Method validation

Introduction

Method validation is the process used to confirm that the analytical procedure employed for a specific test is suitable for its intended use. Results from method validation can be used to judge the quality, reliability and consistency of analytical results; it is an integral part of any good analytical practice.

Analytical methods need to be validated or revalidated

• before their introduction into routine use;
• whenever the conditions change for which the method has been validated (e.g., an instrument with different characteristics or samples with a different matrix); and
• whenever the method is changed and the change is outside the original scope of the method.

Method validation has received considerable attention in the literature and from industrial committees and regulatory agencies.

• The U.S. FDA CGMP (1) request in section 211.165 (e) methods to be validated: The accuracy, sensitivity, specificity, and reproducibility of test methods employed by the firm shall be established and documented. Such validation and documentation may be accomplished in accordance with Sec. 211.194(a). These requirements include a statement of each method used in testing the sample to meet proper standards of accuracy and reliability, as applied to the tested product. The U.S. FDA has also proposed an industry guidance for Analytical Procedures and Methods Validation (2).
• ISO/IEC 17025 includes a chapter on the validation of methods (3) with a list of nine validation parameters. The ICH (4) has developed a consensus text on the validation of analytical procedures. The document includes definitions for eight validation characteristics. ICH also developed a guidance with detailed methodology (5).
• The U.S. EPA prepared a guidance for method’s development and validation for the Resource Conservation and Recovery Act (RCRA) (6). The AOAC, the EPA and other scientific organizations provide methods that are validated through multi-laboratory studies.

The USP has published specific guidelines for method validation for compound evaluation (7). USP defines eight steps for validation:

• Accuracy
• Precision
• Specificity
• Limit of detection
• Limit of quantitation
• Linearity and range
• Ruggedness
• Robustness

The FDA has also published a guidance for the validation of bioanalytical methods (8). The most comprehensive document is the conference report of the 1990 Washington conference: Analytical Methods Validation: Bioavailability, Bioequivalence and Pharmacokinetic Studies, which was sponsored by, among others, the American Association of Pharmaceutical Scientists (AAPS), the AOAC and the U.S. FDA (70). The report presents guiding principles for validating studies of both human and animal subjects. The report has also been used as a basis for the FDA industry guidance document (8).

Representatives of the pharmaceutical and chemical industry have published papers on the validation of analytical methods. Hokanson (9,10) applied the life cycle approach, developed for computerized systems, to the validation and revalidation of methods. Green (11) gave a practical guide for analytical method validation, with a description of a set of minimum requirements for a method. Renger and his colleagues (12) described the validation of a specific analytical procedure for the analysis of theophylline in a tablet using high-performance thin layer chromatography (HPTLC). The validation procedure in this particular article is based on requirements for EU multistate registration.

Wegscheider (13) has published procedures for method validation with a special focus on calibration, recovery experiments, method comparison and investigation of ruggedness. Seno et al. (14) have described how analytical methods are validated in a Japanese QC laboratory. The AOAC (15) has developed a Peer-Verified Methods validation program with detailed guidelines on exactly which parameters should be validated. Winslow and Meyer (16) recommend the definition and application of a master plan for validating analytical methods. J.Breaux and colleagues have published a study on analytical methods development and validation (17). The key point is to develop methods for easy validation and revalidation. O. Krause published a guide for analytical method transfer, comparability, maintenance and acceptance criteria for the testing of biopharmaceuticals (18).

This primer gives a review and a strategy for the validation of analytical methods for both methods developed in-house as well as standard methods, and a recommendation on the documentation that should be produced during, and on completion of, method validation. It also describes what is important when transferring a method.

Strategy for the Validation of Methods

The validity of a specific method should be demonstrated in laboratory experiments using samples or standards that are similar to unknown samples analyzed routinely. The preparation and execution should follow a validation protocol, preferably written in a step-by-step instruction format. Possible steps for a complete method validation are listed in Table 1. This proposed procedure assumes that the instrument has been selected and the method has been developed. It meets criteria such as ease of use; ability to be automated and to be controlled by computer systems; costs per analysis; sample throughput; turnaround time; and environmental, health and safety requirements.

1. Develop a validation protocol, an operating procedure or a validation master plan for the validation
2.  For a specific validation project define owners and responsibilities
3. Develop a validation project plan
4. Define the application, purpose and scope of the method
5. Define the performance parameters and acceptance criteria
6. Define validation experiments
7. Verify relevant performance characteristics of equipment
8.  Qualify materials, e.g. standards and reagents for purity, accurate amounts and sufficient stability
9.  Perform pre-validation experiments
10. Adjust method parameters or/and acceptance criteria if necessary
11.  Perform full internal (and external) validation experiments
12.  Develop SOPs for executing the method in the routine
13.  Define criteria for revalidation
14.  Define type and frequency of system suitability tests and/or analytical quality control (AQC) checks for the routine
15.  Document validation experiments and results in the validation report

Table 1. Steps in Method Validation

Successful acceptance of the validation parameters and performance criteria, by all parties involved, requires the cooperative efforts of several departments, including analytical development, QC, regulatory affairs and the individuals requiring the analytical data. The operating procedure or the Validation Master Plan (VMP) should clearly define the roles and responsibilities of each department involved in the validation of analytical methods.

The scope of the method and its validation criteria should be defined early in the process. These include the following questions:

•  What analytes should be detected?
•  What are the expected concentration levels?
•  What are the sample matrices?
•  Are there interfering substances expected, and, if so, should they be detected and quantified?
•  Are there any specific legislative or regulatory requirements?
•  Should information be qualitative or quantitative?
•  What are the required detection and quantitation limits?
•  What is the expected concentration range?
• What precision and accuracy is expected?
• How robust should the method be?
• Which type of equipment should be used? Is the method for one specific instrument, or should it be used by all instruments of the same type?
• Will the method be used in one specific laboratory or should it be applicable in all laboratories at one side or around the globe?
• What skills do the anticipated users of the method have?

