Prognostic influence of 3-dimensional tumor volume on breast cancer compared to conventional 1-dimensional tumor size.
10.4174/astr.2018.95.4.183
- Author:
Ki Tae HWANG
1
;
Wonshik HAN
;
Sang Mok LEE
;
Jaewoo CHOI
;
Jongjin KIM
;
Jiyoung RHU
;
Young A KIM
;
Dong Young NOH
Author Information
1. Department of Surgery, Seoul Metropolitan Government - Seoul National University Boramae Medical Center, Seoul, Korea. kiterius@snu.ac.kr
- Publication Type:Original Article
- Keywords:
Breast neoplasms;
Prognosis;
Survival analysis;
Tumor burden
- MeSH:
Breast Neoplasms*;
Breast*;
Classification;
Cohort Studies;
Humans;
Multivariate Analysis;
Prognosis;
ROC Curve;
Survival Analysis;
Tumor Burden*
- From:Annals of Surgical Treatment and Research
2018;95(4):183-191
- CountryRepublic of Korea
- Language:English
-
Abstract:
PURPOSE: The prognostic influence of 3-dimensional tumor volume (Tv) on breast cancer compared to conventional 1-dimensional tumor size (T) was investigated. METHODS: Analysis was performed on a cohort of 8,996 primary breast cancer patients who were initially diagnosed with TNM stage I–III. Tumor size was defined as the maximum tumor dimension, and Tv was calculated by the equation of (4π× r1 × r2 × r3)/3; r1, r2, and r3 were defined as half of the largest, intermediate, and shortest dimension of the tumor, respectively. Tv was classified into Tv1, Tv2, and Tv3 according to the cut off values of 2.056 cm3 and 20.733 cm3. RESULTS: The survival curves according to both the T and Tv categories were clearly differentiated (all P < 0.001), as were those for staging by T and Tv (all P < 0.001). In T1 and T2 tumors, the Tv1 group showed superior survival over the Tv2 group (T1, P < 0.001; T2, P = 0.001). Univariate and multivariate analysis both indicated that Tv was a significant prognostic factor (both P < 0.001). The receiver operating characteristic curve showed that the area under the curves were 0.712 (P < 0.001) for Tv and 0.699 (P < 0.001) for T. Positive correlations were observed between the number of positive nodes and T (coefficient = 0.325; P < 0.001), and between the number of positive nodes and Tv (coefficient = 0.321; P < 0.001). CONCLUSION: Tv classification works well for predicting the prognosis of breast cancer, and it is a better predictor than conventional T classification in several aspects. Further studies are needed to validate the practical usefulness of Tv classification in clinical settings.
