Hi,

"The LLOQ defines the method sensitivity..." sensitivity for different ranges of the same calibration curve could be different for example for calibration curve between 4-50. LLOQ could be different for 4-10, 4-25 or 4-50. Especially if slope of calibration curve is done with weighting or fit is not linear (for example ligand binding assays - LBA). "For most LBAs, calibration (standard) curves are inherently nonlinear ...." to make correct fit it is recomended that " in general, more calibration standards are needed to define the fit over the calibration curve range for LBAs" If calibrator standards cover geometrical, or log etc distance between concentrations then you can expect that LLOQ could be different depending on calibration curve range because fit of the calibration curve will depend on range (different ranges represent different fit and slopes and different numbers of concentration points for example).

Best regards

Tomasz

citations from: https://www.fda.gov/files/drugs/published/Bioanalytical-Method-Validation-Guidance-for-Industry.pdf

Tomasz Grabowski

Hi，Tomasz:

Thanks for you answer and the attachment.

So according to your answer, may I take it that, statistically different PSA range would affect the PSA test performance?

Yours Yan

Hi,

If it's a case related to differences in the calibration curve of the method you cant because it's not the whole story. Generally not (only) range of PSA values "per se" has an impact on some methods features like LLOQ but the variability of the method in each calibrator concentration point. Its key element and if the calibration curve is not linear then the method of the fitting could be a key determinant of the differences between LLOQ. I think it's your case...

Best regards

Tomasz