A network of regulatory innovations to improve FDA quality assessments of human drug applications

Abstract

A network of regulatory innovations brings a holistic approach to improving the submission, assessment, and lifecycle management of pharmaceutical quality information in the U.S. This dedicated effort in the FDA’s Center for Drug Evaluation and Research (CDER) aims to enhance the quality assessment of submissions for new drugs, generic drugs, and biological products including biosimilars. These regulatory innovations include developing or contributing: (i) the Knowledge-Aided Assessment and Structured Application (KASA), (ii) a new common technical document for quality (ICH M4Q(R2)), (iii) structured data on Pharmaceutical Quality/Chemistry, Manufacturing and Controls (PQ/CMC), (iv) Integrated Quality Assessment (IQA), (v) the Quality Surveillance Dashboard (QSD), and (vi) the Established Conditions tool from the ICH Q12 guideline. The innovations collectively drive CDER toward a more coordinated, effective, and efficient quality assessment. Improvements are made possible by structured regulatory submissions, a systems approach to quality risk management, and data-driven decisions based on science, risk, and effective knowledge management. The intended result is better availability of quality medicines for U.S. patients.

Graphical abstract

1. Introduction

The U.S. FDA’s Center for Drug Evaluation and Research (CDER) regulates prescription drugs, including new drugs, generic drugs, and certain biological therapeutics including many recombinant DNA biotechnology products. Other human biological therapies are regulated by FDA's Center for Biologics Evaluation and Research (CBER). CDER's Office of Pharmaceutical Quality (OPQ) establishes patient-focused quality standards and integrates the assessment of regulatory drug submissions with the evaluation of manufacturing facilities to assure quality drugs are available to the American public. CDER is implementing a network of regulatory innovations to adapt to changes in the global public health landscape and the development and manufacture of human medicine. This paper describes the regulatory innovations CDER is developing or using, some in conjunction with other global regulators and FDA centers (e.g., CBER), to improve the quality assessment of human drug applications in the U.S.

Quality information in regulatory submissions to CDER ranges from formulation design and analytical testing to manufacturing development, scale-up, and postapproval manufacturing changes. This diverse information requires a multidisciplinary approach to the quality assessment of drug marketing and licensing applications. OPQ's quality assessment of a drug application employs a team of applicable experts in drug substance, drug product, manufacturing, and biopharmaceutics. Additional technical advisers for the quality assessment may contribute expertise in quality surveillance, research, policy, and other regulatory functions.

OPQ's human drug portfolio includes new drugs, generics, over-the-counter drugs, and biologics including biosimilars. More than 1000 original New Drug Applications (NDAs), Abbreviated New Drug Applications (ANDAs), and Biologics License Applications (BLAs) are typically submitted annually to CDER. A growing percentage of these applications include complex drug products and complex manufacturing processes. Such original marketing or licensing applications do not constitute OPQ's full workload; OPQ also receives several thousand investigational new drug (IND) applications and postapproval change submissions in a year (FDA, 2019). Even in fully electronic submissions, all data may not be amenable to analysis if submitted in an unstructured format (e.g., in PDF files) that challenges data access. Regulators are often asked to prioritize and expedite the evaluation of applications with the goal of accelerating access to important clinical advancements or addressing urgent public health issues such as drug shortages (Panzer et al., 2023). FDA needs innovations to modernize quality assessments in response to technological advancements and growing expectations from the public and pharmaceutical industry.

In the past two decades, FDA has driven continual improvement in the quality assessment function with progressive movement toward a more digitalized landscape. Before the introduction of the Electronic Common Technical Document (eCTD) in 2002, individuals performed assessments using a summary-based approach to document decision-making, with a focus on the adequacy of quality testing, and applications were paper-based. Since 2002, the eCTD has guided industry stakeholders in organizing and submitting electronic regulatory submissions around the globe. Compared to paper submissions, eCTD adopters shortened their total time to approval and demonstrated cost savings (Suchanek and Ostermann, 2012). The shift to eCTD was a steppingstone for regulators to develop new, streamlined approaches for improving quality assessment efficiency and effectiveness.

