A study being published Thursday in the New England Journal of Medicine says Genentech's breast cancer drug, Herceptin, improved the life expectancy of patients with late-stage tumors but raised safety concerns that could affect whether the drug is used earlier, when cancers are first diagnosed.

This safety issue, being debated by cancer specialists, has financial implications for Genentech, which grossed $275.9 million last year on Herceptin, its second-best-selling drug.

Herceptin has been on the market since 1998, when the U.S. Food and Drug Administration approved it to treat breast cancers that have recurred in patients who have undergone surgery, radiation therapy and/or chemotherapy.

If Herceptin is eventually shown to be safe and effective against early stage breast cancers, it would expand Genentech's market, while the reverse would limit use of the remedy.

The data reported in the New England Journal of Medicine come from the same clinical trials that led to FDA approval in 1998.

Thursday's publication, however, marks the first time a scientific journal has taken a detailed look at those clinical trials.

The most notable finding of the study was Herceptin's effect on survival. The 235 patients who got Herceptin plus chemotherapy in the clinical trial lived 25.1 months, compared with 20.3 months for the 234 patients who got chemotherapy alone.

``Drugs that prolong survival are rare, rare, rare in cancer treatment,'' said Gwen Fyfe, senior director of oncology at Genentech.

Herceptin was designed to be effective against tumors with an excess of the HER2 gene. Thus, when patients have a breast cancer relapse, only those who test positive for HER2 are treated with Herceptin.

But the controversial details in the paper revolve around the side effects Herceptin showed when given with the common chemotherapy agent doxyrubicin.

Lead author Dr. Dennis J. Slamon, a cancer researcher at the University of California at Los Angeles, explained that administering Herceptin and doxyrubicin together seemed to affect the hearts of test patients, causing fatigue, shortness of breath and other symptoms.

Though such side effects might be tolerable when treating advanced and aggressive breast cancers, Slamon said oncologists, for whom doxyrubicin is a favorite chemotherapy agent, would be reluctant to use it in combination with Herceptin for early stage cancers.

Slamon said evidence in the New England Journal of Medicine, coupled with other Herceptin studies whose results have not yet been published, suggest these side effects do not occur when Herceptin is given with two other chemotherapy agents, taxotere and carboplatin.

However, Slamon said leading oncologists, the specialists who administer chemotherapy drugs, have resisted suggestions that it would be worthwhile to switch agents in order to use Herceptin safely right off the bat, when breast cancer is first diagnosed.

``The uphill battle we're struggling with is that some people don't want to accept these changes,'' Slamon said.

Dr. Larry Norton, a specialist at New York's Memorial Sloan-Kettering Medical Center, is against using Herceptin to treat early stage cancer at this time, especially if it means pushing doxyrubicin aside.

``We don't know yet if it's effective; we don't know yet if it's safe,'' he said, adding that early administration of Herceptin might breed tumors resistant to the drug, rendering it useless when current treatments fail and Herceptin is needed for late-stage cancers.

Slamon said studies are under way to test Herceptin on newly diagnosed cancers.

One test will involve 3,000 patients in a three-way comparison. One thousand will get doxyrubicin. An equal number will get doxyrubicin and Herceptin. The third group will get Herceptin and the other chemotherapy drugs, taxotere and carboplatin.

But it could be 2004 before those studies settle the debate over the safety and effectiveness of Herceptin as a front-line treatment for HER2 breast cancers, Slamon said.

-----

• More news on Breast Cancer

Related MEDLINEplus Pages:

• Breast Cancer