Robustness testing factors

A tentative list of factors that may be investigated in the robustness test is presented in Table 8. This list is not complete and additional factors may be added. The limits for the factor levels are proposals and should be evaluated case by case. The significance of the effects of the factors on the responses such as the resolution of all peak pairs, the tailing factor, retention times, analysis time, etc., is evaluated. 

A tentative list of factors to be evaluated in robustness testing for sample preparation is presented in Table 9. Additional factors can be added. The limits for the factor levels are typical and should be considered as proposals. Responses to be considered are usually related to the quantitative performance of the method. 

For example, if an HPLC system does not pass system suitability because of a capacity factor (k ) that is too low, the results of robustness testing could tell the analyst that the amount of organic can be changed by 10% without affecting the result. If evaluated at the time of validation, the analyst can make the adjustment with no additional validation. If this particular study has not been performed, the analyst has no other choice but to verify the validity of the method at the new condition. 

AIMS/OBJECTIVES AND STEPS IN A ROBUSTNESS TEST SELECTION OF FACTORS AND LEVELS... 

Initially, robustness testing was performed to identify potentially important factors, which could affect the results of an interlaboratory study.Therefore, the robustness test was executed at the end of the method validation procedure, just before the interlaboratory study. Flowever, a method found to be non-robust should be redeveloped and revalidated, leading to a waste of time and money. For these economical reasons, nowadays, method robusmess is verified at an earlier stage in the lifetime of the method, i.e., at the end of method development or at the beginning of the validation procedure. ... 

A robustness test examines potential sources (factors) of variability in one or more responses of the considered method. In pharmaceutical analysis, chromatographic and electrophoretic separation methods are most commonly evaluated. 

Defining and selecting the factors to be examined should be carefully thought through before starting a robustness test. Usually, the factors in a robusmess test are operational or environmental factors. The former are selected from the operating procedure of the method, while the latter are not necessarily specified in this procedure. Those factors, which are most likely to vary when a method is transferred between different laboratories, analysts or instruments, are selected. 

Table 1 gives an overview of potential factors to be considered when performing a robustness test on a capillary electrophoretic (CE) method. In references 6, 9, 17, and 19, lists of factors for high-performance liquid chromatographic (HPLC), gas chromatographic (GC), and/or thin-layer chromatographic (TLC) methods, can be found. 

TABLE I Potential Factors in the Robustness Testing of Capillary Electrophoretic Methods... 

Suppose a buffer system in CE consists of methanol/buffer 10 90 (v/v). When the methanol fraction is selected as factor in a robustness test, the buffer fraction serves as adjusting component to sum the fractions to 1. 

Qualitative factors are also frequently considered in a robustness test. " For CE methods, factors such as the batch or manufacturer of the capillary, reagent or solvent can be selected. When evaluating the influence of such qualitative factor, the analyst should be aware that the estimated effect is only valid or representative for the examined discrete levels and not for any other level of that factor, and certainly not for the whole population. For example, when examining two capillaries X and Y, the estimated effect only allows drawing conclusions about these two capillaries and not about other capillaries available on the market. Such approach allows evaluating whether capillary Y is an alternative for capillary X, used, for instance, to develop the method. 

In robustness tests, usually the factors are examined at two extreme levels.For mixture-related and quantitative factors, these levels usually are chosen symmetrically around the nominal. The range between the extreme levels is selected so that it represents the variability that can occur when transferring the method.However, specifications to estimate such variability are not given in the ICH guidelines. Often the levels are chosen based on personal experience, knowledge, or intuition. Some define the extreme levels as nominal level +x%. However, this relative variation in factor levels is not an appropriate approach, since the absolute variation then depends on the value of the nominal level. Another possibility is to define the levels based on the precision or the uncertainty, with which... 

For qualitative factors, only discrete values are possible, e.g., capillaries X, Y, or Z. As already indicated, this means that only conclusions can be drawn about the examined capillaries and no extrapolation to other capillaries can be made. The most logical in a robustness test is to compare the nominal capillary with an alternative one. 

In robustness tests, peak measurementfanalysis parameters can also be considered. Such parameters are related to the measurement of the detector signal and they affect responses, such as peak areas, peak heights, retention time, and resolution. They allow improving the quality of these responses. These factors can be found in the data treatment software of an instrument, where often only the default settings are used by the analyst. 

The factors and their levels examined in the robustness testing of some chiral separations are given in Tables 2 and 3. 

TABLE 2 Factors and Their Levels Investigated in the Robustness Testing of the Chiral Separations of Propranolol, Praziquantel, and Warfarin ... 

The analyst should avoid creating impossible factor combinations. This occurs, for instance, when choosing both the batch number and the manufacturer of the capillary as factors in a robustness test by means of a two-level design. It is impossible to define two unique batch numbers that exist for both manufacturers. The way to examine both factors is by using nested designs. ... 

FIGURE 4 Factor level intervals examined during (a) method optimization and (b) robustness testing. 

The two-factor interaction effects and the dummy factor effects in FF and PB designs, respectively, are often considered negligible in robustness testing. Since the estimates for those effects are then caused by method variability and thus by experimental error, they can be used in the statistical analysis of the effects. Requirement is that enough two-factor interaction or dummy factor effects (>3) can be estimated to allow a proper error estimate (see Section VII.B.2.(b)). 

