Interference of cytokeratin-20 and mammaglobin-reverse-transcriptase polymerase chain assays designed for the detection of disseminated cancer cells.

Several reverse-transcriptase polymerase chain reaction (rtPCR) assays have been designed for the detection of disseminated cancer cells. The specificity of these discussed molecular approaches is controversial. Biological interference of the cytokeratin-20 and mammaglobin rtPCR assays has been investigated. Cell lines of different lineages and bone marrow and peripheral stem cells from patients without epithelial cancer have been examined for the transcription of the cytokeratin-20 (CK20) and mammaglobin messages prior to and after stimulation with different cytokines in a total of 370 liquid cultures. Amplification of both messages from clinical samples prior to stimulation does not support the high specificity for the detection of disseminated epithelial cancer cells as reported. Cytokeratin-20 was amplified from the chronic myeloic leukemia (CML)-derived line K562. Transcription was not influenced by cytokines, either in cell-line experiments or in clinical samples. The thesis of a low-level background transcription in granulocytes is supported. Mammaglobin was induced in cell lines significantly by GM-CSF and in clinical samples additionally by several more cytokines. These results indicate that under certain conditions involving cytokine production, the use of mammaglobin rtPCR for the detection of epithelial cancer cells could be limited. In conclusion, the mechanism of interference of both rtPCR assays are completely different and further research is necessary before the cytokeratin-20 or mammaglobin rtPCR could become standard methods for the detection of disseminated epithelial cancer cells. These factors leading to so-called false-positive results have to be considered in future applications of rtPCR for the detection of minimal residual disease.