Studying Binding in BIosimilars

By Michael Sadick , Catalent | June 3, 2014

Bio-Layer Interferometry can be a useful addition to the orthogonal analysis toolkit required for characterizing biosimilars

Assessment of biosimilar molecules requires an orthogonal analytical approach. Any one particular analytical technique is not able to fully define the physical and biological characteristics of original biotherapeutic or its potential biosimilar. Instead, a panel of analytical techniques must be used that can assess various and related, and sometimes overlapping, characteristics of the molecule. This allows a more complete and integrated understanding of the biotherapeutic molecule’s form and function. This goal can be achieved by a combination of physicochemical analyses, such as mass spectrometry, size exclusion HPLC, SDS-electrophoresis (gel or capillary) isoelectric focusing (gel or capillary) and glycoanalysis, as well as more biological analyses such as potency (bioassay or ELISA) and binding assays (such as surface plasmon resonance or bio-layer interferometry). This type of approach is the essence of “orthogonal.” Using an orthogonal approach, a more exact comparison may be made between a biotherapeutic molecule and its potential biosimilar.

One of the orthogonal characteristics mentioned above is therapeutic molecule/target molecule interaction, i.e., binding. For such studies, past strategies have utilized Scatchard binding analysis (labor-intensive, specialized, and requiring radioisotopes). More recently, Surface Plasmon Resonance (SPR; BiaCore) has provided more exact and reproducible analyses. However, BiaCore technology, as a whole, is expensive and still quite specialized. To provide a binding assay more robust and economic than is available with BiaCore, we measure binding using bio-layer interferometry (BLI), as offered by ForteBio OCTET RED96.

BLI is an optical analytical technique that analyzes the interference pattern of white light reflected from two surfaces: a layer of immobilized protein on the biosensor tip, and an internal reference layer. Any change in the number of molecules bound to the biosensor tip causes a shift in the interference pattern that can be measured in real-time.

Similar to SPR techniques, there exist multiple immobilization strategies to bind the initial binding partner to the sensor tip, including reactive amine-based binding, biotin/streptavidin binding, Protein A, etc. The binding between the binding partner immobilized on the biosensor tip surface and the target binding partner in solution produces an increase in optical thickness at the biosensor tip, which results in a wavelength shift, Δʎ, which is a direct measure of the change in thickness of the biological layer.

Interactions are measured in real time, providing the ability to monitor binding specificity, rates of association and dissociation.

The OCTET system can be used both for quantitative binding assays (similar to ELISA) and for binding kinetic assays (KON/KOFF/KD determinations).

As an example of using OCTET-analyzed binding kinetics as part of an orthogonal program, this article will discuss the assessment of a biosimilar monoclonal antibody (mAb). The mAb has ligand neutralizing activity, and, although it has Fc-binding capability, it is not associated with any effector function.

The originator molecule was characterized using, among other analyses, oligosaccharide mapping, peptide mapping (by MS), size exclusion HPLC, ion exchange HPLC, hydrophobic interaction HPLC, SDS and IEF capillary electrophoresis, cell-based bioassay (ligand inhibition assay) and OCTET analyses of the binding kinetics of both the Fc and CDR portions of the mAb to their appropriate targets. We were able to establish successful strategies for both a ligand binding assay and an FcƴRIIIa binding assay. Purified ligand was immobilized to the sensor tips using Second Generation Reactive Amine chemistry. The purified FcƴRIIIa (HIS-tagged) was immobilized to nickel-coated sensor tips. KD values were determined, for both binding events, for several lots of originator molecules.

The values derived for both mAb/ligand binding and mAb/ FcƴRIIIa binding were consistent with known values from historical Scatchard and/or BiaCore determinations. Additionally, bioassay analysis and the physicochemical analyses sited above were performed on these same lots of originator material. All analyses, including the Octet binding analyses (see Tables 1 and 2), showed reproducible values.

Several potential biosimilar candidates were then tested in the same fashion (Tables 3 and 4).

Thus, it would appear that there is more variability with Fc/FcƴRIIIa interaction than with mAb/ligand interaction, which is quite consistent. The consistency of the mAb/ligand interaction is supported by the fact that the (inhibition) bioassay gave virtually identical results for all originator lots as well as all biosimilar candidates.

When the binding data are compared to the corresponding physicochemical analyses, the reduced binding by Biosimilar mAb#1 to FcƴRIIIa correlates with a marked change in glycosylation patterns (although fucosylation, itself, was not determined, as the mAb is not and effector mAb). Differences in glycosylation (generally present in the Fc region of the mAb) would be expected to impact mAb-Fc interaction with the FcƴRIIIa. Thus, successful discrimination between different molecules/activities was achieved, and was consistent with alternative (orthogonal) physicochemical analyses.

As alluded to above, the KON, KOFF and KD values derived via OCTET analysis were consistent with known values from historical Scatchard and/or BiaCore determinations. However, use of BLI analysis to determine binding kinetics is considerably less labor-intensive than similar studies using Scatchard analysis. It is also far less expensive and both easier and more rapid to perform than similar studies using SPR (BiaCore). Thus, determination of binding kinetics (as well as development of quantitative binding assays) has rapidly focused, in the Large Molecule Analytical Chemistry department of Catalent Pharma Solutions in Kansas City, MO, on the use of OCTET/BLI.

As a continued effort to establish a robust OCTET platform for characterization of mAb biotherapeutics and associated biosimilars, a test panel is now being established to test mAb interaction/binding with FcƴRI, FcƴRIIa, FcƴRIIb/c, FcƴRIIIa, FcƴRIIIb and FcRN. 

BiaCore is a registered trademark of GE Healthcare Bio-Sciences AB. OCTET is a registered trademark of ForteBio

Acknowledgements:  I would like to acknowledge the invaluable work and contributions of the members of my team, especially Tiffany Walker, M.S. and Dan Papa, Ph.D.

Dr. Sadick is presently a senior manager at Catalent Pharma Solutions in Kansas City, MO, in the Large Molecule Analytical Chemistry (LMAC) department, overseeing and leading the development, growth and maturation of the Biopharmaceutical Characterization group (Bioassay, ELISA, Binding Kinetics and Molecular Biology). He has worked at Catalent Pharma Solutions, and before that, Aptuit, Inc, since 2007. Prior to that, Michael had been recruited to Eli Lilly and Co. as a Research Advisor in 2001 to help lead biotechnology efforts, leading the Bioassay Groups, Molecular Biology and Virology in support of Phase I-III projects, as well as providing guidance for commercial bioassay testing, all on a global level, working with FDA and EMEA regulatory agencies. Before his tenure at Eli Lilly, Michael worked for 10 years (1991 – 2001) at Genentech in South San Francisco. For five years he supported immunochemistry (PAGE and ELISA) and cell-based bioassays for research and discovery efforts, including bioassay development and support. During his latter 5 years at Genentech, Michael took a position of Senior Scientist in the Bioassay Group supporting Pharmaceutical Science efforts (Phases I – III). Dr. Sadick has an extensive background in cellular biology, cellular immunology and receptor signaling, molecular biology and biochemistry. He received his MS and Ph.D. in immunology from University of Washington in Seattle, and his BA in biology from Johns Hopkins, in Baltimore. He worked as research faculty at UCSF Medical Center for five years, characterizing many of the cellular responses which arise during progressive or resolving Leishmania infection. He was integral to some of the original characterizations of the roles TH1 or TH2  T cell responses in in vivo models of immune response and infection.

Related Contract Manufacturing: