ms_03045838122V19.

CA 15-3 II
Cancer Antigen 15-3

Elecsys 2010
MODULAR ANALYTICS E170
03045838 122 100 cobas e 411
cobas e 601
cobas e 602

English M Streptavidin-coated microparticles (transparent cap), 1 bottle, 6.5 mL:

Please note Streptavidin-coated microparticles 0.72 mg/mL; preservative.
The measured CA 15‑3 value of a patient’s sample can vary depending R1 Anti-CA 15-3-Ab~biotin (gray cap), 1 bottle, 10 mL:
on the testing procedure used. The laboratory finding must therefore Biotinylated monoclonal antibody (115D8; mouse) 1.75 mg/L;
always contain a statement on the CA 15‑3 assay method used. CA 15‑3 phosphate buffer 20 mmol/L, pH 6.0; preservative.
values determined on patient samples by different testing procedures
R2 Anti-CA 15-3-Ab~Ru(bpy) (black cap), 1 bottle, 10 mL:
cannot be directly compared with one another and could be the cause of
erroneous medical interpretations. If there is a change in the CA 15‑3 Monoclonal anti-CA 15-3 antibody (DF3; mouse) labeled with
assay procedure used while monitoring therapy, then the CA 15‑3 values ruthenium complex 10 mg/L; phosphate buffer 100 mmol/L, pH 7.0;
obtained upon changing over to the new procedure must be confirmed by preservative.
parallel measurements with both methods.
Precautions and warnings
Intended use For in vitro diagnostic use.
Immunological in vitro assay for quantitative determination of CA 15‑3 in Exercise the normal precautions required for handling all laboratory
human serum and plasma to aid in the management of breast cancer reagents.
patients. In conjunction with other clinical and diagnostic procedures, serial Disposal of all waste material should be in accordance with local guidelines.
testing with this assay is an aid Safety data sheet available for professional user on request.
Avoid foam formation in all reagents and sample types (specimens,
▪ in the early detection of recurrence in previously treated stage II and III calibrators and controls).
breast cancer patients
▪ for monitoring response to therapy in metastatic breast cancer patients Reagent handling
The electrochemiluminescence immunoassay “ECLIA” is intended for use The reagents in the kit have been assembled into a ready‑for‑use unit that
on Elecsys and cobas e immunoassay analyzers. cannot be separated.
All information required for correct operation is read in from the respective
Summary reagent barcodes.
The CA 15‑3 values measured are defined by the use of the monoclonal
antibodies (MAb) 115D8 and DF3 in a sandwich assay. MAb 115D8 is Storage and stability
directed against human milk fat globule membranes,1,2 whereas MAb DF3 Store at 2‑8 °C.
is directed against the membrane fraction from human metastatic breast Do not freeze.
cancer.3
Store the Elecsys reagent kit upright in order to ensure complete
The 115D8 and DF3‑reactive determinants detected by the test are located availability of the microparticles during automatic mixing prior to use.
on a glycoprotein having a molecular weight of > 400 kDa. The antigen is
termed MAM‑6 and belongs to a subgroup of sialylated glycoproteins Stability:
designated polymorphic epithelial mucin (PEM). Polymorphic epithelial
mucins (PEMs) are normally found in the luminal secretion of glandular unopened at 2‑8 °C up to the stated expiration date
cells and do not circulate in the blood. When cells become malignant and after opening at 2‑8 °C 12 weeks
their basal membranes permeable, PEMs are detectable in serum using the
CA 15‑3 assay.4 on the analyzers 5 weeks
Test principle Specimen collection and preparation
Sandwich principle. Total duration of assay: 18 minutes. Only the specimens listed below were tested and found acceptable.
▪ 1st incubation: 20 µL of sample are automatically prediluted 1:10 with Serum collected using standard sampling tubes or tubes containing
Diluent Universal. The antigen (in 20 µL of prediluted sample), a separating gel.
biotinylated monoclonal CA 15‑3‑specific antibody, and a monoclonal Li-, Na-, NH ‑heparin and K3‑EDTA plasma.
CA 15‑3‑specific antibody labeled with a ruthenium complexa) react to
form a sandwich complex. Results obtained when using sodium citrate plasma are about 24 % lower
than compared to those obtained with serum.
▪ 2nd incubation: After addition of streptavidin-coated microparticles, the Criterion: Recovery within 90‑110 % of serum value or slope
complex becomes bound to the solid phase via interaction of biotin and 0.9‑1.1 + intercept within < ± 2x analytical sensitivity (LDL) + coefficient of
streptavidin. correlation > 0.95.
▪ The reaction mixture is aspirated into the measuring cell where the Stable for 5 days at 2‑8 °C, 3 months at ‑20 °C.5
microparticles are magnetically captured onto the surface of the
electrode. Unbound substances are then removed with The sample types listed were tested with a selection of sample collection
ProCell/ProCell M. Application of a voltage to the electrode then induces tubes that were commercially available at the time of testing, i.e. not all
chemiluminescent emission which is measured by a photomultiplier. available tubes of all manufacturers were tested. Sample collection systems
from various manufacturers may contain differing materials which could
▪ Results are determined via a calibration curve which is instrument- affect the test results in some cases. When processing samples in primary
specifically generated by 2‑point calibration and a master curve provided tubes (sample collection systems), follow the instructions of the tube
via the reagent barcode. manufacturer.
a) Tris(2,2'-bipyridyl)ruthenium(II)-complex (Ru(bpy) ) Centrifuge samples containing precipitates before performing the assay.
Reagents - working solutions Do not use heat‑inactivated samples.
The reagent rackpack is labeled as CA 15‑3 II.