Secondly, with the OVAT approach the importance of interactions is not taken into account. An interaction between two factors is present when the effect of one factor depends on the level of another factor. Since only one factor at a time is varied, the presence or absence of interactions cannot be verified. However, this is not dramatic, since in robustness testing the interaction effects are considered negligible. The evaluation of such interactions is more important in method optimization. 

In references 71 and 72, SST limits are defined based on experience, and the examined responses should fall within these limits. The two papers do not provide much information concerning the robustness test performed. Therefore, it is not evident to comment on the analysis applied, or to suggest alternatives. In reference 73, a graphical analysis of the estimated effects by means of bar plots was performed. In reference 74, a statistical analysis was made in which an estimation of error based on negligible two-factor interaction effects was used to obtain the critical effects between levels [—1,0] and [0,4-1]. 

In reference 88, response surfaces from optimization were used to obtain an initial idea about the method robustness and about the interval of the factors to be examined in a later robustness test. In the latter, regression analysis was applied and a full quadratic model was fitted to the data for each response. The method was considered robust concerning its quantitative aspect, since no statistically significant coefficients occurred. However, for qualitative responses, e.g., resolution, significant factors were found and the results were further used to calculate system suitability values. In reference 89, first a second-order polynomial model was fitted to the data and validated. Then response surfaces were drawn for... 

In reference 20, a typical robustness test is not performed, but a study on the influence of peak measurement parameters is reported on the outcome. The study is special in the sense that no physicochemical parameter in the experimental runs is changed, but only data measurement and treatment-related parameters. These parameters can largely affect the reported results, as shown earlier, and in that sense they do influence the robusmess of the method. The different parameters (see above) were first screened in a two-level D-optimal design (9 factors in 10 experiments). The most important were then examined in a face-centered CCD, and conclusions were drawn from the response surfaces plots. 

In this chapter, the possibilities to set up and treat the results of a robustness test were reviewed (Sections I-VIII). Robusmess usually is verified using two-level screening designs, such as FF and PB designs. These designs allow examining the effects of several mixmre-related, quantitative, and qualitative factors, on one or several responses, describing either quantitative and/or qualitative aspects of the analytical method. 

A check of robustness includes preliminary experiments on precision. During robustness testing, a single standard is repetitively analyzed before starting the actual calibration. Without sufficient precision at a single concentration, it is fruitless to calibrate. However, the RSD obtained from repeated injections underestimates the overall error by a factor of up to 3. ... 

MDOE embraces a number of tools which permit experiments to be conducted in the most efficient possible way, achieving several interesting results such as screening of the important factors, optimization of manufacturing and analytical procedures, minimization of costs and pollution, and robustness testing of products and processes. 

A simple example, focusing on the analytical procedure, will illustrate the type of experimental design used to investigate three key factors in an HPLC method. Detailed discussion of experimental designs for robustness testing can be found in Morgan and Hendriks et Riley and Rosanske provide an... 

Most of these steps are similar to the screening at the beginning of method optimization (see Section 6.4.2). In the following we will highlight the main differences between both. The variables tested in a robustness test could be the same as those screened. However, occasionally additional factors of which it is thought that they could affect the content determination of a method but not the. separation are also examined. Examples are variables related to the sample pre-treatment or the detection. 

The levels selected in a robustness test are different from those at which factors are evaluated in method optimization. For optimization purposes the variables are examined in a broad interval. In robustness testing the levels are much less distant. They represent the (somewhat exaggerated) variations in the values of the variables that could occur when a method is transferred. For instance, in optimization the levels for pH would be several units apart, while in robustness testing the difference could be 0.2 pH units. The levels can for instance be defined based on the uncertainty with which a factor level can be set and re.set 36 and usually they are situated around the method (nominal) conditions if the method specifies pH 4.0, the levels would be 3.9 and 4.1. The experimental designs used are in both situations the same and comprise fractional factorial and Plackett-Burman designs. 

THE FACTORS AND THEIR LEVELS. EXAMINED IN THE ROBUSTNESS TEST OF (82) ON A CHROMATOGRAPHIC METHOD FOR THE SEPARATION AND ASSAY OF A DRUG SUBSTANCE AND TWO RELATED COMPOUNDS IN TABLETS... 

The responses of main interest are different during both applications. In optimization, responses related to the separation of peaks (Section 6.2) are modelled. In robustness testing the quantitative aspect (the content determination) of the method is of most interest, since it is the one that should remain unaffected by small variations in the variables. Responses related to the separation (resolution, relative retention) or describing the general quality of the chromatogram (capacity factors, analysis times, asymmetry factors, and column efficacy) are often also studied. As recommended by the ICH guidelines the results of a robustness test can be used to define system suitability test limits for some of the responses [82]. 

In addition to validation of the automation, full validation of the chromatographic procedure, as described in Chapter 12, should be conducted for late-phase methods. This should include specification of system suitability parameters to ensure that the performance obtained during method development and validation is maintained during routine use. The system suitability parameters may include specification of acceptable injection repeatability, criteria for resolution between critical pairs, maximum allowable tailing factors, and a means of verifying that the requisite sensitivity is obtained. As recommended by Vander Heyden et al., system suitability limits are best set following robustness tests. 

Generally, if robustness is designed into the method development process, the methods should transfer more readily. The successful performance of a test method may be sensitive to the setting of some operational parameters. In robustness testing, a variety of parameters are evaluated to determine the extent to which they can be varied without affecting the performance of the method. In an HPLC experiment, the following representative parameters (factors) may be evaluated ... 

---