Abstract 4483: Establishing Patient-Derived Colorectal Cancer Stem Cell Models with a PIK3CA Mutation for the Development of Inhibitory Drugs As Targeted Therapies

Abstract Emerging evidences suggest that cancer stem cells (CSC) may be critically responsible for tumor initiation, progression, metastasis, and drug resistance. It becomes important to ask whether anti-cancer agents are able to target the tumor-initiating subpopulation in relevant CSC models. In this study, we first established xenograft tumors in NOD/SCID mice from a colorectal cancer patient specimen and demonstrated that CD133/EpCAM-expressing CSC population was highly tumorigenic. We then sought to propagate the CSC population under a serum-free condition. In culture, the tumor cells formed non-adherent, 3-dimensional spheroids, a fraction of which retained expression of the CSC markers. When exposed to a serum-containing medium, tumor spheroid cells differentiated into epithelial-like adherent cells with an increase in cell proliferation rate. In comparison with the differentiated progeny, tumor spheroid cells exhibited resistance to the standard-of-care agent oxaliplatin and, in limiting dilution assays in mice, displayed substantially higher tumorigenic potential. In contrast to the tumors originated from the differentiated cells, tumor spheroid cell-derived tumors recapitulated not only the CSC frequency marked by CD133/EpCAM expression, but also the histological characters of the original tumor. Similarly, only were the fragments of spheroid cell-derived xenograft tumors capable of regenerating highly proliferative tumors in secondary transplantation. Thus, the tumor spheroid culture is indeed enriched of drug resistant, self-renewing, and tumor-initiating CSC populations. Mutation profiling of frequently mutated oncogenes using Sequenom OncoCarta™ panel identified a mutation in the kinase domain of PIK3CA (H1047R) in the cultured CSCs. This mutation has been reported present in a large number of colon cancer patients and likely functions as an oncogene (Samuels et al., Science 304:554; 2004). We further demonstrated that a dual mammalian target of rapamycin (mTOR)/phosphoinositide 3-kinase (PI3K) inhibitor (PF-04691502) exhibited a more potent effect on inhibition of in vitro proliferation of the mutated CSCs compared to the chemotoxic agent oxaliplatin. Collectively, our findings suggest that CSC models provide a novel avenue to drug sensitivity and efficacy studies. The well-characterized CSC model systems may assist in the development of more effective therapy against the subpopulation of tumors driven by the CSCs bearing specific mutations. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 4483.