NIH, National Cancer Institute, Division of Cancer Treatment and Diagnosis (DCTD) NIH - National Institutes of Health National Cancer Institute DCTD - Division of Cancer Treatment and Diagnosis

Isolation by size of epithelial tumor cells : a new method for the immunomorphological and molecular characterization of circulatingtumor cells.

Author(s): Vona G, Sabile A, Louha M, Sitruk V, Romana S, Schütze K, Capron F, Franco D, Pazzagli M, Vekemans M, Lacour B, Bréchot C, Paterlini-Bréchot P

Publication: Am J Pathol, 2000, Vol. 156, Page 57-63

PubMed ID: 10623654 PubMed Review Paper? No

Purpose of Paper

This paper compared the sensitivity and specificity of two methods of circulating tumor cell (CTC) detection: isolation by size of epithelial tumor cells (ISET) and RT-PCR amplification of α-fetoprotein (AFP) or prostate-specific antigen (PSA). Potential morphological effects on cells introduced by ISET was also investigated.

Conclusion of Paper

ISET demonstrated greater sensitivity and specificity in comparison to RT-PCR for both analysis of peripheral blood from patients diagnosed with hepatocellular carcinoma and blood spiked with cultured carcinoma cells. CTC morphology was unaffected by ISET.


  1. Study Purpose

    This study compared the sensitivity and specificity of two methods of CTC detection: ISET, which is based on filtration, and RT-PCR amplification of tumor-specific transcripts. This study also investigated potential morphological effects introduced by ISET. Fifteen ml of blood was collected in tubes containing EDTA from eight healthy volunteers, eight patients with chronic hepatitis, and seven patients with hepatocellular carcinoma that were undergoing surgical liver resection.  Blood specimens were processed within 5 hours. To determine if morphology was compromised by ISET, tumor cell lines from hepatocellular carcinomas (HepG2 and Hep3B), breast adenocarcinoma (MCF-7), cervix epithelioid carcinoma (HeLa), and metastatic prostatic adenocarcinoma (LNCaP) were examined post-ISET isolation from spiked peripheral blood specimens using a transmission microscope after H&E and May-Grumwald-Giemsa staining. The same spiked peripheral blood specimens were also subjected to immunostaining for the tumor cell markers α-fetoprotein (AFP) and cytokeratin (KL1) after ISET. ISET was also performed on 6 ml of peripheral blood collected from patients with carcinoma or from healthy volunteers. Briefly, after filtration by gentle aspiration under vacuum the membrane was washed, allowed to air-dry, and was H&E-stained, after which individual cells were collected by laser microdissection. For RT-PCR analyses, nucleated cells were isolated by Ficoll from 6 ml of peripheral blood, RNA was extracted using TRIzol, amplified with a nested protocol using liver-specific (AFP) and prostate-specific (PSA) primers, and analyzed by gel electrophoresis. FISH was performed on ISET-recovered MCF-7 and HepG2 cells using a biotinylated or fluorophore-labeled probe specific for the chromosome 1 centromere. To compare sensitivity, one ml of blood was spiked with one or three HepG2, LNCaP, or MCF-7 cells and ISET was performed followed by counting of tumor cells after cytokeratin immunostaining or analyzed by RT-PCR analyses performed with AFP (α-fetoprotein)-specific primers for HepG2 cells and PSA (prostate-specific antigen)-specific primers for LNCaP cells.

    Summary of Findings:

    Morphological analysis of H&E stained cells revealed that morphology was unaffected by the ISET method. ISET captured tumor cells from tumor cell lines that were immunopositive for cytokeratin and tumor-specific markers (AFP and PSA), while lymphoctes were negative for cytokeratin and positive for LCA.  All immunostaining demonstrated appropriate sub-cellular localization. FISH analysis was also successful on ISET-captured cells, as appropriate staining for the chromosome 1 centromere was successful in MCF-7 and HepG2 cells. In 5-fold replicate tests using one cell spiked in 1 ml blood, ISET detected the LNCaP cell 100% of the time, HepG2 80% of the time, and MCF-7 60% of the time; although when 3 tumor cells were spiked in 1 ml of blood, at least one cell was detected 100% of the time for HepG2 and LNCaP cells, and 80% of the time for MCF-7 cells. Comparatively, RT-PCR was unable to detect one or three HepG2 or LNCaP tumor cells in 1 ml blood. Clinical validation of CTC detection revealed ISET was more sensitive than RT-PCR, as 100% of patients diagnosed with hepatocellular carcinoma were positive with the ISET method compared to 0% with RT-PCR when blood was collected prior to surgery. When blood was collected during tumor resection, 85% of hepatocellular carcinoma patients were positive with the ISET method compared to 43% with RT-PCR. All controls, which included eight healthy volunteers and eight chronic hepatitis patients, were negative by both ISET and RT-PCR CTC detection methods.

    Preservative Types
    • None (Fresh)
    • Neoplastic - Carcinoma
    • Normal
    AnalyteTechnology Platform
    Cell count/volume Light microscopy
    Protein Immunohistochemistry
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Biospecimen Aliquots and Components Cell capture method ISET
    Biospecimen Aliquots and Components Cell number 1 tumor cell
    3 tumor cells
    FISH Specific Targeted nucleic acid chromosome 1 centromere
    Immunohistochemistry Specific Targeted peptide/protein Cytokeratin
    Prostate-specific antigen
    RT-PCR Specific Targeted nucleic acid α-Fetoprotein
    Prostate-specific antigen

You Recently Viewed  

comments powered by Disqus

News and Announcements

  • BRD Visitor Survey is Now Open!

  • NEW Expert-Vetted BEBP Now Available!

  • Most Popular SOPs in 2019

  • More...