IVIG: The Evolution of the Manufacturing Process

The process of isolating “gamma” globulins from blood dates back to the 1940s when Cohn and Oncley were able to use a method of cold ethanol fractionation to separate plasma from precipitated antihemophilic factors and use cold ethanol in a secondary step to isolate the “gamma” globulins. The resulting product was crude by today’s standards.

The original purpose of the Cohn-Oncley plasma fractionation process (Table 1) was to separate albumin from plasma. The motivation was to supply a plasma “expander” for soldiers in World War II. The shortage of the blood supply could be supplemented with albumin and led to better healing rates in blood-loss-related trauma. The original standard immune globulin (IG) also was used to boost the immune system when a patient was exposed to certain viruses such as hepatitis A and rubella, as well as for blood clotting and to treat primary immune deficiency disease (PIDD) patients.

These latter byproducts of the fractionation process (antihemophilic factors, fibrin clotting factors and IG clotting factors and IG), however, were considered secondary products, and little research was done to “purify” these products compared with the research conducted to increase the yield and safety of albumin. One modification by Kistler and Nitschmann (Table 2) that was intended to purify albumin used a lower ethanol concentration precipitate A (equivalent to Cohn Fraction II and III). With this, there also was a higher extraction and utilization of 40% ethanol at a lower pH for precipitate B and C, which improved the yield and purity of the end product of albumin.

At this time during the fractionation process, the gamma globulins that were isolated were mixed with numerous proteins other than monomeric IgG to the extent of more than 10%. This posed a problem for PIDD patients. The biggest issue was that these IGs, when administered intravenously, would aggregate, forming what would be perceived as a foreign antigen, resulting in a significant adverse reaction that was anaphylactoid in nature due to activation of complement, which put a patient at risk. Therefore, IG could safely be administered to a PIDD patient only by intramuscular or subcutaneous routes. The tissue would act as a filter to prevent the aggregates from crossing into the bloodstream and prevent a systemic reaction. Yet, local reactions were common with pronounced inflammation, swelling and pain. The maximum dose also was limited by these routes, which limited the overall increase in IgG level. A dose of 60 ml of a 5% product would require six to 12 injections at one time to provide only 3 grams of IgG. And, this dose was very low and inadequate to afford protection from infection.

The primary effort to improve the yield and safety of albumin may account for the nearly 40-year gap between the original fractionation process and the commercial availability of the first intravenous IG (IVIG). In order to prevent these adverse reactions, the immune serum globulin (ISG) needed to be stabilized from aggregation, and purification of the end product to monomeric IgG was necessary. The eventual development of a safe IVIG formulation enabled PIDD patients to receive therapeutic doses of IgG and marked an incredible change in their quality of life. To understand how this came about, it is necessary to understand the evolution of IVIG products and the rationale for each ingredient to improve the product by improving the process.

First-Generation IVIG Products (Figure 1)

While most of the products mentioned here are available in the United States, many have origins outside the U.S. The first IVIG product available in the U.S. was Gamimune produced by Cutter Laboratories. This first commercially available IVIG was a product of reduction and alkylation, which cleaved the disulfide bridges holding the Fc component of the IgG molecule to the Fab (fragment antigen binding) arms. Even though there was biological activity in prevention of infection and protection for these PIDD patients, it did not provide any immunomodulation effects. Gamimune was a 5% solution stabilized in liquid form with maltose, and refrigeration was required for storage.

The second IVIG in the U.S. market, Sandoglobulin (Sandoz) also was the first “intact” IgG preparation. The Kistler-Nitschmann fractionation process was further modified with an extraction at low pH (4.2) and stabilization with sucrose. This carbohydrate stabilizer was selected to prevent an increase in the glycemic index for diabetic patients. This product is lyophilized and required reconstitution with normal saline. The osmolality of a 3% solution was 498 mOsm/kg, while the 6% solution was 690 mOsm/kg and the 12% was 1074 mOsm/kg. These hypertonic solutions create a solution that is hyperviscous and can increase risk of renal insufficiency and thromboembolic adverse events. Even though this product contained the IgG subclass distribution in a close proximity of normal serum, its IgA content was approximately 720 mcg/ml. For patients who had an IgA-specific deficiency, there was concern that an anaphylactic reaction could occur if this product was administered. Yet, while the occurrence of having an IgA deficiency that was mediated by an anti-IgA (IgE or IgG) antibody was extremely rare, it still concerned physicians.

The third product that would still be classified as a firstgeneration IVIG was Gammagard (Baxter). This product also was lyophilized, but it used glucose as the stabilizer. The plasma fractionation method was Cohn-Oncley, but it also employed anion exchange chromatography for purification. The main focus of this product was to provide a low-IgA product. However, in doing so, the IgG3 and IgG4 subclasses were lower than seen in Sandoglobulin. Another European product, IVeegam, also was a low-IgA product, but it had no IgG3 subclass in the final product. For passive immunization, it was presumed that all four subclasses should be represented in the proportions seen in human serum. This theory was based on the observation that IgG1 and IgG3 worked as pairs as do IgG2 and IgG4.

