Ensuring Quality: Key Steps in Medical Device Validation

By Alessio Rindi, MD Compliance Manager @ PQE Group

The following article is intended to define some primary concepts in the field of process validation that, in recent years, have become increasingly essential for medical device industries in order to prove the effectiveness and safety of their products being reviewed by regulatory agencies.

We start by providing a definition of the process validation and proceed to describe the regulatory and formal aspects that offer official evidence of its importance. Afterwards we focus the attention on when to perform a process validation, explaining the characteristics that a production process must have to become eligible for it.

At the end the classical approach is illustrated, describing the three phases of a typical process validation: Installation Qualification (IQ), Operational Qualification (OQ) and Performance Qualification (PQ).

In addition, stricter documentary aspects involving the Validation Plan (VP) and Validation Report (VR) are addressed. For each phase, a guide is provided for the essential document content. 

Ensuring Quality_Site Banner_1

What is Process Validation and Why it is Important 

 

Definition 

Process validation is a fundamental part of an approved Quality Management System (QMS) and its goal is to ensure that a specific manufacturing process works correctly in producing a product that fulfils its predefined specifications. As a result, it provides evidence that the production process is under control and plays a crucial role in mitigating the risks for the patient connected to the manufacturing quality of the product. 

 

Regulatory and formal aspects 

The importance of this practice can be also surmised by the fact that it is also referred to by a regulatory framework that includes FDA 21 CFR part 820, Regulation (EU) 2017/745 and ISO 13485:2016.  

Despite the concept of process validation being reviewed at the regulatory level, there is not a recognized standard for its implementation that illustrates its requirements. On the other hand, there is a non-binding guidance, the GHTF/SG3/N99-10:2004 (Edition 2), which is widely accepted by regulatory agencies and is used as a guideline by industry for the definition of the main requirements and for the planning of an effective process validation. 

Some exceptions exist where specific process validation requirements are set in dedicated standards. Examples are sterilization process and sterile barrier system manufacturing. The first is addressed by several standards, depending on the sterilization method, for example: 

  • ISO 20857 for dry heat sterilization; 

  • ISO 11135 for ethylene oxide sterilization; 

  • ISO 11137 for radiation sterilization. 

The process validation of the sterile barrier system is instead addressed by ISO 11607-2.  

As part of a QMS, process validation is not only about the collection and analysis of data from the inspected process but it also has connections to the document management, design and development process, change control management, risk management process, supplier qualification, in-process control, and quality assurance. All of these procedures must be in place before starting a process validation. 

 

When to perform a process validation 

Indications about when to perform a process validation are given in 21 CFR 820.75 and similarly in ISO 13485:2016 section 7.5.6. The statements are similar and basically ask for a process validation when the output of a production process cannot be completely controlled by subsequent monitoring and tests. Usually, the control of a process, only by means of in-process controls, results in a highly time consuming and expensive activity; however, in other cases it is also intrinsically impossible to check all samples, as in the case of destructive tests. In this case, the only option is to perform a process validation.  

Other cases that require a process validation can be: 

  • When the types of tests that can be implemented for in-process controls are not sufficient to check all the aspects of the device safety; 

  • When the tests used for in-process controls do not have enough sensitivity. 

Because all these cases do not provide enough risk mitigation, they must be preceded by a more in-depth challenge of the process that can be feasible only within the frame of a validation activity. 

In this case, the in-process controls will be used to check that a validated process is able to maintain its stability over time.  

Of course a process validation is also needed before a new piece of equipment or machine becomes operational or when an abnormal trend is identified in the specifications of a product produced by an existing process. 

Typical processes for which a process validation is required, with respect to medical devices, are: 

  • Sterilization processes; 

  • Sterile barrier systems; 

  • Injection moulding / extrusion / automatic assembly; 

  • Clean room ambient conditions; 

  • Aseptic filling processes. 

Process validation consist of 3 phases: 

  • Installation Qualification (IQ); 

  • Operational Qualification (OQ); 

  • Performance Qualification (PQ).  

All these phases are required and must be performed in this order. Only in the event of change can the possibility to repeat only some parts of the validation be evaluated. 

 

Process Validation Approach 

 

Validation Plan

 A process validation typically starts with the release of a so-called Validation Plan, through which all the preparatory information that is needed to precisely identify the process to be validated and define the validation strategy are reported. Examples of the content of a validation plan can be: 
  • Identification of the process to be validated and why; 

  • Identification of the equipment; 

  • Identification of the product to be manufactured; 

  • Identification of the validation team; 

  • Identification of the main references (internal procedures, regulatory requirements, standards); 

  • Identification of the risk management documents needed to plan the validation strategy (e.g., severity of the product, test to be performed for risk mitigation); 

  • Definition of the statistical rationale for the sampling plan for both OQ and PQ. Usually the sampling plan is based on a risk approach, linking the severity level, identified in a dedicated risk analysis, to defined statistical parameters; 

  • Description of the validation phases to be implemented (IQ, OQ and PQ); 

  • When a revalidation is needed. 

