An analytical result does not necessarily have to exceed the specification limit in order to warrant concern regarding product quality. In the pharma industry, a result may be comfortably within the specification limits and yet evidence the occurrence of a change.
Such a situation is often referred to as being out of trend (OOT).
OOT scenarios are particularly relevant in stability studies, analytical testing, environmental monitoring, and process monitoring. A single analytical result may seem inconsequential on its own, but when it is viewed in relation to historical data, it may show a trend pointing to an impending problem.
For that reason, any good quality system in pharma will not only look at whether an analytical result passes or fails. It will also determine whether the result has behaved as expected.
Take an example of a drug that has stability assay specification ranging from 95.0% to 105.0%. Suppose the stability results are as follows: 99.6%, 99.3%, 99.1%, 98.8%, 96.9% recorded at the different points of time.
Although the latest result is within specification, this does not mean it is automatically classified as OOS; nevertheless, a significant decline in the value as compared to previous records requires explanation.
In particular, it does not mean that 96.9% is not acceptable; however, the question is why the value changed so much.
This is the main goal of Out of Trend evaluation.
For instance, if a given product has been historically producing results of assays in the range of 98.5-100.5%, an unusual shift to the values of 96.5-97.0% may still meet the specifications but indicate that process has started to drift.
Should the trend continue, the product will eventually fail.
This is how an appropriate OOT system works by providing an early warning. Such system will allow quality unit and manufacturing teams to identify possible issues on early stages.
This methodology is of great importance for drugs with long stability studies, since delay in response due to OOS results might cause loss of very important information.
An OOS result means an unacceptable outcome that does not conform to an established specification, or not meeting acceptance criteria.
An OOT result is a result that is specified, but represents an unexpected trend or variation in comparison with other results that were obtained previously.
For example, if a laboratory has an assay specification of 95.0% to 105.0% and the result obtained is 94.6% then the result will be regarded as OOS.
On the other hand, if the obtained result equals to 97.0% but historically the other results were close to 99.5%–100.5%, the result may serve as a potential OOT signal.
Investigation processes can also differ, based on the already approved company procedures. The company OOS investigation requirements are regulated and the company has strict expectations, while the OOT is usually analyzed in the framework of company procedures for trends analysis, stability analysis of the products and pharmaceutical quality monitoring.
Results from the testing of stability are generated after a specified period of time. This gives the opportunity to see how a product acts over time.
OOT trends could come up in the following areas:
Analytical reasons sometimes consist of things like the flaws in the column used in chromatography, the inefficiencies of the measurement tool, defects in the preparation of calibrators, diversified ways of analyzing samples, or changes in working conditions.
Manufacturing causes include changes in the materials used, parameters of manufacturing or the working conditions of the technology, the effect of time on the process, drying or forming parameters, or others.
As regards stability samples, conditions of packaging and storage as well should be taken into account.
Thus, it is important to understand that the laboratory is not always to blame for an OOT result.
The analyst and the reviewer must look at the entire analytical scheme and supporting documents. This includes calculations, chromatograms, spectra, system suitability tests, instrument status, standard preparation, reagents preparation, sample preparation, and any other pertinent laboratory records.
The main goal is to find any laboratory-related reason.
If there is no laboratory-related reason found, then the investigation may go on to the area of production and the history of the product.
The focus on previous batches and time points becomes very important at this point.
The investigation may include manufacturing records, deviations, equipment history, lots of raw materials, process parameters, packaging materials, and storage conditions.
The trend of the result is also very important to analyze.
If an OOT investigation is carried out, trying to achieve a result closer to the historical average by repeating the test many times does not solve the issue and the initial result must be seen as part of the analytical history and evaluated scientifically.
Additional tests might be acceptable if the scientific reasons exist and the tests are conducted according to the proper protocol. But retesting should never be done simply for getting a more convenient result.
The purpose of investigation is to understand the first result, not to substitute it.
