A comparative study of the relative sensitivity and specificity of radiolabelled monoclonal antibody and computerised tomography in the detection of sites of disease in human malignant melanoma.

A monoclonal antibody raised against the high molecular weight melanoma antigen was labelled with indium-111 and injected intravenously into 25 patients with malignant melanoma. The results obtained from images at 24 and 96 h post i.v. administration of the antibody were compared with results obtained from computerised tomography studies with regard to detection of previously unrecognised sites of metastatic disease and apparent false positive localisation. Detailed study of the patients' clinical condition and detection rates using the two methods suggest that both methods detect approximately 80% of clinically and pathologically confirmed metastases. Of 62 known metastases, the antibody detected 50 (81%), with 17 false positive results. False negatives were most common in the lung. In eight patients the two methods were considered of equal value, in 10 the monoclonal gave a greater amount of clinically relevant information, and in seven the CT was superior. In three patients clinically significant metastatic lesions were detected by the radiolabelled monoclonal and had not been previously recognised either by CT scanning or on clinical grounds. No patients had any adverse reaction to the antibody and in the course of our study the dose of antibody was reduced from 20 mg to 200 micrograms with no apparent loss of sensitivity. In at least two patients uptake of the labelled monoclonal into tumour sites would have been adequate for effective targeted radiotherapy.