 

Installation Qualification (IQ)

 The Installation Qualification is the first phase of the process validation and gives us the evidence that the manufacturing equipment or machines have been installed correctly as per supplier specifications. Usually, the main features of an IQ document are the following: 
  • Description of the equipment; 

  • Identification of the personnel involved in the tests; 

  • Personnel training; 

  • Identification of the instruments used (e.g. multimeter, manometer); 

  • Instrument calibration; 

  • Equipment verification.   

  • The following list includes some typical tests, without the presumption to be all-inclusive: 

    • Calibration of sensors, if present; 

    • Utilities verification (e.g., compressed air pressure, electric current voltage/frequency); 

    • Environment of the installation (e.g., clean room, temperature); 

    • Supplier documentation (e.g., user manuals, spare parts list, drawings); 

    • Safety features (e.g., warning signs, grids, emergency stop, alarms); 

    • Functionality of the equipment;  

    • Maintenance schedule and extraordinary maintenance history; 

    • Software validation (when required).

  • Conclusions 

This phase must be initially planned in the Installation Qualification Plan (IQP) and then executed as a report (IQR). IQP issuing can be done at the same time for the Validation Plan. 

 

Operational Qualification (OQ) 

Operational Qualification is the second phase of the process validation and its activities are initiated only after the IQ has been completed. It is defined by GHTF/SG3/N99-10:2004 (Edition 2) as “establishing by objective evidence process control limits and action levels which result in a product that meets all predetermined requirements.” More simply, the OQ is the phase where we challenge our process. At first, we must identify the critical process parameters (or the combination of the process parameters) that have higher impacts on the process, and then identify a window in the critical process parameters, within which it is possible to produce a product that meets its predetermined requirements. This is done by identifying an upper and lower limit per each parameter, the lower being the worst cases (WCs), from which samples shall be taken and then tested.  

Usually during this phase, a Design of Experiment (DOE) technique can be used to gain more insight about the definition of the critical process parameters and process windows. 

A typical format of the OQ document includes: 

  • Identification of the machine; 

  • Identification of the product to be produced; 

  • Personnel performing the activities (e.g., WC definition, samples collection, laboratory tests); 

  • References (internal procedures, regulations, standards); 

  • WCs definition - this section should not include just the list of the parameters to be challenged but also a rationale stating why some parameters are considered to have more impact on the product than others. This can include the reference to preventive DOE activities; 

  • Evidence that the previous validation phase, IQ, has been successfully concluded before starting the OQ activity; 

  • Identification of the test to be performed. Usually input comes from risk analysis and regulatory requirements; 

  • Pre-requirements for tests: 

    • Training records 

    • Presence of test method validation (TMV) for the tests to be carried out 

    • Clear and quantifiable acceptance criteria 

  • Number of samples per each test and reference to the statistical rational; 

  • How to perform data analysis; 

  • Test results; 

  • Raw materials lots; 

  • Conclusions. 

This phase must be initially planned in the Operational Qualification Plan (OQP) and then executed as a report (OQR). OQP issuing can be done at the same time as for VP. Raw data and data analysis must be attached to the OQR whenever possible. 

 

Performance Qualification (PQ) 

The performance qualification is the last stage of the process validation. According to the GHTF/SG3/N99-10:2004 (Edition 2) it is defined as “establishing by objective evidence that the process, under anticipated conditions, consistently produces a product which meets all predetermined requirements.” In other words, during this phase, the process must be challenged at its nominal condition, that is the condition that will be used for mass production. This condition shall be included in the parameters windows previously defined during the OQ phase and will represent the optimal setting of the process. The PQ is definitely a simulation of the real production process; therefore, repeatability and reproducibility must be assessed. To do this, three production runs, at a minimum, are usually performed. Repeatability is assessed within the runs as is reproducibility between runs. Each lot shall be identified by a unique lot number. 

A typical format of the PQ documents usually includes: 

  • Identification of the machine; 

  • Identification of the product to be produced; 

  • Personnel performing the activities (e.g., samples collection, laboratory tests); 

  • Number of runs to be performed. Samples shall be collected in the same way from each run; 

  • Evidence that the previous validation phase, OQ, has been successfully concluded before starting the PQ activity; 

  • Identification of the test to be performed. Usually input comes from risk analysis and regulatory requirements; 

  • Pre-requirements for tests: 

    • Training records 

    • Presence of test method validation (TMV) for the tests to be carried out 

    • Clear and quantifiable acceptance criteria 

  • Number and identification of samples per each test and reference to the statistical rationale; 

  • How to perform data analysis; 

  • Test results; 

  • Raw materials lots; 

  • Conclusions. 

This phase must be initially planned in the Performance Qualification Plan (PQP) and then executed as a report (PQR). PQP issuing can be done at the same time as for VP. Raw data and data analysis must be attached to the PQR whenever possible. 

 

Validation Report (VR) 

Usually at the end of the validation activity a final report is issued. It should contain the reference to all the protocols and reports. It summarize all the findings of the previous stages, deviations, and rational in order to derive conclusions on the process and assess if it is under control and can be considered validated.

This report must be reviewed and approved by the team and management.

More on the Regulatory scope for Medical Devices?

One of PQE Group's unique features, is its global presence, with a dedicated local touch. We'll support you at 'home', providing your business with a rich experience from all over the world. 

Learn more about our compliance support for Medical Device