Introduction: CAR T-cell therapy has revolutionized the treatment of patients with relapsed/refractory (R/R) acute leukemia, NHL, and multiple myeloma. However, there are still areas of improvement in their clinical activity, source of the effector cells, prevention, and management of adverse events that require particular attention. Because of those reasons, NK cells appear as a viable effector cell alternative that can help address these challenges. NK cells offer a profile of activation, expansion, persistence, and cytotoxicity that is different from T cells and, when modified to bear CAR constructs, may provide significant advantages. However, the preclinical development of NK-CARs is challenging mainly because of the difficulty of generating large quantities of cells for testing and well-established pathways for CAR optimization before in vivo evaluation. Therefore, we developed a CAR optimization platform using the NK-92 cell line. NK-92 cells conserve their cytotoxic ability and can be easily expanded in vitro and used for functional and phenotypical evaluations of novel CAR-NK constructs. Here we present a rechallenge cytotoxic assay that mimics repetitive in vivo effector interactions with the target cells and its use for optimization, comparison, and development of NK-based cellular therapies.

Methods: We generated lentivirus transduced CD19 CARs (FMC63-41BB-z) using T cells from healthy donors and NK-92 cells for comparison.T cells were expanded for 12 days, and a 41.9% CAR+ expression was achieved (CART19). Transduced NK-92 cells were sorted by FACS to obtain a population of 98.3 % CAR+ cells (CARNK19) and subsequently expanded for 12 days. JeKo-1 cells were used as CD19+ targets and BxPC3 cells as CD19 neg control (both cell types were GFP-Luc-PuroR). We developed a Luciferase-based rechallenge cytotoxicity assay. For this, we diluted the effector to target (E/T) ratio to obtain a logarithmic trendline of the cells' cytotoxicity. E/T ratio to get viability of 50% (IC50) measured at 4h (for CARNK19) and 24h (for CART19) was used as a proxy of the product's potency. Both CAR Immune Effector Cells (IECs) were co-cultured with their targets at an E/T ratio to obtain 70% cytotoxicity. After 24 hours with the target, we estimated the remaining IEC amount in the culture using GFP exclusion in flow analysis (IEC cells/mL = total cells/mL x GFP neg%). We repeated the plating of E/T ratio dilutions to perform daily IC50 curves using this rechallenge strategy for a total of 5 days. CAR and PD1 expression were measured on Day 0 and Day 5 by flow cytometry.

Results: CART19 showed a higher IC50 than CARNK19 at baseline, 1.7 vs. 0.19 (Figure 1A). The IC50 trend of both IECs over time showed an uptrend that suggests progressive functional exhaustion (Figure 1B). At 5 days of rechallenge, it was 29 times higher in T cells than in NK-92 (12.07 vs. 0.42) and with a slope 265 times higher (10.6 vs. 0.04). Furthermore, we observed that when comparing the levels of CAR expression on Day 0 vs. Day 5, CART19 showed a decrease in CAR expression that was not present in CARNK19 (41.9 to 10.9% vs. 98.3 to 95.5%) (Figure 1C). In addition, there was a higher increase in PD1 expression in CART19 cells than CARNK19 cells from Day 0 to Day 5 of the in vitro rechallenge (9.9 to 46.8% vs. 0.88 to 8.88%) (Figure 1D).

Conclusion: Our data shows the use of NK-92 cells as a tool for optimization and preclinical development of NK cell-based cellular therapies. We demonstrated that it is feasible to set up repetitive cytotoxic challenges that mimic closer in vivo E/T engagement. Moreover, using the cytotoxic IC50 calculated with this platform, we show increased cytotoxicity, less functional exhaustion, and less expression of PD1 in CARNK19 than in its T cell counterpart. Overall, the NK-92 rechallenge cytotoxicity assay platform constitutes a helpful tool for research, development, and optimization of cellular therapies based on NK cell effector function.

Disclosures

No relevant conflicts of interest to declare.

Sign in via your Institution