Table 1.

Comparison between GLP and biomarker assay validation parameters

Parameter	GLP assay	Biomarker assay
Specificity/selectivity	Drugs as xenobiotics are not present in study sample matrix and samples are almost always subjected to clean-up and analyte recovery	Biomarkers as endogenous macromolecules are present in sample matrix. Samples not subject to clean up; thus detection occurs in a complex matrix resulting often in specificity issues
Sensitivity	LC–MS highly sensitive	Many biomarker assays lack sensitivity
Calibration standard	Certified chemical standards readily available	Often composition of the analyte not known or fully characterized. Thus certified standard not available
Calibration model	Mostly linear, extending over several orders of magnitude	Many biomarker assays have limited dynamic range and baseline patient values not known
Quality controls	Certified standard and blank patient sample matrix available	Adequate supplies of blank matrix and certified standard problematic—surrogate matrices utilized
Precision/accuracy/reproducibility	Robust technology subjected to internationally recognized acceptance criteria	Inherently variable because of methodological and biological issues. No recognized ‘acceptance criteria'
Stability	Drug standards, QCs and sample analyte stability often very good	Stability of biological standards and biological matrices analyte often very poor
Parallelism		Studied due to of use of surrogate standards and matrices for calibration purposes
Dilution linearity		Studied due to complex analyte and matrix
Reagents		Method dependent on the performance of reagents that are themselves derived from biologic sources where supply, quality and stability issues become important

Abbreviations: GLP, Good Laboratory Practice; LC–MS, liquid chromatography–mass spectrometry.