Tolerance of the semi-allogeneic fetus presents a significant challenge to the maternal immune system. The effect of pregnancy on maternal cellular immunity was established by assessing maternal effector and regulatory T-cell subsets during human pregnancy. This demonstrated that an increase in maternal peripheral regulatory T-cells or a shift from a Th1 to Th2 phenotype was not a requirement for normal pregnancy. We also determined the profound impact of maternal Cytomegalovirus seropositivity on maternal T- cell dynamics. T-cells with specificity for fetal epitopes have been detected in women with a history of pregnancy but it has been thought that such fetal specific cells were deleted during pregnancy. We identified, using MHC-peptide multimers, fetal-specific CD8 T-cells in half of all pregnancies. The fetal-specific response increased during pregnancy and persisted in the post natal period. Fetal-specific cells demonstrated an effector memory phenotype and retained functional potential. These data show that the development of a fetal-specific adaptive cellular immune response is a normal consequence of human pregnancy. Women with recurrent miscarriage were found to have abnormal T-cell function, with increased IFN\(\gamma\) and Il-17 production. Fetal specific T-cells were also detected in this cohort and progesterone attenuated their function, which may have therapeutic implications.

Myeloid and lymphoid malignancies are potentially curable through a graft versus leukaemia (GvL) effect following allogeneic haematopoietic stem cell transplantation. Whilst donor T cell are thought to be the main mediators of GvL, the effect of donor NK cells within HLA matched T cell depleted transplant setting is more unclear. Patient blood samples were analysed during the first month post-transplant, with higher reconstitution of NK cells at two weeks conferring a relapse protection association. Donor stem cell graft samples, from which NK cells within the patient at two weeks are thought to be derived, similarly displayed a strong association between high NK cell dose and protection from disease relapse. CD56dimDNAM+ NK cells were found to be the population with the most significant association. The ability of NK cells to kill AML blasts in a DNAM dependent manner was shown indicating that direct killing of residual tumour cells may be a valid mechanism of GvL. These findings suggest that optimising the number of NK cells within stem cell grafts should be considered as a means to prevent disease relapse.