Process Validation Failure Investigation and Root Cause

Process validation failure takes place when a process does not satisfy the requirements established beforehand for design validation and does not show stable process performance or sufficient supporting evidence for validation. The investigation of the process validation failure is the record-keeping and scientifically rational process of understanding what happened, why it happened, if the quality of product has been affected and what steps should be taken next to regain control of the process.

A failed validation batch should not be considered a separate testing problem without any regard to the context of a manufacturing process. It is important to access the whole complex of process parameters, components, apparatus, personnel, analysis results and sampling methods applied for carrying out the original validation experiment.

The goal is not to achieve a satisfactory success by means of numerous testing. Why do we need to know the root cause, what consequences are associated with it and whether the process can produce perfect products all the time?

First Response: Contain the Problem and Evaluate the Risk

If a validation run fails, the relevant department must immediately record the incident under the site's deviation management procedures. Management, Quality Assurance, Production, Quality Control, Engineering and Validation departments must be involved according to the nature of failure.

The initial assessment must determine:

• The batch, process stage, equipment and acceptance criteria concerned.
• If the failure affects a critical process parameter, critical quality attribute, analytical outcome, or procedural requirement.
• If any materials which could have been affected had been manufactured, transferred, packed, or released.
• If any constraints on the process should be applied.
• If similar events have taken place during the past validations or during the production process.

The disposition of products must be determined by means of a documented evaluation of quality and in accordance with relevant procedures. A failed validation test does not mean that all batches have to be considered unacceptable, although it does not allow for the automatic approval of the product just because further tests are satisfactory.

In case the granulation endpoint has not been achieved in the validated operating range, the concerned team should evaluate the material in question, the subsequent processing steps, the quality attributes and the impact on previous batches. All decisions should be made based on the pharmaceutical quality system, not simply Production.

Investigate the Evidence Before Choosing a Root Cause

An effective investigation is based on facts. Do not blame human error, equipment malfunction, or variation in raw materials before evidence has been reviewed.

1. Review process and batch records

Check the records for the executed batch, validation protocol, equipment log and the environmental records if it is necessary. Compare observed conditions with accepted parameters of the process and acceptance criteria.

Check for unexplained correction of course, interruptions, alarms, long delays, deviations from the sampling plan, missing or inconsistent information. Be sure that the process has been carried out according to the agreed and authorized instructions.

2. Examine equipment and utility performance

Engineering must ascertain the status of the equipment, including all calibration logs available, preventative maintenance, alarms, control-system logs and relevant utility information. A control panel measurement may not give true process performance unless one can confirm the performance of the appropriate sensor, the control loop, or the data acquisition system.

Inconsistent drying results, for instance, could be due to air flow characteristics, sensor failures, loading configuration, or other differences in the properties of wet granules. The investigation should demonstrate what possibilities are backed by evidence.

3. Evaluate materials, employees and analytical data

Investigate the raw material batch, material characteristics, dispensing records, provider changes and storage conditions of desired materials. Do not assume that human intervention automatically causes human error; rather check the training of the operator and its execution.

Quality control should analyze the original analytical data, calculations, suitability of the instrument and the conditions of sample preparation. If any results arise concern, retesting should be done according to established procedures and accompanied by scientific justification. The original result must be part of the investigation.

Root Cause Analysis of Process Validation Failure

In the context of process validation failure, the aim of root cause analysis is not just to describe what happened but also to find out why it happened. To do that, the analysis can be structured in terms of processes, technologies, materials, techniques, measurements and efforts. Various tools can assist in achieving this goal, such as the 5 Whys method, fishbone diagram and FMEA. The analysis technique should be selected based on the complexity of situations.

Take a hypothetical case of the validation run of tablet compression where the variation in tablet weight is beyond the approved acceptance standard. At this point, the investigator is only able to observe the fact of failure, not the root cause.

