Why Data Analytics Is Essential Throughout a Product Development Lifecycle

In the biopharma industry, data analytics is an essential tool to accelerate product development, reduce manufacturing costs and manage quality control. Enlisting advanced data analytics at every stage of the product lifecycle helps companies develop processes to deliver high quality products – even minimizing impacts of environmental or raw material deviations – and directly impact the bottom line.

^{This article is posted on our Science Snippets Blog}

Current trends in bioprocessing development are heavily influenced by the push toward greater digitalization and process efficiency. This includes methods such as:

Quality by Design (QbD) – developing processes that use data models rather than relying on final product testing to meet quality standards
Continuous bioprocessing – intensification of both upstream and downstream operations that requires a greater level of control during production
High-throughput process development - using miniaturization, automation, and parallelization (digital twins) to create a more systematic workflow
In-silico experimentation – leveraging kinetic and computational models to guide process development, reduce experimental burdens, and enhance process understanding

In regulated industries such as biopharmaceuticals, data analytics is especially important, because developing and following well-documented processes is essential.¹ Advanced data analytics helps biopharma companies develop robust processes that follow regulatory-approved methods and keep a product’s critical quality attributes within the acceptable control limits.

With advanced data analytics, including multivariate data analysis and real-time process monitoring, biopharma companies can take process control to the next level and actually start predicting where a process is heading based on statistical models. This can help with planning downstream delivery and shorten the process steps (leading to faster market delivery).

As a company’s use of data analytics becomes more advanced, they are able to move from understanding, or describing what happened with a process in the past, to gaining information about whether a process is performing optimally at the moment, to predicting where the quality (and other) parameters (like volume) will be in the future.

^{Advanced data analytics is part of a continuum that helps you understand your processes and move from understanding what happened in the past (hindsight) to why something happened (insight) to predicting what will happen (foresight) to achieving a level of prescriptive analytics in which you understand how to make something happen.}

Data Analytics Creates More Robust Processes

Advanced analytics can help pharmaceutical businesses reduce the costs and speed up production timelines while staying within the defined critical quality attributes² Data analytics is an essential tool to help with:

Process development
Quality by Design (QbD)
Process optimization
Process control including PAT
Batch management
Continuous process manufacturing

According to McKinsey, applying data analytics in the pharma industry could improve earnings before interest, taxes, depreciation, and amortization (EBITDA) as much as 45-75%.³

QbD uses DOE to create a robust process that can be well documented. PAT relies on MVDA to evaluate data from sensors (important for automation).

Process Development

Creating an optimal manufacturing process that minimizes variation, utilizes resources optimally, and that can tolerate deviations without compromising quality, relies heavily on Design of Experiments (DOE).

DOE is a statistical method of testing variables to understand the affect they have on your process and product quality. DOE helps scientists reduce the number of experiments needed to create robust processes. This can, in turn, reduce the time to market for products.

DOE helps to:

Minimize the number of experiments you have to run to find the ideal recipe and control strategy
Create a robust process (one that holds up to changes in environment, process hiccups, etc.)

Quality by Design (QBD)

Quality by Design (QBD) is an approach to process development that follows a robustness testing framework designed to meet stringent regulatory standards. It relies heavily on a DOE approach and can be used to predict the critical quality attributes at end of shelf life using a few months of historical data. An example of how the QbD framework can be used in biopharmaceutical development is in final formulation development. All biopharmaceutical products are formulated to a specific dosage to be effectively delivered to patients. A good formulation must be stable – during both manufacturing and for its product shelf life.

For a company to successfully create a robust formula, it’s necessary to take steps to know which factors in the formula composition are critical to quality and what the acceptable ranges of variation are. In other words, to create a product design space that is as robust or broad as possible while still ensuring the product produces the desired (and specified results) under the stated shelf-life conditions. Using the right tools can make this process more accurate, as well as faster and less resource intensive.

The goal of QbD in pharmaceutical development, including formulation and manufacturing process development, is to consistently deliver the intended quality of the product within allowable ranges.

Process Optimization

Data analytics in biopharma isn’t only used for monitoring upstream processes like a bioreactor. Every stage of drug manufacturing can benefit from using advanced data analytics. From cell culture harvest to syringe filling lines, data analytics is essential to ensure consistency and that critical quality attributes are met.