To analyse the typical variability, a business should possess enough relevant historical data. A proper comparison must include such aspects as a product, strength, lot number, method of production, the way of analysis, conditions of storage, and terms of testing.
Statistical methods can be applied to identify the unusual events. Depending on the purpose, companies can either use control charts or historical averages, standard deviation, regression analysis, limit values, or any other statistically valid method.
However, this means that statistical data must facilitate investigations but must not replace scientific approach.
Statistically unusual outcome does not mean that the product is defective. Similarly, the absence of a statistical alert does not imply that the outcome is acceptable, if other quality criteria raise a number of questions.
The first is saying that the results can be ignored because they remain within specification. This means that trend monitoring becomes meaningless.
The second mistake is equating OOT to OOS automatically. Those situations are not the same and should be branched as per the appropriate processes.
Another frequently committed mistake is the establishment of random OOT thresholds where there's no sufficient amount of historical data or scientific reasoning for that.
In addition, there may be a tendency to focus on the laboratory completely. The real OOT signal may have its origin in manufacturing, raw materials, equipment, packaging, or storage.
Lastly, there are organizations that see repetitive OOT results but fail to analyze them on the whole. The single result may be insignificant but obtaining many results may generate a trend and be very important.
Suppose five batches are observed showing a gradual decline in dissolution results and they all satisfy specifications. While they are analyzed separately, nothing alarming is revealed.
Nevertheless, trend analysis can be utilized to indicate the very direction of the process.
The quality team is, after that, able to investigate, whether some adjustments have been made in formulation, granulation, compression, coating, raw materials, equipment, analytical methods, and/or other possible influences.
This technique allows taking actions before process gets to the point of real specification violations.
Last, but not least, this technique is in line with the principles of Quality Risk Management and Pharmaceutical Quality System reflected in ICH Q9 and ICH Q10.
FDA guidelines exist regarding OOS investigations, they already include good principles for laboratory investigation, such as the assessment of laboratory work, and the potential causes related to manufacturing when a laboratory error cannot be assigned. These investigation principles should also be useful for creating a strong procedure for OOT investigations as well.
ICH Q9(R1), is a document that describes the framework for discovering, evaluating, controlling, communicating, and reviewing risks related to the quality of pharmaceutical products. ICH Q10 includes the description of the pharmaceutical quality system and emphasizes monitoring, corrective actions, and continuous improvement.
The written procedure should describe how trends are recognized, which historical information is employed, who evaluates the results of the investigation, when the investigation starts, how laboratory and manufacturing factors are considered, and how the conclusions are recorded.
Moreover, the program must have the capability to make a difference between real trends and usual analytical variations.
Quality teams should conduct periodic reviews of OOT data within the framework of continuous monitoring of both processes and products. The multiple occurrences of OOT calls for taking additional actions, which may include monitoring processes again, reviewing processes, deriving analytically sound methods, carrying out the stability studies, and conducting CAPA depending on results of the investigation.
While the Out of Trend investigation might not seem relevant because the value fits within acceptable limits, it’s important not to overlook its potential significance.
Complete OOT investigation does not look for a laboratory result only but assembles the whole picture as regards historical data, laboratory performance, manufacturing conditions, raw materials, devices, packaging, storage and earlier batches.
When done as it should be, OOT monitoring is no longer statistical analysis but becomes an actual quality tool able to detect problems at an early stage and maintain control during the life of a product.
Such a situation is often referred to as being out of trend (OOT).
For that reason, any good quality system in pharma will not only look at whether an analytical result passes or fails. It will also determine whether the result has behaved as expected.
What is the Meaning of Out of Trend?
Out of Trend means a result that signifies a certain deviation from a historically established trend, although such a result can regard applicable specifications.Take an example of a drug that has stability assay specification ranging from 95.0% to 105.0%. Suppose the stability results are as follows: 99.6%, 99.3%, 99.1%, 98.8%, 96.9% recorded at the different points of time.
Although the latest result is within specification, this does not mean it is automatically classified as OOS; nevertheless, a significant decline in the value as compared to previous records requires explanation.
