Sponsored content brought to you by
Surprises that delay manufacturing programs tend to arrive late in the day. A comparability package will not close if reactor scale-up, site transfer, or cell-line improvement produces a new analytical signal that cannot be reconciled with the material dosed in the clinic. Each of those changes was deliberate, yet the divergence it produced was not. What looks like an analytical problem can reflect an underlying product or process change.
A council of separate assays
The quality of a biologic is conventionally assessed not by one instrument but by a council of specialists, each fluent in one dialect. Charge variants speak through imaged capillary isoelectric focusing, size variants through size-exclusion chromatography, glycans through released-glycan mapping, and identity and modifications through peptide mapping.
Each specialist instrument is an expert, and each offers a partial answer. Across the years from candidate to commercial supply, that council is reconvened as methods, instruments, and qualified reference standards change. What survives is not a story but a stack of testimonies; the seams between them are where late surprises are born.
One framework, one language
The multi-attribute method (MAM) offers a more coherent approach: a peptide-mapping LC-HRMS (liquid chromatography-high resolution mass spectrometry) workflow that can identify and monitor multiple product quality attributes, including selected critical quality attributes, and resolve modifications to specific sites. In a single analysis, MAM can monitor deamidation, oxidation, site-specific glycoforms, sequence variants, and other product-related features, potentially streamlining quality control across the product lifecycle.
MAM is built in two movements: a broad characterization phase that identifies measurable attributes and assembles a product-specific peptide library anchored by accurate mass and retention time; and a monitoring phase that tracks the relative abundance of selected attributes, batch after batch.
Catching what no one thought to ask
MAM’s distinctive capability is new peak detection, a threshold-based comparison that aligns mass, retention time, and intensity features against a product-specific reference and flags new or significantly changed peaks for review. A conventional release assay reports within its intended analytical dimension. New peak detection asks a broader question: has an unexpected peptide-level feature changed?
Detecting that signal while the process is still being developed, rather than after it is locked, can be valuable because the attribute that ultimately matters may not be the one predicted.
When the process changes
Consider a typical crisis: a perfusion process replaces fed-batch, a step is redesigned, and a campaign moves to a second manufacturing site. Regulators require evidence that relevant quality attributes remain highly similar and that observed differences do not adversely affect safety or efficacy. Assembled from scattered legacy assays, that evidence can require substantial time to build and still show its seams.
When the same peptide mapping-based framework has traveled with the molecule from its earliest characterization, comparability can become less of an emergency and more a continuation of an established data stream. The peptide-level reference and historical attribute data already exist. That continuity can strengthen the broader comparability package.
What MAM does not solve
MAM is not a universal solution. By separating the molecule into peptides, MAM does not directly assess aggregation, particles, higher-order structure, biological activity, or modification combinations on the same intact molecule. It also depends on robust data analysis. MAM has been implemented for release in specific applications, and USP <1060> now provides a practical framework, while broader implementation continues to evolve. MAM works best as a backbone for directed attribute monitoring, supported by orthogonal methods.
Analytical continuity by design
At Catalent’s Kansas City analytical center of excellence, MAM and high-resolution mass spectrometry are applied as independent analytical services, supporting programs wherever the molecule is manufactured. Against a decade of process change, the aim is not more testimony, but a single coherent account, and fewer places for the story to break.
Comparability is easier to defend when it never had to be reconstructed.
Learn more about Catalent Biologics Analytical Services.
Facts Only
* Catalent operates an analytical center of excellence in Kansas City.
* Multi-attribute method (MAM) utilizes a peptide-mapping LC-HRMS workflow.
* MAM monitors deamidation, oxidation, site-specific glycoforms, and sequence variants.
* The MAM process consists of a broad characterization phase and a monitoring phase.
* MAM utilizes threshold-based comparison for new peak detection.
* USP <1060> provides a practical framework for MAM implementation.
* Standard biologic quality assessment uses capillary isoelectric focusing, size-exclusion chromatography, released-glycan mapping, and peptide mapping.
* MAM does not directly assess aggregation, particles, higher-order structure, biological activity, or modification combinations on intact molecules.
* Regulators require evidence of similarity in quality attributes during process changes, such as moving from fed-batch to perfusion or transferring to a second manufacturing site.
Executive Summary
Biologic manufacturing often faces late-stage delays when process changes—such as reactor scale-up or site transfers—create analytical signals that diverge from clinical material. Traditional quality assessment relies on a "council" of separate assays, each providing a partial view of the product. Because these methods often change over a product's lifecycle, the resulting data can be fragmented, making comparability studies difficult to defend during regulatory review.
The multi-attribute method (MAM) proposes a more integrated approach by using liquid chromatography-high resolution mass spectrometry to monitor multiple critical quality attributes in a single analysis. By establishing a peptide-level reference early in development, MAM allows for continuous data streams and the detection of unexpected peptide-level features. While MAM streamlines monitoring and enhances comparability, it is not a universal replacement; it must be supported by orthogonal methods to assess attributes it cannot detect, such as aggregation and biological activity.
Full Take
The strongest version of this narrative is that analytical continuity is a risk-mitigation strategy. By replacing a fragmented set of legacy assays with a single, high-resolution digital backbone (MAM), developers can avoid the "emergency" of reconstructing comparability data during late-stage regulatory hurdles.
However, this is a vendor-sponsored advertorial. It employs a specific persuasive architecture: it frames the traditional "council of assays" as a fragmented set of "testimonies" with "seams" where surprises are born—creating a narrative of inevitable failure for those not using an integrated system. This fear of the "late surprise" is then resolved by the introduction of the vendor's specific service offering. The argument relies on the premise that the "coherent account" provided by MAM is inherently superior to the specialized "dialects" of traditional assays, though it admits MAM cannot see everything.
Patterns detected: ARC-0043 Motte-and-Bailey (shifting from the broad claim of "analytical continuity by design" to the narrower, defensible technicality of peptide mapping), ARC-0024 Ambiguity (using "story" and "testimony" to characterize scientific data), ARC-0012 Authority Game (leveraging the USP <1060> framework to validate a specific commercial service).
The root cause is the commodification of regulatory risk. The assumption is that the primary value of high-resolution science is not just better data, but a "defensible" package for regulators. This shifts the focus from biological truth to regulatory expediency.
Bridge Questions:
1. Does the efficiency of a "single coherent account" risk overlooking anomalies that are only visible through the "dialects" of specialized, orthogonal assays?
2. How does the cost and data-complexity of implementing MAM early in development weigh against the risk of late-stage comparability surprises?
Counterstrike Scan: A bad actor would use "Fear, Uncertainty, and Doubt" (FUD) regarding regulatory rejection to force a transition to a proprietary analytical platform. While this content follows that marketing playbook, it remains grounded in established mass spectrometry capabilities and USP frameworks.
Sentinel — Human
This text reads as expert analysis synthesizing complex concepts in biopharmaceutical quality control, demonstrating high coherence and domain-specific knowledge, suggesting human authorship.