《仪器与方法验证专著》（英文版） Analytical Method Validation and Instrument Performance veriticatig

VALIDATION PRACTICES able 2.1. Guidelines for Drug potency assay Characteristic Requirement Characteristic Requirement Precision Repeatability + Li Intermediate precision +, Signifies that this characteristic is normally evaluated; -, signifies that this characteristic is not normally evaluated. PIn cases where reproducibility has been achieved, intermediate precision is not needed 2.3 VALIDATION PRACTICES Different approaches may be used to validate the potency method. However, it is important to understand that the objective of validation is to demonstrate that a procedure is suitable for its intended purpose. With this in mind, the scientist will need to determine the extent of validation required. It is advisable design experimental work such that the appropriate validation characteristics be considered simultaneously to obtain overall knowledge of the capabilities of the analytical procedure 2.3.1 Types of Quantitation Quantitation by External Standard. This quantitation technique is the most straightforward. It involves the preparation of one or a series of standard solu- tions that approximate the concentration of the analyte Chromatograms of the standard solutions are obtained, and peak heights or areas are plotted as a tion of concentration of the analyte. The plot of the data should normally yie straight line. This is especially true for pharmaceuti forms of mathematical treatment can be used but will need to be justified. There are some potential instrumental sources of error that could occur using this quantitation technique. It is critical to have minimal variability between ach independent injection, as the quantitation is based on the comparison le sample and standard areas. However, the current autosamplers are able to minimize this variability to less than 0.5% relative standard deviation(RSD) Quantitation by Internal Standard. Quantitation by internal standard provides the highest precision because uncertainties introduced by sample injection are avoided. In this quantitation technique, a known quantity of internal standard is introduced into each sample and standard solutions. As in the external standard quantitation, chromatograms of the standard and sample solutions are integrate to determine peak heights or peak areas. The ratio of the peak height or area of the analyte to an internal standard is determined. The ratios of the standards

VALIDATION PRACTICES 13 Table 2.1. Guidelines for Drug Potency Assay Characteristic Requirementa Characteristic Requirementa Accuracy + Detection limit − Precision Quantitation limit − Repeatability + Linearity + Intermediate precisionb + Range + Specificity + a+, Signifies that this characteristic is normally evaluated; −, signifies that this characteristic is not normally evaluated. b In cases where reproducibility has been achieved, intermediate precision is not needed. 2.3 VALIDATION PRACTICES Different approaches may be used to validate the potency method. However, it is important to understand that the objective of validation is to demonstrate that a procedure is suitable for its intended purpose. With this in mind, the scientist will need to determine the extent of validation required. It is advisable to design experimental work such that the appropriate validation characteristics be considered simultaneously to obtain overall knowledge of the capabilities of the analytical procedure. 2.3.1 Types of Quantitation Quantitation by External Standard. This quantitation technique is the most straightforward. It involves the preparation of one or a series of standard solu￾tions that approximate the concentration of the analyte. Chromatograms of the standard solutions are obtained, and peak heights or areas are plotted as a func￾tion of concentration of the analyte. The plot of the data should normally yield a straight line. This is especially true for pharmaceuticals of synthetic origin. Other forms of mathematical treatment can be used but will need to be justified. There are some potential instrumental sources of error that could occur using this quantitation technique. It is critical to have minimal variability between each independent injection, as the quantitation is based on the comparison of the sample and standard areas. However, the current autosamplers are able to minimize this variability to less than 0.5% relative standard deviation (RSD). Quantitation by Internal Standard. Quantitation by internal standard provides the highest precision because uncertainties introduced by sample injection are avoided. In this quantitation technique, a known quantity of internal standard is introduced into each sample and standard solutions. As in the external standard quantitation, chromatograms of the standard and sample solutions are integrated to determine peak heights or peak areas. The ratio of the peak height or area of the analyte to an internal standard is determined. The ratios of the standards

