Ionizing radiation (IR) is the most common treatment used to control localized primary prostate cancer (PC). However, for a significant number of patients, radiotherapy fails to adequately control the tumor. Thus, a m...Ionizing radiation (IR) is the most common treatment used to control localized primary prostate cancer (PC). However, for a significant number of patients, radiotherapy fails to adequately control the tumor. Thus, a main clinical problem today is the lack of a specific marker that may be used to predict the treatment outcome and to identify prostate cancer patients who are unlikely to respond to radiation therapy. In this study, we used human PC xenografts with predetermined radioresistant/sensitive phenotypes, and gene expression microarrays, correlated their specific transcripttional profiles with response to radiation. Employing unsupervised two-way hierarchical clustering, we identified four gene clusters displaying different expression patterns. Two clusters showed higher expression levels in the resistant xenografts and the other two clusters showed higher expression levels in the sensitive xenografts. Expression levels of 113 genes differed by at least 3 fold between sensitive and resistant xenografts. These genes represent members of several cellular pathways, some of which are known to be associated with response to radiation. All or several of these genes could serve as predictive tools to determine at biopsy the expected response of a particular tumor to radiotherapy. Indeed, the profiles we identified enabled us to predict the degree of radiosensitivity of a panel of established PC cell lines. Importantly, irradiation of the PC xenografts did not induce any significant changes in gene expression, regardless of their susceptibility phenotype. These data strongly support the first of two models: a: a random effect of irradiation on a homogeneous population of cells, rather than b: of a tumor comprised of a mixture of radioresistant and radiosensitive cell subpopulations. Our findings imply that each of the radio-phenotypes represents different intrinsic characteristics that affect the ability of a tumor to survive radiotherapy.展开更多
文摘Ionizing radiation (IR) is the most common treatment used to control localized primary prostate cancer (PC). However, for a significant number of patients, radiotherapy fails to adequately control the tumor. Thus, a main clinical problem today is the lack of a specific marker that may be used to predict the treatment outcome and to identify prostate cancer patients who are unlikely to respond to radiation therapy. In this study, we used human PC xenografts with predetermined radioresistant/sensitive phenotypes, and gene expression microarrays, correlated their specific transcripttional profiles with response to radiation. Employing unsupervised two-way hierarchical clustering, we identified four gene clusters displaying different expression patterns. Two clusters showed higher expression levels in the resistant xenografts and the other two clusters showed higher expression levels in the sensitive xenografts. Expression levels of 113 genes differed by at least 3 fold between sensitive and resistant xenografts. These genes represent members of several cellular pathways, some of which are known to be associated with response to radiation. All or several of these genes could serve as predictive tools to determine at biopsy the expected response of a particular tumor to radiotherapy. Indeed, the profiles we identified enabled us to predict the degree of radiosensitivity of a panel of established PC cell lines. Importantly, irradiation of the PC xenografts did not induce any significant changes in gene expression, regardless of their susceptibility phenotype. These data strongly support the first of two models: a: a random effect of irradiation on a homogeneous population of cells, rather than b: of a tumor comprised of a mixture of radioresistant and radiosensitive cell subpopulations. Our findings imply that each of the radio-phenotypes represents different intrinsic characteristics that affect the ability of a tumor to survive radiotherapy.
