Extensive Metabolic Correction of Hurler Disease By Hematopoietic Stem Cell-Based Gene Therapy: Preliminary Results from a Phase I/II Trial

Allogeneic hematopoietic stem cell transplantation (HSCT) performed early in life is the current standard of care for patients with severe type 1 mucopolysaccharidosis (Hurler disease), a metabolic disorder caused by mutations in the alpha-L-iduronidase (IDUA) gene, leading to impaired breakdown of glycosaminoglycans (GAG). Secretion of IDUA by donor-derived hematopoietic cells may cross-correct non-hematopoietic cells, slowing progression of tissue damage and cognitive decline. Nevertheless, Hurler patients undergoing HSCT manifest substantial residual disease burden, e.g. on the skeleton and central nervous system (CNS). We conducted a phase I/II clinical study (NCT03488394) to test whether infusion of autologous CD34⁺ hematopoietic stem and progenitor cells (HSPC) transduced ex vivo with a lentiviral vector coding for the IDUA gene was feasible, safe and capable of restoring enzymatic activity in the patients' blood and tissues, up to supraphysiologic levels. The trial originally planned to enroll 6 Hurler patients with preserved neurocognitive function (DQ/IQ>70) that had no access to a suitable allogeneic donor. Sample size has recently been increased to 8 patients. By July 2019, six patients have been treated at a median age of 24 months (range: 14-34), with a median follow up of 4 months (range: 1-13). In all patients, we collected a high number of autologous HSPC by leukapheresis following mobilization with lenograstim and plerixafor, resulting in drug products with a median of 21 million CD34⁺ cells/kg (range: 13-29). Transduction efficiency was high with a median above 80% and a vector copy number (VCN) of 1.7 (range: 1.0-5.2), employing a shortened, 2 day transduction protocol that included prostaglandin E2. All patients showed rapid hematopoietic recovery following myeloablative conditioning with busulfan (targeted to an AUC of 80mg*h/L), fludarabine (160mg/sqm) and rituximab (375mg/sqm). Median duration of grade 4 neutropenia associated with conditioning was 15.5 days (range: 13-19). Also associated with conditioning, Grade 3 thrombocytopenia lasted 4 days, while only 2 out of 6 patients experienced a platelet drop below 20,000/mcL on a single day, in the absence of transfusion support. Adverse events were mild and compatible with myeloablative conditioning, with the exception of patient 3 who experienced an anaphylactic reaction on day+12, which promptly responded to antihistamines, IV fluids and steroids. All evaluable patients showed sustained, supraphysiologic blood IDUA activity (dried blood spot), which was on average 3 fold above the upper limit of normal (evaluable patients: n=5 at 1 month, n=4 at 2 months, n=3 at 3 months). Notably, in n=4 Hurler patients treated with allogeneic HSCT, we detected IDUA activity that ranged within the lowest quartile of normal in spite of full donor chimerism, suggesting substantial gain achieved by overexpressing IDUA in ex vivo genetically-modified autologous HSPC. Urinary GAG excretion fell to normal levels within 3-6 months. IDUA activity was also detected in the cerebrospinal fluid (CSF) of treated patients, accompanied by a logfold reduction in CSF GAGs in the 2 patients with longest follow up. This suggests that gene therapy accomplishes full metabolic correction of tissues, including the CNS. Gene therapy did not induce antibodies against the IDUA protein, while pre-existing antibodies induced by enzyme replacement therapy before gene therapy rapidly disappeared. Patient 1 who reached the 1-year follow-up demonstrated a stable cognitive score, improved findings on brain and spine MRI, resumed growth velocity and an improvement of his skeletal phenotype. The preliminary results from our phase I/II study compare favorably with the standard of care in terms of safety and efficacy, and highlight the potential of genetic engineering of HSPC grafts for therapeutic gain-of-function.