What FDA Draft Guidance Tells Us About In-Process Control Strategies
By Steven Erb, Crown Point Biotech Consulting
The FDA released a draft guidance “Considerations for Complying with 21 CFR 211.110” in January 2025 to describe the agency’s thinking on requirements for drug product manufacturing per 211.110, the regulations for sampling and testing of in-process materials and drug products. These regulations ensure batch uniformity and drug product integrity under current good manufacturing practice (cGMP) regulations (21 CFR parts 210 and 211).
This guidance was released as part of the Framework for Regulatory Advanced Manufacturing Evaluation (FRAME) initiative to prepare a regulatory framework to support the adoption of advanced manufacturing technologies. One of the FDA’s stated missions is supporting and enabling pharmaceutical innovation and modernization, and this guidance appears to reinforce that.
The organization of the content is parsed into three primary focus areas that track advancements in the field of manufacturing control strategies. Firstly, the agency describes the foundations for cGMP control strategies, then provides considerations for advanced manufacturing technology control strategies, followed by use of process models for executing a system that enables maintaining a state of control under cGMP.
Upon first reading this draft guidance, it doesn’t seem to be telling us anything we don’t already know. That is, in order to comply with 21 CFR 211.110, we are required to design and follow a defined written strategy that stipulates the controls required for an effective monitoring system to ensure things are in a state of control during manufacture. That, of course, is the intent of a control strategy, to define specific testing points and methods that monitor if our API is transforming into the final drug product that will provide therapeutic relief to patients. While we may think we are sufficiently familiar with the function of a well-designed control strategy, this guidance serves to modernize the regulations in 211.110 and provides applicability to contemporary advancements in the field.
To reiterate the foundational tenets ensuring batch uniformity and drug product integrity, the Background section of the guidance provides a concise tutorial on how the use of control strategies ensures drug products are manufactured with desired quality as dictated by the regulations in 211.110. Of course, it is difficult to provide specific examples of how to comply with the regulations when they are applicable to the cGMP manufacture of all drug products. To this end, the FDA repeats a few overarching concepts frequently throughout the draft guidance, that cGMP regulations were written to provide manufacturers flexibility to design control strategies as they deem fit, and that knowledge and understanding gained through robust product and process development form the basis for establishing and maintaining control strategies.
Room To Flex: The What, The When And Where, And The How
Without doing so explicitly, the guidance puts the context of 211.110 into the framework of the quality by design approach for drug development (ICH Q8(R2) Pharmaceutical Development). To monitor process performance and maintain a state of control, you must know what to test and the guidance defines this as critical quality attributes (CQAs), which are applicable to all categories of drug products and are identified by process and product understanding, a concept that we should be sufficiently familiar with in the industry. With the identification of the what (i.e. CQAs) to test, the manufacturer needs to define and justify when and where the proposed in-process controls and testing or examinations that are used to monitor those attributes should occur; per 211.110, the when and where is “at commencement or completion of significant phases.”
The FDA does not define the term “significant phase,” again stating that regulations allow flexibility for this determination, depending on the process and drug product. Therefore, it is up to the manufacturer to define and the quality unit to approve the significant phases in the process that require monitoring. In the end, the FDA will evaluate the adequacy of these determinations, so it is important that they are based on scientific- and risk-based approaches (again defaulting to the “it’s up to the manufacturer to decide” motto and aligning with quality by design product development).
The FDA offers specific advisement on process monitoring and control decisions when there are minor equipment and process adjustments that do not require approval by the quality unit. These minor adjustments need to be within pre-established and justified limits which need to have been approved by the quality unit, and production data needs to be reviewed by the quality unit before approval or rejection of a batch. Having addressed the when and where in-process sampling and testing should occur, the specifics of how material and drug product testing is conducted are considered next. The guidance states that the collection of physical units for testing or a schedule by which examination of some sort is done is permitted.
