| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Divisions of
1
Endocrinology and
2
Informatics, National Institute for Biological Standards and Control, Blanche Lane, South Mimms, Potters Bar, Herts EN6 3QG, United Kingdom.
3
Department of Urology, Stanford University School of
Medicine, Stanford, CA 94305-5118.
a Author for correspondence. Fax 44-01707-646730; e-mail brafferty{at}nibsc.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Coded vials of the candidate materials and serum preparations containing PSA in the clinically important range were provided to the 10 laboratories in the study, and participants were asked to perform PSA assays currently in use in their laboratories. Data from 89 immunoassays by 26 different method-laboratory combinations were contributed to the study and analyzed centrally at the National Institute for Biological Standards and Control.
Results: Potency estimates of the preparations relative to the in-house calibrators were in good agreement with the target value of 1 µg of total PSA/vial, the preparation of free PSA giving 1.10 µg/vial (95% confidence interval, 0.99–1.21 µg/vial) and PSA 90:10, 1.11 µg/vial (95% confidence interval, 1.04–1.18 µg/vial). No immunoreactivity was detected in ampoules containing the recombinant material. Use of a common standard of PSA 90:10 significantly reduced the between-laboratory geometric coefficients of variation for serum samples included in the study and gave a much narrower range of potency estimates.
Conclusions: The preparation of free PSA was established by WHO as the First International Standard for PSA (free) with an assigned content of 1 µg of total PSA per vial. In addition, the preparation of bound PSA was established as the First International Standard for PSA (90:10) with an assigned content of 1 µg of total PSA per vial.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-antichymotrypsin (PSA-ACT)
(2)(3), are widely used in the screening,
diagnosis, and monitoring of patients with prostate cancer. However,
because of differential epitope recognition by some of the assay
systems currently in use (4), serum samples containing
equivalent amounts of total PSA but different proportions of the
immunoreactive PSA forms may show significantly different values. To
address this problem, purified preparations of the main immunoreactive
components in serum were made available as PSA reference materials, the
value of which had been assigned by mass spectroscopy (1)
and amino acid analysis (5). Subsequently, a mixture of 90%
PSA-ACT and 10% free PSA (PSA 90:10), representing the average
proportion of PSA-ACT and free PSA in sera of patients with cancer of
the prostate (6), was approved as a primary calibrator by
the NCCLS, along with preparations of PSA-ACT and PSA
(7). Recently, recalibration studies with a panel of human
serum samples in which PSA 90:10 was used as a standard led to marked
decreases in interlaboratory variation (8). This report describes the results of an international collaborative study in which preparations of free PSA and PSA 90:10, along with recombinant DNA-derived PSA were evaluated as candidate WHO reference reagents. The aims of the study were (a) to compare the immunoreactivity of the preparations in immunoassay systems representative of those commonly used in clinical practice or research and assess their suitability as WHO international standards; (b) to assess the stability of the PSA in the lyophilized preparations by assay of the contents of vials that had undergone accelerated thermal degradation; and (c) to compare the PSA immunoreactivity of different serum samples in the immunoassay systems included in the study in terms of both local calibrators and the candidate preparations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The vials of free PSA and PSA 90:10 were donated to WHO by
Prof. T. Stamey, Stanford University (Stanford, CA), and their
preparation has been described previously (8). Recombinant
DNA-derived human PSA (baculovirus-derived) was kindly donated to WHO
by Jenner Technologies (Tiburon, CA). The preparation was received as a
frozen solution containing PSA at 0.33 g/L in 50 mmol/L
Tris-glycine, 150 mmol/L NaCl (pH 7.5), and 6 mol/L urea, and was
stored at 4 °C before ampouling. After dilution to 2 L in a solution
containing 25 mmol/L Tris (pH 7.5), 8.5 g/L NaCl, 5 g/L human serum
albumin, and 1 mL/L Triton X-100, the material was distributed
into glass ampoules in 1-mL aliquots, lyophilized, and sealed according
to procedures described by WHO for International Biological Standards
(9). The human serum samples, kindly donated by Dr. A.
