General Updates | Importance of long-term surveillance after gene therapy in patients with hemophilia

I think the central message is that surveillance after AAV gene therapy should not be viewed as a short post-treatment safety window. This is a one-time treatment, but its biological consequences may remain with the patient for the rest of their life, even when the clinical benefit diminishes or the therapy does not work as well as expected. There are several reasons why prolonged follow-up is necessary. The first is efficacy: factor expression after gene therapy is highly variable and its durability remains uncertain, particularly in hemophilia A. We therefore need to continue monitoring factor levels, bleeding outcomes, and inhibitor development, rather than assuming that an initial response will necessarily be maintained. So that’s number one. And the second issue is liver safety. Liver enzyme elevations have been considerably more common following AAV factor VIII gene therapy than following AAV factor IX gene therapy. And in some individuals, these abnormalities have continued beyond the first year and as far as five years after treatment. These later elevations are sometimes attributed to alcohol, diet, medications, or other common causes. But we do not see the same frequency after hemophilia B gene therapy. And recent liver biopsy findings have also provided preliminary evidence of cytotoxic T-cell infiltration, hepatitis, and unfolded protein response shortly after the AAV factor VIII treatment. These findings were reported in only two individuals, so we should be cautious about over-interpreting them. Nevertheless, they support the possibility that producing factor VIII in hepatocytes, which are not its natural cellular source, may create cellular stress. That is an important reason to continue studying liver health beyond the early treatment period. And the third concern is vector integration and the potential risk of malignancy: AAV is often described as a non-integrating vector because most vector genomes remain as episomes; however, AAV can integrate into the human genome at a low frequency. Now, given the enormous number of vector particles and cells exposed to a clinical vector dose, even a low integration rate may translate into a very large number of individual integration events.

We now also have evidence from a different AAV gene therapy setting of a central nervous system tumor that was driven by vector integration. There are important differences between that therapy and hemophilia gene therapy, including the vector serotype, route of administration, and regulatory elements used. We therefore cannot directly extrapolate the level of risk. However, the case establishes that AAV-mediated tumorigenesis is mechanistically possible in humans. It is no longer purely a theoretical concern. At the same time, we should not overstate the current evidence. Other cancers reported after AAV gene therapy, including a case of hepatocellular carcinoma in a person treated for hemophilia B, have been investigated without finding convincing evidence that vector integration caused the malignancy. In that hepatocellular carcinoma case, the individual had several established background risk factors, and a comprehensive molecular analysis supported a conventional non-vector-related origin for that tumor.

So, the lesson is not that every cancer occurring after gene therapy was caused by the vector; the lesson is that every such case deserves careful investigation, including appropriate tissue collection and high-quality molecular analysis, whenever possible. In terms of duration of follow-up, current World Federation of Hemophilia guidelines recommend monitoring factor VIII activity, liver health, inhibitor development, and bleeding for at least 15 years, with a minimum of annual assessment. Some elements of surveillance, particularly surveillance for hepatocellular carcinoma – again, that’s what gene therapy using AAV vectors targets, that may need to continue throughout the patient’s lifetime. I would therefore regard 15 years as the minimum period of structured follow-up, rather than an expiration date, so to speak. The intensity of monitoring may change over time, but the responsibility does not disappear. Ultimately, “once and done” describes the administration of gene therapy, but it should not describe the relationship with the patient. At this point, the vector remains with the patient, regardless of how well the treatment works, and delayed adverse events may emerge many years later. So, treatment centers, clinicians, and manufacturers therefore need to remain vigilant, need to maintain a long-term and, in some respects, lifelong commitment to surveillance and patient safety.