Issue link: https://www.e-digitaleditions.com/i/1394390
www.biopharminternational.com July 2021 BioPharm International eBook 33 into the impact of excipient choice and processing conditions that result in reduced stability and the onset of aggregation long before aggregates are formed. Comparability studies based on the measurement of secondar y structure can be equally beneficial in manufacturing for confirming batch- to-batch consistency and the validity of post-approval process modifications. These studies may also directly support the demonstration of biosimilarity for a therapeutic in development. There is considerable potential to use FTIR to quantify the relative amounts of different elements of secondary structure and apply the resulting data across the drug development lifecycle, and it is an established characterization technique for this reason. But FTIR is not a workhorse tool. Conventional FTIR inherently pos- sesses several practical limitations; how- ever, within current industrial settings, there are instances where only FTIR can generate the information required. UNDERSTANDING THE LIMITATIONS OF FTIR A closer look at conventional F TIR instrumentation and workflows is helpful to understand the practical limitations of the technique. Because protein therapeutics are labile, the preference in biopharmaceuti- cal analysis is to measure structure using samples in their native state, which can be achieved by keeping preparation to a minimum. This approach also bene- fits analytical productivity. In the ear- lier stages of development, samples are relatively simple and dilute due to drug scarcity, but as formulation progresses, it is necessary to measure at clinically rep- resentative concentrations in the presence of the excipients and additives needed to optimize formulation properties. Therapeutics are also developed in aqueous conditions. For FTIR analysis, background reference absorbance spec- tra are subtracted from sample spectra to enhance protein characterization and improve signal to noise. This subtraction is complicated by the presence of water vapor due to strong absorbance by oxy- gen–hydrogen (O–H) bonds at a wave- length within the Amide I band. Digital processing routines facilitate this task, but accuracy relies on a constant measure- ment temperature and water absorbance spectra are extremely temperature sen- sitive. Furthermore, IR detectors suffer from non-linearity at high absorbance values associated with aqueous samples. As a result, the pathlength is typically limited to 10 µm or less, and the concen- tration of protein required to achieve an acceptable signal-to-noise ratio is corre- spondingly high, optimally within the range 10 –150 mg/mL. These issues largely define the recog- nized drawbacks of conventional FTIR. Due to the combination of reference and water vapor subtraction, FTIR measure- ments also tend to exhibit background drift, poor repeatability, and resulting low sensitivity. The technique is demanding with respect to temperature control and is admittedly laborious with the need for subjective manual intervention. As such, traditional F TIR instrumenta- tion is poorly amenable to automation. Perhaps more crucially, FTIR is ill-suited to the dilute aqueous samples associated with early development. Because of con- centration limitations, FTIR analysis is often used in conjunction with far-UV CD which has an even narrower optimal operating concentration range of approx- imately 0.2 – 2.0 mg/mL. Finally, the alternative approach of increasing sam- ple concentration via manual concentra- tion methods adds the risk of generating unrepresentative data. MMS OVERCOMES HISTORICAL LIMITATIONS The new, commercially available tech- nique of MMS was purposely developed to exploit the inherent utility of IR spec- troscopy while simultaneously addressing the practical limitations of FTIR. MMS automates the requirement for background subtraction by rapidly modulating the sample solution with a matching reference stream across the path of the laser beam at a 1 Hz–5 Hz frequency, thereby producing real-time differential spectral scans across the Amide I band. A schematic of the flow cell highlighting the design of the micro- fluidic path is shown in Figure 1. The resulting differential spectra are instan- taneously auto-referenced, background compensated, and essentially drift-free. This automated approach adds signif- icant efficiency and repeatability to the analytical workf low, versus that with FTIR, by removing an important source of inaccuracy and variability. The concentration limitation typical of FTIR is a second limitation that is addressed by use of a tunable mid-infra- red quantum cascade laser (QCL), which generates an optical beam approximately 100 times brighter than that used in con- ventional FTIR systems. With a QCL laser that is set to an optimized opti- cal configuration for the Amide I band, it is feasible to generate highly sensitive secondary structure information over a concentration range that extends from 0.1 mg/mL–200 mg/mL. The broad dynamic range of MMS spans that of combined far-UV CD and FTIR and makes it possible to now measure sec- ondary structure with a single technique from late-stage development where far-UV CD has greater utility, through formulation and into manufacturing, where FTIR is feasible, by offering the advantage of measurement at clinically representative protein concentrations in complex formulations across the entire development pipeline. The third critical advantage of MMS is the automated advanced spectral pro- cessing engine and software developed to interpret the spectral results and reduce manual input. The data analysis software package rapidly and efficiently converts measured spectral absorbance into infor- mational insight in the form of area of overlap plots, % similarity values, and fractional contribution data for specific HOS motifs of secondary structure, to name a few. These tools make it easy Biopharmaceutical Analysis Protein Characterization