Background: Cytokeratin 20 reverse transcriptase polymerase chain reaction (CK20 RT-PCR) of blood and bone marrow specimens has been suggested for assessment of hematogenously disseminated tumor cell (DTC) spread in colorectal cancer (CRC) patients. Considerable discrepancies among the studies reported indicate a need for better evaluation procedures. We investigated whether mononucleated cell (MNC) enrichment by Ficoll density gradient centrifugation followed by immunomagnetic depletion of CD45-positive cells (extended enrichment) allows better detection of DTC-associated CK20 mRNA compared to MNC enrichment by Ficoll density gradient centrifugation alone (Ficoll enrichment).
Methods: We analyzed 53 samples [38 peripheral blood (PB), 15 bone marrow (BM)] from 38 CRC patients. Extended enrichment was performed for 30 specimens (PB and BM, n=15 each), and Ficoll enrichment for 23 blood specimens. Total RNA was extracted, reverse-transcribed and analyzed by real-time RT-PCR using a LightCycler instrument.
Results: Despite extended enrichment, 10 PB and 8 BM samples could not be analyzed because of low cellular yield. The depletion efficiency of CD45 separation was 2 log. RT-PCR of the housekeeping gene PBGD resulted in high and varied crossing point values (mean 37.1+3.0) for five PB and seven BM specimens. Ficoll enrichment yielded 23 analyzable blood specimens for which the mean crossing point value was 26.7+0.5 in PBGD RT-PCR. CK20 RT-PCR of 23 blood samples (all from Dukes D patients) revealed CK20 transcripts in four cases (17%).
Conclusions: Extended enrichment was not superior to Ficoll enrichment; in fact, the sensitivity was lower. Improvement of the reported CK20 RT-PCR assay of Ficoll-enriched MNC populations is warranted.