2021 Volume 5 Pages 72-83
Background: We evaluated 12 biomarkers as potential prognostic predictors in Japanese breast cancer patients: EGFR, PTEN, IGF-1R, Bcl-2, c-Kit, c-Met, HIF-1α, PDGFRA, survivin, VEGFR2, VEGF-A, and Ki67.
Methods: A total of 466 surgical patients were divided into six subgroups in accordance with St. Gallen 2017 classification: triple-negative, HER2, triple-positive, luminal-low risk, luminal-intermediate risk, and luminal-high risk. Immunohistochemical staining of the biomarkers was compared among the subgroups in relation to 10-year disease-free survival (DFS) and cancer-specific survival (CSS), calculated by the Kaplan-Meier method and evaluated by Cox regression.
Results: In the triple-negative group, by multivariate analysis of Cox model for DFS, c-Kit (hazard ratio [HR], 5.941; 95% confidence interval [CI], 1.171-30.137; p = 0.0315) and HIF-1α (HR, 5.631; 95%CI, 1.014-31.275; p = 0.0482) emerged as significant. In the CSS model, c-Kit (HR, 8.541; 95%CI, 1.396-52.266; p = 0.0203) and HIF-1α (HR, 11.103; 95%CI, 1.227-100.429; p = 0.0322) were significant. In other subgroups, several biomarkers and clinical factors remained significant with DFS model but none with CSS.
Conclusions: We found c-Kit and HIF-1α with a statistical effect on outcomes in the triple-negative group. A combination of these can be useful in predicting prognoses.
