What is robustness?

Written by Anindya Ghosh Roy Posted in Method validation

Has it ever occurred to you what would happen if the conditions applied during the development of an analytical method would vary? In this article we will address the same question through the parameter called robustness, which can be evaluated during method validation if not yet done earlier e.g. during method qualification.

Robustness is the evaluation of an analytical method wherein the results obtained are found to be reliable even when performed in a slightly varied condition. It is the ability of a method to remain unaffected when slight variations are applied. A more general example of this parameter would be to imagine a footballer practising spot kicks using certain technique under given conditions. Now the same footballer is asked to take the kicks using a different sized ball, smaller target, higher atmospheric temperature or different boots. If the results are similar (statistically), the technique could be termed as robust.

The term robustness might be different from the term ruggedness, depending on the definition. In older literature, ruggedness has been used in the USP chapter <1225> to describe (method) variations known as “intermediate precision” as defined by the ICH Q2(R1) guideline. In the current USP chapter this is harmonized. Others propose to use the term ruggedness when analyzing external factors (like different analyst and instruments, which points to the same direction as intermediate precision) and robustness for internal factors characterizing method stability (like e.g. mobile phase flow rate or column temperature of HPLC methods). As the ICH Q2(R1) doesn’t discriminate in such a way and gives a proper explanation about what is “intermediate precision” and “reproducibility” all other variable factors influencing method performance might be considered as robustness parameters.

Figure: (left) Footballer practices spot kicks under given conditions. (right) The same footballer tries to repeat the kicks under slightly varied conditions; a bigger sized ball, a sunnier day and higher temperature.

 

For analytical methods, robustness depends on the type of analytical method under study. The method validation guideline ICH Q2(R1) as well as the ZLG’s Aide mémoire AiM 07123101 has clearly defined the type of variations expected for specific methods:

In the case of liquid chromatography, examples of typical variations are:

  • influence of variations of pH in a mobile phase;
  • influence of variations in mobile phase composition;
  • different columns (different lots and / or suppliers and / or of different age);
  • temperature;
  • flow rate;
  • potentially a different wavelength of the UV detector.

In the case of gas-chromatography, examples of typical variations are:

  • different columns (different lots and/or suppliers);
  • temperature;
  • flow rate.

The determination of the parameters to be examined for robustness can be risk-based.

 

It must be understood that liquid chromatography techniques such as HPLC are often quite sensitive to slight changes. A slightly higher viscous mobile phase or a lesser diameter of the column can lead to pressure fluctuations, non-consistent separations, and in turn cause damage to the analytical column or “just” affect the results. Similarly, change in the column temperature can affect the viscosity of the flow rate and thus influence the results by e.g. different retention times.

Ideally, robustness refers to changes made to the method within the same laboratory. However, robustness can also be described as the feasibility to reproduce the analytical method in different laboratories (then it is part of the parameter reproducibility).

As mentioned earlier, robustness is not considered as a validation parameter in the strictest sense because usually it is already investigated during method development, once the method is (at least partially) optimized. Hence, evaluation of robustness during development makes sense as parameters that affect the method can be identified easily when manipulated for optimization purposes. Keep in mind, that particularly critical materials such as the column type used in a HPLC method or precast gels in a SDS-PAGE from a certain supplier might be no longer available one day in the future. It may therefore be advisable right from the beginning to consider such materials from (two) different suppliers during initial validation. The Aide-mémoire AiM 07123101 of the ZLG also mentions that in addition to sample processing (e.g. influence of temperature and time during extraction or dissolution processes), the stability of the analyte over the measurement time, as well as the exclusion of absorption and adsorption effects on filters, vessel walls, tubings etc. can be examined in the context of robustness.

One good way to analyse robustness was shown by M. Jimidar et al. during the studyMethod Validation and Robustness Testing of an Enantioselective CE Method for Chemical Quality Control”. The authors identified all possible factors that in principle could affect the method and ran robustness studies on them (shown below):

  Factor Unit Limits Level (-1) Level (+1) Nominal
 1  Concentration of cyclodextrin in the buffer electrolyte  mg / 25 mL ± 10 mg 476 496 486
 2  Concentration of the buffer  mg / 100 mL ± 20 mg 870 910 890
 3  pH of the buffer  - ± 0.2  2.8 3.2 3.0
 4  Injection time s ± 0.5 s 2.5 3.5 3.0
 5  Column temperature °C ± 2°C 18 22 20
 6  Rinse time solvent 1 (water) min ± 0.2 min  1.8 2.2 2.0
 7  Rinse time solvent 2 (buffer electrolyte) min  ± 0.2 min  3.8 4.2 4.0

In this case the method was found to be very robust and was successfully transferred to different operational laboratories in Europe, USA, Japan, and China.

 

In conclusion, robustness is a strong parameter used during the development and validation of analytical methods that proves the functionality of the method under slightly different conditions. Additionally, a robust method is easier to transfer.