An external quality assurance trial to assess mass spectrometry protein testing facilities for identifying multiple human peptides.

The application of proteomic liquid chromatography mass spectrometry (LC-MS) for identifying proteins and peptides associated with human disease is rapidly growing in clinical diagnostics. However, the ability to accurately and consistently detect disease-associated peptides remains clinically uncertain. Variability in diagnostic testing occurs in part due to the absence of appropriate reference testing materials and standardised clinical guidelines for proteomic testing. In addition, multiple proteomic testing pipelines have not been fully assessed through external quality assurance (EQA). This trial was therefore devised to evaluate the performance of a small number of mass spectrometry (MS) testing facilities to (i) evaluate the EQA material for potential usage in a proteomic quality assurance program, and to (ii) identify key problem areas associated with human peptide testing. Five laboratories were sent six peptide reference testing samples formulated to contain a total of 35 peptides in differing ratios of light (natural) to heavy (labelled) peptides. Proficiency assessment of laboratory data used a modified approach to similarity and dissimilarity testing that was based on Bray-Curtis and Sorensen indices. Proficiency EQA concordant consensus values could not be derived from the assessed data since none of the laboratories correctly identified all reference testing peptides in all samples. However, the produced data may be reflective of specific inter-laboratory differences for detecting multiple peptides since no two testing pipelines used were the same for any laboratory. In addition, laboratory feedback indicated that peptide filtering of the reference material was a common key problem area prior to analysis. These data highlight the importance of an EQA programme for identifying underlying testing issues so that improvements can be made and confidence for clinical diagnostic analysis can be attained.