Fertility-preserving myeloablative conditioning using single-dose CD117 antibody-drug conjugate in a rhesus gene therapy model

Abstract

Hematopoietic stem cell (HSC) gene therapy has curative potential; however, its use is limited by the morbidity and mortality associated with current chemotherapy-based conditioning. Targeted conditioning using antibody-drug conjugates (ADC) holds promise for reduced toxicity in HSC gene therapy. Here we test the ability of an antibody-drug conjugate targeting CD117 (CD117-ADC) to enable engraftment in a non-human primate lentiviral gene therapy model of hemoglobinopathies. Following single-dose CD117-ADC, a >99% depletion of bone marrow CD34 + CD90 + CD45RA- cells without lymphocyte reduction is observed, which results are not inferior to multi-day myeloablative busulfan conditioning. CD117-ADC, similarly to busulfan, allows efficient engraftment, gene marking, and vector-derived fetal hemoglobin induction. Importantly, ADC treatment is associated with minimal toxicity, and CD117-ADC-conditioned animals maintain fertility. In contrast, busulfan treatment commonly causes severe toxicities and infertility in humans. Thus, the myeloablative capacity of single-dose CD117-ADC is sufficient for efficient engraftment of gene-modified HSCs while preserving fertility and reducing adverse effects related to toxicity in non-human primates. This targeted conditioning approach thus provides the proof-of-principle to improve risk-benefit ratio in a variety of HSC-based gene therapy products in humans.

Fig. 1. CD117-ADC depletes both human and non-human primate CD34+ cells.

A Escalating doses of CD117-ADC were added to primary human and rhesus CD34+ cells in culture, and cell viability at day 6 was evaluated to calculate the half maximal effective concentration (EC₅₀) for killing (n = 3 biologically independent human donors, n = 2 biologically independent rhesus donors). Data are presented as mean +/− standard deviation. The standard deviation was shown as error bar. **p < 0.01 evaluated by one-tailed student’s t-test for human CD34+ cells. p = 1.72e−1, 2.47e−1, 7.29x−2, 9.96e−2, 8.26e−2, 1.31e−4, 5.81e−6, 9.63e−6, 1.78e−4, 3.60e−5, 3.37e−6, 2.89e−1, and 2.98e−1, respectively (from the lower concentration). B Cynomolgus macaques were treated with intravenous injection of CD117-ADC (0.1 or 0.3 mg/kg × 1 day, n = 3 per dose) or busulfan (6 mg/kg × 4 days, n = 3), and bone marrow CD34 + CD90 + CD45RA- cells were quantified by flow cytometry 7 days after drug administration (n = 3). C To evaluate drug clearance, plasma concentrations of ADC in the circulation were measured after a single intravenous injection of CD117-ADC (0.2, 0.3, 0.4, and 0.6 mg/kg) in rhesus macaques (n = 1). The ADC concentrations for 0.2 and 0.3 mg/kg at day 3, and 0.2, 0.3, and 0.4 mg/kg at day 4 were below the limit of quantification. IgG-ADC: immunoglobulin G isotype control-conjugated ADC. Source data are provided as a Source Data file.

Fig. 2. A single dose of CD117-ADC conditioning robustly depletes bone marrow of rhesus macaques.

A Experimental design for escalating-dose CD117-ADC administration into rhesus macaques. The animals received mobilization and had CD34+ cells collected by apheresis, but these cells were not transplanted into the autologous macaques following CD117-ADC injection. B Peripheral blood counts in rhesus macaques before and after CD117-ADC conditioning. C, D Peripheral blood levels of liver transaminases (aspartate aminotransferase (AST) and alanine aminotransferase (ALT)), lactate dehydrogenase (LDH), total bilirubin (TBIL), albumin (ALB), blood urea nitrogen (BUN), creatinine (CREA), and potassium (K) before and after CD117-ADC administration. Source data are provided as a Source Data file.

Fig. 3. Efficient gene marking and robust HbF induction after CD117-ADC conditioning in a rhesus gene therapy model compared to busulfan.

A Schema of autologous CD34+ cell transplantation with thEpoR and shmiR-BCL11A lentiviral vector transduction (MOI 50). Transplantation occurred either 1 day after myeloablative busulfan conditioning (5.5 mg/kg × 4 days) in rhesus macaques (n = 2), 12U018 and 12U020, or 6 or 10 days after a single injection of 0.3 or 0.4 mg/kg CD117-ADC in rhesus macaques (n = 4) (0.3 mg/kg in ZL13 and ZJ62, 0.4 mg/kg in H635 and H96G), respectively. B Blood counts in rhesus macaques before and after transplantation. C Vector copy number (VCN) in granulocytes and lymphocytes post-transplantation, evaluated by quantitative polymerase chain reaction (qPCR). D HbF-positive red blood cells (F-cells) post-transplantation, evaluated by flow cytometry. E HbF (γ-globin) amounts at the protein level in red blood cells post-transplantation, evaluated by reversed-phase high-performance liquid chromatography (HPLC). pA: a polyadenylation signal, ANKp: an erythroid-specific ankyrin-1 promoter, +51: an erythroid-specific enhancer in the BCL11A gene. Source data are provided as a Source Data file.

Fig. 4. Pregnancy in transplanted macaques following CD117-ADC conditioning.

A The menstrual cycle per month after rhesus HSC transplantation with 0.3–0.4 mg/kg CD117-ADC (n = 4 biologically independent animals), myeloablative busulfan (5.5 mg/kg × 4 days, n = 4 biologically independent animals), or myeloablative total body irradiation (TBI, 4.5–5.0 Gy × 2 days, n = 6 biologically independent animals) conditioning. Data are presented as mean +/− standard deviation. The variances of menstrual cycle per month are equivalent among three groups. **p < 0.01, *p < 0.05 evaluated by Tukey’s HSD test (two-tailed). p = 8.84e−3 between CD117-ADC and Busulfan, p = 1.44e−2 between CD117-ADC and TBI, and p = 8.01e−1 between Busulfan and TBI. B Ultrasound image and schema of the fetus in a pregnant macaque (ZL13). C The sex hormones (estradiol, progesterone, and testosterone) in rhesus serum after mating following transplantation with CD117-ADC conditioning. ZL13 and H635 were pregnant. Source data are provided as a Source Data file.