Clinical Chemistry 43: 505-510, 1997;
(Clinical Chemistry. 1997;43:505-510.)
© 1997 American Association for Clinical Chemistry, Inc.
HPLC method for monitoring SDZ PSC 833 in whole blood
Mitchell G. Scott1,a,
Karl G. Hock1,
Daniel L. Crimmins2 and
Paula M. Fracasso3
1
Division of Laboratory Medicine (Box 8118),
2
Department of Molecular Biology and Pharmacology, Protein and Nucleic Acid Chemistry Laboratory (Box 8103), and
3
Department of Medicine (Box 8056), Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, MO 63110.
a Author for correspondence and reprint requests. Fax 314-362-1461; e-mail mscott{at}labmed.wustl.edu
 |
Abstract
|
|---|
P-glycoprotein (Pgp) is a 170-kDa membrane transporter that mediates
drug efflux and is an effector of multidrug resistance. SDZ PSC 833
(PSC), a nonimmunosuppressive cyclosporine that potently modulates Pgp,
is currently under clinical evaluation in patients with cancer. We have
developed a reversed-phase HPLC assay for determining PSC blood
concentrations that utilizes a step gradient with linear segments to
resolve PSC into two distinct peaks (likely to be keto and enol
isomers). To clinically validate the assay, PSC concentrations were
obtained by HPLC from nine patients receiving oral doses of 5 mg/kg
every 6 h. Values ranged from 0.91 to 5.4 mg/L during the dosing
period, comparable with concentrations of PSC that modulate Pgp in
vitro. In addition, we investigated the immunoreactivity of the Abbott
TDx cyclosporin A (CsA) monoclonal whole-blood assay for PSC. The TDx
CsA assay cross-reacts ~17% with PSC as determined by adding known
amounts of PSC to whole blood. When PSC concentrations obtained by the
TDx CsA assay were divided by 0.17, we found agreement between the TDx
CsA assay and the HPLC PSC assay for samples from nine patients.
Key Words: indexing terms: P-glycoprotein • multidrug resistance • cyclosporine
|
Introduction
|
|---|
A frequent limitation of successful chemotherapy in the treatment
of cancer is the development of resistance to antineoplastic agents.
Often these cells are resistant to multiple, unrelated chemotherapeutic
agents. This phenomenon, referred to as multidrug resistance (MDR), may
be caused by overexpression of the mdr1 gene, the product of
which is a 170-kDa transmembrane glycosylated protein, P-glycoprotein
(Pgp) (1)(2)(3)(4).1
Pgp is an ATP-dependent transporter that
"pumps" hydrophobic molecules, including many chemotherapeutic
agents, out of cells. The physiological role of Pgp is unclear, but it
is related to the ATP-binding family of transporters that includes the
cystic fibrosis transmembrane regulator (5).
The role of Pgp in MDR has led to efforts to modulate Pgp activity. A
number of drugs modulate Pgp-mediated efflux, including calcium channel
blockers, local anesthetics, calmodulin antagonists, and cyclosporins
(6)(7). Of these agents, cyclosporin A (CsA)
has clinical potential, as the concentration required to modulate
Pgp-mediated efflux in vitro is achievable in patients
(8)(9)(10). Because CsA is immunosuppressive and exhibits
known toxicities during long-term therapy
(11)(12), other nonimmunosuppressive analogs
of CsA have been examined for their Pgp modulating activity. SDZ PSC
833 (PSC) is a nonimmunosuppressive analog that is ~10-fold more
potent than CsA in reversing MDR in vitro (13)(14)(15)(16).
One–two mg/L of PSC fully reverses Pgp in vitro, and these
concentrations are achievable in the blood of healthy volunteers
without significant side effects (13)(14)(15)(16)(17).