With the entry of two IVIG “intact” products on the market, Gamimune was reformulated without using reduction/alkylation methods to provide the third “intact” IVIG. Gamimune N also was stabilized in glycine, but it did not contain maltose. It was stable at pH 4.2 as a liquid product that required refrigeration for stability. This product also contained no sodium chloride and had a relatively lower osmolality than the other two products.

pH Incubation

Fractionation methods that employed low pH would not only isolate the IgG monomers but also provide an antimicrobial effect. The lyophilized products would be reconstituted in either normal saline or sterile water, and the final pH of the solution for administration was in a normal range of 6.8 to 7.2. The liquid products would stay in solution in pH ranges of 4.2 to 5.5, a lower pH that was necessary to keep the IgG in a predominantly monomeric form. The lower pH did not induce metabolic acidosis in adults because it was a buffered solution, but caution in neonates had to be observed. In addition to this concern, administration of low pH IVIG in small peripheral veins could cause irritation and phlebitis. The use of catheters eliminated this concern.

Carbohydrate Stabilizers

The carbohydrate acts to prevent aggregation of the IgG molecules. The first IgG stabilizer was maltose. As a complex sugar, maltose does not change the glycemic index for diabetic patients and, therefore, does not need to be covered by insulin. However, some glucose monitors do not distinguish the difference between maltose and glucose and can lead to serious false-positive readings. This problem has led to serious adverse reactions when insulin is administered to a hypoglycemic with a false reading.

Sucrose was the next stabilizer used for IVIG. Like maltose, it does not require insulin coverage. Even though the concentration of sucrose is listed as 5%, that is only true when the IVIG concentration is 3%. Normally, Sandoglobulin was administered as a 6% solution and, therefore, the sucrose concentration was 10%. At that concentration, the product was hyperosmolar. Increasing reports of renal insufficiency and renal failure were associated with sucrose-stabilized products.

Glucose also is used as a stabilizer. For diabetic patients, the impact on the glycemic index needs to be accounted for and adjusted during administration.

Viral Risk for First-Generation Products In the 1950s, the Cohn method was an open-vessel production that would not be considered sterile, and it was not pasteurized until the 1960s. The cold ethanol treatment did provide a method of protection from microbial contamination. The strongest evidence for this statement was that there were no cases of HIV transmission through the use of IVIG during the 1980s, when HIV had been transmitted through plasma products and antihemophilic factor products. These products later were heat-treated to prevent transmission of HIV.

In the mid 1990s, there were reports of hepatitis C being transmitted through IVIG products, but they were the ones that used anion exchange chromatography as a method to purify products rather than low pH. These products were recalled and reformulated by using solvent-detergent (S/D) to prevent the transmission of lipid-coated viruses such as hepatitis C. Because S/D does not affect non-lipid-coated viruses such as hepatitis A or parvovirus, other antiviral steps needed to be considered.

Second-Generation IVIG Products

The focus of the second-generation products was to eliminate the risk of viral or even prion transmission from IVIG products. The risk-reduction strategy looked at sequential steps that would remove, partition or destroy the virus. Keeping in mind that new viruses and viral mutations can and do occur, the use of methods with different mechanisms would make the process more robust.

All of the current products use either the Cohn-Oncley or Kistler-Nitschmann method for plasma fractionation. The cold ethanol process itself does provide some measure of viral protection. Low pH or incubation at pH 4 to 4.2 also will have an additive impact on antiviral methods. This method increased the efficiency of the process and reduced product protein loss. Anion exchange chromatography is employed to target the extraction of the IgG subclasses and remove IgA, but it also serves to partition the viruses from the protein. This method alone, however, is not sufficient to prevent all viral transmission.

The second-generation products often indicated a viral safety methodology in their brand name such as Gammagard S/D and Gamimune N S/D. Sandoglobulin, which changed its name to ZLB-globulin and now Carimune, was a secondgeneration product, but it did not employ the use of solvent detergent. This product was never implicated in hepatitis C transmission, and the company felt that its method of low pH incubation and trace enzymatic (pepsin) treatment was sufficient for currently known viral risks.

Third-Generation IVIG Products

When it was determined that humans could be exposed to Creutzfeldt-Jakob disease (CJD) by ingesting meat from cows, with the scare of mad cow disease, there was great concern that blood donors with CJD could expose this risk to recipients of fractionated blood products. Better methods of screening incubation and PCR (polymerase chain reaction) testing were employed on the blood or plasma collection side, but there needed to be a way to ensure that TSE (transmittable spongiform encephalopathy agent) would not contaminate the IVIG products. There was equal concern about new viruses such as West Nile and the risk of transmission.

