How CDMOs Can Implement a Modular PAT Strategy
Process analytical technology (PAT) is playing an increasingly important role in the pharmaceutical and biopharmaceutical industries due to regulatory pressure and the push to reduce costs while managing quality. For CDMOs, any investments into PAT technology must be versatile enough to accommodate a variety of products and processes.
That’s why a modular approach to PAT – including robust data analytics tools – can help CDMOs meet regulatory requirements while managing product quality and reducing costs.
This article is posted on our Science Snippets Blog.
A versatile and modular approach to PAT implementation can help CDMOs meet regulatory requirements while improving product quality and reducing development and production costs.
PAT Supports Continuous Manufacturing at CDMOs
PAT and other types of advanced models for process control are pivotal to support the industry trend towards continuous manufacturing. In a 2020 survey by BioPlan Associates [1], CDMOS rated real-time and inline testing, continuous bioprocessing, automation for bioprocess control and monitoring, and PAT/DOE tools as the top new product development areas needed.
Source: 17th Annual Report and Survey of Biopharmaceutical Manufacturing Capacity and Production, BioPlan Associates, Inc., April 2020.
Yet, CDMOs seem to be facing some perceived hurdles when it comes to implementing the technologies that they know are important to ensure future competitiveness. Some of the top hurdles CDMOS reported in a recent survey were time required to implement, insufficient expertise, cost, and uncertainty how about how regulators will deal with PAT information.
So, what can CDMOs do to overcome some of these PAT implementation hurtles?
By implementing PAT in a modular fashion, CDMOs can actually help reduce the time required for implementation. Modular PAT also helps to overcome hurdles around insufficient expertise and cost.
In addition, many of the regulatory concerns expressed by survey participants could actually be an argument in favor of PAT, as the sort of documentation, data management and audit trails that PAT provides is what regulatory bodies are increasingly looking for to show compliance.
These topics and more are addressed in a related webinar.
Reasons PAT Adoption Is Increasing
In spite of the hurdles, there are a number of reasons that PAT adoption is increasing in the biopharma and pharma industries, including at CDMOs. Some of the top reasons include:
- PAT and other quality programs are expected by the FDA and other regulatory agencies
- Increasing push towards continuous process manufacturing
- PAT is accepted as providing for more robust bioprocessing and overall being cost-effective
- Strong business cases can be made for wide adoption to maximize yields, obtain consistent high-quality product, and minimize quality defects
- Continued improvement in sensors, probes, software, and analytical equipment
As PAT becomes more commonly accepted in the industry, an increasing number of strong business cases have been made for wide adoption. These includes its usefulness in reducing errors, eliminating batch failures, and diminishing loss from processes gone astray. This is especially important for CDMOs, as the decision to adopt PAT is typically based on ability to prove ROI.
The continued improvement and reduced costs of sensor probes over the last five years or so is another reason for increasing levels of PAT adoption. The availability of analytical software associated with these technologies has also become more widespread, both for upstream and downstream bioprocessing, and covers a number of different kinds of therapeutics, as well.
Trends Affecting PAT Implementation
The most obvious thing we can say about the growing need for PAT is that customer needs are evolving. Processes are becoming more complex and automation is deemed as essential. The past use of performance-based metrics around reliability, scalability or specific features, especially in single-use manufacturing platforms, are giving way to more complex processes and automation. The future means more connectivity, remote access and automation.
We can summarize three trends affecting the PAT automation and data analytics market:
1. Intensified processes. If we look at changes over time in the market, one thing that stands out is the need for speed: speed to market, speed to clinic and speed between the different development stages. But the need for quality doesn’t go away. We are, after all, producing substances that are going to be used by patients, and so a high quality is required. That means defining, in very specific terms, what high quality means at each step in the process as you hand over the bioproduct from upstream to downstream. At the same time, we are trying to reduce costs, both development costs and implementation costs. So, what this all leads to is a fundamental trend toward intensification.
2. Data-drive process development. Deployment of new business models will require secure access (and measurement) of business relevant data. This supports the full implementation of QbD and PAT with monitoring control based on Critical Quality Attributes (CQAs). These are methods that rely on accurate data from every stage of the process and use data to inform the decisions.
3. End-to-end connectivity. Complex new processes require higher degrees of sensing and control, namely for CQAs and modular automation that connects seamlessly to other platforms. We are seeing more of a call for end-to-end connectivity of the whole process from equipment analyses to software using standard interfaces. Looking for tools and software that enable you to obtain a modular, plug-and-play approach could be seen as best practice, then.
Selecting the Right PAT Strategy at your CDMO
For CDMOs, the decision to adopt a PAT strategy is one that can improve not only quality, and time to market, and but also profitability. However, this can only be achieved if the right type of PAT strategy is implemented. The important considerations for implementing PAT at your CDMO are:
- Versatility. CDMOs are not product or application specific, so your PAT strategy and tools must be versatile.
- Modularity. Having plug-and-play functionality will help you maximize the use of the technology across multiple programs and reduce operating costs.