A network of quality assessment innovations was developed for CDER to leverage information technology capabilities, structured team-based assessments, and increasing knowledge management. This network of regulatory innovations includes the Knowledge-Aided Assessment and Structured Application (KASA), FDA's contributions to a revised common technical document for quality (International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) M4Q(R2)), approaches to structured data on Pharmaceutical Quality/Chemistry, Manufacturing and Controls (PQ/CMC), the Integrated Quality Assessment (IQA), a Quality Surveillance Dashboard (QSD), and CDER's implementation of the Established Conditions (ECs) tool in ICH Q12. Each regulatory innovation brings a solution to a quality assessment challenge (Table 1). Though this paper focuses primarily on CDER's efforts related to new drugs, generic drugs, and biotechnology products, several innovations are being developed in collaboration with CBER and other global regulators (i.e., ICH M4Q(R2), PQ/CMC, and ECs).

Table 1.

Regulatory innovations that address CDER's quality assessment challenges.

CDER Challenge	Solution
A high volume of quality assessments that rely on freestyle narratives results in a laborious review process.	KASA improves overall efficiency and consistency of regulatory decisions by capturing and managing knowledge, establishing rules and algorithms for risk assessment and communication, offering a structured assessment that reduces reliance on text-based narratives, and using computer-aided analyses to assess regulatory standards and quality risks.
The current structure of eCTD Module 2 lacks coherence and leads to assessors seeking information scattered across the submission.	ICH M4Q(R2) improves organization of the information in eCTD to improve the accessibility of quality information for registration, assessment, and lifecycle management of pharmaceuticals.
A lack of structured data in the eCTD poses challenges to risk assessment, efficient application assessment, and knowledge management.	A structured format for submitted PQ/CMC data will contribute to a more efficient regulatory assessment and enable content consistency by creating a standardized data dictionary.
Perceived burden of postapproval change reporting can disincentivize industry from continual improvement while large numbers of application supplements can overwhelm CDER's quality assessment resources.	ECs can reduce CDER's supplemental submission burden by offering regulatory flexibility to applicants that use robust pharmaceutical quality systems and demonstrate strong scientific understanding of elements necessary to assure product quality. This can streamline FDA's focus on high-risk changes and facilitate continuous improvement by manufacturers.
Facility-related information is stored in multiple disparate systems or databases which makes the assessment process inefficient.	QSD provides a framework for consistent and up-to-date assessments of facilities and identification of potential product quality signals through predictive analytics and natural language processing. The dashboard incorporates interactive visualizations that give assessors the ability to discover and share insights regarding quality system effectiveness and manufacturing capabilities.
Lack of a multidisciplinary approach on an assessment team leads to inefficient and redundant review processes.	IQA process facilities a multidisciplinary, team-based quality assessment on an aligned team framework. This allows assessors from different quality disciplines to communicate clearly to applicants and drive effective, collaborative, and efficient quality assessments.

Even as the volume and complexity of quality assessment work continue to increase, patients and consumers expect and deserve rapidly available, high-quality medicine. These innovations work together to create a regulatory assessment network within CDER. This network can handle quality assessment challenges in different phases of regulatory application submission, assessment, and lifecycle management, both now and in the future (Fig. 1).

Fig. 1.

CDER's Network of Regulatory Innovations to Improve Quality Assessment. The network of innovations is a collaborative effort to improve quality assessments. ICH M4Q(R2) modernizes the eCTD Quality section of an application in Modules 2 and 3 by organizing the information in a structured format. ICH M4Q(R2) will enable the connection between PQ/CMC and KASA. PQ/CMC establishes and implements electronic standards for submitting data. KASA captures and manages knowledge during the application lifecycle by providing a framework for a structured quality assessment. QSD provides facility information to KASA and augments KASA by providing a framework for consistent evaluation of facilities. QSD incorporates interactive visualizations that enable assessor access to current and historical information regarding facilities, manufacturing capabilities, and product quality issues. IQA teams use information funneled through KASA and the QSD to conduct effective and efficient assessments of drug applications. ECs are specifically identified in an application and informed by the QSD. ECs enable lifecycle management and provide information to be captured in KASA.

2. Knowledge-aided Assessment and Structured Application (KASA)

One of the first FDA regulatory innovations was KASA, a system that takes advantage of information technology to streamline regulatory submission, assessment, and registration using structured data, analytics, and knowledge management (Lawrence et al., 2019). Previous approaches to quality assessment relied on an assessor's freestyle narrative regarding the information from a PDF document provided by an applicant. Such a system can result in lengthy text, inconsistent and encumbered ability to share knowledge, difficulty extracting data, and hindered decision-making. The KASA system captures and manages information about risk and control strategies for drug substance, drug product, manufacturing, and facilities in a structured format. The information submitted to and generated by FDA is organized and accessible throughout the lifecycle of the product. KASA enables a more efficient assessment by reducing information redundancies and enhancing communication across CDER assessment disciplines. The launch of KASA marks significant progress in realizing the full potential of knowledge and risk management.

