INSERTION OF MICRORNA TARGET SEQUENCES INTO RETROVIRAL REPLICATING VECTORS EFFECTIVELY RESTRICTS TRANSGENE EXPRESSION AND VIRAL REPLICATION IN HUMAN HEMATOPOIETIC STEM AND PROGENITOR CELLS

Abstract Prodrug activator gene therapy with a retroviral replicating vector (RRV) has shown a highly favorable safety profile and long-term survival in early-phase trials for recurrent high-grade glioma. Overall endpoints were not met in a recent Phase 3 trial, but highly statistically significant survival was observed in prespecified patient subgroups as compared to randomized matched control patients receiving standard-of-care treatments, and further clinical evaluation is being focused on these subgroups. Additional strategies to enhance therapeutic potency may require two or more RRVs to deliver multiple transgenes simultaneously, but RRVs encoated with the same envelope protein will compete for the same cellular receptors, interfering with efficient co-infection. The current clinical vector (formerly Toca511, now DB107) is encoated by amphotropic murine leukemia virus (MLV) envelope, which binds to inorganic phosphate transporter PiT-2/SLC20A2. To switch RRV tropism, we developed RRVs pseudotyped with a heterologous envelope from Gibbon ape leukemia virus (GALV), which utilizes an alternative phosphate transporter, PiT-1/SLC20A1, for cell entry. Efficient co-infection of established and primary human glioblastoma cells with MLV- and GALV-pseudotyped RRV was achieved, without receptor competition. However, human hematopoietic stem/progenitor cells (HSPC) also express high levels of PiT-1, which may increase potential genotoxicity of GALV-pseudotyped RRV. Accordingly, as a novel strategy to restrict gene expression and replication in HSPC, we have developed and tested new RRV designs incorporating microRNA target sequences (miRT). Insertion of miRT recognized by hematopoietic lineage-specific microRNA-142-3p resulted in complete suppression of RRV replication in primary human CD34+ HSPC, with the percentage of infected cells below 1% after vector inoculation at multiplicities of infection (MOI) 0.05 and 0.1, and remaining below 3% even after inoculation at MOI 0.5 and 1.0. In vivo models have been developed for on-going studies to evaluate miRT-mediated restriction strategies for avoidance of potential genotoxicity by RRV in normal hematopoietic cells.