Preclinical Evaluation of a Novel Lentiviral Vector Driving Lineage-Specific BCL11A Knockdown, β-Globin Induction and Simultaneous Repression of β-Globin for the Potential Treatment of Sickle Cell Disease

Ex vivo lentiviral vector-based gene therapy has shown promise as a treatment for severe sickle cell disease (SCD) (Ribeil, NEJM 2017). This approach relies on the regulated expression of a therapeutic anti-sickling β-globin (T87Q) at sufficient levels to reduce or prevent the polymerization of mutant hemoglobin S (HbS). Reactivation of fetal hemoglobin (HbF), by reducing expression of BCL11A with a lineage-specific miRNA adapted shRNA (shRNA (miR)), is an alternative strategy to generate therapeutic hemoglobin in transduced cells (Brendel, JCI 2016). Erythroid-specific silencing of BCL11A has an added benefit of simultaneously reducing expression of adult β-globin (which is mutated to HbS in SCD). The resulting increase in HbF (which itself has anti-sickling properties) and concomitant reduction in HbS may provide an extremely powerful approach to attenuating the sickling tendency of red blood cells. Here we provide new preclinical data generated to support the initiation of a first-in-human trial studying BCL11A silencing. Specifically, we report the characterization of a lentiviral vector construct encoding an erythroid-specific BCL11A shRNA(miR), termed BCH_BBLCR-shRNA(miR), adapted from our previously reported vector, that was developed to overcome poor vector titers observed in the manufacturing scale-up of our original vector. Healthy donor human hematopoietic stem cells transduced with lentiviral vector BCH_BBLCR-shRNA(miR) produced at scale under GMP conditions achieved high vector copy numbers (VCNs; >5 copies/diploid genome) and a 5-fold higher induction of HbF than seen with mock transduced cells (65.5% vs 13.1%). SCD human HSCs transduced with high-titer BCH_BBLCR-shRNA(miR) vector achieved >80% transduction efficiency, resulting in HbF induction between 70% and 81% of all β-like globin chains (with concomitant suppression of HbS, with residual HbS ranging from 19% to 30%). In addition, we observed no impact on in vitro growth and/or differentiation of erythroid cells derived from CD34+ cells. Analysis of individual colonies showed that >96% of individual BFU-E cells have HbF >50%. Repopulation studies in NSG mice transplanted with healthy donor human HSCs transduced (>80%) with BCH_BBLCR-shRNA(miR) showed no deleterious effects in vivo of targeting BCL11A in a lineage-specific fashion even at higher VCN (>3 copies/diploid genome). Lastly, consistent with the expected behavior of current generation lentiviral vectors, in vitro immortalization assays confirmed that the BCH_BBLCR-shRNA(miR) vector did not induce immortalization in this setting. Together these data demonstrate that LVV BCH_BBLCR-shRNA(miR) is (i) safe and efficacious in preclinical studies, and (ii) can be generated at scale in a GMP setting to support a clinical study of this potentially attractive approach to the treatment of SCD.