Despite its utility, KASA is not a replacement for human quality assessors. Instead, as its name implies, it aids the assessor by establishing rules and algorithms for the assessment disciplines to facilitate risk analysis, mitigation, and communication. Risk identification involves computer-aided analyses of applications against regulatory standards and the determination of quality risks throughout the lifecycle of a product based on prior knowledge. KASA also structures the assessment document to minimize unnecessary text-based narratives and summarization of information already provided by applicants.

KASA's development has been the subject of two FDA Advisory Committees, in 2018 and 2022 (FDA, 2018; FDA, 2022a). In each instance, the advisory committee unanimously supported KASA's further development and implementation. KASA was also recognized as part of two FedHealth IT Innovations Awards, in 2021 and 2022, honoring peer-selected programs that drive innovation across federal departments (FDA, 2023). FDA anticipates further improvements to KASA as the technology evolves. Maximizing the benefits of KASA requires information and data that is well-structured and organized in regulatory submissions over a product's lifecycle.

3. New Common Technical Document for Quality (ICH M4Q(R2))

The Common Technical Document, established in the ICH M4Q(R1) guideline, and the resulting eCTD, provide the standardized format enabling KASA's knowledge management functionality (ICH, M4Q(R1), 2002). Although it harmonized the quality information for the registration of pharmaceuticals for human use, several ICH regions have not fully implemented the guideline (ICH, 2021a). ICH M4Q(R1) is now due for revision to further improve registration and lifecycle management efficiency, leverage digital technologies, and accelerate patient and consumer access to pharmaceuticals. Submissions using the structured format outlined in the eCTD streamline communication between applicants and FDA, foster transparency for FDA assessors, and improve global regulatory knowledge management over the product lifecycle. The desire to leverage digital technologies, such as KASA, to further improve lifecycle management and align submissions with modern quality guidelines ICH Q8-Q14, propels ICH M4Q's current revision (ICH, 2021a). Updating the ICH M4Q guideline aims to provide industry with regulatory expectations, enable industry to communicate their chemistry, manufacturing and controls (CMC) information in a consistent and accessible way, and harmonize the submission and assessment of applications.

The ICH M4Q(R2) revision intends to capture quality information needed to support registration and to specify the location in the application for submitting lifecycle management elements (ICH, 2021b). Further, it reduces diversity in requirements for quality information across ICH regions and aspires to streamline requests for pharmaceutical quality system (PQS) and certain relevant Current Good Manufacturing Practice (CGMP) information. The revision intends to facilitate information supporting emerging concepts such as advanced manufacturing, digitalization, data management, artificial intelligence, and advanced analytical tools.

eCTD format enhancements aim to improve submission and assessment efficiency, resulting in accelerated patient access to pharmaceuticals. An efficient and effective quality assessment using KASA requires submissions that incorporate data organization and structured data standards via ICH M4Q.

4. Structured Data on Pharmaceutical Quality/Chemistry, Manufacturing, and Controls (PQ/CMC)

Although ICH M4Q(R2) provides organization for CMC information that facilitates science- and risk-based quality assessment, it does not address standards for data structure and exchange. Without standards for data structure, it can be challenging to conduct consistent risk assessments and quantitative analyses using information from applications. The PQ/CMC effort focuses on establishing such standards to enable a seamless connection and transfer of data from the submitted dossier to the regulatory authority for assessment. Such data includes manufacturing and quality data for pharmaceutical products, for example, data related to drug stability, product specification, and batch analysis. As defined by ICH M4Q, structured data are intended to be submitted in eCTD Module 3 and serves as the information and data repository that supports Module 2. Structured data is integral to the foundation of regulatory assessment and lifecycle management.

FDA has led an effort to identify and prioritize the PQ/CMC information that would benefit from a structured submission approach (FDA, 2022b). The goal of this effort was motivated by provisions, that authorized FDA to require certain submissions to be submitted in a specified electronic format (Office of the Federal Register, National Archives and Records Administration, 2012). Although still under development, PQ/CMC will provide standard data elements and data exchange standards to industry so that future submissions will contain structured quality information that can be used by KASA, QSD, and other FDA systems.

