Built to last: gene therapy for ADA SCID

In this issue of Blood, Reinhardt et al report the long-term clinical benefit and safety in patients with adenosine deaminase deficient severe combined immunodeficiency (ADA SCID) after gammaretroviral gene therapy using autologous bone marrow-derived CD34⁺ stem and progenitor cells.¹ 

Biallelic defects in the ADA gene cause deficiency of ADA, the enzyme that catalyzes the conversion of adenosine to inosine. Lack of ADA activity results in accumulation of deoxyadenosine mono-, di-, and tri-phosphate, which are toxic to lymphocytes, leading to a SCID phenotype and death from opportunistic infection. In contrast to enzyme replacement therapy, which can raise the lymphocyte count somewhat, allogeneic hematopoietic stem cell transplant and autologous gene therapy are definitive treatments that restore meaningful immunity. The technology used in this trial, gammaretroviral vectors to transduce bone marrow-derived cells infused fresh, is already considered outdated. Yet, the detailed analysis of the outcome of 10 patients in this trial makes important contributions to the field.

Reinhardt et al show sustained efficacy from one of the first successful trials of gene therapy for inherited immune deficiency, and indeed for any disease. After multiple attempts that failed because of poor efficiency of gene transfer and failure to sustain gene marking at a level with clinical benefit, groups in the United States, Italy, and London made 2 key changes in the protocol that made the difference between success and failure.^2-4 Enzyme replacement therapy was paused before infusion and, importantly, patients received low-dose busulfan (or melphalan in London) to partially ablate recipient hematopoietic stem cells. These trials all used gammaretroviral vectors in which the viral genes were replaced with a complementary DNA expressing the human ADA gene, leading to strong expression of the ADA enzyme controlled by the gammaretroviral regulatory elements. Patients in this trial reported by Reinhardt et al received transduced bone marrow cells that were infused fresh without cryopreservation, and transduction efficiency (measured as average vector copy number in bulk cultured CD34⁺ cells) was assessed in retrospect, with great variability from product to product. All 10 patients show stable multilineage gene marking 8 to 11 years post-treatment.¹ The group also demonstrates that lymphocytes carrying the gene exhibit a strong selective advantage (see figure), with higher and more even gene marking (10-fold difference between patients) compared with granulocytes, a surrogate for gene marking in long-term hematopoietic stem cells (∼100-fold difference between patients). All patients expressed the ADA enzyme, with improvement in metabolic profile, T, B, and natural killer cell counts, immunoglobulin production, and 9 of 10 remained off enzyme replacement therapy. These biochemical and immunological parameters were strongly correlated with vector copy number in granulocytes with the 10 patients falling into low, medium, and high gene-marking categories. The 3 patients with the highest gene marking in granulocytes consistently had the best cellular and clinical correction, the highest number of unique vector insertions, and the most polyclonal repertoire (see figure).

ADA SCID stands in stark contrast to other inherited immunodeficiencies where gammaretroviral gene therapy was used, in that insertional oncogenesis has been very rarely seen. Gammaretroviral vectors were the first vectors used successfully to transduce murine and human long-term hematopoietic stem cells, and as integrating vectors, were ideal to ensure production of multilineage gene-marked cells. The integration of gammaretroviruses into the DNA of the transduced cells is different for each individual cell, is not controlled, and is not random. Gammaretroviruses by nature tend to insert near the transcriptional start site of genes, leading to the possibility that the strong regulatory elements driving expression of the transgene of interest may inadvertently transactivate genes near the insertion site in that individual cell. The consequences of this biological property of gammaretroviruses manifested dramatically through the development of insertion-driven malignancies. Insertions near LMO2 were largely responsible for T-cell leukemia or lymphoma (6/20 patients with X-linked SCID, 6/9 patients with Wiskott-Aldrich syndrome) and insertions near MECOM associated with myelodysplastic syndrome or acute myelogenous leukemia (4/4 patients with X-linked chronic granulomatous disease, 3/9 patients with Wiskott-Aldrich syndrome), a total of 17/33 patients (51%).⁵ In contrast, only 1 of at least 50 patients with ADA SCID who underwent gene therapy with a gammaretroviral vector on clinical trials^3,4,6,7  or received the commercial product based on the Milan vector, Strimvelis, has developed leukemia, reported in a press release in fall 2020.⁸ Among the 10 patients in the study by Reinhardt et al reported here, 6 (including the 3 patients with the highest granulocyte marking) had integrants detected at more than 1 timepoint in proto-oncogenes implicated in insertional oncogenesis (MECOM, LMO2, HMGA2, IKZF1), yet none have had clinical evidence of leukemia.¹ The presence of integrants near proto-oncogenes was not limited to patients with low transduced cell number, and, conversely, patients with robust and polyclonal repertoire were not protected from having such integrants. The mystery of why ADA SCID patients thankfully appear to have a much lower risk of insertional oncogenesis despite the use of gammaretroviral vectors remains unsolved.

Gene therapy has advanced considerably since the time this trial was conducted. Trials in recent years, including for ADA SCID, use lentiviral vectors, which boast a more neutral insertion pattern, shorter transduction time, and higher titer vectors. Some vectors use cell- or gene-specific promoters and codon-optimized transgenes that lead to regulated and higher expression. Peripheral blood stem cells are more plentiful than bone marrow stem cells, even in older patients. Cryopreservation of transduced cells and transduction enhancers improve uniformity and quality of cell products, and portability. The gammaretroviral vector approach reported here by Reinhardt et al lacks the bells and whistles of today’s approaches, but like a Ford truck, was built to last, and got the job done.