Metabolite-to-parent drug concentration ratios in hair may provide a convenient tool to study the metabolism of certain drugs in a non-invasive way or even post mortem. CYP1A2 is a drug metabolizing enzyme, characterized by large inter-individual differences in its activity. To predict the in vivo activity of CYP1A2, phenotyping using a selective substrate proved to be more informative than genotyping, probably due to the combined effects of environmental and genetic factors on the enzyme activity. The standard approach for CYP1A2 phenotyping is to determine the paraxanthine/caffeine ratio in plasma, at a fixed time point after intake of a dose of the CYP1A2 substrate caffeine. The aim of this study was to evaluate whether paraxanthine/caffeine ratios measured in hair samples reflect the plasma-based CYP1A2 phenotype. Therefore, an LC-MS/MS method for the determination of caffeine and paraxanthine in hair was developed and validated.

All analyses were performed using a UPLC™-API 4000™ system. Separation was achieved on an HSS-T3 column, followed by detection in MRM-mode after positive ESI. Caffeine-¹³C₃ and paraxanthine-¹³C₄-¹⁵N₃ were used as internal standards. The optimized sample preparation comprised double decontamination, manual grinding, proteolytic digestion and extraction of hair samples in an aqueous buffer, followed by clean-up using RP-SPE. Validation was performed based on FDA and EMA guidelines. Caffeine and paraxanthine concentrations were measured in the proximal 3cm segments of hair samples from 60 healthy volunteers and resulting paraxanthine/caffeine ratios were correlated with CYP1A2 phenotyping indices measured in plasma.

Linear calibration curves for caffeine and paraxanthine ranged from 0.02 to 0.50ng/mg. Precision (within-run and between-run) was below 12% (%CV) and accuracy below 7% (bias) for all QC's (0.02, 0.06, 0.20 and 0.40ng/mg). Ion suppression was compensated by the internal standards for both caffeine and paraxanthine. No interferences with both compounds were seen in the chromatogram. Incurred and processed samples were stable under the evaluated conditions. Paraxanthine/caffeine ratios in hair ranged from 0.12 to 0.93 (median 0.40), corresponding ratios in plasma ranged from 0.09 to 0.95 (median 0.40). A statistically significant correlation was found between ratios in hair and plasma (r=0.41, p=0.0011). However, ratios in both matrices showed a considerable variability, compromising the interpretation of hair ratios in individual cases.

Although paraxanthine/caffeine concentration ratios in hair and plasma showed overall a statistically significant correlation, large deviations between hair and plasma ratios in individual cases impede interpretation. Therefore, in the context of CYP1A2 phenotyping, metabolite-to-parent drug concentration ratios in hair need to be handled with caution.