Aromatase Inhibitors in Adjuvant Therapy of Breast Cancer: Before, Instead of, or Beyond Tamoxifen

Abstract

The trial by Boccardo et al, which is published in this issue of the Journal of Clinical Oncology, represents the fourth published randomized trial of an adjuvant aromatase inhibitor used either instead of or after 2 to 3 or 5 years of tamoxifen. Before these data became available, 5 years of adjuvant tamoxifen had been considered standard based on the results of individual trials and the Early Breast Cancer Trialists’ Collaborative Group Overview. These trials showed that 5 years of tamoxifen therapy, compared with either 2 years of tamoxifen or none, was associated with greatly improved relapse-free and overall survival. Several trials of 5 years compared with indefinite tamoxifen or 5 years compared with 10 years of tamoxifen therapy subsequently suggested that longer therapy might be less beneficial, whereas one small trial suggested the opposite conclusion. The Oxford Overview Database, including data from the ongoing randomized ATOM (Adjuvant Tamoxifen or More) and ATLAS (Adjuvant Tamoxifen Longer Against Shorter; C. Davies, personal communication, April 2004) trials of 5 compared with 10 years of tamoxifen, has not yet shown any clear significant difference between 5 years of tamoxifen therapy and more prolonged tamoxifen (http://www.ctsu.ox.ac.uk/ ebctcg/). Based on these aggregate data, most clinicians around the world have adopted 5 years of tamoxifen therapy as the optimal duration. However, it is well known that women with a previous diagnosis of breast cancer have a long-term ongoing risk of breast cancer recurrence. Data from the long-term follow-up of a group of Eastern Cooperative Oncology Group trials and from the Oxford Overview have shown that there are as many recurrences during years 6 through 15 after primary surgery as between years 1 through 5 in both women treated with tamoxifen and controls. The estimated recurrence risk beyond year 5 is approximately 4% per year for women with node-positive disease at the time of primary surgery and approximately 2% per year for women with node-negative disease. Consideration of these data before the design of the current group of adjuvant endocrine trials strongly suggested that additional therapy beyond 5 years of tamoxifen could prove useful. Now, for women nearing completion of 5 years of tamoxifen, the decision to consider additional years of endocrine therapy with an aromatase inhibitor seems fairly straightforward. Issues for discussion should include the risk of recurrence over the next 5 years based on prior nodal status and other risk factors. Patients should be informed of the efficacy of letrozole in reducing recurrence risk and, in the subset of node-positive women, in increasing overall survival. This information, together with a review of the potential short-, intermediate-, and long-term side effects of this aromatase inhibitor, should allow clinicians and patients to make a relatively informed choice regarding the desirability of this treatment. Although the MA-17 trial of 5 years of letrozole compared with placebo after tamoxifen for 5 years has, to date, a median follow-up of only 2.5 years, more data will be forthcoming. Such data should be available before the women we are currently starting on letrozole after 5 years of tamoxifen reach that 2.5-year point. Thus, more mature information regarding efficacy and longer term side effects can be provided as patients move through these phases of treatment. Currently, it seems that the added benefit of letrozole year by year, compared with the previous year, remains constant, at least through the first 3 to 4 years of therapy. Furthermore, women from the MA-17 trial who have completed 5 years of letrozole will be offered the opportunity to undergo another random assignment between a further 5 years of letrozole or placebo, thus JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 22 AUGUST 1 2005