In particular, it does not mean that 96.9% is not acceptable; however, the question is why the value changed so much.
This is the main goal of Out of Trend evaluation.
Importance of Out of Trend (OOT) Results
Specifications provide the acceptable range for the values of a particular characteristics but do not cover all possible changes within that range.For instance, if a given product has been historically producing results of assays in the range of 98.5-100.5%, an unusual shift to the values of 96.5-97.0% may still meet the specifications but indicate that process has started to drift.
Should the trend continue, the product will eventually fail.
This is how an appropriate OOT system works by providing an early warning. Such system will allow quality unit and manufacturing teams to identify possible issues on early stages.
This methodology is of great importance for drugs with long stability studies, since delay in response due to OOS results might cause loss of very important information.
OOT and OOS Are Not the Same
The reason why OOT and OOS are sometimes confused and used indistinguishably.An OOS result means an unacceptable outcome that does not conform to an established specification, or not meeting acceptance criteria.
An OOT result is a result that is specified, but represents an unexpected trend or variation in comparison with other results that were obtained previously.
For example, if a laboratory has an assay specification of 95.0% to 105.0% and the result obtained is 94.6% then the result will be regarded as OOS.
On the other hand, if the obtained result equals to 97.0% but historically the other results were close to 99.5%–100.5%, the result may serve as a potential OOT signal.
Investigation processes can also differ, based on the already approved company procedures. The company OOS investigation requirements are regulated and the company has strict expectations, while the OOT is usually analyzed in the framework of company procedures for trends analysis, stability analysis of the products and pharmaceutical quality monitoring.
Stability Studies: A Major Area for OOT Evaluation
One of the most influential areas to detect OOT trends is the field of stability studies.Results from the testing of stability are generated after a specified period of time. This gives the opportunity to see how a product acts over time.
OOT trends could come up in the following areas:
- Assay
- Dissolution
- Contaminants
- Degradation products
- Water content
- pH
- Preservatives
- Physical characteristics
- Microbial characteristics
What Can Cause an OOT Result?
An OOT result is brought on by any number of factors. Hence, it may begin at the laboratory stage or during manufacturing or through materials, tools, packages, or conditions of storage.Analytical reasons sometimes consist of things like the flaws in the column used in chromatography, the inefficiencies of the measurement tool, defects in the preparation of calibrators, diversified ways of analyzing samples, or changes in working conditions.
Manufacturing causes include changes in the materials used, parameters of manufacturing or the working conditions of the technology, the effect of time on the process, drying or forming parameters, or others.
As regards stability samples, conditions of packaging and storage as well should be taken into account.
Thus, it is important to understand that the laboratory is not always to blame for an OOT result.
How Should an OOT Investigation Be Done?
The investigation should start with the examination of the original data.The analyst and the reviewer must look at the entire analytical scheme and supporting documents. This includes calculations, chromatograms, spectra, system suitability tests, instrument status, standard preparation, reagents preparation, sample preparation, and any other pertinent laboratory records.
The main goal is to find any laboratory-related reason.
If there is no laboratory-related reason found, then the investigation may go on to the area of production and the history of the product.
The focus on previous batches and time points becomes very important at this point.
The investigation may include manufacturing records, deviations, equipment history, lots of raw materials, process parameters, packaging materials, and storage conditions.
The trend of the result is also very important to analyze.
Do Not Use Retesting to Make the Trend Disappear
One of the weaknesses that might be found in investigations is the excessive use of retesting.If an OOT investigation is carried out, trying to achieve a result closer to the historical average by repeating the test many times does not solve the issue and the initial result must be seen as part of the analytical history and evaluated scientifically.
Additional tests might be acceptable if the scientific reasons exist and the tests are conducted according to the proper protocol. But retesting should never be done simply for getting a more convenient result.
The purpose of investigation is to understand the first result, not to substitute it.
