Comparison of Formulation Analysis by UPLC FINAL
- 1. Comparison of Formulation Analysis by UPLC/MS and UPLC/UV Jessica Sitko1, Kim Navetta2, Russell Drago2, Anthony Perretta2, Jennifer Colangelo2 Colgate University, Hamilton, NY1, Drug Safety Research and Development, Pfizer, Groton, CT2 Timely and accurate formulation analysis is critical for all drug safety studies to ensure that the test animals receive the intended dose. In the past year, the number of GLP samples sets for formulation analysis in the analytical lab has doubled and there has been a 3-fold increase in formulations being out of specifications. The increase in out-of-specification formulations more than triples the number of samples received for analysis, therefore, there is a great need to increase efficiencies. Currently, formulation samples are analyzed using ultra high pressure liquid chromatography (UPLC) coupled with ultraviolet (UV) detection. Mass spectrometers (MS) could be used as alternative detectors and are widely used in biomarker, metabolism and clinical pathology labs. In general, MS assays are faster to develop, more specific, more sensitive and have fewer problems with interferences. Chromatography can be simplified in MS assays, resulting in little or no need to make complex mobile phases. The purpose of this project is to bring on-line a simple MS assay using the newly purchased single quadrupole mass spectrometer and measure the formulations submitted to the analytical lab over the summer. The results will be compared to those obtained with the UV method to determine if the MS assay produces comparable results. 1. This experiment used a Waters Acquity Classic UPLC with PDA detector (UPLC/UV) and a Waters Acquity I-Class UPLC with a QDa detector (UPLC/MS) (Figure 1). 2. Different mobile phases were compared to determine which produced chromatographic peaks that met system suitability and linearity requirements with each compound on the UPLC/UV and UPLC/MS (Table 1). 3. Working standards were made from stock solutions and analyzed by UPLC/UV, then analyzed by UPLC/MS. 4. Peak integration was performed using Empower 2 Build 2154 to determine system suitability. 5. System suitability, accuracy and precision, and linearity results were compared between the two systems. 6. Individual standard curves for each compound assessed were constructed by plotting the peak area versus concentration of five standards ranging from 0.001 mg/mL to 0.02 mg/mL. Slope and y-intercept were calculated using a linear fit with no weighting. The concentration of drug in vehicle was calculated relative to the regression line. Figure 1. The UPLC/MS system Chromatogram shown is from a cisapride standard at 0.01 mg/mL obtained with UPLC/MS and the generic method. Figure 2. Representative Chromatogram Produced by UPLC/MS is Comparable to Chromatograms Produced by UPLC/UV Conclusions • The generic method developed and tested on 8 different compounds demonstrated that assay development can be simplified by using the same mobile phase and column as a starting point and only modifying the column temperature and gradient to achieve the required capacity factor and peak shape. Time and resource gains are achieved because scientists will not need to identify, purchase, maintain and test different HPLC/UPLC columns, mobile phase additives, buffers and modifiers. • The UPLC/MS system produced lower limits of quantitation ranging from 5 to 200 fold when compared to the results from the UPLC/UV. Enhanced detection limits can be critical for high potency compounds since they are generally used at lower concentrations. Lower detection limits also allow scientist to further dilute the low dose formulations if a vehicle effect is observed. Obstacles • Data processing with Empower was challenging due to a newly discovered software bug and batch processing did not work on most UPLC/MS data sets. Calculations for these data sets had to be done manually. • Signals detected with UPLC/MS were often so high that carry over to subsequent injections became an issue. To address this, all standards from compound 1, compound 8, propafenone, quinidine, and cisapride were diluted by a factor of 10 and re-injected. Future Experiments • Run more compounds on both systems using the generic method. • Investigate why the signal is suppressed by vehicle in some instances and amplified in others. • Compare total time spent on one formulation analysis for a new compound, including methods development, using UPLC/UV to total time spent using the generic method and UPLC/MS. Acknowledgments Carol Fritz, Lina Luo, Annie LoGuidice, and Jiri Aubrecht Results MethodsProject Rationale Methods MS assay produced results comparable to those obtained by the PDA system, determined by system suitability1 and linearity2 requirements. 