Search

Lentiviral (LV)-based hematopoietic stem and progenitor cell (HSPC) gene therapy is becoming a promising alternative to allogeneic stem cell transplantation for curing genetic diseases. To potentially improve the efficacy, safety and economic sustainability of HSPC transduction, we reasoned to genetically manipulate only the more potent CD34+CD38- HSPC, thereby improving HSPC maintenance in culture in the absence of differentiating cells and downscaling the cell therapy product by a factor of ten without compromising long-term engraftment. This approach would also decrease the total load of vector integration infused in the patients, thus improving its overall safety.First, we determined the engraftment kinetics of CD34+ mobilized peripheral blood (mPB) subpopulations over a 24wk xenotransplantation period. We sorted CD34+ mPB into 4 fractions with increasing expression levels of CD38 and marked each fraction with a specific fluorescent protein allowing to track the population of origin driving hematopoietic reconstitution. Differentially labeled fractions were mixed, and various combinations of CD38-, CD38int and CD38hi HSPC were injected into NSG mice (2 exp, n=30). Almost all long-term repopulating capacity (>90%) was contained within CD34+CD38- cells, and these cells took over hematopoiesis by 9wks. Instead, early reconstitution was mainly driven by CD34+CD38int progenitor cells.In the prospect of a clinical translation, we then modeled the co-administration of gene-modified CD34+CD38- mPB cells with uncultured CD34+CD38int/+ supporter cells aimed to drive fast hematopoietic recovery (3 exp, n=38). Repopulation by gene-modified CD34+CD38- cells was slower (15wks) and incomplete ( 5 wks and re-established long-term (24wks) gene marking up to 85%, thus allowing to benefit from prompt hematopoietic recovery driven by transiently repopulating CD38int/+ supporter cells.Last, we optimized LV transduction in the framework of an improved culture protocol. Exposing CD34+ or CD34+CD38- mPB cells to prostaglandin E2 (PGE2) increased transduction efficiency 1.5-2.5x, allowing to markedly reduce pre-stimulation and LV exposure times better preserving HSC functions. Importantly, the higher gene-transfer efficiency was maintained for up to 24 wks following xenotransplantation (n=33), suggesting that PGE2 facilitated LV transduction in long-term HSC.In summary, these results support the clinical development of novel HSPC gene therapy protocols based on the modification of highly purified HSC subsets, with the prospect to improve the efficacy, safety and feasibility of future ex vivo gene therapy studies.

Ex vivo gene addition into CD34+ hematopoietic stem and progenitor cells (HSPC) followed by autologous transplantation has proven a safe and efficacious therapy for immunodeficiencies, storage diseases and hemoglobinopathies. According to xenograft models and population size estimates of vector-marked cells in gene therapy-treated patients, less than 0.01% of infused CD34+ cells drive long-term (LT) repopulation. Advances in clinical-grade cell sorting technology may make HSC-enriched CD34+ subpopulations accessible for gene therapy, with advantages in terms of lentiviral vector (LV) cost, safety (lowering of integration load) and, potentially, efficacy. By differentially marking mobilized peripheral blood (mPB) CD34 subpopulations distinguished by increasing levels of CD38 expression in order to quantitatively assess their hematopoietic output in an NSG xenograft model over 6 months, we previously mapped most (>90%) LT repopulating capacity to CD34+CD38- cells (lowest 10% CD38 staining), while CD34+CD38int cells drove short-term (ST) reconstitution during the first 2 months after transplantation. We now characterize these subpopulations in terms of CD90 expression, a marker, which has been used in conjunction with CD34 for HSC purification in past clinical trials. CD34+ mPB cells were sorted into CD38-90+ (5% of CD34+), CD38-90- (5%), CD38+90+ (30%) and CD38+90- (60%) fractions and exhaustively transduced with GFP-, OFP-, BFP- and mCherry-expressing LVs, respectively. Differentially marked subfractions were pooled maintaining their original proportions and transplanted into NSG mice. ST engraftment mainly came from CD38+ cells, with equal contribution from the CD90+ and CD90- compartment. LT engraftment was almost exclusively derived from CD34+CD38- cells, of which 70% came from CD90+ and 30% from CD90- cells. Hence, CD34+CD38- is a more sensitive and specific marker combination than CD34+CD90+ to purify LT-HSC. CD34+CD38- cells can be purified by a sequential bead-based selection (CD34 selection of CD38-depleted cells) potentially applicable to clinical practice. We show that CD34+CD38- cells can be efficiently transduced with clinical grade LVs using shortened ex vivo manipulation protocols, reaching similar gene marking levels as with the standard protocol currently used in clinical trials that comprises a double dose of LV. Transduction was stable for at least 5 months when serially measured in xenotransplanted mice, and mice showed multi-lineage hematopoiesis indistinguishable from CD34+ grafts. Based on these results, we are aiming towards clinical development of a new gene therapy protocol based on CD34+CD38-HSPC efficiently transduced with minimum ex vivo culture time (