Abstract Background: Adjuvant endocrine therapy (AET) approximately halves the risk of recurrence among the two-thirds of premenopausal breast cancer patients whose tumors overexpress the estrogen receptor (ER+). AET is recommended for a minimum five-year duration, but many premenopausal women discontinue AET prematurely. Adherence to AET is dynamic and influenced by many behavioral, societal, clinical, and genetic factors. Describing longitudinal patterns of AET adherence is an important first step towards identifying patients at highest risk of nonadherence–-and therefore recurrence–-who may benefit from adherence-enhancing interventions. We used group-based trajectory models to (1) describe AET adherence patterns, and (2) estimate associations between AET adherence and breast cancer recurrence in a Danish premenopausal, ER+ breast cancer cohort. Methods: Our cohort included 4,487 premenopausal women diagnosed with ER+, stage I–III primary breast cancer registered in the Danish Breast Cancer Group (DBCG) clinical database and treated with AET (2002–2011). We followed all patients from initiation of AET until recurrence, mortality, 10 years of follow-up, another primary malignancy, or the end of available follow-up. We excluded individuals with less than 1.5 years of follow-up as breast cancer recurrences that occur in this time are unlikely to be related to AET. At semi-annual follow-up visits, women were registered as having received (or not) a six-month supply of AET. We created daily indicators to determine if a woman had at least 80% of days covered by an AET fill within each six-month period for the first 4.5 years following AET initiation, yielding 9 variables indicating time-varying adherence. We used these indicators to characterize adherence trajectory groups. We defined recurrence as local, regional or distant metastases or contralateral breast cancer registered in the DBCG using follow-up data through 2017. We fit Cox regression models, adjusted for age and clinical factors, to estimate the association between adherence trajectory groups and recurrence. Results: We identified distinct trajectory groups in the cohort (Table 1)—high adherence (69%), slow decline (23%), and quick decline (8%). Women with quick decline were more likely to have stage I disease, have no positive lymph nodes, to not be treated with chemotherapy, and to have lumpectomy instead of mastectomy when compared to women with high adherence. Compared with high adherence women, the estimated rate of breast cancer recurrence was higher among those with slow decline (HR=1.12, 95%CI=1.04–1.21) and those with quick decline (HR=1.38, 95%CI=1.22–1.56). Conclusions: Group-trajectory modeling facilitated empirical description of AET adherence patterns in a premenopausal breast cancer cohort. As expected, women who more rapidly became AET-nonadherent were at increased risk of breast cancer recurrence compared with those who remained adherent. Future analyses will explore the contribution of specific socioeconomic, genetic, and clinical factors to AET adherence patterns, which will inform the design of interventions to improve AET adherence and reduce recurrence risk in breast cancer patients. Table 1. Trajectory group and the observed proportion of individuals adherent to their adjuvant endocrine therapy (AET) over 4.5 years of follow-up Citation Format: Kirsten M. Woolpert, Julie A. Schmidt, Thomas P. Ahern, Timothy L. Lash, Lindsay J. Collin, Bent Ejlertsen, Deirdre Cronin-Fenton. Group-based trajectories of endocrine therapy adherence and risk of recurrence in a Danish premenopausal breast cancer cohort [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-03-18.