BioWorld International Correspondent

Agendia BV, a recently formed Dutch start-up, is planning three large-scale clinical trials of a genomics-based method for predicting the chance of primary breast cancers to recur and metastasize.

If successful, the approach could alter the management of breast cancer by eliminating the need for adjuvant chemotherapy in many women.

In the U.S., more than 95 percent of women with breast cancer who are younger than 55 and who have no detectable tumor cells in local lymph nodes receive chemotherapy to lower the risk of recurrence. The equivalent statistic in Europe is greater than 85 percent. In either case, just 30 percent of patients actually need to undergo such therapy, Agendia's chief operating officer and head of clinical research, Laura van't Veer, told BioWorld International. "Yet we are giving 95 percent or 85 percent [of patients] chemotherapy to reduce the risk of recurrence," she said.

Up to now, it had not been possible to distinguish at the primary tumor stage who will go on to have a poor prognosis from those who will experience a less-severe form of disease. Current clinical and histologic criteria do not provide sufficient information.

Agendia's founding team, based at the Netherlands Cancer Institute (NKI) in Amsterdam, collaborated with bioinformatics firm Rosetta Inpharmatics, now a wholly owned subsidiary of Merck & Co. Inc., of Whitehouse Station, N.J., on identifying 70 genes that have collective expression patterns that provide a predictive indicator of how breast cancer will progress. In two retrospective studies of archived tissue samples from breast cancer patients who underwent long-term follow-up, they defined genetic signatures that correlated with a "good prognosis" and a "bad prognosis."

The first study, reported in the Jan. 31, 2002, issue of Nature, was based on an assessment of genetic expression patterns in 117 outliers, van't Veer said. A follow-up study, which appeared in the Dec. 19, 2002, issue of New England Journal of Medicine, applied the same analysis to 295 randomly chosen subjects and confirmed that the findings appeared to have general application.

The forthcoming prospective trials will involve about 13,000 patients in all. The beginning of the first two is imminent.

"Both of them will start with a pilot phase in October," van't Veer said. The Health Council of the Netherlands, based in the Hague, is funding one study involving 3,000 subjects, while the New York-based Avon Foundation is supporting a second, which will recruit 5,000 patients.

A third study, which also will involve 5,000 subjects, will get under way next year, van't Veer said. The study will receive support from the Brussels-based European Organization for the Research and Treatment of Cancer and from the Sixth Framework Program of the European Commission.

Agendia, which was formed in Amsterdam last month, is planning to introduce its first commercial diagnostic services later this year, van't Veer said. It is collaborating closely with Agilent Technologies Inc., of Palo Alto, Calif., in the development of its microarray-based platform.

"Besides these 70 genes, there are over 1,000 control genes present," van't Veer said. Rosetta Inpharmatics remains a partner as well, and the companies will have joint ownership.

The company initially will perform testing services in-house, but aims eventually to outlicense its technology to third parties. It also plans to establish analogous genetic signatures for melanoma, colon cancer, lymphoma and ovarian cancer. It has access to an NKI biobank containing tissue samples from about 20,000 cancer patients whose clinical outcomes have been documented.

In addition to van't Veer, Agendia's founding team includes CEO Bernhard Sixt and Chief Scientific Officer Rene Bernards. The company has raised debt financing from a number of sources and is looking to secure its first investment round before the year's end, van't Veer said.