Microarray analysis of factors determining apoptosis in human rectal cancers
01 / 2002 - 01 / 2006
Website Nederlandse Kankerbestrijding
PURPOSE: Treatment of cancer patients with chemo- and radiotherapy is aimed at the induction of adequate tumor cell death, which is hampered by intrinsic or acquired resistance of the tumor. The cytotoxic activity of anti-cancer agents on tumor cell lines is often manifested by an apoptotic response. We demonstrated the relevance of apoptosis for the prognosis of patients by showing that apoptotic levels in rectal tumors correlate with the local recurrence-free interval and with survival of the patients. This was achieved by employing a unique collection of pre-operatively irradiated rectal tumors and control tumors from patients treated in a randomized trial coordinated by the Dutch Colorectal Cancer group. The purpose of this grant proposal is to acquire a comprehensive overview of genetic factors that determine intrinsic and radiation-induced tumor cell apoptosis in vivo. We will make use of the above-described collection of more than 1500 clinically evaluated rectal tumor samples, of which a proportion has been already analyzed for several apoptotic and oncogenic markers. Apoptosis is regulated by a large number of genes that are predictable targets for mutations during malignant progression and thereby contribute to treatment failure. The identification of these genes in specific tumor types will provide insight into the mechanism underlying this process and lead to improvements in prevention and treatment. The systematic discovery of apoptotic factors will ultimately make the development of tailor-made therapeutic interventions possible. PLAN OF INVESTIGATION: In order to achieve the above goal, we will make use of expression profiling by microarray analysis. Preliminary results have shown that frozen rectal tumors from this sample collection are suitable for microarray analysis. 1. We will start with the expression analysis of pools of RNA samples derived from tumors grouped according to their apoptotic phenotype (relatively high or low incidence of apoptosis) from either irradiated or non-irradiated patient cohorts. To this aim, commercially available genome-wide oligonucleotide microarrays will be employed. 2. Clustering of the expression patterns according to treatment, apoptotic response and clinical outcome will allow the identification of a presumably very large set of genes whose differential expression is associated with (treatment-induced) apoptosis and clinical results. 3. The resulting data will be used to develop smaller cDNA arrays to be employed for the analysis of the expression profiles of a larger series of individual tumors (>200). Again, state-of-the art statistical techniques will be employed to pinpoint genes or groups of genes whose coordinated expression profiles correlate with intrinsic apoptosis, radiation-induced apoptosis and improved clinical outcome. 4. A selection of the newly identified genes will be the subject of both functional and preclinical studies. Promoter and expression analyses will be conducted both in vitro and in vivo to determine their identity as primary radiation-induced targets or downstream targets of activated apoptotic pathways. Based on these results, we will determine which of these genes are applicable as (routine) prognostic markers and which are potential targets for the development of future cancer therapies. POSSIBLE RESULTS / RELEVANCE FOR CANCER RESEARCH: The proposed comprehensive investigation of genes differentially expressed in tumors according to the incidence of apoptosis will provide ample information on the molecular mechanisms of (treatment-induced) apoptosis in vivo, and possibly result in the identification of new apoptotic pathways. We will investigate which of these genes/pathways represent potential predictive markers and which may be employed as targets for therapeutic intervention. Improved prognostic markers will allow the selection of patients who may benefit from (radio)therapy and facilitate their clinical management. As a large collection of clinical parameters (grade, staging, occurrence of local residual disease and distant metastases, survival) relative to this tumor cohort is readily available, the microarray-derived expression data will also lead to the improved classification of sporadic rectal cancers based on gene expression profiles.