A potential serum panel for breast cancer diagnosis: ongoing clinical validation for its clinical implication.

Abstract

Abstract Abstract #5006 Although every effort has been made to discover single potent biomarkers in oncology, the promising single biomarker often loses its clinical potency in the course of clinical validations. We aimed to construct and verify a biomarker panel with multiple proteins for breast cancer diagnosis. Using Surface-Enhanced Laser Desorption/ Ionization Time-of-Flight mass spectrometry (SELDI-TOF-MS) and microbead array system, epidermal growth factor (EGF), vascular cell adhesion molecule-1 (VCAM-1), CD40L, vitronectin, plasminogen activation inhibitor-1 (PAI-1), vitamin-D binding protein (VDBP), C-reactive protein (CRP), kininogen, apolipoprotein A1 and proapolipoprotein A1 were identified and validated as useful biomarkers for detection of breast cancer. The multiplexing panel (MP) which was constructed by combining algorithm (random forest, support vector machine) with above biomarkers showed that the diagnostic accuracy approached 91% in 216 test samples. A following double blind test verified that the diagnostic accuracy of the MP was 68% with an additional validation set including sera from 49 patients with breast cancer and 51 healthy women (sensitivity=74%; 95% confidence interval (CI)=58.9–85.0%, specificity=62.7%; 95% CI=48.1–75.9%, positive predictive value=65.5%; 95% CI=55.7-75.3%, negative predictive value=71.7%; 95% CI=61.7-80.35%). We suggest that these results support the potential use of the MP as an early detector of breast cancer. In order to validate the role and potency of the MP in routine clinical practice, a large scale clinical validation in a cohort consisting of 500 patients with breast cancer and the same number of healthy women are underway and the results will be released in the near future. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 5006.