Ontologies

Chemical development has been supported by the ISA 88 standard for decades.  A lot has changed in that time, and a more flexible, comprehensive representation of process data is needed. A contemporary ontology provides the data representation that is needed to facilitate data integration, data exchange and data insights. 

The Pistoia Alliance has completed Phase 1 / Proof of Concept to build a pharmaceutical CMC process ontology based on the ISA88/95 framework.

The aim is to standardize laboratory and plant production process recipes to establish standardized definitions, facilitate digital technology transfers and integration with execution systems to capture structured process data for material lot genealogy tracking, streamline technology transfers, and advanced process analytics, and thereby enhance efficiency and transparency throughout the pharmaceutical production lifecycle.

For this project’s initial phase, we worked on three use cases: Process Definition, Process Monitoring, and Comparative Analysis. The focus was on the processes for making the APIs and the drug products, initially focusing on biologics and synthetics.

In this video, we show how we have imported the developed core CMC process ontology as well as two controlled vocabularies for referring to the different parameter measurements as well as the different recipe steps that make up a manufacturing process to create a knowledge graph to answer our competency and business questions.

Hear directly from our Project Manager, and ontology expert, Birthe Nielsen and Cameron Gibs from Crown Point, about the “CMC Process Validation Walkthrough” below…

The scope and key deliverables of this work will include:

• CMC laboratory & manufacturing scale API & Pharmaceutical Product processes 
• Biologics & Synthetics with an initial focus on protein and chemical processes
• Develop domain lexicon and domain taxonomy to accomplish seamless convergence of the protein and chemical processes
• An advanced semantic architectural design for a Process Ontology, featuring seamless integration with the established Product (IDMP-O), Analytical (AFO), and Unit of Measure (QUDT) ontologies
• A taxonomy and controlled vocabulary based on the ISA88/95 framework for Process/Stage /Operation/Action, Process Parameters, Process Performance Indicators, and their definitions

Initially, we will work to cover the following use cases:

• Standardize and digitize laboratory production process (stage) definitions, enabling structured process data capture. 
• Establish digital API & Pharmaceutical Product process definitions for automated CMC dossier authoring 
• Enable digital Process Technology Transfer from laboratory to plant (scale up) and between plants 
• Enable advanced process analytics (process performance) across laboratory and process batches (across scale) 

The European Medicines Agency (EMA) will be the first health agency to mandate compliance with ISO IDMP (Q4, 2024), with the FDA not far behind. Governance of essential IDMP standards and implementations is not assigned to a specific, overarching governing body. Diverging implementations of IDMP across geographical regions and jurisdictional domains are already causing inconsistencies in the interpretation across implementing organizations. Given that there is no semantic alignment between regulatory bodies, there exists a risk that regulatory compliance needs will lead to large integration and interoperability costs and that the benefits from IDMP in drug safety, innovation, and other areas will not be fully realized. Rather than concentrating on the discovery of new medicines, organizations will be struggling with data issues, e.g., the need to map product data across the organization, throughout the product lifecycle.

Due to our robust framework for pre-competitive collaboration. The Pistoia Alliance was selected to manage this initiative with the goal of creating an ontology that demonstrates added value to the ISO IDMP standards for data usability across organizational boundaries and regulatory jurisdictions. We have pulled together a core team consisting of Pharmaceutical stakeholders, regulatory bodies, standards organizations, non-profit groups, and solutions providers for this project.

The key objectives  for the project include:

1. Provide a digital, machine-processable standard.
   – The Pdf document provided by EMA is not enough as different groups implement it differently
2. Solve ambiguities of the ISO IDMP standards enabling feed improvements back to ISO through systematic reviews
3. Bridge different views with ONE product data model between internal pharma departments and between industry groups
4. Provide a vendor-agnostic, and open-source model. The ontology is fully standards-based without any proprietary aspects
5. Reduce implementation effort through a common core