Abstract B49: Using ionizing radiation and primary human esophageal adenocarcinoma xenograft models to interrogate tumor cell characteristics associated with tumor-initiating cells

Abstract

Abstract Radiotherapy is critical to the treatment of esophageal adenocarcinoma (EAC). However, with frequent recurrence and metastasis, five-year survival rates are only 13%. Clinical radioresistance is further evidenced by the fact that escalated doses increase toxicity without improving outcome. The cancer stem cell model may provide insight on the EAC cell of origin and thus, on clinically-observed radioresistance. Recent evidence has demonstrated a differential ability of EAC cells to seed tumors in mice, however a correlation between tumorigenicity and radioresistance in EAC has not been reported. Recently, sequencing and cross-checking of 14 commercially-available EAC cell lines revealed that these cell lines were in fact derived from other cancer types. These findings highlight the need for reliable models of EAC. We have developed mouse xenograft models of primary human EAC that recapitulated the original tumors in cellular differentiation and tumor architecture. Using these models, we compared the in vivo tumorigenicity of irradiated versus non-irradiated (control) tumors. We hypothesized that ionizing radiation would enrich the population of tumor-initiating cells (TIC) in xenograft tumors compared to non-irradiated tumors. Thus far, we are evaluating five primary xenograft models of human EAC that have been established in NODSCID mice. Treatment sensitivity testing has been performed on these five xenograft lines, and limiting dilution assays (LDA) have been performed on the first two. Initial experiments have revealed that xenograft Line 2 is significantly more radioresistant than the other four lines. LDA results on xenograft Line 1 revealed a TIC frequency of 1 in 20,089 in non-irradiated xenograft tumors and 1 in 40,620 in irradiated xenograft tumors. Line 2 revealed a TIC fraction of 1 in 434 non-irradiated tumors and 1 in 31,631 irradiated xenografts. These results suggest two Conclusions: (1) TIC frequency may be associated with relative radioresistance, since the radioresistant Line 2 appeared to be very tumorigenic; and (2) the comparison of TIC fractions in non-irradiated and irradiated xenograft tumors raises concerns that infiltrating mouse cells may dilute the TIC frequency through a radiation-induced inflammatory response. In subsequent LDAs, cells positive for the mouse-specific antigen H2k were depleted from the tumour cell suspension. Results from these LDAs are forthcoming, and will be presented at the meeting. In conclusion, preliminary LDA results have shown that the baseline TIC frequency in EAC xenografts is highly variable and might be related to radiosensitivity; our results suggest that this variability can be several orders of magnitude wider than previously reported. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the Second AACR International Conference on Frontiers in Basic Cancer Research; 2011 Sep 14-18; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2011;71(18 Suppl):Abstract nr B49.