Abstract 2935: Defining variation within a next gen humanized immune system mouse model of human lung cancer in the context of a checkpoint inhibitor efficacy study

Abstract

Humanized immune system mice are critical immuno-oncology tools, and next gen models using human cytokine-expressing mice, such as the NOG-EXL expressing hGM-CSF and hIL-3, drive differentiation of specific cell types. The huNOG-EXL (engrafted with human CD34+ hematopoietic stem cells) is widely used as it develops both myeloid and lymphoid lineages. Our goal was to determine the degree of variance between stem cell donors, tumor types, and immune cell reconstitution (tissue, phenotype and/or kinetics) within a huNOG-EXL lung cancer model. We studied tumor growth kinetics, human immune reconstitution, and tumor infiltrating leukocytes (TIL) in a human cell-derived model of non-small cell lung cancer (NSCLC; A549) and a patient-derived model of small cell lung cancer (SCLC; LU5173). Method: huNOG-EXL were inoculated (n=5 x 2 donors x tumor) with A549 (expanded in vitro to 5x106 cells per inoculum) or LU5173 (P5, warm tumors 600-1000 mm3 total size derived 3 mm3 per inoculum) in the flank subcutaneously (SQ). Tumor growth and body weight were measured for 80+ days. The main study used n=20 huNOG-EXL per donor across three different donors and both tumors for a total n=120. Mice were randomized at a tumor volume of 80-120 mm3 (A549) or 150-250 mm3 (LU5173) into a vehicle (human IgG4) or test article group (Pembrolizumab; humanized anti-PD1 antibody). 24 h later, mice were given (BIW x 7) drug or vehicle (10 mg/kg; 5 ml/kg; IP). 24 h after final dose, animals were sacrificed and blood, spleen, and tumor collected for a 2-panel 16 color FACS analysis. Results: Notable differences were seen between tumor types in slope and late-stage impact on body weight. A549 had accelerated growth but at late time points (day 70) tumor volume and body weight declined for both donors. LU5173 had delayed growth but larger sustained volume sizes, with no adverse effects on animal health or tumor volume reduction in both donors. In the main study, aAcross all three donors, huNOG-EXL reconstituted human lymphoid and myeloid lineages in blood, spleen, and tumor at ~23 wks post engraftment. Donor differences were seen in the overall reconstitution and distribution across the 16 cell types measured. There was a significant difference in the extent of human immune cell infiltration between tumor types. A549 tumors had significantly more hCD45+ cells than LU5173 tumors. The tumor growth inhibition did not significantly differ between treatment groups. Conclusion: Our study provides a comprehensive look at a HIS mouse that reconstitutes multiple human lineages in the context of a human lung cancer model. The significant power and large-scale FACS analysis in multiple tissues demonstrates the impact that tumor type, mouse strain, timeline, and donor can impart on an in vivo efficacy study. Proper power analysis accounting for these is required for successful immuno-oncology studies. Citation Format: Janell R. Richardson, Megan MacBride, Esther Andino, Marla Wilwol, SueEllen Bethmann, Taylor Knapp, Emily Shako, Paul Volden. Defining variation within a next gen humanized immune system mouse model of human lung cancer in the context of a checkpoint inhibitor efficacy study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2935.