Introduction:

Several novel therapeutics have been approved for bladder cancer treatment based on empiric observations of overexpression in tumors, without much significance attributed to their overall role in altering the circuitry within cancer cells. This study employed metagenomics to interrogate previously described prognostic biomarkers to identify molecules that represent the confluence of cellular networks affected during bladder carcinogenesis and progression. The hypothesis is that central molecules may be ideal therapeutic targets regardless of differential expression as they coordinate several cellular circuits crucial to the tumor's survival.

Methods:

Prior expression profiling studies were reviewed to identify annotated transcripts associated with bladder cancer prognosis. After excluding non-mappable transcripts, curated pathway analyses were used to enumerate direct and indirect interactions between genes within the context of merged biological networks. Top ranking molecules, defined by the highest percentile of interactions, were evaluated for their value as therapeutic targets.

Results:

825 genes were used to construct an interactome. 8,776 direct and indirect molecular interactions were noted. Identified genes were percentile-ranked based on the number of direct and indirect interactions with neighboring genes within curated pathways. 78 molecules contributed to the top 90 percentile of all interactions. 27%, 14%, 19% and 28% of these molecules are typically located in the extracellular, plasma membrane, cytoplasmic and nuclear compartments, respectively. Top canonical pathways included intranuclear and death receptor signaling, T cell modulation, and interleukin signaling. Associated cellular functions included cell death, cell-cycle regulation, cell proliferation, and cellular movement. Merged analysis identified eight molecules in the top 90th percentile with known associated monoclonal antibodies or small molecule inhibitors (see Figure; top tox pathway p < 3.84E-10); several of these have not undergone clinical testing in bladder cancer. Another nine molecules that are variously influenced by chemotherapeutics were also identified.

Conclusion:

This study describes a metagenomic approach using prior prognostic biomarkers that may be used to identify druggable targets in bladder cancer. This paradigm may be used to investigate novel combinatorial therapies, potentially in the context of adaptive trials, which are aimed at consensus targets while minimizing the probability of tumor escape.

Funding: N/A