Abstract PO4-02-05: Elevated tumor to aorta ratio of Hounsfield units and late recurrence in patients with estrogen receptor positive breast cancer

Abstract

Abstract Background) More than half of the recurrent disease in breast cancer will happen more than 5 years after initial diagnosis in breast cancer, especially in patients with ER positive disease. Higher proportions of late recurrence lead to escalated treatment and/or or extended treatment for high risk patients. However, it is unclear what characteristics make whom as high risk patients. This study aimed to assess initial tumor vascularity as a prognostic marker for late recurrence. Methods) Female patients with estrogen receptor positive breast cancer who were diagnosed breast cancer between 2003 and 2018 and disease-free at 5 years after primary breast cancer at Wonju Severance Hospital, Korea, were included. Clinocopathological characteristics were collected. Hounsfield units(HU) on contrast-enhanced computed tomography(CT) was used as a marker indicating tumor vascularity. Tumor to aortic arch ratio(TAR) of HU on contrast enhanced CT was applied to enhance objectivity of measurement. Patients were categorized according to the cut-off values retrieved from the receiver operating characteristic curve. Kaplan-Meier curves were generated to compare recurrence-free survival (RFS) and overall survival (OS). Hazard ratio(HR) with confidence interval(CI) was derived with Cox’s proportional hazard model to analyze univariate and multivariate risk factors. Results) The final cohort included 451 patients with a mean age of 56.21 ± 11.2 (22-83) years. Two third of patients had estrogen receptor positive disease and a quarter of patients had HER2 overexpressed disease. Initial TAR was 0.345 ± 0.108 (range, 0.062 - 1.114). Patients with recurrence free survival(RFS)-related events had significantly higher TAR than patients who did not(0.39 ± 0.097 vs. 0.34 ± 0.108, p=0.012). Patients with overall survival(OS)-related events also had higher TAR than patients who did not(0.408 ± 0.096 vs. 0.342 ± 0.107, p=0.002). Cutoff value of 0.408 was driven from area under the receiver operating characteristic curve. Patients with TAR higher than 0.408 showed significantly worse RFS(p=0.001) and OS(p< 0.001) than patients who had TAR equal or lower than 0.408. Hazard ratio(HR) of TAR in late recurrence was 6.17 (CI, 4.89 – 9.23, p=0.003), suggesting it is an independent factor. Other independent factors were age (HR, 1.502; CI, 10.14 – 10.91; p=0.006), tumor size (HR, 1.268; CI, 1.071 – 1.502; p=0.006), and metastatic nodal disease (HR, 2.770; CI, 1.176 – 6.526; p=0.020). Conclusion) TAR of primary tumor was significantly related with patients RFS and OS. Patients with high TAR larger than 0.408 showed worse RFS and OS than patients who did not. According to the Cox proportional hazard model, TAR was an independent factor along with age, tumor size, and metastatic nodal disease, suggesting late recurrence may rather be influenced by clinical factor Citation Format: Airi Han, Hyang Suk Choi, Seok Hahn, In-jeoung Cho, Seung Taek Lim, Jong-In Lee. Elevated tumor to aorta ratio of Hounsfield units and late recurrence in patients with estrogen receptor positive 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 PO4-02-05.