Correlation between numerical and categorical immunohistochemical score of Ki-67 and HER2 with clinicopathological parameters of breast cancer

Abstract

<p><strong>Aim</strong> <br />To evaluate the relationship between numerical and categorical immunohistochemical score of Ki-67 and human epidermal growth factor of receptor 2 (HER2) with clinicopathological parameters of breast cancer (BC).<br /><strong>Methods</strong> <br />The study included 311 patients with invasive BC diagnosed at the Department of Pathology, School of Medicine in<br />Sarajevo, Bosnia and Herzegovina, during the period 2015-2019. The expression level of Ki-67 and HER2 was detected by immunohistochemical analysis.<br /><strong>Results</strong> <br />The expression of Ki-67, as a numerical variable correlated significantly with tumour grade (p=0.025), progesterone receptor (PR) (p=0.034) and categorical score of HER2 (p=0.028). When Ki-67 was categorized into high (>14%) and low (≤14%) level groups, a statistically significant association was found between Ki-67 level groups and HER2 status (categorical and numerical) (p=0.001 and p=0.043, respectively), as well as significant negative linear correlation with PR (p=0.037). The expression of HER2, as a numerical variable, showed a statistically significant correlation with tumour grade (p=0.038), PR (p=0.025) and categorical Ki-67 (p=0.043). Categorical score of HER2 correlated significantly with age (p=0.025), histologic type (p=0.039), tumour grade (p=0.016), estrogen receptor (ER), (p=0.002) progesterone<br />receptor (PR) (p=0.0001), and categorical and numerical value of Ki-67 (p=0.0001 and p=0.0001, respectively).<br /><strong>Conclusion</strong> <br />The results demonstrated that the categorical immunohistochemical score of HER2 provided a greater association<br />with clinicopathological parameters than numerical score of BC. Furthermore, a slightly better correlation with clinicopathological parameters was shown by the numerical value than by the categorical score of Ki-67 by applying a cut-off value of 14%</p>