Looking for factors predicting outcome of breast carcinoma using tissue microarrays

PhD ceremony: Mr. B. van der Vegt, 14.30 uur, Academiegebouw, Broerstraat 5, Groningen

Dissertation: Looking for factors predicting outcome of breast carcinoma using tissue microarrays

Promotor(s): prof. H. Hollema, prof. G.H. de Bock

Faculty: Medical Sciences

'Classical' prognostic indicators like tumour size, tumour grade, estrogen- and Her2/neu receptor expression adequately predict clinical outcome for large groups of breast carcinoma patients, but poorly predict outcome for the individual patient. The tissue microarray (TMA) technique is one of the techniques introduced in recent years to quickly assess the expression of many genes or proteins in one patient to search for better predictors of breast carcinoma. With TMAs it is possible to compile material of tumours of up to 300 patients into one paraffin block. This technique was applied to a series of ductal in situ carcinomas (DCIS) and a series of invasive ductal carcinomas (IDC). We evaluated expression patterns of a candidate biomarker, MUC1, and found differences in outcome and tumour characteristics, suggesting that MUC1 expression patterns reflect tumour differentiation. Screen detected- and interval breast carcinomas were evaluated using TMAs. Screen-detected DCIS showed more markers of tumour biological aggressiveness, suggesting that screen-detected DCIS should not be regarded as over-diagnosis. Interval IDC were larger, without showing signs of increased tumour biological aggressiveness, suggesting that part of the tumours found in the screening are biologically somnolent. TMAs were also used to successfully validate a new anti-Her2/neu monoclonal antibody in breast carcinoma patients. We conclude that the TMA technique can be used to validate 'key' markers from genome wide arraying studies in larger cohorts and may aid the development of clinically applicable prognostic scores to improve 'tailored' treatment for a specific breast carcinoma patient.