According to the FDA, a biosimilar product must be highly similar to the reference product and possess no clinically meaningful differences. This means that approval is based on the totality of the comparison of the biosimilar to the original reference product. Neil Chaudhary, PhD, CEO of Neocytex, said the challenge is developing biologic drugs as biosimilar products with accurate glycosylation patterns that meet FDA guidelines.
Biosimilar manufacturers don’t have the luxury of veering far from the originator product. Divergence from critical quality attributes, of which glycosylation has been listed by the FDA as a key variable, would prompt additional and costly bioassays. The problem in biosimilar production is the lack of trade secret cell lines and precise manufacturing conditions that can markedly alter sugar processing and composition. Thus, biosimilar manufacturing must begin by high-quality analysis of originator mAbs, followed by years of trial and error in testing cell lines that can produce similar products.
Currently, there are a number of mAb therapeutics that have been developed as biosimilars, including Humira, Avastin, and Remicade. A gaggle of others are coming off patent and are tempting candidates for exploitation. However, according to Chaudhary, monoclonals that are heterogeneously glycosylated may be extremely difficult to replicate, constituting a huge barrier to entry for competitors. This can work to the benefit of the original developer of an antibody therapeutic, whose efforts are protected by the difficulty of reproducing an acceptable copy.
“The best strategy for companies working on new drugs would be to use ‘trade secret’ cell lines and production methods that result in extremely heterogeneous glycosylation patterns,” Chaudhary suggested. “These would be quite difficult to reproduce, therefore the best strategy for the developer to protect his investment would be to favor complex glycosylation patterns.”
But for the biosimilar competitor, the best tactic would be to employ modern analytical methods to identify the major bioactive glycosylated variants. Once defined, the competitor would need to engineer cell platforms that can produce those variants. This is extremely challenging, as new cell lines would have to be created.
Chaudhary noted that the challenges posed by the needs of appropriate antibody glycosylation are ideal for attack through the application of the “Bioprocessing 4.0” paradigm. “The features of these innovative approaches, which include automated data processing and addressing very large number of possible parameter combinations constitute important support for designing the strongest combination of antibody structure and glycosylation patterns,” he stated. “My colleagues and I are considering how we can apply large inputs of robotics data to achieve optimal bioprocessing parameters for engineering our glycosylated monoclonals.”