Tumour necrosis factor-alpha (TNF alpha) is a pleiotropic cytokine which is overproduced in rheumatoid joints primarily by macrophages. This cytokine has a potential pathogenic role in the establishment of rheumatoid synovitis, in the formation of pannus tissue and in the process of joint destruction, as it increases synoviocyte proliferation and triggers a cascade of secondary mediators involved in the recruitment of inflammatory cells, in neo-angiogenesis and in the process of joint destruction. These findings made TNF alpha a potential target for anticytokine therapy. Experimental studies have shown that TNF alpha blockade by monoclonal antibodies or by soluble TNF receptor reduced the extent and severity of arthritis both in collagen-induced arthritis in mice and in transgenic mice overexpressing TNF alpha, which develop a rheumatoid-like destructive arthritis. Clinical studies based on the use of anti-TNF alpha antibodies or soluble receptors have suggested a potential beneficial effect of TNF alpha-blocking therapy in inducing amelioration of inflammatory parameters in patients with long-standing active disease. In these patients anti-TNF alpha therapy induces a rapid improvement in multiple clinical assessment of disease activity, including morning stiffness, pain score, Ritchie articular index and swollen joint count. The clinical benefits are associated with an improvement in some serological parameters, such as C-reactive protein and serum amyloid-A, erythrocyte sedimentation rate, blood cytokine levels, haemoglobin, white cells and platelet counts, rheumatoid factor titre and histological features of the synovium. However, it remains to be determined whether anti-TNF alpha therapy may be useful in the long term management of rheumatoid patients and in the achievement of better outcomes of disease. Because TNF alpha production also serves a specific function in host defence against infections and tumours, the adverse effects of long term anti-TNF alpha therapy must be carefully evaluated. In addition, targeting a single mediator may be not sufficient to block the complex inflammatory response in rheumatoid arthritis. For these reasons therapeutic strategies aimed at concomitantly interfering with multiple pathogenic pathways are currently under investigation.