Studies in murine models of allogeneic hematopoetic stem cell transplantation (HSCT) and in humans have brought evidence that Y-chromosome–encoded gene products can elicit a full immune response and act as minor histocompatibility antigens. Nearly 5 decades after the first observation of graft rejection in inbred murine female recipients of male skin grafts or of increased graft-versus-host disease (GVHD) in murine male recipients of female bone marrow,1 and after many controversies, a better understanding of immune phenomena associated with sex-mismatched HSCT has now been reached. A comprehensive study by the Leiden group2 showed this year that H-Y–based rejection was associated with full immune response. In a female patient with aplastic anemia who rejected her brother's stem cell transplant, they identified specific HLA-restricted T-cell helper cells and cytotoxic T cells coexisting with Y-antigen–specific antobodies.2 The clinical observation of increased GVHD in male recipients of female stem cells3 has found its counterpart in rejection.
In this issue, 2 papers extend these observations. Miklos and colleagues (page ) looked for antibodies against the well-defined Y-chromosome–encoded dead box RNA helicase Y (DBY) protein by Western blot analysis and enzyme-linked immunosorbent assay (ELISA) techniques. They found antibodies against DBY in 50% of male patients who received stem cell grafts from female donors. The same antibodies were detected in about 15% to 20% of healthy females, but were not found in healthy males. Antibody responses were primarily directed against areas of amino acid disparity between DBY and its female homologous counterpart DBX. Clearly, the Y-linked DBY protein can trigger a B-cell–mediated immune response. Randolph and colleagues (page ) analyzed 3238 individuals who received HLA-identical sibling HSC transplants for various hematologic malignancies and looked at incidence of GVHD and relapse. GVHD was, as expected, higher in male recipients of female grafts. The same recipients did exhibit a lower hazard of relapse compared with other sex combinations. This graft-versus-leukemia effect (GVL) was preserved even after controlling for GVHD as a time-dependent covariate. Hence, Y-antigen–encoded gene products can act as minor histocompatibility antigens and trigger rejection, GVHD, and GVL. These Y-chromosome–associated effects reflect the full repertoire of the immune response.
These observations have direct clinical consequences. Donor-recipient sex combination has to be considered in any donor-selection algorithm. Rejection is primarily of concern when transplantation is planned for aplastic anemia or congenital disorders. If there is a choice, a male donor for female recipient should not be selected. The decision is more complex in hematologic malignancies. Transplantation-related mortality is more of a concern than rejection. In the past, the strategy was clear: a female donor for a male recipient was to be avoided.4 The decision has now become more difficult. In the short term, survival is worse for males with female grafts. The detrimental effect of GVH reaction is, early after HSCT, higher than the benefit from GVL. However, if with longer follow-up the GVL benefit outweighs early loss, we might have to reconsider the decisions of today.
The findings raise even more intriguing questions. What are the mechanisms behind the recognition of male gene products by women? If nearly 20% of healthy women and 50% of male recipients of female transplants have anti-Y antibodies, why are the clinical consequences so rare? Why are so many HLA-identical sex-mismatched transplants accepted with no rejection? Why are male fetuses safe during pregnancy? There is a hint to these questions in the work of Miklos et al. In all situations when they were able to identify the peptide sequences, there was a difference between the male DBY and female DBX protein. It is unknown whether the differences in these peptides represent spontaneous mutations or a basic inherent sex-linked difference. Obviously, we received clear answers but are left with many open questions.
