Lately, I’ve been feeling that the only case law engaging enough to write about has involved S. 101 issues, so it was a welcome break to comment on some classic principles of organic chemistry.
Apotex filed an IPR petition that Wyeth’s U.S. Patent No. 8,879,828 was obvious in view of four references. The Board granted the petition but went on to find that the antibiotic compositions claimed in the patent were not obvious. The Fed. Cir. agreed that the Board’s final decision was supported by substantial evidence (Appeal No. 2015-1851 (Fed. Cir., August 6, 2016) —Judges Lourie, Wallach and Hughes, Lourie writing).
Claim 1 was directed to an a composition comprising tigecycline HCL or gentisic acid, lactose, having a pH “in a solution” between about 3 and 7. The relatively low pH lessens while the lactose inhibits the epimerization of the 4-dimethyl amino substituent that is present in all three compounds.
At trial, Apotex did not cite any art disclosing this antibiotic, which is a member of the tetracycline family, but rather relied on 4 references disclosing various aspects of minocycline and tetracycline. The primary reference was “CN ‘550” which discloses minocycline in combination with a “supporting agent” that could be lactose and a pH adjusting agent such as HCl. The composition was disclosed to be resistant to oxidation. Another reference was cited to show that minocycline is oxidized above pH5. Another reference disclosed that tetracycline was stable to oxidation at low pH’s and that certain solubilizers like polysorbate 80 help stabilize the drug against epimerization. The Board found that Apotex failed to explain why the POSA would be motivated to replace minocycline in the Chinese reference with tigecycline and agents that would inhibit epimerization.
Apotex must have expected the Fed. Cir. to reverse, since it had presented evidence that minocycline was known “to have failed” [as an antibiotic] and that minocycline, tetracycline and tigecycline are all tetracyclic antibiotics with identical A and B rings. (The phenolic C rings all look identical to me.)
But the Fed. Cir. found the expert testimony on the failure of minocycline unpersuasive and agreed that the Board that there was no evidence that tigecycline “would be as stable in the CN ‘500 composition” based simply on the identical A and B rings. The Fed. Cir. also quoted from Wyeth’s expert’s testimony about differences in oxidation rates and epimerization among the tetracyclines, but it is difficult for me to read this as negating motivation to try prior art compositions comprising actives in the same family to improve tigecycline’s properties.
Tigecycline, tetracycline and minocycline all have a phenol ring that permits oxidation. More importantly, they all share the same A ring which is where the epimerization occurs. All the differences between the structures occur due to differing substituents on the D ring (reading the structures in the Merck Manual from right to left.) It is not clear why the Board found Apotex’s expert testimony as “not supported by objective evidence or analysis” while finding Wyeth’s expert testimony to be persuasive, although the quoted portions seem to relate to unpredictability of success. The only semi-concrete takeaway here is that structural similarity per se is not enough to satisfy the requirement for motivation to make analogous compositions, even if the active compound is in the same family as the compounds in the cited art.