Radioimmunodetection of colorectal carcinoma using technetium-99m-labeled Fab' fragments of the IMMU-4 anti-carcinoembryonic antigen monoclonal antibody

Background. Radioimmunodetection of cancer usingmonoclonal antibody fragments offers certain potentialadvantages over that with whole monoclonal antibodies,including the ability to image early (i.e., to provideimages at an early time after injection of the radio- antibody) while minimizing the incidence of human antimouseantibody response. This paper reports a prospectivetrial comparing radioimmunodetection withIMMU-4 (a murine anti-CEA monoclonal antibody)[Tc.sup.99m]-labeled Fab' fragments to conventional imaging in 35 colorectal cancer patients.Methods. All patients were investigated by conventionaldiagnostic methods (CDM) within 4 weeks ofradioimmunodetection. Surgical corroboration of findingswas obtained in 26 patients (15 with evidence of diseaseon CDM [CDM+] and 11 with abnormal serum CEA[CDM-] as the only evidence for recurrence). After 1 mgIMMU-4 [Tc-.sup.99m]-Fab' was injected (19.3 mCi on average), patients underwent planar/SPECT radioimmunodetection2-5 hours later and planar radioimmunodetection18-24 hours later. Three patients underwent a secondradioimmunodetection study 16, 20 and 23 months afterthe first.Results. Radioimmunodetection was superior toCDM, accurately predicting disease distribution in sixnonsurgical and ten CDM+ surgical patients, and wascomplementary to computed tomography in two nonsurgicaland two CDM+ surgical patients. Radioimmunodetectionwould have directed or changed management decisionsin 6 of the 15 (40%) CDM+ surgical patients.Radioimmunodetection correctly identified all recurrenttumor in 8 of 11 CDM- surgical patients and was negativein one patient with cirrhosis and no recurrence, representing a potential clinical benefit of 82%. Analyzed ona regional basis, radioimmunodetection was found to besuperior to CDM in extrahepatic abdomen and pelvisimaging and was complementary to (although not as accurateas) CDM in the liver. Human anti-mouse antibodydid not develop in any of the patients, including threewho were injected twice.