In Vivo Expansion of NK Cells Stimulated with PM21 Particles Under Ultralow IL-2

Cancer immunotherapy with NK cells is promising, but a key challenge is the development of a method that can consistently expand NK cells from a small fraction (~5%) of unselected PBMCs to a therapeutically effective dose while using “off the shelf” reagents, not requiring long-term complex co-cultures, and with low doses of IL-2. Methods that use feeder cells do expand NK cells, but require complex procedures with prolonged ex vivo co-culture with the feeder cells, are costly, are logistically difficult, and require subsequent administration of high dose IL-2 for persistence after injection. We developed a particle-based method to specifically expand cytotoxic NK cells from unselected PBMCs. These particles, prepared from plasma membrane (PM) of engineered K562 cells that express cytokines such as IL-21 or IL-15 and 41BBL, stimulate in vitro and in vivo expansion of NK cells. In vitro, these PM-particles containing membrane bound IL-21 (PM21-particles) cause highly specific expansion of cytotoxic NK cells from unselected PBMCs by achieving >90% NK cells in 14 days and 10⁴ fold expansion of NK in 21 days. NK cell expansion is consistent with different preparations of PM-particles or leukocyte sources, and the PM-particles retain expansion efficacy with -80 ºC storage. The PM21-particles also stimulate in vivo NK cell expansion in NSG mice under ultralow IL-2 (1,000 U per injection, 3 x weekly). The efficacy of ex vivo pre-activation with PM21, prior to injection, was tested by injecting PBMCs (2 x 10⁶ PBMCs per mice, i.p., 6 mice per group), which were either activated in culture for two days with 200 µg/mL PM21 (PM21-PBMCs) or not pre-activated (PBMCs), and then the abundance of hCD56⁺CD3^- NK cells and other hCD45⁺ lymphocytes were monitored in peripheral blood of mice. Clear efficacy for pre-activation was observed where the mice in the group given PM21-PBMCs had 66 fold higher human NK (hNK) cells as compared to the group injected with PBMCs. The efficacy of administering PM21-particles in vivo was examined also where 2 x 10⁶ PM21-PBMCs, were injected i.p. into mice (4 mice per group) that then received 0, 400, 800 or 1600 µg of PM21 (i.p., 2 x weekly). The mice injected with 800 µg of PM21 had the highest amount of hNK cells averaging 890,000 hNK/mL of mouse blood on day 16 and reaching 95 ± 2% of total hCD45⁺ cells by day 12. By comparing the averages between day 5 and 16 in this experimental group, hNK cell increased a minimum of 390 fold in the mouse blood. Furthermore, examination of organs from mice sacrificed on day 16 after PM21-PBMC injection shows significant numbers of hNK cells in bone marrow, spleen, lung, liver and brain. The extent of NK cells observed in the peripheral blood (>50,000 NK cells/mL blood, 14 days post injection) and biodistribution to sites of disease would be relevant for clinical cancer treatment, and thus shows that PM-particle technology could be suitable for NK cell immunotherapy.