Abstract Despite recent advances in the treatment of patients with pancreatic adenocarcinoma (PDAC), survival remains low. Pre-clinical models of human malignancies often fail to capture the complexity of tumors and as such lack clinical predictive power. Patient-derived tumor xenografts (PDXs) are an emerging pre-clinical platform, capable of retaining the molecular heterogeneity of their originating sample and able to predict the response to therapies in some cases. However, the lack of immune system of the mouse PDX host is a major drawback for a preclinical model, especially with the advent of promising cancer immunotherapies (e.g. anti-PD-1/PD-L1, CXCR4 antagonist). Genetically Engineered Mouse models (GEMMs) carry mutations in genes that are associated with the human disease and can thus mimic the genetic, phenotypic and physiological aspect of the human disease, in an immunocompetent background. The Lox-Stop-Lox (LSL)-Kras(G12D); LSL-Trp53(R172H); Pdx1-cre (KPC) mouse is a favoured model of PDA, recapitulating the desmoplastic tumor histology, immune profile and drug resistance of human PDAC. However, KPC tumor development speed is variable and tumor monitoring requires ultrasound imaging, so therapeutic studies are long, complex and expensive. We describe here a new model, Genetically Engineered mouse-Derived Allograft (GEDA), that applies the methods developed for PDXs to KPC tumors. Pieces of KPC tumor were implanted subcutaneously in the flank of recipient LSL-Trp53 (R172H); Pdx1-cre (PC), immunocompetent mice. At primary passage, 10 mice were implanted from a single primary tumor, and the engraftment rate is 90%. Tumors are palpable 2 weeks after the tumor implantation and can be monitored by calliper measurement for up to 10 weeks, with a tumor doubling time of 14.4 ± 1.5 days. The incidence of metastases was similar to KPC mice, but organ distribution was different, with 57% of 45 recipient mice developing lung metastases, and 0% liver metastases. Histologically the GEDAs appeared more like primary KPC tumors than subcutaneous allograft of KPC cell lines, with intratumoral heterogeneity including desmoplastic regions identified by the presence of collagen fibres and αSMA-expressing cells. Implantation of GEDAs onto GFP-mice revealed that the original donor stroma disappeared within 2 weeks and was replaced by host stroma, but the stromal composition remained comparable even over 5 passages in vivo, with ±15% of variation in αSMA-expressing area between donor and recipients. Typically, around 20% of the GEDA cells were positive for Ki67, remaining stable over 5 passages, suggesting that aggressiveness of the tumor was stable. Qualitatively, by CD3 IHC, the T cell distribution is similar to that reported in KPC tumors (Feig et al, PNAS 2013) with T cell exclusion from tumor cell nests. A therapeutic study was performed with gemcitabine. Tumor growth was not inhibited by gemcitabine (100 mg/kg IP, twice per week); tumor volume change at day 60 (%): 1544 ± 263 (treated) vs. 1350 ± 156 (control), n= 6 per group, p=0.25 ). Few KPC primary tumors show significant response to gemcitabine, so the PDA GEDA was more like KPC than subcutaneous allografts of KPC cell lines which tend to be gemcitabine sensitive. A biobank of KPC GEDAs is now being generated, which will be tested to determine the heterogeneity of therapeutic response to gemcitabine and other agents.In conclusion, we have generated PDAC GEDAs, which combine advantages of both GEMMs and PDXs, cost-effective generation of cohorts of mice for therapeutic studies. The stromal content and intact immune system should enable the evaluation of immunomodulatory therapies and also agents targeting the stromal components of PDAC. Citation Format: Séverine Mollard, Kristopher K. Frese, Aarthi Gopinathan, Frances M. Richards, Duncan I. Jodrell.{Authors}. Genetically engineered mouse-derived allografts (GEDAs): an immunocompetent mouse model of PDAC for the evaluation of novel therapeutic strategies. [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Advances in Science and Clinical Care; 2016 May 12-15; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2016;76(24 Suppl):Abstract nr B54.