Ultra-high performance supercritical fluid chromatography in impurity control II: Method validation

In our previous study we proposed a screening approach using ultra-high performance supercritical fluid chromatography for the determination of 10 pharmaceutical quality control mixtures. Most of resulting methods offered baseline separation of all analytes. However, some of these methods had to be further optimized to ensure their successful validation and applicability to impurity control in drug substance and drug products. Several challenges occurred during the optimization including: (i) the necessity of the resolution of active pharmaceutical ingredient and following impurity equaling at least 3, which was especially difficult to achieve for mixtures of structurally close compounds, (ii) unrepeatable elution of compounds eluting close to the dead volume or at the end of the gradient elution, and (iii) shifts in retention times due to the column aging and effects of additive. The most frequent optimization adjustments involved changes in gradient program. Other adjustments such as the substitution of Viridis UPC2 HSS C18 SB column with a slightly different Acquity UPLC HSS C18 SB column, the addition of acetonitrile in the modifier, and the column coupling also led to beneficial changes in selectivity. Subsequently, validation of all 10 methods was carried out to prove the applicability of ultra-high performance supercritical fluid chromatography methods for the impurity control in pharmaceuticals. Parameters recommended by ICH guidelines Q2 and Q3 including specificity, linearity, range, lower and upper limit of quantification, limit of detection, accuracy, and precision were examined. In addition, intermediate precision and the accuracy profiles were determined for selected methods. Overall, only two impurities did not meet the validation criteria due to low resolution and low sensitivity, respectively. Only identification threshold and not reporting threshold was met for this impurity. (C) 2020 Elsevier B.V. All rights reserved.