Abstract
Gene therapy product release requires reliable and consistent demonstration of biopotency. In hemophilia B vectors, this is usually determined in vivo by measuring the plasma levels of the expressed human factor IX (FIX) transgene product in FIX knockout mice. To circumvent this laborious assay, we developed an in vitro method in which the HepG2 human liver cell line was infected with the vector, and the resulting FIX activity was determined in the conditioned medium using a chromogenic assay. The initial low sensitivity of the assay, particularly toward adeno-associated viral serotype 8 (AAV8), increased approximately 100-fold and allowed linear measurement in a broad range of multiplicities of infection. Statistical parameters indicated high assay repeatability (relative standard deviation (RSD) < 5%) and intra-assay reproducibility (RSD < 20%). To compare the performance of the in vitro and in vivo biopotency assay, we applied statistical analyses including regression techniques and variation decomposition to the results obtained for 25 AAV8-FIX vector lots (BAX 335). These showed a highly significant correlation, with the cell culture-based assay demonstrating less variation than the in vivo test. The in vitro assay thus constitutes a viable alternative to using animals for lot release testing.