Hoffmann-La Roche Inc. v. Apotex Inc.
748 F.3d 1326
Fed. Cir.2014Background
- Roche holds patents ('634 and '957) claiming a method of treating postmenopausal osteoporosis by orally administering ~150 mg ibandronic acid once monthly (Boniva®). FDA approved a 150 mg once-monthly Boniva® in 2005.
- Defendants (generic manufacturers) filed ANDAs; Roche sued under 35 U.S.C. § 271(e)(2). District court granted summary judgment that the challenged claims are invalid for obviousness; Roche appealed.
- Prior art disclosed intermittent/less-frequent bisphosphonate dosing and taught a "total-dose" concept (Riis) that efficacy depends on total dose over a period rather than dosing schedule. Studies (Ravn, Daifotis, Chen, Schofield, Geddes, Mockel, Riis, Recker) discussed various doses and schedules.
- District court found once-monthly dosing and a 150 mg monthly total dose were taught or obvious to try from the prior art; it rejected Roche’s secondary considerations (unexpected results, bioavailability, safety) as insufficient to overcome obviousness.
- This panel majority affirmed summary judgment of obviousness; Judge Newman dissented, arguing the combination was not suggested by the prior art, unexpected results and long-felt failure weigh against obviousness, and invalidation reflected hindsight.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Was once-monthly oral dosing of ibandronate obvious? | Roche: prior art taught away; intermittent quarterly/long-interval dosing was thought ineffective; Recker failure showed non-efficacy. | Defendants: prior art (Lunar News, Krause, Chen, Schofield, Geddes, Riis) taught monthly or intermittent dosing; reasonable expectation of success. | Held: monthly dosing was known in the art or obvious to try; Riis and others provided a reasonable expectation of success. |
| Was selecting ~150 mg/month obvious? | Roche: no suggestion to scale to 150 mg; FDA refused 5 mg/day approval; 150 mg was unexpected and potentially unsafe. | Defendants: Ravn and Daifotis disclosed 5 mg/day and weekly equivalents; Riis total-dose concept made 150 mg (5 mg×30) an obvious, predictable choice. | Held: 150 mg/month was obvious or obvious to try given prior art guidance and predictable scaling. |
| Did safety concerns teach away from 150 mg? | Roche: prior studies showed toxicity concerns and FDA preference for 2.5 mg/day; safety would deter choosing 150 mg. | Defendants: prior art did not show safety limits below 150 mg; Mockel and Daifotis disclosed higher or similar single-dose ranges and tolerability. | Held: no genuine dispute that prior art taught away; safety references did not preclude 150 mg. |
| Do unexpected results (superior BMD, nonlinear bioavailability) negate obviousness? | Roche: MOBILE and bioavailability data show unexpected superior efficacy and nonlinear uptake, supporting nonobviousness. | Defendants: unexpected superiority does not negate a strong prima facie case; bioavailability not shown to explain efficacy. | Held: unexpected results and nonlinear bioavailability insufficient to overcome obviousness on summary judgment. |
Key Cases Cited
- KSR Int'l Co. v. Teleflex, 550 U.S. 398 (2007) (guidance on "obvious to try" and predictable solutions)
- PharmaStem Therapeutics, Inc. v. ViaCell, Inc., 491 F.3d 1342 (2007) (reasonable expectation of success standard)
- Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348 (2007) (reasonable expectation of success in obviousness analysis)
- In re Merck & Co., 800 F.2d 1091 (Fed. Cir. 1986) (superior results alone do not defeat obviousness)
- Graham v. John Deere Co., 383 U.S. 1 (1966) (framework for obviousness inquiry, including secondary considerations)
- In re Cyclobenzaprine Hydrochloride Extended-Release Capsule Patent Litig., 676 F.3d 1063 (2012) (consider all evidence relating to obviousness before invalidating)
- In re Kubin, 561 F.3d 1351 (2009) (limits on "obvious to try" where prior art gives no reason to expect success)
- In re Soni, 54 F.3d 746 (1995) (surprising results weigh against obviousness)