Prediction model of the unbalanced expression of hormone receptor in HER2-negative Breast Cancer

Abstract

Abstract Purpose: Estrogen and Progesterone receptors(ER, PR) are essential indicators for Breast cancer(BC) therapy, but the unbalanced expression effect remains unclear. This study aimed to explore the features of HER2-negative BC with different expressions of ER and PR. Methods: 106,742 women patients with BC in SEER from 2010 to 2015 were divided into four groups: ER-negativePR-negative (NN), ER-negativePR-positive (NP), ER-positvePR-negative (PN), ER-positivePR-positive (PP). The significant results of Cox Hazard regression were used to build the nomograms of NP and PN. Results: Groups of ER-negative (median age<60) were younger than ER-positive. Proportions about Grade III-IV of NP and NN were significantly higher than PN and PP (III: 81.41%, 81.71%, 44.26%, 18.63%; IV: 0.73%, 0.61%, 0.29%, 0.10%). Metastasis rates in NN, NP, and PN were higher than in PP (3.45%,3.02%,3.26%, 2.07%). The survival line of NP was equal to NN (P=0.65) but lower than PP (P=0.00011); PN was lower than PP but significantly higher than NN(P<0.0001). C-index of NP and PN prediction models was 0.721 (95%CI: 0.679-0.761) and 0.750 (95%CI: 0.736-0.763). Conclusion: HER2-negative BC with uneven expressions of ER and PR differs from those with consistent presentations. NP and NN have similar clinical features: worse grading, larger tumor diameter, and higher incidence of peripheral invasion and metastasis rates. The survival of NP and PN was worse than PP; NN was consistent with NP but lower than PN from the curve. The two nomograms were validated effectively and had a moderate ability.