Full validation using β-content, γ-confidence tolerance interval: Application for LC-MS/MS determination of Doxycycline in human plasma

- To ensure the reliability of a bioanalytical method, and to control the risk associated with future use, a statistical strategy for full analytical validation (analytical validation and measurement uncertainty) has been proposed.
- This approach based on the concept of the total error allows to combine the systematic error and random error to build a tolerance interval as an evaluation of the measures' accuracy.
- This approach allows to calculate the measurement uncertainty of the method without any additional effort, by using data coming from the analytical validation when we respect as best as possible the intermediate precision conditions at each concentration level.
- The total error approach which can be expressed as an interval βγ-CCTI, has been demonstrated by evaluating the LC-MS/MS method for the determination of Doxycycline in human plasma.
- We conclude that the approach applied in this work is absolutely flexible as long as the values of β-content and γ-confidence can be modified according to the analyst's requirements and government regulations; it is also effective since it greatly facilitates the calculation of the measurement uncertainty at each level of concentration as aid decision-making tool.

A statistical strategy for full analytical validation has been applied to ensure the reliability of a bioanalytical method and to control the risk associated with future use. To this end, the applicability of this approach, called uncertainty profile, has been demonstrated by evaluating the liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) bioanalysis, for the determination of Doxycycline in human plasma. This innovative procedure combines two main concepts, namely analytical validation and measurement uncertainty; its aim is to guarantee that a known amount of future results obtained with the bioanalytical method will be within the acceptance limits set beforehand. In addition, this approach allows to calculate the measurement uncertainty of the method without any additional effort, by using data coming from the analytical validation when we respect as best as possible the intermediate precision conditions at each concentration level.Thus, the LC-MS/MS method for the determination of Doxycycline in human plasma was found to be valid in the studied concentrations range since the tolerance intervals of type β-content, γ-confidence fell into the acceptable limits of ±15%, and the relative expanded uncertainty did not exceed 11% of the values of β-proportion and α-risk equal to 90% and 5% respectively.