| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Automation and Analytical Techniques |
1
Department of Clinical Chemistry, AZR Daniel den Hoed Cancer Center, Groene Hilledijk 301, 3075 EA Rotterdam, The Netherlands.
2
Department of Pathology, Johns Hopkins Medical
Institutions, Baltimore, MD 21205.
3
Institut für Klinische Chemie, Krankenhaus Dresden
Friedrichstadt, D-1067 Dresden, Germany.
4
Department of Clinical Pathology, Toukai University
School of Medicine, Isehara, Japan.
5
Unitat de Recerca del Cancer, Hospital Clinic I
Provincial, Barcelona E-08034, Spain.
6
Abteilung für Klinische und Experimentelle
Endokrinologie, Frauenklinik, Georg-August-Universitaet Goettingen,
D-37075 Goettingen, Germany.
7
Centro de Patologia de la Mama, E-28003 Madrid, Spain.
8
Laboratoire Central de Biochimie, Hopital Henri Mondor,
Creteil Cedex F-94010, France.
9
Boehringer Mannheim GmbH, D-68305 Mannheim, Germany.
a Author for correspondence. Fax 031-10-4391009; e-mail ingen{at}kclh.azr.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
190
kilounits/L; 63% of patients with newly diagnosed ovarian carcinoma
had values >190 kilounits/L. A comparison of CA 125 values obtained
with the Elecsys test and with other common CA 125 tests in monitored
patients being treated for ovarian cancer showed identical patterns. In
conclusion, the Elecsys CA 125 II assay is linear over a broad range,
yields precise and accurate results, is free from interferences, and
compares well with other assays. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
The CA 125 test is calibrated against the Enzymun-Test® CA 125 II (Boehringer Mannheim GmbH). Reagent-specific application data and a master calibration curve are stored on two-dimensional bar codes supplied with the reagents. Two-point calibration in the user's laboratory adapts the master calibration curve to the analyzer. Reagents and calibrators are ready to use.
analytical evaluation of elecsys ca 125 ii
The test was evaluated according to a standardized protocol based
on the concepts of a European Committee for Clinical Laboratory
Standards document (4) in eight clinical laboratories on
fully automated Elecsys 2010 systems.
Imprecision.
Imprecision studies were carried out using
control materials and human pool sera with defined concentration ranges
of CA 125. Within-run imprecision (CV; 21 replicates per analysis) was
determined in seven laboratories, and between-day imprecision
(measurement once daily for 21 days) was determined in five
laboratories. Total imprecision was measured in the laboratory of the
manufacturer following the NCCLS EP5-T protocol.
Accuracy.
Five laboratories participated in a small
interlaboratory survey by measuring CA 125 (twice daily for 5 days) in
two human pool sera that were shipped to the participants in frozen
aliquots. The results obtained in the individual laboratories were
compared with the overall median.
Method comparison studies with the Elecsys assay were carried out in the participating laboratories with the Enzymun-Test CA 125 II on ES 300, 600, and 607 enzyme immunoassays, two microparticle enzyme immunoassays (MEIA 1 on AxSYM® and MEIA 2 on IMx® immunoanalyzers, Abbott Laboratories) and two IRMAs (IRMA 1, ELSA-CA 125 II; CIS bio international; IRMA 2, Centocor® CA 125 II; Centocor Inc.). The tests were performed according to the manufacturers' protocols. Regression equations were calculated according to Passing and Bablok (5).
Analytical range.
The lower limit of detection was determined
by two different methods. In the one, the calibrator matrix without
antigen was measured in nine independent analyses (21 replicates per
analysis) in three laboratories. The lower limit of detection was
defined as the mean +2 SD of these measurements. In the other, a human
serum sample with a very low CA 125 concentration was used for the same
kind of measurements in one laboratory. The linearity of the test was
analyzed by the dilution of human pool sera with CA 125 concentrations
up to 4500 kilounits/L with either low-concentration human pool sera or
Elecsys Diluent Universal according to a recently described
protocol (6). The method was considered linear if the
measured concentration deviated <10% from the expected concentration.
The possible occurrence of a high-dose hook effect was studied by
measuring serial dilutions (with the Elecsys diluent) of samples with
very high concentrations of CA 125.
Interferences.
Possible interference by hemolysis, icteria, or
lipemia was analyzed by addition experiments performed according to a
Société française biologie clinique protocol
(7) as well as by dilution experiments. The effects of
rheumatoid factors, biotin, and dysproteinemia were checked in the
laboratory of the manufacturer. Results were rated acceptable if the
diluted sample or the sample containing a potentially interfering
substance showed a recovery range of 90–110% of the undiluted sample
or the sample without the interfering substance. Eighteen widely used
drugs and 11 cytotoxic drugs were tested as described previously
(8). The comparability of results obtained in serum and
various types of plasma (heparin, EDTA, or citrate plasma) derived from
the same donors was studied in four laboratories.
healthy individuals/reference range
Five laboratories participated in the determination of the
reference range, using sera from 593 nonpregnant women (195 pre- and
293 postmenopausal) and 289 men. Individuals included were >18 years
of age and judged apparently healthy by clinical and clinical chemistry
parameters. In one laboratory, CA 125 was also measured in sera derived
from 32 women in the first trimester of pregnancy (weeks 2–9). The
results were analyzed by nonparametric fractile limits, and the 95th
percentile was taken as the upper limit.
benign diseases
Samples from 342 patients with benign diseases were analyzed.
