The development of analytical methods is a continuous process that accompanies the method throughout its lifecycle, from initial development to implementation in quality control and transfers between laboratories. The traditional trial-and-error approach is time-consuming and limited in detecting interactions between parameters. An alternative is the quality by design approach, which utilizes statistical methods and experimental design experiments and focuses on systematic development, where quality is planned and understood at the beginning of the development process. The 2024 International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use Q14 guideline introduced quality by design principles to analytical method development.
In silico method development relies on mathematical models to predict the effects of chromatographic conditions. In the master's thesis, an HPLC method was developed and tested for the analysis of ibuprofen and ketoprofen, as well as an HPLC method for a formulation containing lidocaine, 2,6-xylidine, methyl paraben and propyl paraben. In this thesis, a systematic method development is presented, including the evaluation of the physicochemical properties of the analytes, setting the requirements for method suitability, selecting initial conditions, performing a scouting gradient, and optimizing the method.
Within the master's thesis, we evaluated the Practical HPLC Simulator tool by examining the hypotheses of matching retention time, column efficiency, and separation factor between in silico predictions and experimentally determined results on the system. The simulator accurately predicted some parameters, such as the separation factor and peak symmetry, but often incorrectly estimated the retention time, with the degree of deviation depending on the individual analyte.
According to the obtained results the simulator is useful for rapidly testing method modifications however, its limited accuracy restricts its use as a primary tool.
We concluded that the simulator has potential as a tool in the initial stages of developing simple methods and as an educational aid. Further research should include a broader range of molecules and investigate the discrepancies between predictions and actual measurements.
|