A stepwise method development strategy has been employed to develop a robust HPLC method to resolve several closely eluting structurally related impurities in an active pharmaceutical ingredient (API). This strategy consisted of automated column screening, optimization of the most critical chromatographic parameters, DryLab modeling, and experimental verification of optimized separation conditions. DryLab was used to predict an optimized gradient profile and separation temperature and these predictions were verified experimentally. A discussion of the accuracy of these predictions is presented. The robustness of the method was verified and the ability of DryLab to predict, with reasonable accuracy, the outcome of such robustness studies was also examined. Once the robustness was established by the DryLab predictions the remainder of the subsequent verification by experiment becomes a simple reiterative exercise. It is critical to adopt a rational strategy, as demonstrated here, to evaluate the interplay of these factors, thereby greatly enhancing method development efficiency.

The current paper presents a novel approach to applying Quality by Design (QbD) principles to the development of high pressure reversed phase liquid chromatography (HPLC) methods using DryLab. Four common critical parameters in HPLC - gradient time, temperature, pH of the aqueous eluent, and stationary phase - are evaluated within the Quality by Design framework by the means of computer modeling software and a column database, to a satisfactory degree. This work proposes the establishment of two mutually complimentary Design Spaces to fully depict a chromatographic method, one Column Design Space (CDS) and one Eluent Design Space (EDS) to describe the influence of the stationary phase and of the mobile phase on the separation selectivity, respectively. The merge of both Design Spaces into one is founded on the continuous nature of the mobile phase influence on retention and the great variety of the stationary phases available.

This paper describes systematic method development strategies used with different instrumental set-ups according to Quality by Design principles for a sample of toxicological interest. Experiments for DryLab modeling were run in an automated fashion on three Shimadzu instruments, and the accuracy of all models was confirmed to be excellent. Generated models were employed to reduce the original method time of 45 minutes to just over 3 minutes. Additionally, DryLab models were shown to successfully emulate experimental changes typically encountered during a method transfer process (dwell volume, extra column volume, flow rate, column dimensions).

This paper builds a bridge between the pioneering work of Csaba Horváth at the dawn of high pressure liquid chromatography during the 1970s and the present state of the art technologies in the field more than four decades later. In recognition of a lifetime of achievement, a small piece in Csaba Horváth's large work together with Imre Molnár is specifically honored. The work emphasizes the importance of understanding retention phenomena, which started with Csaba Horváth’s Solvophobic Theory and nowadays is the base of the ever-growing Quality by Design movement. It is shown how the latest advances in modeling software can be used to gain insight into retention behavior and how this knowledge can be applied to the development of robust reversed phase liquid chromatography methods.

A multifactorial optimization of three critical HPLC parameters is described. Gradient time (tG), temperature (T), and ternary composition (B1:B2) are optimized for the separation of 22 pharmaceutically relevant analytes based on 12 experimental runs. The effects of these variables on critical resolution and selectivity were examined as all three factors were varied simultaneously. Robust conditions for separation were defined using the Design Space and then verified. This paper demonstrates the applicability of this approach to rapid development of high-quality LC methods using DryLab.

This paper describes the use of "knowledge space" and "design space" for Quality by Design of chromatographic methods using DryLab. The authors look specifically at the selection of the proper range of gradient steepness for experimental input data and its effect on predicted results. They demonstrate that the interpolated simulation as well as the extrapolated simulation results, when the initial slopes were close in range to the target slope, provided good predictions with overall percent error difference of experimental retention times of less than 1%.