Real-world pathologic complete response after neoadjuvant chemotherapy in a safety-net breast cancer population.
Abstract
556 Background: While clinical trials report subtype-specific pathologic complete response (pCR) rates, real-world data from safety-net health systems remain limited and underrepresented in major neoadjuvant trials. Our urban safety-net patient population in Oakland, CA, provides a unique view of real-world neoadjuvant outcomes which is an important marker of outcomes in breast cancer. Methods: Non-metastatic breast cancer patients who had neoadjuvant chemotherapy (NAC) were retrospectively reviewed from Highland Hospital between January 2022 and September 2025. Demographic, tumor, and treatment data, time to treatment initiation [TTI] and adherence (≥10 missed appointments) were analyzed. Results: The total study cohort of 53 patients were predominantly Hispanic/Latino 27(50.9%) and Black African Descent 15 (28.3%), followed by White 6 (11.3%), Asian 4 (7.5%) and Pacific Islander 1 (1.9%). Most were Hormone-Receptor-positive/HER2-negative (HR+/HER2-) 24 (45.3%), while 21 (39.6%) were HER2-positive, and 8 (15.1%) had Triple-negative breast cancer (TNBC). The overall pCR rate was 32.1% (17/53; 95% CI 19.9–46.3). By subtype, pCR was achieved in 50.0% of TNBC (n = 4/8), 47.6% of HER2-positive (n = 10/21), and 12.5% of HR+/HER2- (n = 3/24) patients. Notably, among the patients with HR+/HER2- disease who achieved pCR, two-thirds (66.7%, n = ⅔) exhibited low estrogen receptor expression ( < 10% ER+ staining). Overall treatment completion was high at 94.3%. However, substantial care barriers were identified with 39.6% of total patients experiencing TTI > 60 days (median TTI: 56 days), and 34% of patients had ≥10 missed clinic visits. For HER2-positive disease, a TTI > 60 days was significantly associated with a failure to achieve pCR (63.6% vs. 0%, p = 0.004). Conclusions: In this real-world cohort from a safety-net hospital, although pCR rates for HR+ was comparable to landmark clinical trials, response rates were lower than expected for TNBC and HER2 + disease (50% vs. 63% on keynote 522 or 47% vs. 57-66% in Her2+disease ). Notably, pCR rates were similar between TNBC and HER2+ subtypes in our cohort, in contrast to clinical trial data where HER2+ disease often achieves higher pCR rates with optimized targeted therapy. This efficacy gap likely reflects real-world treatment barriers, prolonged TTI, and distinct tumor biology unique to safety-net populations. These findings highlight actionable quality improvement targets such as fast-track navigation protocols to shorten TTI, optimizing access to guideline-concordant therapy.