A nomogram to predict HER2 status in breast cancer patients with HER2-borderline disease as determined via immunohistochemistry

// Qianqian Guo 1 , Kai Chen 2, 3 , Xiaojie Lin 1 , Yi Su 4 , Rui Xu 1 , Yan Dai 1 , Chang Qiu 1 , Xue Song 1 , Siying Mao 1 and Qianjun Chen 1 1 Department of Mammary Disease, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, Guangdong, P.R. China 2 Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, Guangdong, P.R. China 3 Breast Tumor Center, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, Guangdong, P.R. China 4 Department of Intensive Care, Foshan Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, P.R. China Correspondence to: Qianjun Chen, email: cqj55@163.com Kai Chen, email: chenkai23@mail.sysu.edu.cn Keywords: breast cancer, HER2 status, IHC, calibration, nomogram Received: October 27, 2016      Accepted: March 11, 2017      Published: July 17, 2017 ABSTRACT This study aimed to develop a nomogram to predict fluorescence in situ hybridization (FISH) assay results for HER2-borderline breast cancer as determined via immunohistochemistry (IHC) among patients in China. We reviewed a database of breast cancer patients diagnosed between January 2007 and April 2013 at our institutions. We used logistic regression to develop a nomogram and we used receiver operating characteristic curve analysis and calibration plots to validate our nomogram. In total, 1138, 301 and 344 patients had IHC-determined HER2-negative, HER2-borderline and HER2-positive disease, respectively. Within the training cohort, univariate and multivariate analyses suggested that estrogen receptor (ER) status, progesterone receptor (PR) status and tumor grade were significantly associated with HER2 status (P<0.01). A nomogram was developed and the AUCs for the training and validation cohorts were 0.795 and 0.749, respectively. The calibration plots suggested that the model was well calibrated. This new nomogram can be used to predict HER2 status in HER2-borderline breast cancer patients and will be particularly helpful to resource-limited countries.