Open Access Highly Accessed Open Badges Research article

Comparison of molecular and immunocytochemical methods for detection of disseminated tumor cells in bone marrow from early breast cancer patients

Bjørnar Gilje12*, Oddmund Nordgård12, Kjersti Tjensvoll12, Elin Borgen3, Marit Synnestvedt4, Rune Smaaland12 and Bjørn Naume45

Author Affiliations

1 Department of Hematology and Oncology, Stavanger University Hospital, Stavanger, Norway

2 Laboratory for Molecular Biology, Stavanger University Hospital, Stavanger, Norway

3 Division of Surgery and Cancer Medicine, Department of Pathology, Oslo University Hospital, Oslo, Norway

4 Division of Surgery, Transplantation and Cancer Medicine, Department of Oncology, Oslo University Hospital, Oslo, Norway

5 K.G. Jebsen Center for Breast Cancer Research, Institute for Clinical Medicine, University of Oslo, Oslo, Norway

For all author emails, please log on.

BMC Cancer 2014, 14:514  doi:10.1186/1471-2407-14-514

Published: 15 July 2014



Disseminated tumor cells (DTCs) have potential to predict the effect of adjuvant treatment. The purpose of this study was to compare two methods, reverse transcription quantitative PCR (RT-qPCR) and immunocytochemisty (ICC), for detecting breast cancer DTCs in bone marrow (BM) from early breast cancer patients.


We investigated a subset (n = 313) of BM samples obtained from 271 early breast cancer patients in the “Secondary Adjuvant Taxotere Treatment” (SATT)-trial. All patients in this study had node positive or intermediate/high-risk node negative non-metastatic disease. The DTCs were detected by ICC using AE1-AE3 anti-cytokeratin monoclonal antibodies. Patients with DTCs detected in their BM by ICC after standard adjuvant fluorouracil, cyclophosphamide, epirubicin (FEC) chemotherapy were offered docetaxel treatment. For comparison, 5 × 106 mononuclear cells from the aliquoted BM samples were also analyzed by RT-qPCR using a multimarker (MM) assay based on the tumor cell mRNA markers keratin 19 (KRT19), mammaglobin A (hMAM), and TWIST1. In the MM-assay, a sample was defined as positive for DTCs if at least one of the mRNA markers was positive.


The MM RT-qPCR assay identified DTCs in 124 (40%) of the 313 BM samples compared with 23/313 (7%) of the samples analyzed by ICC. The concordance between the MM RT-qPCR and ICC was 61% (Kappa value = 0.04) and twelve of the BM samples were positive by both methods. By RT-qPCR, 46/313 (15%) samples were positive for KRT19, 97/313 (31%) for TWIST1, and 3/313 (1%) for hMAM mRNA. There were no statistically significant associations between the individual mRNA markers.


The RT-qPCR based method demonstrated more DTC-positive samples than ICC. The relatively low concordance of positive DTC-status between the two different assessment methods suggests that they may be complementary. The clinical relevance of the methods will be evaluated based on future clinical outcome data.

Trial registration NCT00248703.

Disseminated tumor cells; RT-qPCR; Immunocytochemistry; Breast cancer; Bone marrow