The usefulness of monoclonal antibodies (mAbs) in the transplantation field has become evident over the last couple of years. Different mAbs have been used as a prophylactic treatment after transplantation, in a therapeutic way against acute organ rejection and new diagnostic tools to predict clinical rejection immerge. One can even hope that with humanised mAbs or human mAbs obtained by repertoire cloning the formation of human anti-mouse antibodies will be solved although on the one hand this appeared not to be a big problem and on the other hand anti-idiotypic antibodies can still be expected. However, the most puzzling question is how the mAbs modulates the immuno-response and this not only in organ rejection but also in auto-immune diseases. Only one out of many CD25 mAbs with seemingly similar epitope recognition can be used in therapeutical treatment of acute Graft versus Host Disease. The same mAb is not, however, very efficient in the prophylactic treatment of kidney transplantation without association of suboptimal doses of cyclosporin A. Another example is a CD4 mAb which is efficient in the treatment of polyarthritis with no side effects but which provokes transient but clear side effects when used in psoriasis or multiple sclerosis patients. A second CD4 mAb with high inhibitory activity in several bioassays compared to the first CD4 mAb has no beneficial effect at all on polyarthritis. Also the question why there is a percentage of "no response" patients among apparently identical "good response" patients remains unanswered. However it becomes clear from these experiences that: 1) mAbs recognizing a similar epitope and being of the same isotype will not automatically have the same effect in therapy. 2) side effects may be depending of the disease treated. 3) the activities of mAbs in bioassays and even animal models very often do not reflect the in vivo situation in human. 4) efficiency of the treatment with mAbs can be increased by a better understanding of the mode of action and increased efficiency can be expected by association of several mAb or mAb with drugs for the "no response" patients and should be the next step in the therapeutical use of mAbs.