The method’s performance characteristics should be based on the intended use of the method. It is not always necessary to validate all analytical parameters that are available for a specific technique. For example, if the method is to be used for qualitative trace level analysis, there is no need to test and validate the method’s limit of quantitation, or the linearity, over the full dynamic range of the equipment. Initial parameters should be chosen according to the analyst’s experience and best judgment. Final parameters should be agreed between the lab or analytical chemist performing the validation and the lab or individual applying the method and users of the data to be generated by the method. Table 2 gives examples of which parameters might be tested for a particular analysis task.

The scope of the method should also include the different types of equipment and the locations where the method will be run. For example, if the method is to be run on a specific instrument in a specific laboratory, there is no need to use instruments from other vendors or to include other laboratories in the validation experiments. In this way, the experiments can be limited to what is really necessary.

	Major compounds	Major compounds and traces	Traces	Traces
	quantitative	quantitative	qualitative	qualitative
limit of detection	no	no	yes	no
limit of quantitation	no	yes	no	yes
linearity	yes	yes	no	yes
range	yes	yes	no	no
precision	yes	yes	no	yes
accuracy	yes	yes	no	yes
specificity	yes	yes	yes	yes
ruggedness	yes	yes	no	may be

Table 2. Validation parameters for specific tasks

The validation experiments should be carried out by an experienced analyst to avoid errors due to inexperience. The analyst should be very well versed in the technique and operation of the instrument. Before an instrument is used to validate a method, its performance specifications should be verified using generic chemical standards. Satisfactory results for a method can be obtained only with equipment that is performing well. Special attention should be paid to those equipment characteristics that are critical for the method. For example, if detection limit is critical for a specific method, the instrument’s specification for baseline noise and, for certain detectors, the response to specified compounds should be verified.

Any chemicals used to determine critical validation parameters, such as reagents and reference standards, should be

1. available in sufficient quantities,
2. accurately identified,
3. sufficiently stable and
4. checked for exact composition and purity.

Any other materials and consumables, for example, chromatographic columns, should be new and be qualified to meet the column’s performance criteria . This ensures that one set of consumables can be used for most experiments and avoids unpleasant surprises during method validation.

Operators should be sufficiently familiar with the technique and equipment. This will allow them to identify and diagnose unforeseen problems more easily and to run the entire process more efficiently.

If there is little or no information on the method’s performance characteristics, it is recommended to prove the suitability of the method for its intended use in initial experiments. These studies should include the approximate precision, working range and detection limits. If the preliminary validation data appear to be inappropriate, the method itself, the equipment, the analysis technique or the acceptance limits should be changed. Method development and validation are, therefore, an iterative process. For example, in liquid chromatography, selectivity is achieved through the selection of mobile phase composition. For quantitative measurements, the resolution factor between two peaks should be 2.5 or higher. If this value is not achieved, the mobile phase composition needs further optimization. The influence of operating parameters on the performance of the method should be assessed at this stage if this was not done during development and optimization of the method.

There are no official guidelines on the correct sequence of validation experiments, and the optimal sequence may depend on the method itself. Based on the author’s experience, for a liquid chromatographic method, the following sequence has proven to be useful:

1.  Selectivity of standards (optimizing separation and detection of standard mixtures if selectivity is insufficient)
2.  Linearity, limit of quantitation, limit of detection, range
3.  Repeatability (short-term precision) of retention times and peak areas
4.  Intermediate precision
5.  Selectivity with real samples
6.  Trueness/accuracy at different concentrations
7.  Ruggedness (interlaboratory studies)

The more time-consuming experiments, such as accuracy and ruggedness, are included toward the end. Some of the parameters, as listed under (2) to (6), can be measured in combined experiments. For example, when the precision of peak areas is measured over the full concentration range, the data can be used to validate the linearity.

During method validation, the parameters, acceptance limits and frequency of ongoing system suitability tests or QC checks should be defined. Criteria should be defined to indicate when the method and system are beyond statistical control. The aim is to optimize these experiments so that, with a minimum number of control analyses, the method and the complete analytical system will provide long-term results to meet the objectives defined in the scope of the method.

Once the method has been developed and validated, a validation report should be prepared that includes the following:

•  Objective and scope of the method (applicability, type).
•  Summary of methodology.
•  Type of compounds and matrix.
•  All chemicals, reagents, reference standards, QC samples with purity, grade, their source or detailed instructions on their preparation.
•  Procedures for quality checks of standards and chemicals used.
•  Safety precautions.
•  A plan and procedure for method implementation from the method development lab to routine analysis.
•  Method parameters.
•  Critical parameters taken from robustness testing.
•  Listing of equipment and its functional and performance requirements, e.g., cell dimensions, baseline noise and column temperature range. For complex equipment, a picture or schematic diagram may be useful.
•  Detailed conditions on how the experiments were conducted, including sample preparation. The report must be detailed enough to ensure that it can be reproduced by a competent technician with comparable equipment.
•  Statistical procedures and representative calculations.
•  Procedures for QC in routine analyses, e.g., system suitability tests.
•  Representative plots, e.g., chromatograms, spectra and calibration curves.
•  Method acceptance limit performance data.
• The expected uncertainty of measurement results.
• Criteria for revalidation.
• The person(s) who developed and validated the method.
• References (if any).
• Summary and conclusions.
• Approval with names, titles, date and signature of those responsible for the review and approval of the analytical test procedure.

 

UPDATED´ USP CHAPTER <1226> for Verification of Compendial Methods

In Oct 2009 USP had published a Stimuli article “Transfer of Analytical Procedures". Based on comments received, USP now proposes a new general information chapter <1224>. The procedure-transfer process focuses on qualifying the receiving laboratory to perform an analytical procedure that was developed and validated in another laboratory within the same or in a different organization. One of the major differences to the stimuli paper chapter is that <1224> suggests a risk based approach for type and extent of transfer activities, e.g., for comparative testing. To learn everything about the new intended chapter, attend the audio seminar  "Transfer of Analytical Procedures According to the New USP <1224 span=""> and receive SOPs, templates and examples for easy implementation.

