Hormone Replacement Therapy and Breast Cancer

Janet M., a fifties-something woman, entered my office and said as she sat down, "I've read that if I take hormones I'll increase my breast cancer risk. I'm going crazy without sleep and with these mood swings, but I don't want to increase my breast cancer risk by taking hormones."

Like many women, Janet had heard that a recent study, the Women's Health Initiative (WHI), definitively showed that hormone replacement therapy (HRT) increases breast cancer risk. Janet, like most people, didn't realize that this study found no statistically significant increase in breast cancer risk to women who took HRT.

When differences are not significant, an increase in risk may well be due to other factors, not the one being studied, such as HRT use. As often happens when a medical story is reported, the emphasis was on the increase in risk, not whether the increase was likely to be due to the agent being studied or to the size of the risk.

The actual size of a risk is important in any woman's decision making process. In this case the risk was exceedingly small -- only 8 in 10,000 women a year -- which is 0.08% or eight hundredths of one percent! Janet was amazed to learn the actual size of the increase, and said, "You mean I was getting all concerned for a risk that small!"

"And," I pointed out, "even this very small difference in risk may not be due to hormone use." I explained that breast cancers take an average of eight years to reach about half an inch in size. This means that breast cancers started in the first year of the study would not be detected for eight or more years. The study followed women for only about five years, so all or most of the breast cancers found were probably present in an undetected state before the study began.

Janet asked if HRT use might have caused some breast cancers to grow more rapidly and therefore be detected sooner than eight years. This is unlikely. A number of studies find that breast cancers in women who were using HRT were not larger and were not dividing more rapidly than breast cancers in non hormone users. Since breast cancers grow more slowly in older women and the average age in this study was 63, breast cancers in this group would tend to grow more slowly and so take even longer than the eight year average to be detected.

Women in the WHI study used a particular type of hormone Prempro. The results of this study therefore do not apply to other, newer approaches in which more natural hormones are used and a woman's menstrual cycle is more closely approximated.

Janet was surprised to learn that many studies find that women who use HRT do not have an increase in breast cancer risk compared to women who don't use hormones, even when hormones are used for twenty years. Also, in another large study in which some women were assigned to take Prempro and others not, women who used Prempro had no significant increase in breast cancer risk.

As Janet left, she said, "I can see now that when I hear about a study I need to know how big a risk is and not just that it is increased. I'll also ask how long a study it was. This discussion has given me a whole different perspective."