Haiwen IP Memo of August 25， 2016
Does the PRC Supreme People's Court Diverge from the Doctrine of Equivalents in Eli Lilly v. Watson?
The patent infringement dispute between Eli Lilly and Watson Pharmaceuticals (Changzhou) Co., Ltd. ("Watson") on Olanzapine came to an end on July 16, 2016. The PRC Supreme People's Court ("SPC") rendered a judgment on the case after 13 years of litigation procedures between the parties before three tiers of Chinese courts.
I. The '346 patent
Olanzapine was invented by Lilly in the Great Britain in 1990. It was firstly manufactured and marketed by Eli Lilly in the U.S. in 1996 for the treatment of schizophrenia and bipolar disorder. The compound is:
In China, the invention is covered by a Chinese process invention patent No.91103346.7 (" '346 patent "), filed on April 24, 1991 and issued on May 17, 1995. At that time drugs were precluded from patentable subject matter.
Claim 1 of the '346 patent is about reaction paths, which provides:
II. Procedural History
Watson, a subsidiary of a U.S. pharmaceutical company and a generic drug producer, manufactured and marketed Olanzapine upon approval by China Food and Drug Administration ("CFDA") on May 09, 2003.
Based on the facts disclosed in the judgment, we sketch the Watson reaction path as follows:1,2
Eli Lilly sued Watson before Nanjing Intermediate People's Court on September 29, 2003 and alleged that Watson infringed path (a) of claim 1 of the '346 patent ("Path (a)"). Eli Lilly lost the case on April 7, 2008, and appealed to Jiangsu Provincial High People's Court ("Jiangsu Court"). On December 29, 2011, Jiangsu Court reversed the first instance judgment and imposed an injunction to bar Watson from manufacturing Olanzapine by the method recited in Path (a) and awarded Eli Lilly damages of RMB500,000, on the ground that Watson failed to prove that it utilized its own procedure (the procedure that Watson registered with CFDA in 2003) ("2003 Procedure") to manufacture the product, which was not a literal copy of that of Eli Lilly. Jiangsu Court's conclusion was based on a technical appraisal conducted by an agency that failed to obtain Olanzapine by the 2003 Procedure.
Eli Lilly filed before Jiangsu Court the second lawsuit against Watson for continuing infringement between September 29, 2003, the date of filing of the first case, and April 24, 2011, the expiration date of the '346 patent. In this case, Watson stated that since 2003 it had used a manufacturing procedure registered with CFDA in 2008 as a supplement to the original procedure ("2008 Procedure"), with a sacrifice to concede its violation of the relevant administrative law. Consequently, Jiangsu Court held: (a) that the statement of using 2008 Procedure could not be proved to be true due to the inconsistency with the fact found in the first case; (b) that the 2008 Procedure remained in suspicion as to its workability, as when compared with the 2003 Procedure, the alteration in the solvent of the 2008 Procedure was not a substantial change; and (c) that Watson's evidence on its manufacturing record and manufacturing protocol failed to prove that it followed the 2008 Procedure. For the foregoing reasons, Jiangsu Court awarded Eli Lilly damages in the amount of RMB3,500,000.
Both Watson, dissatisfied with the decision of infringement, and Eli Lilly, dissatisfied with the damages, appealed to the SPC.
III. The SPC's Findings
In this instance, the SPC summarized that "the comparison between the respective reaction paths of procedures of the parties is the key to determine whether the Watsons procedure has fallen into the scope of the '346 patent, while the starting reactant, solvent, reaction condition should not be taken into account." The SPC found that the 2008 Procedure is workable based on an in-site inspection performed by FDA of Jiangsu Province in 2009. Further, as to the reaction path in the 2008 Procedure, the SPC found it remained the same as the one recited in a "new drug clinical application" filed in 1999, which is evidence newly gathered by the SPC in response to Watson's application in the instance. The SPC concluded that "Watson had been manufacturing Olanzapine via the reaction path recited in the supplemental procedure registered in 2008, i.e., the 2008 Procedure, from 2003 to the expiration date of the '346 patent."
The SPC further determined that the Watson reaction path does not literally infringe the patent, nor does it constitute an equivalent. The SPC summarized the following differences between the Watson reaction path and the patented reaction path:
Reaction intermediate and reaction steps. Benzylated Three-Ring (i.e., thienobenzodiazepine) intermediate ("Benzylated Three-Ring") is a compound different from a non-benzylated one ("Non-Benzylated Three-Ring") in terms of characteristics of chemical reactions. Specifically, for a Non-Benzylated Three-Ring, both leaving Q radical and amine group can react with N-methyl piperazine, while for Benzylated Three-Ring, the amine group cannot - this avoids unexpected substitution on the amine group and allows only substitution on Q radical. Accordingly, the '346 patent does not comprise a benzylation and a de-benzylation steps ("Additional Steps") before and after the substitution reaction respectively.
Technical effects. Because of the Additional Steps, Watson's procedure would have a decreased yield, while the '346 patent will not. So the technical effects of the two technical solutions, such as high and low yield, are of obvious difference.
