Alloreactivity in stem cell transplantation

Abstract

Acute graft-versus-host disease (GVHD), relapse and graft rejection are the main complications after allogeneic hematopoietic stem cell transplantation (HSCT). The aim of this thesis was to achieve a better understanding of the alloreactivity seen after HSCT, focusing on graft rejection, but also including the graft-versus-leukemia (GVL) effect and GVHD. We prospectively evaluated the GVL effect seen in 199 patients with acute lymphoblastic leukemia (ALL) after HSCT. Independent risk factors for relapse were the absence of chronic GVHD, absence of herpes simplex virus (HSV) infection after HSCT, GVHD prophylaxis with methotrexate (MTX) and cyclosporine A (CsA) and >6 weeks from the diagnosis to complete remission (CR). The association between HSV infection and a low relapse is a new observation and may indicate that viral antigens play a role in the induction of an antileukemic effect.

We also studied whether certain cytokine gene polymorphism were associated with severe GVHD after HSCT. We analyzed 196 patients and their corresponding donors for TNF-308, TNFd, IL-10(-1064) and IL-10(-1082) gene polymorphism. Serum analysis of TNF and IL-10 levels during conditioning therapy was also performed. Our results showed that among patients with sibling donors, the TNFd allele 4 was significantly correlated with acute GVHD grades II-IV. Acute GVHD grades II-IV were more common among patients homozygous for the IL-10-1064 allele 13. Patients homozygous for the TNF-308 allele (AA) correlated with higher TNF-alpha serum levels during conditioning.

We analyzed whether antibody-mediated rejection with antibodies against an important subpopulation of hematopoietic stem cells may cause rejection after HSCT. Between year 2000 and January 2006 20 patients rejected their grafts and we were able to analyze 11 patients with rejection. These 11 patients underwent totally 20 transplantations. In the study we also included 30 patients without rejection and 20 non-transplanted healthy individuals as controls. Ninety-three sera taken pre and post-transplantation from patients receiving HSCT were studied for the presence of donor CD34+/VEGFR-2+ cell-specific antibodies. Patients with rejection and controls were analyzed with FACS and microcytotoxicity assay.

We provide evidence that significantly higher numbers of patients with rejections 9/11 while 1/30 (p=0.001) without rejections had antibodies against donor CD34+/VEGFR-2+ cells, but not CD34-/VEGFR-2-cells. In eight transplantations, antibodies against donor CD34+/VEGFR-2+ cells were detected prior to transplantation.

We treated three patients with antibody-mediated rejection with immune modulation; i.e. plasmapheresis, intra venous immunoglobulin (IVIG) and rituximab prior to re-transplantation. With FACS and microcytotoxicity assay we could follow the pattern of antibodies of concern in sera during the immune modulatory treatment. Two patients had antibodies against donor CD34+/VEGFR-2+ cells and the third patient had anti-HLA-antibodies due to massive blood transfusions before HSCT. The immune modulatory regimen was well tolerated without any major side effects. In one patient with antibodies against CD34+/VEGFR-2+ cells, plasmapheresis resulted in elimination of the antibodies according to microcytotoxicity assay but the patient did not have complete donor engraftment until after development of severe acute GVHD. In the patient with high levels of anti-HLA-antibodies, receiving cord blood HSCT, plasmapheresis decreased the levels of anti-HLA-antibodies. Following cessation of plasmapheresis, the antibody titers increased again after HSCT and the patient never engrafted.

We also evaluated the value of cytotoxic T- and B-cell crossmatch testing before HSCT in 157 patients receiving grafts from unrelated donors. Eleven patients rejected their grafts. One of 11 patients with rejection was positive in a T-cell crossmatch before HSCT and 4/11 in B-cell crossmatches. This method showed a low sensitivity but high specificity concerning rejection. A positive T- and/or B-crossmatch before SCT had no predictive value for survival in this study as compared to patients with a negative crossmatch.

To conclude, HSV infection may decrease leukemic relapse after HSCT, the TNFd4 allele and IL-10 (-1064) allele 13 in patients was correlated to acute GVHD grades II-IV. Antibodies to donor CD34+/VEGFR-2+ cells were correlated to rejection. Immune modulation including plasma exchange, IVIG and rituximab may eliminate CD34+/VEGFR-2+ antibodies and pave the way for engraftment. The cytotoxic crossmatch analysis did not predict graft rejection.