AFFILIATIONS
1 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States
2 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States, Emory University, Atlanta, Georgia, United States
3 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States
4 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States, Emory University, Atlanta, Georgia, United States
5 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States, Emory University, Laney Graduate School, Atlanta, Georgia, United States
6 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States
7 Aflac Cancer & Blood Disorders Center - Children's Healthcare of Atlanta, Atlanta, Georgia, United States, Emory University School of Medicine, Atlanta, Georgia, United States, Emory University, Atlanta, Georgia, United States
ABSTRACT
Background: Acute myeloid leukemia (AML) remains difficult to treat, with 5-year overall survival rates of ~30-50% depending on age group and risk factors. While chimeric antigen receptor (CAR) T-cell therapy for AML is a promising strategy, it faces unique challenges, including tumor heterogeneity, antigen escape, and CAR T-cell dysfunction within an immunosuppressive tumor microenvironment. To overcome these challenges, we propose combining this immunotherapy with cytotoxic chemotherapy. CAR T cells can be rendered chemotherapy-resistant by knocking out deoxycytidine kinase (dCK), a key enzyme in the nucleoside salvage pathway that activates the commonly used AML chemotherapeutics cytarabine and fludarabine. We hypothesize that the use of concurrent chemotherapy and AML-directed CAR T-cell therapy will be more effective than either approach alone in targeting this aggressive disease.
Methods: Six single guide RNAs (sgRNAs) against dCK were designed using CRISPick, Benchling, and IDT sgRNA design programs, prioritizing on-target efficacy and minimizing off-target effects. Each candidate sgRNA was electroporated with recombinant Cas9 protein into the Jurkat immortalized T-cell line to knock out dCK. Knockout (KO) efficiency was quantified at the DNA level by Sanger sequencing comparisons of dCK-edited and unedited cell DNA, at the mRNA level using RNA sequencing, and at the protein level using Western blotting. To test the functional differences of dCK-edited cells versus unedited cells, each group was exposed to fludarabine and/or cytarabine for 72 hours, and a chemiluminescent assay quantifying ATP was used to calculate IC25 doses of each chemotherapeutic. CD33 and CD70 expression on AML cell lines was assessed by flow cytometry after treatment with sublethal doses of chemotherapy. AML-directed CAR-T products were created by transducing healthy donor T cells with lentiviral vectors encoding a CD33-CAR and/or a CD70-CAR. Both pooled and co-transduced approaches were tested using these scFv-based CAR constructs, which have previously been validated in our lab. To create a heterogeneous AML model, luciferase-tagged Molm-13 cells underwent CD33 KO or CD70 KO using the CRISPR/Cas9 knockout methods as described above. In vivo efficacy of our CAR T-cell products was assessed in a heterogeneous Molm-13 AML murine xenograft model. Chemotherapy-resistant CAR T cells are now being generated using the most effective dCK sgRNA.
Results: Of the six sgRNAs tested, two sgRNAs achieved dCK KO in Jurkat cells as indicated by insertion/deletion analysis via Sanger sequencing deconvolution. Both sgRNAs targeted exon 3 of the dCK gene and resulted in +1 and +2 insertions in the nucleotide sequences of edited cells, causing frameshift mutations. These findings were supported by RNA sequencing, which demonstrated a ~50% decrease in dCK mRNA transcript expression, and by Western blot analysis, which revealed a 27-30% reduction in dCK protein levels compared to unedited controls. Despite a partial reduction in protein expression, functional resistance to cytarabine was markedly enhanced. dCK-edited cells demonstrated an 18-fold increase in cytarabine resistance compared to unedited controls, based on IC25 values calculated from chemiluminescent ATP assays. Furthermore, a 30-40% increase in both CD33 and CD70 expression was seen in AML cells after treatment with cytarabine and fludarabine at LD50 doses, highlighting the potential for a synergistic benefit. Single- and dual-targeting CD33/CD70 CAR-T products were successfully generated and exhibited potent in vitro cytotoxicity against AML cell lines. Dual-targeting CAR-T products significantly prolonged survival in our heterogeneous AML murine xenograft model, with the co-transduced CD33/CD70 CAR-T product proving more effective than the pooled CAR-T product (p<0.05). A workflow to generate a chemotherapy-resistant, AML-directed CAR-T product has now been established, in which dCK is knocked out of primary T cells, which subsequently undergo transduction with CAR-encoding lentiviral constructs. These products will be evaluated in vitro and in vivo as described above with and without concurrent chemotherapy.
Conclusions: dCK KO is a feasible strategy to confer chemotherapy resistance in CAR-T products, enabling a combination chemo-immunotherapy approach to target AML. This combination approach has the potential to overcome the current deficiencies in AML-directed CAR-T therapy.
DISCLOSURES
Nabil Saleem, MD: No relevant conflicts of interest to declare.