The doctrine of equivalents is an equitable exception to the basic concept that the patent claims define the limit of claim protection. It is intended to prevent one from making an insubstantial change to a claimed invention to avoid infringement which is defined by the function-way-result test. The doctrine is limited by the prior art, dedication to the public, and prosecution history estoppel. The prior art limitation is straightforward, the doctrine cannot be applied to find infringement if it also results in the prior art infringing the claim, rendering to unpatentable. Dedication to the public is similarly straight forward, if the equivalent was disclosed but not claimed in the patent, the public is free to use the unclaimed but disclosed equivalent. Prosecution history estoppel while seemingly simple, a narrowing amendment to secure patentability surrenders the subject matter lying between the original claim and the amendment, has exceptions as explained in Festo:
The equivalent may have been unforeseeable at the time of the application; the rationale underlying the amendment may bear no more than a tangential relation to the equivalent in question; or there may be some other reason suggesting that the patentee could not reasonably be expected to have described the insubstantial substitute in question. In those cases the patentee can overcome the presumption that prosecution history estoppel bars a finding of equivalence.
In Eli Lilly & Co. v. Hospira, Inc., appeal nos. 2018-2126, 2018-2127, and 2018-2128, the Federal Circuit applied the doctrine of equivalents where the original claims had been amended to avoid prior art. The original claim had been literally infringed but the amended claim was not. The claims were directed to reducing the toxicity associated with administering an antifolate agent by administering prior to the antifolate, folic acid and vitamin B12. As a result of prior art suggesting methatrexate, the antifolate was limited to pemetrexed disodium. Hospira’s formulation used ditromethamine salt of pemetrexed instead of the disodium salt. The Lilly application only described either an antifolate broadly or pemetrexed disodium, no suggestion of other salts even though they were known. The Lilly application did cite to prior art for other antifolates but did not incorporate the disclosures by reference. It was not disputed that the active component was not the salt form but rather the pemetrexed which resulted from the dissociation of the salt in solution or in the body.
Hospira’s FDA application was not for an ANDA since the pemetrexed ditromethamine was not identical to Lilly’s pemetrexed disodium. Instead, Hospira filed a 505(b)(2) application with the FDA. In its application it averred that its compound did not differ in the mode action from that of Lilly’s since the active moiety was the same – pemetrexed - making success on its prosecution history estoppel essential to its non-infringement position.
Since Lilly had only two choices for the antifolate, either the generic antifolate or the species pemetrexed disodium, to avoid the prior art it had no choice but to limit the claims to the specie. Lilly argued that the limitation to the disodium salt was tangential to patentability. On this point Lilly had evidence in the rejection of a dependent claim directed to pemetrexed disodium over prior art teaching the use of a pemetrexed disodium formulation which lacked the folic acid and vitamin B12. This rejection was overcome by the amendment requiring both vitamin B12 and folic acid. This meant that the salt form was not a basis for patentability, i.e., it was tangential.
In this case Lilly was fortunate to have the dependent claim rejected as it provided objective evidence to support its argument that the specific salt was tangential to patentability, it was the pemetrexed that made it patentable over the methatrexate prior art not the salt form. In the absence of this evidence Lilly’s tangential to patentability argument would have been weaker.
The lesson is simple, even when seeking to protect a species of a drug salt form, never limit the disclosure or all the claims to the specific salt. Always describe other salt forms including the generic term “pharmaceutically acceptable salt.” Here such an approach would have made Lilly’s case much stronger and easier.