October 25, 2016
Case Name: Forest Labs. Holdings Ltd. v. Mylan Inc., No. 13–1602–SLR (Consolidated), 2016 U.S. Dist. LEXIS 89268 (D. Del. July 11, 2016) (Robinson, J.)
Drug Product and Patent(s)-in-Suit: Savella® (milnacipran hydrochloride); U.S. Patents Nos. 6,602,911 (“the ’911 patent”), 7,888,342 (“the ’342 patent”) and 7,994,220 (“the ’220 patent”)
Nature of the Case and Issue(s) Presented: Forest is the exclusive licensee of the patents-in-suit, and markets Savella for the treatment of fibromyalgia. Milnacipran hydrochloride is a serotonin reuptake inhibitor (SSRI) and a noradrenaline (norepinephrine) reuptake inhibitor (SNRI) that has been approved in France since 1996 for the treatment of depression. It is not approved for the treatment of depression in the U.S. Savella is designated by the FDA as a new chemical entity. Mylan filed an ANDA seeking approval to make and sell generic milnacipran hydrochloride.
The asserted claims of the ’991 patent claim methods of treating fibromyalgia syndrome (“FMS”) with milnacipran. The same is true of the asserted claims of the ’342 patent, except that the method excludes administering phenylalanine, tyrosine or tryptophan. The ’220 patent claims a method of treating fibromyalgia with milnacipran dosed according to a “fast titration schedule,” which runs from 12.5 mg on day 1 to 100 mg/day for days 5 and beyond.
The court addressed issues of indirect infringement and validity, and ultimately held the asserted claims valid and infringed.
Why Forest prevailed: Mylan argued that it did not intend to induce infringement of the asserted claims of the ’911 patent because its proposed labelling did not instruct physicians to use milnacipran only as a monotherapy treatment. But the court found that the patentee had disclaimed combination therapy to overcome rejections of the asserted claims, and accordingly construed the claims of the ’911 patent to exclude combination therapy. In light of the product labelling, Mylan could not instruct doctors that milnacipran was approved for combination therapy. Therefore, the court concluded that Mylan’s label will “inevitably lead some consumers to practice” the claimed monotherapy, and that Mylan had the requisite intent to induce infringement of claims 1, 2 and 7 of the ´911 patent.
With respect to the ’342 patent, Mylan argued that because its proposed label did not specifically instruct patients to avoid phenylalanine, tyrosine or tryptophan when using milnacipran, it could not induce infringement. The court disagreed based on the warning in Mylan’s label that those amino acids, in combination with milnacipran, could lead to potentially life-threatening side effects for fibromyalgia patients.
With respect to the ’220 patent, Mylan argued that its proposed ANDA label did not instruct physicians or patients to follow the claimed titration schedule. In particular, Mylan relied on the following language from its proposed label: “Based on efficacy and tolerability, dosing may be titrated according to the following schedule.” Mylan argued that the permissive language in the label allowed it to escape induced infringement. But the court reasoned that it was irrelevant that some users may not specifically perform the patented method, and instead focused its attention on the fact that Mylan’s proposed doses were the same as the claimed doses.
Next, Mylan argued that it did not contributorily infringe because its proposed non-infringing uses required off-label use. Rejecting Mylan’ argument, the court said that an off-label use cannot constitute a substantial non-infringing use. Accordingly, the court rejected Mylan’s non-infringement arguments.
The court next addressed the issue of validity. First, Mylan argued that the Horrobin reference anticipated the asserted claims of the ’911 and ’342 patents. Horrobin generally teaches drugs, including milnacipran, in combination with certain amino acids to treat neurologic diseases, including FMS. Mylan’s anticipation argument rested on one sentence, which states that the drug effects of monotherapy “are important but relatively modest.” The court rejected Mylan’s anticipation argument for the following reasons: (i) in view of the court’s claim construction—the ’911 patent claims only monotherapy and the ’342 patent recites a method of treating FMS in a patient without disclosing the specific amino acids described in Horrobin—Horrobin actually teaches away from the claimed inventions; (ii) the single sentence relied on by Mylan refers to the treatment of stroke and brain injury, not fibromyalgia; (iii) the Horrobin reference and the same arguments that Mylan made during the litigation was raised and considered by the Examiner during prosecution of the ’911 and ’342 patent applications.
Second, the court addressed Mylan’s obviousness arguments. Before doing so, it assessed the parties’ respective experts and ultimately found that Forest’s experts had the more relevant and extensive expertise and so offered the more credible evidence. Mylan’s first obviousness argument combined the Horrobin reference with Dwight, Barkin and WO ’223 references. Those references purportedly taught the use of SNRIs to treat FMS. Mylan argued that since milnacipran was one of three known SNRIs (the two others being venlafaxine, discussed by Dwight and Barkin, and duloxetine, discussed by WO ’223), milnacipran was one of a “finite number of identified, predictable solutions.” The court stated that Mylan failed to provide a meaningful explanation for any motivation to make the claimed combination. More particularly, the court concluded that altering the Horrobin reference as proposed would alter the very teaching of that reference. The court then concluded that there was no basis for the proposition that Dwight (which taught treating fibromyalgia with venlafaxine in a monotherapy) when combined with the later-filed Horrobin (which taught treating fibromyalgia in a combination therapy) would lead to an effective monotherapy treatment with milnacipran. In addition, the court found Barkin’s teachings inadequate, as that reference considered venlafaxine to have “unique” properties. In particular, the court found that Barkin taught that at the relevant dose ranges, venlafaxine acts as an SSRI and not as an SNRI: milnacipran “has the ‘opposite’ pharmacology of venlafaxine.”
Mylan’s second obviousness argument combined references that teach the use of tri-cyclic antidepressants (“TCAs”) to treat FMS combined with references that teach that milnacipran is an SNRI and a suitable replacement for TCAs. The court said that the fundamental assumption underlying Mylan’s argument was that TCAs were known to work through dual inhibition. The court rejected Mylan’s second invalidity argument, characterizing it as “overly stated.” For example, it was acknowledged in the prior art that there was uncertainty as to how TCAs impart efficacy in fibromyalgia. Some references ascribed efficacy to sleep improvement. Still others disputed that serotonin or noradrenaline reuptake was important.
As to the ´220 patent, Mylan argued that it was obvious in combination with three references. But the court found that the claimed fast titration schedule was directly contrary to the prior art. It noted that Mylan admitted that no single reference encompassed the claimed fast-titration schedule. The court further concluded that Mylan used hindsight to find fast titration in the prior art and used prior-art titration schedules directed to the treatment of depression but not fibromyalgia.
Finally, the court did not credit Forest’s secondary considerations argument of an allegedly unexpected result (comparable tolerability under the claimed titration schedule), calling the result “a difference in degree rather than a difference in kind.”