循环肿瘤DNA微小残留病灶作为乳腺癌临床试验终点的真实世界全景分析

Background: Minimal Residual Disease (MRD) assessment via circulating tumor DNA (ctDNA) offers superior sensitivity over imaging for relapse detection and monitoring. In breast cancer, MRD identifies molecular relapse with a median 8.9-month lead time from prior studies, serving as an independent prognostic and potential predictive marker for outcome and therapy escalation. Methods: A systematic ClinicalTrials.gov search (July 2025) yielded 133 trials; 124 met inclusion. Studies were stratified by endpoint hierarchy (primary vs. secondary), status, and methodology (tumor-informed, panel-based, methylation, or CTC/protein/tissue). Results: Among 124 trials, MRD was primary endpoint in 56 (45%) and secondary in 66 (53%). Only 4 primary-endpoint trials have published results, while 52 are ongoing. Tumor-informed/patient-specific assays were the dominant modality (38%), followed by CTC/tissue (20%), fixed panels (17%), and methylation (7%); 18% were unreported. Published clinical utility (primary-endpoint trials). In metastatic guidance (NCT05079074), 85% of patients had baseline ctDNA positivity, and those receiving druggable mutation-guided therapy achieved improved progression-free survival (PFS) versus physician's choice (HR 0.45). In a prognostication cohort (NCT03792529), genomic profiling identified frequent TP53 (44%), PIK3CA (28.4%), and ERBB2 (24.8%) alterations; high ctDNA fraction and blood tumor mutational burden (bTMB) predicted shorter PFS in TNBC and HER2+ subtypes. In the relapse-prediction setting (NCT02797652), baseline ctDNA was detected in ~70% (21/30) of operable patients; post-operative ctDNA positivity was 100% predictive of distant metastasis, yielding 71.4% sensitivity for clinical relapses. In an early-detection setting (NCT05227261), among 9,024 individuals screened (0.48% positive), a multi-cancer test demonstrated 99.71% specificity, 70.83% sensitivity, 39.53% positive predictive value, and 99.92% negative predictive value. Monitoring and MRD-adaptive strategies. Serial ctDNA clearance in immunotherapy trials (e.g., NCT02644369) tracked survival, with 100% of patients achieving clearance remaining alive at a median 25-month follow-up. Active MRD-adaptive investigations now deploy randomization to escalation at molecular relapses (e.g., DARE/NCT04567420), single-arm escalation in MRD-positive HER2+ disease (e.g., NCT05388149), and clearance-based de-escalation (e.g., NCT06970912). Conclusions: ctDNA-MRD integration in breast cancer trials is expanding rapidly, shifting to tumor-informed assays. Published data validate ctDNA as a potent prognostic tool (71.4% relapse sensitivity) and predictive biomarker (HR 0.45 intervention benefit). As evidence matures with ongoing randomized trials, assay standardization and actionability remains critical for clinical translation.