Milford-Ward, UK NEQAS, and which had been tested and found negative
for hepatitis B antigen, anti-hepatitis C virus, and anti-HIV,
were filled as 1-mL aliquots under the same procedures.
|
Each participant received duplicate sets of coded vials/ampoules comprising free PSA (96/668), PSA 90:10 (96/670), recombinant PSA (97/590), and human serum samples containing low, medium, and high concentrations of PSA (97/570, 97/568, 97/566, respectively). In addition, participants received accelerated degradation samples of 96/668, 96/670, and 97/590 selected on the basis of assay capacity, sample availability, and a study design giving information across as wide a range of assay systems as feasible. Participants were also asked to include their in-house calibrators in the assays.
participants in the study
Ten laboratories in six countries took part in the study and are
listed alphabetically, by country, below. Throughout the study, each
participating laboratory is referred to by a code number, which was
assigned randomly and which does not reflect the order of listing:
design of the study and assay methods contributed
Participants were asked to perform the PSA immunoassay(s)
currently in use in their laboratories and to carry out at least two
independent assays (i.e., using freshly reconstituted vials/ampoules)
of each assay type, each assay to include a set of coded preparations
selected with regard to assay system capacity and to provide maximum
information for the study. (Participants were advised that the free PSA
component retained some enzymatic activity and might react with
protease inhibitors, and therefore, serum-based matrices were to be
avoided for initial reconstitution of the vial contents.) Each set of
preparations included coded duplicates of 96/668 and 96/670 to provide
an independent assessment of assay accuracy and precision. All
preparations and any local calibrators were to be included at several
doses to provide information on linearity and parallelism of the
dose–response relationship. All data were reported to the National
Institute for Biological Standards and Control (NIBSC) for analysis.
statistical analysis
As far as possible for this study, an "assay" was defined as
an independent test for each preparation beginning with freshly opened
ampoules. The methods of analysis used were dependent on the form of
the data returned by the participating laboratories. The term
"potency" has been used to describe the relative immunological
activities of the preparations.
For those assays where raw data were supplied for both an in-house reference (IHR) preparation and for the ampouled preparations, the principles of parallel-line bioassay analysis were applied (10). Response data were transformed to percentages relative to the estimated upper and lower limits of the dose–response lines for each assay. An in-house program, WRANL (11), was used to provide weighted regression analysis of logit response on log dose with an analysis of variance giving tests for linearity and parallelism of the fitted lines. Estimates of relative potency were then calculated from the fitted models. For some assays, a log transformation or a square root transformation was found to be more satisfactory, in which case unweighted regression analysis was used.
For some assays, no IHR data were available, but raw data for the ampouled preparations were returned, together with the laboratories’ own estimates of potency. In these cases, the potency estimates were calculated directly from the laboratory estimates (adjusting for dilution factor), and the principles of parallel line bioassay were applied to the raw data to obtain relative potencies and to perform statistical checks on linearity and parallelism.
Finally, for those assays where only laboratory estimates of potency were returned, the potency estimates were once again calculated directly from the laboratory’s own estimates. Where estimates were given at different dilutions, tests for satisfactory linearity and parallelism were carried out as described above, using the actual estimates in place of raw assay responses.
Estimates of the relative activity (to ampoules of the same preparation stored continuously at -20 °C) remaining in the ampoules of the PSA preparations after storage at increased temperatures were used to fit an Arrhenius equation relating degradation rate to absolute temperature, assuming first-order decay (12), and hence were used to predict the degradation rate of the preparations when stored at -20 °C.
All mean estimates quoted are unweighted geometric mean estimates. Geometric coefficients of variation (GCVs), which allow the concise expression of the size of the multiplicative variation (13), were used to measure the variability present between laboratories and assays.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. No PSA activity was reported in ampoules of the recombinant
material 97/590 by any of the participants.
|
The transformed response lines were generally found to be linear and
parallel to one another. A pooled "between-assay" GCV was used to
demonstrate the variability between assays of the same type performed
by the same laboratory. A "between-laboratory" GCV was also
calculated, treating assays performed using different assay systems as
being from separate "laboratories". For the
between-laboratory GCV, only the total PSA results were used
because the free-PSA and PSA-ACT results are not directly comparable
with total PSA. A full listing of estimates of activity for all
preparations (shown in
Figs. 1–5
) may be obtained from the authors
upon request.
