September 2010
Environmental Technology
Verification Report
ABRAXIS
MICROCYSTIN TEST KITS:
ADDA ELISA TEST KIT
DM ELISA TEST KIT
STRIP TEST KIT
Prepared by
Battelle
Baitene
the Business of Innovation
Under a cooperative agreement with
U.S. Environmental Protection Agency
ET1/ET1/ET1/
-------
September 2010
Environmental Technology
Verification Report
ETV Advanced Monitoring Systems Center
ABRAXIS
MICROCYSTIN TEST KITS:
ADDA ELISA TEST KIT
DM ELISA TEST KIT
STRIP TEST KIT
by
Ryan James, Anne Gregg, and Amy Dindal, Battelle
John McKernan, U.S. EPA
-------
Notice
The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed, or partially funded and collaborated in, the research described herein. It
has been subjected to the Agency's peer and administrative review and has been approved for
publication. Any opinions expressed in this report are those of the author(s) and do not
necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred.
Any mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
11
-------
Foreword
The EPA is charged by Congress with protecting the nation's air, water, and land resources.
Under a mandate of national environmental laws, the Agency strives to formulate and implement
actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, the EPA's Office of Research and
Development provides data and science support that can be used to solve environmental
problems and to build the scientific knowledge base needed to manage our ecological resources
wisely, to understand how pollutants affect our health, and to prevent or reduce environmental
risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA to
verify the performance characteristics of innovative environmental technology across all media
and to report this objective information to permitters, buyers, and users of the technology, thus
substantially accelerating the entrance of new environmental technologies into the marketplace.
Verification organizations oversee and report verification activities based on testing and quality
assurance protocols developed with input from major stakeholders and customer groups
associated with the technology area. ETV consists of six environmental technology centers.
Information about each of these centers can be found on the Internet at http://www.epa.gov/etv/.
Effective verifications of monitoring technologies are needed to assess environmental quality
and to supply cost and performance data to select the most appropriate technology for that
assessment. Under a cooperative agreement, Battelle has received EPA funding to plan,
coordinate, and conduct such verification tests for "Advanced Monitoring Systems for Air,
Water, and Soil" and report the results to the community at large. Information concerning this
specific environmental technology area can be found on the Internet at
http://www.epa.gov/etv/centers/centerl.html.
in
-------
Acknowledgments
The authors wish to acknowledge the contribution of the many individuals, without whom, this
verification testing would not have been possible. Quality assurance (QA) oversight was
provided by Michelle Henderson, U.S. EPA, and Zachary Willenberg, Battelle. We thank David
Schumacher of the Nebraska Department of Environmental Quality and Robert Waters from the
New York, Suffolk County Department of Health Services for their support of this verification
by providing recreational water samples, and also for peer review of the test/QA plan. We
acknowledge the support from Daniel Snow and David Cassada from the University of Nebraska
Water Sciences Laboratory for developing and validating the reference method in addition to
analyzing the samples for the verification. Finally, we want to thank Andrew Lincoff, U.S. EPA
and Daniel Snow of the University of Nebraska for their review of the test/QA plan and/or this
verification report.
IV
-------
Contents
Page
Foreword iii
Acknowledgments iv
List of Abbreviations ix
Chapter 1 Background 1
Chapter 2 Technology Description 2
2.1 Microcystin ADDA ELISA Test Kit 2
2.2 Microcystin DM ELISA Test Kit 3
2.3 Microcystin Strip Test 4
Chapter 3 Test Design and Procedures 6
3.1 Test Overview 6
3.2 Experimental Design 6
3.3 Test Procedures 7
3.3.1 QC Samples 7
3.3.2 PT Samples 7
3.3.3 RW Samples 9
Chapter 4 Quality Assurance/Quality Control 10
4.1 Reference Method Quality Control 10
4.2 Audits 12
4.2.1 Performance Evaluation Audit 12
4.2.2 Technical Systems Audit 13
4.2.3 Data Quality Audit 13
Chapters Statistical Methods 15
5.1 Accuracy 15
5.2 Linearity 15
5.3 Precision 16
5.4 Method Detection Limit 16
5.5 Inter-Kit Lot Reproducibility 16
5.6 Matrix Effects 16
Chapter 6 Test Results for the Abraxis ADDA ELISA Test Kit 17
6.1 ADDA Test Kit Summary 17
6.2 Test Kit QC Samples 17
6.3 PT Samples 18
6.3.1 Accuracy 19
6.3.2 Precision 22
6.3.3 Linearity 24
6.3.4 Method Detection Limit 25
6.3.5 Inter-Kit Lot Reproducibility 26
6.3.6 Matrix Effect 26
6.4 RW Sample Results 30
6.5 Operational Factors 32
6.5.1 Ease of Use 32
6.5.2 Cost and Consumables 32
-------
Chapter? Test Results for the Abraxis DM ELISA Test Kit 33
7.1 DM Test Kit Summary 33
7.2 Test Kit QC Sample 33
7.3 PT Samples 34
7.3.1 Accuracy 34
7.3.2 Precision 38
7.3.3 Linearity 40
7.3.4 Method Detection Limit 41
7.3.5 Inter-Kit Lot Reproducibility 41
7.3.6 Matrix Effects 42
7.4 RW Sample Results 45
7.5 Operational Factors 47
7.5.1 Ease of Use 47
7.5.2 Cost and Consumables 47
Chapters Test Results for the Abraxis Strip Test Kit 48
8.1 Abraxis Strip Test Summary 48
8.2 Test Kit QC Sample 48
8.3 PT Samples 49
8.3.1 DI Water Samples 49
8.3.2 Matrix Interference Samples 52
8.4 RW Samples 55
8.5 Operational Factors 57
8.5.1 Ease of Use 57
8.5.2 Cost and Consumables 57
Chapter 9 Performance Summary for the Abraxis ADDA, DM, and Strip Test 58
9.1 Performance Summary for the ADDA ELISA Test Kit 58
9.2 Performance Summary for the DM ELISA Test Kit 59
9.3 Performance Summary for the Strip Test Kit 61
Chapter 10 References 63
APPENDIX A Reference Laboratory Method Detection Limit Memo 64
APPENDIX B Abraxis Test Kit Raw Data 66
VI
-------
Tables
Table 1. Summary of Test Samples 8
Table 2. DQIs and Summary of Reference Method QC Results 11
Table 3. Summary of Reference Method CCV Percent Recoveries 12
Table 4. PEA Results: Analytical Comparison of Microcystin Standards 12
Table 5. PEA Results: Evaluation of Extracted Low Level Water Sample 13
Table 6 RB Sample Results for the Abraxis ADDA Test Kit 18
Table 7. Positive Control Sample Results for the Abraxis ADDA Test Kit 18
Table 8. Abraxis ADDA Test Kit Sample Results and Reference Method Results for LR 19
Table 9. Abraxis ADDA Test Kit Sample Results and Reference Method Results for LA 20
Table 10. Abraxis ADDA Test Kit Sample Results and Reference Method Results for RR 21
Table 11. Abraxis ADDA Test Kit Precision Results 23
Table 12. Detection Limit Results for the Abraxis ADDA Test Kit 25
Table 13. Inter-kit lot Comparison of Kit Calibration Standards for the ADDA Test Kit 26
Table 14. RW Matrix Interference Sample Results for the Abraxis ADDA Test Kit 28
Table 15 Chlorophyll-a Interferent Sample Results for the Abraxis ADDA Test Kit 29
Table 16. Statistical Comparisons between Interference Samples for the Abraxis
ADDA Test Kit 30
Table 17. Recreational Water Sample Results for the Abraxis ADDA Test Kit 31
Table 18. RB Sample Results for the Abraxis DM Test Kit 34
Table 19. Positive Control Sample Results for the Abraxis DM Test Kit 34
Table 20. Abraxis DM Test Kit Sample Results and Reference Method Results for LR 35
Table 21. Abraxis DM Test Kit Sample Results and Reference Method Results for LA 36
Table 22. Abraxis DM Test Kit Sample Results and Reference Method Results for RR 37
Table 23. Abraxis DM Test Kit Precision Results 39
Table 24. Detection Limit Results for the Abraxis DM Test Kit 41
Table 25. Inter-kit lot Comparison of Kit Calibration Standards for the DM Test Kit 42
Table 26. RW Matrix Interferent Sample Results for the Abraxis DM Test Kit 43
Table 27. Chlorophyll-a Interferent Sample Results for the Abraxis DM Test Kit 44
Table 28. Statistical Comparisons between Interference Samples for the Abraxis
DM Test Kit 45
Table 29. Recreational Water Sample Results for the Abraxis DM Test Kit 46
Table 30. RB Sample Results for the Abraxis Strip Test Kit 49
Table 31. Abraxis Strip Test Kit Microcystin-LR DI Water Sample Results 50
Table 32. Abraxis Strip Test Kit Microcystin-LA DI Water Sample Results 51
Table 33. Abraxis Strip Test Kit Microcystin-RR DI Water Sample Results 52
Table 34. RW Matrix Interferent Sample Results for the Strip Test Kit 53
Table 35. Matrix Interferent Sample Results for the Abraxis Strip Test Kit 54
Table 36. Recreational Water Sample Results for the Abraxis Strip Test Kit 56
Table 37. Abraxis ADDA Test Kit Performance Summary 58
Table 38. Abraxis DM Test Kit Performance Summary 59
Table 39. Abraxis Strip Test Kit Performance Summary 61
vn
-------
Figures
Figure 1. Microtiter Plate ELISA Test Kit 3
Figure 2. Microcystin Strip Test 5
Figures. Linearity for the Abraxis ADDA Test Kit for LR 24
Figure 4. Linearity for the Abraxis ADDA Test Kit for LA 24
Figure 5. Linearity for the Abraxis ADDA Test Kit for RR 25
Figure 6. Linearity for the Abraxis DM Test Kit for LR 40
Figure 7. Linearity for the Abraxis DM Test Kit for LA 40
Figure 8. Linearity for the Abraxis DM Test Kit for RR 41
Vlll
-------
List of Abbreviations
ADQ audit of data quality
AMS Advanced Monitoring Systems
ASTM American Society for Testing and Materials
CCV continuing calibration verification
CR cross reactivity
CV coefficient of variation
DI deionized
DQI data quality indicator
DQO data quality objective
ELISA Enzyme-Linked Immunosorbent Assay
EPA Environmental Protection Agency
ETV Environmental Technology Verification
HPLC high pressure liquid chromatography
HRP Horseradish Peroxidase
LC-MS-MS liquid chromatography tandem mass spectrometry
LFM laboratory fortified matrix
LOQ limit of quantification
MDL method detection limit
mg/L milligram per liter
mL milliliter
NDEQ Nebraska Department of Environmental Quality
run nanometer
NRC National Research Council
NRMRL National Risk Management Research Laboratory
OD optical density
ppb parts per billion
%D percent difference
PEA performance evaluation audit
PT performance test
QA quality assurance
QAO quality assurance officer
QC quality control
QMP quality management plan
%R percent recovery
r2 coefficient of determination
RB reagent blank
RPD relative percent difference
RSD relative standard deviation
RW recreational water
S standard deviation
SOP standard operating procedure
SPE solid phase extraction
TMB tetramethylbenzidine
ix
-------
TQAP Test/Quality Assurance Plan
TSA technical systems audit
(ig/L microgram per liter
WHO World Health Organization
WSL Water Sciences Laboratory (University of Nebraska)
-------
Chapter 1
Background
The U.S. Environmental Protection Agency (EPA) supports the Environmental Technology
Verification (ETV) Program to facilitate the deployment of innovative environmental
technologies through performance verification and dissemination of information. The goal of the
ETV Program is to further environmental protection by accelerating the acceptance and use of
improved and cost-effective technologies. ETV seeks to achieve this goal by providing high-
quality, peer-reviewed data on technology performance to those involved in the design,
distribution, financing, permitting, purchase, and use of environmental technologies.
ETV works in partnership with recognized testing organizations; with stakeholder groups
consisting of buyers, vendor organizations, and permitters; and with the full participation of
individual technology developers. The program evaluates the performance of innovative
technologies by developing test plans that are responsive to the needs of stakeholders,
conducting field or laboratory tests (as appropriate), collecting and analyzing data, and preparing
peer-reviewed reports. All evaluations are conducted in accordance with rigorous quality
assurance (QA) protocols to ensure that data of known and adequate quality are generated and
that the results are defensible. The definition of ETV verification is to establish or prove the
truth of the performance of a technology under specific, pre-determined criteria or protocols and
a strong quality management system. The highest-quality data are assured through
implementation of the ETV Quality Management Plan. ETV does not endorse, certify, or
approve technologies.
