Personalised medicine promises to transform health care through innovative diagnostics that more precisely stratify patient populations, enabling health care providers to offer more tailored medical therapies, and will better inform treatment decisions. The objective is improved patient outcomes and, presumably, lower overall health care costs. A number of legal decisions however, may alter the relevant intellectual property landscape.

In particular, a lawsuit filed recently by the American Civil Liberties Union could have broad repercussions throughout the entire biotechnology sector, but especially for DNA diagnostic companies and the nascent personalised medicine industry.

The suit, filed on behalf of women’s health groups, individuals and scientific associations, against a diagnostics company that markets a genetic test to assess risks of breast or ovarian cancer, alleges that the patented genes and diagnostics are “products of nature” and “abstract ideas or basic human knowledge”, rendering them invalid and unconstitutional.

Current uncertainty regarding the control of the underlying genomic sequence and gene-disease correlation information potentially limits research and development opportunities and follow-on diagnostic tests. But, at the same time, intellectual property rights often provide incentives to discover fundamental correlations to achieve more advanced diagnostics to improve patient outcomes.

Patent Law and Personalised Medicine

For decades the U.S. patent doctrine held patentable “anything under the sun that is made by man” excluding only laws of nature, natural phenomena, abstract ideas and mental processes. Recently, U.S. courts have, arguably, further constrained this broad expanse of patentable subject matter, threatening previously patentable items such as biologics, proteins and DNA. Importantly, like subject matter jurisdiction, patentable subject matter—a question of law—is a threshold criterion that can completely invalidate a patent.

Lab Corp. v Metabolite

Justice Breyer’s dissent in the Supreme Court of the United States’ reversal of certiorari arguably set off this recent patentable subject matter focus. Lab Corp, generating 20 amicus briefs, involved a process to diagnose vitamin def iciency through correlating the patient’s homocysteine levels. Justice Breyer noted several important policy reasons for limiting these types of process patents, as they threaten “…to leave the medical profession subject to the restrictions […that] may force doctors to spend unnecessary time and energy to enter into license agreements … [that] may raise the cost of health care while inhibiting its effective delivery.”

In re Bilski

In Bilski, the patentee’s patent on a method of arbitrage was found by the U.S. Court of Appeals for the Federal Circuit to fall outside of the realm of patentable subject matter. In outlining the boundaries of 35 USC Section 101, the en banc court ruled that a method claim must be tied to a particular machine, or “…by showing that his claim transforms an article, [and that] use of a specific machine or transformation of an article must impose meaningful limits on the claim’s scope to impart patenteligibility [and that] the involvement of the machine or transformation in the claimed process must not merely be insignificant extra-solution activity.”

Applying the new Bilski test, the Federal Circuit later affirmed a district court’s decision finding that an examination of the correlation between vaccination schedules and the risk of developing chronic immune mediated disorders is unpatentable as a discovery of natural phenomena.

Prometheus v Mayo

The Federal Circuit may likely affirm a further limitation on patentable diagnostics as unpatentable discoveries. Prometheus involves patents for a process of measuring metabolised drugs to determine an optimal patient dosage.

Post-Bilski

The repercussions of these decisions are still unclear, particularly in the biotechnology sector where Bilski seems an unnatural fit. Further, the case law has yet to show that Bilski can be applied beyond its particular facts to include composition patents as well as methods. The U.S. Patent and Trademark Office (USPTO), in its broad application of Bilski, has since rejected claims in numerous patent applications. Similar claims have nevertheless survived in the district courts.

Under a simple reading of the case law, a patented correlation that might be seen in the application of a patent relevant to personalised medicine, (e.g., a method for predicting a patient’s response to a drug in relation to a level of biomarker) would be unpatentable under the post-Bilski understanding of 35 USC Section 101. The Supreme Court of the United States recently granted certiorari and, given its recent history in patent cases, this newly constituted Supreme Court could alter or completely dismantle the bright line rule established in Bilski. Moreover, future patents may be written to skirt adeptly the Bilski decision. Regardless, the current gene diagnostic patent population remains at risk.

Non-Obviousness and Anticipation

In addition to subject matter jurisprudence, patents related to personalised medicine are further and potentially more seriously threatened by the raising of the 35 US Section102 bar to non-obviousness and the lowering of the 35 USC Section 102 (b) bar to anticipation.

In re Gleave recently upheld a USPTO rejection of a DNA-based therapy as anticipated under a generic genetic sequence disclosure. The court noted that the prior publication of the sequence itself was sufficient to allow one skilled in the art to clone the gene. Utility need not be disclosed in the art to make a genetic sequence anticipatory, although it is still required to patent those genes. As a result of the numerous genome projects that sequenced, annotated and published the overwhelming majority of many medically relevant genomes, it is now more likely than not that a gene sequence will in exist in the art, likely anticipating most gene patents.

In re Kubin affirmed a USPTO rejection of a protein patent for obviousness over the combined prior art teachings and the known conventional techniques for cloning proteins. Not only re-invigorating pre-In re Duell’s obvious to try doctrine, the court’s ruling also applied the Supreme Court’s revised non-obvious doctrine to the non-predictable arts, i.e., biotechnology, making it possible for more of these patents to be found to be obvious and also, potentially, more difficult to initially obtain patents on genes.

Conclusion: Effect on Personalised Medicine

Determinations that genes, gene disease correlations and diagnostic tests (necessary within the practice of personalised medicine) may fall outside the scope of patentable subject matter, are anticipated by earlier disclosure or are obvious in light of the prior art, disrupts the intellectual property landscape of personalised medicine.

Previously, the development of personalised medicine might have required the often costly transactions associated with licensing gene-disease correlation patents by diagnostic and pharmaceutical companies, researchers and, perhaps, even the prescribing physician. Now, however, the emerging intellectual property regime may limit the licensor’s ability to exact substantial and restrictive control over such patents, effectively lowering costs and enabling more development in the field. However, as pointedly stated in the Bilskidissent, “…denial of patent protection … would undermine and discourage future research for diagnostic tools … [diverting] unprotectable investments away from discovery of ‘scientific relationships’ within the body that diagnose breast cancer or Lou Gehrig’s disease or Parkinson’s or whatever.”

Personalised medicine will not be fully realised without achieving the appropriate balance.