Design of experiments as a tool for LC-MS/MS method development for the trace analysis of the potentially genotoxic 4-dimethylaminopyridine impurity in glucocorticoids

The present study reports on a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method development strategy supported by design of experiments (DoE) for the trace analysis of 4-dimethylaminopyridine (DMAP). The conventional approaches for development of LC-MS/MS methods are usually via trial and error, varying intentionally the experimental factors which is time consuming and interactions between experimental factors are not considered. The LC factors chosen for the DoE study include flow (F), gradient (G) and injection volume (V-inj) while cone voltage (E-con) and collision energy (E-col) were chosen as MS parameters. All of the five factors were studied simultaneously. The method was optimized with respect to four responses: separation of peaks (Sep), peak area (A(peak)), length of the analysis (T) and the signal to noise ratio (S/N). A quadratic model, namely central composite face (CCF) featuring 29 runs was used instead of a less powerful linear model since the increase in the number of injections was insignificant. In order to determine the robustness of the method a new set of DoE experiments was carried out applying robustness around the optimal conditions was evaluated applying a fractional factorial of resolution III with 11 runs, wherein additional factors - such as column temperature and quadrupole resolution - were considered. The method utilizes a Phenomenex Gemini NX C-18 HPLC analytical column with electrospray ionization and a triple quadrupole mass detector in multiple reaction monitoring (MRM) mode, resulting in short analyses with a 10 min runtime. Draw-backs of derivatization, namely incomplete reaction and time consuming sample preparation, have been avoided and the change from SIM to MRM mode resulted in increased sensitivity and lower LOQ. The DoE method development strategy led to a method allowing the trace analysis of DMAP at 0.5 ng/ml absolute concentration which corresponds to a 0.1 ppm limit of quantification in 5 mg/ml mometasone furoate glucocorticoid. The obtained method was validated in a linear range of 0.1-10 ppm and presented a %RSD of 0.02% for system precision. Regarding DMAP recovery in mometasone furoate, spiked samples produced %recoveries between 83 and 113% in the range of 0.1-2 ppm. (C) 2012 Elsevier B.V. All rights reserved.