Biosimilars, which are new versions of innovator biopharmaceutical products that are marketed after expiration of patents, have emerged as one of the fastest growing development opportunities in the biopharmaceutical sector. In the U.S. alone, industry analysts estimate that biologics worth $80 billion are slated to go off patent by 2015.
Regulatory agencies evaluate biosimilars based on their level of similarity to, rather than the exact replication of, the innovator drug. In the U.S., recent guidance by the FDA says it will "consider the totality of the evidence" when assessing follow-on products. This approach requires sponsors to demonstrate robust chemical comparability to the innovator compound.
The need for increased analytics and the desire for compressed timelines in biosimilars development demands in particular that developers must invest early in Chemistry, Manufacturing, and Controls (CMC)-type analysis to demonstrate comparability to the reference molecule at every stage, particularly during manufacturing.
Due to the complexity of biologics, a product can only be made that is similar to the innovator drug, not identical. Basically, it's impossible for two different manufacturers to produce two identical products even identical host expression systems, processes, and equivalent technologies are used. Therefore, we have to rely on analytics to compare the biosimilar to the innovator product on the market.
Advances in current state-of-the-art analytical methods enhance the likelihood that a product will be highly similar to another product through better targeting of the original product's physicochemical and functional properties. Sponsors with compelling comparability data observe a reduced regulatory burden.
Europe has the best-established framework for biosimilars, with which the U.S. is now becoming aligned. In February 2012, the FDA issued formal draft guidance on biosimilars titled "Scientific Considerations in Demonstrating Biosimilarity to a Reference Product," in which it states that since a one-size-fits-all pathway is not possible, it will "consider the totality of evidence" when assessing follow-on products. The cornerstone of this approach is the structural and functional analyses of the proposed molecule demonstrating comparability with the reference drug.
The FDA has the discretion to determine that certain studies are unnecessary in a 351 (k) application, the approval pathway for biosimilars, which is one of the reasons that early engagement with regulatory authority is vital to ensure expedited approval.
Biosimilars must be systematically engineered to match the reference product. A comparability exercise must be followed with the innovator product at all levels of product development, including: physicochemical attributes, biological activity, preclinical in vivo comparability, Phase I PK and safety, and Phase III efficacy and safety.
This can be difficult because data for the innovator product will be lacking. The only way to get information about the components of the innovator product is from material that is already out in the marketplace. Having multiple batches of the innovator's product, spanning a number of years, can be extremely helpful during the characterization process.
Another important thing to keep in mind during comparability studies is that there is not only inherent variability in biosimilars, but also in the innovator product. This inherent variability means that the biosimilar will never be identical to the innovator product. However, what's important to decipher is how these "differences" impact the clinical outcome. For instance, some differences will have no impact on approval, while other cases might undergo a change that triggers a concern about the clinical impact.
A 2011 paper by Sandoz Pharmaceuticals titled "Acceptable changes in quality attributes of glycosylated biopharmaceuticals," provides excellent data in support of the "inherent variability" concept in all biologic molecules. The paper demonstrates through three case studies that simple process changes which occur in the manufacturing process of biosimilar products can be deemed acceptable. It provides examples of the characterization of batches of darabeopetin alfa from the EU, rituximab, and enbrel, all of which resulted in no change in licensure.
In order to demonstrate that you have a similar product, you need to have your analytical assays in place during the process of development. These assays will help to best replicate what the innovator has done. Overall, the characterization you need to do for a biosimilar will always be much higher than that for a new biological entity (NBE).
The analytical characterization of a biosimilar should include primary, secondary, tertiary, and quaternary structural assessment, biological activity, and analysis of product and process impurities. All of these components must be understood and characterized during your comparability studies of the biosimilar to the innovator product.
A number of innovations that have been surfacing over the last couple of years to assess these protein parameters are now an expectation for a comparability study. For instance, mass spectrometry has become integral to protein structural analysis. In addition, separation technology, C-IEF, plus HDX and immobility are opening up new frontiers in how we're looking at protein structures in complex biologics. Overall, there is no single analytical technique that will tell you structurally what the protein has to offer, a variety of tools are required.
Following are a few real-life examples of analytical tools we used at Covance to compare a biosimilar to the innovator product:
Mass Spectrometry: In one case, we observed a molecular shift that indicated there was something different about the biosimilar in relation to the innovator product. This particular case was due to an incorrect amino acid sequence for the biosimilar, requiring subsequent analytical analysis.
Hormone Receptor Binding by Biacore: An epitope we were probing showed good consistency. For the drug substance the difference was 4% and for the drug product, 2%. With an acceptance criterion of 20% variability, we assessed similarity between the biosimilar and innovator product.
C-IEF: C-IEF is a work-horse for a lot of the biosimilars. It's a good assay to use with in-process samples, as it provides quick turnaround.
Forced Degradation: In addition to structural analysis, it's important to show stability comparability between the biosimilar and the innovator. In one case we did a forced degradation to see how the two products behaved. We discovered different degradation rates in the products, which required us to perform subsequent analytical testing.
N-Glycan Similarity: The N-Glycan is a more refined assay that we use as a platform. It provides more resolution in where the actual differences are than some other methods.
Peptide Map: The most highly resolving, differentiating assays for enzymes or combination of enzymes.
In an ideal world, product heterogeneity would be clearly understood, with variants easily isolated and characterized. Also, no process changes would occur during manufacturing. Even better, you'd have a 'crystal ball' to know what to look for in the innovator product.
Of course, this is not what the real world of biosimilars looks like. In the real world, biosimilars are like snowflakes, no two are the same. For instance, if you get a biosimilar for an innovator product from one company vs. another, there are going to be similarities, but more often a lot more differences. Therefore, there are no canned assays for biosimilars, as you have to address the snowflake effect. At Covance, we're seeing a full range of comparability work from our clients - from very detailed comparability characterization to the innovator all the way to incomplete comparability characterizations.
Time-to-market is critical for all of these products, which results in rapid time pressure situations, including scaling-up for manufacturing, turning around in-process sample analysis, plus developing and validating discriminating assays.
Some other real-world observations include: 1) stability testing in addition to structural analysis should be required to determine rate of degradation is similar between the two products; 2) you need to understand impact of product shelf-life on results; 3) many times the drug product will need to be isolated to demonstrate that the integrity was not compromised; 4) comparability continues after release and 5) meet with Boards of Health (BOH) as quickly as possible.
In terms of protein analysis, the real world requires an integrated set of analytical methods that can evaluate all domains and protein modifications. State-of-the-art techniques are expected to be used, with each analytical method having its own strengths and weaknesses. There is absolutely no crystal ball that will tell you the exact components of an innovator product.
It's an exciting time for biosimilars. The regulatory guidelines will continue to evolve as we get more experienced and biosimilars continue hitting the market. We'll also see demand for biosimilar CMC development continue to grow as it plays a critical role in demonstrating comparability to the reference product. As we're dealing with an inherently variable system, we need to look at biosimilars development from an integrated standpoint.
Covance can help you with your biosimliar development. Visit our biosimilars page for more information.
About Raymond Kaiser, Ph.D.
Dr. Raymond Kaiser is the head of BioPharmaceutical CMC Solutions at Covance. Dr. Kaiser is an expert in the research, development, QA/QA and technical support of biologic, biosimilar and vaccine products. He has spent over 20 years in the biopharmaceutical industry and published over 50 papers and patents on various aspects of the development, characterization and manufacture of biologics and vaccines. Prior to joining Covance, Dr. Kaiser was Executive Director in the Bioprocess R&D Department at Merck Research Laboratories.