The Supreme Court’s Sketchy Science

Their BRCA patent ruling reads like an earnest seventh grader’s book report.

A scientist of Brazilian Enterprise for Agricultural Research analyzes vegetables samples at the National Center for Genetic Resources laboratory in Brasilia on Dec. 19, 2012.

A scientist of Brazilian Enterprise for Agricultural Research analyzes vegetables samples at the National Center for Genetic Resources laboratory in Brasilia on Dec. 19, 2012.

Photo By Evaristo Sa/AFP/Getty Images

In a unanimous ruling yesterday, the Supreme Court held that patenting a gene violates Section 101 of the Patent Act. Though Myriad Genetics had indeed made “new and useful” discoveries, these fell into a long-held exception that “laws of nature, natural phenomena, and abstract ideas are not patentable.” However, the court found cDNA (complementary DNA) to be patentable. The meaning of this ruling is complicated significantly by the court’s sketchy understanding of molecular biology. Justice Scalia, in a concurring opinion, was honest (or exasperated) enough to make his lack of expertise explicit, describing his inability to understand on “[his] own knowledge or even [his] own belief,” the “fine details of molecular biology” relevant to the case.

At issue were several patents held by Myriad Genetics on BRCA1 and BRCA2 genes. Myriad discovered that certain mutations in these genes are closely linked with an increased risk of breast cancer. This was an extraordinary medical discovery and one that came at a great expense. (Incidentally, not all of the expense was Myriad’s—much of the work was done at research universities and paid for by public grants.) In order to monetize this discovery, Myriad patented both the genes themselves and the cDNA of those genes. cDNA stands for complementary DNA (not “composite DNA” as the court wrote inconsistently in its introduction to the opinion), and the court’s misunderstanding of it demonstrates that science policy should not be dictated by people who don’t have a firm grasp of the science. What the patentability of cDNA means for the biotech industry is not obvious and may require significantly more litigation to sort out. Since the justices’ understanding is a patchwork of loosely connected facts, asking them to decide a case like this is a gamble.

Fortunately, this time the court made a decision scientists, doctors, and patients will be happy with. Justice Thomas, writing for the court, pointed out that “groundbreaking, innovative, or even brilliant discovery does not by itself satisfy the §101 inquiry. … Myriad found the location of the BRCA1 and BRCA2 genes, but that discovery, by itself, does not render the BRCA genes ‘new … composition[s] of matter,’ §101, that are patent eligible.”

From a biological perspective, the issues in this case seem strange. The court spent a lot of time during oral arguments trying to find the right analogy to explain exactly what the scientists are doing and whether or not the product of their work is patent-eligible. Removing a gene from the genome became taking a branch off a tree or a medicinal leaf from a plant in the Amazon. In the ruling itself, there was a lengthy opening section explaining the central dogma of molecular biology. It reads like an earnest seventh grader’s book report. The section explains, in a simple way, the process of protein production, focusing on which reactions are and are not done “naturally.” This was used to set up the ruling that cDNA is patentable because it is not naturally present within human cells. The court ruled that the cDNA patents do not violate Section 101 of the Patent Act (though they “express no opinion whether cDNA satisfies the other statutory requirements of patentability”). Well, what exactly is cDNA and why isn’t it natural?

The process by which that information in DNA is converted into proteins involves two basic steps, transcription and translation. In bacteria, DNA is transcribed into RNA, which can be translated directly into proteins. In humans, things are more complicated: The RNA that is transcribed has some sequences that will be translated into proteins (exons) and others that must be spliced away (introns). Once the splicing is complete, the RNA can be translated properly. If a scientist wanted to produce a human protein in bacteria (a common molecular biology technique), the introns would have to be removed from the original DNA before it is inserted into the bacteria because bacteria cannot splice RNA. This cleaned-up, exon-only DNA is called cDNA. To make it, scientists take the fully spliced RNA and “reverse transcribe” it. That’s exactly what it sounds like: The information in the RNA is converted back into DNA form without the original introns.

This method is still used today, but it’s not the only way to get an intron-free DNA sequence. There are companies that will make the DNA for you if you just send them the sequence you want, though it’s possible the patent on cDNA may eventually be ruled to cover such DNA as well. This ambiguity underscores how outdated these patents are. Cases like this one are the death throes of 1990s biotech. Fortunately, patents on genes and cDNA from the 1990s are set to expire pretty soon.

The focus on cDNA is particularly frustrating because the difference between cDNA and natural DNA is a purely technical matter. As some of the amici curiae pointed out, the information content is the real issue. In his amicus brief, James Watson, the co-discoverer of the structure of DNA, uses the word “information” dozens of times but doesn’t mention cDNA once. cDNA is a tool, and barring its use is entirely arbitrary. The information content of the genes is what the patents should or should not cover.

The court’s strain to understand this science is manifest in the overly respectful, declarative language used in the ruling. Statements such as “the study of genetics can lead to valuable medical breakthroughs” are simply adorable.

Once the court established that genes are products of nature and therefore ineligible for patent, it destroyed Myriad’s ability to be the exclusive beneficiaries of the BRCA discoveries. Myriad provides genetic testing for mutations in the BRCA1 and BRCA2 genes. Myriad will tell you, for a fee, if you have a mutation that increases your risk of breast cancer. This service made a lot of sense in 1994, when Myriad first filed a patent for BRCA1. Today the sequencing of genes and even whole genomes is routine. The idea of discovering a gene after the completion of the human genome project is like discovering a new island: It made sense before we had satellite images but today it is outdated.

The true benefit of this ruling will come from the ability of more and more scientists to look at large-scale datasets of BRCA mutations and cancer rates. Right now only Myriad genetics has the right to sequence the BRCA genes. This has had several effects. Myriad sets a price for its testing that is exorbitantly high—they charge $3,400 to test two genes. The price of sequencing an entire genome (which is about 75,000 times bigger) was already only $5,000 in 2011.

Because all patients must rely on a single company for both their test and analysis, other researchers have not had access to Myriad’s data. Some BRCA mutations have “unknown significance” and researchers have complained about the “chilling effect” this lack of data has had. It becomes personal very quickly when you have to make a decision whether or not to get a mastectomy based on research coming out of one, and only one, company.

When technical issues are at the heart of a case with broad public implications, a nuanced understanding of those issues is essential. Not everyone needs to understand molecular biology, but the court should have more than a haphazard understanding if it is going to continue to sort out biotech cases, especially ones with the potential to have so much impact on medical research and patient care.