PSC is a cyclic undecapeptide differing from CsA at amino acids 1 and 2
(13). At position 1, PSC has a 3'-keto group modification
(3'keto Bmt) of the unusual amino acid
N-methyl-4-butenyl-4-methyl threonine (MeBmt) present in
CsA. At position 2, PSC contains valine, like CsD, rather than
-aminobutyric acid, which is present in CsA. The other nine amino
acids in PSC are identical to those in CsA. Because of their similarity
in structure, it is postulated that PSC may cross-react with some
CsA-specific immunoassays and, thus, the use of CsA immunoassays may
ultimately be acceptable for monitoring PSC. Nevertheless, it is
important to initially develop a PSC-specific method such as HPLC to
determine the extent of any cross-reactivity and then to confirm the
feasibility of CsA immunoassays to monitor PSC
(18)(19). Indeed, because of its complex
metabolism and the cross-reactivity of CsA metabolites with
CsA-"specific" immunoassays, HPLC has remained the established
"gold standard" for monitoring blood concentrations of CsA
(18)(19). We report the development of a
simple robust HPLC method for quantification of blood concentrations of
PSC and examine the cross-reactivity of PSC in the Abbott TDx CsA
monoclonal immunoassay.
|
Materials and Methods
|
|---|
Reagents.
PSC (batch number 90902) was kindly provided
by Sandoz Pharmaceuticals Corp., East Hanover, NJ. The HPLC internal
standard, CsD, was also obtained from Sandoz by the Barnes Hospital
Clinical Laboratories. Burdick and Jackson HPLC-grade acetonitrile
(ACN), methanol (MeOH), and hexane were obtained from Baxter Healthcare
Corp., McGaw Park, IL. Deionized H2O in all procedures was
obtained from a MilliQUF system (Millipore Corp., Bedford,
MA).
PSC and CsD solutions (1 g/L) were prepared by adding 5 mg of PSC or
CsD to 5 mL of MeOH. Stock solutions (100 mg/L) of PSC and CsD were
prepared by making 1:10 dilutions of the 1 g/L solutions in MeOH. These
solutions were aliquoted and stored at -70 °C. Whole-blood (K+EDTA
anticoagulant) pools used for preparation of calibrators and
quality-control (QC) material were made with excess sample received by
the Barnes Hospital Clinical Laboratories before discarding. Samples
were restricted to those from private physician office outpatients that
did not have requests for therapeutic drug monitoring. PSC calibrators
containing 10, 7.5, 5, 2.5, and 1 mg/L were prepared before each run by
adding 100, 75, 50, 25, and 10 µL of the 100 mg/L stock PSC to
aliquots of the whole-blood pool to a final volume of 1 mL. Fifty
microliters of the 100 mg/L CsD stock was added to 1 mL of all samples
(calibrators, QC material, and unknowns) to provide a 5 mg/L internal
standard. Control samples containing 7.5 and 2.5 mg/L PSC were prepared
in whole blood and 1-mL aliquots frozen at -70 °C.
The CsA-specific immunoassay is a fluorescence polarization immunoassay
performed on the Abbott TDx analyzer (Abbott Labs., Abbott Park, IL)
(20). To establish the cross-reactivity of PSC in the TDx
CsA assay, CsA-free whole-blood samples were prepared to contain 2.5
and 5.0 mg/L PSC from the 100 mg/L PSC stock solution.
Patient material.
Whole blood was obtained from patients
with advanced cancers enrolled in a phase I trial for paclitaxel and
PSC at Washington University School of Medicine. This trial received
approval of the Washington University School of Medicine Human Studies
Committee, and patients enrolled in the trial provided informed
consent. All patients were given a total of 10 or 12 oral doses of 5
mg/kg of PSC every 6 h. Patients were given 86–210 mg of
paclitaxel (40.5–105 mg/m2) by continuous intravenous
infusion over a 3-h period beginning 2 h after the fifth dose of
PSC.
Chromatography.
Chromatography was performed on a
gradient HPLC system consisting of two Waters 510 pumps, a Waters 717
plus autosampler, a column heater, and a Waters 486 tunable absorbance
detector set at 203 nm (Waters Corp., Milford, MA). Proprietary
Millennium software from the Waters Corp. was used to control the
system and to perform integration of peak areas. Reversed-phase
chromatography was performed with a 25 cm x 4.6 mm C18
silica column containing 5-µm particles with 30-nm pore size (Vydac
Separations Group, Hesperia, CA). The mobile phase for the isocratic
C18 chromatography was prepared by combining 500 mL of ACN,
300 mL of MeOH, and 200 mL of H2O (50:30:20). Isocratic
chromatography was performed at 2 mL/min with an injection volume of 35
µL as previously described for CsA (18)(21).
