Transcript Journal Club - Clinical Chemistry

Alternative Calibration Strategies for
the Clinical Laboratory: Application
to Nortriptyline Therapeutic Drug
Monitoring
M.T. Olson, A. Breaud, R. Harlan,
N. Emezienna, S. Schools, A.L. Yergey,
and W. Clarke
June 2013
www.clinchem.org/content/59/6/920.full
© Copyright 2013 by the American Association for Clinical Chemistry
Background
 Use of LC-MS/MS is growing in the clinical
laboratory
 LC-MS/MS instrument is expensive, so a
single instrument often must be used for
multiple quantitative assays
 Additional challenges include assay
throughput and meeting turnaround time goals
for some assays
© Copyright 2009 by the American Association for Clinical Chemistry
Background
 A major driver behind LC-MS/MS assay cost
and time is the generation of a calibration
curve with every assay run
 CLIA requires linearity proof biannually
 Practice of measuring calibration line with every
assay is unique to mass spectrometry
 Practice of measuring calibration line with every
assay originated when the instruments were less
stable than they are now
© Copyright 2009 by the American Association for Clinical Chemistry
Background
 Mass spectrometry quantification works on the
“response ratio”; the peak area for the analyte
divided by the peak area for the internal standard
 In example below, analyte area is twice the internal
standard. Internal standard concentration is known, so
analyte concentration is twice the internal standard.
© Copyright 2009 by the American Association for Clinical Chemistry
Background
 Instead of making a calibration curve, the
authors used the response ratio
 There is a correction for day to day differences, the
“response factor” (RF) which is the response ratio of a
1:1 analyte:internal standard mixture
 RF can be measured with each batch:
contemporaneous RF or cRF
 Alternatively, RF can be measured with QC failures or
after instrument manipulation: sporadic RF or sRF
© Copyright 2009 by the American Association for Clinical Chemistry
The equations
ƒ is the response factor (RF)
This is a well behaved assay, there should not be an intercept (k=0).
Rearranging…
© Copyright 2009 by the American Association for Clinical Chemistry
Experimental Design
Figure 1. Nortriptyline concentrations were measured for 68 Patients on 16 days over a 2
month period. As these specimens were obtained for routine clinical therapeutic drug
monitoring, the standard clinical protocol was observed (shaded box). Both the cRF and sRF
schemes were performed on each specimen.
© Copyright 2009 by the American Association for Clinical Chemistry
Results: Is the Assay Well Behaved?
(i.e., is k=0? Is ƒ=1?)
Figure 2. The dot-dash blue line representing the calibration curve with an internal standard
concentration of 36ng/mL shows the most overlap with the solid unity line. The highest
proportional and constant error is seen with 6ng/mL.
© Copyright 2009 by the American Association for Clinical Chemistry
Results: Does this work?
 Graph A shows patient results at the
four internal standard concentrations
plotted against calibration curve results
 Best agreement is seen with 36 ng/mL
 Bland-Altman (Graph B) shows mean
bias of 3.69% for cRF versus
calibration curve with a range of -15.8
to 23.2%
Figure 3. The dot-dash blue line representing the
calibration curve with an internal standard
concentration of 36ng/mL shows the most overlap
with the solid unity line. The highest proportional
and constant error is seen with 6ng/mL.
© Copyright 2009 by the American Association for Clinical Chemistry
Results: Is there clinical impact?
Table 1. If the cRF numbers had been used, they would have resulted in the reporting of 3
concentrations in different clinical categories than were resulted using the calibration curve in
the existing protocol. If the sRF numbers had been used, 4 results would have been in
different clinical categories.
© Copyright 2009 by the American Association for Clinical Chemistry
Discussion
 Measurement of multiple analytes with a single
LC-MS/MS assay is commonplace
 Done to minimize calibration and QC
 No clinical benefit in the context of prescription drugs
with no abuse potential
 Multiplexing risks ion suppression and
augmentation for coeluting analytes
 cRF or sRF calibration can cut time spent
calibrating instrument and may make single
analyte assays more feasible
© Copyright 2009 by the American Association for Clinical Chemistry
Discussion
 LC-MS/MS assays function best when analyte
concentration is close to internal standard
concentration (A/IS ~ 1)
 For clinical values for this assay (mostly in 100150 ng/ml range) the ideal internal standard
concentration was 36 ng/mL
 Thus, selection of internal standard for an assay
should reflect expected analyte concentrations
in the patients
© Copyright 2009 by the American Association for Clinical Chemistry
Thank you for participating in this month’s
Clinical Chemistry Journal Club.
Additional Journal Clubs are available at
www.clinchem.org
Download the free Clinical Chemistry app
on iTunes for additional content!
Follow us
© Copyright 2009 by the American Association for Clinical Chemistry