Gene Therapy Using a Self-Inactivating Lentiviral Vector Improves Clinical and Laboratory Manifestations of Wiskott-Aldrich Syndrome

The X-linked disorder Wiskott-Aldrich Syndrome (WAS), characterized by thrombocytopenia, eczema, recurrent infections, autoimmunity and cancer, is typically treated with allogeneic transplantation. Somatic gene therapy (GT) using autologous CD34+ cells is a promising treatment alternative for high-risk patients lacking matched allogeneic donors, as it avoids graft versus host disease. GT using a γ-retroviral vector with intact viral enhancers was efficacious but had a high rate of leukemia due to insertional oncogenesis. We report here preliminary results of 4 WAS patients who underwent GT using a self-inactivating lentiviral (SIN-LV) vector, in which viral enhancers have been deleted and the human WAS cDNA is controlled by a 1.6kB fragment of the human WAS promoter (w1.6_hWASP_WPRE SIN-LV). WASP expression per cell induced by this vector was lower than in normal cells when examined in murine models and human trials. We hypothesized that the SIN-LV would readily correct immune abnormalities due to selective advantage for WAS protein (WASP) expressing T cells whereas correction of myeloid and platelet abnormalities would require high vector copy number (VCN) in the infused cells.

Patients with severe WAS (clinical scores 3-5) were enrolled at a median age of 32 months (17 months-8 years). Patients 1 and 3 had detectable but low WASP expression. Patients 2 and 4 carried mutations that abrogated WASP expression but had evidence of somatic reversion in T and/or NK cells. CliniMACS purified CD34+ mobilized peripheral blood or bone marrow cells were transduced with the vector and infused after busulfan (12-15mg/kg) and fludarabine (120mg/m2) conditioning. CD34+ cell doses ranged from 6.3-24.91 x 10⁶ cells/kg. VCN of the infused cells was variable (3.37, 1.34, 0.54, 1.01 copies/cell). Busulfan exposure was myeloablative or near-myeloablative in patients 1, 3, and 4 (81.2, 77.2, 84.5 mg*h/L) and submyeloablative in patient 2 (48.8 mg*h/L).

All patients are alive with median follow-up of 13.5 months (9-24 months). All patients had improvement in eczema, remain platelet transfusion independent and have had no severe bleeding events. WASP expression in T cells was increased post-GT over baseline. Selective advantage for WASP expressing T cells was apparent in patients 1, 2 and 4, who had higher VCN in T cells at 6 months post-GT (0.93-2.21) than in B (0.48-1.7) or myeloid (0.13-0.89) cells. The presence of revertants in patients 2 and 4 did not appear to interfere with T cell reconstitution. In contrast patient 3 who had the highest WASP expression at baseline and the lowest VCN of infused cells (0.5 copies/cell), had the lowest VCN in T cells at 8 months (0.1 copies/cell). Defective T cell proliferation in response to anti-CD3 stimulation, characteristic of WAS, was improved post-GT. Next generation sequencing of T cell receptors in sorted naïve, memory and regulatory T cells revealed profound abnormalities of diversity, decreased entropy and marked clonal expansions pre-GT; most of these improved at 6 months post-GT. Cytoskeletal function in myeloid cells was highly abnormal pre-GT, as shown by absence of podosome formation in monocyte-derived dendritic cells (0-1% vs. 61% in controls). Podosome formation at 6 months post GT was improved but subnormal (4-40%). Only patient 1, who received the highest cell dose, the highest VCN, and myeloablative busulfan exposure had robust platelet reconstitution (pre-GT 24 versus 110 x 10⁹/L 6 months post-GT) and high level gene marking in myeloid cells 0.89 copies/cell. At the same timepoint, patients 2, 3 and 4 had platelet counts of 20-30 x 10⁹/L and correspondingly lower VCN in myeloid cells (0.13-0.27 copies/cell).

No severe adverse events related to GT have occurred to date, with relatively short follow-up. Integration site analysis of sorted cells showed highly polyclonal reconstitution, with distributions of integration acceptor sites as expected for the lentiviral vector backbone. In summary, GT using a SIN-LV that induces expression of WASP at levels below wild type improved the clinical and laboratory manifestations of WAS, with better reconstitution in patients receiving cells with high VCN.