Skip to main content


We're creating a new version of this page. See preview

  • Meeting abstract
  • Open Access

Cytokeratin and mammaglobin as tumor markers in patients with high risk breast cancer

  • 1,
  • 1,
  • 1,
  • 1,
  • 1,
  • 1,
  • 1 and
  • 1
Breast Cancer Research20002 (Suppl 1) :P7.05

  • Published:


  • Breast Cancer
  • Risk Breast Cancer
  • Prognostic Relevance
  • Cell Collection
  • Internal Sequence

Aims of the study

To investigate the incidence and prognostic relevance of tumor micro-contamination in BM and/or G-CSF-mobilized peripheral blood progenitor cells collections of stage II-III breast cancer patients.

Patients and methods

Patients were enrolled from September 1998 through May 1999 and underwent high-dose chemotherapy with autologous PBPC transplantation. We analysed a total of 71 patients for TC on PBPC apheresis (28 of them had previously frozen samples). 43/71 patients also had their BM evaluated on the day of the first PBPC collection after 5 days of G-CSF administration (BM-STIM) and immediately before HDCT (BM-PRECT). Cytokeratin (CK) expression was evaluated in all samples by either immunocytochemistry (ICC, sensitivity 1 × 10-6 cells, 9-16 × 10-6 cells for each sample) and reverse transcriptase nested PCR (RT-PCR, sensitivity 1 × 10-7cells). Amplified products were then annealed to a 32P-labelled internal sequence probe to confirm specificity. 41 patients were also evaluated by nested RT-PCR for Mammaglobin (MAM, sensitivity 1 × 10-6 cells) gene expression.


(1) PBPC frequency of CK+ was 11% by ICC and 66% by RT-PCR on 71 patients' samples; (2) BMSTIM and BMPRECT frequency of CK+ was 7-14% by ICC and 60-65% by RT-PCR on 43 patients' samples; (3) all CK ICC+ samples were MAM RT-PCR+; (4) 53% of patients with CK RT-PCR+ BM-STIM had their PBPC and BM-PRECT CK RT-PCR+; (5) after a median follow-up of 21 months on 28 patients with frozen samples, 25% of patients relapsed and 43% of them had contaminated PBPC; (6) 74% of CK ICC+ samples were CK RT-PCR+; (7) BMSTIM, PBPC and BMPRECT frequencies of MAM+ were 15, 17 and 22% respectively by RT-PCR on 41 patients' unfrozen samples; (8) PBPC frequency of MAM+ was 20% by RT-PCR on 25 patients' frozen samples.


(1) MAM gene evaluation could add a lot of sensitivity and specificity to the overall results. (2) G-CSF administration for PBPC mobilization does not increase PBPC contamination. (3) A longer follow-up of these patients is needed to evaluate the prognostic relevance of different markers of tumor cell contamination in HDCT for breast cancer.

Authors’ Affiliations

European Institute of Oncology, Milan, Italy


© Current Science Ltd 2000