A Myriad of Loose Ends

This past June, the Supreme Court held in Association for Molecular Pathology v. Myriad Genetics Inc., 569 U.S. __ (2013) that DNA segments encompassing genes cannot be patented. The ACLU and other pro-access organizations rejoiced in victory [1]. Quietly, so did the biotechs [2]. The details of the Supreme Court decision leave many major concerns unaddressed for both sides, perhaps explaining this discrepancy in responses. By examining the two holdings and an important dictum of this case, and contextualizing their biotechnology implications, it is apparent that this case is far from a final say in the question of “patenting life.”

The Court’s two important holdings are: “a naturally occurring DNA segment is a product of nature and not patent eligible merely because it has been isolated,” and “cDNA is patent eligible because it is not naturally occurring.” Id. at 1. Importantly, the court also established the dictum that in analyzing similar cases, they will apply a standard “to determine whether… patents claim any new and useful… composition of matter… or instead claim naturally occurring phenomena.” Id. at 2. All three of these pronouncements are problematic in real-world biotechnology settings.

The First Holding

Scientists have made tremendous progress in the last 30 years with biotechnologies. Whether by introducing useful genes into new host organisms (i.e. cold- [3] or drought- resistant [4] crops), mass producing vital medical proteins (i.e. insulin [5]), or biometrically tracking animals [6] and humans [7], molecular biology has made a tangible and positive impact on humanity. At the heart of this revolution is the role of DNA manipulations, which is exactly what the Myriad decision focuses on as well.

 DNA [8] is extremely accessible to various technological manipulations. DNA is very stable (half-life of 521 years [9]), it only has 4 “codes” (A, T, C, and G), and its coding pattern is invariable across all known life (A pairs with T, and C pairs with G, in an elegant weave that spirals into a double-helix). But what gives DNA its biggest technological relevance is its ability to be amplified.

An individual DNA strand is only 2.5 billionth of a meter wide. For most of the aforementioned applications, the individual DNA strands naturally found within a cell are technologically intractable. A Nobel-winning technique called PCR (polymerase chain reaction [10]) allows for the exponential amplification of the minute quantities of naturally occurring DNA. Using PCR, within 3 hours, naturally occurring DNA segments that are 2.5 billionth of a meter wide can be amplified into a half-inch band visible to the naked eye. And it is these bulk, amplified copies of DNA that are most commonly utilized in biotech applications. Therefore, in practice, it is almost always the copies of “naturally occurring DNA” that are being used, not the “naturally occurring” segments themselves. This technique is now so fundamental to biology that every molecular biology laboratory has a PCR machine.

Unfortunately, this particular innovation seems lost on the Supreme Court’s analysis. The Court focused on the isolation of “naturally occurring DNA” as the prohibitive limits of what is not patentable. Under this narrow interpretation, one could argue that a PCR amplification of a DNA segment may be patentable. That is, when amplifying DNA by PCR, scientists add in synthetic ingredients to make trillions of copies of the “naturally occurring” segment. Because in the subsequent biotech applications, it is these synthetic copies which are being utilized, this holding creates a substantial loophole for biotechs: as long as they use synthetically copied (or otherwise synthetically generated) DNA segments, could that still be patent eligible?

 The theoretical misunderstanding that underpins this loophole is the conceptual classification of DNA as tangible matter rather than intangible information. The value of a book is largely not in the actual ink or paper, but rather in the ideas conveyed by the ink on the paper. Similarly, the worth of four letter codes of DNA as a physical molecule is trivial compared to their worth in storing genetic information. In the book analogy, not only is stealing the physical original copy of the book against the law, but so is the unauthorized dissemination of the information within that book. Analogizing to the contrapositive, if the isolation of the physical molecule of information storage (“naturally occurring DNA segments”) is not patentable, then the information stored within shouldn’t be either. While that may have been the intent of the holding, DNA is very easy to physically manipulate. Therefore, limiting the holding to what can or cannot be done with “naturally occurring DNA” misses the central problem in patenting life.

The Second Holding

To the Court’s credit, it partially addresses this problem in its second holding when it states that cDNAs are patentable because “a cDNA sequence… is not naturally occurring.” This is a much more apropos framework: to examine the sequence of information, rather than the molecular entity. However, this holding has an even bigger problem.