The benefits of structured data are ensuring “same data” for FDA and industry and eliminating the need for manual manipulation of data prior to analytical system consumption. PQ/CMC is an enabler of KASA, it enhances ICH M4Q, accelerates digitization efforts, and ultimately improves lifecycle knowledge management. KASA, ICH M4Q, and PQ/CMC work together to address the management of information related to a drug product across its lifecycle, including when changes are made.

5. Implementation of ICH Q12 Established Conditions (ECs)

Quality assessments continue beyond the initial application approval process. Applicants desiring to make a change to the quality aspects of a product or its manufacturing process evaluate the potential of the change to impact product quality. In the context of the regulations, they then determine the need to report the change with appropriate supporting data. The concept of ECs was first described in the FDA draft guidance for industry entitled “Established Conditions: Reportable CMC Changes for Approved Drug and Biologic Products,” issued May 2015, and has been further discussed in the ICH guidance for industry entitled “Q12 Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management; International Council for Harmonisation” issued May 2021. ECs are those CMC elements of the drug application (e.g., manufacturing process, facilities and equipment, and elements of the associated control strategy) that are necessary to assure product quality. The ICH Q12 guidance provides a framework to facilitate the management of postapproval CMC changes in a more predictable and efficient manner by taking into consideration the applicant's product and process understanding and the effectiveness of the manufacturing site's PQS. This framework promotes continual improvement and innovation throughout the product lifecycle and reduces product availability risks.

ECs can assist FDA by lessening the volume of supplement submissions for evaluation and improving effective knowledge management throughout the application lifecycle. Each application submitted to FDA contains ECs; however, they are not always identified in the application at the time of approval. Confusion as to what constitutes an EC could have a negative impact on change management activities by disincentivizing manufacturers to make changes that could improve quality. Furthermore, this can result in the unnecessary submission of postapproval supplements to FDA for changes that could have been managed solely by a manufacturer's PQS. On the other hand, such confusion could lead to a failure to submit the required supplement for a change. The ICH Q12 guidance describes how an applicant can identify and propose specific ECs and reporting categories for changes to those ECs based on risk. Once approved as part of the application, ECs and the associated reporting categories facilitate efficient risk-based change management and promote effective postapproval strategies.

The modern approach to risk assessment of postapproval changes, enabled by ECs, is not limited to information provided in an application. To enable more informed assessments and decisions, risk management relies on all relevant information, especially the risk analytics provided by KASA and QSD. KASA will capture the risk assessment of ECs throughout the product lifecycle, while the QSD will facilitate knowledge management and facility assessment for their ability to effectively implement proposed changes.

6. Quality Surveillance Dashboard (QSD)

Facility assessment entails a comprehensive evaluation of product attributes, facility manufacturing processes, capabilities, and inspection history. However, the relevant data for such an assessment have historically been collected and maintained in multiple disparate data systems. Hence, when conducting facility assessments, quality assessors can spend extensive time manually checking, extracting, and assessing data from each system. Performing an efficient facility assessment requires a centralized and automated system that provides comprehensive and up-to-date quality-related information on facilities.

QSD was developed to provide a framework for consistent assessment of CDER-regulated facilities through reporting, data exploration, and analytics conducted using multisource facility information in one location. The dashboard was designed to provide the assessor with the ability to navigate details about each facility's products (e.g., application status, supply chain role, route of administration, product availability) and an overall timeline of the facility's history. It uses both predictive models and natural language processing (automated techniques to detect and extract relationships and sentiments in text data) to enable efficient risk-based facility assessments. For example, assessors can quickly obtain an overview of a facility's performance to make data-driven decisions. A quality assessor can then conduct a comprehensive assessment and consider relevant risks by combining the information on the facility and the application.

Feedback gathered from users of a dashboard prototype honed data visualization and reporting for QSD. The user interface incorporates interactive visualizations that enable users to discover and share insights regarding facilities, manufacturing capabilities, and product quality information. Some of the key features of QSD enable an assessor to integrate and govern facility and postmarket product quality data across multiple systems. Currently, QSD enables efficient, objective, risk-based facility assessments for pending applications. Soon, the integration of data collected from KASA will enhance analytics within QSD. As QSD matures, it will expand the number of users, expand data acquisition (including data from KASA), and enhance data cleaning methods. However, even with ideal data, computers do not perform application assessments. Applications are assessed by a team of FDA experts who must be organized and equipped to handle the diverse regulatory submissions of the future.

