gms | German Medical Science

In the development of breast cancer, the transition of DCIS (ductal carcinoma in situ) to IDC (invasive ductal carcinoma) is followed by hematogenic metastasis which is the major cause of mortality from malignant tumours. Therefore, it is imperative to understand the molecular events accompanying this crucial step to improve the classification in high- and low-risk groups for DCIS patients according to risk of relapse or progression. In order to identify new prognostic markers and drug targets correlated with breast cancer progression, our group combined Affymetrix microarray analysis with laser capture microdissection (LCM) to analyze matched DCIS and IDC tissue from nine patients. Subsequent to extensive statistical analysis (rank product, paired t-test, paired Mann-Whitney test) 548 Affymetrix probe sets were identified to be differentially expressed in DCIS and IDC of the nine tumors (446 up-regulated and 102 down-regulated in IDC, p-value: 0.01). Examples for known genes already associated with breast cancer invasion are PLAU/uPa and MMP11, confirming the robustness of the approach. Differential gene expression of 18 candidate genes (15 up-regulated and 3 down-regulated in IDC, respectively) was validated performing triplicate real-time PCR experiments. To localize expression, 4 PCR-validated candidates were selected for in situ hybridization experiments and additionally, proteins of 5 PCR-validated candidates were detected using immunohistochemistry. Results of both methods confirmed the localization for all transcripts and proteins to epithelial tumour cells (representative example in Figure 1 [Fig. 1]). In conclusion, we identified progression-specific candidate genes using an accurate approach, combining LCM and microarray analysis. New candidates genes not yet correlated with breast cancer progression, could be validated using real-time PCR. Transcript and protein of selected candidate genes could be localized to the epithelial tumour cells, respectively. Among them are transmembrane receptors, transcription factors and tumour antigens, thus gene products with potential clinical importance.