The EPA's National Risk Management Research Laboratory (NRMRL) and its verification
organization partner, Battelle, operate the Advanced Monitoring Systems (AMS) Center under
the ETV program. The AMS Center evaluated the performance of three technologies offered by
Abraxis: Microcystin (ADDA) ELISA Test Kit, the Microcystin (DM) ELISA Test Kit, and
Microcystin Strip Test Kits.
-------
Chapter 2
Technology Description
This verification report provides results for the verification testing of three Abraxis Test Kits.
Following are descriptions of the Microcystin (ADDA) ELISA Test Kit, the Microcystin (DM)
ELISA Test Kit, and Microcystin Strip Test Kit (hereafter these technologies will be referred to
as the ADDA, DM, and Strip Test, respectively), based on information provided by the vendor.
The information provided below was not verified in this test.
2.1 Microcystin ADDA ELISA Test Kit
The ADDA Test Kit is an enzyme-linked immunosorbent assay (ELISA) for the congener
independent determination of microcystins and nodularins in water samples. The assay utilizes
polyclonal antibodies that have been raised against the ADDA moiety of the molecule, allowing
for the detection of microcystins and nodularin variants (over 80 variants are currently known) in
drinking, surface, and groundwater at levels below World Health Organization (WHO)
guidelines.
The test is an indirect competitive ELISA and is based on the recognition of microcystins,
nodularins and their variants by a polyclonal sheep antibody. When present in a sample,
microcystins and nodularins compete with a microcystins-protein analog that is immobilized on
wells of a microtiter plate for the binding sites of antibodies in solution. After a washing step, a
second labeled antibody is added and incubated, antibody- Horseradish Peroxidase (HRP). After
a washing step and addition of a substrate/chromogen solution, a color signal is generated. The
intensity of the color is inversely proportional to the concentration of the microcystins/nodularins
present in the sample. The color reaction is stopped after a specified time and the color is
analyzed using a plate photometer to obtain the optical density (OD) at a wavelength of 450
nanometer (nm).
The ADDA Test Kit is not able to distinguish between different microcystin variants. Results
from the ADDA test kit are calibrated with respect to a single variant, microcystin-LR.
However, other microcystin variants are known (based on information provided by Abraxis) to
react to different extents with the antibodies used for detection which is referred to as the cross
reactivity (CR) of the variant. For this verification test, cross reactivity values were used to
quantify results for different variants based on the LR calibration.
The ADDA ELISA Test Kit (Figure 1) contains:
• Microtiter plate coated with an analog of microcystins conjugated to a protein;
• Standards (6) and positive control (1): 0, 0.15, 0.40, 1.0, 2.0, 5.0 parts per billion (ppb);
Positive control has a concentration of 0.75 ppb;
• Antibody solution (monoclonal anti-Microcystins);
• Anti-Sheep-HRP Conjugate;
• Wash Solution 5X Concentrate;
• Color Solution, tetramethylbenzidine (TMB);
-------
• Stop Solution;
• Diluent/zero solution, 25 mL.
The Microcystins ADDA microtiter plate ELISA Test Kit is shown in Figure 1 and measures 9 x
6x3 inches (22.9 x 15.2 x 7.6 centimeters) and weighs 4.8 pounds (2.2 kilograms). The cost is
$440 for a 96-test kit. Other materials and equipment not provided with the kits are pipettes,
pipette tips, a plate photometer capable of reading at 450 nm and distilled or deionized (DI)
water. These materials can be purchased separately from the vendor.
Figure 1. Microtiter Plate ELISA Test Kit
2.2 Microcystin DM ELISA Test Kit
The Microcystins (DM) Test Kit is an ELISA for the determination of microcystins and
nodularins in water samples. The assay utilizes monoclonal antibodies that have been raised
against the ADDA moiety of the molecule, allowing for the detection of numerous microcystins
and nodularin variants in drinking, surface, and groundwater at levels below WHO guidelines.
The test is a direct competitive ELISA method and is based on the recognition of microcystins,
nodularins and their variants by a monoclonal antibody. Microcystins and nodularins, when
present in a sample and a microcystins-HRP analog, compete for the binding sites of anti-
microcystins antibodies in solution. The microcystins antibodies are then bound by a second
antibody (goat anti-mouse) immobilized on the plate. After a washing step and addition of a
substrate/chromogen solution, a color signal is generated. The intensity of the color is inversely
proportional to the concentration of the microcystins/nodularins present in the sample. The color
reaction is stopped after a specified time and the color is evaluated using a plate photometer at a
wavelength of 450 nm.
The DM ELISA Test Kit differs from the ADDA ELISA Test Kit in the coating of the microtiter
plate. The Microcystins DM microtiter plate ELISA Test Kit is packaged the same as the ADDA
kit (see Figure 1) and contains:
-------
• Microtiter plate coated with a second antibody (goat anti- mouse);
• Standards (6) and positive control (1): 0, 0.15, 0.40, 1.0, 2.0, 5.0 ppb; the positive control
samples had a concentration of 0.75 ppb;
• Antibody solution (monoclonal anti-Microcystins);
• Microcystins-HRP Conjugate;
• Wash Solution 5X Concentrate;
• Color Solution (1MB);
• Stop Solution;
• Diluent/zero solution, 25 mL.
The Microcystins DM microtiter plate ELISA Test Kit has a cost of $400 for a 96-test kit. Like
the ADDA kit, other materials and equipment not provided are pipettes, pipette tips, a plate
photometer capable of reading at 450 nm and distilled or deionized water. These materials can
be purchased separately from the vendor.
2.3 Microcystin Strip Test
The Abraxis Microcystin Strip Test is a rapid immunochromatographic test, designed for the use
in the qualitative screening of microcystins and nodularins in recreational waters. A rapid cell
lysis step (QuikLyse™) performed prior to testing is required to measure total microcystins
(dissolved or free, plus cell bound). The Strip Test provides preliminary qualitative test results.
If necessary, positive samples can be confirmed by one of the ELISA test kits described above,
high pressure liquid chromatography (HPLC) or other conventional methods. The test is
designed for field use, requiring no instrumentation or other equipment, no power sources, and
no refrigerated storage.
The QuickLyse™ procedure includes shaking the sample in a lysing vial continuing dried lysis
reagent. After an eight minute period, a reagent paper is added to the lysis vial. After another
eight minute period, the sample is transferred to the Strip Test conical flip-top reagent vial. The
reagent vial is shaken vigorously, then the membrane strip is inserted into the vial to obtain a
reading.
The test is based on the recognition of microcystins, nodularins and their variants by specific
antibodies. The toxin conjugate competes for antibody binding sites with
microcystins/nodularins that may be present in the water sample. The test device consists of a
conical flip-top vial with specific antibodies for microcystins and nodularins labeled with a gold
colloid and a membrane strip to which a conjugate of the toxin is attached. A control line,
produced by a non-microcystin antibody/antigen reaction, is also present on the membrane strip.
The control line is not influenced by the presence or absence of microcystins in the water
sample, and therefore, it should be present in all reactions. In the absence of toxin in the water
sample, the colloidal gold labeled antibody complex moves with the water sample by capillary
action to contact the immobilized microcystins conjugate. An antibody-antigen reaction forms
one visible line in the test line region.
If microcystins are present in the water sample, they compete with the immobilized toxin
conjugate in the test area for the antibody binding sites on the colloidal gold labeled complex. If
a sufficient amount of toxin is present, it will fill all of the available binding sites, thus
-------
preventing attachment of the labeled antibody to the toxin conjugate, therefore preventing the
development of a second line in the test line region. If a second line is not visible in the test line
region, or if the test line is lighter than the negative control line, microcystins are present at the
levels of concern (> 10 ppb). Semi-quantitative results in the range of 0 to 10 ppb can be
obtained by comparing the test line intensity to those produced by solutions of known
microcystins concentrations (control solutions). During this verification, the results were
reported 0 to 10 ppb, or greater than 10 ppb.
The Microcystin Strip Test (Figure 2) contains:
• Microcystin membrane strips (test strips) in a desiccated container
• Sample collection vessels
• Lysis vials
• Graduated disposable pipettes (marked at 1 mL increments)
• Forceps
• Reagent papers
• Conical test vials
• Disposable transfer pipettes
• User's guide and interpretation guide
The Microcystin Strip Test is shown in Figure 2 and measures 9x6^3 inches (22.9 x 15.2 x
7.6 centimeters) and weighs 2.2 pounds (1 kilogram). The Strip Test (containing 20 test strips)
measures 16x6x3 inches (40.6 x 15.2 x 7.6 centimeters), packaged in two boxes and weighs
8.8 pounds (4 kilograms) total. The cost is $150 for a five-test kit and $480 for a 20-test kit.
Figure 2. Microcystin Strip Test
-------
Chapter 3
Test Design and Procedures
3.1 Test Overview
This verification test was conducted according to procedures specified in the Test/Quality
Assurance Plan for Verification of Microcystin Test Kits1 (TQAP) and adhered to the quality
system defined in the ETV AMS Center Quality Management Plan (QMP)2. Evaluating
microcystin test kits was identified by the AMS Center stakeholders as a priority area in 2005.
With stakeholder input to the design, reference method selection, and submission of recreational
waters to be evaluated, the test assessed the performance of microcystin test kits relative to key
verification parameters including accuracy, precision, and method detection limit (MDL). This
verification test took place from July 26 through August 12, 2010. The reference analysis was
performed the week of August 16, 2010.
3.2 Experimental Design
The objective of this verification test was to evaluate the performance of the microcystin test kits
against a known concentration of each microcystin variant in ASTM International Type IIDI
water, as well as microcystin variants in unknown proportions from recreational water (RW)
samples. Battelle conducted this verification test with recreational samples provided from the
Nebraska Department of Environmental Quality (NDEQ), with the University of Nebraska Water
Sciences Laboratory (WSL) providing reference analyses. The technologies were used to
analyze a variety of water samples spiked with the variants microcystin-LR, microcystin-LA, and
microcystin-RR. Because none of the three technologies can specify between the different
variants, the samples were spiked with individual variants. The quantitative results from the
microcystin test kits were compared to the results from the reference method by calculating
percent differences between the results. The reference method for microcystin was based on
direct injection liquid chromatography tandem mass spectrometry (LC-MS/MS)3 for the
determination of microcystins. To attain lower levels of detection, a sample preparation method
was developed by the WSL to extract the microcystins from the water samples and concentrate
the samples using solid phase extraction (SPE)4. The ELISA kits provided a quantitative or
semi-quantitative determination of microcystins. The ADDA (quantitative), DM (quantitative)
and Strip (semi-quantitative) test kits were evaluated for:
• Accuracy - comparison of test kit results (samples prepared in DI water) to results
from a reference method
• Precision - repeatability of test kit results from three sample replicates analyzed in DI
water, matrix interference, and recreational water samples
• Linearity - determination of whether or not the test kit response increases in direct
proportion to the known concentration of microcystin
• Method detection limit - the lowest quantity of toxin that can be distinguished from
the absence of that toxin (a blank value) at a 95% confidence level
• Inter-kit lot reproducibility - determination of whether or not the test kit response is
significantly different between two different lots of calibration standards within the
kits
-------
• Matrix Interference - evaluation of the effect of natural recreational water matrices
and chlorophyll-a on the results of the test kits
• Operational and Sustainability factors - general operation, data acquisition, setup,
consumables, etc.
Each microcystin test kit was operated according to the vendor's instructions by a vendor-trained
Battelle technician. The samples were tested according to the kit instructions, and in compliance
with the microcystin TQAP.
3.3 Test Procedures
The ability of each microcystin test kit to determine the concentration of microcystin was
challenged using quality control (QC) samples, performance test (PT) samples and RW samples.
These sample results were also compared to reference method results. Table 1 presents the test
samples analyzed during this verification test.
QC, PT, and RW samples were prepared by Battelle technical staff the day before testing began.
The test samples were prepared in glass volumetric flasks and stored in amber glass vials at 4 °C
± 3 °C until use. The reference samples that were prepared from the test solutions were stored in
amber glass bottles at < -10°C until analysis approximately 2 weeks later. Replicate samples for
the test kits were taken from the same sample bottle. The QC, PT, and RW samples were
prepared blindly for the operator by coding the sample labels to ensure the results were not
influenced by the operator's knowledge of the sample concentration and variant.
