Introduction

Women with early-stage breast cancer that's been shown to be 'hormone-dependent' are usually given the drug tamoxifen for 5 years after their surgery. Tamoxifen prolongs survival when given for 5 years; after that time, further treatment hasn't any measurable benefits. Although the protective effect of 5 years of tamoxifen continues after the actual treatment period, women are still at risk of for a recurrence of their cancer in the same or the opposite breast, or in a more distant location (a metastasis). About 2-3% of these patients each year will have a recurrence.

Letrozole

Letrozole (Femara®) is a drug that inhibits an enzyme, aromatase, and in this way blocks the production of estrogen; levels of estrogens in post-menopausal women are reduced by 95%. The drug is therefore used in such women with advanced breast cancer. It seemed logical to see whether lowering estrogen levels in women after their 5-year tamoxifen course would reduce or abolish recurrences of hormone-dependent tumors.

The study

Researchers have now reported results from this type of study in the New England Journal of Medicine. ¹ The international trial, which was led by the National Cancer Institute of Canada, included over 5,000 post-tamoxifen women. They were given either 2.5 mg letrozole daily, or a placebo (dummy tablet). The aim was to measure the disease-free survival rate after 5 years.

'Early-stopping' rule

The study contained an important 'early-stopping' rule. The consent form for women taking part stated that "If new side-effects or information about my disease or treatment are discovered during the study, I will be told". (Almost all consent forms for clinical studies contain this sort of guarantee.) To ensure that there were no ill effects of letrozole, an interim analysis was scheduled when the participants in the study had been on letrozole (or placebo) for an average of 2.4 years. An independent committee would decide if the study should be stopped early.

This interim analysis revealed there had been 207 local (same or opposite breast) or metastatic recurrences - 75 in the letrozole group and 132 in the placebo group. The rates of recurrence were reduced by letrozole for both local and distant sites.

These numbers show that letrozole almost halved the rate of recurrence during the 2.4 years. The researchers stopped the study, promptly informed the participants of the findings, and gave those who had been on placebo the chance to switch to letrozole, if they wished.

Side effects?

The women taking letrozole were more likely to experience low-grade hot flashes, arthritis, pain in the joints, and muscle pains, than those on placebo. Osteoporosis was newly-diagnosed in 5.8% of the women on letrozole compared with 4.5% of the placebo patients. There was a trend towards an increased likelihood of cardiovascular events (heart attack, angina, stroke) with letrozole, but this wasn't significant; if the study had run longer, this might have been a significant problem.

What does this mean?

Obviously, it's very good news to have a drug available that will halve the risk of late recurrence of breast cancer, even if the actual risk is not very high. Unfortunately, however, the early-stopping rule, though of clear benefit to the participants, means that we don't know as much as we'd like to. For instance, how long should letrozole be given? More than 2-3 years? Will cardiovascular effects, or the frequency of osteoporosis, limit letrozole's usefulness?

Another problem: Can other aromatase inhibitor drugs be tested in placebo-controlled studies, or would that be unethical? (Some are already running - should they be stopped?) And it will be important to find out if letrozole could be used effectively immediately after the initial treatment of hormone-dependent tumors (i.e. after surgery and radiation, or chemotherapy).

This important study demonstrates that progress in cancer therapy is made in steps - some small, some large - that lead to further studies to test new concepts, and, hopefully, the creation of new steps of progress.