In recent years the use of liquid chromatography tandem
In recent years, the use of liquid chromatography/tandem mass spectrometry (LC–MS/MS) has become a preferred approach to measure drugs and metabolites in biological fluids due to its very high selectivity and sensitivity , , , , , , , , . The great sensitivity of LC–MS/MS permits the use of small blood volumes in patients while its specificity allows measurement of metabolites that often co-elute.
We describe herein the development and validation of three sensitive and specific LC–MS/MS assays allowing the determination of major human CYP450 isoenzyme activities following administration of a drug cocktail in healthy volunteers and patients. This seven-drug cocktail is composed of caffeine (CAF), bupropion (BUP), tolbutamide (TOL), omeprazole (OME), dextromethorphan (DM), midazolam (MDZ) and chlorzoxazone (CZX) (which is administered separately) to phenotype for CYP1A2, 2B6, 2C9, 2C19, 2D6, 3A4/5 and 2E1, respectively (Fig. 1). Our objective was to develop highly sensitive, robust and fast assays to determine CYP450 phenotypes in a real clinical setting using small blood volumes and low oral doses of mixed probe drugs in order to minimize risk of side effects.
Materials and methods
Results These methods were validated according to U.S Food and Drug Administration guidelines for bioanalytical methods validation . All probe-drugs tested in our CYP450 cocktail were evaluated for selectivity, sensitivity, linearity, precision, accuracy, recovery, matrix effect, and stability (freeze and thaw cycles, short term, long term, stock and auto sampler stabilities). Probe drugs and their metabolites can be measured in human plasma and urine, except for OME and CZX which were analyzed in human plasma and urine only, respectively.
Discussion The most relevant CYP450 isoenzymes for drug bupropion hydrochloride manufacturer are CYP3A4/5, CYP2D6, CYP2C9, CYP2C19, CYP2B6, CYP1A2 and CYP2E1 since they are heavily involved in the biotransformation of a large number of drugs. Therefore, most of previously developed probe cocktail methods were intended to target most of these isoenzymes although CYP2B6 and CYP2E1 were often not included , , , , , , , . Several multiprobe cocktail methods such as the Pittsburg cocktail, the Karolinska cocktail and the Cooperstown 5+1 cocktail were successfully used to phenotype and investigate drug–drug interactions in vivo; however, their sample bioanalyses were performed by HPLC, needed large volume of samples, lacked sensitivity or showed analytical interference between compounds , , . The GW cocktail reported by Scott et al.  was the first validated cocktail using LC–MS/MS methods to phenotype for six isoenzymes and was composed of caffeine, diclofenac, mephenytoin, debrisoquine, chlorzoxazone and midazolam as probes drugs. Mephenytoin and debrisoquin have been commonly used for phenotyping CYP2C19 and CYP2D6 activities, respectively. However, their administration is non-longer possible since these drugs were withdrawn from the market in most countries. Three cocktails described by Zhang et al. , Turpault et al.  and Yin et al.  have been also developed but, they were associated with some limitations including; lower detection sensitivity, use of higher doses of probe drugs increasing risk to adverse drug effects, and finally the assay methods were restricted to the analysis of probe drugs only hindering an accurate estimation of CYP450 phenotyping indices since specific metabolite measurements are often required. Many assays described in the literature were developed and validated in either plasma or in urine for specific CYP450 probes. Our methods are more versatile and will allow assessment of various pharmacokinetic parameters such as metabolic clearance and renal clearance. Our methods (Supplemental Fig. 2) require only 50μL of plasma for CAF and 200μL for BUP, DM, MDZ, OME and TOL to assure a sufficient sensitivity compared to 300–1000μL of plasma commonly used in most methods from literature , , , , , , . Zhang et al.  developed an assay using only 50μL of plasma. However, their method was limited to four probe drugs and did not measure any metabolites. Indeed, sample bioanalysis being less complex makes possible the use of smaller plasma volume while maintaining a good precision at low concentrations (LLOQ) for the probe drugs. The CYP450 cocktail described by Oh et al. used microdoses of 5 probe drugs . Although their assay was associated with good sensitivity, it also required a higher volume of plasma than our methods (500 vs. 250μL, respectively) and used propranolol as internal standard while isotopically labeled compounds are employed for all analytes in our methods. The use of labeled analytes as internal standards enhances the robustness of the method by minimizing possible matrix effects; this is an advantage particularly, while performing sample bioanalyses from polymedicated patients. De Andres et al.  and Wohlfarth et al.  have recently developed two methods using comparable oral doses as given in our cocktail and showed similar range of LLOQ. These methods were developed to phenotype for CYP1A2, 2C9, 2C19, 2D6 and 3A4 but did not measure probe drugs for CYP2B6 and CYP2E1 which are involved in the metabolism of approximately 7% and 3% of major commercial drugs, respectively .