Because the reference method is specific to individual microcystins, PT samples for each of the
three different variants were combined into a volumetric flask and brought up to a known
volume with DI water before being sent to the reference laboratory. Then the calculated dilution
factor was applied to the reference method result to determine the PT sample concentration of
each variant. The RW samples were sent for reference analysis without dilution.
3.3.1 QC Samples
Reagent blank (RB) samples were prepared from DI water and exposed to identical handling and
analysis procedures as other prepared samples, including the addition of all reagents. These
samples were used to ensure that no sources of contamination were introduced in the sample
handling and analysis procedures. At least 10% of all the prepared samples were RBs.
As specified in the test kit procedure, at least one positive and one negative control were
analyzed with each ELISA plate. For the Strip Test, a control line is provided on the strip.
3.3.2 PT Samples
PT samples were used to verify the accuracy, precision, linearity, MDL, and inter-kit lot
reproducibility of the test kits. All PT samples were prepared at Battelle using DI water as the
water source. PT samples were individually spiked with microcystin-LR, microcystin-LA, and
microcystin-RR and analyzed in triplicate. The concentration levels were 0.10, 0.50, 1.0, 2.0,
and 4.0 ppb. These concentration levels were used for microcystin-LR and, because of the CR of
the LA and RR microcystin variants, a 7.0 ppb concentration level was also included to evaluate
the dynamic range of the test kits for these two variants. For the semi-quantitative Strip Test Kit,
7
-------
a 15 ppb PT sample was also used to test the semi-quantitative capability of indicating a
concentration higher than 10 ppb. EPA Guidelines were followed to estimate the MDL of the
quantitative test kits. In doing so, a solution with a concentration five times the vendor's
reported detection limit (DL) was used. A minimum of seven replicate analyses of this solution
were made individually for each variant to obtain precision data with which to determine the
MDL.
Table 1. Summary of Test Samples
Type of Sample
QC Samples - Kit
Positive Controls
QC Samples- Laboratory
Reagent Blank (RB)
PT Samples - DI Water
PT Samples - RW Matrix
Interference Samples:
RW sample and tenfold
dilution
PT Samples -
Chlorophyll-a Matrix
Interference Samples:
Chlorophyll-a sample and
tenfold dilution
PT Samples - Inter-kit lot
reproducibility
RW Samples- Through
freeze-thaw lysing
procedure
RW Samples- Through
vendor recommended
procedure
Microcystin
Variant
LR
none
LR
LA
RR
LR
LA
RR
LR
LA
RR
LR
LA
RR
Microcystin
Concentration
(ppb)
0.75
0
0.10,0.50,1.0,2.0,
4.0 ppb
0.50, 1.0,2.0,4.0,
7.0 ppb
0.50, 1.0,2.0,4.0,
7.0 ppb
5 times the vendor
stated MDL
5 times the vendor
stated MDL
5 times the vendor
stated MDL
4.0 ppb or 2.0 ppb*
4.0 ppb or 2.0 ppb*
4.0 ppb or 2.0 ppb*
4.0 ppb or 2.0 ppb*
4.0 ppb or 2.0 ppb*
4.0 ppb or 2.0 ppb*
Replicates
1
3
3
3
3
7
7
7
3
3
3
3
3
3
Total Number
of Samples per
Test Kit
1
10% of total test
samples, 2
15
15
15
7
7
7
6
6
6
6
6
6
A second set of vendor provided calibration standards from a different lot
analyzed following the vendor's procedure
Unknown
Unknown
3 samples >20 ppb, 3
samples >10 ppb, 3
samples ND
3 samples at
unknown
concentrations
3
3
27
9
*concentration that is within the calibration range of the test kit
-------
Additional performance testing was conducted to verify the impact of possible matrix
interferences. Two types of possible matrix interferences, RW water and chlorophyll-a, were
tested. Testing was performed using a RW sample with a low level of native microcystin
concentration (based on information from NDEQ). This RW sample was serially diluted by a
factor of 10 with DI water to provide a less concentrated level of the RW matrix. Then both the
original RW sample and diluted RW samples were fortified with 4.0 ppb (for the Strip Test) or
2.0 ppb (for the ADDA and DM test) of microcystin LR, LA, or RR. The spike level chosen was
dependent on the detection range of each kit. The test kit results in each of the matrices were
compared to determine the impact of the matrix concentration on the test kit results. In addition,
the results from the matrix samples were compared with the PT sample in DI water of the same
microcystin concentration.
To evaluate the effect of chlorophyll-a as a possible interference, a DI water sample that was
fortified with 10 milligram/Liter (mg/L) of chlorophyll-a (Sigma Aldrich, Cat # C5753-5MG
chlorophyll-a from spinach) was prepared by adding a known amount of chlorophyll-a into a
volumetric flask and diluting to volume. The chlorophyll was insoluble. Therefore, the resulting
solutions were clear solutions containing small black pieces of solid chlorophyll-a. These
solutions were then treated in an identical fashion as the above RW sample. The solution of
chlorophyll-a was serial diluted by a factor of 10 to provide solutions of 10 and 1 mg/L
chlorophyll-a. Then, each of these concentration levels was fortified with 4.0 or 2.0 ppb of
microcystin-LR, -LA, or -RR. The test kit results in each of the matrices were compared to
determine the impact of the chlorophyll-a on the test kit results.
Lastly, the calibration standards provided with the microcystin test kits from different lots could
cause variability in the results across test kits. Therefore, two separate lots of calibration
standards were analyzed using the kits and compared to determine the inter-kit lot
reproducibility.
3.3.3 RWSamples
RW samples were obtained from lakes in and around Lincoln, Nebraska to assess kit
performance in recreational waters. The procedure for collecting and preparing the samples for
verification testing and reference analysis is described in the NDEQ standard operating
procedure for microcystin analysis (SOP# SWS-2320.1A)6. In summary, staff from NDEQ
collected the water samples from lakes where there is a potential for human exposure to
microcystins. The RW samples were collected in brown plastic bottles with head space
remaining and returned to the laboratory where they were frozen and thawed three times to lyse
the cyanobacteria and free the microcystin into solution, making it available for analysis. Then
the samples were split for verification testing and reference analysis. Using analytical data
generated by NDEQ, samples used for ETV testing were selected from lakes that had both
detectable and not-detectable microcystin concentrations. Because not all possible variants are
monitored by the reference method, there could be a discrepancy between the test kit results and
the total microcystin determined by the reference method.
The Strip Test kit contains a specific lysing procedure to analyze for microcystin. For this test
kit, three of the RW samples were split before the freeze-thaw process to compare the results
using the two lysing procedures. The Strip Test Kit was used to analyze the three RW samples
with and without the freeze-thaw lysing.
-------
Chapter 4
Quality Assurance/Quality Control
QA/QC procedures were performed in accordance with the AMS Center QMP2 and the TQAP1
for this verification test. QA level III, Applied Research, was specified for this test by the EPA
Project Officer. These procedures and results are described in the following subchapters.
4.1 Reference Method Quality Control
To ensure that this verification test provided suitable data for a robust evaluation of performance,
a variety of data quality objectives (DQOs) were established for this test. The DQOs indicated
the minimum quality of data required to meet the objectives of the verification test. The DQOs
were quantitatively defined in terms of specific data quality indicators (DQIs) and their
acceptance criteria. The quality of the reference method measurements were assured by
adherence to these DQI criteria and the requirements of the reference methods, including the
calibration and QA/QC requirements of the method. Blank samples were required to generate
results below the detection limit and the Laboratory Fortified Matrix (LFM), duplicate, and
Performance Evaluation Audit (PEA) sample results were required to be within 30% of the
expected results. Continuing calibration verification (CCV) standards were required to be within
20% of the expected result. Battelle visited the reference laboratory prior to initiation of the
reference analysis and audited the data package provided by the reference laboratory following
analysis. More details about the audits are provided in Section 4.2. Table 2 presents these DQIs
and the reference method QC sample results. A total of 22 samples were analyzed by the
reference method; 17 were extracted prior to analysis and five were analyzed by direct injection.
One sample duplicate was processed with the 17 extracted samples to assess the DQI. No
sample duplicate was included for samples analyzed via direct injection.
The calibration of the LC-MS/MS method was verified by the analysis of a CCV at a minimum
of every 10 samples. All of the calibration standards were used as CCVs and were interspersed
throughout the run every five samples. The percent recoveries (%R) of CCVs were calculated
from the following equation.
C
%/? = —xlOO (1)
s
Cs is the measured concentration of the CCV and s is the spiked concentration. If the CCV
analysis differed by more than 20% from the true value of the standard (i.e., %R values outside
of the acceptance window of 80 to 120%), the instrument was recalibrated before continuing the
analysis. As shown in Table 3, all reference CCV analyses were within the required range.
Spiked samples were analyzed to assess the efficiency of the extraction method. There was a
LFM spike performed every 20 samples and was assessed by calculating the spike percent
recovery (%Rs) as below.
10
-------
%Rs = -
C-C
•xlOO
(2)
C^is the measured concentration of the spiked sample, C is the measured concentration of the
unspiked samples, and s is the spiked concentration. The spike %R was required to be within
30% of the spiked amount. The two LFM sample results were within this range for all three of
the variants.
The relative percent difference (RPD) of the duplicate sample analysis was calculated from the
following equation.
RPD = -
C -C,
CD)/2
-xlOO
(3)
C is the concentration of the sample analysis, and CD is the concentration of the duplicate sample
analysis. If the RPD was greater than 30%, then the extraction method and the analytical
methods were investigated. Reference method CCV RPD results are provided in Table 2.
Reference method precision of laboratory samples was not determined because the duplicate
extraction was performed on the reagent blank sample.
Table 2. DQIs and Summary of Reference Method QC Results
DQI
Performance
Evaluation Audit
(PEA)
Method
contamination
check
Method
Calibration
Check
Method precision
Method accuracy
Method of Assessment
(Frequency)
PEA Samples (Once before testing
begins)
Method Blank (Once every 20
samples)
Continuing Calibration Verification
(CCV) (Once every 5 samples)
Laboratory Duplicates (Once
every 20 samples)
Laboratory Fortified Matrix (LFM)
Spikes (Once every 20 samples)
Acceptance
Criteria for
Microcystins
70% - 130%
Recovery
< Lowest Calibration
Standard
80% - 120%
Recovery
< 30% Difference
70% - 130%
Recovery
Results
See Tables 4 and 5 in Section
4.2.1
ND for all three variants
See Table 3
See Table 3
LRM1
93% LR
79% LA
97% RR
LRM2
103% LR
105% LA
88% RR
11
-------
Table 3. Summary of Reference Method CC V Percent Recoveries
CCV Cone.
(ppb)
Variant % Recovery
LR
10 99.5
30 109
30
96.5
60 | 97.6
60 103
75 | 98.7
LA
98.2
104
97.1
94.2
109
91.8
RR
96.1
112
98.7
93.5
108
101
Variant RPD
LR
NA
12%
5%
NA
LA RR
NA NA
7% 13%
14% 14%
NA
NA
4.2 Audits
Three types of audits were performed during the verification test: a performance evaluation audit
(PEA), a technical systems audit (TSA) of the verification test procedures, and an audit of data
quality (ADQ). Audit procedures are described further below.
4.2.1 Performance Evaluation Audit
A PEA was conducted to assess the quality of the reference measurements made in this
verification test. National Institute of Standards and Technology traceable standards of
microcystin are not available; however, the Canadian National Research Council (NRC) offers
standards that have gone through the most validation of any commercially available standards
and were recognized by the vendors and stakeholders reviewing the TQAP as the most reliable
standards. The microcystin-LA variant was not available through the Canadian NRC and
therefore was obtained from Abraxis. The approach of using the microcystin-LA variant
standard from Abraxis was approved by all participating vendors prior to use. The standards
obtained from both sources were prepared at 50 ppb in DI water and sent blindly to the reference
laboratory for analysis. These PEA samples were analyzed directly (i.e., without additional
preparation) and were in the mid-level of the calibration range of the reference method. The
stock solutions used to prepare the calibration standards by the reference laboratory were
prepared by dissolving neat standards (not solutions) obtained from EMD Biosciences
(microcystin-LR), Sigma Aldrich (microcystin-LA), and ENZO Life Sciences (microcystin-RR).
The results from the analyses are presented in Table 4.