Various gradient conditions were examined, and the optimal condition
for PSC reported here was a binary step gradient from
ACN:H2O (57:43) to ACN:H2O (68:32) over 11 min,
followed by a gradient to ACN:H2O (90:10) over the next 9
min. The flow rate was 1.5 mL/min, the injection volume was 250 µL,
and the column was maintained at 75 °C during chromatography. Column
regeneration was performed by holding the mobile phase at
ACN:H2O (90:10) for 7 min, returning to ACN:H2O
(57:43) over the next 2 min, and holding an additional 13 min before
the next injection.
Extraction procedures.
Several extraction methods were
examined to determine optimal recovery of PSC from whole-blood lysates.
Solid-phase extraction used in these studies was performed by using a
1-mL Sep-Pak® Plus C18 cartridge (Waters
Corp.). Briefly, 1 mL of a whole-blood lysate sample (patient, QC, or
calibrator) was added to 2 mL of ACN:MeOH (90:10), vortex-mixed, and
centrifuged at 1000g for 5 min to remove precipitated
protein. H2O (500 µL) was added to the supernatant,
mixed, and applied to a cartridge preconditioned with 3 mL of MeOH. The
cartridge was then washed with 1 mL of 700 mL/L MeOH followed by 1 mL
of hexane. Internal standard (CsD) and PSC were eluted from the column
with 2 mL of ACN:H2O (80:20). The eluted material was
evaporated to dryness and reconstituted in 400 µL of mobile phase A
(ACN:H2O, 57:43) before injection.
|
Results
|
|---|
Optimization of chromatographic conditions
. Initial
studies of chromatographic conditions examined resolution and
separation of PSC and CsD at 10 and 5 mg/L dissolved in the initial
mobile phase without extraction. None of the isocratic methods examined
with either biphenyl or C18 columns, including the
isocratic mobile phase of ACN:MeOH:H2O commonly used for
CsA (18)(21), resolved PSC into a distinct
peak (not shown). Therefore, a variety of gradient conditions were
examined, with the C18 gradient conditions described in
Materials and Methods providing optimal resolution. Fig. 1
a depicts a typical chromatogram in which, under these
conditions, CsD elutes at 14.3 min and PSC elutes as two distinct peaks
at 15.5 and 16.5 min.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 1. Chromatogram of CsD internal standard at 5 mg/L and PSC
calibrator (a) at 5 mg/L in mobile phase A and
(b) at 7.5 mg/L after extraction from drug-free whole blood.
Full scale is 0.09 and 0.075 absorbance units in (a) and
(b), respectively.
|
|
PSC extraction from whole blood and quantification by
chromatography
. Solid-phase extraction was chosen for convenience
and resulted in similar recoveries as a common double (acid/base)
liquid extraction procedure used for CsA (18) (data not
shown). Recovery of PSC from whole blood with the Sep-Pak Plus
C18 cartridges was 44% ± 4.1% and 39% ± 3.2% (n
= 3) at PSC concentrations of 20 and 5 mg/L, respectively, whereas
recovery of CsD was 48% ± 6.4% at 5 mg/L (n = 3). This was
determined by comparing peak areas after extraction from whole blood
with the peak areas of the same concentration of drug in the initial
mobile phase but not extracted. The ratio of peak areas of PSC to CsD
was used to develop a calibration curve. Linear regression was
performed on each calibration curve to quantify PSC in QC and patient
samples. Fig. 1b
depicts a typical chromatogram of a whole-blood
calibrator containing 5 mg/L PSC and 5 mg/L CsD. Fig. 2
depicts a chromatogram from a whole-blood sample from a patient
receiving PSC to which 5 mg/L CsD, the internal standard, was added.
The ratio of peak 2 (~16.5 min) area to peak 1 (~15 min) area was
remarkably consistent regardless of PSC concentration or whether PSC
was from calibrators or patient samples. The peak 2:peak 1 ratio was
1.36 ± 0.12 determined from three each of the 1.0, 2.5, 5.0, and
7.5 mg/L calibrators, and 1.46 ± 0.13 for 12 patient samples from
7 patients.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 2. Typical chromatograms of extracts from whole blood of a
patient receiving PSC, with a whole-blood PSC concentration of 3.3
mg/L.
