Regulation of Biosimilars: A Matter of Variability, Similarity, and Comparability
The European Union is strengthening its pioneering role in the regulation of biosimilars by further developing the basic rules for determining the levels of compatibility for this group of drugs. There are, however, some key issues that are not easy to resolve, as evident in a recent workshop on biosimilars organized by the European Medicines Agency (EMA). The meeting, held in October 2013, was called to consider issues raised by three of EMA’s revised guidelines on biosimilars, one covering general principles, the second on non-clinical and clinical matters, and the third on quality issues. The deadline for comments on the first two guidelines was the end of October, whereas for the third issue regarding quality, it was a year ago, indicating that despite the time lapse, there was still a need for further discussions to deal with the difficulties of publishing a final draft.
Revised guidelines on general principles
In its revised guidelines on the general principles for biosimilars, EMA states that the reference product must be a medicine authorized in the European Economic Area (EEA) (1). However, the EU also confirms that “it may be possible” for the biosimilar to be compared, “where needed,” with a non-EEA authorized product in clinical and non-clinical studies. Because the non-EEA authorized product would have to follow similar scientific and regulatory standards as those of the EMA, it would have to be in a country that is a member of the International Conference on Harmonization (ICH), according to the guideline. Also, the biosimilar applicant using the non-EEA route would have to provide bridging data from structural and functional analysis, and possibly from pharmacokinetic and pharmacodynamic studies, to show that the non-EEA drug was “representative” of the EEA reference product. Eugene Corretge, speaking on behalf of European Biopharmaceutical Enterprises (EBE), representing both makers of original biologics and biosimilars, said that the revised guideline was unclear about whether bridging data would be needed when a comparability study was conducted only partially or in totality with a non-EEA authorized medicine.
Karl Heinz Emmert, vice-chair of the European Biosimilars Group (EBG), part of the European Generic Medicines Association (EGA), stated that the revised guidelines put forward “strong scientific principles” to support the use of non-EEA drugs sourced from ICH regions as reference products. There could be more consistency between products from different licensing jurisdictions than those from a single licensing jurisdiction, he explained. “Reference products from one original manufacturer and sourced from a different jurisdiction are often not distinguishable,” Emmert said. “A reference product supplied in one jurisdiction is often manufactured in different facilities.” Yet, it would not require non-clinical or clinical bridging studies, he added.
Comparability of the reference products of one original manufacturer but from different jurisdictions could be established by similarity of quality based on stringent analytical and biological assays, explained Emmert. “Non-clinical and clinical studies are not suitable to establish differences between the ‘same’ reference products of one original manufacturer,” he said. “Analytical and biological assays are much more sensitive and are, thus, much more suitable.”
“For structurally more simple biological medicinal products, a comparative clinical efficacy study may not be necessary if similarity of physicochemical characteristics and biological activity or potency of the biosimilar and the reference product can be convincingly shown,” said Martina Weise of the Federal Institute for Drugs and Medical Devices (BfArM), the German regulatory agency. In this respect, the drawing up of a quality target product profile (QTPP), which summarizes the required quality characteristics of the biosimilar, was regarded as being crucial. “It would be primarily based on extensive characterization data from the reference products,” explained Nicklas Ekman of the Finnish Medicines Agency. “It would be detailed at an early stage of development, forming the basis of the development of the biosimilar product and its manufacturing process.”
Beverley Ingram, an EBE representative, said her association recommended reinforcing the role of the QTPP in a stepwise approach with it being quantitative by nature. However, setting quantitative, meaningful criteria is a technically complex topic, which she suggested should be investigated further in a reflection paper proposed by EMA on the use of statistics in comparing biological drugs, including biosimilars.
Joerg Windisch, EBG chair and chief science officer of Sandoz Pharmaceuticals, stressed that comparability between biosimilars and their reference products included having similar variabilities, particularly those between batches. “No batch of any biologic is ‘identical’ to other batches,” he explained. “Variability is natural even in the human body and is usually not problematic.”
Significant differences would be those that have a clinical impact, Windisch said. It is, therefore, important to have a systematic and quantitative assessment of quality attributes to pick out those that are crucial based on their clinical relevance. These critical attributes are one of the key elements that determine whether a drug is acceptable as a biosimilar, he said.
Another potential source of unacceptable incompatibilities is the use of contrasting expression systems in biosimilar proteins. Windisch pointed out that a distinction had to be made between different types of existing expression systems and those that are novel. Different kinds of existing systems could provide similar quality attributes. Also, because they have been applied for some time, these systems would have “equal or better” safety track records. On the other hand, novel expression systems were by definition unknown entities with an absence of safety data. “The use of expression systems with no safety record is a lot more challenging and is something you do not want to risk doing in the development of a biosimilar,” Windisch said. Differences in a drug that go beyond the range of variability in the reference product, particularly when they have clinical significance, are likely to be rejected as being a biosimilar by regulators.
Workshop attendees were told about the complications of setting general limits on acceptable differences between biosimilars and reference products. “You cannot provide that sort of clarity,” said Weise. “It would be very difficult to set one margin that fits everything. These limits have to be set on a case-by-case basis.”
In respect of an individual-item approach to products, some speakers pointed out that with some biosimilars, comparability should not be confined to the active substances. They should cover the whole formulation, including excipients and diluents because of their effect on the medicine’s safety and efficacy. A similar procedure was suggested for dealing with the issue of immunogenicity. “There are molecules on the market whose immunogenicity has not been established,” said Windisch. “It’s up to the applicant for authorization of a biosimilar to justify that no immunogenicity studies are necessary. In most cases, this seems to be unnecessary but we’ll have to see how the science develops on this issue.” Because of the relatively short history of biosimilars, there is a lack of knowledge about some of the important aspects of their safety. Hence, there is still a lot to be learned about how they should be properly regulated.