These diseases could be summarized as follows: 80 gynecological
diseases (ovarian cysts, ovarian metaplasia, endometriosis, uterine
leiomyoma, cervicitis, and squamous metaplasia), 87 liver diseases
(cirrhosis, hepatitis, necrosis, cysts, cholecystitis, and
hemochromatosis), 46 gastrointestinal diseases (acute and chronic
pancreatitis, colitis ulcerosa, Crohn disease, diverticulosis, and
colon polyps), 33 renal insufficiency patients, and 96 other benign
diseases.
malignancies other than ovarian carcinoma
Sera from 505 patients were included: 56 patients with endometrial
carcinoma, 169 patients with gastrointestinal carcinoma (colonic,
pancreatic, gastric, rectal, or esophageal carcinoma), 58 patients with
bronchial carcinoma, 128 patients with breast carcinoma, and 94
patients with other malignancies (genito-urinary, liver, prostate,
hematological, and thyroid).
ovarian carcinoma patients
Samples drawn at the time of primary diagnosis from 100 patients
with ovarian carcinoma were measured [35 stage I, 17 stage II, 25
stage III, and 11 stage IV according to the International Federation of
Gynecology and Obstetrics (FIGO) criteria (9)]. The
majority of patients had nonmucinous ovarian carcinoma. Additional sera
(n = 171) were derived from patients during treatment and follow-up.
monitoring of ovarian carcinoma
Measurements with the Elecsys test and IRMA 2 (Centocor CA 125 II;
Centocor Inc.) were carried out on samples obtained serially from 50
clinically well-characterized patients during treatment and follow-up.
Measurements with Elecsys and Enzymun-Test reagents were performed in
sera collected serially from 10 patients. At least four points per
patient at different monitoring intervals were tested.
ethics considerations
All procedures followed in this study were in accordance with the
guidelines of the various local ethics committees.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
). Total imprecision CVs ranged from 2.4% to 7.8%. CVs >10%
were found in low-concentration samples (<13 kilounits/L). There were
no major differences at comparable concentrations in the CVs for
control sera and human sera.
|
Accuracy.
Measurement of CA 125 in two human pool sera
revealed median recoveries of 94–110% for the low-concentration serum
pool (32 kilounits/L) and 86–112% for the high-concentration serum
pool (332 kilounits/L) by referring the results of the individual
laboratories to the median of all laboratories (data not shown). These
results indicate a good interlaboratory transferability of the Elecsys
CA 125 II test.
The results of the method comparison studies are summarized in Table 2
. Representative examples of method comparison plots are shown
in Fig. 1
. The data presented include only values up to 500 kilounits/L
(measuring range of the comparison methods) to exclude dilution
effects. Similar results were obtained if higher values were included
in the calculation. Although some outliers were detected, good
comparability of the Elecsys CA 125 II test with Enzymun-Test CA 125
II, MEIA 1, and IRMA 2 was obtained. Systematic differences were found
in the comparison with MEIA 2 and IRMA 1, which can be explained by
differences in the standardization of these methods. Comparable
systematic deviations between MEIA 2 and IRMA 2 as well as another CA
125 test were described in a recent paper (10).
|
|
Analytical range.
The limit of detection calculated from
repeated measurements of the zero calibrator was found to be <0.6
kilounits/L in all analyses. This is equivalent to the readable limit
of the analyzer. One laboratory performed measurements in a serum
sample with a CA 125 value of 0.8 kilounits/L. The limit of detection
obtained for serum was 1.1 kilounits/L, which is highly sufficient for
diagnostic purposes.
According to the manufacturer, the assay is linear up to 5000 kilounits/L. This was checked in four laboratories in a total of 24 dilution series covering three decades of CA 125 concentrations, using either human sera or diluent for dilution. On the basis of an acceptance range of 90–110% recovery, the linearity of the assay when serum was used as diluent was confirmed throughout the range studied (highest serum concentration ~4500 kilounits/L) with the exception of the highest dilution, which exceeded the acceptance criteria in 5 of 12 experiments. The acceptance criteria were not met with the low-concentration human pool sera (5–10 kilounits/L) used for dilution. Analytical imprecision at these very low concentrations was a probable cause for this phenomenon. The Elecsys diluent was found to be suitable provided that the concentration of the diluted sample was not below ~500 kilounits/L.
No high-dose hook effect was observed up to 26 000 kilounits/L, the highest concentration available. Higher concentrations can occasionally be encountered in fluids obtained from ovarian cysts; however, in such cases the laboratory will already be alerted to the possible presence of extremely high concentrations by the nature of the sample. It should be noted that a high-dose hook effect can occur when samples with extremely high concentrations are measured. Such samples are encountered very rarely.