|
|
|
|
|
variability of immunoassay estimates
Within-assay variability was demonstrated by the potencies of the
coded duplicates, relative to each other, of 96/668 and 96/670. The
potencies also provided a measure of the minimum variability between
laboratories using these assay systems because the preparations
compared were known to be identical and hence comparisons were not
affected by any differences in assay specificity. Overall, the
estimates showed good agreement with their expected value of 1.0,
giving mean values of 0.98 for the duplicates of 96/668 and 1.03 for
96/670.
comparison of the various local calibrators with free psa
preparation 96/668
The geometric mean potency estimates calculated for each assay
method in each laboratory are plotted in Fig. 1
. The unweighted geometric mean of estimates from assays for
total PSA was 1.15 µg/vial (95% confidence interval, 1.01–1.31
µg/vial). The results obtained by laboratory 7, which used a
nonequimolar assay, were noticeably higher than the remaining
laboratories. An analysis of variance on the logs of estimates from
individual assays was carried out and Duncan’s multiple range test
(significance set at 1%) was used to check for outliers
(14). Laboratory 7 was shown to be an outlier under this
test. The geometric mean estimate excluding this laboratory was 1.10
µg/vial (95% confidence interval, 0.99–1.21 µg/vial). The
between-assay GCV (6.9%) was of similar magnitude to that observed for
the coded duplicates (8.2%), whereas the between-laboratory GCV
(29%) was large when compared with that for the coded
duplicates (6.9%). Assays measuring free PSA gave a potency estimate
of 0.90 µg/vial (95% confidence interval, 0.84–0.98 µg/vial).
comparison of the various local calibrators with psa 90:10
preparation 96/670
The geometric mean laboratory/assay estimates of potency are
plotted in Fig. 2
. The unweighted geometric mean of estimates from assays for
estimation of total PSA was 1.17 µg/vial (95% confidence interval,
1.06–1.29 µg/vial). Under Duncan’s multiple range test,
laboratories 4 and 7 were shown to be outliers. The geometric mean
estimate excluding these laboratories was 1.11 µg/vial (95%
confidence interval, 1.04–1.18 µg/vial). The between-assay and
between-laboratory GCVs (7.5% and 14%) were of similar magnitude to
those obtained for the coded duplicates (10% and 9.4%). Assays
specific for free PSA gave a potency estimate of 0.11 µg/vial (95%
confidence interval, 0.10–0.12 µg/vial).
comparison of free psa preparation 96/668 with psa 90:10
preparation 96/670
Assays for total PSA gave an unweighted geometric mean estimate
for the potency of 96/668 relative to 96/670 of 1.05 (95% confidence
interval, 0.94–1.18), excluding laboratories 4 and 7. The
between-assay and between-laboratory GCVs (8.4% and 28%) were similar
to those for the potency of 96/668 calculated using the in-house
reference preparations (6.9% and 28%). Assays measuring free PSA gave
a relative potency of 8.10 (95% confidence interval, 7.63–8.61).
stability of free psa preparation 96/668 and psa 90:10 preparation
96/670
Estimates of the activity of samples of 99/668 that had been
stored for 485 days at 4 and 20 °C relative to the samples stored
continuously at -20 °C showed no detectable loss of activity, and
samples stored at 37 °C showed only limited loss, although after
storage at higher temperatures (45 and 56 °C), there was a clear
deterioration in activity. Using these data, we estimated a predicted
degradation rate (12) of 0.042% per year for samples stored
at -20 °C. Similar data were obtained for 96/670 from assays for
total PSA, giving a predicted degradation rate of 0.027% per year at
-20 °C. Limited data from assays for free PSA (laboratories 1 and
9) and PSA-ACT (laboratory 1) were in general agreement with these
results.
comparison of the various local calibrators and psa 90:10
preparation 96/670 with human serum samples (97/570, 97/568, and
97/566)
The geometric mean potencies of the human serum samples are listed
in Table 3
, and potencies from individual assays are plotted in Figs. 3
, 4
, and
5
. Separate values are given for the potencies calculated using
the IHR preparations and 96/670.
|
The between-laboratory GCVs for estimates from assay systems measuring
total PSA (46%, 28%, and 23% for 97/570, 97/568, and 97/566,
respectively) were appreciably smaller when the PSA 90:10 preparation
96/670 was used (27%, 12%, and 18%; Table 3
).
