Original Article
Correlation of nuclear morphometry with clinicopathological features in breast cancer
Wu Boda, Long Meijun, Yang Mei
Published 2020-09-20
Cite as J Chin Physician, 2020, 22(9): 1316-1321. DOI: 10.3760/cma.j.cn431274-20200723-01023
Abstract
ObjectiveUnder the new development of breast cancer diagnosis and treatment technology, the relationship between nuclear morphology and different clinicopathological characteristics of breast cancer was analyzed, and its clinical significance was evaluated.
MethodsIn TCGA (the Cancer Genome Atlas) BRCA (fast invasive carcinoma) data, 443 patients were divided into three groups according to the nuclear morphology: ⑴ the nuclear size and morphology were basically normal; ⑵ the nuclear size slightly increased; ⑶ the nuclear size slightly increased or the difference was obvious. Practical clinicopathological features including American Joint Committee on Cancer (AJCC) tumor stage, AJCC tumor score, AJCC lymph node score, estrogen receptor (ER) status by immunohistochemistry (IHC), progesterone receptor (PR) status by IHC, human epidermal growth factor receptor 2 (HER2) status by IHC and PAM50 subtypes are compared among groups defined by nuclear size. And nuclear size related morphometric parameters, including nuclear area, perimeter, roundness, width and height, are subsequently retrieved in 132 samples and analyzed for their relationship with clinicopathological features.
ResultsThe nuclear size in breast cancer tissues was significantly correlated with survival, especially disease-free survival (P=0.039). Other clinicopathological parameters, except AJCC T stage (P=0.006), the immunohistochemical results of molecular markers ER (P=0.002), PR (P=0.047) and molecular typing of PAM50 (P<0.001) were not significantly associated with the nuclear size of cancer tissues. The area, perimeter, roundness, length and width were correlated with each other; the roundness was the most stable parameter and negatively correlated with other parameters; the perimeter was the most sensitive index for identification, while roundness was not.
ConclusionsOur research indicates that nuclear size, especially nuclear morphometric parameter, perimeter, provides a valuable clinicopathological index, which is useful not only in differentiating breast cancer cells from normal cells, but in differentiating molecular subtypes.
Key words:
Breast neoplasms; Cell nucleus size; Cell nucleus shape; Molecular signature
Contributor Information
Wu Boda
Department of Medical Genetics, the Second Xiangya Hospital, Central South University, Changsha 410011, China
Long Meijun
Breast Cancer Center and Department of Thyroid and Breast Surgery, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou 510630, China
Yang Mei
Hunan Key Laboratory of Tumor Models and Individualized Medicine, the Second Xiangya Hospital, Central South University, Changsha 410011, China
Department of Orthopedics, the Second Xiangya Hospital, Central South University, Changsha 410011, China
