Expression of Soluble Form of Aurora A as a Predictive Factor for Neoadjuvant Therapy in Breast Cancer Patients: A Single-Center Pilot Study

SIMPLE SUMMARY: AURKA tissue expression has been shown to be a predictive and prognostic factor in many solid tumors, including breast cancer. To improve the outcome of breast cancer (BC) treatment, further personalization and the exploration of the use of new biomarkers is needed to assess the effects of the applied treatment. New predictive markers may reduce the cost and toxicity of therapy. In our single-center prospective study, we determined the serum levels of Aurora A, thymidine kinase 1, and human epidermal growth factor receptor type 3 (HER3) using ELISA kits in 119 women with BC before neoadjuvant treatment (NAT). We showed that in a biologically heterogeneous group of BC patients, the pretreatment serum Aurora A levels were of significant value in predicting the response to NAT. The response to NAT was assessed in the postoperative material after systemic treatment as the pCR, that is, an indicator of the complete pathological response to neoadjuvant treatment. To our knowledge, this is the first study of the predictive value of serum AURKA in patients with breast cancer. ABSTRACT: Purpose: To search for new predictive breast cancer biomarkers. We analyzed the serum concentrations of biomarkers involved in carcinogenesis, which can also be targeted by therapy. Methods: In a single-center prospective study, the serum levels of Aurora A, thymidine kinase 1, and human epidermal growth factor receptor type 3 (HER3) were determined in 119 women with BC before neoadjuvant treatment using ELISA kits. Results: The following clinical data were analyzed: age; TNM; the expression of ER, PGR, HER2, and Ki67; histological grade (G); and the response to neoadjuvant treatment (NAT) in the residual tumor burden classification (RCB). A complete pathological response (pCR) was achieved after NAT in 41 patients (34%). The highest proportion of the patients with a confirmed pCR was found for triple negative breast cancer (TNBC) (62.5%); non-luminal HER2-positive (52.6%) cancer subtypes (p = 0.0003); and in the G3 group (50%; p = 0.0078). The patients with higher levels of Aurora A were more likely to achieve pCR (p = 0.039). In the multivariate analysis, the serum Aurora A levels ≥ 4.75 ng/mL correlated with a higher rate of pCR (OR: 3.5; 95% CI: 1.2–10.1; p = 0.023). Conclusions: We showed that in a biologically heterogeneous group of BC patients, the pretreatment serum Aurora A levels were of significant value in predicting the response to NAT.