Pharmaceutical Quality (Industry), published by HL7 International / Biomedical Research and Regulation. This guide is not an authorized publication; it is the continuous build for version 1.0.0 built by the FHIR (HL7® FHIR® Standard) CI Build. This version is based on the current content of https://github.com/HL7/uv-dx-pq/ and changes regularly. See the Directory of published versions

Home Page

Official URL: http://hl7.org/fhir/uv/pharm-quality/ImplementationGuide/hl7.fhir.uv.pharm-quality				Version: 1.0.0
Active as of 2024-05-08				Computable Name: PharmaceuticalQualityIndustry
Other Identifiers: OID:2.16.840.1.113883.4.642.40.41

Purpose

This implementation guide defines a universal realm FHIR specification for the creation and exchange of structured pharmaceutical quality data internationally within or between biopharmaceutical companies and their stakeholders.

Scope

In-scope

• Products regulated as drugs (Small and large molecule medicinal products together with their functional and non-functional packing components; prescription, over the counter and hospital products)
• Developmental and authorized medicinal products
• Data exchange within a biopharmaceutical company, between companies, or between companies and their industry stakeholders (e.g., contract labs, contract manufacturers) (See Background/Considerations for Implementation for further detail on this topic.)

Out of scope

• Products regulated as devices
• Veterinary drugs
• Data exchange with National Competent Authorities / national regulatory authorities for medicinal product authorization and maintenance.
• Instructions on how to transform from the default FHIR XML or JSON described in this IG to another format required by National Competent Authorities / national regulatory authorities (e.g., Word, PDF).

Note:

• Devices and veterinary drugs are only out of scope for version 1.0.0 of the implementation guide. New product lines will be brought in scope in future iterations.
• This implementation guide is designed to serve as a standard for structuring and exchanging pharmaceutical quality content; i.e., a content exchange standard. This implementation guide is not a submission standard and is therefore not intended for submission to National Competent Authorities / national regulatory authorities for medicinal product authorization.
• Biopharmaceutical sponsors are expected to reference and adhere to the relevant national regulatory guidance when preparing data for submission to National Competent Authorities / national regulatory authorities responsible for medicinal product authorization. E.g., Health Canada Filing submissions electronically webpage; U.S. Food & Drug Administration (FDA) Electronic Regulatory Submission and Review webpage (and the developing project for PQ/CMC FHIR Submissions to U.S. FDA); or the European Medicines Agency (EMA) eSubmission website.

Technical Overview

This implementation guide is based upon FHIR version 5.0.0 and is a Universal Realm Specification.

Audience

This guide is primarily meant for but not necessarily limited to organizations involved in the following activities:

• Medicinal product manufacturing
• Laboratory Information Management
• Contract manufacturing
• Biopharmaceutical regulatory affairs
• Developers (structured content authoring solutions, content management solutions, Laboratory Information Management Systems [LIMS])
• Business analysts and policy managers can also benefit from this guide.

Background

Introduction to Pharmaceutical Quality

Pharmaceutical quality, also known as Chemistry, Manufacturing, and Controls (CMC) or just “Quality”, describes how a medicinal product is developed and manufactured; what suppliers and sites are involved; and the controls established to maintain the quality of the medicinal product over its lifecycle.

The scope, best practices, definitions, and content for pharmaceutical quality data are defined by the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH), ICH Harmonised Tripartite Guideline: The Common Technical Document for the Registration of Pharmaceuticals for Human Use: Quality –M4Q(R1); Module 3: Quality, (12 September 2002), also known as the Common Technical Document (CTD).

The exchange and maintenance of pharmaceutical quality data account for the majority of all internal and external data interactions, internationally, within companies and between a company and stakeholders. Despite much of the content being sourced from internal systems that support structured data (e.g., pharmaceutical development systems, manufacturing site systems), and much of it being available for reuse across many medicinal products, pharmaceutical quality content has typically been authored, managed, and transmitted in unstructured Word files or PDFs. This is not sustainable, given the increasing volume, complexity, and frequency of quality work.

Strategic Goals

In an effort to adopt a more efficient operating model, the biopharmaceutical industry is about to transition into its third era or paradigm and, in addition to other benefits, this implementation guide will help accelerate this transition (see figure below). These eras are summarized as follows:

• The first was the pre-2003 era, it was defined by paper documents printed, published into binders and physically delivered to international health authorities via courier. Supported by manual and effort intensive processes. Limited throughput.
• The second was the 2003 to 2023 era, it was defined by electronic paper documents (E.g., (PDF) or Microsoft Word) published and delivered to international health authorities via electronic submission gateways. Supported by simple/automation and effort intensive processes. Increased throughput but still limited throughput.
• The third, which is about to begin, is the post-2023 era, it will be defined by structured data, in its most granular form, delivered via high-frequency application programming interfaces (APIs) and processed by artificial intelligence (AI) agents. Advanced automation, low effort, high volume, high frequency, and high throughput.

Figure: Advancement of the Submission and Decision-Making Ecosystems: The biopharmaceutical industry has undergone a previous transformation from paper-based submission to electronic paper submission. The industry is at the beginning of the next iteration of transformation, as signified by a transition to FHIR API data submission. This will be followed by further advancement as AI algorithms become increasingly prevalent. As the industry progresses in its journey towards digital maturity, faster submission and review processes will be made possible.

Currently, the end-to-end timescale for the pharmaceutical regulatory workflow is measured in months and years. This new paradigm, facilitated by this implementation guide, will use FHIR APIs and other supporting technologies to reduce the potential time for data exchange from months to days, hours, minutes, and eventually sub-seconds.

