De-escalation of breast cancer treatment in older adults from diverse racial and ethnic backgrounds: A retrospective review.

268 Background: De-escalation of breast cancer treatment in older adults with early-stage, hormone receptor-positive disease is supported by evidence showing reduced treatment burden without compromising outcomes. However, little is known about how often these strategies are discussed or implemented in racially and ethnically diverse populations. This study aimed to evaluate the frequency of de-escalation discussions and identify associated clinical factors. Methods: We conducted a retrospective chart review of women aged ≥65 with localized breast cancer treated at the University of Colorado system in 2024. Inclusion was limited to patients self-identifying as members of underrepresented racial or ethnic groups. Abstracted data included demographics, tumor characteristics, comorbidities, use of geriatric decision tools, and documentation of treatment de-escalation. The primary outcome was documentation of a de-escalation discussion, defined as provider-patient communication regarding potential omission of surgery, chemotherapy, radiation, or staging procedures. Logistic regression was used to identify predictors. Results: Fifty-nine patients met inclusion criteria. The mean age was 70.6 years (range 65–85), and median comorbidity count was 3 (range 1–10). Racial/ethnic distribution was 33% Hispanic, 22% Black, 11% Asian, 7% Native American, 8% mixed, and 10% other. Most tumors were Stage I (49%), hormone receptor-positive (92%), and HER2-negative (88%). Only 12% of patients had documentation of a geriatric assessment. Oncotype DX was used to guide therapy in 34% of cases. Axillary staging was omitted in 27% of HR+ patients, while radiation was omitted in 42% of patients with Stage I disease. Overall, 57% of patients had documented discussions about de-escalation with members of the multidisciplinary team (46% with radiation oncology, 29% with medical oncology, and 19% with surgical oncology). In univariate analysis, lower tumor stage (p = 0.03) and availability of Oncotype DX (p = 0.047) were associated with higher likelihood of discussion. Race and ethnicity were not significantly associated with the likelihood of de-escalation discussions (p > 0.1). On multivariate analysis, only stage remained significant (OR 0.40, 95% CI 0.16–0.99, p = 0.04). Age and Oncotype availability were not statistically significant. Conclusions: De-escalation of breast cancer care is discussed in just over half of older adults from underrepresented racial and ethnic groups, mirroring rates seen in majority populations. Tumor stage was the key determinant of whether de-escalation was addressed. Greater integration of geriatric principles and structured shared decision-making could promote more equitable and individualized care.