[Categorization of occult tumour cells in lymph nodes in patients with colon cancer not reliable enough]

Ned Tijdschr Geneeskd. 2011:155:A2697.
[Article in Dutch]

Abstract

Objective: To assess interobserver agreement between pathologists in judging photographs of lymph node preparations of occult tumour cells of patients with colon cancer.

Design: Descriptive and comparative study of interobserver variability.

Methods: All lymph nodes of 82 pN0 patients with colon cancer were analysed using three monoclonal antibodies against epithelial cells. Digital pictures of the 37 lesions detected were placed on a secured website. Forty pathologists selected at random were asked to examine the pictures and to categorize the lesions into 'micro metastases', 'isolated tumour cells' or something else. The degree of agreement was calculated by the Kendall W coefficient (with a range of 0.0-1.0).

Results: Thirty-five pathologists (88%) categorized the 37 lesions. Five lesions (14%) were categorized unanimously as micro metastases or isolated tumour cells. In 26 pictures (70%) the degree of agreement was poor to moderate. When the analysis was performed only on those diagnoses of which the pathologists were confident about their judgment, the percentage of lesions with good agreement rose to 49%. Differences in agreement were principally associated with multifocal lesions, clusters of tumour cells < 0.2 mm with proliferation characteristics in the parenchyma of the lymph node and lymphangio invasion.

Conclusion: The differentiation between micro metastases and isolated tumour cells in lymph nodes of patients with colon cancer was not uniform. If this classification has clinical relevance to colon cancer then better definitions are needed.

Publication types

  • English Abstract

MeSH terms

  • Biomarkers, Tumor
  • Colorectal Neoplasms / classification
  • Colorectal Neoplasms / pathology*
  • Humans
  • Immunohistochemistry / methods*
  • Lymphatic Metastasis
  • Observer Variation*
  • Pathology / standards*
  • Predictive Value of Tests
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Sentinel Lymph Node Biopsy

Substances

  • Biomarkers, Tumor