Verification of Standard Methods

A laboratory applying a specific method should have documented evidence that the method has been appropriately validated. This holds for methods developed in-house, as well as for standard methods, for example, those developed by organizations such as the EPA, American Society for Testing and Materials (ASTM), ISO or the USP.

A number of questions usually arises about the validation of standard methods: Firstly, should these methods be revalidated in the user’s laboratory and, if so, should method revalidation cover all experiments, as performed during initial validation? Secondly, which documentation should be available or developed in-house for standard methods? Official guidelines and regulations are not explicit about validating standard methods. Only CITAC/EURACHEM guide (19) includes a short paragraph that reads as follows:

The validation of standard or collaboratively tested methods should not be taken for granted, no matter how impeccable the method’s pedigree - the laboratory should satisfy itself that the degree of validation of a particular method is adequate for the required purpose, and that the laboratory is itself able to match any stated performance data.

There are two important requirements in this excerpt:

1. The standard’s method validation data are adequate and sufficient to meet the laboratory’s method requirements.
2. The laboratory must be able to match the performance data as described in the standard.

Further advice comes from FDA’s 21 CFR 194 section(a)2: “If the method employed is in the current revision of the United States Pharmacopeia, National Formulary, Association of Official Analytical Chemists, or in other recognized standard references, or is detailed in an approved new drug application and the referenced method is not modified, a statement indicating the method and reference will suffice. The suitability of all testing methods used shall be verified under actual conditions of use.” The spirit of this text is in line with the two requirements listed above.

This section elaborates on what these statements mean in practice, and it gives a strategy for validating standard methods. Like the validation of methods developed in-house, the evaluation and verification of standard methods should also follow a documented process that is usually the validation plan. Results should be documented in the validation protocol. Both documents will be the major source for the validation report.

Figure 1. Workflow for evaluation and validation of standard methods

An example of a step-by-step plan for the evaluation and validation of standard methods is shown as a flow diagram in Figure 1. As a first step, the scope of the method, as applied in the user’s laboratory, should be defined. This should be done independently of what is written in the standard method and should include information such as

• the type of compounds to be analyzed,
• matrices,
• the type of information required (qualitative or quantitative),
• detection and quantitation limits,
• range,
• precision and accuracy as specified by the client of the analytical data and
• the type of equipment—its location and environmental conditions.

As a second step, the method’s performance requirements should be defined in considerable detail, again irrespective of what has been validated in the standard method. General guidelines on validation criteria for different measurement objectives and procedures for their evaluation are discussed later in this chapter.

The results of these steps lead to the experiments that are required for adequate method validation and to the minimal acceptance criteria necessary to prove that the method is suitable for its intended use. Third, required experiments and expected results should be compared with what is written in the standard method.
In particular, the standard method should be checked for the following items:

1. Have the reported validation results been obtained from the complete procedure or from just a part of it? Sometimes the validation data from the published method have been obtained from the chromatographic analysis but have not included sample preparation steps. The diagram in Figure 2 can be used for this check. A complete validation of the analytical procedure should include the entire process from sampling, sample preparation, analysis, calibration and data evaluation to reporting.
2. Has the same matrix been used?
3. Did the validation experiments cover the complete concentration range as intended for the method in the user’s laboratory? If so, has the method’s performance been checked at the different concentration ranges?
4. Has the same equipment (brand, model) been used as available in the user’s laboratory, and, if not, was the scope of standard method regarding this item broad enough to include the user’s equipment? This question is very important for a gradient HPLC analysis, where the HPLC’s delay volume can significantly influence the method’s selectivity.
5. Have performance characteristics, e.g., the limit of quantitation, been checked in compliance with the most recent guidelines, as required for the user’s laboratory (e.g., the ICH guideline (5) for pharmaceutical laboratories)? If not, does the test procedure have equivalency to the guideline?

Figure 2. Steps for validating complete analytical procedures. Standard methods should be checked if all steps are included in the validation data.

If either the scope, the validation parameters or the validation results do not meet the user’s requirements, adequate validation experiments should be defined, developed and carried out. The extent of these experiments depends on the overlap of the user requirements with the scope and results, as described in the standard method. If there is no overlap, a complete validation should be carried out. In the case of a complete overlap, validation experiments may not be necessary.

If method validation experiments are unnecessary, the user should prove the suitability of the method in his or her laboratory. This evidence should confirm that the user’s equipment, the people, the reagents and the environment are qualified to perform the analysis. The experiments may be an extract of the full method validation and should focus on the critical items of the method. Guidelines for these tests should have been developed during method development. If not, they should be developed and carried out at this stage. Typical experiments may include precision of amounts and limits of quantitation. The validation report should include a reference to the standard method.

Validation of Non-routine Methods

Frequently, a specific method is used for only a few sample analyses. The question should be raised as to whether this method also needs to be validated using the same criteria as recommended for routine analysis. In this case, the validation may take much more time than the sample analysis and may be considered inefficient, because the cost per sample will increase significantly. The answer is quite simple: Any analysis is worthwhile only if the data are sufficiently accurate; otherwise, sample analysis is pointless. The suitability of an analysis method for its intended use is a prerequisite to obtaining accurate data; therefore, only validated methods should be used to acquire meaningful data. However, depending on the situation, the validation efforts can be reduced for non-routine methods. The CITAG/ EURACHEM guide (19) includes a chapter on how to treat non-routine methods. The recommendation is to reduce the validation cost by using generic methods, for example, methods that are broadly applicable. A generic method could, for example, be based on capillary gas chromatography or on reversed phase gradient HPLC. With little or no modification, the method can be applied to a large number of samples. The performance parameters should have been validated on typical samples characterized by sample matrix, compound types and concentration range.

