Guidance 003 – Analytical Test Method Validation – System Suitability

System suitability criteria from compendial general chapter methods may be used for some test methods but should be evaluated against the intended use of the test method as to applicability.

Resolution can be calculated between the major component and an internal standard, the major component and an expected impurity found in the reference standard, two impurity peaks in the reference standard (resolution material), or two peaks that are the most difficult to separate (often referred to as “the critical pair”). Since resolution is the primary criterion for specificity and robustness, it serves as a rigorous parameter for suitability. If there is no critical pair to establish a resolution criterion, a retention time window may be used in its place. Minimum criteria should be that there is baseline resolution between the peaks.

Efficiency, as measured by theoretical plates, may be of importance as a measurement of column performance, especially in cases where there are no other peaks with which to perform resolution determinations.

Peak symmetry, as measured by the tailing factor, may be of importance to report, especially in impurity testing. If tailing is too extensive, it may mask other impurities. When the tailing factor increases, integration becomes less reliable and precision can be compromised. Peak fronting may also cause integration problems and may become a factor as columns age.

It is considered that repeatability is normally used as an essential criterion for system suitability testing but this may not be possible for all types of IPC test methods. For example, Area % methods do not require repeatability. System repeatability is regarded as the contribution of the instrumental variability to the precision.

Demonstration of specificity may be required for certain applications and may involve resolution between two significant peaks, peak efficiency by theoretical plates or peak symmetry by tailing factor. It is recommended that the specificity be demonstrated as part of the SST criteria where variability of sample make up is possible (e.g. for a chromatographic method or TLC method, the sample diluent is prepared on day of analysis or may be of a different batch/lot of solvent to that used during validation). It may be of benefit to demonstrate adequate specificity between the diluent (blank solvent) and the sample solution. This is particularly useful when investigating low-level impurities as the detection and attribution of a novel impurity to the sample may be discounted by its presence in the sample diluent.

For non-chromatographic testing, the use of control samples or day/ time of use calibration may sometimes be appropriate for some technologies. Examples include but are not limited to KF, LOD (day/time of use calibration), and particle size (control samples).

Other pharmacopoeias should be consulted if required, however, the US Pharmacopoeia recommends that to determine system suitability % RSD, 5 replicate injections if the % RSD is 2.0 or less and if the % RSD is greater than 2.0, six replicate injections are recommended. While the USP recommends the above % RSDs, these criteria may not be adequately low to assure method performance (e.g. when the %RSD of the assay is 1% or when the specification is tight such as 99.0 – 101.0%). Therefore, it may be relevant to consider using the EP Pharmacopoeia recommendation for a tighter % RSD of 1/3 of the specification range to have a 95% confidence that the result is within the limit. The EP recommends that system suitability for repeatability is based on the limit range and number of standards used in the test, where n can vary from three to six.