2017-08, V 19.0 English 1/4

ms_03045838122V19.0

CA 15-3 II
Cancer Antigen 15-3

Do not use samples and controls stabilized with azide. ▪ after 12 weeks when using the same reagent lot
Ensure the samples, calibrators and controls are at 20‑25 °C prior to ▪ after 7 days (when using the same reagent kit on the analyzer)
measurement.
▪ as required: e.g. quality control findings outside the defined limits
Due to possible evaporation effects, samples, calibrators and controls on
the analyzers should be analyzed/measured within 2 hours. Quality control
For quality control, use PreciControl Tumor Marker.
Materials provided
In addition, other suitable control material can be used.
See “Reagents – working solutions” section for reagents.
Controls for the various concentration ranges should be run individually at
Materials required (but not provided) least once every 24 hours when the test is in use, once per reagent kit, and
▪  03045846122, CA 15-3 II CalSet, 4 x 1 mL following each calibration.
▪  11776452122, PreciControl Tumor Marker, for 2 x 3 mL each of The control intervals and limits should be adapted to each laboratory’s
PreciControl Tumor Marker 1 and 2 individual requirements. Values obtained should fall within the defined
limits. Each laboratory should establish corrective measures to be taken if
▪  11732277122, Diluent Universal, 2 x 16 mL sample diluent or values fall outside the defined limits.
 03183971122, Diluent Universal, 2 x 36 mL sample diluent
Follow the applicable government regulations and local guidelines for
▪ General laboratory equipment quality control.
▪ Elecsys 2010, MODULAR ANALYTICS E170 or cobas e analyzer Calculation
Accessories for Elecsys 2010 and cobas e 411 analyzers: The analyzer automatically calculates the analyte concentration of each
▪  11662988122, ProCell, 6 x 380 mL system buffer sample (either in U/mL or kU/L).
▪  11662970122, CleanCell, 6 x 380 mL measuring cell cleaning Limitations - interference
solution The assay is unaffected by icterus (bilirubin < 1112 µmol/L or < 65 mg/dL),
▪  11930346122, Elecsys SysWash, 1 x 500 mL washwater additive hemolysis (Hb < 1.9 mmol/L or < 3.0 g/dL), lipemia (Intralipid < 1500 mg/dL)
and biotin (< 409 nmol/L or < 100 ng/mL).
▪  11933159001, Adapter for SysClean Criterion: Recovery within ± 10 % of initial value.
▪  11706802001, Elecsys 2010 AssayCup, 60 x 60 reaction vessels Samples should not be taken from patients receiving therapy with high
▪  11706799001, Elecsys 2010 AssayTip, 30 x 120 pipette tips biotin doses (i.e. > 5 mg/day) until at least 8 hours following the last biotin
Accessories for MODULAR ANALYTICS E170, cobas e 601 and administration.
cobas e 602 analyzers: No interference was observed from rheumatoid factors up to a
concentration of 1500 IU/mL.
▪  04880340190, ProCell M, 2 x 2 L system buffer
Typically, no high‑dose hook effectb) can be observed at CA 15‑3
▪  04880293190, CleanCell M, 2 x 2 L measuring cell cleaning concentrations up to 20000 U/mL. However, due to the heterogeneous
solution nature of the CA 15‑3 antigen a high-dose hook effect below this value
▪  03023141001, PC/CC‑Cups, 12 cups to prewarm ProCell M and cannot be completely excluded. In case of an unexpected low result, the
CleanCell M before use sample should be diluted 1:10 (refer to chapter “Dilution”) and tested again.