If, for example, a new compound with a similar structure in the same matrix is to be analyzed, the validation will require only a few key experiments. The documentation of such generic methods should be designed to easily accommodate small changes relating to individual steps, such as sample preparation, sample analysis or data evaluation.

The method’s operating procedure should define the checks that need to be carried out for a novel analyte in order to establish that the analysis is valid. Detailed documentation of all experimental parameters is important to ensure that the work can be repeated in precisely the same manner at any later date.

Quality Control Plan and Implementation for Routine

For any method that will be used for routine analysis, a QC plan should be developed. This plan should ensure that the method, together with the equipment, delivers consistently accurate results. The plan may include recommendations for the following:

1. Selection, handling and testing of QC standards
2. Type and frequency of equipment checks and calibrations (for example, should the wavelength accuracy and the baseline noise of an HPLC UV detector be checked after each sample analysis, or on a daily or weekly basis?)
3. Type and frequency of system suitability testing (for example, at which point during the sequence system should suitability standards be analyzed?)
4. Type and frequency of QC samples (for example, should a QC sample be analyzed after 1, 5, 20 or 50 unknown samples, and should there be single or duplicate QC sample analysis, or should this be run at one or several concentrations?)
5. Acceptance criteria for equipment checks, system suitability tests and QC sample analysis
6. Action plan in case criteria 2, 3 and/or 4 are not met.

In many cases, methods are developed and validated in service laboratories that are specialized in this task. When the method is transferred to the routine analytical laboratory, care should be taken that the method and its critical parameters are well understood by the workers in the departments who apply the method. A detailed validation protocol, a documented procedure for method implementation and good communication between the development and operation departments are equally important. If the method is used by a number of departments, it is recommended to verify method validation parameters and to test the applicability and usability of the method in a couple of these departments before it is distributed to other departments. In this way, problems can be identified and corrected before the method is distributed to a larger audience. If the method is intended to be used by just one or two departments, an analyst from the development department should assist the users of the method during initial operation. Users of the method should be encouraged to give constant feedback on the applicability and usability of the method to the development department. The latter should correct problems if any arise.

Transferring Validated Routine Methods

Validated routine methods are transferred between laboratories at the same or different sites when contract laboratories offer services for routine analysis in different areas or when products are manufactured in different areas. When validated routine methods are transferred between laboratories and sites, their validated state should be maintained to ensure the same reliable results in the receiving laboratory. This means the competence of the receiving laboratory to use the method should be demonstrated through tests, for example, repeat critical method validation experiments and run samples in parallel in the transferring and receiving laboratories. The transfer should be controlled by a procedure, The recommended steps are:

•  Designate a project owner
•  Develop a transfer plan
•  Define transfer tests and acceptance
  criteria (validation experiments, sample
  analysis: sample type, #replicates)
•  Describe rational for tests
•  Train receiving lab operators in transferring lab on equipment, method, critical parameters and troubleshooting
•  Repeat 2 critical method validation tests in routine lab
•  Analyze at least three samples in transferring and receiving lab
•  Document transfer results

Revalidation

Most likely some method parameters have to be changed or adjusted during the life of the method if the method performance criteria fall outside their acceptance criteria. The question is whether such change requires revalidation. In order to clarify this question upfront, operating ranges should be defined for each method, either based on experience with similar methods or else investigated during method development. These ranges should be verified during method validation in robustness studies and should be part of the method characteristics. Availability of such operating ranges makes it easier to decide when a method should be revalidated. A revalidation is necessary whenever a method is changed, and the new parameter lies outside the operating range. If, for example, the operating range of the column temperature has been specified to be between 30 and 40°C, the method should be revalidated if, for whatever reason, the new operating parameter is 41°C.

Revalidation is also required if the scope of the method has been changed or extended, for example, if the sample matrix changes or if operating conditions change. Furthermore, revalidation is necessary if the intention is to use instruments with different characteristics, and these new characteristics have not been covered by the initial validation. For example, an HPLC method may have been developed and validated on a pump with a delay volume of 5 mL, but the new pump has a delay volume of only 0.5 mL.

Figure 3. Flow diagram for revalidation

Part or full revalidation may also be considered if system suitability tests, or the results of QC sample analysis, lie outside preset acceptance criteria and where the source of the error cannot be traced back to the instruments or any other cause.

Whenever there is a change that may require part or full revalidation, the change should follow a documented change control system. A flow diagram of such a process is documented in Figure 3. The change should be defined, authorized for implementation and documented. Possible changes may include

• new samples with new compounds or new matrices,
• new analysts with different skills,
• new instruments with different characteristics,
• new location with different environmental conditions,
• new chemicals and/or reference standards and
• modification of analytical parameters.

An evaluation should determine whether the change is within the scope of the method. If so, no revalidation is required. If the change lies outside the scope, the parameters for revalidation should be defined. After the validation experiments, the system suitability test parameters should be investigated and redefined, if necessary.

Parameters for Method Validation

The parameters for method validation have been defined in different working groups of national and international committees and are described in the literature. Unfortunately, some of the definitions vary between the different organizations. An attempt at harmonization was made for pharmaceutical applications through the ICH (4,5), where representatives from the industry and regulatory agencies from the United States, Europe and Japan defined parameters, requirements and, to some extent, methodology for analytical methods validation. The parameters, as defined by the ICH and by other organizations and authors, are summarized in Table 3 and are described in brief in the following paragraphs.

