Analysis of Hemophilia A Gene Therapies Avoiding Single-Class Pooling Reveals Distinct Gaps in Product-Specific Durability and Safety
Background
Hemophilia is a rare genetic disorder characterized by recurrent spontaneous bleeding or joint damage due to clotting factor deficiency. Patients have managed lifelong prophylaxis with periodic intravenous administration of deficient clotting factors. To alleviate this burden, gene therapies using Adeno-Associated Virus (AAV) vectors have been developed. The principle involves delivering normal genes to liver cells via a single intravenous injection to induce the body to produce clotting factors directly.
However, questions have persisted in clinical practice. It has been difficult to find clear answers regarding how long the initial therapeutic effects observed at the start of administration would safely persist and what differences exist among various treatments. Existing meta-analyses or literature reviews have often committed the error of averaging values by grouping drugs with different vector designs and dosages into a single AAV category. This has obscured the unique characteristics and limitations of each pipeline.
Key Findings
Researchers comprehensively analyzed prospective systemic AAV clinical trials for hemophilia reported up to May 15, 2026. The analysis included a total of 14 independent clinical cohorts, combining 6 Hemophilia A cohorts and 8 Hemophilia B cohorts. Duplicate reports were strictly filtered by comparing trial identifiers, gene vector structures, dosages, enrolled patient numbers, and follow-up periods, and the level of evidence was verified using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool.
The analysis revealed distinct differences between products in the expression levels and persistence of long-term coagulation factors. For example, in the GS001 Phase 1 study, the median Factor VIII (FVIII) activity at week 144 in the low-dose cohort remained at 5.6 IU/dL, whereas the high-dose cohort recorded 42.7 IU/dL at week 104. This clearly demonstrates the variation in systemic activity depending on dosage and formulation composition.
Rates of discontinuation of existing prophylaxis and achievement of zero-treated-bleed were not combined into a single metric but were instead synthesized through product-specific descriptive synthesis. This was because baseline therapies, criteria for re-administration of treatment, and observation periods varied significantly across clinical trials, which could lead to severe distortion if simply summed. Safety indicators also showed large variations between products. Liver abnormalities such as elevated liver enzyme levels, the duration and intensity of immunosuppressant administration, the frequency of treatment-related serious adverse events, and patterns of loss of efficacy varied by formulation. Rare cases of neutralizing antibody formation or thrombosis reported outside the eligible cohorts confirmed the necessity of long-term surveillance tailored to individual products.
Implications and Outlook
This systematic analysis demonstrates that selected patient populations can derive multi-year hemostatic benefits from a single administration, while simultaneously highlighting the risks of hasty generalization. Some Hemophilia B treatment programs targeting Factor IX (FIX) deficiency demonstrated relatively superior durability by maintaining stable endogenous expression for several years. Numerous cases were also confirmed where the frequency of hemostatic agent administration was dramatically reduced, significantly improving patient quality of life.
Conversely, not all products guarantee permanent cures. In some formulations, a gradual decrease in clotting factor activity over time or loss of expression due to immune responses was observed. This highlights the need for complementary strategies to regulate T-cell-mediated immunity in response to gene delivery vectors and molecular design technologies to overcome in vivo gene silencing. Biological constraints, such as the exclusion of patients possessing pre-existing viral antibodies before administration and the difficulty of re-administration due to antibodies formed after the first dose, also remain challenges to be overcome.
BACKGROUND: Adeno-associated virus (AAV) gene therapy can provide sustained endogenous factor expression after a single infusion, but durability and safety differ across products. We synthesised long-term efficacy and clinically interpretable safety outcomes while avoiding inappropriate class-level pooling. METHODS: Five databases were searched from inception to 15 May 2026 for prospective systemic AAV trials in haemophilia A or B. Overlapping reports were linked by trial identifier, vector/transgene, dose, enrolment, sites, sample size and follow-up. One additional eligible report published within the search window was identified during peer review and incorporated. The primary outcome was product-specific factor activity at the latest follow-up with clinical durability. Prophylaxis status, bleeding and safety were synthesised descriptively by product because baseline regimens, outcome definitions and observation windows were not sufficiently harmonised for cross-product pooling. Risk of bias was assessed with ROBINS-I. RESULTS: Fourteen unique cohorts were included: six in haemophilia A and eight in haemophilia B. Reported long-term factor activity remained product-specific. The GS001 Phase 1 study reported median FVIII activity of 5.6 IU/dL at Week 144 in the low-dose cohort and 42.7 IU/dL at Week 104 in the high-dose cohort. Prophylaxis and zero-treated-bleed outcomes were not pooled because baseline treatment, restart criteria, denominators and observation windows differed. Liver abnormalities, immunosuppressive exposure, treatment-related serious adverse events and loss of expression varied by product. Rare inhibitor and thrombotic events reported outside the eligible prospective cohort synthesis reinforce the need for ongoing product-specific surveillance. CONCLUSIONS: Selected patients experienced sustained haemostatic benefit, and several FIX programmes showed stable multiyear expression. Because assays, products, immune-management strategies and follow-up
This study sounds an alarm against the existing practice of prescribing or evaluating hemophilia gene therapies as a single class. Clinicians must comparatively review product-specific long-term activity trajectories and immunosuppressant exposure durations tailored to a patient's age, baseline liver function, and physical activity level. In practice, this enables precision personalized treatment, such as selecting formulations with proven high-dose expression data for active patients who require sustained high activity, or pipelines with shorter immunosuppressant exposure periods for patients concerned about the risk of liver toxicity. In insurance price negotiations and the establishment of reimbursement standards, a practical baseline is expected to be established for designing value-based payment models linked to multi-year follow-up data for each product.