"There are different categories of benign biopsies and they convey different risks," said Mayo Clinic oncologist Lynn Hartmann, who led the study published in Thursday's New England Journal of Medicine. "What we're trying to do is allay concern in women who really don't have an increased risk, and solidify the extent of the increased risk in women who do."

Certain Cell Growth Patterns Are Ominous

Every year, about a million women who have a suspicious lump or mammogram undergo a biopsy -- a tissue sample is removed for microscopic analysis -- that is judged benign.

Doctors have long known that certain cell growth patterns are ominous, even though technically not malignant. However, there were questions about the degree of risk conferred by certain patterns, and whether a family history of breast cancer intensified the risk.

The new study, the largest and most rigorous to date, identified more than 9,000 women who had benign biopsies at the Mayo Clinic between 1967 and 1991. The biopsied tissue was reanalyzed to classify it based on current expert guidelines.

The women were followed for an average of 15 years, during which they developed 707 breast cancers. Their rates of cancer were compared with those from a state cancer registry.

About two-thirds of the women had benign growths made up of "nonproliferative" cells that were not rapidly dividing. These women had a tiny increase in risk, not enough to warrant extra measures to try to reduce it.

Thirty percent of the women had "proliferative" growths with rapidly multiplying cells; they had an 88 percent greater chance of later developing cancer.

The smallest subgroup -- about 4 percent of the women -- had the most worrisome growths, made up of abnormal-shaped, fast-dividing cells. These women had a 324 percent greater risk of developing cancer.

Contrary to previous studies, a family history of cancer did not have a significant impact on risk, even for the women with abnormal cells, called "atypia." However, for women diagnosed with atypia before age 45, the risk quadrupled.

"Just like everything else in breast cancer, it's not simple," Hartmann said.

Discussion of Risk and Options

Indeed, the terminology keeps changing as doctors better understand breast disorders. Women who had fibrous and cystic lumps used to be told they had "fibrocystic disease." Now, these changes, which occur in 60 percent of women, are simply called "fibrocystic changes" because they do not increase breast cancer risk.

In contrast, a benign biopsy that reveals abnormal growth warrants a discussion of risk and options. Putting that risk in perspective is important, Joanne Elmore, a physician at University of Washington in Seattle, wrote in editorial in the journal.

Based on the study, 6 out of 100 women diagnosed with nonproliferative breast changes would be expected to develop breast cancer within 15 years -- not substantially different than 5 out of 100 average women. But the rate jumps to 10 out of 100 for women with fast-dividing breast cells, and to 19 out of 100 for women with atypia.

Anne Rosenberg, a breast surgeon at Thomas Jefferson University, said, "It's the surgeon's responsibility to at least inform them of their risk. At that point, some surgeons feel comfortable doing the risk assessment. But if not, there are counselors trained to do it at virtually every hospital."

Besides routine screening, women may be eligible for ultrasound, MRI (magnetic resonance imaging) or other supplemental screening procedures.

Tamoxifen, a hormonal drug, has been shown to reduce risk by more than half. Newer estrogen-modifying drugs, notably raloxifene and anastrozole, are now being studied as preventive therapy.

Preventive mastectomy may be considered by very high-risk women, such as those with certain genetic mutations.