by Keith Syverson

In 1998, Michael Heller and Rebecca Eisenberg wrote that the proliferation of biotechnology patents created a tragedy of the anticommons where "people underuse scarce resources because too many owners can block each other."  This phenomenon is also referred to as a "patent thicket" where there are multiple different patents from different owners.  Over the past 12 years, many studies have been published evaluating the effect of intellectual property rights in medical research and diagnostic testing.  In October 2009, a group of European scientists led by Isabelle Huys investigated whether the proliferation of gene patents or patents on diagnostic methods created a patent thicket.  The article, Legal Uncertainty in the Area of Genetic Diagnostic Testing, concluded that there was no patent thicket with regard to claims that actually claim DNA sequences, but that there were enough different claims on diagnostic methods to create a chilling effect on medical diagnoses and restrict access to care.

The authors provide an empirical analysis of the nature, extent, and scope of patents relating to 22 of the most commonly tested for genetic diseases in Europe which include cystic fibrosis, breast cancer, and colon cancer.  They then compared the claims to the best practice guidelines in Europe and the United States for testing of susceptibility to these genetic disorders to determine if the patents had any effect on access to genetic testing in either country.  The authors concluded that there was not a "thicket" of patents covering only genes because most of these patents could be easily invented around.

This conclusion was inevitable because of the way the authors categorized biotechnology patents.  The authors found 145 active patents containing 267 independent claims and placed each claim in one of four isolated categories: claims covering a "polynucleotide sequence" were placed in the "gene" category; primers or probes were labeled "oligo"; genetic diagnostic methods were labeled "methods"; and genetic diagnostic kits were placed in the "kit" category.  Conclusions relying on these categories are questionable because there is substantial overlap between them.  A primer or probe, labeled as an "oligo" (which is short for oligonucleotide) is, by definition, a polynucleotide.  Moreover, a claim covering a primer or probe is designed to read on diagnostic methods for that particular gene.  For example, a patent covering a probe for the cystic fibrosis gene not only covers the nucleotide sequence for the probe, but would also prevent an authorized researcher from using that probe as part of a diagnostic of method analyzing the gene itself.

Claims placed in the "method" category can also limit the use of a genetic sequence.  For example, the authors cite the first claim of a patent covering a genetic mutation correlated with a certain type of cancer as an example of a diagnostic method patent, which they distinguish from a "gene patent."  The claims state:

1.    A method of determining a predisposition to cancer comprising testing a body sample of a human to ascertain the presence of a mutation in a gene identified as hMSH2 (human analog of bacterial MutS and Saccaromyces cerevisae MSH2) which affects hMSH2 expression or hMSH2 protein function, the presence of such a mutation indicating a predisposition to cancer.
2.    The method of claim 1 wherein the sample is DNA.
3.    The method of claim 1 wherein the sample is RNA.
4.    The method of claim 1 wherein the sample is isolated from prenatal or embryonic cells.

The specification of the patent also discloses the full cDNA sequence for hMSH2, several cDNA probes which can be used to determine if there is a mutation in the endogenous hMSH2 gene, and the amino acid sequence for the corresponding protein.  The claims read in light of the specification make it reasonably clear that any diagnostic method that relies on testing of a DNA or RNA sample would infringe the patent.  Although this claim is literally for a diagnostic method, it also provides some protection for the nucleotide sequence of the hMSH2 gene itself.  Claims like this highlight the difficulty in placing a claim into a discrete category because patents are rarely designed solely to protect a diagnostic method or a particular nucleotide sequence.   This problem of categorization calls into question the authors' conclusion that there is not a patent thicket with regard to gene patents.

The authors also claim that proposals aimed at banning gene patents are inadequate because they fail to account for diagnostic method patents.  Not only is this conclusion based on a questionable categorization, but it also oversimplifies the argument for banning gene patents.  A recent commentary published by researchers at Duke University defines the term "gene patent" as a "patent that claims rights over nucleic acid sequences encoding proteins, or variants of those sequences."   The Duke commentary points out that gene patents not only underpin the production of therapeutic biological products but also play a pivotal role in diagnostic testing.  The study indicates that a critical issue in the movement to ban gene patents is not necessarily just what precisely is patented, but how the patents are used.  Whether the patent is for a genetic sequence or a diagnostic method, problems can arise when a sole provider has an exclusive license to use the technology.  This creates a host of problems ranging from excessive cost to inability to verify results or to get a second opinion.     

The difficulties with categorizing gene patents present in the Huys study underscore the authors' ultimate conclusion that the uncertainty surrounding the scope of diagnostic method patents plays a key role in restricting patient access to modern genetic testing.  Regardless of how one chooses to categorize patents that relate to genetic sequences, problems can arise when a sole provider holds a monopoly over analyzing a particular genetic disorder.  Whether it be from blocking patents or mere uncertainty, the effect is the same: providers of genetic diagnostics must often take enormous risks in the face of potential infringement actions if they are to provide the most up-to-date and accurate diagnostic testing.