Table 4. PEA Results: Analytical Comparison of Microcystin Standards
Standard
Source
NRC
Canada
Abraxis
#of
Replicates
2
8
2
8
Analysis
Date
27-May
9-Jun
27-May
9-Jun
MC-LR
(%Recovery)
150% ±3%
13 5% ±7%
129% ±2%
121% ±6%
MC-LA
(%Recovery)
Not available
Not available
86% ±2%
86% ± 5%
MC-RR
(%Recovery)
192% ±1%
194% ± 12%
144% ±0%
153% ± 10%
Shading indicates results outside acceptable 30% tolerance based on TQAP
The recoveries of the NRC and Abraxis standards revealed that the reference laboratory method,
using the standards from alternate sources, were outside the acceptance range of ±30%. It was
then discussed with the stakeholders, and accepted by the vendors and the EPA Project Officer,
12
-------
that the reference laboratory use the two available NRC standards (LR and RR) as well as LA
from Abraxis for preparing the reference method calibration solutions. This is not a common
practice for calibration standards and test solutions to be generated from the same source, but
since the objective was to generate comparable vendor and reference data, it was deemed
necessary and appropriate for this verification test due to the difficulties in obtaining certified
microcystin standards.
To achieve the low detection limits required to analyze the test samples, an SPE extraction
method was also developed and used by the reference laboratory for samples expected to be
below 5.0 ppb. The MDL of this method was determined from extraction and analysis of eight
solutions of LR, LA, and RR at 0.375 ppb. The reference method MDLs for LR, LA, and RR
were determined to be 0.095 ppb, 0.141 ppb, and 0.127 ppb, respectively. Appendix A is the
memo from the reference laboratory presenting the MDL data.
A second PEA was performed to evaluate the extraction method efficiency and the analytical
method at a lower concentration relevant for this verification test. Battelle provided WSL with a
blind spiked DI sample at 0.25 ppb that was extracted in triplicate. The results from the second
PEA are presented in Table 5.
TableS. PEA Results: Evaluation of Extracted Low Level Water Sample
LR LA RR
0.25 ppb Spiked Cone, (ppb) %Recovery Cone, (ppb) %Recovery Cone, (ppb) %Recovery
Sample
Replicate 1 0.23 92% 0.21 84% 0.24 96%
Replicate 2 0.25 100% 0.23 92% 0.22 88%
ReplicateS 0.23 92% 0.22 88% 0.26 104%
Average 0.24 95% 0.22 88% 0.24 96%
Standard Deviation 0.01 5% 0.01 4% 0.02 8%
4.2.2 Technical Systems A udit
Battelle's Quality Assurance Officer (QAO) conducted a TSA to ensure that the verification test
was being conducted in accordance with the TQAP1 and the AMS Center QMP2. As part of the
TSA, test procedures were compared to those specified in the TQAP1, and data acquisition and
handling procedures as well as the reference method procedures were reviewed. Two
observations on storage of test records and sample handling and custody were documented and
submitted to the Battelle Verification Test Coordinator for response. The observations from the
TSA were addressed and documented as necessary. The conclusion of the TSA was that
verification testing was performed according to the TQAP. TSA records are permanently stored
with the QAO.
4.2.3 Data Quality Audit
Two ADQs were performed for this verification test. The first was for the data collected on the
first day of testing and the second was on the complete data package generated during verification
test preparation and execution. During the audits, test kit data were reviewed and verified for
completeness, accuracy and traceability.
13
-------
Because the EPA Project Officer designated this as an EPA Category III verification test, at least
10% of the data acquired were audited. The QAO traced the data from the initial acquisition,
through reduction and statistical analysis, to final reporting to ensure the integrity of the reported
results. All calculations performed on the data undergoing the audit were checked.
Observations and findings (mostly related to test record documentation) were reported and
submitted to the Battelle Verification Test Coordinator after the TSA and all observations were
addressed prior to the submission of this final report.
14
-------
Chapter 5
Statistical Methods
The statistical methods used to evaluate the quantitative performance factors listed in Section 3.2
are presented in this chapter. Qualitative observations were also used to evaluate verification test
data.
The microcystin test kits being verified report total microcystin and are also calibrated against
microcystin-LR. Because of this, the kit data were converted from microcystin-LR equivalents
to compare the test kit results to the reference method results for all PT samples. Using cross
reactivity data provided by each vendor (specific to each test kit), the microcystin-LR
equivalents were converted to microcystin concentration by variant as follows:
r = LRequiv (A\
^byvariant
where CLR eqmv is the test kit result in equivalents of microcystin-LR and CR is the mass-based
cross reactivity of the variant.7
For the RW samples, each variant identified through analysis by the reference method was
converted to LR-equivalents, and added together to calculate the total microcystins. The total
microcystin-LR equivalents from the RW reference analyses were compared to the total
microcystin results from the test kits. Because not all possible variants are monitored by the
reference method, there could be a discrepancy between the test kit results and the total
microcystin determined by the reference method.
5.1 Accuracy
Accuracy of the test kits verified was assessed relative to the results obtained from the
reference analyses. The results for each set of analyses were expressed in terms of a percent
difference (%D) as calculated from the following equation:
C —C
%D = — - -xlOO (5)
where CT is the microcystin-LR equivalent results from the test kits being verified and CR is the
concentration as determined by the reference method.
5.2 Linearity
Linearity was determined by linear regression with the toxin concentration measured by the
reference method as the independent variable, and the test kit result being verified as the
dependent variable. Linearity was expressed in terms of the slope, intercept, and the coefficient
of determination (r2). In addition, plots of the observed and predicted concentration values were
constructed to depict the linearity for each variant of microcystin being tested.
15
-------
5.3 Precision
The standard deviation (S) of the results for the replicate samples were calculated and used as a
measure of test kit precision at each concentration. S was calculated from the following
equation:
s =
(6)
where n is the number of replicate samples, Ck is the concentration measure for the kth sample,
and C is the average concentration of the replicate samples. The kit precision at each
concentration was reported in terms of the relative standard deviation (RSD~) presented below.
RSD= =
C
x 100 (7)
5.4 Method Detection Limit
MDL was determined by seven replicate analyses of a fortified sample with the toxin
concentration of five times the vendor's estimated detection limit. The MDL was calculated
from the following equation:
MDL = txS (8)
where t is the Student's value for a 95% confidence level, and S is the standard deviation of the
replicate samples.
5.5 Inter-Kit Lot Reproducibility
Inter-kit lot reproducibility was assessed by calculating the RPD (Equation 3) between OD
results are given to compare between the lots of calibration standards.
5.6 Matrix Effects
Matrix interference effects also were assessed by using a t-test to compare the microcystin test
kit results generated from samples made by spiking undiluted and diluted interference matrices
with the PT sample results at the same spiked concentration (either 2.0 or 4.0 ppb spike
concentration). Each paired t-test was performed using the replicate data from each type of
sample. The null hypothesis is that the difference between the two sets of data is zero.
Therefore, the resulting probability (p)-value gives the likelihood that the null hypothesis would
be true. Therefore, at the 95% confidence interval, p-values less than 0.05 will indicate there is a
small likelihood of the null hypothesis being true and therefore a significant difference between
the two sets of data. Since the number of replicates were predetermined by the test kit
instructions and TQAP, power and sample size calculations were not conducted for this
assessment. It is important to note that strong inference based on the results cannot be
established due to the low power of this study.
16
-------
Chapter 6
Test Results for the Abraxis ADDA ELISA Test Kit
The following sections provide the results of the quantitative and qualitative evaluations of this
verification test for the Abraxis ADDA ELISA Test Kit.
6.1 ADDA Test Kit Summary
As discussed in Chapter 2, the ADDA Test Kit quantifies total microcystins in water based on an
LR calibration. Other variants of microcystins bind differently to the immunosorbent (i.e., cross
reactivity). Therefore, the relative ability for other microcystins to bind has been experimentally
determined by the vendor. For the ADDA Test Kit, the CR of microcystin LA is 125% and the
CR of microcystin RR is 91%. In this report, the test kit data have been reported in both test kit
results as LR equivalents and in CR corrected results by variant, based on Equation 4.
The ADDA Test Kit requires that each standard and sample be analyzed in duplicate and then the
raw data output from the plate reader software reports a mean concentration of the duplicate
analyses. Therefore, a sample indicated in Table 1 to have three replicates corresponded to six
wells being filled as part of the ADDA Test Kit. Each ADDA Test Kit plate contains six
calibration solutions. Following the analysis method, the plate reader measures the absorbance
of the wells containing the calibration solutions at 450 nm wavelengths and the calibration curve
was generated based on the OD of each well. These results were plotted against concentrations
using a four-parameter curve to quantify the samples. According to Abraxis, if a sample was out
of range it was determined to be either above or below the calibration range and either diluted
into the linear range or reported as less than the limit of quantification (
-------
Table 6. KB Sample Results for the Abraxis ADDA Test Kit
Reagent Blank
RB 1
RB2
RB3
RB4
RB5
RB6
Plate
1
1
1
o
J
3
3
Mean Concentration (ppb)
ND
ND
<0.15LOQ
ND
ND
ND
The positive controls for the ADDA Test Kit are presented in Table 7. The vendor stated
acceptable range for recovery of the positive control is between 75% and 125%. In addition, the
coefficient of variation (CV) of the duplicate analyses is reported as a gauge for accurate
quantification of microcystins. The variation between the two data points is considered
acceptable when the %CV is less than 25%. At least one positive control was analyzed at the
end of each plate, and in some instances when space allowed, additional positive controls were
analyzed. All ADDA Test Kit plates used for testing produced a positive control result within
the acceptable range. During verification testing of the ADDA Test Kit, all plates were within
the CV and %R acceptance criteria.
Table 7. Positive Control Sample Results for the Abraxis ADDA Test Kit
Positive Control
1
2
2a
3
4
5
5a
5b
6
Plate
1
2
2
3
4
5
5
5
6
Mean Concentration (ppb)
0.653
0.826
0.808
0.566
0.625
0.717
0.656
0.769
0.903
CV (%)
8.2
4.2
6.2
18
25
2.0
1.7
1.9
16
Percent Recovery (%)
87%
110%
108%
75%
83%
96%
87%
103%
120%
6.3 PT Samples
Tables 8, 9, and 10 present the results for the PT samples for the three variants of microcystin
used during this verification test. In addition, the tables present the sample concentration
corrected for the microcystin cross reactivity, the reference method results and the accuracy
results by variant for the PT samples prepared in DI water.
All samples have at least three results, but some samples include four or more replicate results
because in instances when the %CV were less than 25%, the individual samples were reanalyzed
in duplicate (per the vendor instructions). If the resulting %CV was acceptable for both repeat
samples, they were both included in the result tables, thus resulting in additional data points. In
addition, the 0.50 ppb solutions included all seven replicates from the MDL determination data
in addition to the triplicate analyses of the 0.50 ppb PT samples.
18
-------
6.3.1 Accuracy
Tables 8, 9, and 10 also present the accuracy results for the ADDA Test Kit, expressed as %D
when calculated with the theoretical spike concentration and the reference method concentration.
As shown in Equation 5 (Section 5.1), the reference method value was used for calculation of
accuracy.
