Enhancing Process and Parameter Control for Accelerated Time to Patient

In today’s highly regulated environment for drug development and manufacturing, new challenges have emerged from internal pressures to reduce costs by eliminating waste in the development process, streamlining supply chains and expediting and simplifying technical transfers to deliver therapies to patients more quickly. These challenges are further complicated by the ongoing growth of outsourcing to contract development and manufacturing organizations (CDMOs).

Central to these activities is the development, testing and handoff of control strategies and defined manufacturing processes, with all requirements detailed in the FDA and EMA’s process validation guidelines. This article will discuss the necessity and advantages of adopting a digital approach across all related activities for enhanced control of the design process and its associated parameters and attributes, from development through tech transfer to manufacturing.

In this post-pandemic drug development era, the urgency surrounding time to market, time to filing, time to milestone, and most importantly, time to patient, has never been more pronounced. Accelerating the process of drug identification and design involves research and development, effectively performing a technology transfer (tech transfer) to a manufacturing network with an integrated supply chain while maintaining defendable quality, and finally packaging, shipping and delivering therapies to patients in a timely manner. Given that timelines for this cycle often span years, sometimes up to a decade, and considering that each day removed can save anywhere between $600,000 to $8 million, the potential savings can be substantial.

While there are opportunities for time savings through the entire product lifecycle, one area that is often overlooked and/or simply ignored is the management of the process itself. While the iterative process of development aims to identify and control process parameters and quality attributes, the capture, versioning and management of these parameters and attributes are often handled manually. This contrasts with experiment execution, methods and workflows that may be fully digitized within a mature digital framework.

The guide abides

The FDA’s 2011 Guidance for Industry – Process Validation: General Principles and Practices states, “Process knowledge and understanding is the basis for establishing an approach to process control for each unit operation and the process overall.”  This is broken up into three core stages: process design, process qualification and continued process verification.

Similarly, the EMA Guideline on process validation for finished products – information and data to be provided in regulatory submissions says that “Process validation incorporates a lifecycle approach linking product and process development, validation of the commercial manufacturing process and maintenance of the process in a state of control during routine commercial production,” using the term “ongoing process validation” in lieu of “continued process verification.”

Common to both the EMA’s and FDA’s guidelines is the effective performance of process characterization to drive process control, aiming to understand process variability, establish the control strategy and proactively mitigate risks.

Key deliverables of the process characterization include the critical quality attributes (CQAs), critical process parameters (CPPs), the design space with its acceptable range of operating limits and conditions, the control space or control strategy, the risk assessment with associated mitigation strategies, and the validation plan to detail how the process can be tested and validated to consistently produce a quality product.

The FDA, along with the ICH (International Conference on Harmonization), further introduced the methodology of quality by design (QbD), emphasizing a risk-based approach, and cementing the guidelines in ICH Q8, ICH Q9, ICH Q10 and ICH Q11.

However, despite extensive guidance, QbD and process control are often misunderstood in terms of how they impact each other. Spirited debates can be had for defending both statements that process control supports QbD and QbD leads to process control. In reality, there is a symbiotic relationship between the two (Figure 1).

Figure 1. Process control in a QbD-driven environment to deliver high-quality product. Credit: IDBS.

• Through QbD, CPPs and CQAs are identified and defined, forming the basis for developing effective process control strategies. By maintaining tight control over CPPs and CQAs, process control ensures the product consistently meets its predefined quality targets as specified in the control strategy.

• QbD starts with a thorough understanding of both the product and the process, leading to the design of robust manufacturing processes capable of producing high-quality products. Effective process control involves extensive data collection and analysis, providing valuable insights into process understanding. This data is essential for QbD activities such as risk assessment and continuous improvement.

• QbD promotes a risk-based approach to process development and control. By identifying potential sources of variability and implementing appropriate control strategies, QbD ensures that the process is well-controlled and capable of consistently delivering high-quality products.

In the development of a new drug or therapy, applying QbD and quickly attaining true “process understanding” is a daunting goal. From the receipt of the drug from its preclinical phases, along with the quality target product profile (QTPP), there is intense pressure to complete process development and process characterization, delivering a control strategy into the tech transfer activities.

