Abstract Background: Because of the diagnosis at an earlier stage, and the wider availability of more effective therapies, treatment outcomes of patients with breast cancer are getting much better resulting in an increasing proportion of patients surviving their disease, however, a significant percentage of patients with early-stage disease at initial diagnosis may eventually relapse. Residual tumor cells can remain dormant for many years before causing tumor recurrence. Physical examination and mammography are strongly recommended in surveillance guidelines, but data on routine blood tests or imaging is lacking. The revolution made in recent years, following the approval of novel targeted and immunotherapy agents, may have better outcome if started earlier in the disease course with the lowest tumor burden and in patients with good performance status and adequate organs’ function. Additionally, early detection of oligometastatic disease can give a chance for cure for a subset of these patients. Levels of CA15.3 and CEA are used in decision making in the metastatic setting but their role in surveillance is still controversial. We aim to investigate whether serial measurement of tumor markers, when it correlates with tumor bulk (conditional), can detect early asymptomatic recurrence among subset of patients with high-risk early-stage breast cancer. Methods: Patients with high-risk early-stage breast cancer were invited to participate. High-risk features include any of the following: large tumor size (T3/4), grade-3, node-positive, triple-negative, HER2-positive or hormone receptor (HR)-negative disease. Patients were considered eligible if they have elevated CA15.3 and/or CEA at baseline, that subsequently normalize 6 weeks after surgical resection; upfront or following neoadjuvant chemotherapy. Serial testing of the elevated CA15.3 and/or CEA is done every 2 months thereafter, samples are cryopreserved and will not be processed until disease relapse. As such results will not be used for clinical decisions. The study is still ongoing, we here present an interim analysis evaluating the correlation between certain tumor characteristics and elevated serum marker(s) levels. Results: Since the launch of the study, 367 patients deemed high-risk by the above criteria, accepted to participate. Abnormal baseline CA15.3 and/or CEA was found in 110 (30%); 65 (18%) patients had elevated CA 15.3 while elevated CEA was found in 45 (12%) patients. Abnormal CA 15.3 was found in 27% of cases with T3/4 disease compared with 13% in T1/2 disease (p=0.001), and in 21% of node positive disease versus 7% in node negative disease (p=0.002), in 9% of HR negative disease versus 20% in HR-positive (P=0.038). Tumor grade and HER2 status were not associated with significant difference. Regarding CEA, abnormal levels were detected in 24% of HER2-positive disease compared to 5% of HER2-negative disease (P< 0.0001). Other tumor features were not associated with significant difference. Conclusions: Only a third of patients with high-risk early-stage breast cancer had high level of serum tumor markers at initial diagnosis and thus such markers may correlate with tumor presence. Large tumors (T3/4) and node positive disease were associated with high CA15.3 but not CEA. HER2-stutus and tumor grade had no effect on CA15.3 level, while HER2 positive disease correlated well with high CEA. Citation Format: Sarah Edaily, Baha' Sharaf, Maher Sughayer, Lina Yousef, Hala Abu-Fares, Sarah Abdel-Razeq, Hala Abu-Jaish, Mahmoud Abunasser, Suhaib Khater, Anas Zayed, Osama El Khatib, Rahaf Rahal, Hazem Abdulelah, Maria Shraim, Hikmat Abdel-Razeq. Conditional Use of Serial Serum Tumor Markers CA 15.3 and CEA as Predictors of Disease Relapse in Patients with High-Risk Early-Stage Breast Cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-03-03.