Characterization of post-translational modifications (PTMs) of therapeutic proteins is very important during the bioprocess development to maintain desired product quality and during the submission process to regulatory authorities for product approval. Monitoring glycosylation in pharmacokinetic studies can be useful to evaluate the dependence of clearance rates on different glycoforms. The cost and efficiency of characterization affect the speed to market of biopharmaceutical proteins. A reduction in the number of manual processing steps, cost of reagents and consumption of sample, as well as the time required for chemical analysis, is therefore necessary.

The research presented in this thesis is focused on the potential of using microfluidic discs for automated, miniaturized, parallel and rapid sample preparation for PTM characterization of therapeutic monoclonal antibodies. Paper I describes the method development for N-linked glycosylation profiling. Several sample preparation steps have been performed in an integrated process in the microfluidic compact disc (CD). Paper II demonstrates the use of the method presented in paper I in combination with multivariate statistics for discrimination of glycosylation profiles of different therapeutic antibodies and simulation of a real case of quality control. Paper III is focused on a method for monitoring changes in glycosylation profiles of therapeutic antibodies in serum over time by incubation with an exoglycosidase enzyme. Paper IV describes the method for peptide mapping of therapeutic antibodies. In addition, recent work (unpublished results) assesses the potential of this method for methionine oxidation detection.

The developed methods were fast, robust with low sample/reagent consumption. Generation of glycosylation profile data for one sample was established in approximately 2 h. The amount of samples and antigens loaded into the CD platform for one replicate was less than 0.3 μg and approximately 0.06 μg, respectively. Furthermore, considering the parallel function of the CD, conducting the analysis for 54 samples can be completed within a day.

Data processing tools are valuable to the analytical chemist as they can speed up the analysis, and sometimes solve problems that are not feasible to solve in a traditional manner. However, the complexity of many data processing tools can make their use daunting for the inexperienced user.

This thesis includes two applications and two tools for data processing. The first application focuses on minimizing the manual input, reducing the time required for a simple task. The second application required more manual input, in the form of parameter selection, but process far more data.  The data processing tools both include features that simplify the manual work required. The first by including visual diagnostics tools that helps in setting the parameters. The second via internal validation that makes the tool’s process more robust and reliable, and thereby less sensitive to small changes in the parameters.

No matter how good or precise a data processing tool is, if it is so cumbersome that it is not used by the analytical chemists that need it, it is useless. Therefore, the main focus of this thesis is to make data processing easier.