Utilizing Internal Standard Responses to Assess Risk on Reporting Bioanalytical Results from Hemolyzed Samples

Abstract

Abstract. Bioanalytical analysis of toxicokinetic and pharmacokinetic samples is an integral part of small molecule drugs development and liquid chromatography-tandem mass spectrometry (LC-MS/MS) has been the technique of choice. One important consideration is the matrix effect, in which ionization of the analytes of interest is affected by the presence of co-eluting interfering components present in the sample matrix. Hemolysis, which results in additional endogenous components being released from the lysed red blood cells, may cause additional matrix interferences. The effects of the degree of hemolysis on the accuracy and precision of the method and the reported sample concentrations from hemolyzed study samples have drawn increasing attention in recent years, especially in cases where the sample concentrations are critical for pharmacokinetic calculation. Currently, there is no established procedure to objectively assess the risk of reporting potentially inaccurate bioanalytical results from hemolyzed study samples. In this work, we evaluated the effect of different degrees of hemolysis on the internal standard peak area, accuracy, and precision of the analyses of BMS-906024 and its metabolite, BMS-911557, in human plasma by LC-MS/MS. In addition, we proposed the strategy of using the peak area of the stable isotope-labeled internal standard (SIL-IS) from the LC-MS/MS measurement as the surrogate marker for risk assessment. Samples with peak areas outside of the pre-defined acceptance criteria, e.g., less than 50% or more than 150% of the average IS response in study samples, plasma standards, and QC samples when SIL-IS is used, are flagged out for further investigation.KEY WORDS: hemolyzed sample/haemolyzed sample; hyperlipemic sample; internal standard peak area/internal standard responses; LC-MS/MS; regulated bioanalysis; risk assessment.

Evaluation, Identification and Impact Assessment of Abnormal Internal Standard Response Variability in Regulated LC−MS Bioanalysis

“…The closer structurally an IS is to the analyte of interest, the better it tracks the analyte in LC−MS bioanalysis. SIL ISs, such as 2 H-, 13 C-, and 15 N-labeled analytes, with nearly identical physicochemical properties but a different mass than the analyte, are the best choices and have been extensively employed in the regulated LC−MS bioanalysis [19][20][21][22]. Other advantages of SIL IS have been reported.…”

Hazards in chromatographic bioanalysis method development and applications

“…In these cases, great attention to the amount of ion suppression and its effect on recovery and accuracy of samples is needed. This can be accomplished by using approaches such as monitoring the variability in IS response (Fung, Aubry, Allentoff, & Ji, ).…”

“…This can be accomplished by using approaches such as monitoring the variability in IS response (Fung, Aubry, Allentoff, & Ji, 2015).…”

“…Hemolysis is another phenomenon that occurs quite often during blood sample collection and handling, when lysis of red blood cells results in additional endogenous components being released into the plasma may cause additional matrix effects during the analysis of the real samples (Fung et al, 2015;Hughes, Bajaj, Fan, & Wong, 2009;Tan et al, 2012;Tang & Thomas, 2012). As additional components of red blood cells are normally absent in plasma, their associated matrix effect and even their impact on the stability of target analytes should be evaluated during bioanalytical method development and validation (Bérubé, Taillon, Furtado, & Garofolo, 2011;Fung et al, 2015).…”

“…For assessment of the influence of hemolysis on the accuracy and precision of the method, QC samples can be prepared in plasma with different percentages of hemolyzed blood and analyzed during method development. Additionally, the effects of the degree of hemolysis on the reported sample concentrations can be evaluated by monitoring the variation of internal IS peak areas in hemolyzed QC samples (Fung et al, ).…”