Once the method has been established and validated, it is ready for analytical use. However, as most current analytical methodologies employ automation to increase productivity, analytical runs have become very long (up to days). Therefore it is necessary ensure that the assay continues to perform according to the specifications determined during the validation stage throughout each analytical run. This is accomplished by making and including quality control samples or calibrators (QC) of known concentrations that can be interspersed with the calibration standards and the clinical samples in each analytical run. The QC samples allow the analyst to monitor the accuracy and precision of the method while it is in use. QC samples are standards that are made of known quantities of drug that is spiked into naïve matrix. A minimum of three concentrations that bracket the standard curve should be prepared. The first QC sample should be within 3xof LLOQ, the second QC sample should be mid-range and the third QC sample at the upper end of the standard curve should be included. The QC samples should be run in replicate. The QC samples should be interspersed with the clinical samples and the standard curve calibrators, but there is no consensus on how frequently QC samples should be incorporated. The FDA recommends that 5% of the samples in the run should be QC samples, but six QC samples are the absolute minimum for any run. Both standard calibrators and QC samples should be arranged to detect assay drift. In order to accept the analytical run, two-thirds of the QC samples must be within 15% of their nominal values. For example, if six QC
samples were analyzed (two low QCs, two mid-QCs, and two high QCs) and one replicate each at the mid- and high QC concentrations were greater or less than 15% of nominal, the run would be deemed acceptable. However, if two replicates at the same QC concentration failed (e.g., both mid-QC samples in the example above), or more than two QC samples failed, the analytical run would be rejected.
In addition to monitoring the method performance, it is also good practice to include QC samples with the samples during storage. QC samples can be prepared at the same time the samples are processed, and stored with the samples to monitor storage conditions. This practice is useful to guard against unforeseeable events, such as a power outage that affects freezer function. This use of QC samples, although advisable, is not a requirement of analytical method validation.
Was this article helpful?