7. Integrated Quality Assessment (IQA)

Some innovations for quality assessments are neither electronic nor digital. An efficient quality assessment of an application, even when using advanced technology, requires the seamless integration of different, relevant quality disciplines. The IQA process facilitates a multidisciplinary, team-based approach that follows a defined business process with clear roles and responsibilities for each team member. IQA teams comprise an application technical lead, a regulatory business process manager, discipline assessors, and additional technical advisers as needed. One of the challenges of the team-based IQA approach is that it requires team- and relationship-building, that is, cohesion and consistency within teams, and careful design to bring individuals from across disciplines together to efficiently address a large volume of applications.

OPQ has begun to address this issue by adopting an “aligned teams” strategy when designing IQA teams. This strategy draws from smaller pools of individuals across disciplines to assign aligned IQA teams, resulting in groups of assessors that more frequently work together and therefore have more established working relationships and improved trust. This enables teams to conduct quality assessments as effectively, collaboratively, and efficiently as possible while providing more consistent feedback to applicants. IQA teams foster better relationship-building among assessors from different quality disciplines and therefore maximize the use of each team member's expertise to provide an integrated quality recommendation for marketing applications that is patient-focused and risk-based.

OPQ first adopted the IQA process in 2015, and the positive experience with the IQA helped support CDER's Office of New Drugs implementation of integrated NDA drug review teams in 2020 (FDA, 2021). FDA is driving toward desired future attributes: ensuring organizational agility in an ever-changing regulatory environment and enabling connected cross-functional teams. Looking forward, OPQ will continue to monitor Aligned Teams for opportunities for continual improvement. This ensures OPQ's evolution toward a more agile assessment function to manage applications, regardless of submission variability, and achieve a better workload balance across assessors.

8. Conclusion

The network of quality assessment innovations is improving CDER's ability to handle the increasing volume and complexity of regulatory submissions, review processes, and lifecycle management with consistency. ICH M4Q and PQ/CMC will provide information organization and data standards, respectively, to industry so that future submissions can contain appropriately structured quality information. Incorporating both information organization and data structure standards should empower assessors to effectively and efficiently perform quality assessments via KASA.

Traditional, narrative-based quality assessment using unstructured information is an approach of the past. KASA modernizes application assessment by moving from an output consisting of an unstructured text document to a systemic risk-based regulatory and technical assessment with a uniform output using structured data and information. Along with KASA, QSD supports IQA teams performing facility assessments. Information from the QSD also informs the assessment of ECs, which are specifically identified in an application and captured in KASA to enable lifecycle management. These combined innovations advance application assessment and capture and manage knowledge throughout a product's lifecycle to support the availability of quality medicines to U.S. patients and consumers.

These innovations are meant to assist assessors, not replace them. This network of innovations facilitates efficient and consistent quality assessments by FDA assessors, who will still be responsible for making decisions about the approvability of applications. Further, these tools will continue to improve based on available data and evolving experience, reflecting best practices and standards. Several of these innovations might help move toward a global architecture for regulatory decision-making and enable more collaborative efforts with international regulatory organizations, such as the Pharmaceutical Quality Knowledge Management System Initiative (A regulatory pharmaceutical quality knowledge management system (PQ KMS) to enhance the availability of quality medicines, 2024). As quality is the foundation for assuring safe and effective drugs, CDER's network of innovations is necessary to match the changing landscape of public health and human medicine development and manufacturing.

CRediT authorship contribution statement

Russie Tran: Writing – original draft, Visualization, Conceptualization. Grace Fraser: Writing – original draft, Visualization, Conceptualization. Adam C. Fisher: Writing – review & editing, Visualization, Conceptualization. Sau L. Lee: Writing – review & editing, Visualization, Supervision, Conceptualization. Ashley Boam: Writing – review & editing. Stelios Tsinontides: Writing – review & editing. Jennifer Maguire: Writing – review & editing. Lawrence X. Yu: Writing – review & editing. Susan Rosencrance: Writing – review & editing. Steven Kozlowski: Writing – review & editing. Don Henry: Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data was used for the research described in the article.