Chemistry Reference
In-Depth Information
tAble 4.1
determination of method Accuracy/recovery and precision
Accuracy/recovery
sample concentration
replicate 1
replicate 2
replicate 3
1
98.93%
98.89%
98.54%
2
99.05%
98.55%
98.50%
3
98.88%
98.72%
98.14%
Mean
98.69%
Standard deviation
0.28%
Relative standard deviation
0.28%
Acceptance criteria
Accuracy (mean)
Precision (RSD)
98%-102%
≤2.0%
Assessment
Pass
Pass
determined by comparison to a second validated method, if one exists. In spiking
experiments and the preparation of QC samples, reference standards or standard
reference materials (SRMs) must be used. A reference standard is a highly puri-
fied compound that is well characterized. Because chromatographic methods rely
heavily on reference standards to provide accurate data, the quality and purity of
the reference standard should be well documented. There are two categories of ref-
erence standards: USP reference standards that do not need characterization, and
noncompendial standards. Noncompendial standards must be of the highest purity
that can be obtained by reasonable effort and should be thoroughly characterized
to ensure their identity, strength, quality, and purity. Reference standards are not
always available for every analyte of interest. In the absence of other information,
it may be necessary to calculate the amount of an impurity based on a comparison
of its response to that of the drug substance; the ratio of responses of equal amounts
of the impurity and the drug substance (response factor) should be used if known.
In other words, it is perfectly acceptable to quantitate an impurity against the drug
substance if a suitable impurity reference standard is not available. However, it is
highly recommended that authentic standards be either synthesized or isolated for
all analytes of interest.
There are, in some instances, SRMs of a specific analyte in a known sample
matrix, such as mercury in tuna fish or insulin in a biofluid. These types of SRMs
are often commercially available from various sources. Such SRMs come with a
Certificate of Analysis (CoA), guaranteeing that the reference material contains a
guaranteed amount or level of analyte in a known sample matrix. However, it is not
unusual to find that the exact SRMs to meet all expected needs are not always avail-
able. Sometimes, SRMs must be prepared in-house or through an outside vendor or
contractor, and often require more characterization than the drug substance or prod-
uct itself. Modern chromatography data systems are capable of documenting, track-
ing, or trending accuracy using control charts of the type illustrated in Figure 4.2.
Control charts of this type can be used to easily observe data that falls out of trend,
above or below set predetermined control limits.