POTENCY METHOD VALIDATION are plotted as a function of the concentration of the analyte. a plot of the data should normally yield a straight line Due to the presence of the internal standard, it is critical to ensure that the nalyte peak be separated from the internal standard peak. A minimum of base line separation(resolution >1.5) of these two peaks is required to give reliable quantitation. In addition, to quantitate the responses of internal standard accu- rately, the internal standard should be baseline resolved from any significant related substances and should have a peak height or area similar to that of the standard peak 2.3.2 Standard Plots for Quantitation In many instances in the pharmaceutical industry, drug products may be manu factured in a variety of strengths (e. g, levothyroxine tablets in strengths of 50 100, 150, 200, 500, and 750 ug). To develop and validate these potency methods, three strategies may ollowed Single-Point Calibration. A method may be developed and validated using only tandard analyte concentration. The standard plot generated is used to assay the complete range of tablet strengths. This strategy should be adopted wherever possible due to the simplicity of standard preparation and minimal work for quantitation of the sample. However, this method will require different extraction and dilution schemes of the various drug product strengths to give the same final concentration that is in the proximity of the one standard analyte concentration. Multiple-Point Calibration. Another strategy involves two or more standard con entrations that will bracket the complete range of the drug product strengths. In this strategy it is critical that the standard plots between the two extreme concentration ranges be linear. Therefore, this is a valid calibration method as long as the sample solutions of different strengths are prepared within the con- centration range of the calibration curve. Its advantage is that different strengths can utilize different preparation procedures and be more flexible. Its disadvan- tage is that multiple weighing of standards at different concentrations may give a weighing error One Standard Calibration for Each Strength. The least favored method is to develop and validate using one standard concentration for each product strength This situation will arise when the analyte does not exhibit linearity within reasonable concentration range 2.3.3 System Suitability requirement Prior to injecting a standard solution in m to ensure that the system is performing adequately for its intended purpose. This function is fulfilled by the use of a solution of the system suitability. System

14 POTENCY METHOD VALIDATION are plotted as a function of the concentration of the analyte. A plot of the data should normally yield a straight line. Due to the presence of the internal standard, it is critical to ensure that the analyte peak be separated from the internal standard peak. A minimum of base￾line separation (resolution >1.5) of these two peaks is required to give reliable quantitation. In addition, to quantitate the responses of internal standard accu￾rately, the internal standard should be baseline resolved from any significant related substances and should have a peak height or area similar to that of the standard peak. 2.3.2 Standard Plots for Quantitation In many instances in the pharmaceutical industry, drug products may be manu￾factured in a variety of strengths (e.g., levothyroxine tablets in strengths of 50, 100, 150, 200, 500, and 750 µg). To develop and validate these potency methods, three strategies may be followed. Single-Point Calibration. A method may be developed and validated using only one standard analyte concentration. The standard plot generated is used to assay the complete range of tablet strengths. This strategy should be adopted wherever possible due to the simplicity of standard preparation and minimal work for quantitation of the sample. However, this method will require different extraction and dilution schemes of the various drug product strengths to give the same final concentration that is in the proximity of the one standard analyte concentration. Multiple-Point Calibration. Another strategy involves two or more standard con￾centrations that will bracket the complete range of the drug product strengths. In this strategy it is critical that the standard plots between the two extreme concentration ranges be linear. Therefore, this is a valid calibration method as long as the sample solutions of different strengths are prepared within the con￾centration range of the calibration curve. Its advantage is that different strengths can utilize different preparation procedures and be more flexible. Its disadvan￾tage is that multiple weighing of standards at different concentrations may give a weighing error. One Standard Calibration for Each Strength. The least favored method is to develop and validate using one standard concentration for each product strength. This situation will arise when the analyte does not exhibit linearity within a reasonable concentration range. 2.3.3 System Suitability Requirements for Potency Assay Prior to injecting a standard solution in creating the standard plot, it is essential to ensure that the system is performing adequately for its intended purpose. This function is fulfilled by the use of a solution of the system suitability. System

VALIDATION PRACTICES suitability, an integral part of analytical procedures, is based on the concept that equipment, electronics, analytical operations, and samples constitute an integral system that can be evaluated. System suitability test parameters depend on the procedure being validated. The following notes should be given due consideration when evaluating a system su 1. System suitability is a measure of the performance of a given system on a given day within a particular sample analysis se The main objective of system suitability is to recognize whether or not system operation is adequate given such variability as chromatographic columns, column aging, mobile-phase variations, and variations in instru mentation 3. System suitability is part of method validation. Experience gained dur ing method development will give insights to help determine the system suitability requirements of the final method. An example is the hydrolysis of acetylsalicylic acid to salicylic acid in acidic media. Separation of this degradation peak from the analyte could be one criterion for the system uitability of an acetylsalicylic acid assay 4. A system suitability test should be performed in full each time a system is used for an assay. If the is in continuous use for the same analysi over an extended period suitability should be reevaluated at appro priate intervals to ensure he system is still functioning adequately for its intended use 5. System suitability should be on criteria and parameters collected as a group that will be able to define the performance of the system. So of the common parameters clude precision of repetitive injections (usually five or six), resolution(R), tailing factor (T), number of theoretical plates(N), and capacity factor(k) 2.3.4 Stability Indicating Potency Assay It is important to realize that the pharmaceutical regulators require that all potency assays be stability indicating Regulatory guidance in ICH Q2A, Q2B, Q3B, and FDA 21 CFR Section 211 [1-5] all require the development and validation of stability-indicating potency assays. Apart from the regulatory requirements, it is also good scientific practice to understand the interaction of the drug with its hysical environment. It is logical and reasonable that the laboratory validate methods that will be able to monitor and resolve degradation products as a result of the stability of the product with the environment. For drug substances, we may need to include synthetic process impurities It is common practice to utilize forced degradation studies to accelerate degra- dation of the drug substance or drug product to get an understanding of its egradation profile. Potential environmental conditions that can be used include 40°Cand75% relative humidity(RH,50°Cand75%RH,70°Cand75%RH or 80C and 75%RH. Oxidation, reduction, and pH-related degradations