Interferences.
Addition experiments revealed no effect by
bilirubin up to 850 µmol/L (recovery, 97.2–100.1%), hemoglobin up
to 1.1 mmol/L (recovery, 92.3–101.2%), or triglycerides up to 10.3
mmol/L (recovery, 100.7–102.7%; data not shown). The concentrations
given are the highest tested and are representative of interferences
encountered in practice. Comparable results were obtained in dilution
experiments. No influence on the results of the Elecsys CA 125 II test
was observed by rheumatoid factor activity (1500 kIU/L), biotin (50
µg/L), or in dysproteinemic sera. The addition of 18 widely used
drugs (fivefold therapeutic concentration) and 11 cytotoxic drugs
(one- or fivefold therapeutic concentration) did not affect the
results.
Comparison studies between serum and heparin, EDTA, and citrate plasma yielded slopes from 0.91 to 1.03, intercepts <2.0 kilounits/L, and correlation coefficients of 1.00, thereby indicating that these types of plasma are suitable as sample material (data not shown).
clinical evaluation
Reference range.
CA 125 was measured in 593 apparently healthy
women (pre- and postmenopausal, ages >18 years) in five laboratories.
A cutoff value of 35 kilounits/L (95th percentile) was obtained for the
Elecsys CA 125 II test. This value is identical to the generally
accepted cutoff for other CA 125 assays, which are described in several
publications (11)(12)(13). Slightly higher 95th percentile
cutoff CA 125 concentrations were found in premenopausal women (38
kilounits/L) compared with postmenopausal women (31 kilounits/L). CA
125 concentrations in women in the first trimester of pregnancy were
frequently >35 kilounits/L (median, 38 kilounits/L), as reported in
the literature (1)(11). A 95th percentile cutoff
of 28 kilounits/L was found in 289 men.
Distribution of CA 125 in Different Groups of Patients.
The
distribution of CA 125 values in ovarian carcinoma patients at the time
of primary diagnosis related to the FIGO stage was compared with that
of women suffering from benign gynecological diseases (Fig. 2
). A large overlap existed in the two groups studied. As shown
previously (11), the magnitude of CA 125 increases was
clearly related to the tumor stage.
|
CA 125 concentrations in malignancies other than ovarian carcinoma and
various benign diseases are presented in Fig. 3
. As is known, CA 125 is an unspecific tumor marker [see Ref.
(1)]. Although the highest concentrations measured with
Elecsys CA 125 II were in ovarian carcinoma patients, very high values
frequently occurred in other malignancies studied. Benign diseases were
also frequently associated with CA 125 values >35 kilounits/L.
Especially high CA 125 concentrations were found in the group of benign
liver disease patients (95th percentile, 494 kilounits/L).
|
Diagnostic Sensitivity and Specificity.
For the calculation of
diagnostic sensitivity and specificity, CA 125 values in 100 ovarian
carcinoma patients at the time of primary diagnosis were compared with
80 patients with benign gynecological diseases (clinically relevant
reference group). The maximum diagnostic efficiency was reached at 150
kilounits/L (specificity, 93%; sensitivity, 69%; Fig. 4
). When specificity was fixed at 95%, following the
recommendation of the Hamburg Group for Standardization of Tumor
Markers (14), the sensitivity was 63% (cutoff, 190
kilounits/L). This sensitivity is in agreement with previously
published results for other CA 125 methods, which described
sensitivities from 50% to 60%
(10)(15)(16). At the historical
cutoff of 65 kilounits/L (13), the specificity decreased to
82%.
|
Apparently healthy women were also used as the reference group for ovarian carcinoma patients to simulate the screening situation. At 35 kilounits/L (95% specificity), a sensitivity of 87% was obtained, which confirms other reports (1)(17) that indicated that measurement of CA 125 alone has insufficient diagnostic efficiency to be used as a screening test for the general population [prevalence of ovarian cancer, 0.05% (17)] for ovarian cancer. In this situation the positive predictive value would be only 0.86%.
Follow-up of Ovarian Carcinoma Patients.
Representative
examples of the 60 patients studied with the Elecsys CA 125 II and
either the Enzymun-Test CA 125 II or IRMA 2 are shown in Fig. 5
. When the results obtained by the different methods were
compared, a clear agreement was found. The CA 125 values were correctly
reflecting the status of the disease and the effect of various
therapeutic measures.
|
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  P. Stieber, R. Molina, D. W. Chan, H. A. Fritsche, R. Beyrau, J. M.G. Bonfrer, X. Filella, T. G. Gornet, T. Hoff, W. Jager, et al. Evaluation of the Analytical and Clinical Performance of the Elecsys(R) CA 15-3 Immunoassay Clin. Chem., December 1, 2001; 47(12): 2162 - 2164. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Elecsys CA 125 II; 
, Enzymun-Test CA 125 II; 
, IRMA 2.