(revised September 6, 2009)
The Medicines Company (TMC) has a problem. US 5,196,404, the patent that covers its product bivalirudin (sold in the USA under the name Angiomax®), will expire on March 23, 2010.
Bivalirudin is a 20-amino acid peptide, provided in Angiomax as the trifluoroacetic acid salt, that corresponds to a portion of the naturally occurring protein hirudin. Hirudin is found in the saliva of leeches; it irreversibly binds to thrombin and thus inhibits blood clotting. Bivalirduin, on the other hands, binds thrombin reversibly, and thus resultinig in fewer instances of severe bleeding when administered to people and making it suitable for temporary prevention of blood clotting during catheterization procedures, such as angioplasty.
Sales of Angiomax are on the order of $500 million annually. In principle, TMC could have gotten a patent term extension (PTE) under 35 U.S.C. §156, but Angiomax was approved by the FDA on December 15, 2000, and TMC filed its PTE submission on February 14, 2001, missing the 60-day filing deadline by one day. TMC petitioned the USPTO to accept its PTE submission anyway, but the USPTO said no dice: a deadline is a deadline.
As some readers may recall, TMC then did what any red-blooded, deep-pocketed, well-connected American facing the potential loss of a significant revenue stream would do: it spoke to its Congressman (Rep. William Delahunt of Massachusetts) and asked him to change the law so that the PTO Director could accept TMC’s PTE application as having been timely filed. Although a version of such a bill passed in the House (see, e.g. Kevin Noonan’s post on this bill here), no corresponding bill made it through the Senate, so the expiration date for the ‘404 patent remains March 2010.
Apparently realizing that the game might be up if its lobbying efforts failed, TMC at the same time decided to resort to the tried-and-true method for extending product protection: it filed a new patent application – actually, at least three – on an improvement in the drug product. One of those patent applications issued on September 1, 2009 as US 7,582,727, and by September 3 was listed in the FDA’s Orange Book. According to the patent, the asparagine at position 9 of bivalirudin can undergo de-amination to aspartic acid both during peptide synthesis as well as during compounding of the drug product, resulting in unacceptably high levels of Asp9-bivalirudin, and causing the reconstitution time of the drug (which is sold in dry form after lyophilization and reconstituted before use) to be about 72 seconds. During compounding, this aspartic acid formation results from efforts to raise the pH of the TFA salt to a higher, more physiologically acceptable pH. So TMC developed a new process for reproducibly compounding bivalirudin in a way that limits the amount of Asp9-bivalirudin produced, and facilitates reconstitution of the subsequently lyophilized product in 42 seconds or less.
On July 27, 2008, TMC filed USSN 12/180,553, which matured into the ‘727 patent. The ‘727 patents claims “Pharmaceutical batches of a drug product comprising bivalirudin” and having either certain purity characteristics or certain reconstitution characteristics. Evidently, the concomitantly-filed and still-pending USSN 12/180,550 and 12/180,551 applications (which have not yet been published but which are referred to in the file history of the ‘727 patent) have similar disclosures but claim the process per se and the product when produced by the process.
There are several interesting aspects to this patent and its brethren. First, the patent was filed with a request for accelerated examination. Much of the patent practitioner community has shied away from such requests, in large part due to the greater likelihood of an eventual finding of inequitable conduct versus applications that do not undergo accelerated examination. This increased likelihood is the result of the various disclosure and analysis requirements imposed on the applicant under the accelerated examination process, and the correspondingly increased number of chances for the practitioner to muck something up. Nevertheless, accelerated examination requests have begun to find favor with certain applicants and in certain technologies, specifically technologies where a product can quickly be brought to market, tends to be most valuable shortly after being introduced to the market, and loses commercial value over time. This is generally not the case with pharma patents, where there usually isn’t even a product on the market at the time the first patents covering the putative product are issued. Actually, given the way patent term adjustments under 35 U.S.C. §154 are calculated (which operate without respect to §156), it may even be to a pharma patentee’s advantage to have the USPTO drag its feet during prosecution, so as to get adjusted patent term at the back end of the patent life, when every additional day of exclusivity can be worth tens of millions of dollars.
In the present case, with Anigomax already being sold, and the expiration of TMC’s lone patent looming less than two years after the filing of the ‘553 application, it made sense to request accelerated examination. Until now, a generic drug company could have contented itself with filing an abbreviated new drug application (ANDA) that stated it would not sell its generic product until after the expiration of the ‘727 patent in March 2010, a so-called paragraph III certification. By having the ‘727 patent issue before the expiration of the ‘404 patent, and by immediately listing the ‘727 patent in the Orange Book, TMC has forced generic manufacturers to assess their products in view of the claims of the ‘727 patent. This in and of itself could take time on the part of the generic manufacturers – although as will be discussed below, at least two such manufacturers apparently had no difficulty in quickly making such assessments.