• Specificity (1,2)
•  Selectivity
•  Precision (1,2)
•  repeatability (1)
•  intermediate precision (1)
•  reproducibility (3)
•  Accuracy (1,2)
• Trueness
•  Bias
• Linearity (1,2)
•  Range (1,2)
•  Limit of detection (1,2)
•  Limit of quantitation (1,2)
•  Robustness (2,3)
•  Ruggedness (2)

Table 3. Possible analytical parameters for method validation

(1) Included in ICH publications, (2) Included in USP

(3) Terminology included in ICH publication but not part of required parameters

Selectivity/Specificity

The terms selectivity and specificity are often used interchangeably. A detailed discussion of this term, as defined by different organizations, has been presented by Vessmann (20). He particularly pointed out the difference between the definitions of specificity given by IUPAC/WELAC and the ICH.

Although it is not consistent with the ICH, the term specific generally refers to a method that produces a response for a single analyte only, while the term selective refers to a method that provides responses for a number of chemical entities that may or may not be distinguished from each other. If the response is distinguished from all other responses, the method is said to be selective. Since there are very few methods that respond to only one analyte, the term selectivity is usually more appropriate. The USP monograph (7) defines the selectivity of an analytical method as its ability to measure accurately an analyte in the presence of interference, such as synthetic precursors, excipients, enantiomers and known (or likely) degradation products that may be expected to be present in the sample matrix. Selectivity in liquid chromatography is obtained by choosing optimal columns and setting chromatographic conditions, such as mobile phase composition, column temperature and detector wavelength. Besides chromatographic separation, the sample preparation step can also be optimized for best selectivity.

It is a difficult task in chromatography to ascertain whether the peaks within a sample chromatogram are pure or consist of more than one compound. Therefore, the analyst should know how many compounds are in the sample or whether procedures for detecting impure peaks should be used.

While in the past chromatographic parameters such as mobile phase composition or the column were modified, now the application of spectroscopic detectors coupled on-line to the chromatograph is being used. UV/visible diode-array detectors and mass spectrometers acquire spectra on-line throughout the entire chromatogram. The spectra acquired during the elution of a peak are normalized and overlaid for graphical presentation. If the normalized spectra are different, the peak consists of at least two compounds.

The principles of diode-array detection in HPLC and their application and limitations with regard to peak purity are described in the literature (21). Examples of pure and impure HPLC peaks are shown in Figure 4. While the chromatographic signal indicates no impurities in either peak, the spectral evaluation identifies the peak on the left as impure. The level of impurities that can be detected with this method depends on the spectral difference, on the detector’s performance and on the software algorithm. Under ideal conditions, peak impurities of 0.05 to 0.1 percent can be detected.

Selectivity studies should also assess interferences that may be caused by the matrix, e.g., urine, blood, soil, water or food. Optimized sample preparation can eliminate most of the matrix components. The absence of matrix interferences for a quantitative method should be demonstrated by the analysis of at least five independent sources of control matrix.

Figure 4. Examples of pure and impure HPLC peaks. The chromatographic signal does not indicate any impurity in either peak. Spectral evaluation, however, identifies the peak on the left as impure.

Precision and Reproducibility

The precision of a method (Table 4) is the extent to which the individual test results of multiple injections of a series of standards agree. The measured standard deviation can be subdivided into 3 categories: repeatability, intermediate precision and reproducibility (4, 5). Repeatability is obtained when the analysis is carried out in a laboratory by an operator using a piece of equipment over a relatively short time span. At least 6 determinations of 3 different matrices at 2 or 3 different concentrations should be performed, and the RSD calculated.

The ICH (4) requires precision from at least 6 replications to be measured at 100 percent of the test target concentration or from at least 9 replications covering the complete specified range. For example, the results can be obtained at 3 concentrations with 3 injections at each concentration.

The acceptance criteria for precision depend very much on the type of analysis. Pharmaceutical QC precision of greater than 1 percent RSD is easily achieved for compound analysis, but the precision for biological samples is more like 15 percent at the concentration limits and 10 percent at other concentration levels. For environmental and food samples, precision is largely dependent on the sample matrix, the concentration of the analyte, the performance of the equipment and the analysis technique. It can vary between 2 percent and more than 20 percent.

The AOAC manual for the Peer-Verified Methods program (15) includes a table with estimated precision data as a function of analyte concentration (Table 4).

Intermediate precision is a term that has been defined by ICH (4) as the long-term variability of the measurement process. It is determined by comparing the results of a method run within a single laboratory over a number of weeks. A method’s intermediate precision may reflect discrepancies in results obtained

• from different operators,
• from inconsistent working practice (thoroughness) of the same operator,
• from different instruments,
• with standards and reagents from different suppliers,
• with columns from different batches or
• a combination of these.

Analyte%	Analyte Ratio	Unit	RSD%
100	1	100%	1.3
10	10-1	10%	2.8
1	10-2	1 %	2.7
0.1	10-3	0.1%	3.7
0.01	10-4	100 ppm	5.3
0.001	10-5	10 ppm	7.3
0.0001	10-6	1 ppm	11
0.00001	10-7	100 ppb	15
0.000001	10-8	10 ppb	21
0.0000001	10-9	1 ppb	30

Table 4. Analyte concentration versus precision (Ref. 15)

The objective of intermediate precision validation is to verify that in the same laboratory the method will provide the same results once the development phase is over.

Reproducibility (Table 5), as defined by the ICH (4), represents the precision obtained between different laboratories. The objective is to verify that the method will provide the same results in different laboratories. The reproducibility of an analytical method is determined by analyzing aliquots from homogeneous lots in different laboratories with different analysts, and by using operational and environmental conditions that may differ from, but are still within, the specified parameters of the method (interlaboratory tests). Validation of reproducibility is important if the method is to be used in different laboratories.

•  Differences in room temperature and humidity
•  Operators with different experience and thoroughness
•  Equipment with different characteristics, e.g. delay volume of an HPLC system
•  Variations in material and instrument conditions, e.g. in HPLC, mobile phases composition, pH, flow rate of mobile phase
•  Variation in experimental details not specified by the method
•  Equipment and consumables of different ages
•  Columns from different suppliers or different batches
•  Solvents, reagents and other material with varying quality

Table 5. Typical variations affecting a method’s reproducibility

Table 6 summarizes factors that should be the same, or different, for precision, intermediate precision and reproducibility.

