It has been well established through many studies over recent decades that glycans play key roles in a wide variety of different biological processes and functions. These can either be through direct action, such as receptor binding (e.g. the asialoglycoprotein receptor or inflammatory response pathways), or through influence on overall molecular structure, resulting in a control or modulation of glycoprotein function (think monoclonal antibody fucosylation and ADCC function).
So, armed with this knowledge, can we look a little more deeply into the relationship between glycans and molecular function or properties, at least for certain key biopharmaceuticals, and can the glycan profile be used as a measure of, or have a correlation with, certain aspects of functional testing?
The relationship between glycan structure and molecular interaction was recently investigated as part of a study by Welch et al (1). Here the investigators looked at 21 monoclonal antibody biosimilars that had been approved by the FDA with the focus on high mannose glycosylation. This was of interest to the researchers due to the fact that increased levels of mannosylation in monoclonal antibodies (mAbs) have been associated with increased clearance from the bloodstream via a mannose-specific lectin located in the liver.
Initially the investigators looked at the analytical methods used for glycan characterization across the submitted studies. Most of the applicants whose data were evaluated in this study had used chromatographic analysis of fluorescently labelled glycans, an approach we use at BioPharmaSpec in conjunction with mass spectrometric identification of the labelled glycans.
Data obtained from LC-analysis of 2-AB labeled N-linked oligosaccharides. The N-linked oligosaccharides were released from a monoclonal antibody using PNGase-F.
Where the same reference monoclonal antibody had been used by different applicants, it was found that the data showed a comfortingly high degree of similarity for the relative abundance of components across the laboratories, where the same glycan analytical technique had been used.
The authors then moved on to an investigation of seven specific mAbs within the 21, where it was shown that the biosimilar profile did not overlap the innovator profile for the Man5 high mannose component. What they found was that in each case, despite the high mannose relative percentage being out of range, the PK values were all within the acceptance range. There was very little difference between samples in terms of PK value and range but the sample with the highest relative difference in high mannose glycans between innovator and biosimilar (5.9%) also had the highest PK value (although still within the acceptance criteria range).
Their conclusion from the study was that differences can be detected in high mannose glycan populations through analytical means but that these, up to a certain point, do not translate into differences in PK within a defined acceptance range.
Therefore, analytical data can be key in identifying possible differences between biosimilar and innovator but differences observed will not necessarily translate into clinical differences unless they pass a certain relative threshold, which will be dependent on the structural feature under consideration.
These findings and the application of other forms of cross-correlation between structural and functional data could be very useful going forward in determining when differences observed in sensitive structural investigations are likely to translate to clinical differences or not. Understanding how structural tolerance ranges correlate in terms of clinical outcome could in future be used to evaluate structural data and assess what is really “clinically meaningful” in terms of any differences observed.
It is worth remembering that glycan display to the surrounding environment is very different for mAbs compared to other glycoproteins. Monoclonal antibody glycans are more enfolded in the hinge region of the antibody compared to other glycoproteins whose glycans are more exposed to the surrounding medium and thus freer to interact with other molecules. Therefore, effects seen for one molecule or class of molecules will not directly translate to other molecules. Nonetheless, biosimilar development studies could be further honed to those additional studies beyond the analytical that can be activated if tolerance ranges are exceeded for certain criteria based on those structural investigations.
Of course, there are other factors that must be borne in mind, such as the fact that this investigation focused on one specific structural feature, whereas analytical investigations may reveal more than one feature changed relative to the innovator. How would that then affect any PK decision?
All of this serves to further demonstrate the power of structural data both in its own right as a requirement for understanding the nature of the molecule, but also potentially to provide useful information which can help to define what further clinical investigations may or may not be needed.
1. Welch J, Ausin C, et al. (2023) The mannose in the mirror: a reflection on the pharmacokinetic impact of high mannose glycans of monoclonal antibodies in biosimilar development. Clin Pharmacol Ther., 113(5), 1003-10.
Contact the author or talk to our scientists today about your analytical requirements!
