Conditioning Intensity and Dendritic Cell (DC) Activation: Implications for GVHD Control Using DC Depleting Antibodies.

Graft versus Host Disease (GVHD) and treatment related mortality (TRM) are the major limitations to the widespread application of allogeneic haematopoietic stem cell transplantation (HSCT) for haematological and non-haematological malignancies. Dendritic cells (DC) as the key initiators and directors of the immune response are central to allogeneic transplant interactions. Preparative conventional conditioning (CC) regimens aim to control disease and ablate the host immune response to facilitate normal haematopoietic reconstitution. The conditioning also unleashes a cytokine storm that activates the residual host immune system, driving host DC and donor T cell interactions that result in GVHD. Reducing the intensity of conditioning (RIC) regimens maintains immune anti-leukaemic activity of T replete HSCT, reduces TRM and delays the onset of GVHD, but the overall incidence of GVHD is unchanged. We hypothesise that this is due to increased persistence of host DC. We propose that strategic administration of DC depleting antibodies could be an effective means to control GVHD. Whilst there is some information on the effects of CC on DC; we have shown that mature plasmacytoid (p) DC are increased in mouse spleen after conditioning by radiation, there is no information on the effects of RIC on DC. We have established murine models of conditioning (CC = Cyclophosphamide [CY] + 800cGy and RIC = Fludarabine [FLU] + CY + 200cGy). The effects on DC numbers, activation and subset composition (myeloid (m) DC and DC), cytokine and systemic endotoxin levels were studied on each day of the conditioning regimens in the absence of HSCT. Mice receiving CC have a significantly higher percentage of DC which are pDC compared to RIC (p<0.001) which results in a decrease in the overall percentage of mDC. However, mice that received RIC have significantly higher absolute numbers of host pDC than CC mice. Preliminary data shows no difference in endotoxin levels in mice receiving RIC or CC without HSCT. However, there may be a transient increase in endotoxin levels in mice after 2 FLU injections (p=0.12). No such increase was seen after CC. There were significantly higher levels of TNF-α (p=0.02) and IL1-β (p=0.03) in mice receiving RIC rather than CC without HSCT. The higher absolute numbers of DC, altered subset ratio and cytokine production appears to account for the delayed onset of GVHD in RIC transplant recipients. Intra-peritoneal (ip) injection of N418, a monoclonal antibody to mouse leukocyte integrin CD11c depleted murine DC in vivo. Preliminary experiments show elimination of 50% of DC after injection of N418 (500mg). Subsequent experiments show that 1mg of N418 is sufficient to significantly delay, but not prevent, GVHD in a full MHC mismatched model of HSCT (p=0.025). The action of N418 is specific, as DC depletion was not seen in mice treated with 1 mg hamster Ig. Together, these observations suggest that increasing antibody concentration and prolonged administration may be required to prevent GVHD. The successful application of DC depletion to control GVHD will improve the safety of HSCT for patients with leukaemia.