The advent of multiplexing technologies has raised the possibility that disease states can be defined using discrete genomic and proteomic patterns or signatures. However, this emerging area has been limited by the 'content problem', arising from the uncertainty of which molecules to focus on. The human cluster of differentiation (CD) antigens are expressed on cells of the human immune system (leukocytes) and on other cell types. These heterogeneous molecules perform a host of roles essential to immune function and to the physiology of other lineages. The 339 defined CD antigens and their, as yet, undefined counterparts constitute key components of the expressed human cell surface proteome. We propose that CD antigen expression patterns will form the basis of a rational, discrete and generalized diagnostic and prognostic system. Furthermore, disease-specific CD antigen proteomic signatures are likely to be more robust than corresponding genomic signatures and will also help to identify molecular targets for therapeutic intervention.