Method validation! The key points and difficulties are all here

Thread Content

I. Basic contents of method validation Method validation basically involves the drafting and approval of a protocol, as well as the qualification of testing instruments. It includes four aspects: suitability verification (including accuracy tests, precision determination, linear range tests, specificity tests, etc.), result evaluation, and approval. II. The basic steps for validating a test method are: first, formulating a validation plan; then, qualifying the large-scale precision instruments. The most critical step is conducting the applicability test of the test method. Finally, the test method is evaluated and approved. 1. Development of the validation plan: The validation plan for testing methods is usually proposed by the quality validation team. By reviewing relevant information based on the product’s manufacturing conditions, the chemical structure of raw materials and excipients, intermediates, and decomposition products, specification standards are established, inspection items are determined, and limits for impurities are set, thereby forming a draft of the quality standards. The scope of inspections and tests is determined based on the draft quality standards; specific operational procedures for the testing methods are formulated, and only after approval by the relevant personnel can they be implemented. 2. The testing instruments used in the verification, analysis, and testing of large-scale precision equipment can generally be divided into three categories: A. Ordinary instruments: disintegrators, refractometers, analytical balances, pH meters, melting point determiners, conductivity meters, etc.; B. More precise instruments: polarimeters, automatic titrators, Fehling moisture analyzers, automatic droplet titrators, drug dissolution testers, visible spectrophotometers, electrophoresis devices, etc ； C. Large-scale precision instruments: ultraviolet spectrophotometers, infrared spectrophotometers, gas chromatographs, high-performance liquid chromatographs, thin-layer scanners, etc. To ensure the accuracy and reliability of analytical test data, each testing instrument should be validated before being put into formal use. The validation of the testing instruments is the foundation for the verification of the testing method, and it should be completed first before other verification tests begin. The verification tasks for testing instruments should be determined based on the type of instrument. Dependent on technical performance, it usually includes: installation verification, calibration, suitability pre-testing, and re-verification. 3. Calibration is an important part of instrument validation and testing method verification; it should be carried out prior to the validation tests. UV spectrophotometer calibration includes wavelength calibration, absorbance testing, accuracy testing, and stray light inspection. Both gas chromatographs and high-performance liquid chromatographs require system suitability tests. Determine the minimum theoretical number of plates of the chromatography column under the specified chromatographic conditions. Separation degree and tail factor are specified, with the coefficient of variation required to be no more than 2%. The measuring instruments used in chemical analysis, including volumetric flasks, pipettes, burettes, and analytical balances, should also be calibrated. 4. After the installation of the suitability pre-test instrument has been confirmed, and its functional tests meet the requirements, suitability checks should be carried out using standard samples or controls to determine whether the instrument meets the usage requirements. For example, a preliminary test for the suitability of a melting point apparatus is conducted using metronidazole, which has a known melting point, and the test results are compared with that known melting point. A UV spectrophotometer can be tested using a standard sample with a known concentration; by comparing the measured results with the known values, it can be determined whether the instrument meets the required specifications. While carrying out the aforementioned testing tasks, it is necessary to properly archive the relevant documents and records; each instrument should have a complete set of archival materials. 5. Re-verification: To ensure that the instrument is in good working condition, after the verification process is completed for a newly purchased instrument, it should be done according to the instrument’s category. A plan for reconfirmation is formulated based on the confirmed experience. The interval and content of reconfirmation should be determined based on the type of instrument and its usage conditions; it is generally 3 months, 6 months, or 1 year. The items included in the reconfirmation of instruments usually involve circuit connections, checks on the consumption of accessories and spare parts, cleaning, functional tests, work logs, etc., with the functional tests during installation confirmation being of particular importance. III. Verification of the applicability of testing methods: The purpose of verifying the analytical methods specified in drug quality standards is to prove that the methods used are suitable for the corresponding testing requirements. When drafting drug quality standards, the analytical methods must be validated; when there are changes in the drug manufacturing process, changes in the components of the formulation, or revisions to the original analytical methods, the analytical methods specified in the quality standards also need to be validated. The method validation process and results should both be documented in the instructions for drafting or revising the drug standard. In accordance with the principle requirements of Appendix *XA of the Chinese Pharmacopoeia (2010 edition), the validation of testing methods can be handled in three different ways ； 1. A method that requires no verification. As per the methods specified in pharmacopoeias (including those from various countries such as USP, EP, CP, JP, etc.), a system suitability test is generally conducted solely to confirm that the system meets the requirements (primarily with regard to instrument stability and the resolution of the column). 