A Novel, Codon-Based Approach to Preserve On-Tumor Activity and Reduce Off-Tumor Toxicity of Chimeric Antigen Receptor (CAR) T Cells

Rolando Antonio Vedia, MS (Advisor: Jeffrey Molldrem, MD)

    Chimeric antigen receptor (CAR) T cell therapy has achieved remarkable results in the treatment of various hematological malignancies, ushering in tremendous innovation within the field of immunotherapy and improving patient outcomes. Despite this, CAR-T cell therapy is associated with a variety of toxicities including on-target, off-tumor toxicity which results from the targeting of tumor-associated antigens typically overexpressed on malignant cells but also expressed at lower levels on healthy cells. CAR-T cell-mediated on-target, off-tumor toxicity has been linked to severe adverse effects in patients and thus it is critical to identify and integrate novel approaches to combat these undesirable results. Our group has previously reported the efficacy of a TCR-like CAR named Hu8F4-CAR which targets the leukemia-associated antigen, PR1. PR1 is presented in the context of HLA-A2, is overexpressed on AML myeloblasts, and expressed at low levels in healthy hematopoietic cells. As with most CAR-T cells, the targeting of Hu8F4-CAR against PR1/HLA-A2 renders susceptibility to on-target, off-tumor toxicity. With the goal of reducing the likelihood of such toxicity, we modified Hu8F4-CAR and examined the expression, function, efficacy and toxicity of the resultant CAR-T cells.

      In our approach, we conducted rare codon engineering in Hu8F4-CAR in which select endogenous codons were strategically replaced with rare codons, or codons that are used at the lowest frequency in the human genome. In model organisms, the presence of rare codons has demonstrated the ability to tune gene and protein expression and functionality but this phenomenon has yet to be described in adoptive cellular therapies to our knowledge. We hypothesized that rare codon engineering in Hu8F4-CAR can fine-tune its expression and/or function to reduce the likelihood of off-tumor activity.

        Here we demonstrate that rare codon engineering within Hu8F4-CAR results in reduced cell surface expression and altered overall expression via flow cytometry and immunoblotting, respectively. Importantly, on-tumor efficacy is preserved with rare codon engineered Hu8F4-CAR-T cell variants, S30 and R27. In particular, cytotoxic activity is maintained both in vitro and in vivo against AML cell lines and primary AML. Moreover, we show that the same rare codon engineered Hu8F4-CAR variants, S30 and R27, have reduced on-target, off-tumor activity.  S30 and R27 variants showed decreased activity against healthy expressors of PR1 or HLA-A2 including healthy HLA-A2+ bone marrow, mobilized peripheral hematopoietic cells, and adult peripheral blood mononuclear cells in vitro. Furthermore, in vitro colony-forming unit assays demonstrated improved hematopoietic cell potential with S30 and R27 Hu8F4-CAR compared to wildtype Hu8F4-CAR. Lastly, CAR-T cells expressing the R27 Hu8F4-CAR variant showed reduced activity against healthy human HLA-A2+ cells engrafted in mice in vivo. Collectively, this work demonstrates that rare codon engineering is feasible and effective in tuning of CAR-T cells, and it lends crucial support to the potential of designing CAR-T cells which are able to retain on-tumor activity while reducing off-tumor toxicity.

Advisory Committee:

• Jeffrey Molldrem, MD, Chair
• Gheath Al-Atrash, DO, PhD
• Robert Jenq, MD
• Sijie Lu, PhD
• Yair Reisner, PhD
• Deepa Sampath, PhD

Join via Zoom (Please contact Mr. Vedia for the details of his Zoom meeting.)