以下内容摘自相关书籍，仅供参考。
Analytical Method Validation and Instrument Performance Verification, Edited by Chung Chow
Chan, Herman Lam, Y. C. Lee, and Xue-Ming Zhang
ISBN 0-471-25953-5 Copyright ? 2004 John Wiley & Sons, Inc.
METHOD VALIDATION FOR HPLC ANALYSIS OF RELATED SUBSTANCES IN PHARMACEUTICAL DRUG PRODUCTS
药品中有关物质HPLC方法验证
Y. C. LEE, PH.D.
Patheon YM, Inc.
3.1 INTRODUCTION 介绍
In this chapter we outline the general requirements for analytical method validation for HPLC analysis of related substances in pharmaceutical products. Most of the discussion is based on method validation for pharmaceutical products of synthetic origin. Even though most of the requirements are similar for other types of pharmaceutical drug products (e.g., biopharmaceutical drug products), detailed discussion of method validation for other types of pharmaceutical drug products is outside the scope of this chapter. The discussion focuses on current regulatory requirements in the pharmaceutical industry. Since the expectations for method validation are different at different stages of the product development process, the information given in this chapter is most suitable for final method validation according to the ICH requirements to prepare for regulatory submissions (e.g., NDA). Even though the method validation is related to HPLC analysis, most of the principles are also applicable to other analytical techniques (e.g., TLC, UV).
在本章中，我们主要介绍药品中有关物质HPLC方法验证的通则要求。其中大部分讨论是基于合成药品的方法验证。尽管其它类型药品（例如生物制品）的方法验证要求基本类似，但本章不进行讨论。我们的讨论集中在当前制药行业的法规要求。由于在产品研发过程中，不同阶段对方法验证的期望不同，本章中的信息比较适合于根据ICH要求准备法规注册资料（例如NDA申报）时所进行的最终方法验证。尽管本章讨论的是HPLC方法的验证，但其实大部分原则也适用于其它分析技术（例如TLC和UV）。
3.2 BACKGROUND INFORMATION 背景介绍
3.2.1 Definitions 定义
Definitions for some of the commonly used terms in this chapter are given below.
本章中常用术语定义如下：
? Drug substance (active pharmaceutical ingredient): a pharmaceutical active ingredient.
药用物质（活性药物成分、原料药）：一种药用活性成分
? Related substances: impurities derived from the drug substance and therefore not including impurities from excipients. Related substances include degradation products, synthetic impurities of drug substance, and manufacturing process impurities from the drug product.
有关物质：从原料药衍生出的杂质，不包括辅料中的杂质。有关物质包括降解产物、原料药合成杂质，在制剂生产过程中产生的杂质
? Authentic sample: a purified and characterized sample of a related substance.
认可样品：经过精制和表征确认的某个有关物质的样品。
Unlike reference standards, authentic samples may not be of high purity.
与对照品不同，经认可的样品可能不具有那么高的纯度
However, the purity of an authentic sample has to be determined before use. Authentic samples are used in method development to identify related substances in the analysis. In addition, they are used extensively to prepare the spiked samples in method validation.
但是，经认可的样品的纯度必须在使用前进行检测。认可样品可以用于方法开发阶段，在分析过程中用以识别有关物质，在方法验证被广泛用于制备加样样品。
? Spiked sample: a sample added with a known amount of related substances, prepared from authentic samples during method development or validation.
加样样品：加入了已知数量有关物质的一份样品，在方法开发或验证过程中采用认可样品制备
? Control sample: a representative batch of drug substance (or drug product).
控制样品：原料药（或制剂）具有代表性的批次
Typically, control samples are tested in all analyses to ensure consistency in method performance across different runs. Sometimes, they are used as part of the system suitability test to establish the run-to-run precision (e.g., intermediate precision, reproducibility).
