BioWorld International Correspondent

LONDON - A new cancer drug that specifically targets enzymes involved in cell division is about to enter clinical trials. Initial tests on the drug in animal models suggest that it can shrink certain difficult-to-treat human tumors.

The drug, called VX-680, is being developed for the clinic by Vertex Pharmaceuticals Inc., of Cambridge, Mass. VX-680 inhibits enzymes called Aurora kinases, which are essential for the function of the mitotic spindle, ensuring that cells divide correctly and that daughter cells receive the correct numbers of chromosomes.

Karen Miller, director of biology at Vertex Pharmaceuticals (Europe) Ltd., of Abingdon, UK, told BioWorld International: "We believe that this is an exciting new approach for cancer therapy. VX-680 is one of a new generation of targeted therapies aimed at specific cellular-signaling molecules."

The FDA has given Vertex the go-ahead to pursue Phase I trials of VX-680, which are expected to begin later this year.

An account of the group's studies on VX-680 appears in Nature Medicine (advance online publication Feb. 22, 2004) in a paper titled "VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumor growth in vivo."

The three mammalian Aurora kinases, called A, B and C, were discovered in 1997. Research into their function and activity over the past few years has suggested that they might play multiple roles in the development and progression of cancer by acting as regulators of cell proliferation, by transforming normal cells into cancer cells, and by down-regulating p53, one of the body's natural tumor suppressors.

Liz Harrington, staff investigator at Vertex in the UK and first author of the paper, said it soon became clear that the Aurora kinases were expressed and activated in abnormal ways in many different types of tumors in humans, including leukemias, colon and breast tumors.

Taking those observations into account, the Vertex team hypothesized that inhibiting the Aurora kinases would inhibit cell proliferation. By 2002, they had the crystal structure of Aurora kinase A and set out to design small molecules that would target the enzymes' ATP-binding sites.

Miller said, "Because we knew the structure of these enzymes and the structures of a great number of other kinases, we were able to make the inhibitor very selective for the Aurora kinases over other kinases."

Initial tests on VX-680 showed that it could inhibit all three Aurora kinases.

"We were excited to find that not only could VX-680 halt cell division, but that it could also kill these cells by inducing apoptosis or programmed cell death," Miller said. "Non-dividing cells remained unaffected."

Harrington added: "In vitro tests on multiple human tumor types showed that VX-680 could block proliferation of these cells, including those from breast, prostate, pancreatic, leukemic and colorectal tumors. When we looked at cell death, we found that cells from colorectal, lymphoma and leukemic tumors were particularly susceptible to the cytotoxic effect of VX-680."

VX-680 selectively inhibited Aurora kinase A more than 100 times more strongly than 55 other kinases tested. It also inhibited a kinase called FLT-3. That kinase has a role in hematopoiesis, and a mutant activated form often is found in patients with acute myelogenous leukemia. The authors wrote, "These mutations correlate with poor prognosis, thus making FLT-3 an attractive drug target in its own right." Experiments showed that VX-680 was able to completely inhibit growth of cancer cells taken from patients with acute myelogenous leukemia.

Miller, Harrington and their colleagues next wanted to find out what effect VX-680 had on the growth of tumors in animal models of human cancers. Miller said: "We were very pleased to see that not only did it profoundly inhibit the growth of tumors in vivo, but it was also able to induce regression."

In a mouse model of human leukemia, VX-680 caused 98 percent inhibition of tumor growth. In a mouse model of human colon cancer, the size of the tumor was reduced by 56 percent. Similarly, there was 22 percent shrinkage in average tumor size in a mouse model of pancreatic cancer.

In general, VX-680 was well tolerated by the animals that received it. Miller said: "This agent is likely to inhibit the proliferation of any cycling cell, so the side effects may be similar to those induced by cytotoxic therapies that are already being used in the clinic, such as neutropenia. Clinicians are well used to dealing with these kinds of effects."