Bioanalytical Community

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

Hi Zahir,

  Calibration standards immediately below or above your LQC and HQC must have the same acceptance criteria as LQC and HQC. If they don't, then your run fails, so sorry.

Cheers!

Robert

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

It's a fair question to ask. I agree with Robert, however, that you cannot expand the acceptable range for the calibrators functioning as the new run-specific ULOQ or LLOQ. Expanding the acceptable range for those calibrators would not ensure the assay is performing similarly to validation experiments.

On a side note, this is another benefit of including anchor points for 4PL/5PL curves. If it is a true sporadic error, the anchor points can help the curve retain the correct shape rather than over-emphasizing the exact value of a calibrator that is now the highest or lowest signal in the curve fit, but was not originally intended to fill that role.

Hi Zahir,

  Calibration standards immediately below or above your LQC and HQC must have the same acceptance criteria as LQC and HQC. If they don't, then your run fails, so sorry.

Cheers!

Robert

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

HI, I slightly disagree if I understand the question and answers correctly.

If ULOQ fails, and HQC passes (with original criteria, not ULOQ criteria), the samples with concentrations up to HQC can be accepted, as the range of the curve is reduced, yet the curve up to HQC meets criteria set forth in validation. Only samples with concentrations above the passing HQCs could not be accepted as reliable and would need to be retested

It's a fair question to ask. I agree with Robert, however, that you cannot expand the acceptable range for the calibrators functioning as the new run-specific ULOQ or LLOQ. Expanding the acceptable range for those calibrators would not ensure the assay is performing similarly to validation experiments.

On a side note, this is another benefit of including anchor points for 4PL/5PL curves. If it is a true sporadic error, the anchor points can help the curve retain the correct shape rather than over-emphasizing the exact value of a calibrator that is now the highest or lowest signal in the curve fit, but was not originally intended to fill that role.

Hi Zahir,

  Calibration standards immediately below or above your LQC and HQC must have the same acceptance criteria as LQC and HQC. If they don't, then your run fails, so sorry.

Cheers!

Robert

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

Clarification always good. My reading of the question was to apply the ULOQ acceptance criterion to the new highest calibrator concentration. Based on your response, I think we are aligned.

Your response reminded me of another question, however, that is related but I don't think was asked - can you extrapolate to concentrations outside the new calibrator range. Example, if the LLOQ standard is masked and the only standards below the LQC are anchor points, can you quantitate down to LQC if it passes. My position is that it is not allowed because you cannot extrapolate concentrations outside the calibrators with acceptance criteria. So we typically make sure we put 2 calibrators below LQC and above HQC to avoid rejecting runs if the LLOQ or ULOQ calibrators fail.

HI, I slightly disagree if I understand the question and answers correctly.

If ULOQ fails, and HQC passes (with original criteria, not ULOQ criteria), the samples with concentrations up to HQC can be accepted, as the range of the curve is reduced, yet the curve up to HQC meets criteria set forth in validation. Only samples with concentrations above the passing HQCs could not be accepted as reliable and would need to be retested

It's a fair question to ask. I agree with Robert, however, that you cannot expand the acceptable range for the calibrators functioning as the new run-specific ULOQ or LLOQ. Expanding the acceptable range for those calibrators would not ensure the assay is performing similarly to validation experiments.

On a side note, this is another benefit of including anchor points for 4PL/5PL curves. If it is a true sporadic error, the anchor points can help the curve retain the correct shape rather than over-emphasizing the exact value of a calibrator that is now the highest or lowest signal in the curve fit, but was not originally intended to fill that role.

Hi Zahir,

  Calibration standards immediately below or above your LQC and HQC must have the same acceptance criteria as LQC and HQC. If they don't, then your run fails, so sorry.

Cheers!

Robert

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.

Dear Joleen

Now it is getting complicated :)

If the LLOQ-STD fails then the second lowest STD is the new LLOQ. That one has 20% acceptance criteria. 

If you do not have a STD in between the Low QC and failed LLOQ, then the runs fails. That is LLOQ (STD1) - LowQC -  STD2 - STD3 etc  which now will be LowQC - STD 2(new LLOQ) - STD3 (run fails). QCs should always be 'covered' by a STD point. 

If your second lowest STD is below Low QC then it is still acceptable. That is LLOQ (STD1) -  STD2 - LowQC - STD3 which now will be  STD2 (new LLOQ) - LowQC - STD3. 

But you cannot extrapolate and use the anchor points as STD - as they are not STD but anchor points. So agree to your approach of having two STD below Low QC. 

Best regards

Marianne

Clarification always good. My reading of the question was to apply the ULOQ acceptance criterion to the new highest calibrator concentration. Based on your response, I think we are aligned.

Your response reminded me of another question, however, that is related but I don't think was asked - can you extrapolate to concentrations outside the new calibrator range. Example, if the LLOQ standard is masked and the only standards below the LQC are anchor points, can you quantitate down to LQC if it passes. My position is that it is not allowed because you cannot extrapolate concentrations outside the calibrators with acceptance criteria. So we typically make sure we put 2 calibrators below LQC and above HQC to avoid rejecting runs if the LLOQ or ULOQ calibrators fail.

HI, I slightly disagree if I understand the question and answers correctly.

If ULOQ fails, and HQC passes (with original criteria, not ULOQ criteria), the samples with concentrations up to HQC can be accepted, as the range of the curve is reduced, yet the curve up to HQC meets criteria set forth in validation. Only samples with concentrations above the passing HQCs could not be accepted as reliable and would need to be retested

It's a fair question to ask. I agree with Robert, however, that you cannot expand the acceptable range for the calibrators functioning as the new run-specific ULOQ or LLOQ. Expanding the acceptable range for those calibrators would not ensure the assay is performing similarly to validation experiments.

On a side note, this is another benefit of including anchor points for 4PL/5PL curves. If it is a true sporadic error, the anchor points can help the curve retain the correct shape rather than over-emphasizing the exact value of a calibrator that is now the highest or lowest signal in the curve fit, but was not originally intended to fill that role.

Hi Zahir,

  Calibration standards immediately below or above your LQC and HQC must have the same acceptance criteria as LQC and HQC. If they don't, then your run fails, so sorry.

Cheers!

Robert

Hi All,

For editing a calibration curve (curve reprocessing) in quantitative ligand binding assays, if either LLOQ or ULOQ calibrator had to be masked, the assay limit shifts to the next lowest or highest point (The AAPS Journal (2018) 20: 22). In this case, should that next lowest or highest calibrator point assume the 25% RE or %CV acceptance criteria (like usual LLOQ or ULOQ acceptance criteria)? Thanks for your time.