Radioimmunodetection of cancer: problems and potential.

E M Rankin, J. Gordon McVie

BMJ · 1983 · 14 citations · 29 references

DOIFull text

Open access

Concepts

Abstract

tion of polyclonal goat anticarcinoembryonic antigen anti? bodies. They accurately located 85% of colorectal cancers, 88% of ovarian cancers, 90% of cervical cancers, 71% of lung cancers, 50% of pancreatic cancers, 40% of primary breast cancers, and 78% of metastatic breast cancers.2 Plasma concentrations of carcinoembryonic antigen did not always correlate with the results of radioimmunodetection : tumour was seen in some cases with normal concentrations and missed in some with high concentrations. Most of the sites that were missed were less than 2 cm diameter, and some of the positive sites later confirmed as malignant were not identified with conventional detection methods. Unfortunately, other groups have had difficulty in matching these results. Dykes et al were able to detect four of five primary sites and eight of 11 metastatic sites in 15 patients with gastrointestinal neoplasms using polyclonal sheep anti? carcinoembryonic antigen antibodies.5 In three patients the technique was totally unsuccessful: in two cases technical problems may have contributed, but in the third there seems no explanation for the failure of the method to locate the tumour. Another study using goat and baboon anticarcino? embryonic antigen antibodies gave some disturbing results.6 Deiodination in vivo was rapid, up to half of the administered 131I being excreted in the urine within the first 24 hours, resulting in poor imaging. Four of 27 tumour sites were identified without resort to subtraction techniques, but false positive results were generated in every case with subtraction.

References

29