最典型的做法是在每次检测中均对控制样品进行测试，以保证所使用的在不同次的检测中具有一致的表现。有时，控制样品也作为系统适用性测试的一部分用以建立不同次检测中的精密度（例如，中间精密度，重复性）
? Response factor: the response of drug substance or related substances per unit weight. Typically, the response factor of drug substance (or related substance) can be calculated by the following equation:
响应因子：原料药或有关物质每单位重量所产生的响应。一般来说，原料药（或有关物质）的响应因子可以采用以下公式进行计算
响应因子 =
响应值（响应值单位）
浓度（mg/ml）
Relative response factor: the ratio of the response factor of individual related substance to that of a drug substance to correct for differences in the response of related substances and that of the drug substance. It can be determined using the following equation:
相对响应因子：各有关物质响应因子与原料药响应因子的比值，用以校正有关物质与原料药响应之间的差异。可以采用以下公式计算
相对响应因子 =
单个有关物质响应因子
原料药响应因子
If a linearity curve (Figure 3.1) is constructed for both the related substance and the drug substance by plotting the response versus the concentration, the relative response factor can also be determined by
如果有关物质和原料药的响应与浓度之间均能建立各自的线性关系（见图3.1），则相应校正因子也可以采用以下公式计算
相对响应因子 =
有关物质斜率
原料药斜率
图3.1 相对校正因子
3.2.2 Different Types of Related Substance Analysis 有关物质分析的不同类型
Area Percent. In this approach, the level of an individual related substance is calculated by the following equation:
面积百分比（面积归一法）：采用本方法时，单个有关物质的水平使用以下公式计算
有关物质% =
有关物质面积
×100%
总面积
where the area related substance is the peak area of the individual related substance and the total area is the peak area (i.e., response) of the drug substance plus the peak areas of all related substances. This is one of the simplest approaches for related substance analysis because there is no need for a reference standard.
其中，有关物质面积是该有关物质的峰面积，总面积是原料药加上所有有关物质的峰面积（即响应值）。这是有关物质分析中最简单的方法之一，因为它不需要对照品。
This is particularly important during the early phase of the project when a highly purified reference standard is not available. It is the preferred approach as long as the method performance meets the criteria described below.
在项目早期，无法获得高纯度的对照品时，该方法尤为重要。只要方法表现符合以下标准，这种方法其实是可以优先选择的。
Linearity over a Wide Range of Concentration. Since the areas of the related substances (typically, less than 1%) and drug substance (typically, more than 95%) are summed, it is important that the method is linear from the concentration of related substances (e.g., 1%) to that of the drug substance (e.g., 95%). However, in some cases, the peak shape of the drug substance may not be totally symmetrical at such a high concentration. Therefore, the response may not be linear in such a wide concentration range, and the use of area percentage may not be appropriate. If the response of the analyte is nonlinear at higher concentrations, the related substances would be overestimated. Although this is conservative from a safety perspective, it is inaccurate and therefore unacceptable.
在较宽的浓度范围内具有线性。由于有关物质的面积（典型的结果是低于1%）和原料药的面积（典型的结果是高于95%）要进行相加，因此方法在有关物质的浓度（例如1%）至原料药浓度（例如95%）范围内呈线性就很重要。但是，有些情况下，原料药的峰形在浓度较高时可能并不完全对称，这时，响应值可能在较宽的浓度范围内无法呈线性，可能就不适合采用面积百分比方法。如果被分析物质在高浓度时不呈线性，有关物质结果可能会虚高。尽管从安全角度来说这样更保守，但它毕竟是不准确的，因此是不能接受的。
Sample Concentration (Method Sensitivity). To maintain linearity at the concentration range of the drug substance, scientists may try to lower the sample concentration to improve peak shape for the drug substance. However, if the sample concentration is too low, it will affect the method sensitivity, and the ability to detect low levels of related substances may not be adequate.
样品浓度（方法灵敏度）。为了使原料药保持在线性浓度范围内，研发人员可以尝试降低样品浓度，以改善原料药的峰形。但是，如果样品浓度太低，会对方法的灵敏度产生影响，则对低水平的有关物质检测能力可能就不够了。
Response Factor. The response factors of the related substances should be similar to that of the drug substance (i.e., relative response factors close to unity).
响应因子。有关物质的响应因子应与原料药的相近（即相对响应因子接近统一）。
Otherwise, a response factor correction must be used in the calculation.
如果达不到，则在计算中必须代入响应校正因子。
High–Low. This approach can be used to overcome the limitation of linear range in the area percent method discussed above. In this approach, samples are prepared at a concentration (i.e., high concentration) similar to that of the area percent method (Figure 3.2). In addition, the high concentration sample solutions are diluted further, to low concentrations (Figure 3.3). Samples from both high and low-concentration solutions are injected for analysis. In the injections of the high concentration, the responses of all related substances are determined as these small peaks are detectable. The high sample concentration is used to allow all related substances to be detected and quantitated. In the injection of low-concentration sample, the response of the drug substance is determined. Low concentration is used to ensure that the response of the drug substance is within the linearity range.
