A Redwood City biotech start-up has unveiled new cancer screening tests that are intended to pick the best drugs for treating an individual's tumor and predict whether a patient will be a long-term survivor.

The tests are part of a push toward the latest frontier in medical care: the quest for personalized treatments that match the specifics of the patient's tumor.

Early results, presented at a cancer specialists' meeting in Chicago that ends today, represent the debut for Genomic Health. The 3-year-old privately held company, which has already raised more than $50 million, is launching three large-scale tests to firm up the preliminary results. It hopes to begin offering tests to physicians within a year through its own clinical laboratory.

The same technology can be used by drug companies to determine which patients are most likely to benefit from experimental medications.

If all goes as planned, Genomic Health could emerge as a leader in the long-awaited era of personalized medicine. Its technology is designed to determine which genes are activated in a specific tumor and to make treatment decisions based on that information.

Other larger companies are in the race, including pharmaceutical giant Merck of Whitehouse Station, N.J., and Genomics Diagnostics in Alameda. But four studies released at the annual meeting of the American Society of Clinical Oncology help establish Genomic Health as a leader in this emerging field of medicine.

Small study

In one small study presented Monday, the company's technology was used to establish which lung-cancer patients are most likely to respond to a new-generation cancer drug called Iressa, which blocks the internal machinery of rapidly multiplying cancer cells.

The drug from AstraZeneca won FDA approval in May, even though it shrank tumors in only 10 percent of the cancer patients tested.

But knowing who is mostly likely to respond to the drug could double that percentage, said Dr. Ronald Natale, a cancer researcher and acting medical director at the Cedars-Sinai Comprehensive Cancer Center in Los Angeles.

``In the future, a piece of patient tumor will be used to determine the best and most appropriate treatment for them as an individual,'' Natale said.

Natale and colleagues, using Genomic Health technology, identified eight genes that are activated in tumor cells from patients who later responded to the drug.

In a separate study of patients with breast cancer that had spread to several lymph nodes, researchers from Genomic Health and Rush-Presbyterian-St. Luke's Medical Center in Chicago zeroed in on 16 genes that identify patients with the best chances for long-term, disease-free survival after treatment.

For several years now, researchers have been predicting a new era of personalized medicine in cancer based on knowledge of which genes are activated in a given tumor. But the research has proceeded slowly, partly because it depended on collecting frozen tumor samples and then following patients for several years.

Advantage

Scientists at Genomic Health believe that they have a leg up because their technology allows them to experiment with tumor specimens that have been routinely collected and stored following surgery at virtually every hospital in the country for 20 years or more.

These tumor samples are preserved by chemicals and embedded in a block of paraffin, the same wax used for sealing jars of homemade jams and vegetables. And they are saved for years as part of a patient's treatment record.

The Genomic Health technology makes it possible to test these older tumor samples for small bits of RNA, which are chemicals activated by the genes inside the cancer cells. The corresponding medical records tell how the patient was treated, whether there was a recurrence of the cancer and if the patient survived.

The idea, says Genomic Health Chief Executive Randy Scott, is to ``start with the patient and work backwards'' -- to see what the tumor looked like at the time of diagnosis.

In one of the breast cancer studies Genomic Health released this weekend, for example, many of the patients were first diagnosed more than 15 years ago.

``Biotechnology is where the computer industry was right before the PC,'' said Scott. ``The ultimate end is the development of genetic technology for the individual patient.''