COMBINED HEAT SHOCK PROTEIN 90 AND KINASE INHIBITOR THERAPY: INSIGHTS FROM THE CANCER GENOME ATLAS FOR MUSCLE INVASIVE BLADDER CANCER

INTRODUCTION AND OBJECTIVES: The Cancer Genome Atlas (TCGA) for muscle invasive bladder cancer (MIBC) highlights the contribution of kinases and the heat shock protein 90 (HSP90) in the biology of the disease. Small molecule inhibitors of HSP90 (H90Ins) or oncogenic kinases (KI) have shown limited efficacy as monotherapy in clinical trials. This is due to the development of drug resistance resulting in oncogenic reprogramming. Studies in other tumor models have demonstrated that this can be mitigated through combined H90Ins and KI therapy. Therefore, the purpose of this study is to explore the rationale for combined H90Ins and KI therapy in MIBC. METHODS: Using the HSP90 interactome network database, we identified all the HSP90 interacting genes and profiled these genes in the TCGA for MIBC. We determined the number of interactors that were either overexpressed or mutated in MIBC. These results were subjected to gene ontology analysis. Next, we focused on the HSP90 interacting kinases overexpressed in MIBC and identified an up to date set of clinically actionable targets. RESULTS: Over 130 HSP90 interacting genes were mutated or overexpressed in 10% of the 408 MIBC patients in TCGA. The top 4 mutated genes included TTN at 18%, TP53 at 16%, EP300 at 5% and SPTAN1 at 4% of all patients. The top 100 overexpressed genes observed in 11 to 28% of cases included TTI1, YWHAZ, RAF1 and SETBD1. Comparing the number of patients with mutations to gene overexpression suggests that mRNA expression levels may be the more significant driver of disease. From gene ontology analysis, the biology of MIBC appears to involve alterations in macromolecule metabolism and the regulation of nitrogen compounds. Interestingly, when the oncogenic kinases were specifically profiled, all patients overexpressed at least one HSP90 dependent kinase. In over 40% of the cases, more than one HSP90 dependent kinase was overexpressed. Of the HSP90 dependent kinases combined, RAF1, AKT2, EGFR, YES1, GSK3A PRKAA1 and PTK2 were seen in 68% of the cases and represent targets for which drugs are currently available. CONCLUSIONS: While the altered HSP90 gene network have a modest impact on the biology of MIBC, the more profound influence is exerted through the control of the oncogenic kinases involved in the disease. This data strongly supports the use of combination HSP90 and kinase inhibitor therapy in a patient specific and precision medicine model.