2. Comparison method. The analytical methods that have been validated in a reference laboratory can be confirmed for their reliability through comparative tests. Specifically, the data obtained by this laboratory and the reference laboratory using the same method on the same batch of samples are compared (for instance, at least five batches should be selected, with each batch being tested five times). This comparison helps to determine the feasibility of applying the method in this laboratory. In case of any discrepancies, the causes or design solutions must be identified, and the methods must be re-validated. 3. Methods requiring system verification. IV. Evaluation of the test method and approval: After completing the installation qualification and suitability testing, the test data and information should be summarized and analyzed. A proper evaluation of the test method should be made; the description and conclusion sections of the verification report should be concise. The main deviations in the experiment should be appropriately explained. Then, submit it to the supervising leader for approval. The final product of inspection method validation is a validated method—an inspection method formulated based on the results of the validation and approved by the relevant authorities. V. How to calculate the detection limit in detail? There has long been controversy in the international analytical community regarding how to properly or accurately estimate the detection limit. The special significance of the detection limit lies in its ability to accurately evaluate the detection capability of a given analytical method at low concentration levels. The evaluation of an analytical method’s performance at low concentrations can be approached from various perspectives or with different emphases; for instance, it can be assessed based on the ratio of the minimum detectable signal to instrument noise, statistically estimated from the average variability of blank measurements obtained using the method, or quantitatively evaluated according to the deviation characteristics of the calibration curve of the analytical method, among other approaches. The detection limit is an important indicator for evaluating the performance of an analytical method and testing instruments. By \"detection\" is meant qualitative detection, that is, determining the presence of a substance in the sample at a concentration higher than that of the blank. ACS (American Chemical Society) provides a more concise summary of this definition: The limit of detection is the lowest concentration of an analyte that an analytical method can reliably detect. 1. Definition of limit of detection (LOD): The lowest concentration (amount) of the analyte that can be detected in a sample is referred to as the limit of detection. It is the sample concentration at which the generated signal (peak height) reaches k times the standard deviation of the baseline noise. Generally, this corresponds to the concentration at which the signal-to-noise ratio (S/N) is 2:1 or 3:1. There are no specific requirements regarding the accuracy and precision of its determination. Currently, the detection limit is generally defined as the concentration at a signal-to-noise ratio (S/N) of 3:1. 2. The calculation formula is: D=3N/S (1), where N represents noise; S represents the sensitivity of the detector ； D —— Detection limit. The formula for calculating sensitivity is: S = I/Q (2), where S represents sensitivity ； I——Signal response value ； Q —— Injection volume. By combining equations (1) and (2), the following equation is obtained: D = 3N × Q/I (3). In this equation, Q represents the injection volume ； N——Noise ； I——Signal response value. I/N represents the signal-to-noise ratio (S/N) at that injection volume; this ratio can be obtained through automatic analysis of the chromatogram by the workstation, and any standard chromatography or mass spectrometry workstation is capable of performing such S/N analysis. This makes the calculation of the detection limit very convenient. 3. Calculation method: In actual calculations, there are two ways to express the detection limit: one is the detection limit for the sample in the sampling vial, and the other is the detection limit based on the original sample. A. For the first detection limit, it can be calculated as long as the injection volume and signal-to-noise ratio are known. If the sample concentration in the injection vial is 1 mg/L, and the signal-to-noise ratio at this concentration is 300 (as determined by workstation analysis), then the detection limit is: D = (3 × 1 mg/L) / 300 = 0.01 mg/L. It can also be expressed in terms of the absolute injection volume; if the injection volume is 10 ul, then the detection limit is: D = 3×(1 mgL-1×10 ul)/300 = 0.1 ng. B. For the second representation method, both the sampling volume of the original sample and the final volume of the extracted sample need to be taken into account. Still using the aforementioned sample as an example, if the sample weight is 5 grams and the final volume after dilution is 5 mL, then the method’s detection limit is: D = 0.01 mgL-1 × 5 mL / 5 g = 0.01 mg/kg. That is, when the concentration of the substance to be analyzed in the original sample is 0.01 mg/kg, and if the sample volume taken is 5 g with a final volume of 5 mL after preprocessing, the concentration of the sample in the injection vial can reach 0.01 mg/L (assuming a recovery rate of 100%). Under other given analytical conditions, this will result in a signal with a noise level three times higher. In actual testing work, the second representation method is more common. 