Revisiting Adjuvant Hormone Therapy in Premenopausal Women With Breast Cancer: Escalation and De-Escalation

Abstract

have published guidelines for locoregional breast cancer that suggest in premenopausal women, clini-cians should give adjuvant tamoxifen for 5 years in low-risk breast cancer, extend hormone therapy in high-risk patients, and consider adding ovarian function suppression (OFS) in patients with a higher risk of relapse. However, the guidelines are relatively ambiguous about patient selection. 1,2 The article by Sella et al 3 is of great value in guiding clinical decision making. However, some points remain to be elucidated. Patient Selection for Ovarian Suppression Early meta-analysis studies showed a positive effect of ovarian ablation or suppression, but the population was heterogeneous, and they were conducted at a time in which modern chemotherapy and antihuman epidermal growth factor receptor 2 therapies were lacking. 4-6 In the INT-0142 study, among premeno-pausal women with node-negative, hormone receptor-positive breast cancer, OFS failed to show a statistically significant benefit. However, this trial was under-powered because of early termination of accrual. Moreover, since chemotherapy was not permitted, it is possible that treating physicians only offered patients with the lowest risk of recurrence (even among patients who met the inclusion criteria) the chance to participate in this trial. 7 In the Korean ASTRRA study, the addition of 2 years of OFS to tamoxifen improved disease-free survival (DFS) and overall survival (OS) in patients who remained premenopausal or resumed ovarian function after chemotherapy. There was no evidence of a heterogeneity of treatment effect in the subgroup analysis. 8 The SOFT/TEXT joint analysis is a landmark study that provides detailed information about patient selection. The hazard ratio of risk reduction was similar across subgroups. 9 Because of the small absolute benefit in the low-risk patients, Sella et al 3 provide a useful table, using distant recurrence rate cutoffs of , 5%, 5% to 15%, and. 15%, calculated by the Regan score, as low, intermediate, and high risk. It is extremely helpful in guiding patient selection for OFS and extended hormone therapy. Optimal Duration of Ovarian Function Suppression The optimal duration of OFS is yet to be determined. The most widely used standard, as the protocol of the SOFT and TEXT trial, is OFS for 5 years. 9 The HOBOE trial also demonstrates DFS benefit with 5 years of OFS plus letrozole over tamoxifen alone although OS benefit was not detected. On the other hand, according to the Korean ASTRRA study, adding 2 years of OFS still improves DFS and OS. 8 Similarly, in the earlier ZIPP study, adding 2 years of goserelin provides a significant benefit for event-free survival and OS. 10 The article written by Sella et al 3 suggests that clinicians may consider prolonging OFS after the initial 5 years in high-risk cases. Given the lack of clinical data comparing different durations of OFS, the duration of treatment should be tailored to each patient after shared decision making. The patient's age at the initiation of OFS, and whether they are approaching menopause, may be taken into consideration. The Optimal Agent to Pair with OFS The selection of the endocrine therapy accompanying OFS is also an issue. The ABCSG-12 study showed no benefit of anastrozole over tamoxifen when combined with OFS. 11,12 By contrast, the SOFT/TEXT data provided evidence of a benefit of aromatase inhibitors over ta-moxifen in DFS although not in OS. 9 Ovarian escape may be an underscored issue that explains this inconsistent result. We use fixed dose of aromatase, which may be insufficient for overweight patients, who have higher proportion of fat, leading to a higher level of aromatase production. In ABCSG 12, overweight patients treated with anastrozole were associated with higher risk of disease recurrence and death. 13 In the SOFTEST substudy, among 25% of the patients, the estrogen was not well suppressed. 14 No prior chemotherapy (P 5 .06) and higher body mass index (P 5 .05) were associated with increased on-treatment estrogen level. In the ABCSG-12 study, none of the patients received postoperative chemotherapy and there was a higher percentage of patients younger than 40 years old (18%) compared with the nonchemotherapy population in the SOFT/TEXT data (9.3% in SOFT and 15.6% in TEXT). It is possible that more patients in the ASSOCIATED