Multivariate Data Analysis (MVDA) helps you look for trends in your data and optimize processes. MVDA is a statistical technique used to analyze data that originates from more than one source or has multiple variables. It’s a more robust way of analyzing data than univariate data analysis, which evaluates at one variable at time.

Data analytics helps ensure process changes remain scalable, documented and compliant.

Process Control and PAT

Process Analytical Technology (PAT) enables biopharma companies to collect real-time data for advanced process control strategies, effective automation and process optimization.

In practice, executing a successful PAT strategy requires careful and ongoing monitoring and analysis of data coming from multiple sources: from raw material suppliers to production. Process control requires coordination in how data is collected, shared, and analyzed starting with the raw material.

PAT and other types of advanced models for process control rely on the use of robust sensors and are pivotal to support the industry trend towards continuous manufacturing. The use of PAT become more and more important in the pharma industry. Some of the top reasons include:

PAT and other quality programs are expected by the FDA and other regulatory agencies⁵
The increasing push towards continuous process manufacturing
PAT is accepted as providing for more robust bioprocessing and overall being a cost-effective product, and minimizing quality defects
Continued improvement in sensors, probes, software and analytical equipment

Batch Process Management

Monitoring batch processes requires creating a picture (or model) of what an ideal process looks like. That involves looking at past data from both the process inputs and the process outputs and merging them into single model. The goal is to be able to quickly identify when something goes wrong with the process and rapidly correct the error before the batch is compromised.

For batch process analytics, , two perspectives need to be merged: data over time and data that shows quality or yield parameters of finished batches. These two models are called Batch Evolution Model (BEM) and Batch Level Model (BLM).

For many companies, the general process flow for batch production is limited to measuring the quality of each batch after its completed. Companies are working with fixed batch recipes with few sensors. They have limited real-time awareness of the batch performance and very little room for error.

With data analytics, including multivariate data analysis and real-time process monitoring, companies can take processes to the next level and actually start predicting what the critical quality attributes of a batch on the current production trajectory will be. Data analytics can help identify how raw material quality can impact the final product, and even identify genealogy of individual batches.

Continuous Process Manufacturing

Continuous manufacturing is one of the key trends within the pharmaceutical industry, both for the production of ‘classical’ drugs as well as large molecules. Developing production processes based on statistical models allows companies to shift from traditional batch processing to continuous process manufacturing.
The FDA considers multivariate models to be surrogates for traditional release tests, for example, a stand-in for dissolution.5 This means:

Shorter cycle time
Reduced inventory
Reduction in end-product testing
Reduction in manufacturing cost

Real-time data analytics makes continuous process manufacturing possible. This combination of real-time process monitoring and data analysis provides insights about processes in real-time. When connected digitally with a continuous feedback loop, it can be used to control processes and predict future production quality.

Real-time process control is part of a robust system of digital transformation that uses QbD, DOE, MVDA and real-time data analytics to optimize workflows and help automate production.

DOE helps you generate the right data or reduce the amount of data you need to understand cause and effect in a process.
MVDA helps you find trends in your data over time.
Real-time data analytics helps you keep track of what’s currently happening in your processes.

Data Analytics Grows the Bottom Line

Applying data analytics across the entire product lifecycle can help deliver value to the bottom line. A strong business case can be made for wide adoption of data analytics to maximize yields, obtain consistent high quality (by reducing errors and eliminating batch failures) and diminishing loss from processes gone astray. By coordinating strategic alliances and aligning business objectives across the organization, companies can maximize the value for internal stakeholders.

^{References:
1. Britannica, “Drug discovery and development,” Drug Development Process, Accessed Feb. 18, 2022.
2. DiMasi, Grabowski, Hansen, “Innovation in the pharmaceutical industry: New estimates of R&D costs,” Journal of Health Economics, vol. 47, 2016, p 20-33.
3. McKinley, “How pharma can accelerate business impact from advanced analytics,” Jan. 2018.
4. C. Wurth et al., “Quality by Design Approaches to Formulation Robustness- An Antibody Case Study,”Journal of Pharmaceutical Sciences (2016) 1e9
5. Quality Considerations for Continuous Manufacturing , Guidance for Industry , DRAFT GUIDANCE , U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), February 2019 https://www.fda.gov/media/121314/download}