The Significance of Historical Data
The effectiveness of OOT evaluation is determined by the historical data used in it.To analyse the typical variability, a business should possess enough relevant historical data. A proper comparison must include such aspects as a product, strength, lot number, method of production, the way of analysis, conditions of storage, and terms of testing.
Statistical methods can be applied to identify the unusual events. Depending on the purpose, companies can either use control charts or historical averages, standard deviation, regression analysis, limit values, or any other statistically valid method.
However, this means that statistical data must facilitate investigations but must not replace scientific approach.
Statistically unusual outcome does not mean that the product is defective. Similarly, the absence of a statistical alert does not imply that the outcome is acceptable, if other quality criteria raise a number of questions.
Common Mistakes in OOT Investigation
There are some mistakes that can decrease the efficacy of any OOT program.The first is saying that the results can be ignored because they remain within specification. This means that trend monitoring becomes meaningless.
The second mistake is equating OOT to OOS automatically. Those situations are not the same and should be branched as per the appropriate processes.
Another frequently committed mistake is the establishment of random OOT thresholds where there's no sufficient amount of historical data or scientific reasoning for that.
In addition, there may be a tendency to focus on the laboratory completely. The real OOT signal may have its origin in manufacturing, raw materials, equipment, packaging, or storage.
Lastly, there are organizations that see repetitive OOT results but fail to analyze them on the whole. The single result may be insignificant but obtaining many results may generate a trend and be very important.
OOT as an Early Warning System
A complete pharmaceutical quality process has to be utilizing OOT information in a proactive manner.Suppose five batches are observed showing a gradual decline in dissolution results and they all satisfy specifications. While they are analyzed separately, nothing alarming is revealed.
Nevertheless, trend analysis can be utilized to indicate the very direction of the process.
The quality team is, after that, able to investigate, whether some adjustments have been made in formulation, granulation, compression, coating, raw materials, equipment, analytical methods, and/or other possible influences.
This technique allows taking actions before process gets to the point of real specification violations.
Last, but not least, this technique is in line with the principles of Quality Risk Management and Pharmaceutical Quality System reflected in ICH Q9 and ICH Q10.
Regulatory Expectations
There is absolutely no replacement for a scientifically valid investigation when an unexpected quality signal is found.FDA guidelines exist regarding OOS investigations, they already include good principles for laboratory investigation, such as the assessment of laboratory work, and the potential causes related to manufacturing when a laboratory error cannot be assigned. These investigation principles should also be useful for creating a strong procedure for OOT investigations as well.
ICH Q9(R1), is a document that describes the framework for discovering, evaluating, controlling, communicating, and reviewing risks related to the quality of pharmaceutical products. ICH Q10 includes the description of the pharmaceutical quality system and emphasizes monitoring, corrective actions, and continuous improvement.
Development of an Effective OOT Program
The development of an effective out-of-trend (OOT) program must start with a clearly written written procedure.The written procedure should describe how trends are recognized, which historical information is employed, who evaluates the results of the investigation, when the investigation starts, how laboratory and manufacturing factors are considered, and how the conclusions are recorded.
Moreover, the program must have the capability to make a difference between real trends and usual analytical variations.
Quality teams should conduct periodic reviews of OOT data within the framework of continuous monitoring of both processes and products. The multiple occurrences of OOT calls for taking additional actions, which may include monitoring processes again, reviewing processes, deriving analytically sound methods, carrying out the stability studies, and conducting CAPA depending on results of the investigation.
While the Out of Trend investigation might not seem relevant because the value fits within acceptable limits, it’s important not to overlook its potential significance.
Complete OOT investigation does not look for a laboratory result only but assembles the whole picture as regards historical data, laboratory performance, manufacturing conditions, raw materials, devices, packaging, storage and earlier batches.
When done as it should be, OOT monitoring is no longer statistical analysis but becomes an actual quality tool able to detect problems at an early stage and maintain control during the life of a product.
No comments:
Post a Comment
Please don't spam. Comments having links would not be published.