Peter Feldman

Removal and Inactivation of Enveloped and Non-Enveloped Viruses during the Purification of a High-Purity Factor IX by Metal Chelate Affinity Chromatography

Virus reduction during the copper chelate affinity chromatography stage used during the purification of a new high‐purity factor IX (BPLs 9MC) has been investigated. Virus reduction for the enveloped virus Sindbis was 6.5 log, a value which included approximately 2 log of inactivation due to the use of an acidic wash buffer (pH 4.4) during chromatography. In the case of the non‐enveloped hepatitis‐A‐like poliovirus, which is acid‐resistant, the virus reduction value was 4.0 log and was exclusively due to physical virus removal during the chromatographic process.

Virus removal from factor IX by filtration: validation of the integrity test and effect of manufacturing process conditions.

Virus removal from a high purity factor IX, Replenine-VF, by filtration using a Planova 15N filter has been investigated. A wide range of relevant and model enveloped and non-enveloped viruses, of various sizes, were effectively removed by this procedure. Virus removal was confirmed to be effective when different batches of filter were challenged with poliovirus-1. It was confirmed that intentionally modified filters that failed the leakage test had completely lost the ability to remove virus, thus confirming that this test demonstrates gross filter failure. In the case of the more sensitive integrity test based on gold particle removal, it was found that a pre-wash step was not essential. Planova filters that had been modified by sodium hydroxide treatment to make them more permeable, and filters manufactured with varying pore-sizes over the range of 15-35 nm, were tested. The integrity test value that resulted in the removal of >4 log(10) of poliovirus-1 from the product correlated with that recommended by the filter manufacturer. Virus removal from the product was not influenced by filter load mass, flow-rate or pressure. These studies confirm the robustness of this filtration procedure and allow suitable process limits to be set for this manufacturing step.

Sensitivity of the Fibrinogen Clotting Time: An in vitro Test of Potential Thrombogenicity

The fibrinogen clotting time (FCT) is a measure of thrombin activity, and is used to evaluate the potential thrombogenicity of prothrombin complex concentrates (PCC). We have defined end points for clot formation in this test which allow the measurement in PCC of thrombin concentrations as low as 0.001 IU/ml. The FCT of thrombin and PCC samples which did not contain antithrombin III (ATIII) were the same when measured at 20°C or 37°C. In the presence of ATIII (0.05 or 0.25 IU/ml), samples of PCC which were known to contain thrombin showed shorter FCT at 20°C than at 37°C. Inclusion of both ATIII (0.25 IU/ml) and heparin (4 IU/ml) in PCC ensured the complete inactivation of endogenous thrombin.

Hepatitis A Transmission by Factor IX Concentrates: Control by Severe Dry Heat Treatment at 80°C

In a recent article in this journal, Lawlor et al. [1] provided evidence that a prothrombin complex concentrate (PCC), used for treating factor IX deficiency, had transmitted hepatitis A. However, this product did not include a specific virus inactivation method and the inclusion of a suitable step such as severe dry heat treatment could have prevented virus transmission. Further cases of hepatitis A transmission by another factor IX concentrate have also recently been reported [2] for which compelling evidence of a link between product and recipient was shown by nucleic acid sequence data. However these findings, while cause for concern, should not be used to generalise on the viral safety of PCCs. The conclusion of Lawlor et al, that steps to inactivate both enveloped and non-enveloped viruses are essential in the preparation of concentrates, is valid but this retrospective report should be set in the context of current manufacturing technology. This relates both to the specific virus inactivation method used and to the precise fractionation method by which the concentrate is prepared. The generic description ‘prothrombin complex concentrate’ in reality applies to products with diverse purification methods, each with their own influence on purity, thrombogenicity and virus safety of the final product. Thus the finding that the Irish PCC (type ‘PSB’) contains no detectable IgG [1] contrasts with the PCC produced at BPL (factor IX fraction type ‘9A’) which contains approximately 500 μg IgG/ml, equivalent to about 5% of the total protein in the vial, even though both products are prepared by slight variations of the same general method. For instance, because a proportion of the IgG may be directed against hepatitis A, this may provide some passive immunity to recipients. However it is not clear whether this contribution is significant. An effective method for inactivating the non-enveloped hepatitis A virus has been available for some considerable time: severe dry heat treatment at 80 C for 72 h, as used in the BPL intermediate-purity PCC type 9A, inactivates both enveloped and non-enveloped viruses. Our ability to heat the concentrate under such severe conditions, while retaining adequate functional activity in the product, further reflects differences in the properties of various PCC products. The programme of development, originally initiated to reduce the risk of hepatitis transmission, allowed BPL to introduce the 80 C/72 h-treated 9A product as long ago as 1985, in response to the urgent need to control the spread of HIV [3]. The rapid introduction of 9A allowed clinical studies of virus transmission to precede the availability of laboratory results on virus inactivation, but the process was later shown to reduce the risk of non-A, non-B hepatitis transmission [4]. We have since tested the inactivation of a range of viruses, including hepatitis A, by severe dry heat treatment to complement the lengthy (now 11-year) clinical experience of using 9A. The results of these studies are as follows. (1) 5–6 log of Sindbis and herpes simplex type-1 were inactivated during dry heating at 80 C for 72 h. These viruses are used as models for hepatitis C and HIV. (2) 4–6 log of encephalomyocarditis virus and polio virus type 1 were inactivated during the freeze-drying process. These two picornaviruses are related to hepatitis A. (3) 3–5 log of hepatitis A virus itself were inactivated during freeze-drying. The total inactivation reached 5 log after only 8 h of the standard 72 h 80 C heat treatment cycle. While the levels of hepatitis A virus inactivation due to freezedrying are substantial, they clearly may not be great enough to prevent the transmission of hepatitis A as occurred with the PCC described by Lawlor et al. [1]. However, freeze-drying followed by heat treatment at 80 C is effective at inactivating hepatitis A virus in PCCs such as 9A. Essentially similar findings have been reported for high purity factor VIII and IX concentrates for which the inactivation of 2 log occurred over freeze-drying and reached a total of 5 log after 24 h at 80 C [5]. Thus the report of hepatitis A transmission overlooks specific products and virus inactivation methods which have been available since 1985, albeit unfortunately a few months after the two hepatitis A transmissions described. More recently there has been a move to high purity factor IX products many of which include a solvent/detergent treatment step for the inactivation of enveloped viruses [6]. In addition, the chromatographic processes that are used on some of the high purity products are effective at removing and/or inactivating nonenveloped viruses such as hepatitis A. For instance, in the case of Replenine , the copper chelate affinity chromatography step has been shown to remove 4–5 log of nonenveloped viruses including hepatitis A [7, 8]. Some new approaches such as the use of virus filters [9, 10] have been applied or considered for increasing the safety of factor IX products particularly with regard to small non-enveloped viruses. However severe heat treatment should not be forgotten as it represents a well-established procedure which can effectively inactivate a wide range of viruses including hepatitis A. Nevertheless, each and every defined virus reduction method should be considered on its merits for a specific product. Where yield and safety are compatible with the method, long-established and effective inactivation methods such as severe heat treatment should still be considered, particularly given the security advantage of a terminal treatment step in the final sealed container.