Recent experimental evidence demonstrates that the platelet-specific collagen receptor, glycoprotein (GP)VI is essentially all uncleaved on normal circulating platelets, but is shed from the platelet surface in a metalloproteinase-dependent manner in response to GPVI ligands (including collagen), anti-GPVI antibodies or activation at the platelet Fc receptor, FcgammaRIIa. This raises the question of whether shed ectodomain fragment in plasma could be a useful biomarker of thrombotic risk and/or autoimmune thrombocytopenia. In this study, we developed a sandwich enzyme-linked immunosorbent assay (ELISA) for measuring soluble GPVI in human plasma, using rabbit anti-GPVI polyclonal antibody in the solid-phase, murine anti-GPVI monoclonal antibody (1A12) in the fluid-phase and horseradish peroxidase (HRP)-coupled anti-mouse antibody and enhanced chemiluminescence (ECL) for detection. The ELISA was optimized for sensitivity, reproducibility, inter- and intra-assay precision, addition and recovery and detected GPVI in plasma with a lower detection limit of approximately 1 ng/mL. Effects of different anti-coagulants (trisodium citrate, acid-citrate-dextrose or EDTA) were negligible. In ten healthy donors, soluble plasma GPVI levels were 18.9 +/- 4.1 ng/mL. Treating normal platelet-rich plasma with a GPVI ligand (collagen-related peptide, CRP), calmodulin inhibitor W7 (that induces GPVI shedding without platelet activation) or N-ethylmaleimide (that directly activates platelet sheddases), under conditions previously shown to induce GPVI shedding, also increased plasma GPVI levels by up to approximately 7-fold, compared to previously reported autoimmune (anti-GPVI) patient plasma where soluble GPVI was approximately 10-fold higher than normal. Characterization of this sensitive ELISA should facilitate analysis of functional/diagnostic role(s) for soluble GPVI in human plasma associated with thrombotic/immune dysfunction.