2704 - Treatment Implications of Low Compliance with Endocrine Therapy for Older Women with Early-Stage Breast Cancer

M. Lubas¹, R. M. Shulman², J. E. Meyer³, J. Fredette², R. Bleicher², A. Williams², and J. Hasler²; ¹Fox Chase Cancer Center, Philadelphia, PA, United States, ²Fox Chase Cancer Center, Philadelphia, PA, ³Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA

Purpose/Objective(s): Older patients with hormone receptor-positive, HER2-negative, early-stage breast cancer (HPEsBC) are routinely treated with breast-conserving surgery (BCS) followed by adjuvant radiation therapy (RT), oral endocrine therapy (ET), or both. However, for patients who are unable to complete a full 5-year course of ET, RT may improve oncologic outcomes. Our study of HPEsBC was undertaken to identify predictors of low ET compliance and to assess the role of RT in altering survival for older ET low-compliant patients. Materials/

Methods: Using the nationwide Flatiron Health electronic health record-derived de-identified database, patients = 65 years old with clinical or pathologic T1-2N0 HPEsBC treated with BCS from 2011-2018 were reviewed. Four adjuvant groups were identified: ET, RT, ET + RT, and no therapy. A 60-day landmark Kaplan-Meier (KM) curve and Cox proportional hazards model were used to compare OS. Low ET compliance was defined as ET taken for <4 years (compliance < 80%). Multinomial logistic regression then isolated predictors of ET compliance. Using the Cox model with ET compliance as a time-varying covariate, hazard ratios for compliance and radiation were determined while adjusting for age, race, ethnicity, socioeconomic status quintile, insurance, clinic visits, practice type, ECOG, tumor grade, and histology.
Results: The study group included 1,488 HPEsBC patients (354 ET, 118 RT, 894 ET + RT, and 122 no therapy). Low compliance was recorded for 287 of 1248 ET-treated patients (23%). In an unadjusted analysis, median OS (mOS) was greater for patients treated with ET + RT than for patients receiving ET alone, RT alone, or no therapy (mOS not reached, mOS 125.1 mos, mOS 120.3 mos, mOS 97.4 mos, respectively; p<0.001). When the effect of treatment was adjusted for all covariates, mortality was greater for RT alone and no therapy than for ET + RT (HR=1.77, p=0.011; HR=2.53, p<0.001, respectively). Increasing age and treatment at an academic practice (vs community) were associated with low ET compliance (OR=1.06, p<0.010; OR=2.58, p<0.001, respectively). The Cox model with a time-varying covariate demonstrated an increased risk of mortality when ET compliance declined 10% (HR=1.17, p<0.001). An interaction between compliance and RT showed no differential effect of radiation based on levels of compliance.
Conclusion: Low compliance was observed in almost 1 in 4 HPEsBC patients = 65 years old and was associated with advancing age and treatment at an academic center. Declines in ET compliance as small as 10% were associated with significantly increased mortality. Further study to identify a possible survival benefit from RT for elderly HPEsBC patients with ET compliance levels much lower than 80% is warranted.