The study may look at the functioning of the feeder, the properties of granules flow and the velocity of the turret. Compression settings, powder separation, the sampling method and the efficiency of the balance also may be examined. In case the investigation proves that inconsistency in granule flow occurred as a result of wrong granulation endpoint control, the study will have to find out the reasons of failed endpoint control and the causes of poor process design or weak monitoring strategy.

It is not enough to conclude that "operator error" has occurred unless one can specify what was the action or inaction that took place, what proof supports it and why the existing procedures, training, measures, or means failed to avoid the incident.

Root cause should be called confirmed, probable, or undetermined only where the site procedure allows using such terms and what available evidence supports it. The situation with undetermined root cause should not be blamed on the unproven explanation just to end the investigation.

Determine the Impact on Product Quality and Validation Status

The investigation must ascertain whether the failure is confined to validation or represents an inherent weakness in process control.

The evaluation needs to take into account the batch in question, other batches that have undergone validation, prior batches that were produced commercially and possibly related products or equipment, or anything else that used the same process or method, at the same time.

The analysis should cover trends, variations, complaints, stability data, etc., however, this information should have a positive influence on the evaluation process.

Only passing the finished product quality tests does not indicate that the process was validated. The purpose of validation is having the evidence of consistent performance of the process.

If the failure is connected to a critical parameter, the team must assess the parameter and confirm whether the validation outcomes are valid and whether any adjustments or changing in validating will be needed, if any requalification, further studies, or changing methods have to take place.

CAPA: Correct the Cause, Not Just the Result

The premise of corrective and preventative actions (CAPA) is based on findings that stem from investigations. Corrective action resolves the identified issue while preventive action or additional systemic enhancement is aimed at addressing either the circumstances that led to occurrence of the issue or the initial conditions that could lead to the reoccurrence of the incident.

Depending on the underlying cause, the actions to be taken may include:

• Repairing equipment and checking its performance.
• Adjusting conditions or control methods through change control process.
• Enhancing controls of the raw materials or overseeing suppliers whose raw materials are used.
• Changing the testing procedures or methods.
• Changing the standard operating procedures (SOP) and training the staff.

Just repeating a failed validation exercise without fixing the problem is insufficient for a corrective action and preventative action (CAPA). If the protocol itself has an inappropriate acceptance criterion or the sampling technique is insufficient, this needs to be looked into and any changes need to be appropriately justified before continuing with the process.

It is necessary to verify if the CAPA worked with pre-defined and meaningful evidence. Proof may be provided, depending on circumstances, by a satisfactory outcome in subsequent runs, improved indicators in the course of the process, successful challenge testing or by the problem not happening again during a suitable monitoring period.

When Can Validation Be Considered Successful?

It is not sufficient to declare validation successful solely based on initiation of corrective action. The success of validation must be based on the original protocol, validation records, deviations, investigation results, current status of CAPA and other evidence that is required to assure an acceptable state of the process.

If there is a need for performing repeated or additional studies, the scope of those studies, as well as acceptance criteria, must be scientifically justified and approved by the quality system. The data of unsuccessful studies must remain traceable and should not be eliminated from the record of validation history.

QA has to analyze whether the investigation substantiates the proposed conclusion and whether the final validation conclusion reflects the evidence. The inspector may be interested in such issues as the reasons of original testing failures, the way in which the cause of failure was determined and whether the procedure of investigation of relevant materials was implemented.

The failure to validate a process is an occasion for detecting shortcomings in mastering and controlling the process, but the investigation can yield meaningful insights only if it is performed beyond the manifestation of the failure. A satisfactory investigation is one that correlates findings of the investigation with the underlying cause of the failure, assesses impact on the resulting products and prescribes appropriate corrective actions. The assurance of process control should not be based only on the success of subsequently produced batches; rather, any available evidence should substantiate the notion of assurance in process control.

Get ready to use editable Validation Protocols in MS-Word FormatView List

Share

Tweet

Share

Pin it

Share