Larson S M
Department of Medical Imaging, Memorial Sloan-Kettering Cancer Center, New York, New York 10021.
Cancer Res. 1990 Feb 1;50(3 Suppl):892s-898s.
In the last decade of radioimmunodetection studies the radiolabeled antibody preparations used have gradually changed from polyclonal antibodies labeled predominantly with 131I to monoclonal antibodies labeled with diverse radionuclides including 131I, 111In, 123I, and 99mTc. Over this period progressive improvement in tumor imaging has been observed when one compares the best examples of early studies, performed with 131I labeled heterosera, to the best of modern images, obtained with 123I, 99mTc, or 111In labeled monoclonal antibodies. Important findings in 61 clinical studies reviewed include the reports from several centers which demonstrate occult disease in patients with carcinoma of colon, melanoma, and lymphoma, and the improved sensitivity and specificity of radioimmunodetection in comparison to transmission computerized tomography in the lymph nodes and abdomen, in lymphoma and colon cancer, and ovarian cancer. Evaluation of the liver remains a difficult problem with this technique and standard approaches are superior in most reports. The general principle of targeting radioactivity to tumor with radiolabeled antitumor antibody and the feasibility of developing practical clinical methodology which will add new diagnostic information have clearly been established. Toxicity, particularly for index studies, is reassuringly limited. In all the studies with surgical confirmation after i.v. injection, uptake in tumor is in the range of 0.005% injected dose/g tumor, and this low tumor uptake remains the single greatest limitation of the method. A second important problem is the prompt development of human anti-mouse antibody, which reduces the usefulness of follow-up studies. A serious criticism of the information currently available on radioimmunodetection is that the clinical studies reported to date vary greatly in approach and results. The vast majority of studies are early Phase I clinical trials, from which toxicity information and biodistribution data can be derived but which give limited information about impact on clinical management. Standardization in the study design is needed in order to establish the efficacy of radioimmunodetection in adequate and well controlled clinical trials.
