Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Detection of circulating epithelial cells in the blood of patients with breast cancer: comparison of three techniques.

Author(s): Ring AE, Zabaglo L, Ormerod MG, Smith IE, Dowsett M

Publication: Br J Cancer, 2005, Vol. 92, Page 906-12

PubMed ID: 15714202 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared sensitivities and specificities of three different methods for the detection of circulating epithelial cells in blood using immunomagnetic separation (IMS) and laser scanning cytometry (LSC), cell filtration and LSC, or a multi-marker real-time RT–PCR assay. Blood (18 mL) was collected in EDTA tubes (first 5 mL was discarded to minimize contamination by skin epithelial cells) from 20 healthy volunteers and 40 patients diagnosed with metastatic breast cancer, processed within 30 min, and divided into three 6 mL aliquots. The IMS/LSC aliquot was subjected to density gradient enrichment prior to incubation with a BerEP4-FITC primary antibody, a secondary antibody followed IgG Microbeads. Cells were then harvested via magnetic separation and centrifugation, after which the cytospin spot was stained for cytokeratin and analyses using a Laser Scanning Cytometer. The cell filtration/LSC aliquot was diluted 1:1 in PBS 2 mM EDTA and filtered under gravity through a Poretics polycarbonate Track-Etch-type (PCTE) membrane prior to washing, staining for cytokeratin and LSC analysis. The real-time RT-PCR aliquot was subjected to density gradient enrichment before RNA was extracted with Tri Reagent, quantified using a bioanalyzer, and analyzed by real-time RT–PCR for CK19, mammaglobin, PIP, and ribosomal RNA RPL19.

   Summary of Findings:

   The mean number of positive epithelial cells did not differ significantly when detected using LSC following IMS compared to LSC following cell filtration and were modestly correlated in blood collected from healthy volunteers (0.40 vs. 0.42/6 mL, r=0.432, P=0.007) and strongly correlated in blood collected from patients diagnosed with metastatic breast cancer (6.55 vs. 3.87/6 mL, r=0.742, P=0.0012,). The percentage of breast cancer patients identified as positive for circulating epithelial cells when the upper 95% confidence interval of mean cell number in healthy volunteers was used as a cutoff was greater when cells were detected by LSC following IMS compared to LSC following cell filtration (48% vs. 30%, respectively; P=0.18) and also when the threshold value for positive samples was set to correspond to 100% specificity in healthy volunteers (38% vs.25%, respectively).

   Statistically significant differences were reported between blood collected from healthy volunteers and patients diagnosed with breast cancer for mean normalized expression (MNE) of CK19, mammaglobin, and PIP (P<0.0001, P=0.0011, and P=0.002, respectively) but not for the ribosomal RNA RPL19 (P=0.58). Two (10%) specimens from healthy volunteers were positive for only one marker and none were positive for more than one marker. The percentage of breast cancer patients identified as positive for circulating epithelial cells when the upper 95% confidence interval of MNE in healthy volunteers was used as a cutoff was transcript-specific, as  24/40 (60%) specimens were positive for CK19, 18/40 (45%) were positive for mammaglobin, and 14/40 (35%) were positive for PIP with seven (18%) being negative for all three markers, 16 (39%) positive for only one gene, 11 (28%) positive for two, and six (15%) positive for all three markers. When the cutoff applied was based upon 100% specificity in healthy volunteers, 22/40 (55%) specimens from patients diagnosed with breast cancer were positive for CK19, 18/40 (45%) for mammaglobin, and 7/40 (18%) for PIP with 11 (28%) negative for all three markers, 15 (38%) positive for only one, 10 (25%) positive for two, and four (10%) positive for all three. Concordance as defined by specimens determined as either positive or negative by both markers analyzed was 65% between CK19 and mammaglobin, 53% between CK19 and PIP, and 50% between PIP and mammaglobin; but significantly more specimens were positive for CK19 than for PIP (60 vs 35%, P=0.014).

   Significantly more blood specimens collected from patients diagnosed with breast cancer were positive for at least one of the markers in the real-time RT–PCR assay than by LSC following cell filtration or IMS (83 vs 30%, P<0.001 and 83 vs 48%, P<0.001; respectively) and more specimens were positive for CK19 using the real-time RT–PCR assay than using LSC following cell filtration (60 vs 30%, P=0.006) but not LSC following IMS (60 vs 48%, P=0.06). Significantly more specimens were determined positive for at least one of the markers in the real-time RT-PCR assay than by LSC following cell filtration or IMS (73 vs 25%, P<0.001 and 73 vs 38%, P<0.01, respectively). Concordance as defined by specimens determined as either positive or negative by both techniques used was 65% when analyzed by LSC following cell filtration and IMS (9/12 patients positive by cell filtration were also positive by IMS). When comparing LSC following IMS and real-time RT–PCR, concordances were 65, 70, and 50% for CK19, mammaglobin, and PIP, respectively; and 60, 60, and 50% when comparing LSC following cell filtration with the real-time RT–PCR assay. The concordance between specimens defined as positive by all three markers in the real-time RT-PCR assay and those positive by LSC following cell filtration and IMS was 73 and 60%, respectively.

   Biospecimens

   • Bodily Fluid - Blood

   Preservative Types

   • None (Fresh)

   Diagnoses:

   • Neoplastic - Carcinoma
   • Normal

   Platform:

   Analyte	Technology Platform
   Protein	Fluorescent microscopy
   RNA	Real-time qRT-PCR
   RNA	Automated electrophoresis/Bioanalyzer

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Diagnosis/ patient condition	Healthy Breast Cancer
   Biospecimen Aliquots and Components	Cell capture method	immunomagnetic separation cell filtration
   Real-time qRT-PCR Specific	Targeted nucleic acid	CK19 mammaglobin PIP ribosomal RNA RPL19

   View more