- Scalability. In order for the PAT to be rolled out and expanded within the company at a fast pace, in both process development and manufacturing, you need a tool and process that is scalable.
Modular PAT Methods Exist
The vision is already the reality when it comes to modular PAT technologies.
Modular Upstream PAT Platform
Here’s an example of what a modular PAT approach looks like with the Sartorius BioPAT® portfolio.
BioPAT® MFCS , BioPAT® Trace, BioPAT®ViaMass are being used to control and automate feed and bleed operations in a single-use bioreactor. The Sartorius BioPAT® portfolio can be scaled from micro-bioreactor systems like Ambr® all the way to 2000L production bioreactors, – making it an excellent candidate for CDMOs to consider when implementing their PAT strategy.
Watch a demo and example of how this cell bleed controller system works in this webinar:
Incorporating Spectroscopy PAT Data
Another key area in PAT is how to capitalize on spectroscopic sensor data (i.e Raman, NIR, IR, UV). Such sensors can be interfaced to almost any type of manufacturing process, and by rapid, precise, and nondestructive analysis, it is possible to predict analyte concentration, quantify physical parameters, or fingerprint process streams and raw materials.
The Bioprocess industry has dabbled with Raman for more than a decade – however, it has largely remained on the fringe, mainly adopted by only the largest biopharma manufacturers and other early adopters. A few of the key challenges have been: lack of true SU sensor integration, customer bandwidth challenges in PD with bench scale adapted systems, and significant difficulty in transferring a model from PD to manufacturing
In order to address these challenges, Sartorius developed the BioPAT® Spectro Integrated Spectroscopy Platform, which enables QbD in Ambr® and BIOTSTAT® STR and using SIMCA-Q®.
BioPAT® Spectro was designed to meet three key requirements:
- Enable Raman spectroscopy in high throughput process development
- Facilitate the model building process with seamless transferability from PD to commercial scales
- Full single-use integration and scalability for commercial manufacturing
Plug-and-play Raman port: For example, you can connect either the Tornado Raman Spectrometer or Kaiser Raman Spectrometer through the same single-use port.
Read more about the BioPAT® Spectro Integrated Spectroscopy Platform.
A Few Examples of Modular PAT Implementation
Some data supporting the use of predictive models for Raman spectroscopy include a published report from GSK [3] with models built in Ambr® 15 for glucose, lactate, protein and glutamate predictions. The study shows that across different Ambr® runs, which are represented by different colors in the graphic below, the predictive Raman models are directly comparable to measured concentration levels.
Data source: GSK Stevenage, performed with a prototype spectroscopy integration and a Tornado Raman spectrometer.
The next question is, how scalable can these models be?
Below you see an example from model building exercise that was done in Ambr® 250 and FlexSafe STR® 200L. It shows that the model used at 250mL can be directly transferred to the 200L scale with the same level of prediction confidence.
These examples demonstrate how Raman spectroscopy has grown from a novelty to a proven global PAT strategy for bioprocessing.
Want to know more?
Watch a demo of how PAT works in SIMCA® Multivariate Data Analysis software and see examples of it in action.
Best Practices to Consider for PAT Rollout
When you start to consider how to roll out PAT within you CDMO, there are some important steps to follow.
1. Manage PAT as a strategic initiative. This means that any sort of PAT rollout should have clear sponsorship and it should also have product development and manufacturing involvement. The best outcome for PAT is when a process developed for PD can be scaled to manufacturing without additional testing and resources.
2. Do quantified justification. This is where you create your business case. What are you expecting to achieve? What are your goals? Do you want to maintain or increase quality? Are you looking for a faster product release? Would you like to reduce cycle time? Do you need to reduce labor costs?
3. Do system architecture planning. There needs to be a certain level of system integration of not only the physical hardware itself, but also the software component. You need to consider how the data feeding in or out of the hardware and into the software will be managed, and also the level of process control you need or want to implement. Your architecture planning should be scaled out from development all the way to manufacturing.
4. Have cross-organizational involvement. Involving everyone in the organization from product development to manufacturing to quality control to regulatory is important. Make sure you completely outline the requirements and contingencies at every step of the way. Make note of specific uses for data or steps that might not be obvious.
5. Deploy the right skill sets. Implementing PAT, whether it's the hardware or the software, will require a new set of skills. This will be an additional workload and may call for additional skills that the staff you have on board might not have. Don't be afraid to rely on contractors. Contractors can implement system architecture planning. They can help with model building. Sartorius, for example, has a skilled and experienced team of data scientists who do work on a contractual basis as well.
6. Develop a program plan. You need to have a program plan that outlines all the steps, timelines, contingencies and people involved. Project and program management are important. You might consider taking a phased approach to implementing the architecture itself and to have a resource plan behind that, and also to make sure that the skill sets match the program plan as well.
Try it yourself
References:
Want to know more?
Watch a demo of how PAT works in SIMCA Multivariate Data Analysis software and see examples of it in action.