Chiral separation of four phenothiazines by nonaqueous capillary electrophoresis and quality by design-based method development for quantification of dextromepromazine as chiral impurity of levomepromazine.

The separation of the enantiomers of mepromazine, promethazine, thioridazine and alimemazine was studied by nonaqueous capillary electrophoresis in the presence of cyclodextrins using 1 M acetic acid and 50 mM ammonium acetate in methanol as background electrolyte. Heptakis(2,3-di-O-acetyl-6-O-sulfo)-ß-cyclodextrin, heptakis(2,3-di-O-methyl-6-O-sulfo)-ß-cyclodextrin (HDMS-ß-CD) and octakis(2,3-di-O-methyl-6-O-sulfo)-γ-cyclodextrin were the most effective chiral selectors for mepromazine, promethazine and alimemazine. Subsequently, a method for the determination of dextromepromazine as chiral impurity of levomepromazine was developed employing quality by design principles. Using HDMS-ß-CD as selector, a fractional factorial resolution V+ design was employed for evaluating the knowledge space, while a central composite face centered design provided further method optimization and the basis for the computation of the design space by Monte Carlo simulations. The final experimental conditions included a 30/40.2 cm fused-silica capillary with 75 µm inner diameter and a background electrolyte composed of 0.75 M acetic acid and 55 mM ammonium acetate in methanol containing 27.5 mg/mL HDMS-ß-CD. The applied voltage was 22 kV and the capillary temperature was 15°C. Following method robustness testing via a Plackett-Burman design, the method was validated for dextromepromazine in the range of 0.01 to 3.0 % relative to a concentration of 0.74 mg/mL levomepromazine and applied to the analysis of reference standards of the European Pharmacopoeia and commercial tablets. The assay also allowed the detection of levomepromazine sulfoxide although the quantitation of the compound was hampered by the poor peak shape of the late migrating diastereomer.

Simultaneous determination of dextromepromazine and related substances 2-methoxyphenothiazine and levomepromazine sulfoxide in levomepromazine on a cellulose tris(4-methylbenzoate) chiral column.

A high-performance liquid chromatography method for the determination of dextromepromazine, levomepromazine sulfoxide and 2-methoxyphenothiazine in levomepromazine samples was developed. The separation of the analytes was achieved within 10 min on a stationary phase containing cellulose tris(4-methylbenzoate) as chiral selector. The mobile phase consisted of 0.1% diethylamine in methanol with a flow rate of 1.0 mL/min. The method was validated according to the International Council for Harmonization guideline Q2(R1). The detection limits based on a signal-to-noise ratio of 3 were in the range of 0.002 to 0.005 µg/mL. The method proved to be precise and accurate in the concentration range of 0.025-1.0 % for levomepromazine sulfoxide and 2-methoxyphenothiazine and 0.025% to 3.0% for dextromepromazine relative to a concentration of 0.1 mg/mL of levomepromazine, with the exception of levomepromazine sulfoxide at the 0.1% level. The method was subsequently applied to the analysis of finished pharmaceutical products as well as of reference substances of the European Pharmacopoeia.

A quality by design-based approach to a capillary electrokinetic assay for the determination of dextromepromazine and levomepromazine sulfoxide as impurities of levomepromazine.

Using a quality by design approach, a capillary electrophoresis method for the simultaneous determination of dextromepromazine and the oxidation product levomepromazine sulfoxide in levomepromazine was developed. The analytical target profile was defined that the method should be able to quantify 0.1% of both impurities with a precision of ≤10%. Hydroxypropyl-γ-cyclodextrin was used as chiral selector. The critical process parameters cyclodextrin concentration, buffer pH and concentration as well as temperature and applied voltage were studied using a fractional factorial resolution V+ design for defining the knowledge space. A central composite face centered design was used as response surface methodology for deriving the design space by Monte Carlo simulations. The selected working point was a 100mM citric acid buffer, pH 2.85, containing 3.6mg/mL hydroxypropyl-γ-cyclodextrin, a temperature of 15°C and a voltage of 25kV. Robustness was estimated using a Plackett-Burman design. The method was subsequently validated in the relative concentration range of 0.1%-1.0% of the impurities for a solution containing 0.25mg/mL levomepromazine. The method was applied to the determination of the purity of the reference substance of the European Pharmacopoeia and of the drug in a commercial injection solution.