	Precision	Intermediate Precision	Reprodu-cibility
Instrument	same	different	different
batches of accessories e.g. chrom. columns	same	different	different
Operators	same	different	different
Sample matrices	different	different	different
Concentration	different	different	different
Batches of material, e.g., reagents	same	different	different
Environmental conditions, e.g., temperature	same	different	different
Laboratory	same	same	different

Table 6. Variables for measurements of precision, intermediate precision and reproducibility

Accuracy and Recovery

The accuracy of an analytical method is the extent to which test results generated by the method and the true value agree. Accuracy can also be described as the closeness of agreement between the value that is adopted, either as a conventional, true or accepted reference value, and the value found.

The true value for accuracy assessment can be obtained in several ways. One alternative is to compare the results of the method with results from an established reference method. This approach assumes that the uncertainty of the reference method is known. Secondly, accuracy can be assessed by analyzing a sample with known concentrations (e.g., a control sample or certified reference material) and comparing the measured value with the true value as supplied with the material. If certified reference materials or control samples are not available, a blank sample matrix of interest can be spiked with a known concentration by weight or volume. After extraction of the analyte from the matrix and injection into the analytical instrument, its recovery can be determined by comparing the response of the extract with the response of the reference material dissolved in a pure solvent. Because this accuracy assessment measures the effectiveness of sample preparation, care should be taken to mimic the actual sample preparation as closely as possible. If validated correctly, the recovery factor determined for different concentrations can be used to correct the final results.

The concentration should cover the range of concern and should include concentrations close to the quantitation limit, one in the middle of the range and one at the high end of the calibration curve. Another approach is to use the critical decision value as the concentration point that must be the point of greatest accuracy.

Active ingredient (%)	Analyte Ratio	Unit	Mean Recovery(%)
100	1	100%	98-102
10	10-1	10%	98-102
1	10-2	1 %	97-103
0.1	10-3	0.1%	95-105
0.01	10-4	100 ppm	90-107
0.001	10-5	10 ppm	80-110
0.0001	10-6	1 ppm	80-110
0.00001	10-7	100 ppb	80-110
0.000001	10-8	10 ppb	60-115
0.0000001	10-9	1 ppb	40-120

Table 7. Analyte recovery at different concentrations (Ref 9)

The expected recovery (Table 7) depends on the sample matrix, the sample processing procedure and the analyte concentration. The AOAC manual for the Peer-Verified Methods program (15) includes a table with estimated recovery data as a function analyte concentration.

The ICH document on validation methodology recommends accuracy to be assessed using a minimum of nine determinations over a minimum of three concentration levels covering the specified range (e.g., three concentrations/three replicates each). Accuracy should be reported as percent recovery by the assay of known added amount of analyte in the sample or as the difference between the mean and the accepted true value, together with the confidence intervals.

Linearity and Calibration Curve

The linearity of an analytical method is its ability to elicit test results that are directly proportional to the concentration of analytes in samples within a given range or proportional by means of well-defined mathematical transformations. Linearity may be demonstrated directly on the test substance (by dilution of a standard stock solution) and/or by using separate weighings of synthetic mixtures of the test product components, using the proposed procedure.

Linearity is determined by a series of 3 to 6 injections of 5 or more standards whose concentrations span 80–120 percent of the expected concentration range. The response should be directly proportional to the concentrations of the analytes or proportional by means of a well-defined mathematical calculation. A linear regression equation applied to the results should have an intercept not significantly different from 0. If a significant nonzero intercept is obtained, it should be demonstrated that this has no effect on the accuracy of the method.

Frequently, the linearity is evaluated graphically, in addition to or as an alternative to mathematical evaluation. The evaluation is made by visually inspecting a plot of signal height or peak area as a function of analyte concentration. Because deviations from linearity are sometimes difficult to detect, two additional graphical procedures can be used. The first is to plot the deviations from the regression line versus the concentration or versus the logarithm of the concentration, if the concentration range covers several decades. For linear ranges, the deviations should be equally distributed between positive and negative values.

Another approach is to divide signal data by their respective concentrations, yielding the relative responses. A graph is plotted with the relative responses on the y-axis and the corresponding concentrations on the x-axis, on a log scale. The obtained line should be horizontal over the full linear range. At higher concentrations, there will typically be a negative deviation from linearity. Parallel horizontal lines are drawn on the graph corresponding to, for example, 95 percent and 105 percent of the horizontal line. The method is linear up to the point where the plotted relative response line intersects the 95 percent line. Figure 5 shows a comparison of the two graphical evaluations on a sample of caffeine using HPLC.

The ICH recommends, for accuracy reporting, the linearity curve’s correlation coefficient, y-intercept, slope of the regression line and residual sum of squares. A plot of the data should be included in the report. In addition, an analysis of the deviation of the actual data points from the regression line may also be helpful for evaluating linearity. Some analytical procedures, such as immunoassays, do not demonstrate linearity after any transformation. In this case, the analytical response should be described by an appropriate function of the concentration (amount) of an analyte in a sample. In order to establish linearity, a minimum of five concentrations is recommended. Other approaches should be justified.

Figure 5. Graphical presentations of linearity plot of a caffeine sample using HPLC.

Plotting the sensitivity (response/amount) gives clear indication of the linear range. Plotting the amount on a logarithmic scale has a significant advantage for wide linear ranges. Rc = Line of constant response.

Range

The range of an analytical method is the interval between the upper and lower levels (including these levels) that have been demonstrated to be determined with precision, accuracy and linearity using the method as written. The range is normally expressed in the same units as the test results (e.g., percentage, parts per million) obtained by the analytical method.