Digital maturity

Data availability is directly related to a company’s digital maturity. 

Where product data is concerned, many organizations already have several systems across process development and manufacturing that capture or assist in the capture of data supporting execution, often related to the product itself. E.g., the pH at 7:52 am was 6.9, the feed was started at 8:03 am, etc. However, very little information exists to support the evolution or development of the process itself.

Having process definitions, in their myriad versions, along with supporting data in Excel spreadsheets is indeed “digital” in nature but is very low on the data maturity scale. This forces the manual management of the process definition itself, rarely includes learnings and insights about the process, and does not implicitly foster process improvements. This has led to the current hybrid approach to digital maturity between process and product knowledge.

While process definition stores and recipe management systems are available and can provide much of this capability, they often reside in the manufacturing space in tandem with deployed manufacturing execution systems, focusing on horizontal rather than vertical tech transfer, with an emphasis on manufacturing and regulatory needs like continued process verification and product release.

By implementing these systems earlier in the product lifecycle, they can help document the development of process, while the ELN (electronic lab notebook) documents the product. The combination of these data, collected early in development, not only smooths the tech transfer process, but also creates the data architecture required to support future advanced analytics such as parameter prediction and AI in process control.

These systems, implemented early and optimized for use throughout the drug lifecycle, can support a range of activities:

• In R&D, these systems can jump-start new formulations. 
• In process development and scale-up, they support process optimization and parameter management.
• In manufacturing, they drive batch execution and management.
• For supply chains, they can speed material and related supplier identification and promote harmonization across the manufacturing network.
• QA, QC and Regulatory all benefit from access to digital copies of the process definitions.
• For tech transfer, the data smooths the efficient transfer of process across the organizational membrane.

The tech transfer time tribulation

The development of the control strategy and its subsequent handoff to a manufacturing partner is rife with risk and delays. This is compounded when a sponsor chooses to engage with a new contract manufacturing partner. Challenges that can slow down the tech transfer include:

• Poor or insufficient documentation supporting the control strategy.
• Foundational differences in quality management approaches.
• Differing expectations of data sharing before, during and after the tech transfer.
• Differences in equipment and/or facilities.
• Scale-up and scale-out of the process.
• Differences in global regulatory compliance.

While several of these challenges are driven by people, a common element to many of them is data availability and process understanding.

The ROI

Organizations adopting a digitally mature approach to process design and parameter management gain numerous benefits. Firstly, they achieve significant scientific value through improved process understanding, leading to valuable process improvements and enhanced regulatory readiness.

Additionally, cost savings are realized through fewer experiments and deviations. This digital approach allows for quicker investigation and resolution of deviations, further contributing to cost efficiency. Moreover, the harmonization of processes across the network – from using common equipment and materials to standardized processes, training and even the names of parameters – results in substantial time savings. Thus, a digital strategy not only enhances control but also drives scientific, economic and operational efficiencies.

Attacking the digitalization challenge requires commitment from the business, involving leadership support along with budget, as well as IT contributions to ensure GxP and principles like ALCOA++ are maintained throughout. But the return on this investment can be significant.

The industry likes to tout numbers in the millions of dollars per day for delivering therapies to market faster. But let’s also remember the value of the therapy to an individual patient suffering from these disease states. This time saving is accomplished through a combination of improvements, such as the ability to reduce the number of required experiments in development; speeding up tech transfer and associated qualification runs, reducing deviations in manufacturing that slow product release and accelerating regulatory filings.

Additionally, there are time savings provided by the harmonization of processes across the network, from common equipment, materials, processes and training, down to the names of the parameters.

Effective digitalization for effective tech transfers

As industry keeps pace with constantly evolving regulations, the pressure remains on cutting both time and costs by eliminating waste in the development process, streamlining supply chains and expediting and simplifying tech transfers to deliver therapies to patients faster.

Adopting a digital approach across all related activities for enhanced control of the process can yield more effective tech transfers, improved process understanding throughout the lifecycle, enhanced process optimizations, and, most importantly, get therapies to patients more quickly.