Cardiotoxicity related to doxorubicin chemotherapy is a major late effect in childhood cancer survivors. Serum cardiac troponin concentrations (cTnT) can be elevated during doxorubicin therapy but the cellular associations with this myocardial injury are not well understood. We evaluate a novel nanotechnology-based biomarker discovery approach on a pilot set of serial serum samples from 11 children with acute lymphoblastic leukemia receiving doxorubicin therapy to determine if a proteomic signature of myocardial injury could be identified. This nanoparticle-based biomarker capture technology was utilized to analyze 40 serial serum samples from these children, 3 of whom seroconverted, 2 from cTnT-negative to cTnT-positive and 1 from cTnT-positive to cTnT-negative. High-resolution mass spectrometry analysis of the captured material identified 13 differentially expressed candidate proteins, whose spectral count values reflected changes in cTnT concentrations, which were verified in the serum samples from the 3 consistently cTnT-negative and 5 consistently cTnT-positive children. Of the 13 candidate proteins, 5 were significantly elevated (p < 0.1) in the independent validation set of cTnT-positive samples (serum amyloid A, cardiac muscle actin proprotein, a gamma globulin, HIV-enhancer-element binding protein, and C-reactive protein). These results demonstrate the potential for novel nanoparticle-capture biomarker discovery workflow to be applied to the doxorubicin cardiotoxicity-based setting. The identified candidate biomarkers require further validation in larger cohorts to evaluate clinical impact. Significance Identification of new biomarkers for early detection of chemotherapy-induced cardiotoxicity is of critical importance so that administration of cardioprotectants can be utilized before substantial heart damage has occurred. We developed and utilized a unique biomarker workflow based on a novel nanotechnology method in order to demonstrate the potential for such an approach to uncover low abundance markers that could be useful in a clinical setting once extensively validated.