Establishing a viable, real-time method to exchange product quality information across biopharmaceutical manufacturing organizations can help to facilitate technology transfer, allow enhanced monitoring of facilities, and can potentially support developing advanced manufacturing methodologies, such as point-of-care manufacturing.

Importantly, spurring innovation in the biopharmaceutical sector is only one benefit of this real-time data exchange approach. Ultimately, the approaches highlighted within this implementation guide will help bring high quality medicinal products to market faster, and in a more cost-effective manner. This will also help create more robust manufacturing processes and more manageable global supply chains. This in turn will help reduce manufacturing down time, reduce the chance of shortages, and increase product quality and data compliance. Taken together, each of these benefits will enable speedier patient access to new treatments and minimize the post-approval life-cycle management burden for sponsors, while contributing favorably to overall global patient experience.

Structured Data with High Granularity

Given the Strategic Goals noted above, this implementation guide encourages the biopharmaceutical industry to adopt a new paradigm based on structured data with a high degree of granularity.

For example, each individual test (i.e., ObservationDefinition resources) is designed to be a discrete component. As a discrete component, this allows each test to be managed in a master data library where it can be centrally maintained, version controlled, and reused across different contexts or use cases. Each test, or ObservationDefinition, can be rolled up and combined with other resources into a Bundle that represents a larger concept like a stability study or test specification. It therefore becomes easier for sponsors to trace what tests, or what versions of tests, were used in any given stability studies or test specifications for any given batch over time. The same “rollup” or “bundling” approach is applied across all components used in the 14 data domains described in section 3 Domains Overview.

Once the individual components, like the ObservationDefinitions, are rolled up into bundles with other resources they can easily be exchanged between systems as FHIR compliant XML or JSON by default. However, if needed, these same bundles can be transformed into any other format that the biopharmaceutical sponsor may need. E.g., an XML or JSON bundle can be transformed into a PDF or Word document that maintains the traditional look and feal of monolithic ICH Common Technical Document (CTD) template.

This high degree of granularity maximizes a sponsor’s flexibility and choice in how they choose to manage their data; maximizes data reuse; reduces data redundancy; supports transformation into any desired output format; and makes this data more accessible for data science, ontologies, and artificial intelligence.

Considerations for Implementation

The following figure depicts (1) a potential future data pipeline for implementers; and (2) the intended scope of this IG’s use relative to the regulatory submission preparation workflow. As noted above, data structured according to this implementation guide is designed to be highly granular. The figure, and the following description, are provided to facilitate discussion and understanding of how to use the highly granular data models described in this implementation guide.

Figure: FHIR Data Pipeline for Pharmaceutical Quality (Industry)

In summary, this potential future data pipeline could work as follows,

• Step I: FHIR data is sent to the sponsor’s partners to conduct a particular activity on their behalf (e.g., run a stability study). The sponsor’s partner returns FHIR Data to the sponsor from source systems (e.g., LIMS) into systems designed to manage regulatory information.
• Step II: Regulatory systems collate the data into individual Resources and Profiles (See Domains Overview for further detail). This may include a mix of submittable and non-submittable content (e.g., transactional details). When preparing for a regulatory submission, the submittable content is extracted and bundled to prepare for further transformation.
• Step III: The submittable content needs to be transformed from the default FHIR XML or FHIR JSON format to a submission ready format compliant with the relevant standards prescribed by each regulator (refer to national regulatory guidance for those submission requirements). With the right software, the FHIR compliant XML or JSON can be transformed into any output format stipulated by any regulator. The transformer(s) should be designed to make the output comply with content and style requirements as needed.
• Step IV: The submission ready output undergoes validation, or any other required processing steps, to confirm compliance with a national regulator’s defined technical and business rules.
• Step V: The validated submission ready output is delivered via the regulator’s accepted method of exchange (e.g., FHIR compliant API, Electronic Submission Gateway, physical mail).

The scope of this implementation guide only covers data exchange activities for Step I, Step II, and Step III. The activities depicted in Step IV and Step V are out of scope for this implementation guide since they are determined by national regulations and regulatory guidance.

Project Team

Biopharmaceutical industry:

• Rita Algorri, Amgen Inc.
• Sheetal Gaiki, Johnson & Johnson
• Mike Abernathy, Amgen Inc.
• Craig Anderson, Pfizer Inc.
• Colin Wood, AstraZeneca

Lantana Consulting Group:

• Rik Smithies
• Rob Hausam
• Diana Wright

Acknowledgements

This guide is developed and produced through the efforts of Health Level Seven International (HL7) as part of the Pharmaceutical Quality (Industry) project (Project ID 1800 and Project Scope Statement: PSS-2145).

The project team appreciates the support and sponsorship of the HL7 Public Health Work Group, and all volunteers and staff associated with the creation of this resource. The team appreciates the comments and input from the biopharmaceutical community as well as the HL7 volunteers who participate in the Biomedical Research and Regulation Work Group.

Health Level Seven, HL7, FHIR and the [FLAME DESIGN] are registered trademarks of Health Level Seven International, registered in the US Trademark Office. For acknowledgement of terminology content, please see the IP Statements section.

Providing Feedback

Please suggest improvements and provide notes about issues through HL7’s Jira Software (free account and log-in required).

HL7’s Confluence pages provide guidance on submitting Specifcation Feedback in Jira.

See also the HL7 FHIR Community Chat Zulip stream Pharm Quality Industry IG for discussion about the Pharmaceutical Quality (Industry) FHIR IG and BR&R Pharmaceutical Quality (PQ) - Industry Use Case project.