Recent reports have shown that several proteins have exhibited variations in glycosylation patterns that are specific for a certain disease. Most often, protein biomarkers used in clinical assays are studied based solely on quantitative measurements. We suggest that the reliability of certain assays, as well as a new possibility of finding yet other biomarkers, lies in the use of glycosylation pattern analysis of specific serum proteins or in a combination of the two. This has been shown for alpha-fetoprotein in liver cancer. We have earlier presented methods for discriminating between very similar glycosylation patterns of antibodies extracted from serum samples using the target antigen. Now we intend to extend this to the analysis of glycosylation patterns of disease specific proteins in biological samples. We intend to develop a generic methodology based on affinity extraction, enzymatic release of glycans, MALDI-MS analysis and multivariate data evaluation. The technical platform that we intend to use has been developed for performing quantitative MS-based immunoassays which will serve as a mutual base for both the qualitative and quantitative analysis of the protein biomarker. This will be paired with quantitative analysis for four initial candidate proteins and the clinical specificity and sensitivity of these biomarkers determined. The methodology will also be adapted to an automated microfluidic workflow for unattended operation.