Hemophilia A (HA) is characterized by a decrease in the functional clotting protein factor VIII (FVIII). Current hemophilia treatments consist of either prophylactic or on-demand administration of FVIII protein. Despite the vast improvements and availability of these treatments, which have increased the life expectancy and quality of life of HA patients, up to 30% of patients who receive FVIII infusions develop inhibitors to FVIII, rendering subsequent treatments ineffective and placing the patient at risk of a life-threatening bleeding event. The current method used to eradicate inhibitors is immune tolerance induction (ITI), in which the patient is repeatedly infused with high doses of FVIII protein to induce tolerance. While effective in some patients, ITI is extremely expensive, the mechanism whereby tolerance is induced is not well understood, and as many as 35% of patients fail to become tolerized. In this study, we created a Master Cell Bank of human placental cells (PLC) transduced with a lentiviral vector (0.5 vector copies/diploid genome equivalent) to produce high levels (4.9IU/10^6 cells/24 hr) of mcoET3, an expression/secretion-optimized FVIII protein (PLC-mco). We hypothesized that the immunomodulatory properties of PLC could be exploited to deliver FVIII while preventing inhibitor formation, and to determine whether the route of administration impacted the efficacy and safety of this cell-based treatment. To accomplish these objectives, we administered one of the following to normal healthy juvenile sheep recipients: a) 3 weekly IV infusions of 4x10^6 PLC-mco/kg (calculated to provide ~20IU/kg mcoET3 each 24 hr); or b) a single intraperitoneal (IP) infusion of 10^7 PLC-mco/kg or c) to enable comparison of the relative immunogenicity of mcoET3 when delivered as a bolus protein injection versus when constitutively secreted from transplanted PLC-mco, we also included a control/reference group in which we administered 5 weekly IV injections of 20 IU/kg recombinant mcoET3 protein. Prior to the first injection, and once per week for 5 weeks after the first injection, we collected blood from each animal and isolated plasma and peripheral blood mononuclear cells (PBMC). The plasma was used to determine FVIII activity by aPTT and to assess the presence of mcoET3-specific IgM and IgG by ELISA. PBMC were used to perform ELISpot assays to determine whether mcoET3-specific Th1 and Th2 cells developed following each infusion paradigm. After IV PLC injection, the sheep occasionally exhibited transient labored breathing; this was never observed in animals that received IV protein infusions nor those that received PLC via the IP route. One week after the IP injection of PLC-mcoET3, plasma FVIII levels had increased by 6.3% over baseline and then further increased to remain in a range between 63-108% for weeks 2-5. Following IV PLC infusions, sheep showed a 12.9% increase in plasma FVIII levels at 1-week post-infusion which persisted in weeks 2-4 and rose to a maximum of 30% five weeks after the first injection. As expected, given FVIII's short half-life, the sheep that received mcoET3 protein exhibited baseline plasma FVIII levels when evaluated 1 week after each IV protein infusion. To evaluate humoral immunity in these animals, ELISA analysis demonstrated that all sheep in this study were devoid of mcoET3-specific IgM antibodies at all time points. Similarly, no mcoET3-specific IgG antibodies were ever detected in sheep transplanted IV or IP with PLC-mco. In contrast, the IV infusion of recombinant mcoET3 protein resulted in the generation of a specific IgG response by 2 weeks after the first infusion. With respect to cell-mediated anti-mcoET3 immunity, ELISpot assays performed on PBMC demonstrated that the only cohort that developed Th1 or Th2 cells specific to mcoET3 was the one that received mcoET3 protein. In this group, Th2 cells became detectable by 3 weeks after the first protein infusion, while Th1 cells only became evident by week 4. Taken together, our results show that human PLC can be transduced to produce high levels of FVIII protein, and that these cells can then be transplanted into previously untreated animals to yield a prolonged increase in plasma FVIII levels. Our findings also support our hypothesis that delivering a fVIII transgene via PLC enables avoidance of an immune response to a FVIII protein (mcoET3) that is immunogenic to sheep when delivered as an IV bolus.

Disclosures

Doering:Expression Therapeutics, LLC: Current equity holder in private company, Patents & Royalties, Research Funding; Kilpatrick, Townsend & Stockton: Consultancy.

Author notes

*

Asterisk with author names denotes non-ASH members.