More to the point, ANDA filers will not opt to wait until 2028 when the ‘727 patent expires to introduce their products. Instead, during the life of the ‘727 patent, they will have to file a “paragraph IV” certification with the FDA, stating that their products do not infringe the ‘727 patent, or that this patent is invalid. This will give the Medicines Company the opportunity to file suit against each ANDA filer under 35 U.S.C. 271(e)(2). If the ANDA was only filed after the ‘727 patent was listed in the Orange Book, the filing of such a suit will trigger a 30-month stay in the FDA’s ability to approve the ANDA.
The situation isn’t quite as rosy for TMC if the ANDA was already filed before the ‘727 patent was listed. In that case, TMC can’t get a stay against the ANDA filer who as a result of the OB amends his ANDA to include a paragraph IV certification regarding the ‘727 patent, and who promptly notifies TMC of the paragraph IV certification. TMC will only be able to file suit and attempt to get a preliminary injunction preventing approval of the ANDA while the litigation is pending (or at least for part of the litigation.) While that isn’t quite as nice as a 30-month stay, even if TMC gets a PI that runs for part of the litigation, each additional day of exclusivity will be worth over $10 million.
This, of course, assumes that TMC files suit against the ANDA filer. TMC may choose not to do so, for example if it believes that the ANDA filer doesn’t infringe the patent. (Under 21 U.S.C. 355 (j)(5)(C)(i)(III), and ANDA filer may make an offer of confidential access the ANDA, to allow the patentee to determine if it believes the ANDA filer infringes or not.) We’ll know soon enough if a suit is filed. Nevertheless, even if TMC doesn’t file suit, it will benefit by the filing of a paragraph IV certification, because from the time the first ANDA filer to make a paragraph IV certification begins to market its bivalirudin product, it will be entitled to 180 days of exclusivity versus any other generic manufacturers. Assuming that the first paragraph IV ANDA filer will only begin to market its product after expiration of the ‘404 patent, Angiomax will only have one direct competitor; and TMC may be able to further temporarily mitigate the damage through the sale of an “authorized generic”, viz. Angiomax sold under another label, which is not subject to the 180-exclusivity period.
It may be that the latter outcome was the one desired by TMC, because it appears that in terms of actually protecting TMC from competing bivalirudin products, generic manufacturers can readily avoid infringement of the claims of the ‘727 patent, which may make a colorable claim of infringement difficult for TMC and could thus make the filing of an infringement suit ill-advised. The three independent claims of the patent read,
1. Pharmaceutical batches of a drug product comprising bivalirudin (SEQ ID NO: 1) and a pharmaceutically acceptable carrier for use as an anticoagulant in a subject in need thereof, wherein the batches have a pH adjusted by a base, said pH is about 5-6 when reconstituted in an aqueous solution for injection, and wherein the batches have a maximum impurity level of Asp9-bivalirudin that does not exceed about 0.6% as measured by HPLC.
11. Pharmaceutical batches of a drug product comprising bivalirudin (SEQ ID NO: 1) and a pharmaceutically acceptable carrier for use as an anticoagulant in a subject in need thereof, wherein the batches have a pH adjusted by a base, said pH is about 5-6 when reconstituted in an aqueous solution for injection, and wherein the batches have a maximum reconstitution time that does not exceed about 42 seconds and a maximum total impurity level that does not exceed about 2% as measured by HPLC.
19. Pharmaceutical batches of a drug product comprising bivalirudin (SEQ ID NO: 1) and mannitol for use as an anticoagulant in a subject in need thereof, wherein the batches have a pH adjusted by sodium hydroxide, said pH is about 5-6 when reconstituted in an aqueous solution for injection, and wherein the batches have a maximum reconstitution time that does not exceed about 42 seconds and a maximum total impurity level that does not exceed about 2% as measured by HPLC. [emphasis added - DJF]
Thus, anyone producing a drug product having an Asp9-bivalirudin content greater than 0.6% and a reconstitution time of greater than 42 seconds would not infringe the claims of the ‘727 patent. According to statements repeatedly made by the patentee during prosecution, its own originally-FDA approved product does not meet these limitations (and hence is not invalidating prior art). Thus generic companies can produce products are bioequivalent the original Angiomax and approvable by the FDA, but do not infringe the claims.
