The statutory amendments to US patent law adopted in September 2011 were only the latest (if the most far-reaching) in a series of changes to US law following the major revision of 1952. Another significant change occurred on June 8, 1995, when the USA switched from a 17-years-from-grant system to a 20-years-from-earliest-filing system. The transition was accompanied by a provision that the term of any granted patent in force on that date, or which was subsequently granted on an application filed before but still pending as of that date, would be the longer of 17 years from grant or 20 years from the earliest filing date.
In anticipation of the change, there was a rush of filings in early June 1995, as applicants attempted to file applications that would benefit from this “best of both worlds” situation. As reported by Dennis Crouch on Patently-O earlier this year, there are still a few hundred applications pending that were filed before the change was made that, if allowed, will be in force for 17 years from grant. (Conversely, the filing of a continuation application after June 8, 1995 subjects the application to the 20-years-from-earliest-filing-date provisions; Aaron Feigelson has found more than few patents that as a result of the change were “dead on arrival”.)
Amgen made good use of the change in the law. It has five US patents filed before June 8, 1995 that trace their lineage to the 1980’s, some as far back as 1983, all of which issued later; three of these were filed on the eve of the statutory switch: no. 5,547,933 (filed June 7, 1995, granted August 20, 1996); 5,618,698 (filed June 6, 1995, granted April 8, 1997; 5,955,422 (filed August 2, 1993, granted September 21, 1999); 5,756,349 (filed June 6, 1995, granted May 26, 1998), and 5,441,868 (filed in 1987, granted in August 1995).
Collectively these patents cover Amgen’s blockbuster drug Epogen® (recombinant erythropoietin, “EPO” or epoetin alfa). In 2007, several of the patents were found to be infringed and not invalid, resulting in Roche being unable to bring its competing Mircera® product to market; in 2009, Roche and Amgen settled their dispute, with Roche stipulating to the validity and infringement of all five patents, and agreeing to keep Mircera off the market until July 1, 2014. Epogen was first approved by the FDA in 1989, meaning Amgen secured 25 years of effective patent protection for this drug in the USA.
A related Amgen product is Aranesp® (darbepoetin alfa), which like Epogen is used to treat anemia in certain situations. According to the prescribing information, darbepoetin alfa is a modified version of erythopoietin (EPO); in particular darbepoetin alfa
“is an erythropoiesis-stimulating protein that is produced in Chinese hamster ovary (CHO) cells by recombinant DNA technology. Aranesp is a 165-amino acid protein that differs from recombinant human erythropoietin in containing 5 N-linked oligosaccharide chains, whereas recombinant human erythropoietin contains 3 chains. The 2 additional N-glycosylation sites result from amino acid substitutions in the erythropoietin peptide backbone. The approximate molecular weight of darbepoetin alfa is 37,000 daltons.”
As with EPO, in the case of Aranesp Amgen made good use of the change in US patent term calculations: it filed USSN 08/479,892 on June 6, 1995, claiming continuation and continuation-in-part priority back to a 1989 application. The ‘892 application matured into US Patent No. 7,217,689 on April 25, 2007, with an nominal expiration date 17 years later, viz. April 25, 2024. Aranesp was approved by the FDA in 2001, meaning the drug will enjoy an effective 23 years of patent protection in the USA if the ‘689 patent isn’t invalidated.
Where the story gets interesting, at least from an Israeli perspective, is when we look at what happened in the corresponding case in Israel. Aranesp was approved by the Israel Ministry of Health in 2002; on the patent side, Israel patent no. 110669 was filed in 1994 and claims priority from USSN 08/108016, a 1993 application in the US priority chain for the US ‘689 patent. IL 110669 was granted in 2009 and, absent a patent term extension (PTE), will expire in August 2014.
The patentee filed a request for a PTE within the prescribed time after the grant of IL 110669. But here it ran into a technical problem. The statute defines a “basic patent”, which is the Israel patent for which the PTE is sought and that “protects” the API (or formulation, or method of making either); and the statute defines something called a “reference patent”, which is a patent in any one of several recognized jurisdictions that also protects the API (or formulation, or method of making either). Section 64D(5) of the statute says that if the FDA approved the patent-protected product or process, then in order to receive a PTE in Israel, a “reference patent” in the USA must have received a PTE from the USPTO.
In its Israel PTE application, Amgen stated that the reference patent in the USA is the ‘689 patent. That makes sense: claim 3 of IL 110669, for example, reads,
3. An analog of human erythropoietin which is: Asn30Thr32Val87Asn88Thr90 EPO.
which is quite similar to claim 23 of the US ‘689 patent, which reads,
23. An analog of human erythropoietin comprising the amino acid sequence of human erythropoietin from residues 1–165 as shown in SEQ ID NO: 26 except for the amino acid changes Asn30Thr32Val87Asn88Thr90 EPO.
The problem for Amgen was that no PTE was ever issued for the US ‘689 patent, which according to 64D(5) means that IL 110669 isn’t eligible for a PTE either. Indeed, no PTE for the US ‘689 could have issued, since a PTE in the USA can’t extend the patent more than 14 years from the date of FDA approval, viz. 2015, and here the US ‘689 patent won’t expire until 2024, so the patent wasn’t even eligible for a PTE.
Realizing it had a problem on its hands, Amgen argued in its Israel PTE application, and then in ex parte proceeding before the ILPTO Commissioner, that it was nevertheless entitled to a PTE for IL 110669, inter alia because the term of the US ‘689 “reference” patent had effectively been extended by virtue of the change in US law. Citing some earlier district court and ILPTO decisions, in a written decision from September 27 the Commissioner rejected Amgen’s arguments, on the grounds that the statute was clear and that it wasn’t his place to engage in “purposive construction” of the statute when his task in deciding PTE applications was purely ministerial. Amgen had FDA approval for the drug but no PTE for a US patent protecting the drug, hence he could not issue a PTE in Israel.
Thus Amgen’s choice to file June 6, 1995, and thereby obtain the “best of both worlds” for its US patent term, doomed its chances of obtaining a PTE in Israel. Of course, at the time, there was no PTE legislation in Israel, and even if there was, given the disparities in the two markets, and the billions of dollars Amgen has made and will continue to make with Aranesp, this one would have been a no-brainer.
Interestingly, had Amgen chosen to file the ‘892 application two days later, when it would have been subject to the 20-years-from-earliest-filing limitation, it probably still would have lost out on a PTE in Israel: the US ‘689 patent – the one that issued in 2007 – would then have been set to expire in 2003, four years before it issued, and by dint of the FDA’s 2001 approval of the drug, would have been eligible for a PTE only until 2006, a year before the patent issued. So even with the grant of temporary PTEs by the USPTO until 2006, Amgen may well have found itself in the same boat as far as Israel goes. Then again, maybe not: as noted above, the USPTO isn’t averse to issuing patents that are DOA, so perhaps a 2007 grant with an earlier, albeit temporary, PTE, would have been sufficient to gain PTE eligibility in Israel.
[Editorial comment: the Commissioner reached the right decision in this case, but it would be nice if he and his staff would start being equally circumspect in their reading of the statute on other matters, such as “overlap” between applications, or with regard to the timing of the filing of divisional applications, both areas in which the ILPTO has in recent years invented new laws not found in the statute.]