Collagen-type-II-induced arthritis (CIA) is an autoimmune disease, which involves a complex host systemic response including inflammatory and autoimmune reactions. CIA in C57BL/6 mice resembles human rheumatoid arthritis (RA) in terms of its disease course, histological findings, and in its response to commonly used anti-arthritic drugs. In this study, the goal has been to identify proteins from serum that change in their abundance in CD38-KO versus WT mice which may reflect their distinct response to an antigen-challenge that induces the development of an autoimmune disease.CIA is milder in CD38-/- than in Wild-type (WT) mice. We analyzed the sera from CD38-/- versus WT mice either with arthritis (CIA+), with no arthritis (CIA-), or with inflammation (Complete Freund’s adjuvant (CFA)-treated mice). To decrease the dynamic concentration range of serum a combinatorial ligand library composed of hexapeptides was used (called ProteoMiner). ProteoMiner-equalized serum samples were then subjected to 2D-DiGE and MS-MALDI-TOF/TOF analyses to identify proteins that changed in their relative abundances. Multivariate analyses revealed that a multi-protein signature (n = 28) was able to discriminate CIA+ from CIA- mice, and WT from CD38-/- mice within each condition. Likewise, a distinct multi-protein signature (n = 16) was identified which differentiated CIA+CD38-/- mice from CIA+ WT mice, and lastly, a third multi-protein signature (n = 18) indicated that CD38-/- and WT mice could be segregated in response to CFA treatment This approach allows the identification of multiple protein species, or proteoforms of a given protein in a single analysis, and therefore, to focus the interest in fully characterize just the protein species that differ in abundance.