Regular mammograms to check for breast cancer, widely recommended for women over 50, can also prevent deaths of women in their 40s, according to results of a American Cancer Society study announced yesterday.

In both age groups, women whose breast cancer was found while they were enrolled in a national study that included annual mammograms had survival rates substantially higher than expected from national statistics, suggesting that more cancers were detected at an early, curable stage.

Cancer experts predicted yesterday that the findings will likely lead to national recommendations that women 40 or older have a mammogram annually. A mammogram is a special type of X-ray used to view the breast tissue.

Breast cancer strikes about one out of 10 American women. The society estimates that 41,000 women will die of the disease this year, placing it second only to lung cancer as a cause of cancer deaths among women.

The new report is an analysis of breast cancer survival data among women who participated in the Breast Cancer Detection Demonstration Project (BCDDP), an $80 million program sponsored in the 1970s by the society and the National Cancer Institute (NCI) to study the benefits of regular mammograms and breast examinations.

The project, which involved 280,000 women, provided annual screening examinations, including a physical examination and mammography at each visit, for five years.

The examinations revealed breast cancers in 4,257 women, who were then referred to their own doctors.

The new report, published in the September/October issue of Ca, the cancer society's journal, compares survival data on these women with detailed national breast cancer survival statistics compiled by NCI.

Among study participants with invasive breast cancer (tumors that had spread outside the duct of the milk-producing gland where they originated), 81 percent were alive eight years after diagnosis.

The national survival rate for patients with invasive breast cancers was 65 percent.

This corresponds to a cancer death rate for women in the study that was 46 percent lower than the national figure.

Although an earlier analysis of the BCDDP data showed that regular mammograms improved breast cancer survival among women over 50, the new report for the first time shows similar benefits for women aged 40 to 49.

"What it means to me is that women {in the study} were having their cancers detected earlier," when there was a better chance of cure, said Dr. Charles R. Smart, chief of the NCI's Early Detection Branch.

He said that if doctors across the country would recommend regular mammograms for their female patients over 40, the national death rate from breast cancer might be cut by 46 percent, as was seen in the study.

Sam Shapiro, a professor at Johns Hopkins University School of Hygiene and Public Health, said the fact that the BCDDP program had no control group -- a comparison group who did not receive mammograms -- somewhat weakened its power to draw conclusions. But he said that, nevertheless, experts in the field are "leaning toward . . . screening {women} ages 40 to 49" as well as those over 50.

A mammogram is an X-ray done with special techniques to show the details of breast tissue.

When the test was first being developed, it delivered significant radiation doses, raising concerns that if a woman was tested regularly over her lifetime, the cumulative radiation exposure might actually increase her risk of breast cancer.

But in recent years, technological advances have greatly reduced the radiation doses involved.

At the same time, mammograms can detect many breast cancers when they are too small to feel during a physical examination.

About 40 percent of breast cancers found in the BCDDP study were detectable only by mammography, Dr. William F. Feller of Georgetown University Medical School said yesterday.

The American Cancer Society currently recommends a single, baseline mammogram for every woman between age 35 and 40, regular mammograms every one to two years between 40 and 49, and annual mammograms after 50.

NCI recommends that the test "may be considered" for women 50 and over, and for women over 40 with a family history of breast cancer.

Smart said a board of scientific counselors to the NCI will consider on Monday whether to recommend that women between 40 and 49 have the test every one to two years.