Tissue Microarray Background
Applications in Oncology Research
Michael Gleeson, Darmouth Medical School
Genetics 144: Oncogenomics, February 7, 2005
Genetics 144: Oncogenomics, February 7, 2005
Fundamentals of TMA
Advantages and Limitations of Tissue Microarray Applying TMA to Cancer Research References 
(click on image to link to the Kallioniemi, Wagner, Kononen, & Sauter's
Review in Human Molecular Genetics) (1.).
Tissue Microarray technology is a novel bridging of advanced array-based approaches to data gathering with standardized Medical Pathology Laboratory practices. Every day, hospital Pathology departments process hundreds of samples, analyzing patient tissue samples for aberrant protein expression, cytogenetic abnormalities, improper tissue morphology and many other phenotypic and genotypic markers. Normally, this takes place on a patient by patient, sample by sample, test by test basis.
Tissue Microarray is a methodology developed to allow investigators to query up to 1000 patient samples simultaneously with up to a theoretical 300 assays. This is accomplished by careful loading of the "TMA block" (visualized below) with core biopsies taken from banked tissue samples (A). The TMA block accepts freshly sectioned 0.6mm diameter core biopsy samples that are arrayed into the block at 0.8mm spacing, allowing up to 1000 specimens in a typical block (B). 5um sections are cut using a microtome, thus allowing up to 300 slides to be generated for each block (C) (2). Common practice involves staining every 50th slide with hematoxylin & eosin to ensure morphology remains consistent throughout the biopsy.
(click on image above to link to Nocito, Kononen, Kallioniemi & Sauter's
Mini-Review in the International Journal of Cancer).
Despite the hi-throughput potential of TMA, the analysis and readout of the many assays may be variable. Commonly, rapid scoring by a pathologist using a brightfield microscope ("up to hundreds of tissue spots per hour...") is a standard approach (2).
Automated digital image capture is another readout, followed by pathologist scoring of the image in silico. A further evolution in the analysis of stained TMA sections is automated scoring of staining intensities and features on TMA slides using image analysis software.
In contrast to DNA Microarray, where one sample (patient) is tested for the expression of many thousands of genes or gene products, TMA approaches allow for the screening of one gene or protein across thousands of samples (patient) (1).
ADVANTAGES:
Allows use of archival tissue specimens (DNA FISH, IHC)
Employs classical clinical pathology techniques & stains (rapidly translatable in the medical community)
Expands (300) and Accelerates (100X) the # of questions one might ask of a tumor biopsy sample
Protein/DNA/Morphology changes in tumor context (cellular origin of molecular target)
LIMITATIONS:
Limited # of queries compared with DNA microarray
Population-level research tool; NOT a diagnostic tool
Labor/Sampling Intensive
Although the above listed qualities of tissue microarray are grouped into "Advantages" and "Limitations", it must be stated that this technology was not designed to assay biological material in the same way or for the same needs/applications that DNA based microarray systems were developed. Given the complexities involved in the generation of monoclonal antibodies, and the quality controls required in standard immunohistochemistry, TMA is most certainly a time-intensive tool. But it has tremendous potential for quickly mining data from banked tumor samples. It is not a diagnostic tool, but an assay that may lead to the development of better diagnostic tools by giving oncology researchers the means to rapidly analyze the significance of multiple phenotypic and genotypic changes across a spectrum of tumor samples when that data is correlated with clinical outcomes (1,2).
In an elegant example of integrated DNA Microarray discovery, TMA tumor biopsy analysis and clinical outcomes analysis, Barlund et al. followed CGH studies which described amplification of the chromosomal region 17q23 in up to 20% of primary breast tumor samples. They then went on to sequentially employ DNA Microarray, quantitative real-time PCR and finally Tumor Microarray to identify the overexpression of ribosomal s6 kinase (S6K) in 9% of primary breast tumors. The overexpression of S6K correlated with poor prognosis in the clinic (3).
In a more recent application of Tissue Microarray technology, Jacquemier, Ginestier et al. employed protein expression profiling and the power of hierarchical clustering of array data to identify subclasses of breast cancer and to predict prognosis of patients. Their scoring methodology outperformed both the NIH and St. Gallen Prognosis grading criteria in their sizeable patient population (over 1600 tumor biopsies from 552 patients)(4).
For a more complete review of Jacquemier et al, please follow the link below. Please note: a subscription to Cancer Research is required to view/access the figures, tables and text discussed.
Review of: Jacquemier, J., C. Ginestier, et al. (2005). Cancer Res 65(3): 767-79
1. Kallioniemi, O. P., U. Wagner, et al. (2001). "Tissue microarray technology for high-throughput molecular profiling of cancer." Hum Mol Genet 10(7): 657-62.
2. Nocito, A., J. Kononen, et al. (2001). "Tissue microarrays (TMAs) for high-throughput molecular pathology research." Int J Cancer 94(1): 1-5.
3. Barlund, M., F. Forozan, et al. (2000). "Detecting activation of ribosomal protein S6 kinase by complementary DNA and tissue microarray analysis." J Natl Cancer Inst 92(15): 1252-9.
4. Jacquemier, J., C. Ginestier, et al. (2005). "Protein expression profiling identifies subclasses of breast cancer and predicts prognosis." Cancer Res 65(3): 767-79.
Return to Tissue Microarray in Oncology Research Main
Return to the Oncogenomics Main Page