VALIDATION PRACTICES 15 suitability, an integral part of analytical procedures, is based on the concept that equipment, electronics, analytical operations, and samples constitute an integral system that can be evaluated. System suitability test parameters depend on the procedure being validated. The following notes should be given due consideration when evaluating a system suitability sample. 1. System suitability is a measure of the performance of a given system on a given day within a particular sample analysis set. 2. The main objective of system suitability is to recognize whether or not system operation is adequate given such variability as chromatographic columns, column aging, mobile-phase variations, and variations in instru￾mentation. 3. System suitability is part of method validation. Experience gained dur￾ing method development will give insights to help determine the system suitability requirements of the final method. An example is the hydrolysis of acetylsalicylic acid to salicylic acid in acidic media. Separation of this degradation peak from the analyte could be one criterion for the system suitability of an acetylsalicylic acid assay. 4. A system suitability test should be performed in full each time a system is used for an assay. If the system is in continuous use for the same analysis over an extended period, system suitability should be reevaluated at appro￾priate intervals to ensure that the system is still functioning adequately for its intended use. 5. System suitability should be based on criteria and parameters collected as a group that will be able to define the performance of the system. Some of the common parameters used include precision of repetitive injections (usually five or six), resolution (R), tailing factor (T ), number of theoretical plates (N), and capacity factor (k
). 2.3.4 Stability Indicating Potency Assay It is important to realize that the pharmaceutical regulators require that all potency assays be stability indicating. Regulatory guidance in ICH Q2A, Q2B, Q3B, and FDA 21 CFR Section 211 [1–5] all require the development and validation of stability-indicating potency assays. Apart from the regulatory requirements, it is also good scientific practice to understand the interaction of the drug with its physical environment. It is logical and reasonable that the laboratory validate methods that will be able to monitor and resolve degradation products as a result of the stability of the product with the environment. For drug substances, we may need to include synthetic process impurities. It is common practice to utilize forced degradation studies to accelerate degra￾dation of the drug substance or drug product to get an understanding of its degradation profile. Potential environmental conditions that can be used include 40◦ C and 75% relative humidity (RH), 50◦ C and 75% RH, 70◦ C and 75% RH, or 80◦ C and 75% RH. Oxidation, reduction, and pH-related degradations are

POTENCY METHOD VALIDATION lso utilized for degradation studies. Usually, the target is to achieve 10 to 30% degradation. Creating more than 30%o degradation will not be useful, due to the potential for secondary degradation. Secondary degradation occurs when the first degradation impurity degrades further. Furthermore, degrading the drug substance or drug product beyond 30% will not be meaningful, since this is unacceptable in the market place ICH Q2A suggested validation of the characteristics of accuracy, precision, specificity, linearity, and range for potency and content uniformity assay. A detailed discussion of each of these parameters is presented later in this chapter Some examples of validation data are presented along with a brief critical dis on of the data 2. 4 STRATEGIES AND VALIDATION PARAMETERS The most important consideration for strategies of method validation is to design experimental work so that the appropriate validation characteristics are studied multaneously, thereby minimizing the number of experiments that need to be done. It is therefore important to write some form of protocol to aid the planning process. Executing the experimental work without prior planning will be a disaster for the validation 2.4.1 Linearity The iCh defines the linearity of an analytical procedure as the ability(within a given range) to obtain test results of variable data(e.g, absorbance and area under the curve) which are directly proportional to the concentration(amount of analyte) in the sample. The data variables that can be used for quantitation of the analyte are the peak areas, peak heights, or the ratio of peak areas(heights) of analyte to internal standard peak. Quantitation of the analyte depends on it obeying Beers law and is linear over a concentration range. Therefore, the working sample concentration and samples tested for accuracy should be in the linear range Linearity is usually demonstrated directly by dilution of a standard stock solu tion. It is recommended that linearity be performed by serial dilution of stock solution. Preparing the different concentrations by using different weights of standard will introduce weighing errors to the study of the linearity of the analyte(in addition to adding more work) but will not help to prove the linearity of the analyte. Linearity is best evaluated by visual inspection of a plot of the signals as a function of analyte concentration. Subsequently, the variable data are generally used to calculate a regression by the least squares method. As recommended by the ICh, the usual range for the potency assay of a drug substance or a drug product should be t20% of the target or nominal concentra- tion and +30% for a content uniformity assay. At least five concentration levels ould be used. Under normal circumstances, linearity is achieved when the co ficient of determination(r)is >0.997. The slope, residual sum of squares, and

