Preclinical Evaluation of Genomic-Based Therapies in Pancreatic Cancer and Glioblastoma
Location
Steelcase Lecture Hall
Description
INTRODUCTION: The focus of this study is the testing of biomarker-driven analytical methods to identify targeted therapies in pancreatic cancer and glioblastoma, which are highly invasive and metastatic cancers with poor outcomes and few treatment options. The objective was to make treatment predictions based on the molecular signatures of pancreatic cancer and glioblastoma samples, then to evaluate the efficacy of these therapies using preclinical models. METHODS AND MATERIALS: XenoBase Bio-Integration Suite (XB-BIS) in an informatics platform for the analysis of molecular data using Personalized Medicine (PMED) algorithms. PMED applies four independent methods (Drug Target Expression, Connectivity Map, Parametric Gene Set Enrichment, and GeneGo Network Topological Enrichment Analysis) to a genomic dataset to identify targeted therapies. Affymetrix data was collected from panels of pancreatic cancer cell lines and human glioblastoma specimens and analyzed in XB-BIS to predict therapies, which were evaluated in vivo. RESULTS: Treatment of mice with subcutaneous pancreatic tumors with Chlorpromazine, predicted by CMAP, resulted in a decrease in tumor volume and extended survival compared to control animals. Predictive algorithms identified BCNU, Doxorubicin, and Marimastat as potential treatments for glioblastoma. Combination treatment of mice implanted intracranially with U251 glioblastoma cells showed extended survival compared to control mice and similar survival to standard-of-care treatment, Temozolomide. CONCLUSIONS: We have demonstrated efficacy of therapies identified by the PMED approach in relevant models of pancreatic cancer and glioblastoma. While further investigation is needed, these therapies could prove to be a great resource against two devastating human diseases.
Preclinical Evaluation of Genomic-Based Therapies in Pancreatic Cancer and Glioblastoma
Steelcase Lecture Hall
INTRODUCTION: The focus of this study is the testing of biomarker-driven analytical methods to identify targeted therapies in pancreatic cancer and glioblastoma, which are highly invasive and metastatic cancers with poor outcomes and few treatment options. The objective was to make treatment predictions based on the molecular signatures of pancreatic cancer and glioblastoma samples, then to evaluate the efficacy of these therapies using preclinical models. METHODS AND MATERIALS: XenoBase Bio-Integration Suite (XB-BIS) in an informatics platform for the analysis of molecular data using Personalized Medicine (PMED) algorithms. PMED applies four independent methods (Drug Target Expression, Connectivity Map, Parametric Gene Set Enrichment, and GeneGo Network Topological Enrichment Analysis) to a genomic dataset to identify targeted therapies. Affymetrix data was collected from panels of pancreatic cancer cell lines and human glioblastoma specimens and analyzed in XB-BIS to predict therapies, which were evaluated in vivo. RESULTS: Treatment of mice with subcutaneous pancreatic tumors with Chlorpromazine, predicted by CMAP, resulted in a decrease in tumor volume and extended survival compared to control animals. Predictive algorithms identified BCNU, Doxorubicin, and Marimastat as potential treatments for glioblastoma. Combination treatment of mice implanted intracranially with U251 glioblastoma cells showed extended survival compared to control mice and similar survival to standard-of-care treatment, Temozolomide. CONCLUSIONS: We have demonstrated efficacy of therapies identified by the PMED approach in relevant models of pancreatic cancer and glioblastoma. While further investigation is needed, these therapies could prove to be a great resource against two devastating human diseases.