Wiskott-Aldrich syndrome (WAS) is a primary immunodeficiency characterized by eczema, thrombocytopenia, infections, and a high risk of developing autoimmunity and cancer. In a recent clinical trial, a γ-retroviral vector was used to introduce a functional WAS gene into autologous hematopoietic stem/progenitor cells (HSCs), followed by reinfusion of the gene-corrected HSCs into the patients. This strategy provided clinical benefit but resulted in expansion and malignant transformation of hematopoietic clones carrying vector insertions near oncogenes, thus increasing leukemia risk. We have developed a clinical protocol for WAS based on lentiviral vector (LV) gene transfer into HSCs.

Analysis of common insertion sites (CIS) in gene therapy trials using lentiviral versus γ-retroviral vectors. Word clouds show the intensity of insertion sites clustering in each of the CIS genes (the larger the gene name, the larger the number of insertion sites within or in the proximity of that gene). The names of the CIS genes detected in both gene therapy trials are reported at the intersection between the circles. Analysis of common insertion sites (CIS) in gene therapy trials using lentiviral versus γ-retroviral vectors. Word clouds show the intensity of insertion sites clustering in each of the CIS genes (the larger the gene name, the larger the number of insertion sites within or in the proximity of that gene). The names of the CIS genes detected in both gene therapy trials are reported at the intersection between the circles.

Three patients with WAS were treated in a phase I/II clinical trial with gene-corrected HSCs after pretreatment with a reduced-intensity myeloablative regimen. Autologous CD34+ cells were transduced with an optimized LV carrying the WAS gene under the control of its endogenous promoter. Patients were monitored for up to 2.5 years after gene therapy by molecular, immunological, and clinical tests. We also investigated the genomic distribution of LV integration sites in the patients’ bone marrow and peripheral blood cell lineages.

Administration of autologous HSCs transduced with LV at high efficiency (>90%) resulted in robust (25 to 50%), stable, and long-term engraftment of gene-corrected HSCs in the patients’ bone marrow. WAS protein expression was detected in myeloid cells at similar rates and in nearly all circulating platelets and lymphoid cells. In vitro T cell proliferative responses, natural killer cell cytotoxic activity, immune synapsis formation, and suppressive function of T regulatory cells were normalized. In all three patients, we observed improved platelet counts, protection from bleeding and severe infections, and resolution of eczema. Vector integration analyses on >35,000 unique insertion sites showed distinct waves of HSC clonal output, resulting in highly polyclonal multilineage hematopoietic reconstitution. In contrast to ©-retroviral gene therapy, our LV-based therapy did not induce in vivo selection of clones carrying integrations near oncogenes. Consistent with this, we did not see evidence of clonal expansions in the patients for up to 20 to 32 months after gene therapy.

Our gene transfer protocol provided efficient stem cell transduction in vitro, resulting in robust and stable in vivo gene marking. WAS expression was restored to near-physiological levels in the patients, resulting in immunological and hematological improvement and clinical benefit. Clonal tracking of stem cell dynamics by vector insertions showed details of hematopoietic reconstitution after gene therapy. Comparison with clinical data from ©-retroviral gene therapy in the same disease setting strongly suggests that LV gene therapy offers …