Molecular indicators of tumor progression
Institute for Pathology, Cantonal Hospital Basel, University Clinics, Basel, Switzerland
AIM: Morphologic features alone are often imprecise in predicting the biologic behavior of cancer. Therapy decisions in patients with prostate cancer are complicated by two major dilemmas. First, it is not sufficiently possible to differentiate those tumors that will behave aggressively - and thus require radical treatment – from slowly growing tumors, where watchful waiting may be a choice to avoid overtreatment. Second, there is no effective therapy against late stage hormone-refractory cancers that have escaped the growth suppression of hormone withdrawal. The molecular mechanisms that drive progression and hormone-refractory growth are poorly understood. High-level DNA amplifications can pinpoint to genes that drive tumor progression and may serve as new therapeutic targets. Here, we analyzed the role candidate markers during tumor progression and their potential as therapeutic targets in prostate cancer. METHODS: A progression tissue microarray was constructed to study the prevalence of new molecular markers across the different stages of prostate cancer. It contains samples from 535 prostate specimens including benign prostatic hyperplasia as controls, prostatic intraepithelial neoplasia (PIN), clinically localized prostate cancers, as well as hormone-refractory local recurrences and distant metastases. The expression levels of 13 different proteins were analyzed using immunohistochemistry (Bcl-2, p53, ILK, Syndecan-1, MUC-1, EGFR, HER2/neu, HSP-90, Ep-CAM, MMP-2, CD-10, CD-117, Ki67). FISH with a BAC probe was used to analyze the prevalence of the DNA amplification of KCNMA1, a potassium channel that may be a driving force of the amplification at 10q22 in late-stage prostate cancers. RESULTS: Significant overexpression in hormone-refractory prostate cancer and metastatic tissue compared to localized prostate cancer was found for Ki67 (64% vs. 9%), Bcl-2 (11% vs. 1%), p53 (35% vs. 4%), Syndecan-1 (38% vs. 3%), EGFR (16% vs. 1%), and HER2/neu (16% vs. 0%). Overexpression of CD-117 was restricted to 1 single metastasis. All other markers did not show relevant differences in expression between subgroups. Similarly, FISH showed DNA amplification of KCNMA1 in 13.5% of 141 hormone-refractory and metastatic prostate cancers, but not in untreated primary tumors. Quantitative RT-PCR revealed a marked overexpression of KCNMA1 mRNA in the hormone-insensitive cell line PC-3 with KCNMA1 amplification as compared to non-amplified control cell lines. CONCLUSION: p53, Bcl-2, Syndecan-1, EGFR and HER2/neu are preferentially expressed in hormone-refractory and metastatic prostate cancer. KCNMA1 appears as an attractive amplification target gene at 10q22 to drive progression in late-stage prostate cancer. Selected inhibition of these targets might offer a strategy to treat advanced tumors and prevent further progression. Treatment decisions should not be based on findings in primary tumors but rather on tissues from recurrent or metastatic lesions. Basel, 21/12/03 L. Bubendorf, M.D.
Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Nice, France; February 7 - 10, 2004; in plenary session 803 (Risk & assessment).