Full scale is 0.11 absorbance units.
|
|
HPLC assay performance
. Intrarun precision was
determined by running whole-blood QC samples five times each in a
single run and was 3.7% at a mean concentration of 2.6 mg/L, 4.1% at
a mean concentration of 6.2 mg/L, and 9.1% at a mean concentration of
7.7 mg/L. Interrun precision from 18 runs was 6.5% at a mean of 7.66
mg/L of PSC, 8.2% at a mean of 5.24 mg/L (n = 5), and 8.5% at a
mean of 2.58 mg/L. The HPLC method was sensitive to 800 µg/L as
determined by performing parallel dilutions on a sample containing 2.5
mg/L PSC and noting deviations of >20% from expected values. This was
consistent with a mean value of 1.02 ± 0.16 (n = 18) for the
1.0 mg/L calibrator when this calibrator was calculated as an unknown
against the calibration curve.
To examine potential interferences, the following drugs were added to a
whole-blood sample containing 5 mg/L PSC and CsD: acetaminophen (164
mg/L), amikacin (29 mg/L), amitryptyline (470 µg/L), chloramphenicol
(55 mg/L), CsA (360 µg/L), digoxin (2.3 µg/L), disopyramide (5.2
mg/L), ethosuximide (108 mg/L), gentamicin (6 mg/L), lidocaine (6.5
mg/L), methotrexate (8 µmol/L), phenobarbital (35 mg/L), phenytoin
(22 mg/L), primidone (10.1 mg/L), procainamide (8.9 mg/L),
N-acetylprocainamide (10.5 mg/L), propanolol (128 µg/L),
quinidine (4.7 mg/L), salicylate (330 mg/L), theophylline (24.07 mg/L),
tobramycin (7.3 mg/L), valproic acid (105 mg/L), and vancomycin (36
mg/L). At least seven additional peaks were observed, compared with the
sample without these drugs, but the retention times and recovery of CsD
and PSC were not affected (Fig. 3
).
View larger version (15K):
[in this window]
[in a new window]
|
Figure 3. Chromatogram of extracted whole blood containing 5 mg/L
PSC and CsD, to which 23 other drugs were added at concentrations
indicated in the text.
Full scale is 0.11 absorbance units.
|
|
PSC was stable in whole blood for at least 14 days at 4 °C, as three
samples with initial values between 2.53 and 7.42 mg/L produced values
that averaged 2.6% higher after 14 days at 4 °C. Similarly, five
patient samples with values between 2.7 and 5.3 mg/L produced values
that averaged 2.0% higher after 18 months at -70 °C.
PSC cross-reactivity in the TDx CsA assay.
To determine
the immunologic cross-reactivity of PSC in the TDx monoclonal CsA
assay, CsA-free whole-blood calibrators containing 2.5 mg/L and 5.0
mg/L PSC were assayed in multiple TDx runs. CsA values for these
samples were 0.45 + 0.047 mg/L and 0.86 + 0.10 mg/L (n = 10),
respectively, indicating that PSC exhibits ~17% cross-reactivity
with CsA in the TDx CsA-specific assay and that these samples had a CV
of 10–11% in this assay. Therefore, CsA values obtained from this TDx
method were multiplied by 5.7 to obtain approximate PSC concentrations
in whole-blood samples of patients receiving PSC. PSC values obtained
from the TDx CsA assay compared well by linear regression with those
obtained by the HPLC method for 86 samples from nine patients: TDx
= 0.88HPLC + 0.44 mg/L, r = 0.93 (Fig. 4
). Deming correction (22) of the least-squares
analysis was TDx = 0.93HPLC + 0.32 mg/L. Analysis of the
differences between these methods with the method of Bland and Altman
(23) showed that PSC values averaged 0.14 ± 0.36
mg/L higher by the TDx method than by HPLC (Fig. 5
).
View larger version (11K):
[in this window]
[in a new window]
|
Figure 4. HPLC vs TDx PSC concentrations for 86 whole-blood samples
from nine patients.
Solid line depicts Deming correction of linear regression.
|
|
Whole-blood concentrations of PSC during the dosing regimen ranged from
0.91 to 5.42 mg/L by HPLC and 0.92 to 5.41 mg/L by the Abbott TDx CsA
assay for nine patients in this study. Whole-blood PSC concentrations
for samples taken within 1 h before the next dose ("trough"
concentrations) ranged from 0.96 to 3.72 mg/L by HPLC and 0.99 to 5.41
mg/L by the Abbott TDx CsA assay.
|
Discussion
|
|---|
Development of a precise HPLC assay for PSC was not as
straightforward as initially anticipated. Isocratic conditions that
work well for CsA (18)(21) resulted in very
poor resolution of PSC even at temperatures of 70–75 °C, which are
necessary to minimize peak broadening of CsA due to interconversion of
its conformers (21). Under these conditions, we
consistently observed a very broad bilobed peak that spanned 2–2.5
min. The gradient conditions used greatly enhanced peak resolution but
still consistently produced two peaks. The ratio of areas of the second
peak to the first peak (~16 min) was consistent regardless of whether
the sample was pure PSC in mobile phase or PSC in patient whole blood.
