Developing a Therapeutic Anti - Dendritic Cell Antibody to Prevent Graft Versus Host Disease.

Host and donor dendritic cells (DC) stimulate alloreactive donor T lymphocytes, and initiate GVHD. We have shown that polyclonal antibody to the DC surface activation marker human CD83 (anti hCD83), which depletes activated DC, can prevent human DC and T cell induced lethal xenogeneic GVHD in SCID mice without impairing T cell mediated anti-leukaemic and anti-viral (CMV and influenza) immunity (J Exp Med 2009; 206: 387). Therefore, we made and tested a polyclonal anti mouse CD83 (RAM83) antibody in murine HSCT models and developed a human mAb against hCD83 as a potential new therapeutic immunosuppressive agent.

RAM83 specificity and function was tested by flow cytometry and allogeneic MLRs. It was administered as a single dose on day -1 of murine syngeneic and allo HSCT. Human and mouse single chain variable fragment (scFv) phage libraries were panned on recombinant human CD83 extracellular domain. After screening for specificity and affinity, the clones were reformatted to human IgG1 and expressed by transfected CHO cells. The purified mAb were tested for their ability to block a mixed leucocyte reaction (MLR).

Lin-,Sca1+,Kit+ murine stem or progenitor cells were CD83-ve. RAM83 treated mice receiving syngeneic HSCT survived >35days (normal day 14 WBC). Full MHC mismatched (2× 60mg/kg cyclophosphamide + 2× 500cGy conditioned), B6 [H-2^b] into BALB/c [H-2^d]) recipient mice treated with a single dose of RAM83 developed GVHD later and survived longer (mean 26.9 days ±3.3) compared to controls (RAneg 12.3±1.3days, p=0.001; Nil antibody 7.7±1.5 days, p=0.0008). Daily high dose (50mg/kg) cyclosporin A delayed GVHD by a mean 18.2±3.3 days. Four phage clones that bind to human CD83 have been reformatted. Of the two tested to date, one mouse/human chimeric mAb has lost binding affinity, probably as a result of immunoglobulin variable region glycosylation. The other reformatted fully human anti-CD83 (3C12) mAb blocked a MLR, by an ADCC mechanism.

mAb targeting of activated DC is a promising novel approach to immunosuppression. The 3C12 clone and others will be subjected to preclinical testing in the xenogeneic SCID mouse model. The human mAb, which prevents GVHD without impairing the desired graft versus leukaemia effect will be developed for phase 1 clinical trials in clinical alloHSCT.

Hart:CRC-BT: Consultancy.