Updating the European Pharmacopoeia impurity profiling method for terazosin and suggesting alternative columns

This work was motivated by the demand of European Directorate for the Quality of Medicines and HealthCare (EDQM). A new liquid chromatographic (LC) method was developed for terazosin impurity profiling to replace the old European Pharmacopoeia (Ph. Eur.) method. This new method is published as part of the new Ph. Eur. monograph proposal of terazosin in Pharmeuropa issue 32.2. The aim of the method renewal was to cut the analysis time from 90 min (2 × 45 min) down to below 20 min. The Ph. Eur. monograph method is based on two different chromatographic separations to analyze the specified impurities of terazosin. The reason for the two methods is that two of the impurities are not sufficiently retained in reversed phase (RP) conditions, not even with 100% water as eluent. Therefore, next to RP, an ion-pair (IP) chromatographic method has to be applied to analyze those two impurities. With our new proposed method it was possible to appropriately increase the retention of the two critical compounds using alternative stationary phases (instead of a C18 phase which is suggested by the Ph. Eur. method). Applying a pentafluoro-phenyl (PFP) stationary phase, it was feasible to separate and adequately retain all the impurities. The detection wavelength was also changed compared to the Ph. Eur. method and is now appropriate for the detection and quantification of all impurities using perchloric acid in the mobile phase at low pH. Another goal of the present study was to develop a generic workflow and to evaluate the chromatographic resolution in a wide range of method variables and suggest some replacement columns for terazosin impurity profiling. Retention modeling was applied to study the chromatographic behavior of the compounds of interest and visualize resolution for the different columns, where a given criterion is fulfilled. A zone (set of chromatographic conditions) of a robust space could be then quickly identified by the overlay of the individual response surfaces (resolution maps). It was also demonstrated that two columns from different providers (Kinetex F5 and SpeedCore PFP) can be used as replacement columns, providing sufficient resolution at the same working point and a high degree of robustness.