A Novel Stealth Strategy That Activates Adoptively Transferred Allogeneic Immune Cells and Avoids Rejection for Off-the-Shelf Cell-Based Cancer Therapy

Chimeric antigen receptor (CAR) T-cell therapies have revolutionized the treatment of hematologic malignancies, however, logistical complexities associated with patient-specific CAR T-cell therapies often limit broad accessibility to patients. Many of these challenges can be overcome with an allogeneic cellular product that is available off-the-shelf, and overcoming immune cell-mediated rejection of allogeneic cell therapy is an area of significant research. Conditioning chemotherapies, which are commonly administered with CAR T-cell therapy, suppress a patient's immune system and may create a suitable window of activity for allogeneic cell therapies to elicit clinical response. However, protracted lympho-conditioning has been associated with poor immune reconstitution and increased susceptibility to opportunistic infections. Deletion of human leukocyte antigen (HLA) surface expression is known to abrogate T-cell alloreactivity, but deletion of class I HLA must be combined with other immune-modulating strategies to avoid NK cell-mediated recognition.

To this end, allogeneic models combining class I HLA deletion with NK cell inhibitory molecules, such as HLA-E and CD47, have been shown to abrogate NK cell reactivity in mouse models. However, since HLA-E is the canonical activator of NKG2C, a dominant activating receptor found on human NK cells, and since the ligand for CD47, SIRPα, is known to be expressed on macrophages and dendritic cells and not on human NK cells, the observed effects of these immune-modulating strategies may not translate into the patient-treatment setting. In assessment of their protective effects in a defined human system, we found that HLA-E or CD47 overexpression on class I HLA null human cells offer only partial protection in evading various human NK cell compartments. We found that class I HLA-null K562 cells engineered to over-express CD47 were ineffective in inhibiting NK cells (0 to 7% inhibition). Separately, K562 cells engineered to over-express HLA-E, while effective in inhibiting NKG2A+ NK cells (90.2% +/- 3.7% inhibition), were unable to completely inhibit CD56 dim NK cells (33.2% +/- 29.6% inhibition) and not only failed, but instead activated, NKG2C+ NK cells (167% +/- 69% activation). Our data highlight the limitations of engineered CD47 and HLA-E modalities in suppressing broad populations of NK cells in clinically relevant settings.

We therefore evaluated expression of the alloimmune defense receptor (ADR) that uniquely targets alloreactive immune cells (Mo et al. Nat Biotechnol 2021). We have shown that the expression of ADR has the potential to evade host immune cells without the need for further genetic editing such as class I HLA deletion. To determine its applicability for off-the-shelf cell therapy, ADR expression was engineered into induced pluripotent stem cells (iPSCs) and iPSC-derived CAR-NK (CAR-iNK) cells were generated. CAR-iNK cells carrying the ADR modality (ADR+ CAR-iNK cells) showed normal patterns of differentiation (>99% CD56+ with co-expression of NK cell receptors such as NKG2D, NKp30 and NKp46), suggesting that ADR expression did not disrupt hematopoiesis or the expansion of iNK cells. Additionally, ADR+ and ADR-negative CAR-iNK cells produced similar cytotoxicity against tumor cells. We next tested the ability of ADR to provide resistance to alloimmune rejection by coculturing ADR+ CAR-iNK cells with allogeneic pBMCs in a mixed lymphocyte reaction (MLR) assay. Notably, ADR+ CAR-iNK cells maintained durable persistence throughout the entire duration of the MLR assay and suppressed the expansion of alloreactive T- and NK-cells in comparison to the control arm (Figure 1).

Collectively, initial preclinical studies suggest that ADR-modified CAR-iNK cells resist host immune cell rejection, while eliciting a durable anti-tumor response. Our preliminary data show evidence toward a promising off-the-shelf solution for elimination of broad pools of alloreactive T- and NK- effector cells in the clinical setting without the need for lympho-depleting conditioning or genetic editing strategies.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal