// February 04, 2020

Strengthening Biotherapeutic Innovations through Emerging Analytical Tools


Innovations in biotherapeutic medicines are advancing at an unprecedented pace and offer great promise for improving outcomes for patients. Analytical tools and approaches for understanding quality are central to fully realizing the promise these new therapies hold, and they too are quickly evolving in sophistication and precision. 

Multi-attribute method (MAM) is an emerging analytical approach that is transforming how manufacturers identify and characterize critical quality attributes of biotherapeutics during the early stages of discovery and development as well as monitor product quality attributes and impurities during process development and validation, stability testing, and, ultimately, product release by QC laboratories in a cGMP environment. MAM provides manufacturers with more effective and efficient ways to evaluate product quality and consistency, while potentially saving time and other valuable resources.

At its inaugural Biologics Stakeholder Forum earlier this month, USP explored how manufacturers are currently deploying MAM as well as regulatory perspectives on its use in drug development programs and license applications. Since MAM is still relatively new, USP is also exploring potential standards that could support increased use and adoption of MAM on the path to greater regulatory acceptance.

Strengths of MAM

Biotherapeutic products are large molecules with complex structures. Inherent to this complexity are modifications that can impact these molecular structures and, thus, product purity, quality and biological activity.

Conventional control strategies require multiple tests—in some cases 20 or more—to identify and monitor critical quality attributes of complex biotherapeutics. The most common form of MAM being used by manufacturers combines chromatographic separation with mass spectrometry (MS) detection to simultaneously measure multiple quality attributes. Software is then used to examine this rich source of data and monitor both expected attributes as well as unexpected new peaks throughout a product's lifecycle. MS is better able to discriminate between molecular entities than traditional UV detection of chromatographically separated peaks that can often contain multiple species. Thus, it can provide manufacturers with more detailed insights on the impact of modifications on product quality. 


A presentation by Jette Wypych of Amgen—credited with pioneering MAM a decade ago—provided a historical perspective of the company's evolving journey with the implementation of MAM. Wypych emphasized the importance of engaging regulators about MAM early in the development process for greater receptivity further downstream when filing regulatory dossiers containing proposed product specifications. Wypych described how critical elements such as new peak detection (NPD) need to be well understood and implemented before MAM can be usable in a QC environment. A common challenge with using MAM to assess quality is discerning if a new peak is an impurity that warrants attention or if it is "noise." Establishing appropriate controls to ensure that a MAM method is sensitive enough to support a threshold assignment for identifying NPDs is an area for potential standard development. 

Andrew Dawdy described a somewhat different approach to MAM at Pfizer, which is focusing its resources on extensive characterization and data mining before deploying MAM in QC. Harvesting MAM data on product quality attributes and building robust data libraries allow for continuous product quality attribute (PQA) monitoring in addition to attribute monitoring across products (e.g., a host cell protein impurity that may be common across different MAb products). Successful transfer of methods across manufacturing sites is also an important consideration for manufacturers who plan to scale-up and implement MAM globally. Pfizer has set up standardized MAM workstations in two different analytical laboratories to assess the ability to transfer methods that are reproducible and robust and have the ability to be implemented in a CGMP environment. Pfizer is currently conducting compliance and risk assessment work to implement MAM in Pfizer's QC laboratories within the next 5 years.

The central role of MS in MAM was prominent in both presentations from Amgen and Pfizer and in other forum discussions. Despite the time and effort required to properly conduct MS, Wypych discussed how future adaptation of MS's power and speed could eventually lead to use of MAM in continuous manufacturing processes as well as process analytics in support of Quality by Design. 

Another popular topic of discussion at the forum was software. Different software platforms use different algorithms to analyze data gathered through MAM and many are not compliant with 21 CFR part 11 requirements for use in a cGMP environment. Determining whether MAM data collected using different software are truly comparable can present challenges. Dawdy discussed an effort by the MAM Consortium to develop open-source MAM software to evaluate data gathered across instrument providers and enable sharing between laboratories. 

According to a presentation by Sarah Rogstad of FDA's Emerging Technologies Team, 79 of 80 BLAs approved between 2000-2015 used MS in drug substance characterization. However, MS was less commonly used in QC testing of therapeutic proteins due to the complexity of using MS in a QC environment. Some general benefits of MAM seen in applications include more detailed information at the molecular level (analysis of site-specific modifications allows for tighter control); the ability to differentiate between species that may overlap using conventional chromatographic approaches and control of unexpected new molecular entities through NPD.

Next Steps

USP will continue to engage stakeholders and explore opportunities for developing tools and solutions to support the adoption of MAM. Presentations and notes from the Biologics Stakeholder Forum will be made available on USP's website, and USP plans to form a new USP Expert Panel dedicated to development of an informational chapter containing best practices for adopting MAM for characterization of biologics. USP is already working on MAb-based performance standards that may help ensure the sensitivity of MAM-based methods and support transfer of methods across laboratories. 

To receive copies of the presentations and notes or to find out more about volunteering on a USP Expert Committee or Panel, please contact USPBiologics@usp.org.