Table 8. Abraxis ADDA Test Kit Sam
Sample
Description
0.10LR
Avg ± SD
0.50 LR
Avg ± SD
1.0 LR
Avg ± SD
2.0 LR
Avg ± SD
4.0 LR
Avg ± SD
Kit
Results:
LR
Equivalents
(ppb)
0.096
-------
Table 9. Abraxis ADDA Test Kit Sample Results and Reference Method Results for LA
Sample
Description
0.50 LA
Avg ± SD
LOLA
Avg ± SD
2.0 LA
Avg ± SD
4.0 LA
Avg ± SD
Kit Results:
LR
Equivalents
(ppb)
0.667
0.914
0.718
0.969
1.05
0.927
0.842
0.905
0.740
0.750
0.634
0.828 ±0.134
2.04
1.78
1.55
1.79 ±0.246
3.60
4.32
4.53
4.15 ±0.488
7.70
7.99
7.84 ± 0.206
CR Corrected
Cone. By
Variant (ppb)
0.534
0.731
0.574
0.775
0.837
0.742
0.674
0.724
0.592
0.600
0.507
0.663 ±0.107
1.63
1.43
1.24
1.43 ±0.197
2.88
3.46
3.62
3.32 ±0.390
6.16
6.39
6.27 ±0.165
Accuracy by
Variant for
Theoretical
Concentration (%
Difference)
7%
46%
15%
55%
67%
48%
35%
45%
18%
20%
1%
33% ±21%
63%
42%
24%
43% ±20%
44%
73%
81%
66% ± 20%
54%
60%
57% ± 4%
Accuracy by
Variant for
Reference
Concentration (%
Difference)
33%
83%
44%
94%
109%
85%
68%
81%
48%
50%
27%
66% ±27%
133%
104%
77%
104% ±28%
69%
103%
113%
95% ± 23%
105%
113%
109% ±6%
Reference
Concentration
(ppb)
0.40
0.70
1.7
3.0
20
-------
Table 10. Abraxis ADDA Test Kit Sample Results and Reference Method Results for RR
Sample
Description
0 50 RR
Avg ± SD
1.0 RR
Avg ± SD
2.0 RR
Avg ± SD
4 0 RR
Avg ± SD
Kit Results:
LR
Equivalents
(ppb)
0.490
0.544
0.552
0.568
0.567
0.518
0.514
0.549
0.563
0.552
0.542 ± 0.026
1.10
0.973
0.972
1.01 ±0.071
1.80
2.46
2.20
2.15 ±0.335
4.36
4.07
3.24
3.55
3.81 ±0.504
CR
Corrected
Cone. By
Variant
(ppb)
0.538
0.598
0.607
0.624
0.623
0.569
0.565
0.603
0.619
0.607
0.595 ±0.029
1.20
1.07
1.07
1.11 ±0.078
1.97
2.70
2.42
2.37 ±0.368
4.79
4.48
3.56
3.90
4.18 ±0.554
Accuracy by
Variant for
Theoretical
Concentration
(% Difference)
8%
20%
21%
25%
25%
14%
13%
21%
24%
21%
19% ±6%
20%
7%
7%
11% ±8%
-1%
35%
21%
18% ±18%
20%
12%
-11%
-2%
5% ± 14%
Accuracy by
Variant for
Reference
Concentration
(% Difference)
42%
57%
60%
64%
64%
50%
49%
59%
63%
60%
57% ± 8%
123%
98%
98%
106% ±14%
23%
69%
51%
48% ±23%
50%
40%
11%
22%
31% ±17%
Reference
Concentration
(ppb)
0.38
0.54
1.6
3.2
For the LR spiked samples, the reference method results ranged from 0% to 17% less than the
target spike concentration. For LR, the percent difference ranged from -45% to 58%, with
overall average percent difference values ranging from -12% to 33%. One 0.10 ppb sample was
determined as being less than the LOQ so no %D was calculated, but the other three %Ds were
less than 14% from the reference method result. Only two replicate results are given for the 4.0
ppb samples as the samples were repeated four times after the result was reported as "out of
range". The sample was not diluted because lower concentrations had already been analyzed.
For the 0.50 ppb samples, the %D ranged from -9% to 40%, but the absolute difference from the
reference concentration was no more than 0.170 ppb. For the 1.0 ppb samples, the %D ranged
from 19% to 58%, corresponding to a maximum absolute difference from the reference
concentration of 0.481 ppb. Similarly, for the 2.0 ppb samples, the %D ranged from -45% to
39% and the maximum absolute difference from the reference concentration was 0.860 ppb. For
the 4.0 ppb samples, the two reported results had a %D of less than 10%. For LR, the %D when
21
-------
compared to the theoretical spike concentration range from -48% to 32% with the overall
average %D values ranging from -17% to 10%.
For the LA spiked samples, the reference method results ranged from 15% to 33% less than the
target spike concentration. For LA, the percent difference ranged from 27% to 276%. For the
0.50 ppb samples, the %D ranged from 27% to 109%, corresponding to an absolute maximum
difference from the reference concentration of 0.437 ppb. For the 1.0 ppb samples, the %D
ranged from 77% to 133%, corresponding to a maximum absolute difference from the reference
concentration of 0.930 ppb. For the 2.0 ppb samples, the %D ranged from 69% to 113% and the
maximum absolute difference from the reference concentration was 1.92 ppb. For the 4.0 ppb
samples, the two reported results had %Ds of 262% and 276%, corresponding to a maximum
absolute difference from the reference concentration of 3.39 ppb. For LA, the %D when
compared to the theoretical spike concentration range from 1% to 81% with the overall average
%D values ranging from 33% to 66%.
For the RR spiked samples, the reference method results ranged from 20% to 46% less than the
target spike concentration. For RR, the percent difference ranged from 11% to 123%. For the
0.50 ppb samples, the %D ranged from 42% to 64%, corresponding to an absolute maximum
difference from the reference concentration of 0.244 ppb. For the 1.0 ppb samples, the %D
ranged from 98% to 123% and the maximum absolute difference from the reference
concentration was 0.663 ppb. For the 2.0 ppb samples, the %D ranged from 23% to 69%
corresponding to a maximum absolute difference from the reference concentration of 1.10 ppb.
For the 4.0 ppb samples, the %D ranged from 11% to 50%, corresponding to a maximum
absolute difference from the reference concentration of 1.59 ppb. For RR, the %D when
compared to the theoretical spike concentration range from -11% to 35% with the overall
average %D values ranging from 5% to 19%.
6.3.2 Precision
Precision results for the ADDA Test Kit are presented in Table 11. The RSD was determined as
a percentage according to Equation 7 (Section 5.3) for all DI water, matrix interference and
recreational water samples. The RSDs ranged from 5% to 45% for the LR variant; however,
seven of the nine sample sets had RSDs lower than 16%. For LA, the RSDs ranged from 3% to
25% and from 4% to 16% for the RR variant. The precision for the RW samples ranged from
3% to 47%, but all but two RW samples sets had RSDs less than 12%. The overall average of all
RSDs was 13%, with a minimum of 3% and a maximum of 47%.
22
-------
Table 11. Abraxis ADDA Test Kit Precision Results
Variant
LR
LA
RR
Unknown
Sample Concentration in DI
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb LR in 1 mg/L Chlorophyll-a DI
2.0 ppb LR in 10 mg/L Chlorophyll-a DI
2.0 ppb LR in lOx dilution of RW Matrix
2.0 ppb LR in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb LA in 1 mg/L Chlorophyll-a DI
2.0 ppb LA in 10 mg/L Chlorophyll-a DI
2.0 ppb LA in lOx dilution of RW Matrix
2.0 ppb LA in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb RR in 1 mg/L Chlorophyll-a DI
2.0 ppb RR in 10 mg/L Chlorophyll-a DI
2.0 ppb RR in lOx dilution of RW Matrix
2.0 ppb RR in RW Matrix
RW1
RW2
RW 3 (lOx dilution)
RW 3 (20x dilution)
RW 4 (4x dilution)
RW 5 (4x dilution)
RW 6 (2x dilution)
RW7
RW8
RW9
Precision (%RSD)
10%
15%
16%
45%
7%
11%
5%
7%
31%
16%
14%
12%
3%
18%
6%
19%
25%
5%
7%
16%
13%
14%
15%
4%
8%
10%
5%
47%
8%
12%
12%
22%
NA
5%
3%
NA - Result was less than the LOQ so no calculation of RSD
23
-------
6.3.3 Linearity
The linearity of the ADDA Test Kit measurements was assessed by performing a linear
regression of the ADDA test kit results against the reference method results for the five PT
samples ranging from 0.10 to 4.0 ppb of microcystin LR in DI water and four PT samples
ranging from 0.50 to 4.0 ppb for microcystin LA and RR in DI water. Figures 3, 4, and 5 present
the results of the linear regressions for LR, LA, and RR, respectively. The slope, intercept, and
coefficient of determination (r2) for each regression equation are shown on the charts. The linear
regressions compared to the reference method results had coefficients of determination of 0.906
for LR, 0.990 for LA, and 0.961 for RR.
9 -
8 -
I7
y 6
y = 0.933x +0.087
R2 = 0.906
01234
Reference Cone, (ppb)
Figure 3. Linearity for the Abraxis ADDA Test Kit for LR
7
5 -
1 -
y = 2.66x-0.220
R2 = 0.990
01234
Reference Cone, (ppb)
Figure 4. Linearity for the Abraxis ADDA Test Kit for LA
24
-------
9 -i
8 -
u b
<5
I 4
.!2 3
x
I 2
1
y=1.18x +0.168
R2 = 0.961
01234
Reference Cone, (ppb)
Figure 5. Linearity for the Abraxis ADDA Test Kit for RR
6.3.4 Method Detection Limit
The MDL was assessed by analyzing at least seven replicates of a sample spiked at
approximately five times the vendor-stated detection limit for the microcystin test kit (which was
0.10 ppb). Table 12 lists the replicate results, the %CV of the duplicate ADDA Test Kit analysis
for each individual replicate, the standard deviations for the replicate results, and shows the
calculated MDLs for the three variants. The calculated MDL values were 0.14, 0.22, and 0.05
ppb for LR, LA, and RR, respectively.
Table 12. Detection Limit Results for the Abraxis ADDA Test Kit
Variant
Sample
Concentration
(ppb)
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
Standard Deviation
t value
n
MDL
LR
Mean
Cone.
(ppb)
0.59
0.58
0.59
0.55
0.48
0.54
0.55
0.44
0.38
0.43
NA
0.08
1.8
10
0.14
%CV
8.9
2.5
2.3
11
15
3.6
9.8
18
7.0
1.6
NA
LA
Mean Cone.
(ppb LR
Equivalents)
0.91
0.72
0.97
1.0
0.67
0.93
0.84
0.91
0.74
0.75
0.63
0.13
1.8
11
0.22
%CV
5.5
1.2
3.6
4.5
24
5.6
3.6
2.2
0.60
12
2.1
RR
Mean Cone.
(ppb LR
Equivalents)
0.49
0.54
0.55
0.57
0.57
0.52
0.51
0.55
0.56
0.55
NA
0.03
1.8
10
0.05
%CV
1.3
0.50
3.3
17
8.9
0.80
12
6.3
8.9
4.7
NA
25
-------
6.3.5 Inter-Kit Lot Reproducibility
Two sets of kit calibration standards were analyzed on the sample plate to compare whether or
not the calibration standards from different lots were similar. The data are presented in Table 13.
The OD values were compared by calculation of the RPD between each pair of OD
measurements. All RPDs were less than 13% and all but four were less than 6%.
Table 13. Inter-kit lot Comparison of Kit Calibration Standards for the ADDA Test Kit
Standard
Std 0 ppb
Std 0.15 ppb
Std 0.40 ppb
Std 1.0 ppb
Std 2.0 ppb
Std 5.0 ppb
OD Values
Set A
1.48
1.47
1.14
1.16
0.836
0.841
0.561
0.560
0.434
0.458
0.335
0.340
SetB
1.34
1.29
1.10
1.05
0.786
0.790
0.544
0.551
0.413
0.456
0.337
0.299
RPD
11%
13%
3%
11%
6%
6%
3%
2%
5%
0%
1%
13%
6.3.6 Matrix Effect
Matrix interference effects were assessed by using a t-test to compare the ADDA Test Kit results
generated from samples made by spiking undiluted and diluted interference matrices with the PT
sample results at the same concentration. The two possible interfering matrices included a RW
sample both undiluted and after undergoing a tenfold dilution and chlorophyll-a at 10 mg/L and
1 mg/L. Tables 14 and 15 provide the ADDA Test Kit sample results for the RW matrix
interference samples and chlorophyll-a interference samples, respectively, including the average
and SD for each sample. Because this comparison is made to evaluate only the impact of the
matrix on the sample result, LR equivalents are used.
Each paired t-test was performed using the replicate data from each type of sample. The null
hypothesis is that the difference between the two sets of data is zero. The resulting probability
(p)-value gives the likelihood that the null hypothesis would be true. Therefore, at the 95%
confidence interval, p-values less than 0.05 will indicate there is a small likelihood of the null
hypothesis being true and therefore a significant difference between the two sets of data (at low
power).
Table 16 summarizes the results of a paired t-test for both sets of interference data by showing
the p-values associated with each of the applicable comparisons across both types of possible
interfering matrices. Across both the chlorophyll-a and RW results, three out of 16 comparisons
resulted in statistically significant differences (two comparisons could not be performed because
there were only two replicate results for the LR spiked undiluted RW samples). The 2.0 ppb LA
spike into DI water was significantly different from the 2.0 ppb LA spike into both 1 mg/L
(p=0.005) and 10 mg/L (p=0.006) chlorophyll-a. Table 15 shows that the 2.0 ppb spike into DI
26
-------
water generated an average result of 4.15 ppb compared with an average result of 0.552 and
0.580 ppb for the spike into 1.0 mg/L and 10 mg/L chlorophyll-a, respectively. The other
statistically significant difference was between the RR spikes into undiluted and diluted RW
(p=0.01). These two samples were not significantly different from the PT sample spike in DI
water, but they were different from each other with average concentrations of 1.35 ppb for the
diluted RW and 2.56 ppb for the undiluted RW. There were three other pairs of data that were
close to being statistically significant differences with p-values of less than 0.1. The LR spike
into DI was nearly different from the chlorophyll-a spikes and the RR spike into DI was nearly
different from the RW spikes.
