By Laura Karas
Are manufacturing method patents — patents not on a pharmaceutical drug itself, but on a method of production of a drug — warranted intellectual property protections, or groundless obstacles to competition?
Patents protect and reward innovation by permitting the patent-holder the exclusive right to make, use, and sell the invention for a twenty-year period. Pharmaceutical companies have attracted scrutiny, criticism, and legal challenges for amassing large numbers of patents on pharmaceutical drugs, especially high-priced and high revenue-earning drugs.
Here I explore the topic of pharmaceutical patents on methods of production and translate into layman’s terms some thought-provoking recent scholarship by innovation scholars W. Nicholson Price and Arti Rai.
Are Methods of Production Patentable?
What is — and is not — patentable subject matter continues to be a point of contention among patent-seekers, patent-holders, and those who want to replicate patented inventions.
The Patent Act itself makes clear that “any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof” may be patented (35 U.S.C. § 101). Methods (i.e., processes) are, therefore, among the categories of patent-eligible subject matter.
These categories are subject to the judicially-created exceptions for laws of nature, natural phenomena, and abstract ideas. But a number of gray areas remain. For example, products of nature cannot be patented, but transgenic plants and animals and microorganisms that are the product of human invention may be patented. Naturally occurring human genes cannot be patented, though recent federal bills have sought to change that, whereas synthetic DNA can be patented.
Do Method Patents on Biologics and Other Drugs ‘Follow the Progress of Science’?
Even business methods may be patentable, as established in the landmark 2010 Supreme Court case Bilski v. Kappos. As Professor John Duffy has argued in the Stanford Law Review, the rise of business method patents did not result from judicial activism, but rather followed a technological transformation in business fields like marketing and finance, which were heavily influenced during the later decades of the twentieth century by engineering and the growth of quantitative methods: “[p]atents followed the progress of science and technology,” and “[t]he courts validated that development only later.”
A similar argument could be made for patents on biologic drugs: these drugs are quite complex, both in structure and in method of production, and are quite unlike small-molecule drugs. They represent a more intricate and elaborate form of pharmaceutical innovation, and so warrant a larger number of patents. Consistent with this view, the courts, by not ruling more aggressively against challenged method patents (and other patents) on biologic drugs, are responding to larger developments in biopharma and biotechnology. The patents have followed the progress of science, and the courts have validated that development. A one-patent-per-drug system simply would not reward the complexity of the innovation that biologic drugs (as well as gene therapies and other recent pharmaceutical marvels) encompass.
Indefensible Manufacturing Method Patents
The validity of method patents as a class does not mean that every method patent is defensible. In a recent article, legal scholars W. Nicholson Price and Arti Rai offer a theory as to why invalid manufacturing method patents may be the “culprits” in patent thickets that unjustifiably extend total patent protection on biologics and so block biosimilars from entering the market.
According to Price and Rai’s argument, manufacturing methods in use at the time of a drug’s launch — but for which patents are sought more than one year post-launch — cannot be patented because of the bar on patentability of inventions in commercial use for more than one year prior to filing. The authors write: “No U.S. patent can be granted if the patented invention was used commercially in the United States more than 1 year before the patent was filed.”
This argument is consistent with the most recent revisions to patentability that took effect with the Leahy-Smith America Invents Act (AIA): one cannot obtain a patent on works that are “patented, described in a printed publication, . . . in public use, on sale, or otherwise available to the public” before one files (35 U.S.C. § 102(a)(1)). Such works are considered prior art and cannot be patented, with an exception for a disclosure made by the inventor within one year of filing (35 U.S.C. § 102(b)(1)).
Thus, the authors are correct that if an invention (here a manufacturing method) was in public use more than one year before the filing date, the inventor is not entitled to a patent.
Manufacturing method patents, their argument continues, on methods invented post-launch must not have the capacity to block biosimilar market entry, because such patents “must not be necessary” to produce a biosimilar to the originator drug (presumably since the drug maker that produced the originator used an alternative method). Here, the authors’ argument is less convincing, because post-launch methods may well be necessary to replicate a drug’s production. Drug makers enshroud manufacturing methods, whether pre- or post-launch, in secrecy. A patent on a post-launch method may obstruct production of a biosimilar just as well as a pre-launch method patent when the pre-launch manufacturing method remains unknown or incompletely known to the maker of the biosimilar.
Although the authors call such pre- and post-launch manufacturing method patents “logically impossible patents” for the reasons described above, logic is not really at issue here. The issue is patentability. (Perhaps a better descriptor could be “statutorily indefensible patents.”) The authors cite Helsinnn Healthcare v. Teva Pharmaceuticals, Inc., holding that, under the AIA, a secret sale can qualify as prior art (and, by analogy, secret use of a manufacturing method in commercial production should also qualify as prior art).
But it is difficult, if not impossible, to police a manufacturing method in commercial use more than one year prior to filing, because only the drug maker knows what manufacturing methods were used to produce the drug. Despite the legal soundness of Price and Rai’s objections to the validity of such patents, their arguments have little force in current practice. (For this reason, they suggest in a more recent article establishing cross-talk between the FDA and the USPTO regarding manufacturing processes in use at the time a drug is approved.)
Finally, I question the authors’ statement that “competitors must cut through the entire thicket to win market entry.” Competitors are capable of innovating. They, too, may develop novel methods of production — for which they, too, can seek patents — and those novel methods can bypass an originator company’s method patents and permit an earlier market entry than would otherwise be possible.
Are Manufacturing Method Patents to Blame in Patent Thickets?
In sum, methods are within the realm of patent-eligible subject matter. Thus, drug makers rightly seek to patent methods of production. Whether a particular method is anticipated by prior art (because it was available to the public, or was “in use” by that very drug company, albeit secretly, more than one year prior to filing) is a different issue, and one that is much harder to discern.
Whether invalid manufacturing method patents are the main obstacle in a typical patent thicket — as opposed to method-of-use patents, formulation patents, or other secondary patents — remains an open question. A related article by Price and Rai published last month in Nature Biotechnology reports that nearly two-thirds of patents asserted across 34 cases of BPCIA litigation were filed more than one year after FDA approved the originator biologic, and of these, nearly three-quarters were manufacturing process patents. This suggests a key role for potentially invalid manufacturing process patents in blocking timely approval of biosimilars.