For assay tests, the ICH (5) requires the minimum specified range to be 80 to 120 percent of the test concentration, and for the determination of an impurity, the range to extend from the limit of quantitation, or from 50 percent of the specification of each impurity, whichever is greater, to 120 percent of the specification.

Figure 6. Definitions for linearity, range, LOQ, LOD

Limit of Detection

The limit of detection is the point at which a measured value is larger than the uncertainty associated with it. It is the lowest concentration of analyte in a sample that can be detected but not necessarily quantified. The limit of detection is frequently confused with the sensitivity of the method. The sensitivity of an analytical method is the capability of the method to discriminate small differences in concentration or mass of the test analyte. In practical terms, sensitivity is the slope of the calibration curve that is obtained by plotting the response against the analyte concentration or mass.

In chromatography, the detection limit is the injected amount that results in a peak with a height at least two or three times as high as the baseline noise level. Besides this signal/noise method, the ICH (4) describes three more methods:

1. Visual inspection: The detection limit is determined by the analysis of samples with known concentrations of analyte and by establishing the minimum level at which the analyte can be reliably detected.
2. Standard deviation of the response based on the standard deviation of the blank: Measurement of the magnitude of analytical background response is performed by analyzing an appropriate number of blank samples and calculating the standard deviation of these responses.
3. Standard deviation of the response based on the slope of the calibration curve: A specific calibration curve is studied using samples containing an analyte in the range of the limit of detection. The residual standard deviation of a regression line, or the standard deviation of y-intercepts of regression lines, may be used as the standard deviation.

Figure 7. Limit of detection and limit of quantitation via signal to noise

Limit of Quantitation

The limit of quantitation is the minimum injected amount that produces quantitative measurements in the target matrix with acceptable precision in chromatography, typically requiring peak heights 10 to 20 times higher than the baseline noise.

If the required precision of the method at the limit of quantitation has been specified, the EURACHEM (22) (Figure 8) approach can be used. A number of samples with decreasing amounts of the analyte are injected six times. The calculated RSD percent of the precision is plotted against the analyte amount. The amount that corresponds to the previously defined required precision is equal to the limit of quantitation. It is important to use not only pure standards for this test but also spiked matrices that closely represent the unknown samples.

For the limit of detection, the ICH (5) recommends, in addition to the procedures as described above, the visual inspection and the standard deviation of the response and the slope of the calibration curve.

Figure 11. Limit of quantitation with the EURACHEM (80) method.

Any results of limits of detection and quantitation measurements must be verified by experimental tests with samples containing the analytes at levels across the two regions. It is equally important to assess other method validation parameters, such as precision, reproducibility and accuracy, close to the limits of detection and quantitation. Figure 6 illustrates the limit of quantitation (along with the limit of detection, range and linearity). Figure 7 illustrates both the limit of detection and the limit of quantitation.

Ruggedness

Ruggedness is not addressed in the ICH documents (4,5) Its definition has been replaced by reproducibility, which has the same meaning as ruggedness, defined by the USP as the degree of reproducibility of results obtained under a variety of conditions, such as different laboratories, analysts, instruments, environmental conditions, operators and materials. Ruggedness is a measure of reproducibility of test results under normal, expected operational conditions from laboratory to laboratory and from analyst to analyst. Ruggedness is determined by the analysis of aliquots from homogeneous lots in different laboratories.

Robustness

Robustness tests examine the effect that operational parameters have on the analysis results. For the determination of a method’s robustness, a number of method parameters, for example, pH, flow rate, column temperature, injection volume, detection wavelength or mobile phase composition, are varied within a realistic range, and the quantitative influence of the variables is determined. If the influence of the parameter is within a previously specified tolerance, the parameter is said to be within the method’s robustness range.

Obtaining data on these effects helps to assess whether a method needs to be revalidated when one or more parameters are changed, for example, to compensate for column performance over time. In the ICH document (5), it is recommended to consider the evaluation of a method’s robustness during the development phase, and any results that are critical for the method should be documented. This is not, however, required as part of a registration.

Stability

Many solutes readily decompose prior to chromatographic investigations, for example, during the preparation of the sample solutions, extraction, cleanup, phase transfer or storage of prepared vials (in refrigerators or in an automatic sampler). Under these circumstances, method development should investigate the stability of the analytes and standards.

The term system stability has been defined as the stability of the samples being analyzed in a sample solution. It is a measure of the bias in assay results generated during a preselected time interval, for example, every hour up to 46 hours, using a single solution (Figure 9). System stability should be determined by replicate analysis of the sample solution. System stability is considered appropriate when the RSD, calculated on the assay results obtained at different time intervals, does not exceed more than 20 percent of the corresponding value of the system precision. If, on plotting the assay results as a function of time, the value is higher, the maximum duration of the usability of the sample solution can be calculated.

Figure 9. Schematics of stability testing

The effect of long-term storage and freeze-thaw cycles can be investigated by analyzing a spiked sample immediately after preparation and on subsequent days of the anticipated storage period. A minimum of two cycles at two concentrations should be studied in duplicate. If the integrity of the drug is affected by freezing and thawing, spiked samples should be stored in individual containers, and appropriate caution should be employed for the study of samples.

Which Parameters Should Be Included in Method Validation?

For an efficient validation process, it is of utmost importance to specify the right validation parameters and acceptance criteria. The more parameters, the more time it will take to validate. The more stringent the specifications or acceptance limits, the more often the equipment has to be recalibrated, and probably also requalified, to meet the higher specifications at any one time. It is not always essential to validate every analytical performance parameter, but it is necessary to define which ones are required. This decision should be based on business, regulatory and/or accreditation requirements:

1.  For contract analyses: What does the client request?
2.  For regulatory submission: What do the regulations or guidelines require?
3.  For laboratory accreditation: What do the standard and relevant guidelines recommend?

