Imagine a world where all the data generated in your Chemistry, Manufacturing, and Controls (CMC) processes flows seamlessly across your organization, breaking free from the constraints of siloed systems and proprietary formats. This scenario isn't a far-off dream—it's the promise of "liquid data," a concept rapidly gaining traction in the biopharma industry.
In a recent TetraScience webinar, "The Power of Liquid Data in Biopharma CMC Workflows," industry CMC veterans shared insights that challenge our traditional approaches to data management. As our industry grapples with an expanding universe of modalities—from small molecules to complex biologics and cutting-edge cell and gene therapies—the need for a paradigm shift in handling CMC data has never been more urgent.
But what exactly is liquid data, and how can it revolutionize CMC workflows? Let's dive into the key takeaways from this illuminating discussion and explore how this concept could reshape the future of drug development and manufacturing.
The Challenge: Fragmented Data in a Complex Landscape
The volume and complexity of CMC data continue to grow exponentially, yet much of this valuable information remains trapped in silos, inaccessible to those who need it most.
Matt Burke, the VP and Global Head of R&D (CMC) at the Menarini Group, highlighted this challenge: "You have different groupings of data collected in the CMC development pathway. Even within a company, drug product and drug substance groups are often separate, maybe physically located in different areas of the world."
This fragmentation hampers efficiency and limits the potential for cross-functional insights that could accelerate drug development and improve product quality.
The Solution: Liquid, Engineered Data
Enter the "liquid data" concept - a paradigm shift approaching CMC information. As Ken Fountain, TetraScience’s VP of Scientific Applications, explains, "Liquid data is liberated from silos, extensible, shareable amongst various groups, and can move between instruments and applications."
This concept goes beyond mere data accessibility. Liquid data is part of a broader framework of engineered data designed to be AI-ready and optimized for analytics. As Fountain elaborated, truly liquid data is:
- Centralized: Raw data is moved into a cloud-based location, allowing large-scale computing and drawing relationships between related datasets.
- Contextualized: Proper metadata is added, including system type, project ID, batch ID, and other typical attributes of scientific data.
- Standardized: Data is engineered into a standard, vendor-neutral format, unlocking it from proprietary systems.
- Interoperable: It can interact with data from various systems, including batch information, procedures, and quality management systems.
- Scalable: The approach can handle data from thousands of instruments across an enterprise.
By transforming CMC data into this liquid state, companies can open new possibilities for data analysis, cross-functional collaboration, and process optimization throughout the drug development lifecycle. For example, scientists and process engineers from sponsor companies and contract partners can work together iteratively on the same scientific data in real time.
Liquid data also leads to faster decision-making and improved process understanding. Matt Burke shared an example where complex dissolution data, once visualized effectively, allowed for much quicker consensus among team members. He described a particularly challenging fixed-dose combination product that required bioequivalence testing in ten different dissolution media. The complexity of the data made it difficult to interpret trends. Still, when presented in an interactive dashboard, the team could quickly identify optimal formulation parameters, significantly accelerating the decision-making process.
With liquid data, companies can better assess and mitigate risks across their product portfolio, a capability on which regulatory bodies like the FDA are increasingly focusing. Matt Burke brought up how liquid data strategies could help biopharma companies align better with the FDA's KASA (Knowledge-aided Assessment & Structured Application) initiative, which aims to move away from narrative-based submissions to more structured, data-centric approaches.
Implementing Liquid Data: Challenges and Strategies
While a liquid data strategy's benefits are clear, implementing it comes with its own challenges. Foremost among them is the requirement for a cultural shift. Industry CMC leader Jo Barrett noted on the webinar that "we've all been battling with data for a long time. So there are probably lots of local departmental initiatives to make data more available within that department. Here, we're looking at having centralized data across the whole organization."
This shift requires strong sponsorship at the enterprise level and buy-in from scientists who may be accustomed to their data workflows.
Ensuring data quality at the source is also crucial because building trust in the centralized data system is essential for its adoption. As Matt Burke emphasized, "My instinct is, you should have data quality at the beginning. In the opposite scenario, I imagine you're collecting the data but not scrutinizing it enough... and it points you in a certain direction, and then you try it, and it doesn't work."
There's also often tension between the needs of scientists generating data and those using it downstream. Mark Buswell, an experienced life sciences leader in CMC, observed, "For the scientists at the bench who create the data, that makes their life more complicated... Their experiment setup gets a bit slow because they have to enter many parameters into the ELN."
Finding ways to streamline data entry while maintaining quality is critical.
Regarding technical implementation, Ken Fountain outlined some critical aspects of creating liquid, engineered data. "The first thing we must do is get that raw data into a centralized location and add the proper context... Once that data is centralized in a repeatable way and given context through metadata... it can be rapidly searched and associated with other data." This process, he said, involves careful consideration of data schemas, taxonomies, and ontologies.
The Path Forward: Strategies for CMC Leaders
To harness the power of liquid data in CMC workflows, consider the following strategies:
- Start small. Ken Fountain advises you to start in an area where you see some or all of the benefits of lab data automation and data visualization. “Where would those types of things help with many pain points in the CMC process?"
- Focus on high-impact areas. Identify process steps where AI could significantly impact CMC activities and work backward to determine what data needs re-platforming.
- Invest in change management. Recognize that implementing a liquid data strategy is as much about people and processes as technology. Invest in training and communication to ensure buy-in across the organization.
- Tap Into industry resources. Explore resources and working groups such as ISPE, the Pistoia Alliance's CMC Ontology project, and FDA initiatives like KASA (Knowledge-aided Assessment & Structured Application) to stay informed about best practices and emerging standards.
- Consider the entire product lifecycle. Jo Barrett emphasized the importance of thinking beyond development: "That need to transfer products quickly, to make changes to raw materials... if we've got that data, and we can assess the risk, assess what might be going on, and just be more confident of a right-first-time transfer, a right-first-time material change is going to impact our patients."
A Data-Driven Future for CMC
As the biopharma landscape evolves, CMC leaders have a unique opportunity to drive innovation through more intelligent data strategies. By embracing the concept of liquid, engineered data, companies can break down silos, accelerate decision-making, and ultimately bring life-changing therapies to patients faster and more efficiently.
The journey to liquid data may be challenging, but the potential rewards—accelerated development timelines, improved product quality, and enhanced regulatory compliance—make it a strategic imperative for forward-thinking CMC organizations.
As you embark on this transformation, remember that the goal is better data management and a fundamental shift in how we approach drug development and manufacturing. In this new paradigm, data becomes a byproduct of our processes and a driving force for innovation and excellence in CMC.
To hear the entire roundtable discussion, watch the webinar on demand.