A method for automated quantitative analysis of bupropion and its major metabolites in serum by high performance liquid chromatography (HPLC)

Introduction: Interpreting an individual patient's drug concentration in relation to a therapeutic and a dose-related reference range is suitable for controlling compliance, lack of clinical response, adverse effects at recommended doses, drug interactions and genetic variations of metabolism. Therefore it is an important contribution to complement pharmacovigilance programs [1–2]. Bupropion is an antidepressant and smoking cessation aid. It acts as a dopamine and norepinephrine reuptake inhibitor, as well as α₃β₄-nicotinic receptor antagonist. Bupropion is extensively metabolized. Three pharmacologically active metabolites were identified in human plasma; hydroxybupropion, threo-hydrobupropion and erythro-hydrobupropion. We aim to establish a simple method for automated quantitative analysis of bupropion and its major metabolites in serum using column-switching high performance liquid chromatography (HPLC). Several routine methods were evaluated in reference to their prospective capacity to determine bupropion and its major metabolites, and UV-spectra of bupropion, hydroxybupropion, threo-hydrobupropion and erythro-hydrobupropion were recorded to determine the ideal detection wavelengths. Methods: The chromatographic analyses were performed on a Dionex system with Betasil C6 (250 × 4,6 mm, 5 µm, Method 1 and 2) or PerfectSil 120 ODS-L (250 × 4,6 mm, 5 µm, Method 3) analytical columns. The mobile phase consisted of 0,04 M KH₂PO₄/acetonitrile (pH 3.0) for Method 1, 0,02 M HCOOCH₃/acetonitrile (pH 3.0) for Method 2, and acetonitrile/MeOH/TEMED/H₂O mixture at pH 6.5 for Method 3. For sample clean-up we used online-extraction where serum was injected onto a LiChristopher ADS RP-4 precolumn and retained analytes were eluted by back-flush flow onto the analytical column. Results and Conclusion: Bupropion and hydroxybupropion are easily separated by all tested methods. The separation of threo-hydrobupropion and erythro-hydrobupropion was better for the low pH measurements (Method 1 and 2) but insufficient for those under neutral pH conditions (Methods 3). However, Method 1 and Method 2 yielded almost identical retention times for erythro-hydrobupropion and bupropion. As for the detection wavelengths, bupropion has absorption maxima at 210 nm and 254 nm. The UV-absorption spectra of hydroxybupropion, threo-hydrobupropion, and erythro-hydrobupropion all show absorption maxima at 211 nm and only very weak absorption at higher wavelengths. Thus, a bupropion peak measured at 254 nm would be hardly affected by underlying metabolite peaks. The method development is still in progress. Mobile phase components that are nontransparent above 210 nm – e.g. carbon acids – need to be replaced. The separation problem for threo- and erythro-hydrobupropion is yet to be solved. References: [1] Baumann P, Hiemke C, Ulrich S, et al. The AGNP-TDM expert group consensus guidelines: Therapeutic drug monitoring in psychiatry. Pharmacopsychiatry. 2004 Nov; 37(6): 243–65. [2] Haen E, Greiner C, Bader W, Wittmann M. Wirkstoffkonzentrationsbestimmungen zur Therapieleitung: Ergänzung therapeutischer Referenzbereiche durch dosisbezogene Referenzbereiche. Nervenarzt 2008 May; 79(5): 558–66.