Abstract 5337: Profiling of human CD34+ donors in a myeloid-supportive HIS model provides increased predictability and facilitates study design for evaluation of immuno-oncology therapeutics

Abstract Background: Humanized immune system (HIS) mice are critical immuno-oncology tools for in vivo evaluation of therapeutic target engagement and efficacy. HIS mice require a human donor source of CD34+ hematopoietic stem cells to provide the human immune cell system. However, variance across donors complicates study design and hypothesis testing. To determine if parameters critical to drug discovery are consistent across batches of HIS mice from the same donor, we evaluated two independent cohorts of human tumor xenograft-bearing mice made using a common set of five different CD34+ cell donors. Our study included evaluation of tumor growth kinetics, human immune reconstitution and phenotype, and the pharmacokinetics and anti-tumor efficacy of a therapeutic monoclonal antibody directed against a human target expressed on tumor-associated macrophages (TAMs). Method: HIS NOG-EXL (huNOG-EXL) mice were created using CD34+ cells from five donors in two independent cohorts (~12 weeks apart). Baseline immune reconstitution in blood was assessed at 10 weeks post engraftment. MDA-MB-231 cells were inoculated bilaterally and mice were randomized into groups for immune profiling, pharmacokinetics, and efficacy studies. Body weight, clinical observations, endogenous hIgG levels, and tumor growth were measured. For efficacy, tumors grew to ~100 mm3 and mice were dosed intraperitoneally every five days (6 doses total). Separate cohorts of mice were used for immune phenotyping and the single-dose PK study. Results: Donor consistency was seen for some of the parameters assessed but not all. The overall health outcomes for a given donor appeared to have an impact on the consistency of important parameters (e. g. efficacy). Donors that had poor health outcomes (body condition scores BCS 3) showed less consistency than those that experienced good (BCS=3) outcomes. The health trajectory, peripheral human immune cell engraftment, and PK profiles were similar across experiments for a given donor. Total human immune cell infiltration in tumors was similar across experiments, but there were notable shifts in immune cell composition. Whereas mice in the first cohort showed a dominant TAM phenotype within the tumor (~50% of all hCD45+ cells), mice across donors in the second cohort had reduced TAMs and corresponding increases in several T cell subsets. Efficacy was consistent in donors with good health outcomes (Donors 1 and 4) and was inconsistent in those with poor outcomes (Donors 3 and 5). Conclusion: Our study demonstrates that a return-to-donor approach can provide information that is reproducible across cohorts to help build a donor profile. An understanding of this profile should afford greater predictability and facilitate study design for drug discovery applications. Citation Format: Janell R. Richardson, Megan M. MacBride, Esther Andino, Emily Sack, Nicholas Smith, Po Y. Ho, Fang Xiao, Katelynn A. McEachin, David T. Omstead, Rachel Grgich, Michael N. Alonso, Justin A. Kenkel, Shelley E. Ackerman. Profiling of human CD34+ donors in a myeloid-supportive HIS model provides increased predictability and facilitates study design for evaluation of immuno-oncology therapeutics abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 5337.