高-低浓度法（自身稀释对照法）。高-低浓度法可以用来克服上述面积百分比方法中线性范围的缺点。采用高-低浓度法时，样品溶液配制浓度（即高浓度）与面积百分比方法相近（见图3.2），同时，将高浓度的样品溶液进一步稀释至低浓度（图3.3）。将高浓度和低浓度样品溶液分别进样，采用高浓度图谱中的有关物质响应值，因为这时所有这些小峰均可以被检测到。同时，采用低浓度样品图谱中的原料药响应值。高浓度样品溶液的目的是使用所有有关物质均可被检出及定量，而低浓度样品溶液的目的是保证原料药的响应在线性范围内。
图3.2 高浓度图谱
图3.3 低浓度图谱
After dilution, response of the drug substance in the low-concentration sample is similar to that of related substance in the high-concentration sample. Therefore, only a small linearity range is required for this method. In addition, since high sample concentration is used for the determination of related substances, high method sensitivity can be achieved. The limitation of the high–low approach is that each sample is injected at least twice (i.e., high and low concentrations) and the total analysis time will be doubled. In addition, an additional step is required to dilute the high concentration to a low concentration, and dilution error can occur during the second dilution.
稀释后，原料药在低浓度样品溶液中的响应与有关物质在高浓度样品溶液相近。因此，这种方法只需要较小的线性范围。另外，由于高浓度样品溶液用于检测有关物质，这时可以达到较高的灵敏度。高-低浓度方法的局限性是每个样品必须至少进样两次（即高浓度一次，低浓度一次），总的分析时间会加倍。还有，由于需要多一步稀释步骤，在第二次稀释过程中可能会产生稀释误差。
External Standard. In this approach, related substance levels are determined by calculation using a standard curve. The concentration of related substance is determined by the response (i.e., peak area of individual related substance) and the calibration curve. A reference standard of the drug substance is typically used in the calibration. Therefore, a response factor correction may be required if the response of related substance is very different from that of the drug substance. A single-point standard curve (Figure 3.4) is appropriate when there is no significant y-intercept. Otherwise, a multipoint calibration curve (Figure 3.5) has to be used.
外标法。外标法中，有关物质的水平采用标准曲线进行计算。有关物质的浓度由响应值（即单个有关物质的峰面积）和校正曲线决定，因此，如果有关物质的响应因子与原料药的响应因子差异较大的话，可能会需要响应校正因子。如果没有较大的Y轴截距，则可以使用单点标准曲线（图3.4），否则，需要采用多点校正曲线（图3.5）。
Different types of calibration are discussed in Section 3.2.3.
不同类型的校正在第3.2.3部分进行了讨论。
The external standard approach offers several advantages over the area percent method, as discussed below.
外标方法比面积百分比方法具有几个优点，讨论如下：
Reduced Linear Range. Unlike the area percent and high–low methods, which use the response of the drug substance in sample injections for calculation, an external standard method uses a standard curve. Typically, the concentration range of the calibration curve is similar to that of related substances in the sample (e.g., 1 to 5% of the nominal sample concentration). Therefore, this method requires a small linear range.
降低了线性范围。外标方法与面积百分比和高-低浓度方法不同。在高-低浓度方法中，采用的了进样样品中的响应用于计算，在外标法中采用的是标准曲线。一般来说，校正曲线的浓度范围与样品中有关物质的浓度范围相近（例如，1-5%的名义样品浓度）。因此，本方法所要求的线性范围较小。
图3.4 单点校正
图3.5. 多点校正
Improved Method Sensitivity. In this approach, only the responses of individual related substances are used in the calculation. Since the area of drug substance peak in the sample injections is not necessary for the calculation, high sample concentrations can be used without worrying about the off-scale response of the drug substance. This approach is particularly useful when the scientists want to improve the method sensitivity by increasing the sample concentration.
改善了方法灵敏度。外标法中，只需要使用各有关物质的响应进行计算。由于样品图谱中，原料药的峰面积不需要参与计算，因此可以采用较高的样品浓度，而不需要担心原料药响应过载。在研发人员想要通过增加样品浓度提高方法灵敏度时，外标法尤为有用。
Reference Standard. One of the limitations of the external standard method is that a well-characterized reference standard is essential. In addition, each analysis requires accurate weighings of small quantities (e.g., 10 mg) of reference standard. Therefore, weighing error can affect method precision and accuracy.