Given that the molecular basis on which the test kits operate is well-characterized and
understood from the literature8, Table 15 provided unexpected results. Two variants (LR and
LA) demonstrated an interference effect but the third variant (RR) did not. This could have been
caused by a number of factors, such as chlorophyll-a source and stability. However, due to the
limited number of replicates that were analyzed, additional testing would be required to provide
a better understanding as to whether there is matrix interference due to chlorophyll-a, or another
variable not investigated in this verification testing.
27
-------
Table 14. RW Matrix Interference Sample Results for the Abraxis ADDA Test Kit
Variant
Unknown
LR
LA
RR
Sample
Description
UnspikedRW
Matrix (RW 9)
2.0 ppb LR in
DI
2.0 ppb LR in
tenfold dilution
of RW Matrix
2.0 ppb LR in
RW Matrix
2.0 ppb LA in
DI
2.0 ppb LA in
tenfold dilution
of RW Matrix
2.0 ppb LA in
RW Matrix
2.0 ppb RR in
DI
2.0 ppb RR in
tenfold dilution
of RW Matrix
2.0 ppb RR in
RW Matrix
Mean Kit Results:
LR Equivalents
(ppb)
0.639
0.600
0.621
1.87
2.63
1.10
1.05
1.70
1.90
1.66
4.05
2.60
3.60
4.32
4.53
2.66
3.86
3.03
3.01
2.17
1.89
1.80
2.46
2.20
1.37
1.39
1.29
2.78
2.38
2.50
Average
Result
(ppb)
0.620
1.66
1.76
3.32
4.15
3.18
2.36
2.15
1.35
2.56
SD
0.020
0.749
0.130
1.03
0.488
0.617
0.583
0.335
0.054
0.208
CR
Corrected
Cone. By
Variant (ppb)
1.87
2.63
.10
.05
.70
.90
.66
4.05
2.60
288
3.46
3.62
2.12
3.09
2.42
2.41
1.74
1.52
1.97
2.70
2.42
1.50
1.52
1.41
3.06
2.61
2.75
28
-------
Table 15. Chlorophyll-a Interferent Sample Results for the Abraxis ADDA Test Kit
Variant
LR
LA
RR
Sample
Description
2.0ppbLRinDI
2.0 ppb LR in 1
mg/L
Chlorophyll-a DI
2.0 ppb LR in 10
mg/L
Chlorophyll-a DI
2.0 ppb LA in DI
2.0 ppb LA in 1
mg/L
Chlorophyll-a DI
2.0 ppb LA in 10
mg/L
Chlorophyll-a DI
2.0ppbRRinDI
2.0 ppb RR in 1
mg/L
Chlorophyll-a DI
2.0 ppb RR in 10
mg/L
Chlorophyll-a DI
Mean Kit Results:
LR Equivalents
(ppb)
1.87
2.63
1.10
1.05
0.391
0.432
0.486
0.488
0.532
0.493
3.60
4.32
4.53
0.533
0.669
0.581
0.425
0.576
0.617
0.548
1.80
2.46
2.20
1.98
1.44
1.61
1.57
2.24
1.77
1.57
2.20
2.01
Average
Result
(ppb)
1.66
0.436
0.504
4.15
0.552
0.580
2.15
1.64
1.96
SD
0.749
0.048
0.024
0.488
0.102
0.035
0.335
0.230
0.285
CR
Corrected
Cone. By
Variant (ppb)
1.87
2.63
1.10
1.05
0.391
0.432
0.486
0.488
0.532
0.493
2.88
3.46
3.62
0.426
0.535
0.465
0.340
0.461
0.494
0.438
1.97
2.70
2.42
2.17
.58
.77
.73
2.46
.94
.73
2.42
2.21
29
-------
Table 16. Statistical Comparisons between Interference Samples for the Abraxis ADDA
Test Kit
Description of Comparison
2.0 ppb in DI compared with 2.0 ppb in tenfold
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in undiluted
RW
2.0 ppb in undiluted RW compared with tenfold
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in 1 mg/L
Chlorophyll-a DI
2.0 ppb in DI compared with 2.0 ppb in 10 mg/L
Chlorophyll-a DI
2.0 ppb in 1 mg/L Chlorophyll-a DI compared
with 2.0 ppb in 10 mg/L Chlorophyll-a DI
p-value (D-different, ND-not different)
LR
0.788 (ND)
NA
NA
0.089 (ND)
0.087 (ND)
0.156(ND)
LA
0.084 (ND)
0.102(ND)
0.311(ND)
0.005 (D)
0.006 (D)
0.677 (ND)
RR
0.054 (ND)
0.327 (ND)
0.010 (D)
0.307 (ND)
0.510(ND)
0.242 (ND)
Shading indicates a statistically significant difference
NA- The p-value could not be determined with only 2 replicate results of the undiluted RW LR spiked samples
6.4 RW Sample Results
Table 17 presents the RW results for the ADDA Test Kit and the reference analysis. The
concentrations were determined by the reference method for only three of the approximately 80
variants that are naturally occurring in recreational waters. The total microcystins measured by
the ADDA Test Kit may have other variants present that would not have been detected by the
reference method. Therefore, no quantitative comparison was made between the ADDA Test Kit
and the reference method results. The reference data were converted into LR-equivalents
according to the ADDA Test Kit cross reactivity for the variants. In general, the samples that
were determined to have higher total concentrations by the ADDA Test Kit had higher total
concentrations as determined by the reference method. All of the ADDA Test Kit total
microcystin results were greater than the reference method results that only quantified three
variants. However, the results of the ADDA Test Kit were usually within a factor of two of the
reference method. The LR, LA, and RR variants likely make up a significant proportion of the
microcystins that are measurable by the ADDA Test Kit.
30
-------
Table 17. Recreational Water Sample Results for the Abraxis ADDA Test Kit
Sample
Description
RW 1 (20x
dilution)
RW 2 (lOx
dilution)
RW 3 (lOx
dilution)
RW 3 (20x
dilution)
RW 4 (4x
dilution)
RW 5 (4x
dilution)
RW 6 (2x
dilution)
RW7
RW8
RW 9 (RW
Matrix)
Kit Results:
LR
Equivalents
(ppb)
2.132
1.843
2.235
2.298
2.087
1.742
3.497
1.019
1.202
1.101
2.258
2.667
1.574
.587
.280
.388
.405
.710
.207
0.931
-------
6.5 Operational Factors
During testing activities, the technical operators were instructed to fill out an Ease of Use
Questionnaire. This section summarizes these observations as well as other operational
considerations about the technology.
6.5.1 Ease of Use
The test kit operator reported that the ADDA Test Kit was easy to use. The brochure and flow
charts with illustrations were clear and easy to follow. Solution and sample preparation were
minimal, involving dilution of the samples that were initially above the quantification range.
The procedure includes three incubation periods that total 2.5 hours. Previous knowledge or
training on the use of micro-pipettes and or multi-channel pipettes with 96-well plates is
recommended for consistent readings. The Battelle operator that was trained by Abraxis had
more than 10 years of analytical laboratory experience, but was not experienced with ELISA
analysis. A spectrophotometer plate reader is necessary for obtaining the spectrophotometric
readings that are then analyzed using any commercial ELISA evaluation program (four-
parameters are recommended by the vendor). Once the analysis was complete, the remaining
solutions were disposed in the trash in accordance with local regulations.
6.5.2 Cost and Consumables
The listed price for the ADDA Test Kit at the time of the verification test was $440. The kit has
a 12-month shelf life as received and should be stored at 4 to 8 °C. Of the 96-wells on one plate,
16 wells are needed for calibration and control samples. The remaining 80 wells are for sample
analyses that are performed in duplicate. Other consumables required for the test, but not
included in the kit are pipettes, pipette tips, and distilled or DI water. These can be obtained
from the vendor.
32
-------
Chapter 7
Test Results for the Abraxis DM ELISA Test Kit
The following sections provide the results of the quantitative and qualitative evaluations of this
verification test for the Abraxis DM ELISA Test Kit.
7.1 DM Test Kit Summary
As discussed in Chapter 2, the DM Test Kit quantifies total microcystins in water based on an
LR calibration. Other variants of microcystins bind differently to the immunosorbent (i.e., cross
reactivity). Therefore, the relative ability for other microcystins to bind has been experimentally
determined by the vendor. For the DM Test Kit, the CR of microcystin LA is 48% and the CR of
microcystin RR is 53%. In this report, the test kit data have been reported in both test kit results
as LR equivalents and in CR corrected results by variant, based on Equation 4.
The DM Test Kit requires that each standard and sample be analyzed in duplicate, and then the
raw data output from the plate reader software reports a mean concentration of the duplicate
analyses. Therefore, a sample indicated in Table 1 to have three replicates corresponded to six
wells being filled as part of the DM Test Kit. Each DM Test Kit plate contains six calibration
solutions. Following the analysis method, the plate reader measured the absorbance of the wells
containing the calibration solutions at 450 nm wavelengths and the calibration curve was
generated based on the OD of each well. These results were plotted against concentrations using
a four-parameter curve to quantify the rest of the samples. According to the Abraxis kit
instructions, if a sample was out of range and it was determined to be either above or below the
calibration range, then it would either be diluted or reported as
-------
Table 18. KB Sample Results for the Abraxis DM Test Kit
Reagent Blank
RB 1
RB2
RB3
RB4
RB5
RB6
Plate
1
1
1
o
J
3
3
Mean Concentration (ppb)
<0.15LOQ
<0.15LOQ
<0.15LOQ
<0.15LOQ
<0.15LOQ
<0.15LOQ
The positive controls for the DM Test Kit are presented in Table 19. The vendor kit instructions
stated acceptable range for recovery of the positive control was between 75% and 125%. In
addition, the CV of the duplicate analyses was reported as a gauge for accurate quantification of
microcystins. The variation between the two data points was considered acceptable when the
%CV was less than 25%. A positive control was analyzed at the end of each plate, and in some
instances, there were additional positive controls analyzed. This was done to fill in the final
column of wells on the 96-well plates. All DM Test Kit plates used for testing produced a
positive control result within the acceptable range. During verification testing of the DM Test
Kit, all plates were within the %CV and %R acceptance criteria.
Table 19. Positive Control Sample Results for the Abraxis DM Test Kit
Positive Control
1
2a
2b
3
4
5a
5b
Plate
1
2
2
3
4
5
5
Mean Concentration (ppb)
0.836
0.746
0.740
0.603
0.756
0.616
0.505
CV (%)
9.0
9.1
4.7
0.10
2.7
6.4
1.3
Percent Recovery (%)
111%
99%
99%
80%
101%
82%
67%
7.3 PT Samples
Tables 20, 21, and 22 present the results for the PT samples for the three variants of microcystin
used during this verification test. In addition, the tables present the sample concentration
corrected for the microcystin cross reactivity, the reference method results and the accuracy
results by variant for the PT samples prepared in DI water. All samples have at least three
results, but some samples include four or more replicate results because in instances when the
%CV were less than 25%, the individual samples were reanalyzed in duplicate(vendor test kit
instructions). If the resulting %CV was acceptable for both repeat samples, they were both
included in the result tables, thus resulting in additional data points. In addition, the 0.50 ppb
solutions included all seven replicates from the MDL determination data in addition to the
triplicate analyses of the 0.50 ppb PT samples.
7.3.1 Accuracy
Tables 20, 21, and 22 also present the accuracy results for the DM Test Kit, expressed as %D.
As calculated by Equation 5 (Section 5.1), the reference method value was used for calculation
of accuracy. For LR, the reference method ranged from 0% to 17% less than the target spike
concentration. For LA the reference values ranged from 15% to 33% lower than the target spike
concentration and for RR, they were from 20% to 46% lower depending upon the sample. All
34
-------
data are provided so that the calculation of %D can be calculated relative to the spike value as
well as the reference method if desired by the reader.
