(T1030-06-31) Strategies for Correcting Peak Fronting of Oxycodone Hydrochloride, Naloxone Hydrochloride and Related Substances Observed in Reversed—Phase Liquid Chromatography

Purpose: There is no published method for simultaneous determination of oxycodone hydrochloride (HCl) and naloxone HCl in presence of their impurities. In this work, we developed and validated a reversed-phase ultra-performance liquid chromatographic (RP-UPLC) method to assay oxycodone, naloxone, and their impurities (oxycodone N-oxide, naloxone N-oxide, noroxymorphone and noroxycodone) in oxycodone HCl and naloxone HCl extended-release tablets. During the method development phase, we addressed the issue of peak fronting¹ which prevents baseline separation and accurate peak area calculation. Peak fronting appears in many pharmaceuticals that can undergo keto-enol tautomerization¹. Therefore, we investigated the impact of the use of ion pairing agents with different alkyl chain lengths and the role of common mobile phase solvents in improving the peak shape and resolution.

Methods: Chromatography was done on a Waters® AcQuity system equipped with Waters^TM Ultraviolet (UV) Detector at 245 nm using an XBridge BEH C18, 2.5 mm, 2.1x100 mm column. A sample of manipulated tablets containing oxycodone HCl (40 mg) and naloxone HCl (20 mg) was used in the method development together with impurities that have close relative retention time with oxycodone and naloxone: oxycodone N-oxide, naloxone N-oxide, noroxymorphone and noroxycodone. Oxycodone HCl and naloxone HCl reference materials were obtained from United States Pharmacopoeia (USP) and the impurities were obtained from Cerilliant®. A gradient elution using a ternary system of phosphate buffer pH 2.6, methanol and acetonitrile was used. Three different ion pairing agents were used: sodium salts of 1-heptanesulfonate, 1-octanesulfonate 1-nonanesulfonate were used. The flow rate was set at 0.85 mL/min, column temperature at 40°C and the injection volume was 5 μL. The total run time was 15 minutes. The method validation parameters, specificity, repeatability, intermediate precision, injector precision, accuracy, linearity, range, the limit of detection (LOD) and limit of quantitation (LOQ) were determined. Resolution and symmetry factor were calculated to evaluate the peaks quality.

Results: Peak fronting was not only exhibited by oxycodone but also observed in other compounds having a similar structure such as morphine. Previous reports showed that compounds capable of enol-keto tautomerism exhibit fronting peaks in RP-UPLC. The approaches described in the literature to fix this problem are as follows: the use of ion pairing agents, controlling column temperature, use of ternary mobile phase and/or adjusting the pH of the mobile phase. The variable approaches aim to push the keto-enol equilibrium to one side and this shift in equilibrium improves the peak shape. During method development and from the literature search of liquid chromatography methods of keto-enol systems, it was found that methanol was critical to obtain symmetric peaks. However, methanol does not provide enough resolution and other organic solvent need to be included (Figure 1). Based on oxycodone HCl and naloxone HCl monographs in the USP and European Pharmacopoeia, critical impurities that elute close to these compounds were used to demonstrate the specificity of the developed method. Improved resolution and method specificity were found with the use of ion pairing agents of a longer side chain length such as 1-heptanesulfonate and 1-octanesulfonate(Figure 2). 1-nonanesulfonate resulted in higher resolution than the aforementioned two ion pairing agents. The resolution between any of the critical impurities was more than 3.03 which means higher selectivity for all impurities. Symmetry factor as a measure of peak fronting or tailing were within the desired acceptable range between 0.95 and 1.1 for oxycodone and any of its related peaks. The detector response was linear over the range of: 1.30 to 156.05 mg/mL for oxycodone HCl and 0.65 to 77.98 μg/mL for naloxone HCl. For oxycodone HCl, LOD was 0.032 mg/mL and LOQ was 0.097 mg/mL. For naloxone HCl, the LOD was 0.087 mg/mL and the LOQ was 0.26 mg /mL. The linearity, range, LOD and LOQ covered the expected concentration of oxycodone HCl and naloxone HCl in the milled tablet product and in dissolution samples. The average recoveries were 100.12% for oxycodone HCl and 101.22% for naloxone HCl. The injection repeatability was 0.44% and 0.49% for oxycodone HCl and naloxone HCl, respectively. The repeatability and intermediate precision were measured over three days and the relative standard deviation (RSD% ) for oxycodone HCl were 1.33%, 1.29% and 0.76% for Day 1 through Day 3. For naloxone HCl the %RSD were 1.32%, 0.81% and 1.25% for Day 1 through Day 3. The RSD of injection repeatability was 0.49% and 0.44% for naloxone and oxycodone, respectively.

Conclusion: An RP-UPLC method was developed and validated for separation of oxycodone HCl, naloxone HCl, and their critical impurities in tablets and in dissolution samples. Out of 3 ion pair agents test, the best chromatographic resolution was obtained with 1-nonanesulfonate which has a longer side chain length. Methanol is essential to correct peak fronting; however, it needs to be combined with a ternary organic solvent to improve resolution.

References: Brogle, K.; Ornaf, R. M.; Wu, D.; Palermo, P. J., J. Pharm. Biomed. Anal. 1999, 19

Acknowledgments: The authors would like to thank April Winemiller from FDA and Jennifer Lo from Biopharma Services Inc for their help to make this work possible. This work was funded by contract number HHSF223201610004I-HHSF22301002T.

Figure 1: Use of methanol is associated with more symmetric peak shape as seen in chromatogram B but low resolution while Use of acetonitrile is associated with better resolution as seen in chromatogram A but asymmetric peak shape.



Figure 2: example of chromatogram showing separation of naloxone N-oxide, oxycodone N-oxide, noroxymorphone and noroxycodone HCl at 1 µg/mL from naloxone HCl :70 µg/mL; oxycodone 140 µg/mL.



Table 1 Repeatability and intermediate precision results for determination of oxycodone HCl and Naloxone HCl 40/20 mg manipulated tablets