Let me begin with an eight-sentence primer on molecular genetics. Every cell in the human body has the same DNA. However, the reason a skin cell is different from a liver cell is because different portions of the DNA are turned on in different cells. Our DNA is organized into approximately 20,000 functional units called genes. Each gene gives directions to make a specific protein, which are the cellular machinery that perform the work of life. Therefore, what genes are turned on determines the functional characteristics of a cell, thus creating performance diversity despite the same starting DNA material. Over the last 60 years, scientists have discovered nature’s two-step process of making protein from DNA. First, a gene is copied and its “junk” sequences are excised from this copy to form an intermediate called the mRNA. Then, the mRNA is “read” and translated into amino acids that chain together to form proteins.

Unlike DNA, which is very stable, mRNA is not. The half-life of mRNA typically varies from hours to days, compared to DNA’s robust 521 years. Therefore, in studying genes, scientists frequently copy the mRNA into a DNA format; this copy is called the cDNA. cDNA is easier to store and use in subsequent experiments. This very standardized copying process (reverse transcription) takes about an hour. However, as the petitioners in the case correctly point out, ultimately, “the nucleotide sequence of cDNA is dictated by nature, not by the lab technician.” Supra at 17. To put it bluntly, cDNA is simply a copy of the naturally occurring molecule of mRNA. The holding that this is somehow patentable is extremely problematic because copying is not innovating.

Practically speaking, in almost all biotech settings, the standard practice is to utilize cDNA rather than DNA. This is because cDNA does not contain the “junk” sequences which make downstream production impossible under many settings. Thus, after discovering the DNA sequence of a gene, scientists almost always capture and copy its mRNA to make cDNA for subsequent analysis. To put it another way, almost every DNA gene ever discovered has had its cDNA synthesized due simply to the normal workflow of a molecular biologist. Therefore, to hold that a gene’s DNA cannot be patented but its cDNA can is patently pointless.

Composition of Matter v. Naturally Occurring Phenomenon

The Court then draws on two life sciences precedents in Myriad to delineate its distinction between what is patentable and what is not. In what it describes as an unpatentable “naturally occurring phenomenon,” it points to the example of Funk Brothers Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948). A farmer decided to physically mix several strains of naturally occurring soil-aerating bacteria into a single inoculant, so that the mixed inoculation helps aerate soil better than inoculants of the individual bacteria. The Court held that this was not patentable because mixing “did not alter the bacteria in any way.” Supra at 13. However, in what the Court describes as a patentable “composition of matter,” it points to the example of Diamond v. Chakrabarty, 447 U.S. 303 (1980). In that validly upheld patent, scientists added four genes to a bacterium, which enabled new traits. In the Court’s reasoning, the former was just the work of nature, whereas the latter invoked the work of science to change nature, thus making it patent-eligible.

In my opinion, this distinction seems artificial: what is “natural” and what is “science” is limited simply by our knowledge of science. What if, not-so-hypothetically, it was discovered that certain bacteria in the presence of other bacteria exchange genes with each other (in a now well-known process called conjugation)? A scientist wishing to do what Chakrabarty accomplished may opt to use one bacterium as a vector to add genes to another bacterium. But isn’t she simply performing a mixing of bacteria that Funk Brothers held wasn’t patentable?

More generally, few people will have qualms about giving most “medicine” (i.e. aspirin, Viagra, etc.) some level patent protection when it is first made. But if asked to compare between patenting “medicine” and patenting a gene, common sense would probably guide most people to argue that the former is patentable because it is somehow synthetic; “medicine” is made by science working on nature. Like the Supreme Court in Myriad, seeing something as a molecule or a chemical makes it easier to understand its discreteness as an item, as well as its potential to be evaluated for patent eligibility. However, this theoretical line is blurring with each advance of science. Antibiotics, for example, are products of nature; specifically, there are genes which can naturally synthesize “medicine” like penicillin. If this “medicine” can be patented, can the genes that make it also be patented? Despite our advances in the last half-century, our knowledge of all of life’s biochemistries is still quite limited. Newly discovered genes confer so many previously unfathomable abilities, from oil-eating bacteria [11] to pesticide-secreting plants [12]. As we discover more and more “naturally occurring phenomenon” that can synthesize various kinds of downstream “composition of matter,” does it still make sense to allow only the end product, but not the upstream genetic synthesizer, to be patented?

