7.2 Immunogenicity

be examined as well as factors having an influence on immunogenicity such as the disease state wherein the immune systems may be compromised such as in the use in patients undergoing chemotherapy. Immunogenicity is a problem only when it is clinically relevant—when it has an effect on the safety or the efficacy of the therapeutic protein. Clinically relevant immunogenicity includes when antibodies change how the drug reacts in the body, when antibodies make the protein less therapeutically effective, when antibodies change natural proteins in the body, or when antibodies trigger a severe allergic reaction, which is very rare. Clinically relevant immunogenicity is not common but must be monitored for all therapies.

It is well established that repeated injection of even native human proteins can result in a break in immune tolerance to selfantigens in some patients leading to a humoral response against the protein that is enhanced when the protein is aggregated or partially denatured. Although in most cases an immune response to a biopharmaceutical has little or no clinical impact, ADAs do, however, pose a number of potential risks for the patient, particularly in the case of a neutralizing antibody response. Firstly, an ADA response can adversely affect the PK and the bioavailability of a drug thereby reducing the efficacy of treatment and necessitating either escalating the dose or switching to alternative therapy if such therapy is available. An ADA response can also adversely affect the safety of treatment and cause immune complex disease, allergic reactions and, in some cases, severe autoimmune reactions. Serious and lifethreatening adverse events can occur when ADAs cross-react with an essential, nonredundant endogenous protein such as EPO or TPO. Thus, several cases of PRCA were associated with the development of antibodies to recombinant EPO following a change in formulation. Similarly, the development of antibodies to PEGylated MGDF cross-reacted with endogenous MGDF, resulting in several cases of severe thrombocytopenia. All biosimilar products are evaluated based on the regulatory guidelines such as the FDA guidance for binding antibodies and neutralizing antibodies. Binding antibodies bind to the protein but usually have no effect. Neutralizing antibodies can inhibit the function of the protein in the body. The FDA is more concerned with neutralizing antibodies because they are more likely to have clinical consequences. Because older products may have limited immunogenicity data based on tests with inadequate sensitivity, immunogenicity between a biosimilar and its reference product cannot be compared using data from the package insert of the reference product. Any comparison of immunogenicity will need a side-by-side clinical test of the biosimilar and its reference to Safety similarity

265

Biosimilarity: The FDA Perspective

266 ensure valid comparison. Without a side-by-side comparison, more sensitive tests may get higher antibody-positive results with the biosimilar. Animal models do not predict immunogenicity in humans. Most animals (even primates) can develop a strong antibody response to human proteins. Immunogenicity testing in animals may be useful to evaluate drug functioning or toxicity changes that might result because of antibodies to the human protein, or to see if responses are the same to two different products. Aggregation, which is when proteins clump together, is the most common factor associated with increased immunogenicity. Aggregation should be a key part of analytical testing, but other product changes (e.g., impurities) have not been associated with increased immunogenicity. The amount of premarket and postmarket immunogenicity data needed for the approval of a potential biosimilar product will depend on an analytical assessment of similarity between the biosimilar and its reference product, as well as the rate of clinical consequences of immunogenicity observed with the reference product. If an immune response to the reference is rare, two separate studies may be sufficient to evaluate immunogenicity: (a) premarket study to detect major differences in immune responses between biosimilar and reference and (b) postmarket study to detect subtle differences in immunogenicity. The FDA recommends that immunogenicity tests use the patient population that is most likely to show an immunological response for these studies. Product changes associated with increased immunogenicity can be assessed using analytical tests. Protein aggregation is the primary product change associated with increased immunogenicity. Rigorous analytical testing comparing the biosimilar candidate to the originator reference product in head-to-head studies is the most sensitive way to test for likely immunogenicity in patients. These tests include the measurement of protein aggregates in the drug product and over its shelf life. Thus, analytical testing may be the best way to minimize immunogenicity of a biosimilar. While such tests also need to be done for originator biologics, it is much more difficult to anticipate likely immunological responses with a brand new product. Most clinical comparator trials cannot detect true differences in clinically significant immunogenicity because the frequency of events is so low. With current statistical methods, very large clinical studies, with over 3000 patients each, would be needed to evaluate meaningful differences in immunogenicity. Consequently, a robust pharmacovigilance program, able to capture clinically relevant immunological responses during real-world use of the biologics, may be a better method to detect clinically relevant immunogenicity problems. Differences in antibodies may not be relevant if there are no clinical consequences for patients. It is not expected that the possibly reduced immunogenicity of a biosimilar, through more modern manufacturing

and better control of aggregation, will cause the FDA to remove the reference product from the market. Eprex immunogenicity is often cited as a reason to demand clinical immunogenicity testing in biosimilars. Johnson & Johnson made a manufacturing change to Eprex (recombinant EPO) and removed a protein, human albumin, from the formulation of their product marketed in Europe. This change was overseen by regulators using the standard process of demonstrating high similarity with comparability studies of the pre- and postmanufacturing-changed products. The “new” Eprex induced antibodies to Eprex and to natural EPO found in the body, causing PRCA. Although the individual cases of PRCA were very serious, the actual incidence was low (2/10,000). Clinical studies could not have detected PRCA at such a low incidence. A clinical study to determine a difference in the rate of PRCA would have to have involved a very large number of patients. Instead, a robust pharmacovigilance system with analytical investigations eventually resolved the issue. The predominant immune mechanisms leading to drug hypersensitivity may include from no effect to endogenous cross-reactivity (Table 7.1). The risk of drug hypersensitivity can be increased by some patient-related factors, which include female gender, specific genetic polymorphism, as well as by some drug-related factors, which include the chemical properties, the molecular weight of the drug, and the route of administration. It is known that drugs with great structural complexity are more likely to be immunogenic. However, drugs with a small molecular weight (less than 1000 Da) may become immunogenic by coupling with carrier proteins, such as albumin, forming complexes. Moreover, the route of administration affects the immunogenicity, the subcutaneous route being more immunogenic than the intramuscular and intravenous routes. Table 7.2 lists the incidence of immunogenicity of recombinant drugs. It is noteworthy that about 5% of the U.S. population is allergic to food, and or recombinant proteins and antibodies, many fall within or below this incidence rate. Compounds like sargramostim, aldesleukin, and

Table 7.1 Listed Effects of Immune Responses from Various Drugs

Result of Immune Response Drug

No effect rh-GH PK alteration rh-Insulin Reduced efficacy GN-CSF, interferon alpha, interferon beta (the majority of patients become neutralizing antibody positive (NAbs+) within 6–18 months of treatment, while clinical impact of NAbs is delayed and is not seen until 24 months of therapy, abolishing activity Loss of efficacy Natalizumab (persistent antibodies) Cross-reaction with endogenous Factors VIII and IX, rh-EPO, rh-MDGF Safety similarity

267