4. Precautions: As can be seen from Equation (3), the value of the signal-to-noise ratio is directly related to the size of the detection limit. The ratio obtained varies significantly depending on the method used to calculate the signal-to-noise ratio, and this is related to the way in which the peak value of the baseline noise is defined. There are generally three different definitions: A) peak-to-peak signal-to-noise ratio, which uses the average height of a certain segment of the baseline noise ； B. Peak/half-peak signal-to-noise ratio, which is 1/2 of the average height of the baseline noise over a certain period ； C. Root Mean Square (RMS) signal-to-noise ratio, calculated using the root mean square value of a certain segment of baseline noise. In addition, the calculated signal-to-noise ratio is also highly dependent on the location of the noise considered; whether noise from the baseline on the side of the signal is used, or noise from a certain length of baseline, results vary significantly, and sometimes the differences can be quite large. Generally, the calculation is done using the noise peak values on both sides of the sample peak. 5. Determination of the detection limit: A. The \"Global Environmental Monitoring System Guidelines for Water Monitoring\" stipulate that, at a confidence level of 95%, the detection limit is defined as the value at which there is a significant difference between the measured value of a sample and the measured value of a sample with zero concentration: L = 4.6Sb. Here, L represents the lowest detectable concentration of the method. Sb —— the standard deviation of blank replicates (intra-batch) with n determinations (more than 20 repetitions). B. The International Union of Pure and Applied Chemistry (IUPAC) specifies that for various optical analysis methods, the following formula can be used for calculation: L = KSb/SL – where L represents the minimum detectable concentration of the method; Sb is the standard deviation of multiple blank measurements (in terms of absorbance) ； S —— the sensitivity of the method (i.e., the slope of the calibration curve). To conduct an accurate assessment, a sufficient number of blank measurements are required; 20 measurements is ideal. The International Union of Pure and Applied Chemistry strongly advocates for using a k value of 3; generally, this corresponds to a confidence level of around 90%. If =0, it does not mean that the detection limit is infinitely small. At this point, it is necessary to prepare a series of samples with a concentration slightly greater than zero (so as to generate a measurable signal value) to replace the full procedural blank test; the standard deviation of these samples is then used in its place. Furthermore, sometimes for the sake of convenience in work and easier comparison, a signal value that is acceptable to everyone is also designated as the detection limit; for example, in spectrophotometry, the concentration corresponding to an absorbance of 0.010 is set as the detection limit L. C. The limit of detection in the method specified by the U.S. EPA SW—846 is: L=3.143SbD. Some spectrophotometric methods use a concentration value or absolute amount corresponding to an absorbance (after deducting the blank) of 0.010 as the limit of detection; this is an agreement among laboratories. E. Gas chromatography: The minimum amount of substance that must be introduced into the chromatograph for the detector to generate a response signal that can be distinguished from noise is known as the detection limit, which is generally twice the noise level. F. Ion-selective electrode method: The concentration value corresponding to the intersection point where the extended line of the linear portion of the calibration curve intersects the line passing through the blank potential and parallel to the concentration axis is the detection limit of this ion chromatography method using an electrode. The L value calculated from the measured blank value should not be greater than the minimum detectable concentration specified by the analytical method; if it is higher than this specified value, it is necessary to identify the cause and reduce the blank value before conducting further measurements until the result is satisfactory. 6. Preparation of the calibration curve: A. Follow the steps of the analytical method to determine the upper limit of measurement under laboratory conditions, based on the linear relationship between the measured concentrations and the instrument signal values. When the upper limit of determination is lower than the detection limit of the method, only the measured linear range can be used. B. The analytical steps for plotting the calibration curve should be the same as those for sample analysis, and there should be no fewer than 5 concentration values. C. The preparation of the calibration curve is carried out simultaneously with the testing of each batch of samples. For certain analytical methods, the slope of the calibration curve remains stable and the inter-batch error is small; when using the originally prepared calibration curve, 2 standard samples of moderate concentration and 2 blank controls should be tested simultaneously with the samples. The relative deviation between the measured standard concentration (after subtracting the blank) and the concentration value corresponding to the original calibration curve should be less than 5% for spectrophotometry ； The atomic absorption method should be less than 10%; otherwise, the calibration curve must be recreated. VI. Conclusion Finally, it should be emphasized that the detection limit and the lower limit of determination are two distinct concepts; they are important indicators for evaluating the performance of an analytical method and testing instruments. The detection limit is a qualitative concept, whereas the lower limit of determination is a quantitative one. Understanding these two concepts and their relationship can improve the accuracy and reliability of detection results in chromatographic analysis. This is particularly important in trace analysis, where determining the detection limit and the lower limit of determination is crucial for selecting appropriate analytical methods. It also holds great significance for laboratory quality control.