Identifying Breast Cancer Patients Most Likely to Benefit From Aromatase Inhibitor Therapy After Adjuvant Tamoxifen W e are responding to the recently published study by Freedman and colleagues in which it is suggested that the benefits of adding an aromatase inhibitor after 5 years of adjuvant tamoxifen in patients with early-stage breast cancer are small. 1 First, we wish to emphasize that decisions regarding adjuvant hormonal therapy for individual patients must take into account the absolute risk of future disease recurrence and mortality as well as the proportionate reduction in risk produced by treatment.
Results from the National Cancer Institute of Canada Clinical Trials Group MA-17 trial indicate a 42% reduction in the risk of recurrent breast cancer and contralateral new primary breast cancer among patients who received letrozole after 5 years of adjuvant tamoxifen. 2 In their study of 471 patients who completed 5 years of adjuvant tamoxifen, Freedman et al. reported an absolute event rate of 7% in Years 5 through 10. Accordingly, the authors concluded that the absolute benefit of an aromatase inhibitor after adjuvant tamoxifen is small. They stated that patients with a higher baseline risk of recurrence have more substantial absolute benefits.
The absolute risk of recurrence in the study by Freedman and colleagues was low compared with published figures from other larger studies. In the placebo arm of MA-17 (n ΒΌ 2594 women), 10.2% of patients developed recurrent disease at 4 years after they completed tamoxifen. In a recently published, population-based study (n ΒΌ 1086 women) from the British Columbia Cancer Agency (BCCA), the event rate in Years 5 through 10 after 5 years of adjuvant tamoxifen was 14.6%. 3 Furthermore, the most recent Oxford Overview reported an event rate in Years 5 through 10 of approximately 10%. Consistent with their lower event rate is the fact that the cohort studied by Freedman et al. appeared to have a lower baseline risk of recurrence compared with the MA-17 and BCCA populations. The proportion of women who were lymph node-negative was much higher in the study by Freedman et al. (68%) than in the MA-17 study (49%) and the BCCA study (42%).
Thus, although we do not disagree with the methodology or reasoning employed by Freedman et al., we believe it is critical that clinicians who decide whether to recommend an aromatase inhibitor after 5 years of tamoxifen use the best available data to assess the baseline risk of recurrence. Because it appears that the event rate in the cohort reported by Freedman et al. may be an underestimate, clinicians who are trying to decide whether to advise extended adjuvant therapy would be better advised to rely on tools such as Adjuvant Online 5 that make use of much more substantial datasets to estimate baseline risk.