New potential biomarkers for breast cancer prognosis

Abstract

Background: Breast cancer (BC) is the most frequent oncological disease among women with terrible mortality. Specific prognostic biomarkers could improve treatment choice and BC patient survival by the accurate assessment of the risk of an adverse outcome. We proposed that alterations of 16 genes namely RAD50, KIF15, KIF2C, PARD3, SAP30BP, SPP1, ANKRD11, FAM50A, LGALS3BP, HMGN2, LRRFIP1, ANKRD30A, ABCA4, PDCL, RBPJ, TOP2B encoding tumour antigens defined by SEREX approach in our previous study could be new potential biomarker of BC. The aim of this study was to evaluate the prognostic power of the copy number alterations (CNA) of above-mentioned genes in tumours of BC patients. Methods: The CNA data of target genes and supported clinical data of BC patients (1098 causes) were downloaded using TCGA data portal (https://portal.gdc.cancer. gov/). Overall survival (OS) was defined from the date of surgery to the date of event (death) or cut-off (120 month). For all variants of CNA (deleted, diploid, gained, amplified copy number) of each gene the univariate analysis of OS was performed by the Kaplan-Meier method and log-rank test, also analysis of OS in relation to clinical outcomes was performed using the Cox proportional hazards regression model. All statistical analyses were carried out by RStudio (Ver. 1.0.153), with P values<0.05 considered significant. Results: The univariate analysis of CNA association with overall survival showed that the patients with low copy number (Deletion) of ABCA4 (N=335, P-value = 0,00125) and LGALS3BP (N=172, P-value =0,0483) genes have worse overall survival probability than patients without them (N=725 and N=908 correspondingly). The investigation of clinicopatological characteristics impact on OS revealed that the risk of poor outcome increase significantly with age (P-value = 1,75-10-6) and tumor stage progression (P-value = 1,303-10-11). Decreasing of ABCA4 and LGALS3BP genes copy number both individually (Likelihood-ratio test P-value =3,331-10-16, P-value = 1,11-10-16 correspondingly) and together (P-value = 4,441-10-16) significantly impact on OS in the Cox regression model after adjustment for clinicopathological characteristics namely age and tumor stage. Conclusions: Strong impact of decreased copy number (deletion) of ABCA4 and LGALS3BP genes on OS have been identified in this study however a number of limitations should be considered despite these significant Results. The mechanisms underlying the association of the copy number of ABCA4 and LGALS3BP genes with clinical outcomes in BC remain to be established. More accurate evidence of the prognostic values of ABCA4 and LGALS3BP genes in BC requires additional multicenter studies using more numerous patient populations and consideration to other clinicopathological features.