The introduction of Process Analytical Technology (PAT) can be viewed differently throughout the pharma manufacturing chain. Bill Miles, of Lorien Engineering Solutions, offers a plant designer's view of incorporating PAT into an existing production facility
The majority of articles published on Process Analytical Techniques (PAT) and PAT technology have been generated by either the QA/QC communities or by product makers, and have been aimed at readers in those groups. This article has been written from a facility design standpoint and reflects on how these altered working practices might affect the design of the facility and its services infrastructure.
For pharmaceutical manufacturers, post-approval regulatory factors tend to inhibit the introduction of any change, whether it is a new technology or simply a prudent adjustment within the validated process, thus improvement and innovation in the pharmaceutical industry is currently limited.
Such processes are often finalised during clinical trials, and because of the way we presently adhere to the written word, these processes also become fixed and frozen. The Registration File (the licence) lists and defines process parameters and quality attributes. Most variations to that file require defined change control procedures, thus post approval changes currently involve a great deal of effort, such as registration activities or additional clinical trials.
To implement a process optimisation or change the feedstock suppliers is quite difficult because of the inflexibility within currently validated processes. We already see that variation in bio-feedstock sourcing is difficult to handle under GMP conditions because of regional and seasonal variations.
During the product lifecycle, process innovations to continuously improve product quality are currently associated with high costs to change the registration file with the authorities. Traditionally the relationship between the FDA and manufacturers has been tinged with authoritarianism and the associated fear linked with an ethos of telling what won't pass, but standing back from being part of the solution.
In more recent times, the FDA has shown a more proactive approach, which is crystallised in PAT Guidance. This altered paradigm signals a departure from "end of line testing" to "in flight testing" such that, at the successful end of a process run, the product can be "real time released".
The key to this mind shift seems tethered to a view that manufacturers at "hands-on level" are focused on executing the mantra as set out in the SOPs without any insight into why the SOP was thus worded. The FDA is minded that if greater understanding of the critical issues is achieved by everyone involved in the manufacturing phase (and why they are critical to the product efficacy), then the process will move from being task driven to being quality driven (where the word "quality" incorporates drug performance attributes). This is at the heart of PAT, together with another tool called Quality by Design (QbD).
QbD places emphasis on greater understanding of the upper and lower values of tolerances on the individual ingredients and on the possibilities of their tolerances drifting as the ingredients are mixed, worked (shear), and subjected to variations in temperature (compared with the stored temperature). The addition of purified water or solvents may also contribute to drift.
Another acknowledged variable is that which comes with the raw materials and APIs. The greater understanding of these physical properties and of the design space as a whole will enable a more flexible process control strategy either backward or forward feed loops.
It is important to understand that the FDA may not deem these physical drifts as changes - equally they may not deem an adjustment to mitigate the drift to be a change. This approach encourages continuous improvement.
The end-of-line testing as we know it drove the need for a space to store finished goods that are held pending QC approval real time release would see the product moved directly to the despatch warehouse.
Moving the QC control upstream into the live manufacturing process may well drive a different location for the QC lab and its headcount (space). The current QC output signal is PASS or FAIL, whereas the online tweaking of one process element delivers a report signal on the fine-tuning that gets fed back into an SOP revision. These altered outputs may well alter the benching/desk requirements within QC.
Another consequence of this new reactive manufacturing environment is the need to have space for staff who are more attuned to process control and in-flight adjustments. These people may also possess process equipment knowledge attributes and may be capable of feeding objective comment on the merits of using an alternative type of machine that does not stress the ingredients so much, such as low-shear versus high-shear mixers, or alternatives to microwave drying. PAT offers a better platform for closer working and shared dialogue between r&d, QA, QC, IT and process engineering.
So, design issues that are attendant to implementing a PAT approach into an existing facility are:
An altered headcount in QA/QC will drive changes to locker/change areas
Real time product release also releases space originally needed for pre-despatch product storage
The introduction of operatives of a process/automation/engineer calibre may drive the need for new space within production
Implementing any or all of the above will affect the current process layout and its established flow
Alterations to headcount and their disposition will impact on the design of cleanrooms
Alterations to headcount will also alter the latent and sensible heat gains to the areas concerned - this needs to be accounted for in the HVAC design.
All of the above will drive alterations to heating and cooling water, drainage, and process utilities, such as purified water/WFI.
When embracing the world of PAT, don't forget to include facility designers in the dialogue - remember that they were involved in the original design.