Abstract P2-05-02: Active surveillance with a combination of tumor marker CA27.29 and detection of circulating tumor cells two year after primary diagnosis strongly predicts subsequent prognosis

Abstract Introduction The prognosis of patients with early breast cancer is commonly estimated by prognostic factors obtained at the time of the initial diagnosis. However, patients and physicians alike are seeking for factors evaluating the prognosis years thereafter during follow-up. The identification of a patient group with an unfavourable prognosis could lead to secondary treatment intervention, potentially improving outcome. Aim of the study was to assess the added prognostic value of circulating tumor cells (CTCs) and CA27.29 beyond established predictors. Materials and Methods Patients of the phase III SUCCESS-A study were included into this analysis (n=1005). SUCCESS-A is a chemotherapy study for high risk patients with a comprehensive translational research program, which included the determination of CTCs and CA27.29 two years after the initial diagnosis. A Cox regression model with disease-free survival (DFS) as outcome and well-established predictors (age, BMI, pT, pN, histology, grading, ER, PR, Her2neu) was compared with an extended Cox model with the well-established predictors and additionally CTC (>0 versus 0) two years after randomization, and CA27.29 (in U/mL) measured after chemotherapy and again two years after randomization using a likelihood ratio test. In case of significance, the extended model was applied to predict for each patient the risk of disease recurrence within the next 12 months (0 to 100%). Cross-validated AUC, sensitivity and specificity values were determined to assess clinical usefulness of risk prediction. Results The markers CA27.29 and CTC were both significantly associated with subsequent prognosis (p < 0.000001). The detection of CTCs increased the risk of subsequent DFS events (HR=2.14, 95%CI: 1.31-3.48), while CA27.29 after two years increased the risk for DFS events with a HR of 1.12 per U/mL increase (95%CI: 1.09-1.15). The combination of the two markers significantly empowered the prognostic relevance, with a HR of 6.64 for patients with CTCs and an elevated CA27.29 by 10 U/mL compared to patients without CTCs and without CA27.29 elevation. The mean risk of disease recurrence in the third year after randomization was 2.38%. Discrimination of patients with and without disease recurrence based on risk prediction from the extended Cox model (AUC: 0.80) was better than discrimination based on the clinical model without the markers CTC and CA27.29 (AUC: 0.64). Sensitivity with regard to decision thresholds 1%, 2%, 3%, and 4% predicted risk was 0.89, 0.77, 0.65, and 0.55, respectively. The corresponding specificity was 0.42, 0.69, 0.81, and 0.88. Discussion Both CTCs and CA29.27 values determined 2 years after primary diagnosis are clinically relevant predictors of subsequent prognosis for those patients. This study extends evidence for active surveillance of breast cancer survivors. Identifying a group of women with a high recurrence risk after two years could be the basis for the development of secondary adjuvant treatment. Citation Format: Janni W, Rack B, Häberle L, Friedl TWP, Tesch H, Lorenz R, Jäger B, Fehm T, Müller V, Schneeweiß A, Lichtenegger W, Blohmer J, Beckmann MW, Scholz C, Pantel K, Trapp E, Fasching PA. Active surveillance with a combination of tumor marker CA27.29 and detection of circulating tumor cells two year after primary diagnosis strongly predicts subsequent prognosis [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-05-02.