Anticancer antibodies offer promise of high sensitivity and specificity for the detection of disease. as they bind to tumour-associated antigens usually found only on a given type of tumour. Most studies have shown that radioimmunodetection is superior to other imaging modalities in the detection of disease outside the liver. Indium-111 (In-111)-labelled satumomab pendetide is the first radiolabelled monoclonal antibody to be marketed for detection of disease. It is approved for single-time use for the detection of extrahepatic abdominal disease in colon and ovarian cancer. Optimal radionuclide-antibody conjugation is still not a reality. Antibody labelling with technetium-99m (Tc-99m), although still under development, appears promising. In-111 has suitable half-life and imaging characteristics, but is expensive and has a complex radiation decay scheme, increasing radiation burden; In-111-labelled antibodies result in high hepatic and sometimes bone marrow uptake, precluding their use in the evaluation of hepatic disease. Radioiodine labelling, although easy, is limited to well-equipped nuclear medicine centres. Furthermore, mouse antibodies invariably evoke an immune response in humans, precluding repeated administration. The nontoxicity, specificity and affinity of antibody binding makes it inevitable that antibodies will be used increasingly in the detection of cancer. Several advances still need to be realised: (a) development of nonimmunogenic molecules such as all-human or humanised antibodies or antibody fragments; (b) development of better linkers, allowing stable conjugation of nuclide to antibody; and (c) characterisation of more restricted and better expressed tumour antigens that will permit rapid targeting of antibody to tumour.