A full listing of estimates of activity for all preparations may be obtained from the authors upon request.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Comparison of the ampouled preparations of free PSA and PSA 90:10 and local calibrators showed little evidence of any nonparallelism in the 26 method-laboratory combinations included in the study. However, there was a greater between-assay GCV for all estimates compared with that seen for the coded duplicates, which suggests that there were differences in the assay system specificities. The use of both monoclonal-monoclonal and monoclonal-polyclonal assay systems, which may differ in their recognition of the free and complexed forms (4), was probably the main contributing factor, and it was noted that the GCV was significantly reduced by the exclusion of data from those systems known to be (laboratory 7) or apparently (laboratory 4) "nonequimolar" in recognition of the two forms. Omission of other nonequimolar data made no appreciable difference to the estimates. Differences in calibration of the kits may also have contributed to the variability of the estimates (15).
PSA 90:10 is representative of the ratio of the forms of PSA found in
serum of patients with cancer of the prostate (6), although
the ratio can vary. It was noted in the present study that although the
serum samples were broadly in line with the predicted 90:10 ratio,
there did appear to be differences between the sera. Use of a common
standard of this preparation, 96/670, significantly reduced the
between-laboratory GCVs for the serum samples in the study (Table 3
)
and gave a much narrower range of potency estimates (Figs. 3
, 4
, and 5
). A similar result was obtained using a 90:10 preparation as
calibrator for nine commercial immunoassays (5).
There currently are no internationally recognized reference reagents for the PSA forms that exist in serum. Concentrations in biological fluids, measured by immunoassay, are routinely expressed in gravimetric units, typically µg/L total PSA, using as calibrators either free PSA purified from seminal plasma or more recently, PSA-ACT and PSA 90:10. The contents of the bulk preparations of free PSA and PSA-ACT used in 96/668 and 96/670 were accurately determined by amino acid analysis (5) and mass spectrometry (1), and based on these data, vials were filled with 1 µg of total PSA. The results of this study were in good agreement with that value, 96/668 (free PSA) giving 1.10 µg/vial (95% confidence interval, 0.99–1.21 µg/vial), and 96/670 (PSA 90:10) giving 1.11 µg/vial (95% confidence interval, 1.04–1.18 µg/vial). The preparations appeared to be sufficiently stable to serve as reference reagents with a predicted loss of immunoreactivity when stored continuously at -20 °C of <0.05% per year. At its 50th meeting, therefore, the Expert Committee on Biological Standardization of WHO established the preparation in vials coded 96/668 as the First International Standard, 1999 for PSA (free) with an assigned content of 1 µg of total PSA per vial and established the preparation in vials coded 96/670 as the First International Standard, 1999 for PSA (90:10) with an assigned content of 1 µg of total PSA per vial. The availability of both preparations will allow validation and calibration of those assays that may distinguish between the various forms of PSA in serum and, in the case of 96/668, to provide a primary standard for the increasing number of assays designed to measure free PSA only (16). The preparations can be obtained upon written request to NIBSC (PO Box 1193, Potters Bar, Herts EN6 3QH, United Kingdom) or by e-mail (standards@nibsc.ac.uk.)