A small piece of suggestion

 Life science patents have an especially difficult challenge in fathoming the exact line between nature itself and the products of nature. Unlike engineers who invent or chemists who synthesize, by trade, biologists discover. Therefore, I am of the opinion that because so much hinges on nature as the raw ingredient in the life sciences, it is unfair to deny a patent simply because it rests largely on the products of nature. However, a patent needs to demonstrate a full suite of innovation, from the intangible natural information (i.e. genes) to the tangible downstream effectors (i.e. drugs). Patenting of genes in and of themselves is insufficient because there need to be demonstrated downstream applications. As the Court correctly points out in Myriad, “as the first party with knowledge of the [gene] sequences,… [a company] is in an excellent position to claim applications of that knowledge.” Supra at 17.

For example, if I were to be the first to discover the gene for making the antibiotic qenicillin, under my full suite requirement scheme, my patent should only be granted if I included the gene discovered, its DNA sequences, the isolated and purified qenicillin drug that this gene directs a cell to make, and the demonstrated application of this new drug. A patent scheme that has this full suite requirement would be felicitous for three reasons.

First, it is consistent with previous patent schemes. A patent for a novel drug, such as an antibiotic or any other biologic, would normally be granted through the usual channels. In this scheme, I am simply piggybacking onto a well-established framework of drug patents an additional claim of upstream DNA/genetic information.

Second, my approach gives appropriate incentives for further research. Many “basic science” discoveries today, such as deciphering genes’ DNA sequences (“annotating” a gene), are made in academic settings using tax-payer funding. (The BRCA1 gene in question for Myriad Inc. was first discovered at UC Berkeley and University of Utah.) It would be unfair if a spinoff for-profit company from an academic lab can then deny others from studying this gene after merely a perfunctory effort. Instead, by requiring a company to perform the full gamut of discovery, validation, and application, there is proper incentive for good faith effort and great disincentive for genetic patent trolling.

Undoubtedly, it is always a careful balancing act between providing incentives and obstructing competition. Genetic patent trolling would be especially egregious in the age of computational biology because of how easy it is to simply annotate and then claim a segment of the already-published human genome [13]. Giving a patent for merely rummaging through public information obstructs the rights of others to better investigate the gene in question. That is precisely the reason why I believe taking the suite approach provides the incentives to expend in-depth effort to fully develop a medically relevant discovery and a concomitant applicable product. Admittedly, this will still generate inequities because companies are likely to be the ones to fully actualize a “basic” research discovery made by an academic. However, unless research universities commit themselves to developing, manufacturing, and dispensing drugs based off of their discoveries, one could argue that it is the academics who are obstructing social welfare. That is, if academics won’t develop and manufacture the drugs that society needs (and paid for), why shouldn’t society incentivize someone else who can and will? 

Third, my approach gives the Courts an easier time navigating technological advances for more consistent holdings. Whenever a court makes a holding with reference to a specific stage of life sciences workflow, future advances will inevitably circumvent the technicality (i.e. ease of making cDNA), making the holding futile. But by using a suite approach, the Court only has to focus on the big picture of patent function, with the genetic information only constituting a subcomponent, making life sciences patents more comparable to traditional fields.

I’ll conclude by noting that it’s interesting that the Supreme Court made a point to declare in Myriad that effort does not correlate with innovation and patent-worthiness. Supra at 14. Philosophically, that is absolutely correct; practically and technically, that is questionable. Every glaring loophole I have pointed out is undergirded by the notion of how effortless it is to circumvent the Myriad ruling: an average molecular biologist can easily change what is Myriad-unpatentable to become Myriad-patentable. The Supreme Court should not be drawing lines at the technical level; it’s just not good at it. Determining what is good faith effort to innovate beyond the obvious requires keeping abreast of biology’s advances. Therefore, if the Courts continue to rule on the basis of technology, the landscape of life sciences patents will continue to fluctuate with each advance in technology. Instead, if the courts ruled on the conception of DNA’s stored information and on the paradigm of full-suite product development, then its lines in the sand won’t be washed away so readily by the tides of technological advance.

[1] https://www.aclu.org/blog/womens-rights-free-speech-technology-and-liberty/victory-supreme-court-decides-our-genes-belong