Rates of inadequate ovarian suppression in premenopausal breast cancer patients receiving ovarian function suppression.

Abstract

e12730 Background: Ovarian function suppression (OFS) improves outcomes for premenopausal patients with high-risk estrogen receptor-positive (ER+) breast cancer. However, real-world monitoring of estradiol (E2) levels and the adequacy of biochemical suppression remain understudied. We evaluated patterns of OFS use, E2 testing, and suppression effectiveness across a large academic cancer center network. Methods: We retrospectively analyzed 3,113 patients ≤55 years with stage I-III ER+ breast cancer treated with definitive surgery and adjuvant endocrine therapy from 2018-2024 across 23 sites of the University of Pittsburgh Medical Center (UPMC) Hillman Cancer Center. Demographics, clinic-pathologic information, OFS receipt, endocrine therapy, E2 testing, and suppression inadequacy (E2 ≥15 pg/mL) were collected. Logistic regression identified predictors of OFS use. Survival was assessed using stage-adjusted and time-varying Cox models. Results: Among 3,113 patients, 2,051 were premenopausal, of whom 672 (33%) received OFS. In adjusted logistic models, younger age (odds ratio [OR] per one-year increase: 0.91; 95% CI: 0.89-0.93), nodal involvement (N1: OR 1.85; 95% CI: 1.34-2.56), higher grade (Grade III: OR 1.60; 95% CI: 1.12-2.29), elevated Ki-67 ≥20% (OR 1.33; 95% CI: 1.04-1.70), and chemotherapy receipt (OR 1.42; 95% CI: 1.11-1.80) strongly predicted OFS use. Among 640 patients with known OFS start dates, only 53% underwent any E2 testing. Inadequate ovarian suppression occurred in 50% of monitored patients. Younger age strongly predicted inadequate suppression (50% if < 40 vs 26% if 45-50; p < 0.005). Inadequate suppression was more common with tamoxifen than AIs (54.9% vs 41.2%; OR 1.73, p < 0.005). Both liquid chromatography-mass spectrometry and sensitive immunoassays detected significant rates of inadequate suppression (29% and 14% of tests, respectively), with similar overall distribution patterns. In survival analyses, the low number of events limited the power to detect overall survival or disease-free survival differences in standard Cox models. However, in a time-varying Cox model among premenopausal patients, OFS was associated with 82% reduction in mortality (HR = 0.18, p = 0.001) when accounting for immortal time bias. Conclusions: In this large real-world cohort, OFS was more frequently used in higher-risk patients. However, E2 monitoring was infrequent, and nearly half of the monitored patients experienced inadequate ovarian suppression, particularly younger women and those on tamoxifen. It is unclear if E2 values can be interpreted accurately as Tamoxifen is associated with increased serum E2 levels. These findings highlight the need for standardized E2 monitoring protocols and timely OFS administration to ensure consistent ovarian suppression in real-world clinical practice.