Structural Assessment of Protein Biologics Using 2D-NMR

As Food and Drug Administration (FDA)-approved protein biologics near the end of their patent life, it is expected that biosimilars will come to market; to date, the FDA has approved one biosimilar (a version of filgrastim), while the European Union has approved about 20 biosimilars over the last 10 years. In the U.S., a biosimilar is defined as a protein biologic that is highly similar to an already FDA-approved drug, i.e., reference product, with no clinically meaningful differences from the reference product and only minor differences in clinically inactive components allowed. In order to evaluate and compare reference products and their biosimilars, an array of analytical and biophysical methods is required; the accuracy, precision, robustness, and suitability of which must be determined and advanced.

Structure of the human granulocyte colony-stimulating
factor (G-CSF) determined by NMR spectroscopy [1].
 PDB ID 1GNC.

NIST researchers at IBBR are working to advance the application of NMR methods to the structural characterization of protein biologics. One particular effort, recently published in Nature Biotechnology, has demonstrated that 2D-NMR can yield a precise and unique ‘fingerprint’ of the higher-order structure of a protein biologic that can assist companies and regulators when assessing the critical quality attribute of higher-order structure among biosimilars. 2D-NMR is one of the few approaches that can yield complete assignment of 3D structure across the entire molecule at atomic-level resolution.

The IBBR team, along with collaborators from FDA’s Center for Drug Evaluation and Research (CDER), Sweden’s Medical Products Agency (MPA), and Health Canada’s Center for Biologics Evaluation, reported measurements for four independently manufactured versions of filgrastim—one FDA-approved reference product, and three unapproved biosimilars sourced from India—a granulocyte colony-stimulating factor (G-CSF) analog used to ward off infection in cancer patients undergoing chemotherapy. Samples of the four filgrastim products were analyzed using heteronuclear 2D-NMR correlation at 15N natural isotopic abundance. The technique resolves, in a 2D frequency map, the positions of proton–nitrogen atom pairs from each amide bond of each amino acid in the protein. Spectra were independently acquired on the same samples on six different spectrometers, at four different field strengths ranging from 500 MHz to 900 MHz, in four different laboratories.

The NMR data from the four laboratories were determined to have an experimental precision of 8 parts per billion (ppb), which is well below any chemical shift changes that could be induced by a significant structural change—such as an amino acid substitution or a modification of a residue by oxidation or local conformational change—and so it can be concluded that the atomic structures of all four filgrastim products were the same within the tight precision limits of the data. As a separate effort, the NIST researchers repeated the measurements on all four samples one year after the inter-laboratory comparison and did not find significant structural changes in any of the products, highlighting the applicability of the method for assessing product stability.

In addition to reporting on the utility of 2D-NMR for high-precision measurement of the atomic structure of protein biologics, this study also described statistical methods for assessing large numbers of datasets, which is important for monitoring batch-to-batch variation for one product as well as determining biosimilarity between products. This inter-laboratory collaboration has also revealed the importance of proper instrument calibration and control of laboratory conditions to ensure that results are reliable and that the method can be used as a structure comparability tool. For example, early on in the inter-laboratory study, the research team identified deviations in data from two instruments and then determined that these were caused by variations in sample temperature. Ultimately, it was shown that proper calibration of the instruments removed these discrepancies and resulted in measurements that were well within the determined experimental precision.

In the next phase of the work, the inter-laboratory effort will be expanded to include thirty laboratories on five continents. The IBBR team and their collaborators will compare 2D-NMR measurements of a monoclonal antibody (mAb) that NIST is developing as a reference material, the NIST mAb IgG1κ (NIST RM 8670). mAbs, which are therapeutic proteins developed to bind to specific protein or cell targets, such as cancer cells, are the largest class of approved protein biologics in the world, and the ability to extend 2D-NMR methods to this class of drug would represent an important landmark in their analytical characterization. Beyond determining the precision of 2D-NMR across a larger network of laboratories, the research is expected to yield a catalog of best practices to ensure the reliability and reproducibility of results, and to provide greater assurance of drug product quality to regulatory agencies.

This work was published as: H. Ghasriani, D.J. Hodgson, R.G. Brinson, I. McEwen, L.F. Bushe, S. Kozlowski, J.P. Marino, Y. Aubin and D.A. Keire, “Precision and Robustness of 2D-NMR for Structure Assessment of Filgrastim Biosimilars.” Nature Biotechnology (2016) 34, 139–141. doi:10.1038/nbt.3474

Read more here: http://www.nist.gov/mml/bmd/assessing-the-biosimilarity-of-protein-drugs.cfm

References

1. Zink, T., et al. (1994) Biochemistry 33, 8453-8463.