A widespread reduction to practice for an approach that employs genetically engineered mouse (GEM) cancer models for more predictive cancer drug efficacy evaluation is largely impeded by significant costs intrinsic to operations involving GEM animals but also due to frequent shortage of appropriate expertise in supporting complex genetics projects. To tangibly streamline the most laborious and costly components of preclinical studies (that, for example, involve manipulations with a sizeable breeding colony of GEM models and frequent genotyping to identify animals with appropriate allelic combinations) we will construct and characterize a library of ES and iPSC lines from a collection of well characterized and proven to be clinically relevant GEM models or carcinogenesis, including those for non-small cell lung, brain, ovarian serous epithelium, and melanoma cancer types. In order to achieve optimal outcomes in the efficiency of iPSC derivation while maintaining sufficient genetic diversity of an input cell population subject to in vitro reprogramming, we will apply an adenoviral vectors gene transfer based technology whereby recombinant viruses transiently express multicistronic self-processing polypeptides upon infection of genetically divergent pools of cells. The latter will in turn be derived from fetal tissues collected from embryos that belong to different litters to assure genetic diversity stipulated by random segregation of alleles in mixed background crosses typical for many multi-allelic GEM models. Resulting iPSC clones (at least five per model) that bear desired set of modified loci will undergo extensive in vitro characterization prior to producing cohorts of chimeric animals, now designated as non-germline GEM (NG-GEM) models, to be used both in preclinical drug evaluation studies and to conduct experiments on basic cancer mechanisms. In the first set of studies, the chimeric animals will be exploited for the purpose of tumorigenesis induction in specific tissues, such as brain or ovaries, and subsequent evaluation of tumors by patho-histology and gene expression techniques vis--vis similarly induced tumors appearing in conventionally bred animals. Selected subsets of chimeric animals will be further subjected to a pharmacological intervention with combinations of therapeutics proven to be efficacious in reverting the tumor growth in de novo cancer mouse models. As above, representative clinical, histological and molecular analyses will be performed to comparatively assess anti-cancer drug response in iPSC-derived chimeric animals employing as baseline available data gathered in a similar set of assays from conventionally bred tumor-bearing mice. Such comprehensive cross-evaluation of two similar though technologically divergent approaches is intended to further validate the robust and cost-conscious strategy of preparing preclinical cohorts of tumor bearing animals based on iPSC-derived non-germline GEM models. To extend the benefits of constructed iPSC library, in another set of in vivo experiments, several cohorts of iPSC derived NG-GEMs developed with different iPSC clones (though bearing an identical combination of inducible oncogenic alleles) will be brought in a comparative juxtaposition to identify variations in carcinogenesis outcomes, putatively driven by strain-specific dissimilarities and/or impact of modifier(s). Direct genetic analyses of iPSC clones will be undertaken to identify such modifier loci.
