Ten years ago this month, in August 2002, FDA published its groundbreaking guidance document, Pharmaceutical cGMPs for the 21st century: A Risk-Based Approach, to encourage a quality- and risk-based, approach to pharmaceutical development and manufacturing. With a growing focus on process analytical technology (PAT), industry quickly began to apply the guidance to production as well as development and the concept of quality by design (QbD) emerged. QbD requires the thorough scoping and understanding of key parameters in process development. An outcome of a QbD approach is that the product attains the desired quality every time it is produced. QbD creates a robust and repeatable framework that ensures that quality is designed-in, not tested for, in the product.

There was initial resistance from some industry parties to the implementation of QbD. Many thought that the process would mean an increase in workload during the development phase. Some noted that implementation would be significantly more complicated with biologics compared with traditional small-molecule drugs. With the understanding that quality and improved efficiency were crucial to all drug products, in 2008, the FDA Office of Biotechnology Products launched a pilot program for QbD to evaluate and identify best practice for the risk-based approach with biologics. Work in this area is now ongoing throughout the industry and this article examines how critical quality attributes (CQA)—a key component of QbD—can be defined using analytical tools, with a focus on nuclear magnetic resonance (NMR).

Aims of bioprocess optimization

Optimization tools. Spent media analysis using high-performance liquid chromatography (HPLC) and liquid chromatography–mass spectrometry (LC–MS) techniques is a common approach to understanding the factors limiting cell growth, and therefore, can help identify which parameters of the cell-culture media should be adjusted to optimize protein production. However, several limitations are associated with these methods. Analyses are generally slow, costly, and restricted in the number of components which may be studied, leading to an incomplete picture of the bioprocess and the factors which may affect scale-up.

Recently, application of NMR spectroscopy to the profiling of complex matrices, such as cell culture media and spent media, has shown major advantages when compared with LC methods (1, 2). NMR spectroscopy, a universal detection method, is characterized by a large dynamic range and measurement of the signal intensity can provide fast acquisition of quantitative results.

NMR and the future of bioprocess optimization

Cell-culture process optimization is crucial if production of recombinant protein products is to be commercially viable. Developments in NMR methodology have provided a new and useful tool allowing the biologics industry to rapidly and accurately analyze culture media, thereby elucidating the concentration profile of multiple components. This technique can be used in designing new media, troubleshooting existing culture media problems, standardizing cell-culture media prior to large-scale production, and examining spent media to highlight efficiency and cost issues.

According to FDA, "Quality by Design [QbD] is understanding the manufacturing process and identifying the key steps for obtaining and assuring a pre-defined final product quality" (3). FDA has identified the use of new analytical methods, such as NMR, to monitor and control processes as important in increasing manufacturing quality through QbD (3). Offering the ability to characterize chemically complex media, NMR techniques have the potential to contribute significantly to an understanding of process-critical parameters, helping to reduce performance variability and minimize the risk of process failure in large-scale biopharma production.

Advantages of NMR monitoring. As previously mentioned, one advantage of NMR compared with LC-based techniques is the ability to analyze simultaneously more than 50 (up to ca. 100–150) compounds, including amino acids, saccharides, components of the Krebs cycle and vitamins at one time. Specific sets of NMR peaks relate to specific analytes and each peak can be interpreted quantitatively. Therefore, it is possible to identify molecules—polar, nonpolar, volatile and nonvolatile— not previously reachable when using a targeted approach based on techniques such as HPLC or MS. NMR can help in the evaluation of the components that a cell line is consuming, as well as give insight into the production of metabolites and metabolic pathways. For example, formate, acetate, saccharides, and methanol can be accurately and rapidly monitored at the same time using NMR.

NMR is characterized by a suitable limit of detection (1–10 μM), with a linear signal over a broad concentration range (1μ–500 mM) and can provide highly reproducible results. Systematic calibration is not required. As a consequence, results are comparable across machines and research sites.