BOSTON (The Boston Globe) -- In a new assault on the practice of long-term hormone use, a National Cancer Institute study published Wednesday found that women who took estrogen were significantly more likely to develop ovarian cancer than those not on the hormone.

The finding marks another setback for hormone replacement therapy, with pills once hailed for their youth-preserving quality now implicated as a cause of potentially fatal illness.

The new study tracked thousands of women for nearly two decades, finding that women who took estrogen were, on average, one-and-a-half times more likely to develop ovarian cancer, a particularly lethal form of cancer. That risk soared the longer a woman took the medicine: Women on estrogen for 20 years or beyond were three times more likely to develop ovarian cancer than those who did not take the pills.

The report in Wednesday's Journal of the American Medical Association comes one week after federal authorities halted another study of hormone replacement therapy after research showed that the pills were doing more harm than good, even causing conditions the medicine was once believed to prevent, including heart disease. That study looked at women who took a combination of two hormones, estrogen and progestin.

"This is just further fuel for the fire," said Dr. JoAnn Manson, chief of preventive medicine at Brigham and Women's Hospital. "This study provides further concerns about long-term use of hormone replacement therapy."

Those drugs have been prominent in the medicine cabinets of millions of American women since the 1940s, originally designed to ease the short-term symptoms of menopause, such as hot flashes and night sweats.

Over time, hormones also emerged as a treatment of choice to help women avoid heart disease and osteoporosis, a change that meant women went from using the drugs for a few months to taking them, in some cases, for decades. Today, an estimated 8 million women in the United States regularly take estrogen. Previous studies had presented conflicting evidence about the link between estrogen and ovarian cancer.

Now, with two studies in one week undermining long-held beliefs about hormones, doctors are reevaluating their recommendations to patients who have flooded clinics with urgent phone calls.

"This may very well be sounding the death knell for long-term hormone use," said Dr. Luis E. Mendez, a women's cancer specialist at the University of Miami medical school. "Definitely, you're getting a double-whammy here. It's hard to put a lot of faith in the statements that it's OK for patients to take long-term hormone replacement therapy."

The study tracked 44,241 women for 19 years, asking them about their use of hormones and other medical questions. The increased risk of ovarian cancer was slight for women who used hormones for less than four years, climbing steeply with the passage of time, with women on estrogen for 10 to 19 years having an 80 percent higher chance of developing the disease.

Estrogen alone is the hormone of choice for women who have undergone hysterectomies, while other women on hormone-replacement therapy typically take a hormone drug called Prempro, which includes both estrogen and progestin.

The National Cancer Institute researchers examined whether women taking the estrogen-progestin pill had a higher risk of ovarian cancer and could find no evidence of that. But they were unwilling to state conclusively that there's no link between Prempro and ovarian cancer because they had insufficient data to reach such a conclusion, the federal scientists said.

The number of women on long-term doses of estrogen is extremely small, especially in the Northeast where hysterectomies are less common than in other parts of the nation, said Dr. Maureen Connelly, co-director of the Menopause Consultation Service at Harvard Vanguard Medical Associates.

Ovarian cancer is relatively uncommon among American women, with the average woman having a 1.7 percent chance of developing the disease during her lifetime, compared with a 13.3 percent chance for breast cancer.

"Our study reiterates the message we've been hearing over the years and certainly over the past week - that women should continue to talk to their health-care providers about whether hormones might be right for them," said James V. Lacey Jr., the National Cancer Institute epidemiologist who is lead author of the study. "Don't panic."

Specialists theorize that because ovarian tissue is especially sensitive to hormones, altering them post-menopause may increase cancer risk. An animal study cited in the National Cancer Institute report found that estrogen stimulated growth of cancer cells in rabbits.

Conversely, other hormonal changes - such as taking birth control pills - are known to provide protection against cancer.

While the estrogen study monitored a substantial number of patients for an extended period, it probably will not provide the definitive answer about use of the hormone. That will come from the ongoing Women's Health Initiative, the same study that concluded last week that the combination therapy Prempro could be perilous to patients' health.

The National Cancer Institute review is known as an "observational study," which means scientists simply tracked the health of thousands of women who took hormones as part of their routine health care. The Women's Health Initiative review is following a more rigorous research model, with specific groups of patients being assigned to take either estrogen or a dummy pill. Neither patients nor researchers know who gets which pill.

Dr. Wulf H. Utian, executive director of the North American Menopause Society, said he wants to see the results of that study before he makes a final decision on whether to recommend estrogen use. Still, he conceded, the accumulating evidence warrants that physicians adopt new thinking on hormones.

"If women are going to be on estrogen for greater than 10 years," Utian said, "we should factor that in as a worst-case scenario - if you stay on estrogen, it might increase your risk of ovarian cancer."

Until the Women's Health Initiative study of estrogen is complete, physicians and patients are engaging in a classic risk-benefit analysis of hormones - one weighted toward short-term use of hormones. "For the short-term benefit of relieving hot flashes and other symptoms of menopause," said Manson, of Brigham and Women's, "a small increase in risk may be a price women are willing to pay."

Copyright 2002 The Boston Globe. All rights reserved.