Table 20. Abraxis DM Test Kit Sample Results and Reference Method Results for LR
Sample
Description
0.10LR
Avg ± SD
0.50 LR
Avg ± SD
1.0 LR
Avg ± SD
2.0 LR
Avg ± SD
4.0 LR
Avg ± SD
Kit Results:
LR
Equivalents
(ppb)
0.176
0.157
0.150
0.183
0.194
0.172 ±0.018
0.701
0.729
0.683
0.705
0.734
0.619
0.680
0.777
0.816
0.776
0.722 ± 0.057
1.42
1.36
1.33
1.37 ±0.046
2.44
2.72
2.76
2.64± 0.174
4.61
4.61
4.77
4.66 ±0.097
CR
Corrected
Cone, by
Variant
(ppb)
0.176
0.157
0.150
0.183
0.194
0.172 ±0.018
0.701
0.729
0.683
0.705
0.734
0.619
0.680
0.777
0.816
0.776
0.722 ± 0.057
1.42
1.36
1.33
1.37 ±0.046
2.44
2.72
2.76
2.64± 0.174
4.61
4.61
4.77
4.66 ±0.097
Accuracy by
Variant for
Theoretical
Concentration
(% Difference)
76%
57%
50%
83%
94%
72% ±18%
40%
46%
37%
41%
47%
24%
36%
55%
63%
55%
44% ±11%
42%
36%
33%
37% ± 5%
22%
36%
38%
3 5% ±9%
15%
15%
19%
17% ± 2%
Accuracy by
Variant for
Reference
Concentration
(% Difference)
76%
57%
50%
83%
94%
72% ±18%
67%
74%
63%
68%
75%
47%
62%
85%
94%
85%
72% ± 14%
71%
63%
61%
65% ± 6%
28%
43%
45%
3 9% ±9%
24%
24%
29%
26% ± 3%
Reference
Concentration
(ppb)
0.10
0.42
0.83
1.9
3.7
35
-------
Table 21. Abraxis DM Test Kit Sample Results and Reference Method Results for LA
Sample
Description
0.50 LA
Avg ± SD
LOLA
Avg ± SD
2.0 LA
Avg ± SD
4.0 LA
Avg ± SD
Kit Results:
LR
Equivalents
(ppb)
0.559
0.575
0.631
0.506
0.536
0.474
0.516
0.534
0.523
0.536
0.539 ±0.043
0.998
0.975
1.05
1.01± 0.036
2.06
2.29
2.12
2.16 ±0.120
3.72
3.82
3.75
3.76 ±0.055
CR
Corrected
Cone. By
Variant
(ppb)
1.17
1.20
1.32
1.05
1.12
0.988
1.08
1.11
1.10
1.12
1.12 ±0.089
2.08
2.03
2.18
2. 10 ±0.074
4.30
4.78
4.42
4.50 ±0.249
7.74
7.97
7.81
7.84 ±0.115
Accuracy by
Variant for
Theoretical
Concentration
(% Difference)
133%
140%
163%
111%
123%
98%
115%
123%
118%
123%
125% ±18%
108%
103%
118%
110% ±7%
115%
139%
121%
125% ±12%
93%
99%
95%
96% ± 3%
Accuracy by
Variant for
Reference
Concentration
(% Difference)
191%
199%
229%
164%
179%
147%
169%
178%
172%
179%
181% ±22%
197%
190%
211%
199% ±11%
153%
181%
160%
165% ±15%
158%
165%
160%
161% ±4%
Reference
Concentration
(ppb)
0.40
0.70
1.7
3.0
36
-------
Table 22. Abraxis DM Test Kit Sample Results and Reference Method Results for RR
Sample
Description
0.50 RR
Avg ± SD
1.0 RR
Avg ± SD
2.0 RR
Avg ± SD
4.0 RR
Avg ± SD
Kit Results:
LR
Equivalents
(ppb)
0.501
0.482
0.471
0.506
0.411
0.417
0.399
0.364
0.532
0.468
0.455 ±0.054
1.01
1.03
1.03
1.02 ±0.012
2.07
1.95
1.95
1.99 ±0.066
3.84
3.72
3.67
3.74 ±0.087
CR
Corrected
Cone. By
Variant
(ppb)
0.945
0.909
0.889
0.955
0.775
0.787
0.753
0.687
1.00
0.883
0.859 ±0.103
1.90
1.94
1.93
1.93 ±0.022
3.90
3.69
3.68
3.75 ±0.125
7.24
7.03
6.92
7.06 ±0.164
Accuracy by
Variant for
Theoretical
Concentration
(% Difference)
89%
82%
78%
91%
55%
57%
51%
37%
101%
77%
72% ±21%
90%
94%
93%
93% ±2%
95%
84%
84%
88% ± 6%
81%
76%
73%
77% ±4%
Accuracy by
Variant for
Reference
Concentration
(% Difference)
149%
139%
134%
151%
104%
107%
98%
81%
164%
132%
126% ±27%
252%
260%
258%
257% ± 4%
144%
130%
130%
13 5% ±8%
126%
120%
116%
121% ±5%
Reference
Concentration
(ppb)
0.38
0 54
1.6
3.2
For the LR spiked samples, the reference method results ranged from 0% to 17% less than the
target spike concentration. For LR, the %D ranged from 24% to 94%. For the 0.10 ppb samples,
the %D ranged from 50% to 94%, but the absolute difference from the reference concentration
was never more than 0.094 ppb. For the 0.50 ppb samples, the %D ranged from 47% to 94%,
corresponding to a maximum absolute difference from the reference concentration of 0.396 ppb.
For the 1.0 ppb samples, the %D ranged from 61% to 71% and the maximum absolute difference
from the reference concentration was 0.592 ppb. For the 2.0 ppb samples, the %D ranged from
28% to 45% and the maximum absolute difference from the reference concentration was 0.856
ppb. For the 4.0 ppb samples, the three reported results had %Ds of less than 29%. For LR, the
%D when compared to the theoretical spike concentration ranged from -24% to 94% with the
overall average %D values ranging from 26% to 72%.
For the LA spiked samples, the reference method results ranged from 15% to 33% less than the
target spike concentration, and 20 to 46% lower than the target spike concentration for the RR
spiked samples. For LA and RR, the %D ranged from 147% to 229% and 81% to 260%,
respectively. These %Ds are calculated based on the concentration being corrected for the CR of
the LA and RR variant. However, for both variants, the data suggest that the uncorrected results
in LR equivalents would provide concentrations that were much more similar to the reference
method concentrations.
37
-------
7.3.2 Precision
Precision results for the Abraxis DM Test Kit are presented in Table 23. The RSD was
determined as a percentage according to Equation 7 (Section 5.3) for all DI water, matrix
interferent and recreational water samples. The RSDs ranged from 2% to 11% for the LR
variant, from 1% to 9% for the LA variant, and from 1% to 12 % for the RR variant. The RSD
results for the RW samples ranged from 2% to 9%. The overall average of all RSDs was 7%,
with a minimum of 1% and a maximum of 35%.
38
-------
Table 23. Abraxis DM Test Kit Precision Results
Variant
LR
LA
RR
Unknown
Sample Concentration in DI
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb LR in 1 mg/L Chlorophyll-a DI
2.0 ppb LR in 10 mg/L Chlorophyll-a DI
2.0 ppb LR in lOx dilution of RW Matrix
2.0 ppb LR in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb LA in 1 mg/L Chlorophyll-a DI
2.0 ppb LA in 10 mg/L Chlorophyll-a DI
2.0 ppb LA in lOx dilution of RW Matrix
2.0 ppb LA in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
2.0 ppb RR in 1 mg/L Chlorophyll-a DI
2.0 ppb RR in 10 mg/L Chlorophyll-a DI
2.0 ppb RR in lOx dilution of RW Matrix
2.0 ppb RR in RW Matrix
RWl(10x dilution)
RW 1 (20x dilution)
RW 2 (lOx dilution)
RW 2 (20x dilution)
RW 3 (lOx dilution)
RW 3 (20x dilution)
RW4
RW 4 (4x dilution)
RW5
RW 5 (4x dilution)
RW6
RW 6 (2x dilution)
RW7
RW8
RW9
Precision (%RSD)
11%
8%
3%
7%
2%
10%
3%
5%
6%
8%
4%
6%
1%
3%
9%
3%
4%
12%
1%
3%
2%
3%
4%
2%
2%
7%
3%
6%
5%
3%
4%
8%
5%
2%
4%
9%
2%
NA
6%
5%
NA - Result was less than the LOQ so no calculation of RSD
39
-------
7.3.3 Linearity
The linearity of the DM Test Kit measurements was assessed by performing a linear regression
of the DM Test Kit results against the reference method results for the five PT samples ranging
from 0.10 to 4.0 ppb of microcystin LR in DI water and four PT samples ranging from 0.50 to
4.0 ppb for microcystin LA and RR in DI water. Figures 6, 7, and 8 present the results of the
linear regressions for LR, LA, and RR, respectively. The slope, intercept, and coefficient of
determination (r2) for each regression equation are shown on the charts. All linear regressions
compared to the reference method results had r2 > 0.98.
9
8
Q.
~ 6
u
i *
I 4
U)
I 2
1 -
0
y=1.23x +0.187
R2 = 0.993
0123
Reference Cone, (ppb)
Figure 6. Linearity for the Abraxis DM Test Kit for LR
y = 2.57x +0.135
R2 = 0.997
0123
Reference Cone, (ppb)
Figure 7. Linearity for the Abraxis DM Test Kit for LA
40
-------
9
8
3" 7
Q. '
Q.
I5 -
•Is
1 -
0
= 2.16x + 0.207
R2 = 0.985
0 1 Reference £onc. (ppb) 3 4
Figure 8. Linearity for the Abraxis DM Test Kit for RR
7.3.4 Method Detection Limit
The MDL was assessed by analyzing at least seven replicates of a sample spiked at
approximately five times the vendor-stated detection limit for the microcystin test kit (which was
0.10 ppb). Table 24 lists the replicate results, the %CV of the duplicate DM Test Kit analysis for
each individual replicate, the standard deviations for the replicate results, and shows the
calculated MDLs for the three variants. The calculated MDL values were 0.11, 0.08, and 0.10
ppb for LR, LA, and RR, respectively.
Table 24. Detection Limit Results for the Abraxis DM Test Kit
Variant
Sample
Concentration (ppb)
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
Standard Deviation
t (n=10)
MDL
LR
Mean
Cone.
(ppb)
0.70
0.73
0.68
0.71
0.73
0.62
0.68
0.78
0.82
0.78
0.06
1.8
0.11
%cv
2.0
2.9
4.7
17
11
3.5
2.8
1.4
5.1
7.6
LA
Mean Cone.
(ppb LR
Equivalents)
0.56
0.58
0.63
0.51
0.54
0.47
0.52
0.53
0.52
0.54
0.04
1.8
0.08
%CV
2.2
11
4.4
7.2
6.7
3.4
2.5
2.8
13
4.2
RR
Mean Cone.
(ppb LR
Equivalents)
0.50
0.48
0.47
0.51
0.41
0.42
0.40
0.36
0.53
0.47
0.05
1.8
0.10
%cv
13
4.9
5.5
3.9
4.0
13
6.2
2.6
2.3
2.8
7.3.5 Inter-Kit Lot Reproducibility
Two sets of kit calibration standards were analyzed on the sample plate to compare whether or
not the calibration standards from different lots were similar. The data are presented in Table 25.
The OD values were compared by calculation of the RPD between each pair of OD
41
-------
measurements. In addition, the RPD for each pair of OD results are shown. The RPDs were less
than 10% and all but five were less than 5%.
Table 25. Inter-kit lot Comparison of Kit Calibration Standards for the DM Test Kit
Standard (ppb)
0
0.15
0.40
1.0
2.0
5.0
OD Values
Set A
1.14
1.12
0.975
0.996
0.762
0.719
0.489
0.451
0.294
0.288
0.144
0.146
SetB
1.12
1.16
0.953
0.957
0.772
0.711
0.442
0.439
0.276
0.265
0.154
0.158
RPD
2%
4%
2%
4%
1%
1%
10%
3%
6%
8%
7%
8%
7.3.6 Matrix Effects
Matrix interference effects were assessed by using a t-test to compare the DM Test Kit results
generated from samples made by 2.0 ppb spiking of undiluted and diluted interference matrices
with the PT sample results at the same concentration. The two possible interfering matrices
included a RW sample both undiluted and after undergoing a tenfold dilution and chlorophyll-a
at 1 mg/L and 10 mg/L. Tables 26 and 27 provide the DM Test Kit sample results for the RW
matrix interference samples and chlorophyll-a interference samples, respectively, including the
average and SD for each sample. Because this comparison is made to evaluate only the impact
of the matrix on the sample result, LR equivalents are used.
Each paired t-test was performed using the replicate data from each type of sample. The null
hypothesis is that the difference between the two sets of data is zero. The resulting probability
(p)-value gives the likelihood that the null hypothesis would be true. Therefore, at the 95%
confidence interval, p-values less than 0.05 indicate there is a small likelihood of the null
hypothesis being true and therefore a significant difference between the two sets of data (at low
power).
