ISPO

Microarray assessment: bladder & prostate

Lukas Bubendorf, Tobias Zellweger, Ronald Simon, Thomas C. Gasser, Guido Sauter

Institute for Pathology and Urologic Clinics, University of Basel, Switzerland

AIM: New microarray technologies and public human genome resources offer unprecedented opportunities to uncover the molecular mechanisms of cancer to better predict the progression risk and develop new therapeutic targets. DNA microarray studies have already identified a high number of new genes that may be involved in the development or progression of urological cancers. To explore the prevalence and clinicopathological associations of potentially important genes, a high number of tumors across various stages of progression need to be analyzed. Here, we summarize our recent results from marker analyses in prostate and bladder cancer using tissue microarrays. METHODS: To facilitate large-scale molecular analyses in hundreds of tumors simultaneously, we have recently developed the tissue microarray (TMA) technology. TMAs are paraffin blocks containing up to 1000 cores of individual cancer tissues for simultaneous in situ analyses on TMA sections. TMAs do not only facilitate large-scale studies, but also preserve precious tissue blocks, that would otherwise be used up after only a few studies. We have constructed several types of prostate TMAs so far. For example, we have generated a prostate progression TMA with >350 specimens across all stages of progression including benign controls, PIN, primary localized and advanced cancers, hormone-refractory local recurrences and distant metastases. Prognostic arrays were designed to test the prognostic significance of new molecular markers in >600 prostate and >2000 bladder cancers from patients with long-term clinical follow-up. Marker analyses were performed using fluorescence in situ hybridization (FISH) for the analysis of gene copy numbers or amplifications and immunohistochemistry for the analysis of protein expression. RESULTS: Combining cDNA microarray experiments in the CWR22R xenograft model system with TMA analysis allowed us to rapidly identify genes that are associated with hormone-refractory growth not only in vitro but also in vivo such as IGFBP2. The utility of TMA for efficient FISH analysis was demonstrated by a survey of gene amplifications in prostate cancer. Androgen receptor was amplified in 22% of metastatic and hormone-refractory tumors but almost never in primary prostate cancers, while MYC amplification was preferentially found in distant metastases. Immunohistochemistry on TMA is increasingly used to explore the significance of new molecular markers. For example, we observed that loss of NKX3.1 and Annexin 7 expression is strongly associated with advanced prostate cancers, suggesting a tumor-suppressor function of these genes. Using a prognostic prostate TMA, we were able to fully reproduce the significance of known prognostic markers (e.g. Gleason grade, Ki67 and p53) and suggest a prognostic role of additional genes (e.g. MUC1) in only a few experiments. Similarly, grade and Ki67 information obtained from sections of 4 replicate blocks of a bladder cancer TMA showed almost identical survival curves for each replicate TMA. The results were also highly concordant with large section analysis, suggesting that the small size of the specimens does not significantly affect the ability to detect clinico-pathological associations on TMAs. CONCLUSION: Tissue microarrays are about to become a standard tool in cancer research allowing a massive acceleration of studies correlating molecular in situ findings with clinico-pathological information.

For more information, contact lbubendo@bluewin.ch

Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Paris, France; February 9 - 12, 2002; in the section on Cancer Genomics.

http://www.cancerprev.org/Journal/Issues/26/101/902/4189