Minimal residual disease detection in breast cancer: improved sensitivity using cytokeratin 19 and epidermal growth...

Breast cancer cells pose a difficult target for detection due to their lack of tumor specific markers. Small numbers of metastatic cells circulate through the blood and reach target organs, but current methods are presently not sensitive enough to allow for early detection. Our goal is to develop a method for specific and sensitive detection for breast cancer tumor cells by using cell enrichment coupled to RT-PCR with selective primer pairs. Fifty million normal donor mononuclear cells (NDMNC) were spiked with serial dilutions of a breast cancer cell line, positively selected for surface expression of the epithelial cell adhesion molecule (EPCAM) with anti-Ber-EP4 mAb. Total RNA was isolated and the cDNA was transcribed and measured by quantitative PCR for cytokeratin 19 (CK19), epidermal growth factor receptor (EGF-R), and beta-2-microglobulin (beta2M) transcripts using specific hybridization probes with the LightCycler System. A sensitivity of 1 tumor cell in 5 million NDMNC was consistently achieved with a metastatic breast cancer cell line with target primer sets for CK19 and EGF-R. Less, but appreciable sensitivity was achieved by spiking NDMNC with other breast cancer cells. Detection of both gene targets ranged between 1 in 5,000 to 5 million cells. The method described, a cell enrichment procedure and RT-PCR using specific hybridization probes (spanning introns and hybridizing to areas that discriminate transcribed pseudogenes), results in increased sensitivity while decreasing false positives above those methods previously reported. This general approach may have wide applicability to increasing sensitivity of minimal residual disease detection in other cancers.