Combined measurement of CA 15-3 with novel autoantibodies improves diagnostic accuracy for breast cancer

OBJECTIVE CA 15-3 is a traditional biomarker for advanced breast cancer with limited sensitivity for early stage patients. In order to increase the sensitivity for early detection, in this study, we introduced novel tumor-associated autoantibodies that were measured concurrently with serum CA 15-3 to evaluate their diagnostic advantage in breast cancer. METHODS We investigated a T7 breast cancer complementary deoxyribonucleic acid (cDNA) phage library for tumor-associated antigens using sera from normal and breast cancer patients. Identified novel tumor-associated antigens phage proteins were then used to develop enzyme-linked immunosorbent assays to measure corresponding autoantibodies in 150 breast cancer, 150 normal, and 40 other cancer (non breast) patient serum samples. Meanwhile, the same samples were measured for CA 15-3 concentrations. Receiver operating characteristic curve analysis was used to evaluate the predictive accuracies of single markers as well as combined markers. RESULTS Sequencing analysis revealed that two phage-expressed proteins were within the open reading frame and had significant homology to proteins heterogeneous nuclear ribonucleoproteins F (hnRNPF) and ferritin heavy chain (FTH1). Autoantibodies against hnRNPF and FTH1 alone were significantly higher in patients than in control serum samples (P < 0.01), and the area under the curve for hnRNPF and FTH1 alone was 0.73 and 0.69, respectively. However, when the two autoantibody biomarkers were analyzed in combination with serum CA 15-3 values, the area under the curve increased to 0.93, and the optimal sensitivity and specificity became 89.3% and 93.8%, respectively. Further messenger ribonucleic acid (mRNA) analysis showed that hnRNPF and FTH1 were significantly upregulated in tumor tissues. CONCLUSION Our results indicated that combined serologic biomarkers of tumor-associated antigens with autoantibodies may improve the diagnostic accuracy of breast cancer.