Abstract The ability to profile the genome and transcriptome of bladder cancer on a global scale has fundamentally changed the way bladder cancer is diagnosed, treated, and prognosticated. As large amounts of information continue to accrue from the whole-genome/exome/transcriptome analyses of human materials, is genetic modeling of bladder cancer still necessary and useful? The answer is clearly an affirmative one based on the following reasons. First and foremost, we continue to need genetically tractable experimental systems to validate the biologic potential of putative tumor drivers, distinguish them from the passenger abnormalities, and ascertain whether they are initiators of tumor formation or promoters of tumor progression. There are ample examples where recurrent mutations identified in humans by themselves completely lack tumorigenicity and instead require specific collaborative events to initiate tumors. Perhaps the best example relates to the mutations that activate the receptor tyrosine kinase (RTK)-RAS-PI3K signaling pathway. Despite the fact that most if not all non-muscle invasive bladder cancer (NMIBC) harbors at least one mutation in this pathway, none of the mutations affecting different genes (e.g., FGFR3, RAS, PIK3CA, or PTEN deletion) is overtly tumorigenic in mouse models. By systematically dissecting the biologic effects of the loss of several individual genes residing on 9p21, my laboratory has obtained new data (which I will discuss), implicating the specific cooperative events that are both necessary and sufficient to initiate NMIBC during the activation of RTK-RAS-PI3K pathway. The lack of tumorigenicity of p53 mutation/deletion, alone or in combination with RB1 loss, in the genesis of MIBC is another notable example. For these reasons, additional studies are therefore required to pin down the driver events underlying NMIBC versus MIBC. Second, because bladder cancer is primarily a carcinogen-induced cancer, we continue to need model systems to investigate how lifestyle-associated carcinogens, such as tobacco smoke and e-cigarette smoke, and environment-associated carcinogens, such as aromatic amines, acrolein, and arsenic, inflict damages on our genome and epigenome. Bladder cancer is likely a result of long-term exposure to an assortment of carcinogens, but our understanding about the changes at the carcinogen-genome interface remains extremely scarce. The model systems should be highly instrumental in allowing us to better assess risks in carcinogen exposure and informing us about effective prevention strategies. Third, some of unresolved issues regarding the cell(s) of origin of bladder cancer variants and subtypes and the bases underlying the inter- and intratumor heterogeneity can be greatly facilitated by lineage-specific targeting and tracing in mouse models. Fourth, we continue to need a toolbox with which to critically evaluate new concepts and strategies in diagnosis and treatment. In the post-genome era, the demand for human-relevant model systems in this area will be considerably more, not less, than before. Models worthy of particular attention include those that can be used to test chemosensitivity, immunotherapeutics, and combination-therapy modalities. Finally, the continued need for genetic model systems will be shaped by more sophisticated modeling strategies and methods. For instance, the modelers will likely prioritize genetic and epigenetic alterations that are highly prevalent in human specimens over less prevalent ones, conditional systems over constitutive ones, and lineage-specific targeting over whole-urothelial targeting. There will also likely be an increased use of genome editing techniques, such as CRISPR/Cas9, in creating somatic models of bladder cancer, instead of just germline mutations. Citation Format: Xue-Ru Wu. Genetic modeling of bladder cancer in the post-genome era [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2019 May 18-21; Denver, CO. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(15_Suppl):Abstract nr IA18.
Read full abstract