59 LC-MS/MS-method development quantifying FVIII Validation First, a test was performed to determine whether the LLOQ level of 1 ng/mL was achievable. After performing the sample purification as described above, LLOQ sample was analyzed together with a blank (FVIII deficient plasma). The signal to noise (S/N) ratio at the retention time of the surrogate peptide “GELNEHLGLLGPYIR” was obtained. Here a S/N ratio of 27 was found which is above the acceptance threshold of 5 (Figure 6). The accuracy and precision of the calculated concentrations of standards 1 to 7 were also in agreement with guidelines (Table 2). Within-run and between-run precision and accuracy were validated in three days using four QC levels in fivefold. The overall precision and accuracy for LLOQ, QC low, med and high were found to be within acceptance criteria of <20% and 15%, respectively (Table 3). Combined standard uncertainty, which consists of the bias and CV, was calculated through the Nordtest approach and can be used to derive uncertainty values of sample measurements between de validated calibration range 1 – 500 ng/mL (Table 4). Matrix effect was investigated by spiking known amounts of FVIII to citrate plasma from 3 different volunteers. No matrix effect was observed; all calculated values corresponded well with spiked values (Table 5). Figure 6. Left side, chromatogram of standard 1 ng/mL (A) and chromatogram of blank depleted FVIII plasma (B) both measuring GELNEHLGLLGPYIR SRM transition 561747.93. Right side, chromatogram of internal standard (C) and chromatogram of blank (D) measuring GELNEHLGLLGPYI(R*) SRM transition 564752.93. Table 2. Statistics of the back calculated concentrations of the standard curve analyzed during three days. Nominal concentration [ng/mL] 500 200 80 40 16 4 1 Mean 506.97 194.42 74.81 39.16 15.76 4.34 1.00 Stdev 9.45 3.43 0.56 1.14 0.44 0.26 0.04 Accuracya 1.39 -2.8 -6.5 -2.1 -1.5 8.5 0.1 Imprecisionb 1.9 1.8 0.8 2.9 2.8 6.1 4.2 a Accuracy: determined as (measured conc. – nominal conc.) / nominal conc. × 100%. b Imprecision: expressed as co-efficient of variation. 3
RkJQdWJsaXNoZXIy MTk4NDMw