▪  03005712190, ProbeWash M, 12 x 70 mL cleaning solution for run In vitro tests were performed on 28 commonly used pharmaceuticals. No
finalization and rinsing during reagent change interference with the assay was found.
▪  12102137001, AssayTip/AssayCup Combimagazine M, In rare cases, interference due to extremely high titers of antibodies to
48 magazines x 84 reaction vessels or pipette tips, waste bags analyte‑specific antibodies, streptavidin or ruthenium can occur. These
effects are minimized by suitable test design.
▪  03023150001, WasteLiner, waste bags
For diagnostic purposes, the results should always be assessed in
▪  03027651001, SysClean Adapter M conjunction with the patient’s medical history, clinical examination and other
Accessories for all analyzers: findings.
▪  11298500316, Elecsys SysClean, 5 x 100 mL system cleaning b) High-dose hook effect: A sample with a true concentration clearly above the measuring range,
but found within the measuring range.
solution
Limits and ranges
Assay
Measuring range
For optimum performance of the assay follow the directions given in this
document for the analyzer concerned. Refer to the appropriate operator’s 1.00-300 U/mL (defined by the lower detection limit and the maximum of the
manual for analyzer‑specific assay instructions. master curve). Values below the lower detection limit are reported as
< 1.00 U/mL. Values above the measuring range are reported as
Resuspension of the microparticles takes place automatically prior to use. > 300 U/mL (or up to 3000 U/mL for 10‑fold diluted samples).
Read in the test‑specific parameters via the reagent barcode. If in
exceptional cases the barcode cannot be read, enter the 15‑digit sequence Lower limits of measurement
of numbers. Lower detection limit of the test
Bring the cooled reagents to approximately 20 °C and place on the reagent Lower detection limit: < 1.00 U/mL
disk (20 °C) of the analyzer. Avoid foam formation. The system The lower detection limit represents the lowest measurable analyte level
automatically regulates the temperature of the reagents and the that can be distinguished from zero. It is calculated as the value lying two
opening/closing of the bottles. standard deviations above that of the lowest standard (master calibrator,
Calibration standard 1 + 2 SD, repeatability study, n = 21).
Traceability: This method has been standardized against the Elecsys Dilution
CA 15‑3 assay. This in turn has been standardized against the Use Diluent Universal for automatic sample predilution. Samples with
Enzymun‑Test CA 15‑3 method and CA 15‑3 RIA by Fujirebio Diagnostics. CA 15‑3 concentrations above the measuring range despite predilution
Every Elecsys reagent set has a barcoded label containing specific must be diluted 1:10 with Diluent Universal (either manually for all analyzers
information for calibration of the particular reagent lot. The predefined or automatically by the MODULAR ANALYTICS E170, cobas e 601 and
master curve is adapted to the analyzer using the relevant CalSet. cobas  e 602 analyzers). The concentration of the diluted sample must be
Calibration frequency: Calibration must be performed once per reagent lot > 30 U/mL.
using fresh reagent (i.e. not more than 24 hours since the reagent kit was After manual dilution, multiply the result by the dilution factor.
registered on the analyzer). Renewed calibration is recommended as
follows:

