Proteomic profiling with protein-chip technology has been used successfully to discover biomarkers with potential clinical usefulness in several cancer types. Little proteomic study has been done in B-cell lymphomas. We determined whether the expression of a set of proteins by protein-chip technology coupled with new informatics tools could be used to build a model to molecularly classify B-cell lymphoma subgroups. We used surface-enhanced laser desorption/ionization time-of-flight mass spectrometry to analyze 18 CD10+ B-cell lymphomas, including 6 grade 1 (G1) follicular lymphomas (FLs), 7 grade 3 (G3) FLs, and 5 Burkitt lymphomas. We used 7 reactive follicular hyperplasia cases as a control group. By using SAX2 ProteinChip arrays (Ciphergen Biosystems, Fremont, CA), we found a unique protein expression profile for each type of lesion. Two-way hierarchical clustering analysis of these protein expression profiles differentiated reactive follicular hyperplasia, FL, and Burkitt lymphoma, with 5 major clusters of differentially expressed protein peaks. In addition, we identified histone H4 as a potential differentially expressed protein marker that seems to distinguish G1 from G3 FL. To our knowledge, this is the first proteomic study using protein-chip technology for molecular classification of B-cell lymphoma subtypes with clinical samples.