Gene therapy for beta-thalassemia is based on the transplantation of genetically-modified autologous hematopoietic stem cells (HSC) into patients affected by the severe form of disease. The genetic treatment of the hemoglobinopathies poses the general challenge of efficient level of gene transfer into HSC and high and persistent transgene expression, in the differentiated progeny of a genetically modified stem cell. The validation of a gene therapy approach to thalassemia requires to obtain results of gene correction in a broad number of patients’ cells, since different molecular defects in the beta-globin gene lead to the clinical phenotype. The heterogeneity in the molecular defects and in the proportion of alpha and non-alpha (beta, gamma and delta) chains will represent a key element to set a threshold in the amount of vector-derived beta-chain required to correct a thalassemic phenotype. Additionally, the impact of some biological parameters, such as the degree of BM erythroid hyperplasia, the BM subpopulations proportion and the apoptotic index, on the successful correction of thalassemic phenotype needs to be studied in the perspective of clinical translation. In order to address these issues, we collected samples from BM aspirates and isolated CD34+ cells from 25 beta+ and beta0 thalassemic patients, characterized by different genotypes and biochemical profiles of globin chains synthesis. A novel, erythroid specific LV expressing human beta-globin from a minimal promoter enhanced by only 2 LCR elements (HS2 and HS3) was used to transduce BM derived CD34+ cells at high efficiency (>80%). The efficacy of the beta-globin LV in correcting the human thalassemic phenotype was tested in an in vitro model of erythropoiesis and in the human-mouse hematological chimera. Upon transduction, normal level of HbA expression was achieved in erythroblastic cultures and BFU-E, associated with a progression towards erythroid maturation, which was impaired in mock-transduced thalassemic cells. Molecular analysis showed proviral integrity, with no detectable rearrangements and an average proviral copy number of 2.4. Analysis of specific globin chains proportion and contribution to phenotype correction in the context of different genotypes is under evaluation.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author