Abstract P2-08-13: The prognostic significance of Ki67 and progesterone receptor; could they really have prognostic importance in early breast cancer with estrogen receptor positive, HER-2 negative?

Abstract

Abstract Background: Ki67 has been studied a lot as a prognostic factor, but ambiguity exists as a value of a prognostic factor that conforms to the guideline due to analytical and methodological issues. However, Ki67 is still represented as IHC representing proliferation, and it is evaluated as an important prognostic factor in many studies. Recently, the international Ki67 in breast cancer working group (IKWG) suggested that in early breast cancer with Estrogen receptor positive (ER+) and HER2 negative (HER2-), Ki67 5% or less could predict a good prognosis and 30% or more could predict a bad prognosis. However, the reality and limitation is that it is still difficult to predict the prognosis at Ki67 5% to 30%. The aim of our study intends to identify that if Progestrone receptor (PR) information is added to Ki67, which has such an ambiguous section, it can give additional prognostic information. Methods: We selected 2668 patients with early breast cancer who underwent surgery and were pathologically diagnosed with ER+ and HER2-. We divided these patients into three groups by Ki67 <5%, 5%≤Ki67≤30%, and Ki67>30%. PR was divided into low and high based on 20%. In each of the three groups, multivariate analysis was performed to identify whether PR is an important factor in prognosis. We set the Ki67 cutoff to 15% in the 5%≤Ki67≤30% group. In this group, we made three combinations of Ki67 and PR with Low group (Ki-67<15%/PR≥20%), Intermediate group (Ki-67≥15%/PR<20%, and Ki-67<15%/PR≥20%), and High group (Ki-67≥ 15%/PR<20%). Multivariate analysis was performed to identify an association of Ki-67/PR combinations with other prognostic variables. Results: Among total patients, 747 patients had Ki67 <5%, 1713 patients had 5% ≤Ki67≤30%, and 208 patients had Ki67 >30%. Multivariate analysis was performed in total patients, and hazard ratio(HR) of Ki67 <5%, 5% to 30%, and >30% was 1, 2.978, and 6.203, respectively, and also it was statistically significant (p=0.004). In the clinical risk (combination of T size, N status, and Histologic grade), HR of clinical high was significantly higher than clinical low (HR: Clinical low : Clinical high = 1: 2.234, p<0.001). Also, PR ≤20% was high HR compared with PR>20%, which was statistically significant (HR: PR >20 : PR ≤20% = 1: 1.550, p=0.016). Multivariate analysis was performed for each of the three groups. In the Ki67 <5% and Ki67 >30% groups, PR had no statistical significance, but in 5%≤ Ki67 ≤30%, PR was statistically significant (HR: PR >20% vs PR ≤20% = 1 vs 1.613, p= 0.029). In the multivariate analysis of 5%≤ Ki67 ≤30% performed with the Ki67/PR combination, HR of Low, intermediated, and high was 1, 2.139, and 2,054, respectively, and also it was statistically significant (p=0.003, and 0.030). Conclusion: Ki67 can be used as a single prognostic value in at least <5% and >30% of early breast cancer with ER+ and HER2-. In the case of 5% ≤Ki67≤ 30%, additional prognostic prediction is possible with the combination of Ki67 and PR. Keywords: Ki67, Progesterone receptor, Prognosis, Estrogen receptor positive, HER2 negative, Early breast cancer Citation Format: Soon Bo Choi, Jung Min Park, Jee Hyun Ahn, Jieon Go, Jeeye Kim, Hyung Seok Park, Seung Il Kim, Byeong-Woo Park, Seho Park. The prognostic significance of Ki67 and progesterone receptor; could they really have prognostic importance in early breast cancer with estrogen receptor positive, HER-2 negative? [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-08-13.