Open Access Highly Accessed Open Badges Research article

Preoperative core needle biopsy is accurate in determining molecular subtypes in invasive breast cancer

Xiaosong Chen1, Long Sun1, Yan Mao1, Siji Zhu1, Jiayi Wu1, Ou Huang1, Yafen Li1, Weiguo Chen1, Jianhua Wang2, Ying Yuan3, Xiaochun Fei4, Xiaolong Jin4 and Kunwei Shen1*

Author Affiliations

1 Comprehensive Breast Health Center, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197 Ruijin Second Road, Shanghai 200025, China

2 Department of Biochemistry and Molecular & Cell Biology, Shanghai Jiaotong University School of Medicine, 227 Chongqing Nan Road, Shanghai 200025, China

3 Department of Radiology, Shanghai Ninth People’s Hospital, Affiliated to Jiaotong University School of Medicine, 639 Zhizhaoju Road, Shanghai 200011, China

4 Department of Pathology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197 Ruijin Second Road, Shanghai 200025, China

For all author emails, please log on.

BMC Cancer 2013, 13:390  doi:10.1186/1471-2407-13-390

Published: 19 August 2013



Estrogen receptor (ER), progesterone receptor (PgR), HER2, and Ki67 have been increasingly evaluated by core needle biopsy (CNB) and are recommended for classifying breast cancer into molecular subtypes. However, the concordance rate between CNB and open excision biopsy (OEB) has not been well documented.


Patients with paired CNB and OEB samples from Oct. 2009 to Feb. 2012 in Ruijin Hospital were included. ER, PgR, HER2, and Ki67 were determined by immunohistochemistry (IHC). Patients with HER2 IHC 2+ were further examined by FISH. Cutoff value for Ki67 high expression was 14%. Molecular subtypes were constructed as follows: Luminal A, Luminal B, Triple Negative, and HER2 positive.


There were 298 invasive breast cancer patients analyzed. Concordance rates for ER, PgR, and HER2 were 93.6%, 85.9%, and 96.3%, respectively. Ki67 expression was slightly higher in OEB than in CNB samples (29.3% vs. 26.8%, P = 0.046). Good agreement (κ = 0.658) was demonstrated in evaluating molecular subtypes between CNB and OEB, with a concordance rate of 77.2%. We also used a different Ki67 cutoff value (20%) for determining Luminal A and B subtypes in HR (hormone receptor) +/HER2- diseases and the overall concordance rate was 79.2%. However, using a cut-point of Ki67 either 14% or 20% for both specimens, there will be about 14% of HR+/HER2- specimens that are called Luminal A on CNB and Luminal B on OEB.


CNB was accurate in determining ER, PgR, and HER2 status as well as non-Luminal molecular subtypes in invasive breast cancer. Ki67 should be retested on OEB samples in HR+/HER2- patients to accurately distinguish Luminal A from B tumors.

Breast cancer; Core needle biopsy; Molecular subtype; Ki67; Concordance rate