Originally created 07/29/01

Patenting gene sequences and the proteins they produce



Night and day, the computers at Incyte Genomics churn out gene-sequence data. Almost as quickly, Incyte's computers crank out gene-sequence patent applications.

Once scientists at the Palo Alto, Calif., biotech firm identify a sequence that may help develop profitable new drugs, the data is digitally funneled to Incyte's patent division. There the company's legal staff chooses from among 50 patent application templates that classify new genes and proteins by their function.

"To write these applications from scratch would cost $15,000 to $20,000 and take four weeks," says Diana Hamlet-Cox, manager of Incyte's patent department. "We can file hundreds of sequences" in a week. Indeed, over the past nine years, Incyte has patented 580 gene sequences - more, she claims, than any other company.

The sprint to decode the human genome may be winding down, but another race is heating up: the mad dash by Incyte, Celera Genomics and other biotech companies to patent the sequence itself.

With pharmaceutical firms beginning to design personalized drugs based on different genetic profiles, biotech businesses see a gold mine in licensing to drugmakers patented gene sequences and the proteins they produce. Some biotech companies also want to prevent competitors from commercializing genetic information because they plan to develop their own drugs.

In just the past few years, gene-sequence patents have become the biotech industry's lifeblood. But a serious clot is developing in the patent approval process.

Tens of thousands of biotech patent applications are inundating the U.S. Patent and Trademark Office, lengthening the lag time between application and patent to two years or more.

Such delays threaten to cost biotech companies millions of dollars in lost licensing and royalty fees. And while the government makes patent decisions that will guide medical research for decades, scientific understanding of the genome's complexity changes almost daily. Earlier this year, for instance, scientists radically revised the number of estimated human genes from about 120,000 to fewer than 40,000.

The patent office this year tightened its regulations to require that patent applicants more precisely identify the function of genes and proteins. This shifting landscape presages an onslaught of litigation over ownership of gene sequences and proteins. Adding to the uncertainty: an upcoming U.S. Supreme Court case that could change the rules of the patent process.

The Supreme Court set the stage for the biotech patent explosion two decades ago. In a watershed 1980 case, Diamond vs. Chakrabarty, the court ruled that a patent could be granted for a genetically engineered bacterium. Until then, living matter was generally considered unpatentable.

But the Supreme Court could not have foreseen how the genomic revolution would collide with the patent office. The office, spread across 18 buildings in the suburbs of Washington, received approximately 20,000 biotech patent applications in the fiscal year ending last September. Thousands of applications currently await review.

While the office's overall approval rate is 70 percent, it OKs only 40 percent of biotech-related patent applications. Celera Genomics in Rockville, Md., which sequenced the human genome, is seeking patents on 150 to 300 genes valuable for drug development. Robert Millman, the company's director of intellectual property, acknowledges that the patent office faces intense pressure to limit the scope of gene patents. "The world is worried that patent monopolies will be granted on genes with little proof," he notes.

Drugmakers face the costly prospect of paying biotech companies to license specific gene sequences and proteins needed to develop genetically targeted medicines. To gum up the patent works, 10 drug giants formed the SNP Consortium in 1999 to search for minute genetic differences called single nucleotide polymorphisms. By freely publicizing SNPs - which can be used as drug targets - Big Pharma hopes to make it more difficult to patent those sequences, according to SNP Consortium Chairman Arthur Holden.

This fall, the Supreme Court is slated to hear a case that could shape future litigation over biotech patents. If upheld, Festo Corp. vs. Shoketsu Kinzoku Kogyo Kabushiki Co. could make it harder for biotech companies to bring patent-infringement cases against competitors that artificially alter a patented gene sequence, replacing one bit of DNA with another, without changing the proteins that gene produces.

However, the case won't resolve some fundamental disputes over gene patents, such as whether patenting a gene sequence amounts to a patent on any proteins produced by that sequence. "The law is running so far behind," says John Barton, a professor at Stanford Law School.

In late 1997, Myriad Genetics and Oncormed became embroiled in a dispute over a gene patent that is a portent of litigation to come. Both firms had received patents on the same gene, called BRCA1. When mutated, BRCA1 predisposes women to breast and ovarian cancer. Salt Lake City-based Myriad discovered the gene and received patent rights to its various mutations; Oncormed, now owned by Gene Logic of Gaithersburg, Md., garnered a patent on a variation of the normal version of the gene. Both companies hoped to make money from a BRCA1 gene test and sued each other for patent infringement.

The case eventually settled, with Myriad retaining exclusive rights to the test. But if different mutations in the gene are subsequently discovered by competitors, Myriad could potentially face more litigation.