Nearly three months after raising $135 million to move its lead blood-clotting agent into a pivotal study, ZymoGenetics Inc. started the Phase III trial of recombinant human Thrombin (rhThrombin) in surgery patients.

The randomized, double-blinded study will enroll at least 400 patients at 35 sites, evaluating rhThrombin in four types of surgery: spinal, liver resection, peripheral artery bypass and arteriovenous graft construction.

Seattle-based ZymoGenetics expects to file a biologics license application by the end of 2006 and potentially launch the product in late 2007.

"This is, of course, a very important milestone in the development of ZymoGenetics," said Bruce Carter, the company's president and CEO, in a conference call. "We do believe that in a little more than a year we should be in a position to file for our first product license."

While the company has six marketed products, rhThrombin represents its first candidate brought through clinical development without a partner.

The Phase III trial will compare the efficacy of rhThrombin to bovine-derived thrombin, the current standard of care, and is designed to support broad product labeling of rhThrombin to control blood loss during surgery. The primary objective is to compare the two products by the incidence of hemostasis within 10 minutes. Secondary endpoints will look at safety, as well as testing for the formation of antibodies against each of the two products.

ZymoGenetics expects rhThrombin to perform better on safety and immunogenicity tests, but not necessarily on efficacy. "We feel as though the bovine thrombin will perform comparably to rhThrombin, at least as it relates to the efficacy endpoint," said Douglas Williams, the company's executive vice president and chief scientific officer.

All of the available data seem to support that assumption, he said. The company's Phase II trial for rhThrombin established safety as a primary endpoint and showed the drug also met secondary endpoints that evaluated immunogenicity and developed point estimates of time to hemostasis. The drug was expected to move into a Phase III trial under a special protocol assessment, but the company decided to discontinue that process and begin the trial immediately based on its conversations with the FDA. Waiting for an SPA would have "pushed the start of enrollment into 2006," Williams said.

In the trial, researchers expect to use validated assays to look at antibody reactions to each product. The marketed version of thrombin is derived from bovine plasma, which brings with it impurities that cause the formation of antibodies that cross-react with human blood proteins, leading to serious bleeding complications.

This is an area in which rhThrombin might show an advantage, and become more attractive for prescribing physicians. "We don't anticipate we'll have a black box warning," Williams said. "I think that reflects the most significant difference."

Recombinant products do not contain animal blood products and can be manufactured at high levels of purity, suggesting that rhThrombin could replace the currently marketed product. As history has it, "the replacement of plasma-derived proteins with a recombinant version has been highly successful," Carter said.

Some examples of that success include Novoseven (Novo Nordisk A/S), Benefix (Genetics Institute) and Kogenate (Bayer AG), all of which have replaced animal-plasma human blood products. South San Francisco-based Genentech Inc.'s recombinant human insulin (Humulin) also replaced beef or pork products used to treat diabetics.

"We believe that none of the risks have increased since we embarked on this project, but the commercial attractiveness is much greater now than when we started," Carter said.

Sales of bovine-derived thrombin by King Pharmaceuticals Inc., of Bristol, Tenn., are expected to be about $240 million in the U.S. this year. The sales were $175 million in 2004 and $142 million in 2003.

The company has three other internal products, TACI-Ig, interleukin-21 and interleukin-29. The first is in Phase Ib trials in systemic lupus erythematosus, rheumatoid arthritis, multiple myeloma and non-Hodgkin's lymphoma. IL-21 is in Phase I trials for cancer, and IL-29 is expected to enter the clinic in 2006 for hepatitis C.

ZymoGenetics' marketed products include Novolin and NovoRapid for diabetes, NovoSeven for hemophilia, Regranex for wound healing, GlucaGen for hypoglycemia and gastrointestinal motility inhibition, and Cleactor for myocardial infarction.

The company's stock (NASDAQ:ZGEN) went down 20 cents Thursday to close at $16.27.