Comprehensive analysis of the association of clinically relevant values of Ki-67 labeling index with clinicopathologic and immunohistochemical criteria in female invasive breast carcinoma

Abstract

Saba El-Gendi, Iman Talaat, Ghada Abu-Sheasha

Objective: Breast cancer aggressiveness is related to tumor cell proliferation. Despite this, the Ki-67 index is not recommended for routine use in newly diagnosed breast carcinomas. Material and Methods: A total of 164 invasive breast carcinomas were stratified into the intrinsic molecular subtypes based on estrogen receptor, progesterone receptor (PR), HER2, and Ki-67 immunostaining. We studied the distribution of Ki-67 among the molecular subtypes and correlated it with clinicopathologic parameters. Furthermore, the change in the Ki-67 index with tumor size, grade and lymph node (LN) status among the molecular subtypes was examined. Results: As a continuous variable, the median Ki-67 did not show significant differences with the clinicopathological variables. At a cutoff ≥ 14%, it correlated significantly with the mitotic index. At a cutoff ≥ 20%, it additionally correlated with the PR status. The median Ki-67 level varied significantly between luminal A and all other molecular subtypes. The median Ki-67 level in T1/T2 tumors compared to T3/T4 tumors was slightly higher in luminal B HER2+, slightly lower in HER2 enriched, and nearly similar among luminal A, triple negative and luminal B HER2-subtypes, yet without statistical significance. The median Ki-67 was lower in G1/G2 compared to G3 tumors in all-except luminal B HER2-positive subtype but without statistical significance. The Ki-67 distribution change between N0/N1 and N2/N3 cases among the molecular subtypes was significant. Conclusions: The impact of Ki-67 as a proliferation marker on the biological behavior of breast carcinomas is context dependent, and its clinical utility increases when interpreted in combination with other prognostic markers in the context of the molecular subtypes. Further studies, on larger sample sizes are recommended to unravel how the molecular types can affect the relation between Ki-67 and clinicopathological characteristics, particularly the LN status.