367 CRISPR/Cas9 gene-edited, allogeneic anti-CD83 CAR-T cells demonstrate potent activity in GvHD and AML tumor models

Background

CD83 is expressed on blood cancer cells, including acute myeloid leukemia (AML) blasts, and on allo-activated immune cells that are implicated in graft-versus-host disease (GvHD).¹ Targeting CD83 thus has the potential to improve outcomes in AML, as well as to reduce the risk of GvHD, which is a significant cause of mortality and morbidity in patients who undergo allogeneic hematopoietic cell transplant. The potential benefit of targeting CD83 is supported by literature showing that depletion of CD83+ cells in preclinical models of GvHD and AML is an effective therapeutic strategy.^1,2

Methods

We assessed the effectiveness of CRISPR/Cas9 gene-edited, allogeneic anti-CD83 CAR-T cells against animal models of GvHD and AML, and whether additional gene edits could enhance CAR-T potency in these settings. Anti-CD83 CAR-T cells were made with TRAC disruption to reduce the risk of GvHD, B2M disruption to reduce allogeneic host rejection, and insertion of an anti-CD83 CAR construct into the TRAC locus. To assess whether CD83 expression on activated T cells causes CAR-mediated fratricide, the CD83 gene was disrupted. To increase potency further, two additional gene disruptions were introduced: ZC3H12A (which encodes Regnase-1) and TGFBR2 (which encodes TGFBRII).

Results

We found that CD83 knockout (KO) improved the in vitro expansion of anti-CD83 CAR-T cells and enhanced in vivo activity. In a THP-1 tumor model, treatment with CD83 KO anti-CD83 CAR-T cells improved median survival when compared with wild-type cells (not reached vs. 59 days). In a xenogeneic GvHD model, CD83 KO cells were more effective at delaying GvHD at a dose of 1e6 CAR+ cells (median survival 71.5 vs. 51 days) and prevented GvHD entirely at a dose of 3e6 CAR+ cells. Activity could also be enhanced by combining CD83 KO cells with belatacept, a CTLA4-Fc fusion protein that blocks co-stimulation of T cells. Anti-CD83 CAR-T cells with KO of Regnase-1, TGFBRII, and CD83 (R/T/83 KO cells) maintained robust expansion in vitro, demonstrated increased target cell killing in vitro, and showed enhanced in vivo activity. In a THP-1 tumor model, durable complete responses were observed in all animals (5/5) treated with R/T/83 KO cells and in 60% of animals (3/5) treated with CD83 KO cells. In a xenogeneic GvHD model, a single dose of 1e6 CAR+ R/T/83 KO cells prevented GvHD while CD83 KO cells delayed GvHD onset.

Conclusions

Collectively, these data support the clinical evaluation of gene-edited, potency-enhanced, allogeneic anti-CD83 CAR-T cells in relapsed/refractory AML patients.

References

Shrestha B, Walton K, Reff J, Sagatys EM, Tu N, Boucher J, Li G, Ghafoor T, Felices M, Miller JS, Pidala J, Blazar BR, Anasetti C, Betts BC, Davila ML. Human CD83-targeted chimeric antigen receptor T cells prevent and treat graft-versus-host disease. J Clin Invest. 2020; 130:4652-4662. Wilson J, Cullup H, Lourie R, Sheng Y, Palkova A, Radford KJ, Dickinson AM, Rice AM, Hart DN, Munster DJ. Antibody to the dendritic cell surface activation antigen CD83 prevents acute graft-versus-host disease. J Exp Med. 2009; 206:387-398.