Work

Development of a Cd33 P67 Epitope Editing Platform Toward Hematopoietic Stem Cell Shielding in Anti-cd33 Chimeric Antigen Receptor T-cell Therapy for Acute Myeloid Leukemia

Year: 2026

Type: article

Abstract: Introduction Anti-CD33 chimeric antigen receptor (CAR) T-cell therapy is a promising approach for acute myeloid leukemia (AML), but CD33 is also expressed on normal hematopoietic stem and progenitor cells, causing prolonged myeloablation and dependence on stem cell rescue. Existing strategies, including transient CAR expression, suicide switches, bridge-to-transplant approaches and complete CD33 knockout, may impair physiological CD33 function or increase therapeutic complexity. Objectives To develop the experimental platform required for CD33 P67 epitope editing as a future hematopoietic stem cell shielding strategy for anti-CD33 CAR T-cell therapy. Materials and methods AML cell lines were characterized for CD33 therapeutic epitope expression by flow cytometry using the anti-CD33 clones P67 and WM53, in three independent experiments. Stable luciferase-expressing AML cell lines were generated by lentiviral transduction and puromycin selection and validated by bioluminescence assays for future preclinical studies. Genome-editing strategies targeting the CD33 P67 epitope were designed in silico in Benchling (human reference GRCh38), including single-guide RNA (sgRNA) selection, an adenine base-editing (ABE8e; Addgene #138489) approach for selective epitope editing, and a parallel complete-knockout strategy using Cas9 ribonucleoprotein with a homology-directed repair (HDR) ssODN donor. A workflow for ABE8e mRNA production was implemented to support future CD34+ hematopoietic stem cell editing. Results Stable luciferase-expressing AML cell lines were established and validated by bioluminescence assays, providing models for future preclinical studies. Flow cytometry with P67 and WM53 showed high CD33 P67+/WM53+ double-positivity (∼90-100%) in HL-60, MOLM-13, Kasumi-1, MV4-11 and U937, and lower double-positivity (∼60%) in OCI-AML3. P67-detected CD33 median fluorescence intensity (MFI) was highest in HL-60 and significantly lower in the remaining lines, whereas WM53-detected MFI confirmed overall CD33 expression. HL-60, MOLM-13, MV4-11, Kasumi-1 and U937 were selected as the most suitable models for genome-editing studies. P67 was defined as the therapeutic target epitope, as this clone recognizes the CD33 region targeted by the anti-CD33 agent gemtuzumab ozogamicin (GO). Based on structural and in silico analyses, editing was directed to codons F21 and W22 in exon 2 of CD33, which form the P67 antibody-binding epitope. In contrast to complete CD33 knockout, selective editing of these residues is expected to abolish P67-mediated recognition while preserving endogenous CD33 expression and function. Accordingly, sgRNAs, an HDR donor and an ABE8e strategy were designed, and the ABE8e mRNA production workflow was established. Discussion and Conclusion These preparatory studies establish the methodological framework for subsequent CD33 P67 epitope editing in hematopoietic stem cells and for future functional evaluation of shielding against anti-CD33 CAR T cells, using GO as a positive control. Genome editing of AML cell lines and CD34+ cells, together with these functional assays, is ongoing or planned. This platform represents an essential step toward CD33 epitope editing as a strategy to enable safer anti-CD33 CAR T-cell therapy for AML.

Source: Hematology Transfusion and Cell Therapy

DOI: 10.1016/j.htct.2026.106854