Merck & Cie v. Watson Laboratories, Inc., 125 F. Supp. 3d 503

Background

Merck sued Watson for infringing U.S. Patent No. 6,441,168 (claim 4) covering the Type I crystalline calcium salt (pentahydrate) of 5‑methyl-(6S)-tetrahydrofolate (MTHF) identified by specific PXRD two‑theta peaks and water content.
Watson filed ANDAs to market generic versions of Safyral® and Beyaz®, whose MTHF form Merck alleges infringes claim 4; the parties stipulated that if claim 4 is valid it would be infringed by Watson’s ANDA products.
Key claim elements include PXRD peaks at two‑theta 6.5, 13.3, 16.8, 20.1 and a water of crystallization of at least one equivalent (Type I identified in the spec as ~14.5% water; solubility "sparingly soluble").
Watson challenged claim 4 as invalid under the on‑sale bar (35 U.S.C. §102(b)), anticipation (§102(a)) by a prior '850 patent, obviousness (§103) (alone or with a '655 patent), and lack of written description (§112).
Factual disputes centered on fall 1998 communications between Merck and Weider about a 2 kg MTHF sale (existence/definiteness of an offer given a Confidentiality & Noncompetition Agreement requiring a signed definitive agreement) and on whether prior art processes/products inevitably produce the claimed Type I PXRD/solubility.

Issues

Issue	Merck's Argument	Watson's Argument	Held
On‑sale bar (§102(b))	Fall 1998 communications were non‑binding due to CDA §5.2 requiring a signed definitive agreement; industry custom and missing safety/liability terms show no definite offer.	Sept. 9/16, 1998 communications (price, qty, delivery, payment) were a commercial offer or sale; §5.2 did not preclude a standalone purchase or was waived.	No on‑sale bar: CDA required further assent; communications not sufficiently definite; no waiver.
Anticipation (§102(a)) vs. '850 patent	'850 product differs: solubility ("practically insoluble" vs claimed "sparingly soluble") and experiments Watson cites did not follow '850 process and showed seeding/contamination.	'850 Example 3 produced a pentahydrate; Watson's experiments (Rogers) produced PXRD peaks matching claim 4, so anticipation is inherent.	No anticipation: prior art/product did not inherently disclose claim 4; Rogers’s work non‑probative (different starting material, seeding observed).
Obviousness (§103)	Unknown polymorph discovery is unpredictable trial‑and‑error; no reasonable expectation of success from prior art (no sufficient motivation + unpredictability).	Prior art ('850 and '655) would motivate and reasonably lead a skilled artisan to produce Type I crystals.	Not obvious: lack of reasonable expectation of success; trial‑and‑error nature of polymorph discovery defeats obviousness.
Written description (§112)	Claim requires "at least one" water; specification discloses Type I typically >3 waters but "typically" means fewer waters are encompassed — specification supports claim scope.	Specification does not disclose monohydrate/embodiment with only one water equivalent; thus lacks written description for that breadth.	Adequate written description: disclosure of Type I as "typically" >3 waters does not exclude fewer waters; inventors possessed claimed scope.

Key Cases Cited

Pfaff v. Wells Elecs., 525 U.S. 55 (on‑sale bar requires invention ready for patenting and subject of a commercial offer)
KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (obviousness framework and flexible analysis)
Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336 (written description requirement standard)
Silicon Graphics, Inc. v. ATI Tech., Inc., 607 F.3d 784 (anticipation requires each claim element disclosed expressly or inherently)
Trintec Indus., Inc. v. Top‑U.S.A. Corp., 295 F.3d 1292 (inherent anticipation requires inevitability)
Atlanta Attachment Co. v. Leggett & Platt, Inc., 516 F.3d 1361 (definiteness of commercial offer for on‑sale analysis)
Graham v. John Deere Co. of Kansas City, 383 U.S. 1 (Graham factors and secondary considerations for obviousness)
Medicines Co. v. Hospira, Inc., 791 F.3d 1368 (on‑sale bar application)