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1-antichymotrypsin; NIBSC, National Institute for Biological Standards and Control; IHR, in-house reference reagent; and GCV, geometric coefficient of variation. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-antichymotrypsin is the major form of prostate-specific antigen in serum of patients with prostatic cancer; assay of the complex improves clinical sensitivity for cancer. Cancer Res 1991;51:222-226.1-antichymotrypsin. Clin Chem 1991;37:1618-1625.1-antichymotrypsin: potential reference material for international standardization of PSA immunoassays. Clin Chem 1995;41:1273-1282.1-antichymotrypsin: influence of cancer volume, location and therapeutic selection of resistant clones. J Urol 1994;152:1510-1514.[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  T. Fortin, A. Salvador, J. P. Charrier, C. Lenz, X. Lacoux, A. Morla, G. Choquet-Kastylevsky, and J. Lemoine Clinical Quantitation of Prostate-specific Antigen Biomarker in the Low Nanogram/Milliliter Range by Conventional Bore Liquid Chromatography-Tandem Mass Spectrometry (Multiple Reaction Monitoring) Coupling and Correlation with ELISA Tests Mol. Cell. Proteomics, May 1, 2009; 8(5): 1006 - 1015. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Sturgeon, M. J. Duffy, U.-H. Stenman, H. Lilja, N. Brunner, D. W. Chan, R. Babaian, R. C. Bast Jr., B. Dowell, F. J. Esteva, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for Use of Tumor Markers in Testicular, Prostate, Colorectal, Breast, and Ovarian Cancers Clin. Chem., December 1, 2008; 54(12): e11 - e79. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. H. Jansen, M. Roobol, C. H. Bangma, and R. H. N. van Schaik Clinical Impact of New Prostate-Specific Antigen WHO Standardization on Biopsy Rates and Cancer Detection Clin. Chem., December 1, 2008; 54(12): 1999 - 2006. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Sturgeon, B. R. Hoffman, D. W. Chan, S.-L. Ch'ng, E. Hammond, D. F. Hayes, L. A. Liotta, E. F. Petricoin, M. Schmitt, O. J. Semmes, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for Use of Tumor Markers in Clinical Practice: Quality Requirements Clin. Chem., August 1, 2008; 54(8): e1 - e10. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Lilja, D. Ulmert, T. Bjork, C. Becker, A. M. Serio, J.-A. Nilsson, P.-A. Abrahamsson, A. J. Vickers, and G. Berglund Long-Term Prediction of Prostate Cancer Up to 25 Years Before Diagnosis of Prostate Cancer Using Prostate Kallikreins Measured at Age 44 to 50 Years J. Clin. Oncol., February 1, 2007; 25(4): 431 - 436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A.R. Kort, F. Martens, H. Vanpoucke, H. L. van Duijnhoven, and M. A. Blankenstein Comparison of 6 Automated Assays for Total and Free Prostate-Specific Antigen with Special Reference to Their Reactivity toward the WHO 96/670 Reference Preparation Clin. Chem., August 1, 2006; 52(8): 1568 - 1574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Stephan, M. Klaas, C. Muller, D. Schnorr, S. A. Loening, and K. Jung Interchangeability of Measurements of Total and Free Prostate-Specific Antigen in Serum with 5 Frequently Used Assay Combinations: An Update Clin. Chem., January 1, 2006; 52(1): 59 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Haese, V. Vaisanen, J. A. Finlay, K. Pettersson, H. G. Rittenhouse, A. W. Partin, D. J. Bruzek, L. J. Sokoll, H. Lilja, and D. W. Chan Standardization of Two Immunoassays for Human Glandular Kallikrein 2 Clin. Chem., April 1, 2003; 49(4): 601 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. G. Blijenberg, M. F. Wildhagen, C. H. Bangma, J. A. Finlay, V. Vaisanen, and F. H. Schroder Comparison of Two Assays for Human Kallikrein 2 Clin. Chem., February 1, 2003; 49(2): 243 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Semjonow, F. Oberpenning, C. Weining, M. Schon, B. Brandt, G. De Angelis, A. Heinecke, M. Hamm, P. Stieber, L. Hertle, et al. Do Modifications of Nonequimolar Assays for Total Prostate-specific Antigen Improve Detection of Prostate Cancer? Clin. Chem., August 1, 2001; 47(8): 1472 - 1475. [Full Text] [PDF]
|
|
|
|
|
|
R. M. Kisabeth Laboratory Adaptations--Changing Expectations Clin. Chem., August 1, 2001; 47(8): 1509 - 1515. [Full Text] [PDF]
|
|
|
|
|
|
U.-H. Stenman Immunoassay Standardization: Is It Possible, Who Is Responsible, Who Is Capable? Clin. Chem., May 1, 2001; 47(5): 815 - 820. [Full Text] [PDF]
|
|
|
|
|
|
D. W. Chan and L. J. Sokoll WHO First International Standards for Prostate-specific Antigen: The Beginning of the End for Assay Discrepancies? Clin. Chem., September 1, 2000; 46(9): 1291 - 1292. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
