OR WAIT 15 SECS
Robert Glaser is chief technology officer at Auvesy, Inc.
Software systems can help validate pharmaceutical manufacturing and steer the most appropriate design of processes in the right direction.
Editor’s Note: This article was published in Pharmaceutical Technology Europe’s October 2019 print issue.
Within pharmaceutical manufacturing, the goal of process validation is to build quality into the operation at every step to ensure consistent product quality. It seems straightforward, but how does one achieve this when it involves a myriad of multiple systems (programmable logic controllers [PLCs], human-machine interfaces [HMIs], control systems, etc.) and requires a team effort that encompasses people from various disciplines within the plant?
It is important to design processes that ensure product quality, safety, and efficacy. Today’s software systems can help validate pharmaceutical manufacturing and steer the most appropriate design of processes in the right way. For instance, these systems can deliver clear documentation, provide version control for the multitude of systems in use, and perform backup to ensure all risks are mitigated.
Software systems have become instrumental as validation expands from analytical methods used for the quality control of drug substances and drug products to computerized systems for clinical trials, labeling, or process control. Quality cannot be adequately ensured merely by in-process and finished product inspection or testing. Validation must happen at each step of a manufacturing process, which involves lots of team members from engineers to quality control to compliance officers, etc. to ensure that the finished product meets all quality attributes, including specifications.
Part of the validation process is in the documentation. There needs to be integrated support for documentation with 100% clarity and traceability. In addition to this, version control of all software systems used within the pharmaceutical manufacturing process is necessary to ensure compliance with current good manufacturing practices (CGMPs). These regulations state that manufacturing processes must be designed and controlled to ensure that in-process materials and the finished product comply with quality requirements that have already been predetermined and assure consistency and reliability (1). The regulations (21 Code of Federal Regulations [CFR] Parts 210 and 211) encompass requirements for approving or rejecting processes and specifications that impact drug quality.
In fact, 21 CFR 211.22 requires clear, written documentation for quality control and any anomalies or deviations must be justified and documented as well. To meet such compliance regulations, it is crucial to deploy software specializing in version control, backup, and documentation.
In highly regulated industries such as the pharmaceutical industry, the ‘four eyes principle’, otherwise known as the two-person rule, is a requirement that mandates two individuals must approve an action before it can be taken, and this requirement is also usually managed by the same software system that ensures process validation.
The four eyes principle gives another reason for deploying a validation software system that is able to integrate seamlessly with a full range of automation devices and equipment commonly in use by pharmaceutical manufacturers and not just tied to one single manufacturer. The version control system needs to be vendor-agnostic, which further safeguards that validation is properly addressed.
Validation software systems help manage and keep track of all types of activities that take place over the lifecycle of the product and process: whether a piece of equipment designed to manufacture something, a process/recipe to make something, or a computer program to control something.
With validation, it is instrumental to make sure the process is repeated exactly step by step all the way through and is achieving the end result within the allowed tolerances. For instance, temperature set points for producing certain drugs would have the processes validated and assigned validation thresholds. For the threshold, each individual step would be assigned a certain range, such as the temperature could fluctuate from five to six degrees or anywhere in-between, but if it crosses over to seven degrees, the process would be flagged and product considered out of specification.
The pharmaceutical industry has a need for validation, which is best done not manually but rather using a sophisticated system. For instance, such software systems will automatically store the new versions or copies as they’re made. When a new staff member logs in, the system will know who last notated the file; everything is time-stamped, allowing for a clear audit trail. It also ensures all the correct data are being captured. It can prompt the user to enter notes and/or comments, which are stored by the system and can be provided in real time. This way the user can easily reference why changes were made. Additionally, the system can let the user know if a contractor made changes that were saved to an Âexternal drive. The system can alert the person in charge that they need to get a copy of the drive. Currently, this process is often handled manually. In many instances, pharmaceutical companies will use external contractors and then realize thereafter the need for better controls to be put into place. Team leaders see the need for a smart version control and a backup system to safeguard them from any potential problems.
Such systems as described are typically used on a daily basis. They are especially helpful to developers who are making the code changes. With automated validation in place, many issues related to tracking changes go away. For example, when using the validation system, the user would simply log into the system and use the latest copy of the programming file from the central repository before beginning work. They wouldn’t worry about having the latest version and would have any notes or comments that may need to be addressed. Once done with the day of work, the user would simply save all the changes and essentially put the file back on the shelf and check it back into the system along with any notes or comments.
The validation system should act as an automated backup or scheduler. It should connect to the control network without human involvement. It should automatically reach out to the devices in the network to obtain the current copy that is running on a device and check it against the central repository to ensure its correct. Nothing should happen that is outside of the rules
In a validation system, there is also flexibility, because the user can choose how frequently they would schedule a backup. Most will do it on a daily or shift basis, so that any and all changes will be noted. For example, perhaps somebody walked up to a controller and put some code on it without letting anyone know. A validation system would pick up on that and flag it. The system will catch vulnerabilities through regular scanning intervals.
The system compares what is actually running on the device to what is assumed should be running there. If changes are noted that are not running, the system will conduct a comparison to determine if there are any issues. It takes into account just the code differences. For example, if there are 10,000 lines running on a controller, the system will simply highlight what is changed and will send that information to an email distribution list so that a manager or Âdecision-maker can decide what action to take.
This change-control methodology focuses on managing change to prevent inadvertent consequences. Certain manufacturing changes (i.e., changes that alter specifications of a drug or a critical device attribute or bioavailability) are mandated to have regulatory filings and prior regulatory approval. Change is an inherent part of the lifecycle of a pharmaceutical product. A change can be an addition to, deletion of, or modification to a manufacturing process, material, product, procedures, or equipment.
Validation is used in the process design stage through commercial production and can provide the Âscientific evidence that a process can consistently result in delivering a quality product. When manufacturing pharmaceutical drugs, there must an accurate, repeated degree of certainty in obtaining consistent results. A lot of time and money is spent on certifying one individual way of making a particular drug. Digital validation systems help to set a protocol that requires review, signoff, and signatures. The protocol is always checked and authenticated before the process can go into production. Software systems are, therefore, key in the solution to help pharmaceutical companies get to a validated state.
1. 21 Code of Federal Regulations Parts 210 and 211.
Pharmaceutical Technology Europe
Vol. 31, No. 10
When referring to this article, please cite it as R. Glaser, "Using Software in Process Validation," Pharmaceutical Technology Europe 31 (10) 2019.