iMC enhanced antitumor efficacy of BCMA-CAR NK cells against THP-1 tumors in NSG mice. (A) Schematic retroviral vector encoding iMC; CAR including BCMA-targeting single-chain variable fragment, the minimal CD34 epitope (Q), CD8α stalk and transmembrane region, and the CD3ζ signaling domain; and human IL-15 with the transgenes separated by T2A and P2A ribosomal skipping sequences. (B) Activated NK cells were doubly transduced with γ-retroviral vectors encoding iMC-BCMA.ζ-IL15 and iRC9-ΔCD19 to generate DS.BCMA.ζ/IL15-modified NK cells. Flow cytometric analysis to determine transduction efficiency is displayed using anti-CD34 (iMC-BCMA.ζ-IL15) and anti-CD19 (iRC9-ΔCD19) antibodies compared with nontransduced (NT) NK cells. (C) BCMA expression in THP-1 tumor cells was determined by flow cytometric analysis with anti-BCMA antibody or isotype control. (D-E) NSG mice (n = 5 per group) were engrafted with 10⁷ NK cells NT or transduced with DS.BCMA.ζ.IL15 NKs 3 days after IV implantation of 10⁶ THP-1–eGFPFFluc tumor cells; 1 mg/kg of rimiducid (Rim) or vehicle (Veh) was administrated IP 5 times per week for the first week and 3 times per week thereafter. BLI was monitored by IVIS. Multiple Student t tests were used to compare DS.BCMA.ζ/IL15 NK Rim group with tumor alone group. (F-G) From day 40 to 48, mice from NT, DS.BCMA.ζ.IL15 NK vehicle, and Rim groups were euthanized. THP-1–eGFPluc cells were identified in bone marrow and spleen as GFP⁺ populations. (H-I) Human NK cells were identified in spleen and bone marrow as mCD45⁻GFP⁻hCD45⁺hCD34⁺ populations. Student t test was used for comparisons. *P < .05, **P < .01, ***P < .001. FSC, forward scatter; SSC, side scatter.