A likely explanation for the consistent presence of two peaks is
keto
enol isomerization at the 3' position of the amino acid, MeBmt,
at position one of the undecapeptide.
When PSC values from the HPLC assay were compared with putative CsA
concentrations of patient samples containing PSC from the TDx CsA
immunoassay, we found that PSC cross-reacts ~17% with CsA in this
assay. Both the good correlation and the general agreement in absolute
values suggest that the HPLC and TDx methods are primarily detecting
the parent molecule. The slightly higher values observed with the TDx
assay may be due to detection of PSC metabolites that are similar in
structure to CsA metabolites, which exhibit some cross-reactivity in
the TDx CsA assay (20). For instance, the CsA metabolite
AM9, which shows the highest cross-reactivity (~13%) in the TDx
(20), is a result of hydroxylation of amino acid 9 in CsA.
It is likely that similar, simple metabolites will be present for PSC
and cross-react to a similar extent in this immunoassay. Clearly,
additional studies will be required to identify PSC metabolites, their
elution patterns in chromatography, and their reactivity with anti-CsA
antibodies.
The HPLC method described for quantifying PSC in whole blood is
sufficiently precise and accurate for clinical use. However, the
gradient conditions and subsequent column equilibration restrict sample
throughput to one injection every 60 min. Thus, this HPLC method will
likely be most useful for some limited pharmacokinetic studies during
clinical trials and for determining PSC cross-reactivity in other CsA
immunoassays rather than for routine monitoring of PSC. For instance,
we found that PSC in whole blood from these nine patients had a
reproducible immunologic cross-reactivity of ~17% in a commonly used
immunoassay for CsA (20). Whole-blood PSC concentrations
determined by HPLC correlated well with those determined by the TDx
monoclonal CsA immunoassay after accounting for this cross-reactivity.
Other CsA immunoassays exhibit different patterns of cross-reactivity
with CsA metabolites (19). Thus, it will be important to
determine the extent of PSC cross-reactivity in these assays to assess
their potential utility to monitor this CsA analog. On the basis of our
observations with the TDx CsA immunoassay, it is likely that at least
some CsA immunoassays will be acceptable for routine monitoring of PSC,
when and if this becomes clinically necessary.
|
Acknowledgments
|
|---|
We thank the staffs of the Barnard Cancer Center and the General
Clinical Research Center for obtaining patient samples, and John Turk
and David Piwnica-Worms for critical review of the manuscript. This
research was supported in part by grants from the American Cancer
Society (IN-36–35) and the US Public Health Service (MO1 RR00036).
|
Footnotes
|
|---|
1 Nonstandard abbreviations: MDR, multidrug resistance; Pgp, P-glycoprotein; CsA, cyclosporin A; PSC, SDZ PSC 388; MeBmt, N-methyl-4-butenyl-4-methyl threonine; ACN, acetonitrile; MeOH, methanol; and QC, quality control. 
|
References
|
|---|
-
Van der Bliek AM, Borst P. Multidrug resistance. Adv Cancer Res 1989;52:165-203.
[ISI][Medline]
[Order article via Infotrieve]
-
Goldstein LJ, Pastan I, Gottesman MM. Multidrug resistance in human cancer. Crit Rev Oncol Hematol 1992;12:243-253.
[ISI][Medline]
[Order article via Infotrieve]
-
Pinedo HM, Giacone G. P-glycoprotein—a marker of cancer cell behavior. N Engl J Med 1993;333:1417-1419.
[Free Full Text]
-
Roninson I. The role of MDR1 (P-glycoprotein) gene in multidrug resistance in human cancer. Crit Rev Oncol Hematol 1992;43:95-102.
-
Valverde MA, Diaz M, Sepulveda FV, Gill DR, Hyde SC, Higgins CF. Volume regulated chloride channels associated with the human multidrug resistance P-glycoprotein. Nature 1992;355:830-833.
