Speculating that breast cancer may originate in the uterus, researchers at the Harvard School of Public Health have found that breast cancer risk as an adult is linked to infant birth weight.

Women in the study who weighed less than 5.5 pounds at birth had about half the risk of breast cancer than women who weighed more than 8.8 pounds, according to the study, reported in last week's Lancet, a British medical journal.

"Our study shows a significant relationship between weight and future breast cancer risk," Karin Michels, an epidemiologist at the School of Public Health and the lead author of the study, said in a statement. "But this should not lead women to strive for lower birth weight while pregnant."

The authors suggested that the surge of maternal estrogen to the infant, which is correlated with higher birth weights, may be a factor that increases risk of breast cancer for the female fetus.

Heavier babies are exposed to more estrogen, and the rapid cell division spurred by the exposure may increase the probability of damage to the genetic material of the fetus, researchers suggested.

They also said increased amounts of estrogen may cause breast tissue to be more at weights sensitive to cancer-causing agents later in life.

Cells of mammary glands in the uterus may be especially vulnerable to the effects of high estrogen levels or other pregnancy-related influences.

Those weighing between 7.7 and 8.8 pounds at birth were 14 percent less likely to get breast cancer than those weighing more than 8.8 pounds. The risk was about 32 percent lower at weights between 6.6 and 7.7 pounds and 34 percent lower between 6.6 and 5.5 pounds. Risks dropped 45 percent for weights under 5.5 pounds.

The study, which examined factors in early life that may lead to breast cancer, found that weight in relation to gestational age may be a critical factor. However, prematurity alone was not a predictor of breast cancer.

Other factors that are being studied include the mother's smoking, alcohol-drinking habits, diet during pregnancy and the child's diet during early life.

Michels worked with colleagues at the School of Public Health, the Medical School and Brigham and Women's Hospital.

The study is based on the Harvard Nurses Health Studies, case-control studies of 2,151 women.