Tacrolimus Analyses

Tacrolimus concentrations were determined from 10 mL EDTA blood samples collected at several time points after transplantation. Blood sample collection was scheduled at 8.00 a.m. and patients were instructed to administer tacrolimus at 10.00 a.m. for once-daily dosing regimens, and at 10.00 a.m. and 10.00 p.m. for twice-daily dosing regimens. Blood sampling was assumed at 10 or 22 h after dosing. Blood samples for whole blood tacrolimus monitoring were drawn at the same time as the samples for tacrolimus plasma analysis, and were processed and analysed within 4 h. EDTA blood samples drawn for plasma tacrolimus measurements were kept at room temperature and within 4 h were processed to plasma by temperature-controlled centrifugation (21°C) at 1300g for 10 min, and subsequently stored at − 80°C until analysis. Haemolysed samples, as determined by visual inspection, were not analysed.

All tacrolimus analyses were performed by the Laboratory of the department of Clinical Pharmacy and Pharmacology at the University Medical Center Groningen. The laboratory participates in the LGC proficiency testing programme (LGC Group, Bury, UK) to verify accuracy, precision and specificity. Concentrations were measured using a TSQ Quantiva mass spectrometer with a Vanquish UHPLC system, both from Thermo Fisher Scientific (Waltham, MA, USA).

For whole blood tacrolimus analysis, a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was used as reported previously, with minor alterations [11, 12]. The linear range was 1.00–50.0 ng/mL, with between-day and within-day imprecision, as measured on 3 separate days, of < 10% for all four quality control samples (1.0 µg/L, 5.0 µg/L, 15.0 mg/L, 40.0 µg/L; n = 5), with an overall bias of 1.3–11.3%. All samples were analysed within 4 h, which covers the validated benchtop stability period of 74 h.

For plasma tacrolimus analysis, a validated LC-MS/MS method was used [13]. The linear range was 0.05–5.00 ng/mL, with between-day and within-day imprecision, as measured on 3 separate days, of <10% for all four QCs (0.05, 0.1, 2.0, 4.0 µg/L; n = 5), with an overall bias of −5.1 to 3.0%. Samples were prepared and frozen within the allowed 5-h benchtop stability, and analysed within three freeze-thaw cycles.