In addition, the claims are directed to “batches”. Only 16 patents in the USPTO’s database contain the phrase “pharmaceutical batches”, and the ‘727 patent is the only one that contains this phrase in the claims. The specification defines “batches” as follows:
“As used here, "batch" or "pharmaceutical batch" refers to material produced by a single execution of a compounding process of various embodiments of the present invention. "Batches" or "pharmaceutical batches" as defined herein may include a single batch, wherein the single batch is representative of all commercial batches (see generally, Manual of Policies and Procedures, Center for Drug Evaluation and Research, MAPP 5225.1, Guidance on the Packaging of Test Batches at 1), and wherein the levels of, for example, Asp9-bivalirudin, total impurities, and largest unknown impurity, and the reconstitution time represent levels for all potential batches made by said process. "Batches" may also include all batches prepared by a same compounding process.” [emphasis added – DJF]
According to statements made by the patentee during prosecution, one of the difficulties with its own prior art compounding process was that the levels of Asp9-bivalirudin differed from batch to batch. Thus, given the patentee’s own definition of “batches” and the fact that batches of varying consistency were known, it would seem that a competitor could even avoid infringement by insuring that only some of its commercially acceptable batches fall outside the scope of the claims, even if some of those batches fall within the scope of the claims. (Of course, if the patent is upheld as valid, TMC will be able to tout its product as being purer and/or more quickly reconstitutable than its competitors’, since at least some of the competitors’ products will have a higher impurity level and/or longer reconstitution time.)
I initially posted this post on September 4. Later that day, TMC announced the receipt of of two paragraph IV certification notice letters regarding the ‘727 patent, one from Teva Parenteral Medicines, the other from Pliva Hrvatska, d.o.o. (In fact, Pliva is a subsidiary of Barr, which is a subsidiary of Teva; it’s not clear why the same parent company filed two paragraph IV certifications.) I don’t know if that constitutes a record for the time between the listing of a patent in the Orange Book and the filing of a paragraph IV challenge, but it certainly was quick. I also don’t know if the fast turnaround signifies extreme vigilance and prescience on the part of Teva, or if it signifies something else.
Another interesting aspect to the ‘727 patent is that it was filed with a non-publication request. This isn’t usually done in applications being examined under the accelerated examination, unless they claim priority to an earlier application (which is not the case here), as such applications usually issue as patents within 18 months, rendering the non-publication request superfluous. The only reason for filing a non-publication request would have been to ensure that in the two-and-half months between January 8, 2010 (the earliest date on which the ‘553 application could have published) and March 23, 2010, the application would not publish, thus allowing the patent to function as a submarine patent in the event that prosecution dragged on. But in today’s patent enforcement landscape, submarine patents don’t have quite the bite that they did in the heyday of Jerome Lemelson. Short of preventing competitors from making bivalirudin products that consistently meet the limitations of the claims, it’s not clear what benefit would have obtained for TMC had the patent issued after March 23, 2010, with the non-publication request in place: generic competitors who had already placed product in the marketplace would not be forced to recall that product; since the generic products would already have been approved by the FDA, TMC would be unable to obtain a 30-month stay in the approval process, should it have chosen to sue; and in the wake of the Supreme Court’s eBay decision and the anti-big pharma climate now prevalent in the USA, there is no guarantee that had TMC sued first the court would have issued an injunction preventing the sale of additional generic products.
On the other hand, the risk involved in filing a non-publication request is palpable: if TMC filed corresponding applications abroad and did not withdraw the non-publication request – and if it filed such applications, it would have had to do so by July 8, 2009, in which case it would have already had to notify the USPTO of such filing and withdrawn its non-publication request, an event which according to the file history did not transpire – the ‘727 patent could go down in flames under 35 U.S.C. 122(b)(2)(B)(iii).
So on its face, there would seem to be little benefit, and great risk, in filing the non-publication request in parallel with the accelerated examination request. (There would have been a small upside to having the ‘727 issue as a submarine patent: it would have provided TMC with time before grant of the patent to investigate competing products, and to be ready to file suit against competitors the day the patent was granted, albeit with a decreased ability to obtain an injunction post-eBay.)
Then again, given the ease of working around the ‘727 patent, there would be little incentive to file in jurisdictions that do not have provisions similar to those of Waxman-Hatch (i.e. all jurisdictions outside the USA), so maybe TMC really didn’t file elsewhere and its risk in making the non-publication request was minimal. TMC thus gained the benefit of possible grant of the patent before expiration of the ‘404 patent, thus setting itself up to enjoy a 30-month stay in the approval of competing products, or at least a 180-day period in which it would have only one competing product; or conversely, in the worst case, grant of the patent after competitors had already launched, giving TMC the element of surprise and the possibility of immediately suing them for infringement.
All in all, for TMC this appears to be a reasonable recovery after having fumbled the patent term extension for the ‘404 patent. We’ll be watching to see if TMC sues Teva/Pliva in the next 45 days.
