The preferred method of analysis for a stability indicating assay is reverse-phase high-performance liquid chromatography
(HPLC). Reverse-phase HPLC is preferred for several reasons, such as its compatibility with aqueous and organic solutions,
high precision, sensitivity, and ability to detect polar compounds. Separation of peaks can be carried out by selecting appropriate
column type, column temperature, and making adjustment to mobile phase pH. Poorly-retained, highly polar impurities should
be resolved from the solvent front. As part of method development, a gradient elution method with varying mobile phase composition
(very low organic composition to high organic composition) may be carried out to capture early eluting highly polar compounds
and highly retained nonpolar compounds. Stressed samples can also be screened with the gradient method to assess potential
elution pattern. Sample solvent and mobile phase should be selected to afford compatibility with the drug substance, potential
impurities, and degradants. Stress sample preparation should mimic the sample preparation outlined in the analytical procedure
as closely as possible. Neutralization or dilution of samples may be necessary for acid and base hydrolyzed samples. Chromatographic
profiles of stressed samples should be compared to those of relevant blanks (containing no active) and unstressed samples
to determine the origin of peaks. The blank peaks should be excluded from calculations. The amount of impurities (known and
unknown) obtained under each stress condition should be provided along with the chromatograms (full scale and expanded scale
showing all the peaks) of blanks, unstressed, and stressed samples. Additionally, chiral drugs should be analyzed with chiral
methods to establish stereochemical purity and stability (11, 12).
The analytical method of choice should be sensitive enough to detect impurities at low levels (i.e., 0.05% of the analyte
of interest or lower), and the peak responses should fall within the range of detector's linearity. The analytical method
should be capable of capturing all the impurities formed during a formal stability study at or below ICH threshold limits
(13, 14). Degradation product identification and characterization are to be performed based on formal stability results in
accordance with ICH requirements. Conventional methods (e.g., column chromatography) or hyphenated techniques (e.g., LC–MS, LC–NMR) can be used in the identification and characterization of the degradation products. Use of these techniques can provide
better insight into the structure of the impurities that could add to the knowledge space of potential structural alerts for
genotoxicity and the control of such impurities with tighter limits (12–17). It should be noted that structural characterization of degradation products is necessary for those impurities that are
formed during formal shelf-life stability studies and are above the qualification threshold limit (13).
Various detection types can be used to analyze stressed samples such as UV and mass spectroscopy. The detector should contain
3D data capabilities such as diode array detectors or mass spectrometers to be able to detect spectral non-homogeneity. Diode
array detection also offers the possibility of checking peak profile for multiple wavelengths. The limitation of diode array
arises when the UV profiles are similar for analyte peak and impurity or degradant peak and the noise level of the system
is high to mask the co-eluting impurities or degradants. Compounds of similar molecular weights and functional groups such
as diastereoisomers may exhibit similar UV profiles. In such cases, attempts must be made to modify the chromatographic parameters
to achieve necessary separation. An optimal wavelength should be selected to detect and quantitate all the potential impurities
and degradants. Use of more than one wavelength may be necessary, if there is no overlap in the UV profile of an analyte and
impurity or degradant peaks. A valuable tool in method development is the overlay of separation signals at different wavelengths
to discover dissimilarities in peak profiles.
Peak purity analysis.
Peak purity is used as an aid in stability indicating method development. The spectral uniqueness of a compound is used to
establish peak purity when co-eluting compounds are present.
Peak purity or peak homogeneity of the peaks of interest of unstressed and stressed samples should be established using spectral
information from a diode array detector. When instrument software is used for the determination of spectral purity of a peak,
relevant parameters should be set up in accordance with the manufacturer's guidance. Attention should be given to the peak
height requirement for establishing spectral purity. UV detection becomes non linear at higher absorbance values. Thresholds
should be set such that co-eluting peaks can be detected. Optimum location of reference spectra should also be selected. The
ability of the software to automatically correct spectra for continuously changing solvent background in gradient separations
should be ascertained.
Establishing peak purity is not an absolute proof that the peak is pure and that there is no co-elution with the peak of interest.
Limitations to peak purity arise when co-eluting peaks are spectrally similar, or below the detection limit, or a peak has
no chromophore, or when they are not resolved at all.
Mass balance establishes adequacy of a stability indicating method though it is not achievable in all circumstances. It is
performed by adding the assay value and the amounts of impurities and degradants to evaluate the closeness to 100% of the
initial value (unstressed assay value) with due consideration of the margin of analytical error (1).
Some attempt should be made to establish a mass balance for all stressed samples. Mass imbalance should be explored and an
explanation should be provided. Varying responses of analyte and impurity peaks due to differences in UV absorption should
also be examined by the use of external standards. Potential loss of volatile impurities, formation of non-UV absorbing compounds,
formation of early eluants, and potential retention of compounds in the column should be explored. Alternate detection techniques
such as RI LC/MS may be employed to account for non-UV absorbing degradants.