Abstract PS1-13-29: Carcinoembryonic antigen (CEA), Cancer Antigen 15.3 (CA 15.3), and 18-FDG-PET in Early Breast Cancer Follow-up: Findings from the KRONOS Trial

Abstract Background: Current breast cancer (BC) follow-up guidelines for asymptomatic patients recommend only annual mammography and periodic physical exams, based on two trials published in 1994 (JAMA vol. 271). No subsequent randomized controlled trials (RCTs) have evaluated alternative strategies. Advances in BC biology, treatments, and diagnostic tools like 18-FDG-PET have transformed the landscape, prompting a need to reassess surveillance protocols. The role of serum tumor markers (CEA, CA 15.3) and 18-FDG-PET have never been validated in prospective RCTs. Study Design: The KRONOS trial for the first time over 30 years later the above cited RCTs, compares standard follow-up (annual mammography and physical exam; control arm) with an intensified approach (standard plus quarterly serum CEA and CA 15.3, with 18-FDG-PET performed if markers rise critically; experimental arm). The primary endpoint assesses whether intensified surveillance detects recurrences earlier, measured by the difference in restricted mean survival time (RMST). Assuming a 20% five-year recurrence, a 3-month RMST reduction (diagnostic lead time ∼10 months) was targeted. If early diagnosis of metastases is demonstrated, the impact on survival will be evaluated as the final endpoint. Secondary endpoints include marker predictive value, PET accuracy, and quality of life. Patients and Methods: Eligible patients had stage I-III BC after adequate surgery, only very low-risk cases and rare histologic types were excluded. The study included two cohorts: newly starting follow-up (cohort 1) and those completing five years without relapse (cohort 2). Participants were randomized 1:1, stratified by nodal status, HER2, and ER status. Follow-up was planned for at least 10 years after surgery. The mean diagnostic anticipation achievable in the experimental arm was assessed by calculating the difference in RMST between the 2 arms and testing such a difference non-parametrically by means of bootstrap resampling. Results: Between October 2014 and November 2021,1507 patients were randomized: 742 to control arm, 765 to experimental arm. Median age was 56 years; 52% stage I, 34.9% stage II, 13.1% stage III. Tumor biology: 64.8% ER-positive/HER2-negative, 25.8% HER2-positive, 9.4% triple-negative. Median follow-up was 7 years; 130 recurrences (8.6%) occurred—101 distant metastases (6.7%) and 29 locoregional relapses (1.9%). The 10-year relapse-free survival was 88.5% in the experimental arm vs. 89.2% in the control arm. Additionally, 89 patients (5.9%) developed new primary cancers, and 20 (1.3%) died without recurrence. The RMST over 10 years was 7.7 years (experimental) vs. 8.3 years (control), indicating an anticipatory benefit of 7.4 months (95% CI -3.1-13.4). Overall survival was similar, with 97 deaths (6.4%) and no significant difference between groups. Subgroups analysis by cohort, stage and biological subtypes will be also presented at the symposium. Conclusions: The 7-year recurrence rate (∼8.6%) was lower than expected. Intensified follow-up did not improve overall survival in this unselected BC population. These findings support current standard surveillance, with no evidence of benefit from routine use of serum markers or 18-FDG-PET in asymptomatic patients. Citation Format: C. Zamagni, R. Wirtz, V. Torri, D. Sartori, M. Carapelle, L. Vivona, P. Stieber, J. Hubner, M. Bergamo, L. Tondulli, C. Pizzirani, S. Coccato, N. Cacciari, A. Baldoni, A. Bernardi, T. Dalsass, S. Quercia, S. Prader, E. Haspinger, D. Rubino, A. Mandrioli, M. Cubelli, F. Abbati, M. Massucci, R. Pagani, S. Fanti, M. Gion. Carcinoembryonic antigen (CEA), Cancer Antigen 15.3 (CA 15.3), and 18-FDG-PET in Early Breast Cancer Follow-up: Findings from the KRONOS Trial [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS1-13-29.