16 POTENCY METHOD VALIDATION also utilized for degradation studies. Usually, the target is to achieve 10 to 30% degradation. Creating more than 30% degradation will not be useful, due to the potential for secondary degradation. Secondary degradation occurs when the first degradation impurity degrades further. Furthermore, degrading the drug substance or drug product beyond 30% will not be meaningful, since this is unacceptable in the market place. ICH Q2A suggested validation of the characteristics of accuracy, precision, specificity, linearity, and range for potency and content uniformity assay. A detailed discussion of each of these parameters is presented later in this chapter. Some examples of validation data are presented along with a brief critical dis￾cussion of the data. 2.4 STRATEGIES AND VALIDATION PARAMETERS The most important consideration for strategies of method validation is to design experimental work so that the appropriate validation characteristics are studied simultaneously, thereby minimizing the number of experiments that need to be done. It is therefore important to write some form of protocol to aid the planning process. Executing the experimental work without prior planning will be a disaster for the validation. 2.4.1 Linearity The ICH defines the linearity of an analytical procedure as the ability (within a given range) to obtain test results of variable data (e.g., absorbance and area under the curve) which are directly proportional to the concentration (amount of analyte) in the sample. The data variables that can be used for quantitation of the analyte are the peak areas, peak heights, or the ratio of peak areas (heights) of analyte to internal standard peak. Quantitation of the analyte depends on it obeying Beer’s law and is linear over a concentration range. Therefore, the working sample concentration and samples tested for accuracy should be in the linear range. Linearity is usually demonstrated directly by dilution of a standard stock solu￾tion. It is recommended that linearity be performed by serial dilution of a common stock solution. Preparing the different concentrations by using different weights of standard will introduce weighing errors to the study of the linearity of the analyte (in addition to adding more work) but will not help to prove the linearity of the analyte. Linearity is best evaluated by visual inspection of a plot of the signals as a function of analyte concentration. Subsequently, the variable data are generally used to calculate a regression by the least squares method. As recommended by the ICH, the usual range for the potency assay of a drug substance or a drug product should be ±20% of the target or nominal concentra￾tion and ±30% for a content uniformity assay. At least five concentration levels should be used. Under normal circumstances, linearity is achieved when the coef- ficient of determination (r2) is ≥0.997. The slope, residual sum of squares, and

STRATEGIES AND VALIDATION PARAMETERS y-intercept should also be reported as required by the ICh. The slope of the regression line will provide an idea of the sensitivity of the regression and hence the method to be validated. The y-intercept will provide the analyst with an estimate of the variability of the method. For example, the ratio percent of the y-intercept with the variable data at nominal concentration are sometimes used to estimate the method variability. Figures 2.2 and 2.3 illustrate acceptable and nonacceptable linearity data, respectively 2.4.2 Accuracy The ICh defines the accuracy of an analytical procedure as the closeness of agreement between the values that are accepted either as conventional true val- ues or an accepted reference value and the value found. Accuracy is usually reported as percent recovery by assay, using the proposed analytical proce dure, of known amount of analyte added to the sample. The ICH also rec- ommended assessing a minimum of nine determinations over a minimum of 6 1000 1500 Concentration re 2.2. Linearity with correlation coefficient greater than 0.997 16.000 12.000 10.000 Concentration (ug/mL) Figure 2.3. Linearity with correlation coefficient less than 0.997

STRATEGIES AND VALIDATION PARAMETERS 17 y-intercept should also be reported as required by the ICH. The slope of the regression line will provide an idea of the sensitivity of the regression and hence the method to be validated. The y-intercept will provide the analyst with an estimate of the variability of the method. For example, the ratio percent of the y-intercept with the variable data at nominal concentration are sometimes used to estimate the method variability. Figures 2.2 and 2.3 illustrate acceptable and nonacceptable linearity data, respectively. 2.4.2 Accuracy The ICH defines the accuracy of an analytical procedure as the closeness of agreement between the values that are accepted either as conventional true val￾ues or an accepted reference value and the value found. Accuracy is usually reported as percent recovery by assay, using the proposed analytical proce￾dure, of known amount of analyte added to the sample. The ICH also rec￾ommended assessing a minimum of nine determinations over a minimum of Concentration Response 10 0 2 0.00 5.00 10.00 15.00 4 6 8 Figure 2.2. Linearity with correlation coefficient greater than 0.997. Peak area 18,000 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 Concentration (µg/mL) 0 50 100 150 200 250 Figure 2.3. Linearity with correlation coefficient less than 0.997