2/4 2017-08, V 19.0 English

ms_03045838122V19.0

CA 15-3 II
Cancer Antigen 15-3

After dilution by the analyzers, the MODULAR ANALYTICS E170, Subjects < 25 25-50 > 50-200 > 200
cobas e 601 and cobas e 602 software automatically takes the dilution into total U/mL U/mL U/mL U/mL
account when calculating the sample concentration.
N Classification in percent (%)
Expected values
• Healthy subjects: UICC III 77 53 30 14 3
Results of a reference range study using a panel of samples from 378 UICC IV 24 25 17 37 21
apparently healthy non‑pregnant females (Roche study No. RD000788) Recurrent 75 15 25 36 24
Percentile (%) U/mL Confidence interval (U/mL) Disease
95 26.4 25.2-27.9 Each laboratory should investigate the transferability of the expected values
to its own patient population and if necessary determine its own reference
97.5 28.5 26.9-34.5 ranges.
99 34.5 28.7-57.8 Specific performance data
• Patients with benign diseases and pregnant women: Representative performance data on the analyzers are given below.
Results obtained in individual laboratories may differ.
Relative distribution of CA 15‑3 concentrations in patients with benign
disease and pregnancy Precision
Precision was determined using Elecsys reagents, pooled human sera and
Subjects < 25 25-50 > 50-200 > 200 controls in accordance with a modified protocol (EP5‑A) of the CLSI
total U/mL U/mL U/mL U/mL (Clinical and Laboratory Standards Institute): 6 times daily for 10 days
N Classification in percent (%) (n = 60); repeatability on MODULAR ANALYTICS E170 analyzer, n = 21.
The following results were obtained:
Liver, pancreas, 109 84 16 0 0
gall bladder Elecsys 2010 and cobas e 411 analyzers
Breast 58 88 12 0 0 Repeatability Intermediate
precision
Gynecological 42 83 12 5 0
diseases Sample Mean SD CV SD CV
U/mL U/mL % U/mL %
Renal failure 37 81 19 0 0
Human serum 1 38.0 0.81 2.1 1.38 3.6
Urological 34 82 18 0 0
diseases Human serum 2 85.5 2.72 3.2 3.66 4.3
Bacterial 28 96 4 0 0 Human serum 3 179 4.56 2.6 6.60 3.7
infection PreciControl TMd)1 24.5 0.62 2.5 0.87 3.6
Pregnancy 34 97 0 3 0 PreciControl TM2 67.6 2.48 3.7 2.83 4.2
• Patients with malignant diseases (others than breast): d) TM = Tumor Marker
Relative distribution of CA 15‑3 concentrations in individuals with MODULAR ANALYTICS E170, cobas e 601 and cobas e 602 analyzers
malignancy other than breast
Repeatability Intermediate precision
Subjects < 25 25-50 > 50-200 > 200
Sample Mean SD CV Mean SD CV
total U/mL U/mL U/mL U/mL
U/mL U/mL % U/mL U/mL %
N Classification in percent (%)
Human serum 1 18.9 0.24 1.3 20.1 0.64 3.2
Stomach-Cac) 36 75 14 8 3
Human serum 2 76.4 1.07 1.4 79.0 3.08 3.9
Hepatocellular-Ca 37 60 32 3 5
Human serum 3 148 1.72 1.2 156 7.75 5.0
Lung-Ca 38 82 13 5 0
PreciControl TM1 20.3 0.24 1.2 21.3 0.89 4.2
Ovarian-Ca 34 47 21 29 3
PreciControl TM2 47.6 0.70 1.5 49.6 1.82 3.7
Gynecological-Ca 5 40 20 40 0
Method comparison
Prostate-Ca 48 79 17 4 0
A comparison of the Elecsys CA 15‑3 II assay (y) with the Elecsys CA 15‑3
Colorectal-Ca 40 93 7 0 0 assay (x) using clinical samples gave the following correlations:
Pancreatic-Ca 40 65 33 2 0 Number of samples measured: 52
c) Ca = Carcinoma
Passing/Bablok regression6
• Patients with breast cancer: Slope: 1.06 (95 % confidence range: 1.01‑1.15)
Relative distribution of CA 15‑3 concentrations in patients with breast Intercept: 2.66 (95 % confidence range: ‑0.99‑5.97)
malignancy. The staging of patients according to UICC criteria was Coefficient of correlation: 0.965
performed at primary diagnosis before any treatment. The patients The sample concentrations were between approximately 6 and 280 U/mL.
diagnosed with recurrent disease had developed metastases (M1).
Analytical specificity
Subjects < 25 25-50 > 50-200 > 200 The Elecsys CA 15‑3 II tumor marker assay is based on the monoclonal
total U/mL U/mL U/mL U/mL 115D8 and DF3 antibodies which are only available from Fujirebio
Diagnostics, its licensees and its representatives. The performance
N Classification in percent (%) characteristics of test procedures using these antibodies cannot be
UICC I 56 88 12 0 0 assumed for test methods using other antibodies.
UICC II 126 85 13 2 0