Public knowledge. Despite the fact that the benzylation of amine on the Three-Ring intermediate will reduce side effects is public knowledge in chemical synthesis field, the differences in reaction intermediate and reaction step are material. Given the changed characteristics of chemical reactions of Benzylated Three-Ring and decreased yield rendered by Additional Steps, such public knowledge only means that the improvement of Watson's procedure over the '346 patent is not significant, does not mean that the two technical solutions are substantially identical.
IV. Our Comments
The SPC reaches its conclusion of non-infringement by an analysis under the doctrine of equivalents, which is provided in Article 17.2 of the Several Provisions of the SPC on Issues Relating to Laws Applicable for Trial of Patent Dispute Cases (2015). The doctrine of equivalents in Chinese patent law is the same as its U.S counterpart, i.e., the parties should argue whether a particular element performs substantially the same function in substantially the same way to obtain substantially the same result, and whether the changed element can be easily imagined by those skilled in the art at the time of infringement.
The SPC holds that the Watson's procedure is not in the way same of the '346 patent due to Additional Steps and is not to obtain the result same of the '346 patent due to the decreased yield. The SPC's analysis and conclusion seem a divergence from the doctrine of equivalents analysis.
i) Eli Lilly's Claimed Invention
Far before the invention of Olanzapine, a series of Three-Ring compounds were found to be of anti-mental disease activity as early as the GB patent 1533235, filed on November 26, 1975. However, the compounds were observed toxic in clinical trials. Olanzapine possesses a surprisingly unforeseeable property - to treat disorders effectively and more importantly, safely.
Eli Lilly's invention is completed by the modification of the Three-Ring Core with N-methyl piperazine. Eli Lilly's claim in dispute only contains one element, the substitution of 4'Q with N-methyl piperazine ("Desired Substitution") step. In the claim, the Desired Substitution performs the function to modify the toxic Three-Ring Core in the way to substitute its 4'Q radical with N-methyl piperazine to obtain a safe anti-mental disease drug.
ii) The Function-Way-Result Test
From our perspective, Watson's Additional Steps should be qualified as an equivalent to Eli Lilly's Desired Substitution under the Function-Way-Result Test.
Firstly, the Additional Steps do not change the way and the function. In Watson's procedure, the 4'Q radical is still substituted with N-methyl piperazine and thus the Three-Ring Core is modified. The Additional Steps stem from a concern that the 10' amine on the Three-Ring Core will react with N-methyl piperazine ("Side Reaction") and thus the Desired Substitution is interfered. Watson presumed the Additional Steps would block the Side Reaction and thus facilitate the Desired Substitution. Even if Watson succeeded, this would be a mere improvement invention and would still fall into the protection scope of the '346 patent. Ironically, the Additional Steps are nothing but a roundabout as under Watson's reaction condition, the Side Reaction will almost not occur even if the 10' amine is unprotected.3
Secondly, the result is the same. The introduction of yield into evaluation is unfair.This is inconsistent with the standard set by the SPC with the focus on comparison of reaction paths. Yield, a parameter parallel to "starting reactant, solvent, reaction condition" should also not be taken into account.
iii) The Exchangeability Test
To elect benzyl - a routine protection group - to prevent the presumed Side Reaction is easily imagined for those skilled in the art. We can invoke two official conclusions. The first is the SPC's statement: "[although] it is a public knowledge that to protect the amine in the Three-Ring reductant to decrease the Side Reaction,…" The second comes from the Patent Reexamination Board of SIPO, in its decision4 in the invalidation procedure against Watson's patent No.03136388.1 (with a priority dated March 20, 2003) covering its Olanzapine process, which reads, "[i]n chemical reactions, when the presence of amine leads to the occurrence of side reaction, thus affecting the yield and purity of the product, those skilled in the art are motivated to use amine protecting group to protect amines in order to avoid the occurrence of side reactions ..."
This case gives food for thought on the practical application of doctrine of equivalents, which has been designed for circumventing insubstantial alterations and trivial changes. SPC's pro-defendant decision seemingly diverges from the doctrine and its significance remains to be determined.
1 http://law.wkinfo.com.cn/document/show? aid=MjAwODk4NTc4NDA%3D&bid=&collection=case&language=&tokens=9a95c2eb9b568a26363c4d5984ed6cf5&modules=&showType=0
No 1 (2015) civil 3 final, People's Republic of China Supreme People's court civil judgment
2 Watson's patent CN03136388.1
3 See the record of hearing on April 22, 2015.
http://www.chinacourt.org/article/subjectdetail/id/MzAwNEg1MYABAA%3D%3D.shtml
4 Decision No.11065 on Invalidation Procedure, PRB SIPO.
August 25, 2016
© Haiwen & Partners 2016
This Haiwen IP Memo is for general information only and does not constitute our legal advice or a legal opinion. For further information, please consult your regular contact at our firm.
HAIWEN & PARTNERS
www.haiwen-law.com