending that daratumumab is not a “human” antibody either literally or under the doctrine of equivalents.  Janssen argued that daratumumab is “humanized” and not “human.” Specifically, Janssen argued that “human” and “humanized” are mutually exclusive therefore non-infringement of the “human” claims can be determined as a matter of law. The Court agreed with Janssen that no reasonable juror could conclude that daratumumab is a human antibody. This conclusion follows from the Court’s claim construction and undisputed facts. The Court concluded that an antibody produced using a transgenic mouse is a humanized antibody because it is, as any reasonable factfinder would conclude, “from (either in whole or in part) a nonhuman species.”

MorphoSys contended that even if the “human” antibody claims are not literally infringed, they are infringed under the doctrine of equivalents. Janssen argued that the doctrine of equivalents does not apply for three reasons: (1) the disclosure-dedication rule; (2) the specific exclusion principle; and (3) prosecution history estoppel. In the view of Janssen, the patents note that humanized antibodies could be made from transgenic mice but do not claim (in the relevant claims) such antibodies. MorphoSys counters that the “disclosure-dedication rule” does not apply because the patents do not describe “humanized” antibodies as an alternative to “human” antibodies. But the court concluded that the disclosure-dedication doctrine applies here. A person of ordinary skill reading the patent would identify humanized antibodies as an alternative to human antibodies. Therefore, having claimed only human antibodies in some claims, MorphoSys cannot now argue that those claims should also cover humanized antibodies under the doctrine of equivalents. Under the “specific exclusion” principle, the doctrine of equivalents cannot broaden a claim to cover a feature that is “the opposite of, or inconsistent with, the recited limitation.” The Court concluded that humanized antibodies are specifically excluded from the claims directed solely to human antibodies. This conclusion follows logically from MorphoSys’ admission; given that a human antibody cannot also be a humanized antibody (and vice versa), the terms are “inconsistent with each other.” Moreover, a holding that humanized antibodies were equivalent to human antibodies would render the claim term “human or humanized” redundant. Therefore, daratumumab is specifically excluded from the “human” antibody claims. Thus the Court found that both disclosure-dedication and specific exclusion apply here and therefore will not reach the issue of prosecution history estoppel.

Invalidity: Written Description

Janssen contended that the claims lack sufficient written description because the patents fail to disclose (1) any species of the “Binding Claims” (2) sufficiently “representative species” for any claimed genus and (3) “common structural features” sufficient to visualize or recognize the claimed genera. With respect to Binding claims court said that there are genuine disputes of material fact that preclude summary judgment of lack of written description for the Binding claims. A reasonable factfinder could find that peptide mapping was recognized as sufficiently reliable to demonstrate possession to a POSA. Based on the record evidence, a reasonable factfinder – taking the evidence in the light most favorable to MorphoSys – could find that Figure 7 of the patents conveys possession, notwithstanding the x-ray crystallography data and the Replitope Report. With respect to Representative species court said that there are genuine issues of material fact that preclude summary judgment here. For example, a reasonable factfinder could find that the four disclosed antibodies are representative of all known members of the claimed genera, including daratumumab. Therefore, the Court denied this portion of Janssen’s summary judgment motion. With respect to Common structural features court agreed with Janssen & said that patents fail to disclose any structural elements that would “inform one of skill how to identify antibodies possessing the claimed properties.” The claims “cover any and all CD38 antibodies that satisfy broad functional tests, yet the specification does not teach the necessary correlation between those claimed functional properties (i.e., binding within specific locations or different cell-killing activities) and the structural characteristics (e.g., the amino acid sequence) of antibodies having those properties.” Indeed, multiple MorphoSys witnesses admit that an antibody’s sequence cannot be used to predict its binding properties. Given the undisputed lack of a known relationship between an antibody’s structure (its sequence) and its function (its binding properties), the only reasonable conclusion is that the specification does not sufficiently disclose structural features common to the members of the genus. Thus court granted summary judgment with respect to the “structural features” prong of the written description issue.

Invalidity: Enablement

“To be enabling, the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.” Court said that any reasonable factfinder would have to find that very many unique anti-CD38 antibodies exist. Dr. Bradbury “very conservative[ly]” estimates that there are 10¹⁹ (ten quintillion) anti-CD38 antibodies. Each of the asserted claims limits the recited human anti-CD38 antibodies in multiple ways. According to Dr. Bradbury’s estimates, which he characterizes as “very conservative,” the most limiting claim (claim 3 of the ‘590 patent) would cover 0.5% of all human anti-CD38 antibodies. It follows that any reasonable factfinder would have to find that each claim covers a very large number of antibodies. Moreover, “Non-conservative” variants of known antibodies would have to be screened to determine their effectiveness. And obtaining antibodies within the claims that are not conservative variants of disclosed antibodies would require substantial time and effort by a POSA.

Applying the Wands factors, the Court concluded that undue experimentation would be needed to practice the full scope of the claimed invention. First, with respect to “the quantity of experimentation necessary,” the Court finds that obtaining antibodies within the claims (other than conservative variants of disclosed antibodies) would require substantial time and effort by a POSA. Turning to “the amount of direction or guidance presented” and “the presence or absence of working examples,” the specifications are largely unhelpful. Although the patents provide four working examples, they do not teach a POSA how to predict from an antibody’s sequence whether it will bind to CD38. Nor do the patents improve a POSA’s ability to discover any of the countless antibodies within the scope of the claims that are not a conservative variant of a disclosed antibody. Rather, a POSA attempting to obtain a claimed antibody that is not a variant of a known antibody would have to do essentially the same amount of work as the inventors of the patents-in-suit; like the inventors, a POSA would have to discover these antibodies de novo through phage display or another technique. The specifications’ deficiencies are particularly acute in view of the next four factors: “the nature of the invention,” “the state of the prior art,” “the relative skill of those in the art,” and “the predictability or unpredictability of the art.” Turning to the final factor, the Court finds that “the breadth of the claims” here is vast. There are a very large number of anti-CD38 antibodies, and a very large number of them meet the claims’ other limitations. For all of these reasons, application of the Wands factors leads the Court to conclude that the claims are not enabled. The Court thus granted Janssen’s motion for summary judgment of invalidity for lack of enablement.

Invalidity: Indefiniteness

Janssen contended that the Binding Claims are indefinite because the patents do not make sufficiently clear which measurement technique to use to determine whether an antibody meets the binding properties recited by the Binding Claims. Janssen specifically contended that a POSA would not use PepSpot to measure binding because “there is no dispute that PepSpot cannot definitively measure binding to the actual CD38 protein.” In support of this contention, Janssen relies on binding data for MOR03080 from the Replitope Report and Janssen’s x-ray crystallography experiment, which Janssen argues shows the inaccuracy of peptide mapping techniques like PepSpot. Court held that Janssen’s extrinsic evidence – which would have been unavailable to a POSA at the pertinent date of the patent – is not relevant to the indefiniteness inquiry as the intrinsic evidence unambiguously conveys to a POSA that epitope mapping techniques like PepSpot can be used to measure an antibody’s binding properties. Therefore, Janssen has failed to meet its burden to show, by clear and convincing evidence, that the Binding Claims are indefinite.