Gamimune N was replaced with Gamunex (IVIG-C) when caprylate (octanoic acid) was used to replace solvent detergent. The advantage of caprylate was that it was a natural plant (8 chain fatty acid) that did not have to be completely removed (as opposed to S/D) and it was more effective in a shorter time frame. These changes allowed for a more efficient recovery of highly purified (>98%) IgG. This was the first product to use caprylate in its process.

The next step was the introduction of nanofiltration. Many products go through an ultrafiltration process that removes non-IgG proteins and some larger viruses, but the introduction of nanofiltration was adopted to make Carimune NF. It was essentially the same product as Carimune but with the additional nanofiltration step. It also was the first product to have an approved TSE removal step.

Another concern was the risk of renal failure induced by IVIG. IVIG-associated acute renal failure was first reported in 1987. The U.S. Food and Drug Administration (FDA) received more than 114 reports worldwide (87 in the U.S.) of acute renal failure associated with the administration of various IVIG products. The vast majority of these were related to products that were stabilized with sucrose, but there also were occurrences with other carbohydratebased stabilizers such as glucose and maltose. The patient risk factors included those older than age 60, diabetes mellitus, sepsis, proteinuria, renal insufficiency and concomitant renal risk medications. A black-box warning was issued by the FDA for all IVIG products to warn of such risk.

As the doses of IVIG continued to increase for the treatment of autoimmune diseases, another risk was discovered. The risk of thromboembolism, including myocardial infarction (MI), was reported in the literature, but the cause was not apparent. It is believed that a combination of risks, including former embolic events and age in combination with the administration of hyperosmolar and hyperviscous solutions administered at a rapid rate, could result in thromboembolism. The immediate recommendation was to use lower concentrations at slower rates, but the concentration of the protein may not be reflective of the osmolality of the solution.

Fourth-Generation IVIG Products

It may be debatable whether or not there is a fourth generation, but the distinction is the movement away from carbohydrate-stabilized IVIG products to those that are in liquid form and amino-acid stabilized. Glycine was used in the very first IVIG product (Gamimune) and is still used today in Gammagard Liquid and Gamunex, both 10% liquid products. The latest amino acid stabilizer is L-proline, which is used in Privigen 10% Liquid. L-proline provides a strong hydrophobic environment to increase monomeric IgG. The theoretical advantage of this amino acid would be in the reduction of routine adverse events such as fever, chills and flushing, and headaches. The new liquid products that are at a 10% concentration do not contain sugars and only a trace amount of sodium and, thus, have much lower osmolality than their earlier-generation products. These changes are intended to minimize both the renal and thromboembolic risk associated with IVIG administration.

Conclusions

Between 1940 and 1980, very few changes in the manufacturing process left ISGs as products of minimal impact for passive immunization for patients. It was not until the use of additional steps, which included purifying globulin fraction and preventing aggregation to allow IV administration, that the world of immune therapy changed.

The continuous process of improving the manufacturing process has led to products that are virtually free of microbial risk. As the majority of use has evolved from passive protection for patients with immune deficiency diseases to patients with autoimmune diseases, additional adverse events have developed. Most likely due to administering higher doses at faster rates, such issues as renal failure and thromboembolic risks surfaced. The response to these issues resulted in products that are stabilized with amino acids instead of carbohydrates, resulting in products with a higher concentration that are still iso-osmolar.

What are the next steps to develop the ideal IVIG? A single process that would eliminate the risk of all viruses, prions or future microbial mutations would be ideal. While the current IVIG products are virtually free of risk or microbial transmission, they require multiple and expensive steps to achieve that.

If a higher-concentration IVIG product could be achieved without causing a hyperosmolar product, that would be desirable. The highest concentration of IVIG is currently 10%, which still requires a high volume for administration. The complete elimination of IgA without removing the IgG subclasses would eliminate any concern for anaphylaxis. Even though the risk related to this phenomenon is very rare, it still does exist.

An ideal agent would be one that can be administered in a relatively short time (less than one hour) and that does not cause the usual headache, fever, chills and malaise commonly associated with administration of IVIG. The vast majority of the adverse events related to IVIG are mild and infusion-related. And, the degree of tolerability to treatment depends on the product profile. But, each patient has different degrees of tolerability, and slower infusion rates are often necessary.

Finally, development of an IVIG in its final form ready for administration in common final dosage sizes would be ideal. Currently, since the doses of IVIG usually range from 0.4 to 1.0 grams/kg, the dose exceeds the content of one vial. The largest vial size currently is 20 grams; therefore, either administering sequential vials or pooling the vials into a separate product container is necessary. And, the risk of contamination, error and waste are associated with this common practice.