Abstract Background: Angiogenesis is known to be related with poor survival. However, previous clinical studies with anti-angiogenetic agents failed to show a clear overall survival benefit and one of the reasons was lack of appropriate biomarker. Therefore, the identification of biomarkers that may identify patients who might benefit from target agents against angiogenesis may be the key to determining which patients would benefit from these therapies. This study aimed to find biomarker which can reflect angiogenesis of the tumor and serve as a prognostic factor. Methods: Patients with stage I-III breast cancer who completed planned treatment were included. Data were retrospectively collected from the Wonju Severance Christian Hospital database of Yonsei University and the Korean National Cancer Center database. Results: 534 patients were enrolled in this study. Patients were divided into two groups according to the cut-off value (31.4) of proportional ratio between maximum Hounsfield unit (HU) of tumor and maximum HU of aortic arch (maximum tumor - aorta ratio, TAR), and analyzed with clinico-pathological factors. Tumor stage was significantly related with higher TAR [TI vs T2-T3, 29.5% vs 50.6%, p < 0.001]. While estrogen receptor positivity was not related with TAR. Human Epiermal growth factor Rceptor-2 showed higher frequency in high TAR [high TAR vs. low TAR, 17.6% vs. 28.1%, p = 0.006]. Kaplan-Meier curve with log rank test showed significantly worse distant recurrence free survival in high TAR group (p < 0.001) Cox proportional hazard model indicated that age, ER negativity, and high TAR were significant risk factors for distant relapse. Conclusion: TAR measured by CT performed before treatment can be used as a prognostic factor for overall and distant relapse-free survival in patients with breast cancer. We believe TAR could be a potential biomarker to select patients who will benefit from target agents based on anti-angiogenesis. Key words: breast neoplasms, computed tomography, Hounsfield unit, survival analysis. Table 1.Baseline characteristics of patientsCharacteristicsNumber (n = 534) (%)Age54.4 ± 11.5HistopathologyInfiltrating ductal carcinoma498 (93.3)Infiltrating lobular carcinoma19 (3.6)Others17 (3.2)T stageT025 (4.7)T1280 (52.4)T2198 (37.1)T331 (5.8)Tumor size (cm)2.4 ± 1.8N stageN0356 (66.5)N1116 (20.6)N249 (9.2)N319 (3.6)Unknown1 (0.2)ER (+) *297 (55.6)PR (+) **274 (51.3)HER 2 (+) ***128 (24.0)ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; * Data were missing in 3 patients ** Data were missing in 2 patients *** Data were missing in 4 patients. Table 2.Relationship between TAR and clinico-histologic factorsClinico-pathologic factorsLower TAR(< 31.4 %) (n = 200)Higher TAR(≥ 31.4) (n = 334)p-valueAge56.2 ± 11.253.4 ± 11.60.007T stage (2 – 3)59 (29.5 %)169 (50.6 %)< 0.001Present lymph node metastasis62 (31.0 %)116 (34.7 %)0.395ER (+) *122 (61.3 %)175 (52.7 %)0.053PR (+) **109 (54.8 %)165 (49.5%)0.243HER – 2 (+) ***35 (17.6 %)93 (28.1 %)0.006TAR, Maximum Hounsfield unit of tumor aorta ratio; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2* Data were missing in 3 patients ** Data were missing in 2 patients *** Data were missing in 4 patients Citation Format: Kwangmin Kim, In-Jeong Cho, Hany Noh, Hyangsook Choi, Seok Hahn, Seung Taek Lim, Jong-In Lee, Airi Han. Prognostic significance of Hounsfield unit of tumor in patients with breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-08-19.