0.2 mg of compound was weighed and diluted to concentrations of 0.01 mg/mL (STD A), 0.005 mg/mL (E1 and E2), 0.015 mg/mL (F1 and F2). A check mark indicated that the requirement was met, while an x indicates that the requirement was not met. We expect that the instances in which requirements were not met were not caused by the assay, but from weighing or dilution errors. We are working to resolve these issues. Table 2. UPLC/MS System Suitability and Linearity Compared to UPLC/UV Utilizing the New Generic Method Described Compound Requirements UV MS UV MS UV MS UV MS UV MS UV MS UV MS UV MS drug peak retention time precision1 RSD ≤ 5% peak area precision1 RSD ≤ 2% capacity factor (k')1 ≥ 2.0 resolution1 ≥ 1.5 NA NA NA NA NA NA NA NA NA theoretical plates1 ≥2000 tailing factor1 0.7 ≤ T ≤ 2.0 standard agreement: A and B ± 2% standard agreement: A and C ± 2% correlation ceofficient, R2 ≥0.99 recovery factor2 %RSD ≤ 5% 1 flecainide cisapride2 3 8 propafenone quinidine Addition of vehicle to the formulation standards affects peak area. The addition of vehicle caused signal suppression in some cases, while it caused signal amplification in others. Future experiments should be conducted to determine how different vehicles interact with the mass spectrometer. (Peak area with 20 % vehicle/peak area without vehicle * 100) Table 3. % Agreement Between UPLC/MS Peak Area With and Without Vehicle compound 0.001 mg/mL 0.01 mg/mL 1 98 100 2 no data 73 3 97 100 8 146 136 propafenone 102 112 quinidine 93 141 flecainide 101 102 cisapride 174 110 The UPLC/MS system shows lower detection limits than the UPLC/UV system. Limits of detection were defined as the point at which a peak was seen with a signal-to-noise ratio of 10:1. Table 4. Detection Limit Comparison Compound UPLC/UV (pg on column) UPLC/MS (pg on column) 1 100 2 2 50 1 3 100 20 8 100 1 propafenone 100 1 flecainide 200 1 quinidine 200 5 cisapride 100 1 A generic method using 0.1% formic acid in water as mobile phase A and acetonitrile as mobile phase B, a Waters Acquity BEH C18 1.7µm (2.1 x 100 mm) column, a 3-4 minute run time, and injection volume of 1 or 2 µL was used with all 8 compounds. Using this method for all or most compounds and only adjusting the gradient and/or column temperature will greatly decrease the amount of time spent in methods development Table 1. UPLC/UV and UPLC/MS Conditions Pfizer Compounds 1 2 3 8 All Compounds Conditions Original Original Original Original New Generic Method HPLC/UPLC column Waters CSH C18 1.7µm Phenomenex Kinetex C18 1.3µm Waters BEH C18 1.7µm Waters CSH C18 1.7µm Waters UPLC BEH C18 1.7µm Column dimensions 2.1x100 mm 2.1x50 mm 2.1x100 mm 2.1x100 mm 2.1 x 100 mm run time 4 min 4 min 4 min 3 min 3 to 4 min mobile phase A 50 mM KH2PO4 50 mM KH2PO4 50 mM KH2PO4 50 mM KH2PO4 0.1% formic acid in water mobile phase B Acetonitrile Acetonitrile Acetonitrile Acetonitrile Acetonitrile column temp 40°C 45°C 45°C 50°C 40° - 50°C sample temp 22°C 22°C 22°C 22°C 22° C flow rate 0.4 mL/min 0.5 mL/min 0.4 mL/min 0.4 mL/min 0.400 - 0.500 mL/min injection volume 2 µL 1 µL 2 µL 2 µL 1 - 2 µL wavelength 254 nm 246 nm 277 nm 345 nm same as original molelcular weight 390.1 342.5 286.1 538.9 same as original Safety Pharm Tool Compounds propafenone quinidine flecainide cisapride All Compounds Conditions Original Orignial Original Original New Generic Method HPLC/UPLC column Prodigy Phenyl 3 Shim-Pack XR-ODS Zorbax TMS ODS Hypersil Waters UPLC BEH C18 1.7 µm Column dimensions not defined 2.0 x 75 mm not defined 150 x 4.6 mm 2.1 x 100 mm run time 10 min 10 min not defined 16 min 3 to 4 min mobile phase A 0.1% formic acid 0.1% formic acid in water 1% acetic acid in 0.01 M pentanesulfonate 0.05 M Na2HPO4, pH 8.4 0.1% formic acid in water mobile phase B 0.1% formic acid in MeOH 95% ACN with 0.1% formic acid acetonitrile acetonitrile acetonitrile column temp not defined not defined not defined not defined 40° - 50°C sample temp not defined not defined not defined not defined 22° C flow rate 0.200 - 0.400 mL/min 0.5 mL/min not defined 1 mL/min 0.400 - 0.500 mL/min injection volume 10 µL 20 µL not defined 20 µL 1 - 2 µL wavelength 254 nm 330 nm 308 nm 272 nm same as original molelcular weight 342.2 325.1 415.6 466.1 same as original UPLC/MS produced results as precise as UPLC/UV, but did not produce results as accurate as UPLC/UV. The reason for this is currently unclear, but could be due to dilution errors or compound degradation. Figure 3. UPLC/MS Accuracy Compared to UPLC/UV