Anouk Donners

88 Chapter 5 with 12 blank human plasma samples from 12 different individuals. The linearity of the standard curve was assessed with 1/x weighting. The within-run and between-run accuracy values and precision were evaluated for the quality control (QC) samples of LLOQ, QC low, QC medium, and QC high, corresponding to concentrations of 4, 10, 200 and 400 µg/mL, respectively. Stability was tested using QC low and high samples in the autosampler (after sample preparation, at 10 °C) and for three freeze (-80 °C)-and-thaw cycles. Samples were analyzed in quintuplicate on three different days. Cross-validation study Patient sampling The cross-validation study had a cross-sectional design and was performed on patients from the Van Creveldkliniek (University Medical Center Utrecht, the Netherlands) in accordance with our local institutional Medical Ethics Review Board-approved, optout procedure (study approval number 21-77/C). Patients with haemophilia A received emicizumab loading doses of 3 mg/kg/week for 4 weeks, followed by maintenance doses of 6 mg/kg/4 weeks with varying dosing intervals (from 7 to 28 days) using entire vials according to local clinical protocol [31]. Samples were taken in loading and maintenance phases during clinical visits (usually at week 1, 2, 4, and month 3, then annually) in the period between June 2018 and February 2021. All peripheral blood samples from patients receiving emicizumab were collected through venipuncture in 4.5 mL tubes (BD vacutainer®), containing 1/10 volume of 105 mM trisodium citrate. Plasma samples were prepared from blood samples by two subsequent centrifugation steps at 2000 G for 5 min at room temperature. Samples were aliquoted, stored at -80 °C and analyzed with mcOSA and LC-MS/MS. Modified and calibrated one-stage clotting assay The emicizumab concentration was measured with the mcOSA on a Sysmex CS2500, a coagulation analyzer (TOA Medical Electronics Co., Ltd., Hamburg, Germany) with Actin FS APTT reagent (Siemens, Marburg, Germany). Standard dilutions for CS2500 were applied, and were followed by an extra dilution 1:8 with Owren’s Veronal Buffer (CA system buffer) to minimize FVIII interference, then FVIII-deficient plasma, ActinFS and CaCl2 were added (Siemens, Marburg, Germany). Emicizumab concentrations were deduced from an emicizumab calibration curve, based on the plasma calibrator (R2 Diagnostics; catalog #152-401-RUO, 102 µg/mL, LOT #EC0140). The plasma controls (R2 Diagnostics; catalog #152-401-CE) of Level 1 (26.6 µg/mL, LOT #E10310) and Level 2 (73.4 µg/mL, LOT # E20410) were used as internal quality controls. The calibration curve was linear over a concentration range of 10−200 µg/mL with an R2 of 0.9985. The within-run and between-run precision (relative standard deviation [RSD], %) of the control samples ranged between 3.5% and 5.7%. The RSDs of the two control samples were similar after four freeze-and-thaw cycles. The lower limit of quantification was 2 µg/mL.