Enhancement of ovarian tumor classification by improved reproducibility in matrix-assisted laser desorption/ionization time-of-flight mass spectrometry of serum glycans

Abstract

The serum N-glycome is a promising source of biomarker discovery. Matrix-assisted laser desorption/ionization time-of-flight (MALDI–TOF) mass spectrometry (MS) profiling of serum N-glycans was attempted for differentiating borderline ovarian tumor from benign cases, for which a low data spread is essential. An experimental protocol using matrix-prespotted MALDI plates and fast vacuum drying of the loaded N-glycan samples was developed, thereby minimizing the intensity variations in the replicates to an average relative standard deviation (RSD) of 3.96% for the highest N-glycan peak (m/z 1485.53) of the Sigma–Aldrich serum standard. When applied to sera of ovarian tumors, this procedure exhibited an average RSD of 5.74% for m/z 1485.53 and of 7.28% for all MS peaks. This improved reproducibility combined with the OVA-Beyond® screening software resulted in 75.1% and 79.4% correct classification for benign and borderline tumor samples, respectively, while the classification rates by the conventional ovarian tumor marker CA-125 were 54.4% and 53.1%, respectively. Both true positive rate and true negative rate fluctuated with small numbers of markers and converged as the number of markers increased. Cross-validations were performed in comparison with CA-125. These results suggest that our optimized process for MALDI–TOF MS of the serum glycome has a great potential for the screening of early stage ovarian cancer.