Evaluation of immunohistochemical markers in non-small cell lung cancer by unsupervised hierarchical clustering analysis: a tissue microarray study of 284 cases and 18 markers

Authors

  • NHC Au,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • M Cheang,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • DG Huntsman,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • E Yorida,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • A Coldman,

    1. Population and Preventive Oncology, British Columbia Cancer Agency, Vancouver, BC, Canada
    Search for more papers by this author
  • WM Elliott,

    1. iCapture Centre, St Paul's Hospital and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • G Bebb,

    1. Medical Oncology, British Columbia Cancer Agency, Vancouver, BC, Canada
    Search for more papers by this author
  • J Flint,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • J English,

    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    Search for more papers by this author
  • CB Gilks,

    Corresponding author
    1. Genetic Pathology Evaluation Centre of the Department of Pathology and Prostate Centre—Vancouver General Hospital, British Columbia Cancer Agency and University of British Columbia, Vancouver, BC, Canada
    • Department of Pathology, Rm 1438, 1st Floor JPPN, Vancouver General Hospital, 855 W 12th Avenue, Vancouver, BC, V5Z 1M9, Canada.
    Search for more papers by this author
  • HL Grimes

    1. Institute of Cellular Therapeutics and Department of Surgery, University of Louisville, Louisville, KY, USA
    Search for more papers by this author

Abstract

This study has investigated a panel of immunomarkers in non-small cell lung carcinoma (NSCLC). Unsupervised hierarchical clustering analysis was used to investigate the possibility of identifying different subgroups in NSCLC based on their molecular expression profile rather than morphological features. A tissue microarray consisting of 284 cases of NSCLC was constructed. Immunohistochemistry was used to detect the presence of 18 biomarkers including synaptophysin, chromogranin, bombesin, NSE, GFI1, ASH-1, p53, p63, p21, p27, E2F-1, cyclin D1, Bcl-2, TTF-1, CEA, HER2/neu, cytokeratin 5/6, and pancytokeratin. Univariate analysis of all 18 markers for prognostic significance was performed. Immunohistochemical scoring data for NSCLC were analysed by unsupervised hierarchical clustering analysis. Kaplan–Meier survival curves were plotted for the different cluster groups of lung tumours identified by this method. Analysis of the three different World Health Organization (WHO) subtypes (adenocarcinoma, squamous cell carcinoma, large cell carcinoma) of NSCLC individually showed that different markers were significant in different subtypes. For example, p53 and p63 were significant for squamous cell carcinoma (p = 0.007 and p = 0.03, respectively), whereas cyclin D1 and HER2/neu were significant prognostic markers for adenocarcinoma (p = 0.025 and p = 0.015, respectively). These markers were not significant prognostic predictors for NSCLC as a group. Hierarchical clustering analysis of NSCLC produced four separate cluster groups, although the vast majority of cases were found in two cluster groups, one dominated by squamous cell carcinoma and the other by adenocarcinoma. The clinical outcomes of cases from the four cluster groups were not significantly different. Prognostic indicators vary between different morphological subtypes of NSCLC. Unsupervised hierarchical clustering analysis, based on an extended immunoprofile, identifies two main cluster groups corresponding to adenocarcinoma and squamous cell carcinoma; cases of large cell carcinomas are assigned to one of these two groups based on their molecular phenotype. Copyright © 2004 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Ancillary