[Medline]
[Order article via Infotrieve]
-
Tsuruo T, Iida H, Tsukagosti S, Sakurai Y. Overcoming of vincristine resistance in P388 leukemia in vivo and in vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil. Cancer Res 1981;41:1967-1972.
[Abstract/Free Full Text]
-
Ford JM, Hait WN. Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacol Rev 1990;42:155-199.
[ISI][Medline]
[Order article via Infotrieve]
-
Verweij J, Herweijer H, Oosterom R, et al. A phase II study of epidoxorubicin in colorectal cancer and the use of cyclosporine-A in an attempt to reverse multidrug resistance. Br J Cancer 1991;64:361-364.
[ISI][Medline]
[Order article via Infotrieve]
-
Sonneveld P, Durie BGM, Lokhorst HM, et al. Modulation of multidrug-resistant multiple myeloma by cyclosporine. Lancet 1992;340:255-259.
[ISI][Medline]
[Order article via Infotrieve]
-
Twentyman PR. Cyclosporines as drug resistance modifiers. Biochem Pharmacol 1992;43:109-117.
[ISI][Medline]
[Order article via Infotrieve]
-
Myers BD, Ross J, Neurton L, Luetscher J, Perlroth M. Cyclosporine associated chronic nephropathy. N Eng J Med 1984;311:699-705.
[Abstract]
-
Kahan BD, Flechner SM, Lorber MI, Jensen C, Golden D, Van Buren CT. Complications of cyclosporine therapy. World J Surgery 1986;10:348.
-
Gaveriaux C, Boesch D, Jackez B, Bollinger P, Payne T, Loor F. SDZ PSC 833, a non-immunosuppressive cyclosporine analog, is a very potent multidrug-resistance modifier. J Cell Pharmacol 1991;2:225-234.
-
Boesch D, Muller K, Pourtier-Manzanedo A, Loor F. Restoration of daunomycin retention in multidrug-resistant P388 cells by submicromolar concentrations of SDZ PSC 833, a nonimmunosuppressive cyclosporin derivative. Exp Cell Res 1991;196:26-32.
[ISI][Medline]
[Order article via Infotrieve]
-
Twentyman PR, Bleehen NM. Resistance modification by PSC-833, a novel non-immunosuppressive cyclosporin A. Eur J Cancer 1991;27:1639-1642.
-
Keller RP, Altermatt HJ, Nooter K, et al. SDZ PSC 833, a non-immunosuppressive cyclosporine: its potency in overcoming P-glycoprotein-mediated multidrug resistance of murine leukemia. Int J Cancer 1992;50:593-597.
[ISI][Medline]
[Order article via Infotrieve]
-
Sandoz Research Institute, Sandoz Pharmaceuticals
Corporation. SDZ PSC 833 Investigator's Brochure, East Hanover, NJ,
September 1994..
-
Vine W, Bowers LD. Cyclosporine: structure, pharmacokinetics and therapeutic drug monitoring. CRC Crit Rev Clin Lab Sci 1987;25:275-311.
-
Holt DW, Johnston A, Roberts NB, Tredjer JM, Trull AK. Methodological and clinical aspects of cyclosporine monitoring: report of the Association of Clinical Biochemists task force. Ann Clin Biochem 1994;31:420-446.
-
Yatscoff RW, Copeland KR, Faraci CJ. Abbott TDx monoclonal antibody assay evaluated for measuring cyclosporine in whole blood. Clin Chem 1990;36:1969-1973.
[Abstract]
-
Bowers LD, Matthews SE. Investigation of the mechanism of peak broadening observed in the high-performance liquid chromatography analysis of cyclosporine. J Chromatogr 1985;333:231-238.
[ISI][Medline]
[Order article via Infotrieve]
-
Cornbleet PJ, Gochman N. Incorrect least-squares regression coefficients in method-comparison analysis. Clin Chem 1979;25:432-438.
[Abstract/Free Full Text]
-
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;i:307-310.
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
D. Chen, D. L. Crimmins, F. F. Hsu, F. P. Lindberg, and M. G. Scott
Hemoglobin Raleigh as the cause of a falsely increased hemoglobin A1C in an automated ion-exchange HPLC method
Clin. Chem.,
June 1, 1998;
44(6):
1296 - 1301.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
