Originally published in The Ottawa Citizen October 15, 2003
Original Title: A review of letrozole therapy for breast cancer
The announcement last week that Letrozole reduced breast cancer recurrence represents another step forward in the battle against breast cancer. When an announcement of a “groundbreaking” study, it is understandable to react with hope that a cherished life will be saved and suffering will end.
Methodical evaluation of new research and treatments is critical to provide the public and health care professionals accurate information. Indeed, it is especially important when dealing with common diseases that touch the lives of many families and friends. Television and radio news, because of time constraints, cannot cover all the details and critical analysis of each medical study.
It is necessary to digress for a moment to define two terms integral to providing perspective to study results. Most studies will report disease reduction in one or two ways: relative and absolute difference. It is these terms that can be a source of confusion and misunderstanding.
Suppose a disease incidence in a particular group was initially two per cent and after treatment was reduced to one per cent. One could conclude that the incidence was cut by 50 per cent ((2-1)/2). This is relative difference. Since the rate dropped from two to one per cent, the absolute difference is one per cent (2-1).
The well-designed letrozole produced clinically relevant information. Important limitations of the study need addressing to provide a balanced view of the results. Two editorials in the November 6, 2003 issue of the New England Journal of Medicine (NEJM) reviewed the data with a critical and supportive eye.
In any study, the sample size (study population) must be great enough to provide statistically significant results. The letrozole study met this criterion by enrolling 5187 women eclipsing the minimum statistical requirement of 4800.
The study evaluated post menopausal women over the age of 50 years with estrogen receptor sensitive breast cancer. Each had recently completed five continuous years of Tamoxifen therapy. The investigators wanted to know if Letrozole, introduced following Tamoxifen therapy, would reduce breast cancer recurrence compared to women who did not take letrozole (the placebo group).
The length of the study was to last five years. It was halted at 2.4 years because the reduction in breast cancer recurrence exceeded their expectations. It was decided that it would be unethical to withhold this treatment from the placebo group.
Of the 5157 women, 75 out of 2575 (2.9 per cent) in the Letrozole group and 132 of the 2582 women (5.1 per cent) in the placebo group had a recurrence of their breast cancer. In other words, 57 fewer women compared to placebo developed breast cancer. This result was reported as a 43 per cent relative reduction ((132-75)/132).
The absolute reduction of breast cancer is somewhat different. Indeed, Dr. Harold J. Burstein in a NEJM editorial states, “To date, letrozole has reduced the aggregate number of recurrences of breast cancers from 5.1 per cent to 2.9 percent-a reduction of approximately 1 event per 100 women treated per year.” The one in 100 statistic is calculated by dividing the absolute reduction of 2.2 per cent (2.2 women per 100 women) by 2.4 years.
The study presents other limitations. The investigators admit the lack of a true five-year duration makes it difficult to project the long-term consequences of letrozole treatment. At best the study’s conclusions reflect only two to three years of Letrozole treatment.
It is difficult to extrapolate the information past three years. Indeed, the stated objective of the trial was “to determine the disease-free survival and overall survival for women who has previously received = 5 years of adjuvant (tamoxifen) randomized to receive either letrozole 2.5 mg daily or placebo for five years.”
Overall survival was not significantly different after two years between the placebo and letrozole group. Extrapolation of the data into the third and fourth years indicates five to six per cent more women in the letrozole group would have survived.
John Bryant, Ph. D. and Dr. Norman Wolmark in their NEJM editorial comment that these findings cannot “be used to support a recommendation of five years of letrozole treatment, since none of the patients have been followed that long.”
They further state that the long term safety of this medication with respect to heart and bone health and quality of life could not be determined. Letrozole blocks the production of estrogen. It is associated with the development of osteoporosis, hot flashes, night sweats, muscle ache, arthritis, osteoporotic fractures, potential cardiovascular disease and impaired sexual function.
It is therefore difficult to make clinically appropriate judgments about the risks versus the benefits of the treatment when the full extent of its side effects and efficacy remains unknown.
As with other treatments, time, experience and more clinical trials will help answer some of the unknowns. Letrozole is an option for specific patients but they must be apprised of the limitations of the data. It is imperative that post menopausal women with estrogen receptor sensitive breast cancer discuss her options with her oncologist.
© Dr. Barry Dworkin 2003