对照品。外标法的一个受限问题是它需要进行过结构确证的对照品。另外，每次分析均需要对微量（例如10mg）的对照品进行准确称量，因此称量误差可能会对方法的准确性和精密度产生影响。
3.2.3 Suitability of Related Substance Analysis 有关物质分析的适用性
As discussed in Section 3.2.2, linear range is a critical factor for determining the suitable type of related substance analysis. The following are different situations to illustrate the rationales. Typically, the low end of a linearity curve is about 50% of the ICH reporting limit (e.g., 50% of 0.1% = 0.05%). This is to ensure that the method will be able to calculate results accurately below the ICH reporting limit.
在第3.2.2中已讨论过，线性范围是决定哪一种有关物质分析方法适用的关键因素。以下是对不同情形下合理性的讨论。一般来说，线性曲线低端为ICH报告限度的50%（例如0.1%的50%为0.05%）。这是为了保证方法能够精确计算结果的水平低于ICH报告限。
The high end of the linearity curve is the nominal concentration (i.e., 100%).
线性曲线的高端为名义浓度（即100%水平）。
This is the target sample concentration for the drug substance.
该浓度为原料药的目标样品浓度。
Case 1. Linearity demonstrated from 50% of the ICH reporting limit to a nominal concentration of drug substance in the sample solution. In addition, no significant y-intercept is observed (Figure 3.6). In this case, area percent calculation is suitable because the linearity range covers the responses of related substances and that of the drug substance in the sample solution. Therefore, these responses can be used directly to calculate the area percentage of each related substance.
案例1. 样品溶液线性为ICH报告限的50%至原料药名义浓度，未观察到显著的Y轴截距（见图3.6）。这种情况下，面积百分比是适用的，因为线性范围覆盖了有关物质的响应，和样品溶液中原料药的响应。因此，这些响应可以直接用于计算各有关物质的面积百分比。
图3.6 线性：案例1
Case 2. Linearity demonstrated from 50% of the ICH reporting limit to 150% of the shelf life specification of related substance. No significant y-intercept is observed (Figure 3.7). In this case, a high–low calculation is more suitable, as the response is linear only up to the shelf life specification level. Drug substance concentration in sample solution (high concentration) should be diluted to the linear range to obtain the low-concentration solution. Therefore, the response of drug substance in low concentration will be within the linearity range and suitable for calculation. Alternatively, a single-point external standard calibration of concentration within the linearity range can also be used.
案例2. 线性覆盖有关物质ICH报告限的50%至货架期限度的150%，未观察到显著Y轴截距。这种情况下，高-低浓度法会更适当，因为响应的线性仅达到货架期上限水平。供试液（高浓度）中原料药浓度要稀释至符合线性范围的低浓度溶液。此时，低浓度溶液中原料药的响应会在线性范围内，可以用于计算。另外，也可以采用线性范围内的单点外标校正法。
图3.7 线性：案例2
Case 3. Linearity demonstrated from 50% of the ICH reporting limit to 150% of the shelf life specification of a related substance, and a significant y-intercept is observed (Figure 3.8). Due to the significant y-intercept, a single-point calibration (e.g., high–low or one-point external standard calibration) is not suitable.
案例3. 线性覆盖有关物质ICH报告限的50%至货架期的150%，观察到显著的Y轴截距（见图3.8）。鉴于显著的Y轴截距，单点校正法（即，高-低浓度法或单点外标校正法）均不适用。
In this case, multiple-point external standard calibration is the most appropriate.
在此情况下，多点外标校正法是最适当的。
图3.8 线性：案例3
See Section 3.3.3 for more discussion of the significant y-intercept.
关于显著的Y轴截距的讨论参见3.3.3部分。
3.2.4 Preparation before Method Validation 方法验证的准备工作
Critical Related Substances. Critical related substances are those that may exist at significant levels in the drug product. Authentic samples of these critical related substances should be available for method validation. According to the ICH guidelines, all related substances at a level exceeding the identification threshold have to be identified. These related substances should be considered critical and included in the method validation.
