Meeting New Regulatory Expectations of Characterization Methods Qualification of Biophysical Analyses
Yijia Jiang, Cynthia Li, Ranjini Ramachander, Jie Wen, and Linda Owers Narhi Forensic and Biophysical Analysis, Global Cellular & Analytical Resources Department Amgen Inc.
Abstract
Biophysical techniques including CD, FTIR, fluorescence and DSC etc. have routinely been used by the bio-pharmaceutical industry to study the effect of process, manufacturing, formulation and storage conditions on protein conformation. These analyses are included in regulatory filings and comparability studies in order to verify that the higher order structure of a protein has been retained. However, there has been an increasing demand from the regulatory
agencies for qualification of the characterization methods, including the biophysical methods, used for comparability studies. To meet new regulatory expectations, in this paper, we will discuss criteria identified for the qualification of the general biophysical characterization analyses, the approaches we take to assess the criteria and the results of some experiments.
Abbreviations
DSC, Differential Scanning Calorimetry; CD, Circular Dichroism; FTIR, Fourier Transform Infrared; CxN, 20 mM NaCitrate, 140 mM NaCl, pH x buffer
Introduction
Due to the complex nature (presence of higher order structures, carbohydrate, post-translational modification etc.) of biological products, the characterization of these molecules has been critical in ensuring product quality and efficacy and patient safety. Biophysical techniques including CD, FTIR, fluorescence and DSC (1-4) have routinely been used in characterizing protein higher order structures, conformation and thermal stability. Therefore they have been used by
the bio-pharmaceutical industry to study the effect of process, manufacturing, formulation and storage conditions on protein conformation, to ensure that the higher order structures (secondary and tertiary structure) are not adversely affected by changes in those conditions. These analyses have routinely been included in regulatory filings (5). However, there has been an increasing demand from the regulatory agencies on the qualification of the characterization methods including the biophysical techniques for comparability studies. To meet these new regulatory expectations, in this paper, we would like to identify criteria for the qualification of the general biophysical characterization methods, define the approaches to assess the criteria and show the results of some experiments. The results of these techniques are based on the average molecule in solution. The signal strength and signal to noise ratio depends on the primary sequence of the proteins, the conformation of the proteins, and the buffer they are in, the qualification will need to be specific for each technique, and possibly for each protein, or at least structural family. The buffer components can also affect the signal to noise ratio dramatically, thus some aspects of qualification will be technique, and possibly even protein and formulation specific.
There are a several aspects of the qualification of the biophysical methods, in addition to system suitability tests, that apply to all the techniques. These include the precision of the measurement and the fit for purpose of the analyses.
The precision of the measurement can be assessed by multiple measurements of the same protein sample using the same instrument and/or different instruments by the same/different analyst on the same/different dates at the same/different site. The parameters to be used to evaluate the precision of the measurement will be different depending on the specific techniques and should be defined prior to the start of the experiments and monitored closely during the precision assessment. The precision assessment will help set the acceptance criteria for future comparability studies.
The evaluation of the fit for purpose of the techniques in detecting changes in protein conformation and thermal stability can be achieved through a few approaches. These include in parallel bio-physical and biochemical analyses of accelerated degradation samples, correlating the results if possible; studies including analysis of the effect of pH, heat, or denaturant perturbation and freeze-thaw treatment; blending studies with an irreversibly denatured species or a native protein of a defined, different structure, and computer simulation studies where the spectrum/profile of a native protein will be mathematically mixed with that of a denatured one at various mass ratios and the resulting spectral changes will be assessed and correlated with the percentage of the denatured protein. The appropriate approach depends on the specific protein and technique. The parameters used for precision assessment will also be used for the fit for purpose assessment for the individual techniques.
Examples of some precision and fit for purpose assessments of the biophysical methods are included as well.
CD
Circular dichroism (CD) is observed when optically active matter absorbs left and right handed circularly polarized light slightly differently. CD spectroscopy measures the difference in absorption of left-handed polarized light versus right-handed polarized light as a function of wavelength. Secondary structure can be determined by CD spectroscopy in the far-UV spectral region (190-250 nm). At these wavelengths the chromophore is the peptide bond, and the signal arises when it is located in a regular, folded environment. Alphahelix, beta-sheet, and random coil structures each give rise to a CD spectrum with a characteristic shape and magnitude. The presence of tertiary structure in a protein can be assessed from the CD spectrum of a protein in the near-UV spectral region (250-350 nm). At these wavelengths the chromophores are the aromatic amino acids and disulfide bonds, and the local asymmetric environment they are sensitive to is that defined by the overall tertiary structure of the protein. Each folded protein has a characteristic near UV CD spectrum. In both cases the sample CD spectrum is a difference spectrum. Any buffer components with either absorbance of chiral properties can affect the signal to noise ratio of the measurements.
...To view the complete article, please login or subscribe if you don't already have an account.