The efficacy of IVIG in preventing GvHD has not been definitely demonstrated clinically. Using a xenogeneic model of GvHD in NOD/SCID/ccÀ (NSG) mice, we showed that weekly administration of IVIG significantly reduced the incidence and associated mortality of GvHD to a degree similar to CsA. Unlike CsA and OKT3, IVIG were not associated with inhibition of human T-cell proliferation in mice. Instead, IVIG significantly inhibited the secretion of human IL-17, IL-2, IFN-c and IL-15 suggesting that IVIG prevented GvHD by immunomodulation. Furthermore, the pattern of modification of the human cytokine storm differed from that observed with CsA and OKT3. Finally, in a humanized mouse model of immune reconstitution, in which NSG mice were engrafted with human CD34 þ stem cells, IVIG transiently inhibited B-cell reconstitution, whereas peripheral T-cell reconstitution and thymopoiesis were unaffected. Together these in vivo data raise debate related to the appropriateness of IVIG in GvHD prophylaxis. In addition, this model provides an opportunity to further elucidate the precise mechanism(s) by which IVIG inhibit GvHD.