The autoimmune response is executed via cognate interactions between effector immune cells and antigen presenting cells. Cognate interactions provide the immune effectors with specific signals generated through the antigen receptor as well as with second, non-specific signals, generated from the interaction of pairs of cell-surface molecules (costimulatory molecules) present on their plasma membrane. Disruption of this second, non-specific costimulatory signal results in the interruption of the productive (auto)immune response, leading to anergy, a state of immune unresponsiveness. The CD28:B7 families of molecules and the CD40:CD40L pair of molecules are considered as critical costimulatory elements. Disruption of the CD28:B7 interaction using a genetically engineered soluble form of the inhibitory molecule CTLA4 in vitro, as well as in experimental models of rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE), led to the inhibition of the autoimmune response. Similarly, promising data stem from the use of an anti-CD40L monoclonal antibody (mAb) in murine SLE. While such treatments prevent the development of autoimmunity in animal models, this preventive approach is inapplicable to human diseases. However, the rational bench-to-bedside approach led investigators to clinical trials of CTLA4-Ig and of two different humanized anti-human CD40L mAbs in patients with RA and SLE, respectively. Initial experience with the use of CTLA4-Ig in patients with RA is encouraging, since in one short-term trial the construct was well-tolerated and produced clinically meaningful improvement of the disease in a significant proportion of those treated. Surprisingly, the anti-CD40L mAb treatment approach in human lupus was not fruitful, since short-term administration of the anti-CD40L mAb ruplizumab in lupus nephritis was correlated with life-threatening prothrombotic activity despite initial encouraging data in the serology and renal function of the patients. Also, IDEC-131 anti-CD40L mAb treatment did not prove to be clinically effective in human SLE, despite being well tolerated. Precise tailoring of the administration schemes for these novel therapeutic modalities is awaited.Finally, conceptually different approaches to block costimulation by inhibiting the induced expression of costimulatory molecules or the transmission of their specific intracytoplasmic signal have already produced encouraging preliminary results.