Vector Analysis of Multicentric Lymphoma in a Severe Hemophilia Α Dog after AAV Gene Therapy

Abstract

Background Adeno-associated viral (AAV) vector gene therapy for severe hemophilia A (HA) is approaching a clinical reality with multiple successful clinical trials. Despite this success, questions remain about the long-term safety of AAV vectors, especially regarding genomic integration and resulting oncogenesis. Studies of neonatal mice have demonstrated AAV integrations that drive hepatocellular carcinoma formation (Donsante et al. Science, 2007). Studies in HA canine models have also demonstrated AAV integrations and clonal expansion of hepatocytes with integrated vector genomes, but no evidence of tumor formation (Nguyen et al. Nat Biotechnol, 2021). Additionally, three tumors have been reported in AAV gene therapy recipients with HA or hemophilia B in clinical trials, without evidence of vector integrations driving oncogenesis. Though overall reassuring, these data highlight that the field needs to remain vigilant about the possibility of AAV integrations driving cancer development. Thus, the assessment of tumors in gene therapy recipients remains important for the AAV field. Aims To determine the role of AAV vector in the development of multicentric lymphoma in a severe HA dog that received liver-directed gene therapy. Methods Vector genome copy numbers in relevant tissues were determined by real-time PCR with primers situated in the 3'-end of the promoter and the 5'- end of the transgene. The vector copy number was calculated by comparing the analyte amplification signal against a standard curve of the plasmid DNA. Results A privately owned mixed-breed dog (PC9) with a history of recurrent spontaneous bleeding events presented with canine factor VIII (cFVIII) levels <1% normal, consistent with severe HA. At 9-months of age, he received liver-directed AAV8 gene therapy with a high-expressing canine factor FVIII (cFVIII-ΔF/V3) variant transgene at a dose of 6 x 1012 vg/kg via intravenous administration. Following gene therapy, his cFVIII levels increased to 2-3% normal and remained stable through duration of follow-up. His annualized bleeding rate decreased from 8 pre-gene therapy to 0 post-gene therapy. At 3.5 years after vector administration, PC9 was diagnosed with a widely disseminated cancer and euthanized due to complications of the disease. At autopsy, he was diagnosed with multicentric lymphoma involving peripheral and visceral lymph nodes, spleen, liver, and lungs. The average age of lymphoma diagnosis in dogs is between 6-9 years of age (Edwards et al. Vet Comp Oncol, 2003), so the development of cancer at an early age in a gene therapy recipient warranted additional study. Vector copy number analysis showed that the liver had the highest levels of AAV vector genomes, as expected. Liver samples showed a heterogenous distribution of vector genomes, consistent with previous studies of AAV8 transduction (Bell et al. Hum Gene Ther, 2011). Splenic samples also had a high level of AAV vector genomes, which again is consistent with known targeting of AAV8. Importantly, the lymph nodes had no detectable AAV vector genomes present, suggesting that vector integration could not contribute to lymphoma development. cFVIII levels remained stable before and after diagnosis of lymphoma, supporting lack of vector integration in the malignant cells. Conclusions AAV gene therapy resulted in a 100% reduction in the annualized bleeding rate in this pet dog with severe HA. Furthermore, the multicentric lymphoma that developed 3.5 years after liver-directed gene therapy is unlikely related to the AAV vector. We are unable to rule out other factors that may have contributed, including environmental influences or genetic predispositions for lymphoma. These results support the ongoing efforts to treat HA with AAV-based gene therapy and reinforce the safety profile of AAV vectors outside of a laboratory environment.