2017-08, V 19.0 English 3/4

ms_03045838122V19.0

CA 15-3 II
Cancer Antigen 15-3

Clinical performance in follow-up 6 Bablok W, Passing H, Bender R, et al. A general regression procedure
Patients diagnosed with breast cancer were examined in a retrospective for method transformation. Application of linear regression procedures
study (at least 4 samples/patient during follow-up study) and classified for method comparison studies in clinical chemistry, Part III.
according to progression/recurrence, regression/response or no change of J Clin Chem Clin Biochem 1988 Nov;26(11):783-790.
disease based on the clinical information (medical imaging and other For further information, please refer to the appropriate operator’s manual for
clinical investigations). The CA 15‑3 concentrations were measured in the analyzer concerned, the respective application sheets, the product
parallel and from their relative change, marker-increase (defined as > 25 % information and the Method Sheets of all necessary components (if
and resulting in values > 10 U/mL after increase) and marker-decrease (by available in your country).
> 25 %) were determined.
A point (period/stop) is always used in this Method Sheet as the decimal
Early detection of recurrence separator to mark the border between the integral and the fractional parts of
Of 44 patients previously diagnosed and treated for stage II or stage III a decimal numeral. Separators for thousands are not used.
breast cancer (no evidence of disease at the beginning of the follow-up
study), 20 had recurrence of disease according to clinical symptoms. Of CA 15‑3 is a registered trademark of
these, 18 (90 %) had CA 15‑3 levels that increased by > 25 %. Of the Fujirebio Diagnostics, Inc.
24 patients that did not experience recurrence, 17 (71 %) had CA 15‑3
levels that remained the same (within 25 %) or decreased. Symbols
Table for early detection of recurrence: Roche Diagnostics uses the following symbols and signs in addition to
those listed in the ISO 15223‑1 standard.
Clinical diagnosis: Recurrence of disease
Contents of kit
Yes No Total
Analyzers/Instruments on which reagents can be used
Yes 18 7 25
CA 15‑3 increase Reagent
No 2 17 19
> 25 % Calibrator
Total 20 24 44
Volume after reconstitution or mixing
The corresponding results for positive predictive value (PPV) and negative
predictive value (NPV) with the 95 % confidence interval as derived from
COBAS, COBAS E, ELECSYS, MODULAR and PRECICONTROL are trademarks of Roche. INTRALIPID is a
the table are: trademark of Fresenius Kabi AB.
Positive predictive value: 72 % (50‑87 %) All other product names and trademarks are the property of their respective owners.
Negative predictive value: 90 % (66‑98 %) Significant additions or changes are indicated by a change bar in the margin.
© 2013, Roche Diagnostics
Monitoring response to therapy
Of 60 patients diagnosed with metastatic breast cancer, 23 responded to
therapy as determined by clinical criteria. Of these, 18 (78 %) had CA 15‑3
levels that decreased by > 25 %. Of the 37 patients that did not respond to
therapy, 29 (78 %) had levels of CA 15-3 that remained the same (within Roche Diagnostics GmbH, Sandhofer Strasse 116, D-68305 Mannheim
www.roche.com
25 %) or increased.
Table for response to therapy:
Clinical diagnosis: Response to therapy
Yes No Total
Yes 18 8 26
CA 15‑3 decrease
No 5 29 34
> 25 %
Total 23 37 60
The corresponding results for positive predictive value (PPV) and negative
predictive value (NPV) with the 95 % confidence interval as derived from
the table are:
Positive predictive value: 69 % (48‑85 %)
Negative predictive value: 85 % (68‑95 %)
References
1 Hilkens J, Buijs F, Hilgers J, et al. Monoclonal antibodies against
human milk-fat globule membranes detecting differentiation antigens of
the mammary gland. Prot Biol Fluids 1982;29:813-816.
2 Hilkens J, Buijs F, Hilgers J, et al. Monoclonal antibodies against
human milk-fat globule membranes detecting differentiation antigens of
the mammary gland and its tumors. Int J Cancer 1984;34:197-206.
3 Kufe D, Inghirami G, Abe M, et al. Differential reactivity of a novel
monoclonal antibody (DF3) with human malignant versus benign breast
tumor. Hybridoma 1984;(3):223-232.
4 Sekine H, Ohno T, Kufe DW. Purification and characterization of a high
molecular weight glycoprotein detectable in human milk and breast
carcinomas. J Immunol 1985;135(5):3610-3615.
5 Guder WG, Narayanan S, Wisser H, et al. List of Analytes;
Preanalytical Variables. Brochure in: Samples: From the Patient to the
Laboratory. GIT-Verlag, Darmstadt 1996;8. ISBN 3-928865-22-6:8-9.

4/4 2017-08, V 19.0 English