While in-process controls and tests are required, sampling does not necessarily require unit operations to physically remove material for test, allowing for in-line, at-line, or on-line measurements in place of physical sample removal. This provides freedom to use process analytical technologies (PAT) or other testing (referred to generically as enhanced process monitoring) that does not require physically removing samples for examination, thus expanding on the regulations dictated in 211.110. Specifically acknowledging the use of in-line, at-line, and on-line monitoring as a compliant means of process monitoring per 211.110 is welcomed for conversion of batch manufacturing to continuous manufacturing for well-established manufacturing systems and for new process technologies used in the generation of advanced therapy medicinal products.
Bringing 211.110 Into The Modern Fold
The General Considerations section of the guidance describes what you would assume from a modern comprehensive control strategy for a manufacturing process. The Additional Considerations section digs deeper into next-generation control strategy concepts integrating advanced manufacturing technologies and process modeling, but without many specifics. The FDA restates their commitment to supporting advanced manufacturing technologies and, for the purposes of the guidance, it utilizes continuous manufacturing as a representative example for application of the regulations to compliant control strategy generation. For continuous manufacturing, isolating in-process materials may not be as feasible as with batch manufacturing processes where there are isolated steps.
Expounding further on the previous section of the guidance, a quality unit can make a scientifically sound data driven determination of what constitutes a significant phase where control testing is required. A significant phase here (and it should be noted for batch manufacture) can constitute two or more unit operations. Having a robust understanding of the process permits a manufacturer to determine when appropriate sampling and testing can occur during continuous manufacturing so that sampling frequency adequately controls for in-process and drug product quality.
Again, the manufacturer is in the driver’s seat, as long as they can justify the sampling frequency is sufficiently representative to ensure a statistically valid conclusion about quality of in-process materials and the drug product.
FDA Q13 Continuous Manufacturing of Drug Substances and Drug Products describes process modeling under control strategies but does not expound on how it pertains to complying with 211.110. In this guidance, the FDA describes the virtues of tracking the behavior of a process by a mathematically represented model, declaring and encouraging that models can be a component of the overall control strategy to enhance understanding and monitor performance.
However, the FDA takes an opportunity to declare that “control strategies that rely solely on current process models would be insufficient to satisfy requirements of 211.110.” By their reasoning, they make it clear they have not yet become aware of a process model that can always ensure the continued validity of all the model’s underlying assumptions, particularly during unplanned disturbances. Without being able to account for unplanned disturbances, a control strategy based only on a process model without any integrated testing is unable to prevent nonconforming in-process materials.
In effect, you cannot account for what you cannot predict, and it seems the FDA is weary to accept a process model that cannot account for all variables that may negatively impact product quality and the final end user — the patient. However, the FDA is fully supportive of the use of process models when paired with in-process material testing or enhanced process monitoring, and this pairing can reduce the number of tests required, compared to control strategies that do not utilize process models.
Putting It All Together
The draft guidance on complying with 211.110 spells out much of what we already know and presume for a comprehensive control strategy but does add some color to how the regulations fit into today’s manufacturing paradigms. Regulations stipulated in 211.110 and cGMP in general are flexible to accommodate different drug products and unique processes.
The manufacturer should have a firm grasp on in-process and drug product critical quality attributes, when and where to test specific samples during significant phases of their process and based on these two components, how to test those samples to monitor process performance and maintain control of in-process and drug product material quality.
Advancements in manufacturing, like continuous manufacturing, are covered by 211.110 and the use of process modeling can further support design of compliant and comprehensive control strategies.
Using their own words, the “FDA encourages the use of scientifically valid combinations of modern control strategies to develop and implement effective and innovative approaches in pharmaceutical development, manufacturing, and quality assurance.” In essence, begin with the end in mind and utilize the data you generated during development to implement a comprehensive control strategy using recent and new advances in process monitoring that maintains the quality of your API from start to finish. The more data you have, the more confident you can be in defining and controlling significant phases in your process, eventually leading to modeling your process in full.
Should anyone want to comment on this draft guidance before finalization by FDA, comments are due by April 7, 2025.
About The Author:
Steven Erb, Ph.D., is an independent consultant and founder of Crown Point Biotech Consulting. He has over 13 years of experience in CMC development at small biotech and large pharma companies with significant experience leading analytical development, quality control, technology transfer programs, and managing CDMOs. He has authored numerous regulatory filings supporting IND CMC content, briefing books, annual reports, and amendments. Contact him on LinkedIn.