CRISPR/Cas9 Multi-Editing Enhances CAR NK Cells Therapeutic Potential Against Multiple Myeloma

Despite significant advances in the treatment strategies for Multiple Myeloma (MM), there is still a significant proportion of patients who will progress due to therapy resistances. CAR NK immunotherapy has recently emerged as a cost-effective and safer therapeutic alternative in relapsed/refractory cancer patients that can be applied in an allogeneic context, thus representing a promising off-the-shelf therapy against hematological malignancies. However, although CAR NK cells have demonstrated great antitumoral efficacy against MM in preclinical studies, there are shortcomings that may compromise their cytotoxic potential including their limited in vivo persistence or the presence of some dominant immunosuppressive signals in the tumor microenvironment (TME), triggered by soluble factors (i.e. TGF-β) or inhibitory immune checkpoints (e.g. NKG2A-HLA-E axis). To overcome all these limitations with a simultaneous strategy, we aimed to develop a next generation cord blood (CB)-derived CAR NK product using CRISPR/Cas9 technology to disrupt the expression of NK receptors involved in these signaling pathways.

We first simultaneously disrupted KLRC1 (gene that encodes NKG2A) and TGFBR2 genes in CAR CB-NK cells achieving 76.9% (64.6-84.5%) knockout (KO) efficiency for NKG2A and 63.4% (50.1-73.0%) for TGFβ-RII measured by multiparametric flow cytometry (M-FACS) and confirmed by the analysis of insertion-deletion events by Sanger sequencing (86.9% (70-97%) for KLRC1 and 66.5% (49-92%) for TGFBR2). Genome editing did not affect viability of the cells nor 41BB-BCMA CAR expression. Without any prior cell sorting process, double KO (DKO) CAR CB-NK cells showed higher antitumoral efficacy in vitro against different MM cell lines compared to control cells (RPMI-8226 2:1 E:T; 86.2% vs 53.6% specific lysis; p<0.001), even in the presence of soluble TGF-β. However, DKO CAR CB-NK showed a lower proliferative capacity which may compromise their in vivo persistence.

To address this challenge, we also targeted PRDM1 gene reaching 92.8% (84-96%) KO efficiency measured by Sanger sequencing and confirmed by western blot analysis by 79.4% (69.8-94.2%) protein reduction. As a result of PRDM1 disruption, triple KO (TKO) CAR CB-NK cells exhibited a 5.2-fold increased in vitro proliferative potential at day +26 post-nucleofection compared to non-edited cells. TKO CAR CB-NK cells showed similar in vitro efficacy as DKO cells against MM cell lines and patient-derived primary MM cells without hematotoxicity against healthy PBMCs. Moreover, we confirmed lack of large chromosomal structure aberrations in edited cells using Bionano’s optical genome mapping Stratys™ System.

In vivo persistence, efficacy and potential oncogenesis of DKO and TKO effector cells are being tested in a long-term immunodeficient MM mouse model with promising preliminary efficacy results measuring tumor burden by bioluminiscence (average radiance 54 days after treatment infusion: 1.4x10⁹ p/s/cm²/sr for non-edited CAR NK; 2.3x10⁸ p/s/cm²/sr for DKO NK cells; 3.8x10⁶ p/s/cm²/sr for TKO NK cells). Furthermore, up to date, (day +70 post-infusion), no preliminary signs of oncogenesis has been observed using TKO effector cells.

In summary, combined disruption of different target genes in CAR CB-NK cells by multiplex CRISPR/Cas9 genome editing represents a feasible, efficient and safe strategy to boost their cytotoxic capacity against MM and overcome CAR-independent TME-derived therapy resistances.