Background: Colorectal cancer liver metastases (CRCLM) are a major source of morbidity and mortality. Historically, curative therapy has been limited to surgical resection, but only a small fraction of patients are eligible. Cellular immunotherapy has shown promise in hematologic cancers, but challenges related to solid tumor therapy remain with optimal cell trafficking, elevated interstitial fluid pressures (IFP), and immunosuppression. We hypothesized that engineered natural killer (NK) cells expressing a natural killer group 2, member D (NKG2D) activating chimeric receptor (ACR) and membrane bound IL-15 (NKX101) would increase anti-tumor activity in vitro and in vivo utilizing our established regional delivery strategies. Methods:In vitro cytotoxicity and cytokine release of NKX101 cells or non-transduced NK cells (NT-NK) derived from the same donor were determined by co-culture systems with HCT116 cells that endogenously express NKG2D ligands. CRCLM-bearing NSG™ mice were treated with NKX101, NT-NK, or vehicle (CTRL) via portal vein (PV) for regional delivery (RD) or tail vein (TV) for systemic delivery (SD). Tumor burden (TB) was measured via tumor bioluminescence (TBL) and histopathology (HP). Flow cytometry (FC) determined the quantity of cells delivered. Student’s t-test and Mann-Whitney tests were performed for statistical comparisons. Results: NKX101 transduction efficiencies ranged between 63.5 – 75.6% across 3 separate healthy donors. EC₅₀ values derived from a 4-hour cytotoxicity assay for NKX101 vs. NT-NK were 3-4 fold lower with the greatest difference observed at the 1:1 effector-to-target (E:T) ratio (mean percent cytotoxicity: 72% vs. 20%, p = 0.001). In vitro cytokine assessment revealed 2.0-2.6 fold increases in IFN-γ, GM-CSF, and TNF-α levels compared to NT-NK cells (p < 0.0001 across all groups). In vivo, FC showed 2.89-fold increase in cell delivery using RD vs. SD on PTD1 (n = 3, p = 0.006). TBL was improved with 5 x 10⁶ cells via PV vs. TV (n = 6) from post-treatment day (PTD) 1-7, with greatest difference seen on PTD7 (12.9 vs. 42.6, p = 0.07). HP analysis showed reduction of TB at PTD7 with PV treatment. Conclusions: NKX101 demonstrated improvements in in vitro cytotoxicity and pro-inflammatory cytokine release. RD techniques in vivo revealed increased cell delivery and improved tumor control. Further studies are underway to confirm our initial findings and understand NKX101 cellular kinetics and susceptibility to immunosuppression in the liver, along with planned clinical evaluation in Phase 1 trials.