关键有关物质。关键有关物质是指那些在原料药中可能存在水平较高的有关物质。在进行方法验证时，应获得这些关键有关物质的认可样品。根据ICH指南，所有超出鉴别阈的有关物质均需要进行确证。这些有关物质应作为关键有关物质，并包括在方法验证中。
To determine the critical related substances, one can review the related substance profile when the drug substance (or drug product) is subject to stress testing. The most significant related substances in stress testing should be considered critical. In addition, significant related substances (i.e., greater than ICH identification threshold) observed in stability studies during product development should also be included in the method validation. The related substance method has to be validated with respect to each critical related substance; therefore, the workload associated with method validation will increase drastically if the number of critical related substances is large.
在决定哪些有关物质是关键的时候，可以对原料药（或制剂）强降解实验中有关物质谱进行审核。在强降解试验中最明显的有关物质应作为关键的有关物质。另外，在产品研发的稳定性试验中观察到的明显的有关物质（即，大于ICH鉴别阈）也应该包括在方法验证中。有关物质方法应针对各关键有关物质进行验证，因此，如果关键有关物质很多的话，方法验证的工作量将会大大增加。
Lower and Upper Concentration Range for Method Validation. The concentration range of related substances is typically related to the targeted quantitation limit (QL) at the low end and the proposed shelf life specification at the high end. Therefore, it is important to have a good estimate of these limits; otherwise, inappropriate concentrations may be used in method validation. Even though ICH proposes a method validation range from the ICH reporting limit to 120% of specification, one would want to extend the range to 50% of the ICH reporting limit to 150% of specification to ensure that the method is suitable for most intended uses. The ICH reporting limit is given in Table 3.1. In general, the quantitation limit should be lower than the corresponding ICH reporting limit.
方法验证的浓度范围。有关物质的浓度范围低端一般与目标定量限有关，高端与货架期限度有关。因此，对这些限度有一个恰当的估计就很重要，否则，可能会在方法验证中采用了不适当的浓度。尽管ICH建议方法验证的范围可以从ICH报告限至限度的120%，但公司可能会将范围扩大至ICH报告限的50%至限度的150%，以保证方法可以适用大部分用途。ICH报告限在表3.1中给出。一般来说，定量限要低于相对应的ICH报告限。
This is to ensure that the method is accurate and precise enough to report results at the level of the ICH reporting limit.
这是为了保证方法的准确和精密性足以保证在ICH报告限水平进行结果报告。
Method Procedure. Since the method procedure is undergoing constant modifications during method development, it is very important to define the procedure before method validation. This will ensure that the same method procedure will be used in all method validation experiments.
检验方法。由于在方法研发中，可能会对方法不断进行修订，因此在进行方法验证之前，对检验方法进行界定就很重要，这样可以保证在所有方法验证试验中均使用了同样的检验方法。
Table 3.1. Various ICH Thresholds Regarding Degradation Products in New Drug Products as Stated in the Current ICH Guidelines Q3B(R)
表3.1 现行ICH指南Q3B(R)中要求的新制剂降解产物ICH阈
最大日剂量1
阈值2
报告阈
≤1g
0.1%
>1 g
0.05%
鉴别阈
＜1mg
1.0%或5μg TDI 3，取低者
1-10 mg
0.5%或20μg TDI，取低者
>10mg-2g
0.2%或2mg TDI，取低者
> 2 g
0.1%
定性阈
＜10mg
1.0%或50μg TDI，取低者
10-100mg
0.5%或200μg TDI，取低者
＞100mg-2g
0.2%或2mg TDI，取低者
＞2g
0.1%
1 日摄入原料药数量
2 基于原料药百分比的阈值
3 TDI 总日摄入量
Critical Experimental Parameters for Robustness. Critical experimental parameters should be identified during method development, and they will be investigated in the robustness experiments.
耐用性关键试验参数。方法开发时要识别出关键的试验参数，在耐用性试验中要对这些关键参数进行调查。
System Suitability Tests. The appropriate system suitability tests should be defined before method validation (e.g., precision, resolution of critical related substances, tailing, detector sensitivity). These system suitability tests should be performed in each method validation experiments. System suitability results from the method validation experiment can be used to determine the appropriate system suitability acceptance criteria.
系统适用性测试。在方法验证开始前，要界定适当的系统适用性试验（例如，精密度、关键有关物质的分离度、拖尾因子、检测器灵敏度）。这些系统适用性试验在每次方法验证试验中均要运行。方法验证试验的系统适用性结果可以用于决定适当的系统适用性可接受标准。
----转自Julia
