Teva Pharmaceuticals recently introduced Tbo-filgrastim in the U.S., a drug similar to Amgen’s biologic drug Neupogen. Both drugs boost the production of infection-fighting white blood cells in certain cancer patients receiving chemotherapy. Teva currently sells the drug in the European Union, where they hold five percent of the current market share.
Although Teva used FDA’s standard drug approval process by filing a Biologicals License Application for Tbo-filgrastim, it is precisely for these kinds of drugs that the Public Health Services Act, Section 501k, was introduced. The goal of these draft guidelines, which govern the process for review and approval of drugs deemed interchangeable with original biotech products, is to abbreviate the pathway for biosimilars. This should translate into more affordable drugs. The question of how much more affordable depends on how much heavy lifting needs to be done during preclinical and clinical trials.
The market for these kinds of drugs is growing rapidly. Biosimilars are forecast to account for 7 out of 10 of the top prescription drugs by 2014. And according to IMS Health, biosimilar sales are expected to reach $1.9 billion to $2.6 billion by 2015.
A Shifting Paradigm
Over the past few years many pharmaceutical companies have shifted their attention from small molecules to protein therapeutics. These companies face challenges as the focus on safety has shifted from traditional toxicity endpoints to immunogenicity, with scientists previously working on small molecules now dealing with biologics. Scientists have to change their whole mind set with the development of biosimilars. To put it in perspective, the difference between the development of small molecule generics and large molecule biosimilars is like the difference between driving a car and sailing a boat. Driving is precise; steering and speed can be controlled with ease. Sailing, however, is full of ambiguity. You will have to navigate toward your destination, and your speed very much lies outside your control.
Small molecule drugs are well defined and have a set production process to follow, which makes them reproducible. Pharmaceutical companies can simply obtain the originator patent information from the U.S. Patent and Trademark Office or from published literature.
On the other hand, biologic drugs are large, complex molecules and are much more difficult to characterize and define. Despite years of experience with drugs under patent protection, companies still have difficulty replicating their own production process, and small molecule scientists may be making the transition to the challenge of large molecule production.
Following The Guidance of The FDA
The FDA draft guidance related to “Scientific Considerations in Demonstrating Biosimilarity to a Reference Product” was issued in February 2012. In the draft, the FDA considers the totality of the evidence provided to support a demonstration of biosimilarity. It also includes structural and functional characterization to assess whether a copy is highly similar to a reference product.
The biosimilar development process reflects a sequence of narrowing steps (Figure). The process of creating a biosimilar begins with extensive structural and functional characterization followed by increasingly targeted animal and clinical studies to validate that there are no differences. This stepwise approach can also include: general scientific principles in conducting comparative structural and functional analysis; animal testing; human PK and PD studies; clinical immunogenicity assessment; and clinical safety and efficacy.
Following these steps as well as performing structural and functional characterization to assess whether a copy is highly similar to the reference product is important in minimizing the burden of preclinical and clinical testing. However, we agree with the FDA on the statement that side-by-side clinical testing is essential when done between the copy and the original because variables such as structure and choices made during production can influence the nature of the biologic. Advancements in manufacturing and production methods have increased the likelihood that a biosimilar product can be deemed highly similar, but no matter how sophisticated and advanced analytical tools have become, it still may not be possible to detect relevant and functional differences between two proteins. These complexities make exact replication of the originator’s active molecule nearly impossible.
Within the draft guidance itself, the FDA states that data derived from analytical studies, animal studies, and a clinical study or studies are required to demonstrate biosimilarity unless FDA determines it unnecessary. However, in most cases, human PK and PD profiles of a protein product often cannot be adequately predicted from functional assays and/or animal studies alone.
Two of the most important and most difficult aspects of biosimilar clinical testing are bioequivalency (purity and potency) and immunogenicity (safety).
The bioequivalency of biosimilars must be established through assays using antibodies to extract the biologic from the sample. However, using a biologically derived technique for assessment complicates testing. Because antibodies used by the originator are generally unavailable, the biosimilar producer will have to generate their own antibody. Since these antibodies are critical to obtaining comparative data, they need to be fully characterized and the specificity and affinity of the antibodies for each molecule determined. The end point of the bioequivalency testing is to demonstrate equivalency in pharmacokinetics of the originator and the biosimilar in preclinical and clinical studies.
Biosimilars are highly diverse, results can vary from test to test, and assays can “see” things differently. For example, even if the drug is not bioequivalent, the antibody may not be able to distinguish one from the other. This can happen if the antibody binds to the same molecular component in both the originator and the biosimilar. On the contrary, the antibody could also be seeing the drug differently because of the antibody binding differently to an originator and biosimilar due to their unique glycosylation patterns, when in reality they are actually bioequivalent. These differences in the way antibodies see biosimilars can cause problems and may affect the overall effectiveness of the drug and how the body reacts to it.
A practical approach to determine bioequivalency is to use one assay for both molecules, using the biosimilar as the calibration curve, and quality control samples from the originator and the biosimilar. This provides one assay for the biosimilar the originator. The samples and controls from each will be assayed using the same platform, the same set of reagents and conditions, and will only require one calibration curve.
With regard to immunogenicity, positions range from suggesting that pre-approval testing of follow-on biologics should include the identification of the product factors with the greatest risk for immunogenicity to beliefs that immunogenicity testing should be the same for an innovator and follow-on product.
One of the main difficulties in immunogenicity testing is detecting small immunogenic differences between biologics. Small molecules rarely elicit an immune response, but biologics that are developed from living cells can interact with patients in unforeseeable ways. Although these molecules are very large, the immune response may be elected by epitopes as small as seven to nine amino acids. The typical causes leading to immunogenicity are: impurities, degraded protein, aggregates, formulation, and the route of administration. Reviewing literature of the original drug will provide companies with an indication of the extent and timing of the testing that is required to evaluate immunogenicity.
A very important aspect of immunogenicity is the neutralizing antibody assay. These assays are often cell-based and detect the biological action of the drug. Generally these cell-based assays are modifications of the activity assay used to assess potency of the protein drug and are complex and sensitive to many variables. As a general rule, it is best for companies to have their neutralizing antibody assays prior to start of repeated dosing or at least before the start of patient studies.
Even if you know what you are starting with, neutralizing assays will take six to nine months to develop and validate.
A Look Ahead
Despite the recently issued FDA draft guidance, controversy still remains on many issues including standards for interchangeability, biosimilar naming, the use of foreign data, and the specific circumstances in which clinical trials will be required. In addition there is even controversy in the willingness of biosimilar sponsors to adhere to the requirements for disclosure of their applications to innovator companies. However, the industry will continue to take steps forward to develop biosimilars in the shortest, most cost-effective way possible without any undue risk to human safety.