A phase 1/2 safety and efficacy study of TAK-754 gene therapy: The challenge of achieving durable factor VIII expression in haemophilia A clinical trials

Abstract

Introduction: Haemophilia A is an X-linked bleeding disorder resulting from a deficiency of factor VIII (FVIII). To date, multiple gene therapies have entered clinical trials with the goal of providing durable haemostatic protection from a single dose. TAK 754 (BAX 888) is an investigational AAV8-based gene therapy containing a FVIII transgene. Reduction in CpG motifs was performed to reduce immunogenicity based on prior observations. Here, we describe the results of the first two cohorts treated with TAK 754.

Aim: To report clinical and translational results of the TAK-754 phase 1/2 AAV gene therapy study for the treatment of haemophilia A.

Methods: A phase 1/2 single arm open-label dose escalation study of TAK-754 was performed in participants with severe haemophilia A (NCT03370172). Participants were monitored for safety events, endogenous FVIII activity and bleeding rates. Glucocorticoids were implemented to preserve transgene expression. A transcriptomics analysis was performed to evaluate immunogenicity along with additional post-hoc analyses.

Results: Four participants were dosed in two cohorts. Infusion of TAK 754 was well-tolerated. All participants developed mild transient transaminase elevation and subsequent loss of FVIII expression within the first 12 months of treatment despite use of glucocorticoids. Transcriptomic analysis did not demonstrate significant changes in immunogenicity signals in peripheral blood. One serious adverse event of hypophosphatemia occurred in the second cohort without obvious risk factors.

Conclusions: Sustained FVIII expression remains a challenge in haemophilia A AAV gene therapy trials. Mechanisms of transgene expression loss require further study as clinical studies enter long term follow-up periods.

Keywords: adeno‐associated virus; gene therapy; haemophilia; hypophosphatemia; phase 1–2.

FIGURE 1

(A–E) TAK 754 vector design and individual participant study time course of factor VIII expression, ALT, bleeding, and immunosuppression. TAK‐754 vector design and individual participant time courses. A, TAK‐754 transgene and expression cassette elements. B and C, Baseline characteristics and time course of study participants in cohort 1. D‐E: Baseline characteristics and time course of study participants in cohort 2. Annualized bleeding rates are calculated based the number of reported bleeding events divided by the number of months in the reporting period and multiplied by 12. Per protocol, bleeding events include any spontaneous or traumatic bleed experienced during the interval. Participant 4 resumed prophylaxis with emicizumab at approximately +28 months of follow‐up. ITR, inverted terminal repeat; BDD‐FVIII, B‐domain deleted factor VIII; TTR, transthyretin; AAV, adeno‐associated virus; ABR, annualized bleeding rate; ALT, alanine aminotransferase; cp, capsid particles; CRM, cross‐reacting material; C, central; L, local; FVIII, factor VIII; L, laboratory; OS, one‐stage; SCR1, screening 1; Wk, study week.

FIGURE 2

(A) Participant 4 hypophosphatemia time course. (B) Participant 4 hypophosphatemia risk factor analysis. (C) Participant 4's polygenic risk score (PRS) compared to both controls and individuals from the UK Biobank (UKBB). On the left side, histogram depicting the PRS for blood phosphate levels in a typical test (refer to Supplementary Methods for details). The test included 20% of unrelated UKBB participants, along with Participant 4 from the TAK‐754 trial and two controls, whose PRS percentiles are color‐coded. Participant 4's PRS percentile falls slightly below −1 standard deviation, while both controls fall within ±1 standard deviation of the calculated predisposition index. On the right side, the distribution of PRS percentiles for Participant 4 (the case) and the controls, estimated through over 100 experiments. In each of these experiments, the UKBB dataset was randomly split into training and testing sets at an 80/20 ratio. The distribution of the PRS percentiles for the case is significantly lower (p < 1e⁻⁵; Wilcox test) when compared to the control percentiles. *Genetic analysis on single genes was performed by local laboratory using an NGS panel. W: (study) week; UKBB: United Kingdom Biobank, TRP: Tubular Reabsorption of Phosphate value calculated as: 1‐[(urine phosphorus/serum phosphorus) X (serum creatinine/urine creatinine)]; FAERS (FDA Adverse Event Reporting System).