Overcoming non-specific adsorption issues for AZD9164 in human urine samples: Consideration of bioanalytical and metabolite identification procedures

Abstract

A key challenge in the development of robust bioanalytical methods, for the determination of drug analyte in human urine samples, is the elimination of potential analyte losses as a result of non-specific adsorption to container surfaces in which the samples are collected, stored or processed. A common approach to address adsorption issues is to treat the urine samples with additives that serve to increase analyte solubility and/or minimise interaction with the container surfaces. A series of adsorption experiments were performed on human urine samples containing an adsorption-prone in-house development compound (AZD9164). A roller-mixing methodology was employed to maximise sample interaction with container surfaces and quantification of analyte was performed by LC–MS/MS following minimal sample preparation. In the absence of any urine additive, adsorptive losses averaged 35% but were highly variable between different lots of urine. In the presence of a range of additives, including the surfactants Tween 80, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulphonate (CHAPS) and sodium dodecylbenzenesulphonate (SDBS), analyte adsorption was shown to be eliminated. Of particular academic interest was the finding that adsorptive losses could also be reduced upon the addition of phospholipid. The presence of additive generally had no marked impact on the analyte MS response but the use of an isotopically labelled internal standard satisfactorily compensated for instances in which ion suppression was observed, e.g. in the presence of Tween 80. Since metabolite profiling/identification investigations are often performed on urine samples originating from early clinical pharmacology studies, the elution of selected additives was also monitored by MS. CHAPS, dimethylacetamide (DMA) and HP-β-cyclodextrin eluted as single chromatographic peaks in, or just after, the column void volume whilst polymeric Tween 80, and to a lesser extent SDBS, eluted over a wide retention time window. The potential of the latter surfactants to obscure the detection of unknown metabolites is significant and therefore their use in urine samples, upon which metabolite investigations are to be performed, is not recommended. Upon consideration of other factors such as additive cost and toxicity, CHAPS was selected for use in development of the validated assay.