AbstractBackground:Detection of minimal residual disease (MRD) via circulating tumor DNA (ctDNA) is a promising biomarker for risk stratification and treatment guidance in cancer. This study evaluated the clinical impact of tumor-informed ctDNA MRD monitoring during the perioperative period in stage II-III breast cancer (BC).Patients and Methods:Patients with stage II-III BC were prospectively enrolled at Seoul National University Hospital (Feb 2023 - Jul 2024). In the neoadjuvant systemic therapy (NST) arm, 20mL of whole blood (WB) was collected at eight perioperative time points: baseline (N00), after one cycle of NST (N01), after completion of NST (N03), and at 1, 6, 12, 24, and 36 months post-surgery. In the adjuvant chemotherapy (AC) arm, WB was collected at seven postoperative time points: 1, 2 months post-AC, and at 6, 12, 24, 36, and 48 months post-surgery. Plasma samples were analyzed using the CancerDetectTM assay (IMBdx, Inc), a personalized tumor-informed MRD detection test. Samples with ≥2 detected mutations were classified as ctDNA+.Results:A total of 231 female patients were enrolled, and 189 met eligibility criteria. This analysis included 111 patients with analyzable WES data, focusing on 103 NST patients. Number of patients analyzed were 103 at N00, 100 at N01, 94 at N03, and 46, 52, 10 at 1, 6, 12 months post-op, respectively. Median age was 51 years (25-72). There were 48 (46.7%) HER2-positive (HER+), 31 (30.1%) triple-negative (TNBC), and 24 (23.3%) hormone receptor (HR)-positive/HER2-negative (HR+/HER2-) BCs. 58 (56.3%) were stage II and 45 (43.7%) were stage III. Median number of target variants from WES was 53 (8-100) and higher in TNBC (67) and HER2+ BC (55) than HR+/HER2- BC (31.5, p = 0.006). Baseline ctDNA positivity was 81.6% (84/103), and ctDNA+ rates by subtype were as follows: 87.1% for TNBC, 83.3% for HER2+ BC, and 70.8% for HR+/HER2- BC (p = 0.281). The positive rate by stage were 77.6% and 86.7% for stage II and stage III, respectively (p = 0.356). Median ctDNA fractions were higher in TNBC (0.186) than HER2+ BC (0.020) or HR+/HER2- BC (0.018, p = 0.031), and higher in stage III (0.125) than stage II (0.012, p = 0.003).After NST completion, the overall pathologic complete response (pCR) rate was 57.3% and ctDNA positivity dropped to 20.2% (19/94), and ctDNA negativity at N03 was associated with higher pCR rates (62.7% vs. 26.3%, p = 0.010). Among N00 ctDNA+ patients, 79.7%achieved clearance after NST, which correlated with higher pCR rates (66.1% vs. 12.5%, p < 0.001). In this subpopulation (median follow-up 13.6, 6.7-19.9), only 4.9% (2/41) of the pCR patients were ctDNA+, no recurrences were reported.Conclusions:Tumor-informed ctDNA assays correlated with NST response in stage II-III BC. Variations in target variants recovered from WES may explain differences in ctDNA positivity by subtype, which should be considered for clinical interpretation and application.Citation Format:Jongseong Ahn, Jiwoong Jung, Yongjun Cha, Sunghoon Heo, Eunhye Kang, Woosung Chung, Hwang-Phill Kim, Duhee Bang, Sang-Hyun Song, Han-Byeol Lee, Wonshik Han, Tae-You Kim. Preliminary results of perioperative minimal residual disease monitoring using a tumor-informed ctDNA assay in stage II-III breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4572.