My presentation today was about the use of haploidentical stem cell transplantation as a tool for immunotherapy in children with acute lymphoblastic leukaemia. Our focus is mainly to use allogenic or haploidentical stem cell transplantation as an immunotherapeutic approach and haploidentical means in most of our children we use the parents as the stem cell donor. In order to facilitate a transplant and make it safe we do an in vitro T-cell depletion or ex vivo T-cell depletion. So we take out the T-cells which would otherwise cause graft versus host disease out of the graft and the T-cells which we take out are the so-called T-cell alpha-beta T-cells because these are the main T-cells which cause graft versus host disease. So this allows us to use parents as a donor in our children and then after the transplant we further treat these children with an immune augmenting therapy against the leukaemia in order to avoid relapses.

So our focus is on monoclonal antibodies against the target antigen CD19 which is expressed on older leukaemic blasts, or most of them. We have generated our own chimeric anti-CD19 antibody which we use then for these children after they have received these stem cell grafts from their parents. So far our results are promising so we have done a number of children now who have relapsed after standard transplant. Then we do a second or some children even receive a third transplant from their parents and then we do this antibody therapy against CD19 for up to two years.

So far our overall survival is quite promising in these children in contrast to our previous experience without this antibody treatment after transplantation. The main focus is to activate the so-called natural killer cells with this antibody. So this antibody is engineered to exert a very effective so-called ADCC, antibody dependent cellular cytotoxicity, which activates quite effectively natural killer cells via their Fc receptor, the so-called CD16, and this activation results in quite long relapse free survival of these children.

The tolerability is very good, the children tolerate this treatment very well. We have not seen severe side effects of this antibody treatment and now, of course, are focussing on performing a multi-centre prospective trial using this approach in patients with advanced leukaemia.

Our other focus is on the use of so-called memory natural killer cells. This is an interesting subpopulation of natural killer cells which seems to have some memory, like CD8 memory T-cells. There seems to be a similar feature in NK cells so once the NK cells have seen the leukaemic cells and again they seem to have a memory for them because they are much faster activated, they are much more effective and they produce much more cytokines after the second exposure to these leukemic blasts. So we are also focussing on using this kind of memory NK cells for post-transplant adoptive transfer, so to isolate these memory NK cells and later adoptively transfer these NK cells, so these are donor NK cells after transplant, to adoptively transfer these NK cells to the patient also to avoid or prevent relapses in these high risk patients.

That’s one approach and the other approach is that we use also a CAR T-cell approach. Our approach is a little bit different than from the other approaches. We use an adaptor strategy so we separate the CAR T-cells from the antigen that we use CAR T-cells which are directed against biotin once the biotin is bound to a protein, in this case an antibody. So we generate one CAR T-cell against the biotin and then we can use any biotinylated antibody as the target. Of course this is a lot of still ongoing work in vitro and in vivo in mouse models. It seems to work quite effectively but we have not yet clinical data, this or next year we hopefully can start with some clinical trials, but the system is very flexible because we can use several targets at the same time. The problem at the moment with CAR T-cell therapy is that, especially with CD19 CAR T-cells which are mainly used in acute lymphoblastic leukaemia, is the fact that relapses occur and then these relapses are CD19 negative. So there is a selection process. We can use several targets at the same time, for example we use biotinylated antibodies against CD19, against CD20, against CD22, which are all expressed on the leukaemic blasts. Once we use all these together it seems that at least in our in vitro models this antigen is much less of a problem than we use only one single target antigen. So that’s one of the advantages of using such a flexible system.

Of course we also anticipate that the system is safer because we can switch the CAR T-cells on and off because these CAR T-cells otherwise do not recognise anything unless you don’t give any biotinylated antibody. But we have to wait, of course, for the clinical studies.