Analytical Task	Identi-fication	Impurity quantitative	Impurity qualitative	Assay Cate 3
Accuracy	no	no	yes	yes
Precision
- repeatability	no	yes	no	yes
- interim precision	no	yes	no	yes
- reproducibility	no	no	no	no
Specificity	yes	yes	yes	yes
Limit of detection	no	no	yes	no
Limit of quantitation	no	yes	no	no
Linearity	no	yes	no	yes
Range	no	yes	no	yes*

Table 8. ICH Characteristics

* may be required, depending on the nature of the specific test

Analytical Task	Assay Category 1	Cat 2 quantitative	Cat 3 qualitative	Assay Cate 3
Accuracy	yes	yes	*	*
Precision	yes	yes	no	yes
Specificity	yes	yes	yes	*
Limit of detection	no	no	yes	*
Limit of quantitation	no	yes	no	*
Linearity	yes	yes	no	*
Range	yes	yes	*	*
Ruggedness	yes	yes	yes	*

Table 9. USP Characteristics

* may be required, depending on the nature of the specific test

The validation parameters depend on the analytical task and the scope of the method. For example, both the USP (26) and the ICH (4) contain chapters on validation procedures for different analytical tasks, both of which are included to provide some ideas on what type of validations are required for different tasks (see Tables 8 and 9). For example, according to the ICH, accuracy, any type of precision and limits of detection and quantitation are not required if the analytical task is identification. For assays in USP category 1, the major component or active ingredient to be measured is normally present at high concentrations; therefore, validation of limits of detection and quantitation is not necessary.

Because the type of analysis and the information that should be obtained from a sample have so much influence on the validation, the objective and scope of the method should always be defined as the first step of any method validation.

Summary Recommendations

1. Develop a validation master plan or an operating procedure for method validation.
2. For individual method validation projects, develop a validation project plan
3.  Define intended use of the method and performance criteria.
4.  Check all equipment and material for performance and quality.
5.  Perform validation experiments.
6.  For standard methods: check scope of the standard with your own requirements.
7.  For non-routine methods: develop and use generic methods and customize them for specific non-routine tasks.
8.  Develop an operating procedure for method transfer between laboratories

References

1. U.S. FDA, Title 21 of the U.S. Code of Federal Regulations:
   21 CFR 211—Current good manufacturing practice for finished pharmaceuticals.
2. U.S. FDA - Guidance for Industry (draft) Analytical Procedures and Methods Validation: Chemistry, Manufacturing, and Controls and Documentation, 2000
3.  ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories, 2005
4. International Conference on Harmonization (ICH) of Technical Requirements for the Registration of Pharmaceuticals for Human Use, Validation of analytical procedures: definitions and terminology, Geneva (1996)
5.  International Conference on Harmonization (ICH) of Technical Requirements for the Registration of Pharmaceuticals for Human Use, Validation of analytical procedures: Methodology, adopted in 1996, Geneva
6.  U.S. EPA, Guidance for methods development and methods validation for the Resource Conservation and Recovery Act (RCRA) Program, Washington, D.C. (1995).,
   http://www.epa.gov/sw-846/pdfs/methdev.pdf
7.  General Chapter 1225, Validation of compendial methods, United States Pharmacopeia 30, National Formulary 25, Rockville, Md., USA, The United States Pharmacopeial Convention, Inc., (2007).
8. U.S. FDA - Guidance for Industry, Bioanalytical Method Validation
9.  G. C. Hokanson, A life cycle approach to the validation of analytical methods during pharmaceutical product development, Part I: The initial validation process, Pharm. Tech., Sept. 1994, pp. 118–130.
10. G. C. Hokanson, A life cycle approach to the validation of analytical methods during pharmaceutical product development, Part II: Changes and the need for additional validation, Pharm.Tech., Oct. 1994, pp. 92–100.
11. J. M. Green, A practical guide to analytical method validation, Anal. Chem. News & Features, 1 May 1996, pp. 305A–309A.
12. B. Renger, H. Jehle, M. Fischer and W. Funk, Validation of analytical procedures in pharmaceutical analytical chemistry: HPTLC assay of theophylline in an effervescent tablet, J. Planar Chrom. 8:269–278 (July/August 1995).
13. Wegscheider, Validation of analytical methods, in: Accreditation and quality assurance in analytical chemistry, edited by H. Guenzler, Springer Verlag, Berlin (1996).
14. S. Seno, S. Ohtake and H. Kohno, Analytical validation in practice at a quality control laboratory in the Japanese pharmaceutical industry, Accred. Qual. Assur. 2:140–145 (1997).
15. 9. AOAC Peer-Verified Methods Program, Manual on policies and procedures, Arlington, Va., USA (1998).
    http://www.aoac.org/vmeth/PVM.pdf
16. 16. P. A. Winslow and R. F. Meyer, Defining a master plan for the validation of analytical methods, J. Validation Technology, pp. 361–367 (1997).
17. J.Breaux, K. Jones, and P. Boulas, Pharmaceutical Technology, Analytical Technology and Testing, 6-13, 2003
18. S.O. Krause, A Guide for testing biopharmaceuticals, Part II: acceptance criteria and analytical method maintenance, Pharm. Tech. Europe, Vol 18(6), 29/38, 2006
19. 29. CITAC/EURACHEM, Working Group, International guide to quality in analytical chemistry: An aid to accreditation, 2002.
20. J. Vessman, Selectivity or specificity? Validation of analytical methods from the perspective of an analytical chemist in the pharmaceutical industry, J. Pharm & Biomed Analysis 14:867–869 (1996).
21. L. Huber and S. George, Diode-array detection in high-performance liquid chromatography, New York, Marcel Dekker, ISBN 0-8247-4 (1993).
22. EURACHEM – The Fitness for Purpose of Analytical Methods A Laboratory Guide to Method Validation and Related Topics, 1998
23. AOAC Guidelines for Standard Method Performance Requirements (link)