Table 28 summarizes the results of a paired t-test for both sets of interference data by showing
the p-values associated with each of the applicable comparisons across both types of possible
interfering matrices. Both the RW matrix results for LA and the lOx RW sample were
significantly different from the DI water results and the diluted and undiluted RW were also
significantly different from one another. Table 26 shows the 2.0 ppb LA spike into DI water
generated an average result of 2.16 ppb compared with the average result of 1.71 and 2.02 ppb
for the diluted and undiluted RW samples, respectively. The chlorophyll-a results for LR, LA,
and RR were all statistically different when compared to the DI results except the 1 mg/L
chlorophyll-a solutions spiked with RR (p = 0.169). Table 27 shows the 2.0 ppb LR spike into
DI water generated an average result of 2.64 ppb compared with an average result of 0.470 and
0.480 ppb for the spike into 1.0 mg/L and 10 mg/L chlorophyll-a, respectively. For LA, the 2.0
ppb spike into DI water generated an average result of 2.16 ppb compared with an average result
of 0.340 and 0.320 ppb for the spike into 1.0 mg/L and 10 mg/L chlorophyll-a, respectively.
42
-------
There was no difference determined when comparing the two levels of chlorophyll-a solution
results for all three variants. All of the undiluted and diluted RW samples were significantly
different from one another for all three variants. The spiked undiluted RW samples each
exhibited a higher microcystin concentration than did the diluted RW sample even after the
samples were corrected for any background microcystin present.
Table 26. RW Matrix Interferent Sample Results for the Abraxis DM Test Kit
Variant
Unknown
LR
LA
RR
Sample
Description
UnspikedRW
Matrix (RW 9)
2.0 ppb LR in
DI
2.0 ppb LR in
tenfold
dilution of RW
Matrix
2.0 ppb LR in
RW Matrix
2.0 ppb LA in
DI
2.0 ppb LA in
tenfold
dilution of RW
Matrix
2.0 ppb LA in
RW Matrix
2.0 ppb RR in
DI
2.0 ppb RR in
tenfold
dilution of RW
Matrix
2.0 ppb RR in
RW Matrix
Mean Kit Results:
LR Equivalents
(ppb)
0.44
0.48
0.47
2.43
2.72
2.75
2.34
2.40
2.56
3.48
3.10
3.20
2.06
2.29
2.12
1.69
1.75
1.67
1.93
2.11
2.00
2.06
1.95
1.94
1.81
1.83
1.75
2.13
2.21
2.12
Average
Result
(ppb)
0.470
2.64
2.44
3.26
2.16
1.71
2.02
1.99
1.80
2.16
SD
0.025
0.174
0.111
0.197
0.120
0.045
0.090
0.066
0.040
0.049
0.025
CR Corrected
Cone. By
Variant (ppb)
2.43
2.72
2.75
2.34
2.40
2.56
3.48
3.10
3.20
4.30
4.78
4.42
3.53
3.66
3.48
4.04
4.41
4.17
3.90
3.69
3.68
3.42
3.47
3.32
4.04
4.19
4.02
43
-------
Table 27. Chlorophyll-a Interferent Sample Results for the Abraxis DM Test Kit
Variant
LR
LA
RR
Sample
Description
2.0ppbLRinDI
2.0 ppb LR in 1
mg/L
Chlorophyll-a DI
2.0 ppb LR in 10
mg/L
Chlorophyll-a DI
2.0 ppb LA in DI
2.0 ppb LA in 1
mg/L
Chlorophyll-a DI
2.0 ppb LA in 10
mg/L
Chlorophyll-a DI
2.0ppbRRinDI
2.0 ppb RR in 1
mg/L
Chlorophyll-a DI
2.0 ppb RR in 10
mg/L
Chlorophyll-a DI
Mean Kit
Results: LR
Equivalents (ppb)
2.44
2.72
2.76
0.482
0.503
0.413
0.492
0.477
0.460
2.06
2.29
2.12
0.350
0.343
0.328
0.293
0.323
0.352
2.07
1.95
1.95
1.82
1.92
1.84
2.14
1 99
2.04
Average
Result
(ppb)
2.64
0.470
0.480
2.16
0.340
0.320
1.99
1.86
2.06
CR Corrected
Cone. By Variant
SD (ppb)
0.174 2.44
2.72
2.76
0.047 0.482
0.503
0.413
0.016 0.492
0.477
0.460
0.120 4.30
4.78
4.42
0.011 0.729
0.715
0.683
0.030 0.610
0.673
0.733
0.066 3.90
3.69
3.68
0.054 343
3.62
3.46
0.077 4.04
3 76
3.84
Given that the molecular basis on which the test kits operate is well-characterized and
understood from the literature8, Table 27 provided unexpected results. Two variants (LR and
LA) demonstrated an interference effect but the third variant (RR) did not. This could have been
caused by a number of factors, such as chlorophyll-a source and stability. However, due to the
limited number of replicates that were analyzed, additional testing would be required to provide
a better understanding as to whether there is matrix interference due to chlorophyll-a, or another
variable not investigated in this verification testing.
44
-------
Table 28. Statistical Comparisons between Interference Samples for the Abraxis DM Test
Kit
Description of Comparison
2.0 ppb in DI compared with 2.0 ppb in tenfold
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in undiluted
RW
2.0 ppb in undiluted RW compared with tenfold
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in 1 mg/L
Chlorophyll-a DI
2.0 ppb in DI compared with 2.0 ppb in 10 mg/L
Chlorophyll-a DI
2.0 ppb in 1 mg/L Chlorophyll-a DI compared
with 2.0 ppb in 10 mg/L Chlorophyll-a DI
p-value (D-different, ND-not different)
LR
0.087 (ND)
0.097 (ND)
0.034 (D)
0.003 (D)
0.003 (D)
0.672 (ND)
LA
0.011(0)
0.017 (D)
0.012 (D)
0.001 (D)
0.001 (D)
0.529 (ND)
RR
0.042 (D)
0.090 (ND)
0.002 (D)
0.169(ND)
0.044 (D)
0.111 (ND)
Shading indicates a statistically significant difference
7.4 RW Sample Results
Table 29 presents the RW results for the DM Test Kit and the reference analysis. The
concentrations were determined by the reference method for only three of the approximately 80
variants that are naturally occurring in recreational waters. The total microcystins measured by
the DM Test Kit may have other variants present that would not have been detected by the
reference method. Therefore, no quantitative comparison was made between the ADDA Test Kit
and the reference method results. The reference data were converted into LR-equivalents
according to the DM Test Kit cross reactivity for the variants. In general, the samples that were
determined to have higher total concentrations by the DM Test Kit had higher total
concentrations as determined by the reference method. All of the DM Test Kit total microcystin
results were greater than the reference method results, which only quantified three variants.
However, the results of the DM Test Kit were usually within a factor of two of the reference
method. The LR, LA, and RR variants likely make up a significant proportion of the
microcystins that are measurable by the DM Test Kit.
45
-------
Table 29. Recreational Water Sample Results for the Abraxis DM Test Kit
Sample
Description
RW 1 (lOx
dilution)
RW 1 (20x
dilution)
RW2(10x
dilution)
RW2(20x
dilution)
RW 3 (lOx
dilution)
RW 3 (20x
dilution)
RW4
RW 4 (4x
dilution)
RW5
RW5(4x
dilution)
RW6
RW 6 (2x
dilution)
RW7
RW8
RW 9 (RW
Matrix)
Test Kit Results
Results: LR
Equivalents
(ppb)
2.417
2.194
1.168
1.157
1.226
1.410
1.298
0.721
0.668
0.735
1.427
1.362
1.337
0.737
0.698
0.690
3.914
4.039
4.543
1.079
0.974
1.003
4.568
4.608
4.410
1.128
1.221
1.154
1.682
1.881
2.009
0.904
0.934
0.924
-------
7.5 Operational Factors
During testing activities, the technical operators were instructed to fill out an Ease of Use
Questionnaire that is an appendix in the TQAP1 for this verification test. This section
summarizes these observations as well as other operational considerations about the technology.
7.5.1 Ease of Use
The test kit operator reported that the DM test kit was easy to use. The brochure and flow charts
with illustrations were clear and easy to follow. Solution or sample preparation is minimal,
involving diluting the wash solution or the samples that are above the quantification range. The
procedure includes two incubation periods that total 2 hours. The solutions in the kit produce a
color change in the wells, confirming that those wells contain the solution. This feature is
extremely helpful as technicians can become confused about which wells have had the solution
added and which ones have not when analyzing 96-well plates. Previous knowledge or training
on the use of micro-pipettes and/or multi-channel pipettes is recommended for consistent
readings. The Battelle operator that was trained by Abraxis had more than 10 years of analytical
laboratory experience, but was not experienced with ELISA analysis. A spectrophotometer plate
reader is necessary for obtaining the readings that are then analyzed using any commercial
ELISA evaluation program (four-parameter is recommended by the vendor). Once the analysis
was complete, the remaining solutions and well contents were disposed in the trash in
accordance with local regulations.
7.5.2 Cost and Consumables
The listed price for the DM test kit at the time of the verification test was $400. The kit has a 12-
month shelf life as received and should be stored at 4 to 8 °C. Of the 96-wells on one plate, 16
wells are needed for calibrators and controls. The remaining 80 wells are for sample analyses
that are performed in duplicate. Other consumables required for the test, but not included in the
kit are pipettes, pipette tips, and distilled or DI water. These can be obtained from the vendor.
47
-------
Chapter 8
Test Results for the Abraxis Strip Test Kit
The following subsections provide the results of the Abraxis Strip Test Kit.
8.1 Abraxis Strip Test Summary
As described in Section 2.3, the Strip Test is a qualitative test for microcystins and nodularins.
The Strip Test semi-quantitative results in the range of 0 to 10 ppb can be obtained by comparing
the test line intensity to those that have been produced by solutions of known microcystin
concentrations and are shown in the Strip Test brochure. In summary, the test strip is exposed to
a water sample; as the water moves up the test strip through capillary action, the water does not
inhibit an antibody antigen interaction. This uninhibited antibody antigen interaction results in
the appearance of a control line in the test line region of the test strip. The control line should be
present on the test strip when wetted, and by design, is not influenced by the presence or absence
of microcystins in the water sample. Therefore, it should be present whenever testing is being
conducted. In the absence of microcystins or nodularins in water being tested, a second line that
is the same color, but can vary in darkness in comparison to the control line, appears in the test
line region. However, if microcystins or nodularins are present, the second line appears lighter
in color than the control line or does not appear at all, leaving only the control line. The Strip
Test brochure contains an interpretation figure that depicts the relative darkness of the second
line to the control line over the microcystin detection range. This figure was used to interpret the
observations of the test strips. During this ETV test, when the control line and the test line were
observed to be the same color (denoted as dark line,line, or faint line), the microcystin
concentration was interpreted as 0 to 10 ppb and when the test line was lighter in color than the
control line or absent (denoted as no line) the concentration was interpreted to be greater than 10
ppb. The Strip Test includes a lysing procedure for the determination of total microcystins
called, QuikLyse™. The Strip Test and QuikLyse™ reagents are designed to be used in
combination. According to Abraxis, use of the Strip Test without the QuikLyse™ reagents will
adversely affect the performance of the test, producing inaccurate results.
8.2 Test Kit QC Sample
The QC samples analyzed with the Abraxis Strip Test Kit included KB samples. In addition,
each test strip included the control line as described above. Ten percent of all samples analyzed
were KB samples. The results of all KB sample results were 0 to 10 ppb and are presented in
Table 30. The KB sample was analyzed by the reference method and was determined to be
below the LOQ. The control line on each test strip for the KB samples and all test samples
appeared as expected and therefore, all results were accepted and none of the samples were re-
analyzed.
48
-------
Table 30. RB Sample Results for the Abraxis Strip Test Kit
Sample Description
RB 1
RB2
RB3
RB4
RB5
RB6
RB7
RB8
RB9
Batch
1
1
1
4
4
4
7
7
7
Control line? (Y/N)
y
y
y
y
y
y
y
y
y
Observation
Line
Line
Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
8.3 PT Samples
8.3.1 DI Water Samples
Tables 31, 32, and 33 present the results for the PT samples (concentrations between 0.10 and 15
ppb) for the three variants of microcystin used during this verification test. All of the samples
were analyzed in triplicate and each of them produced a control line on the Test Strip to indicate
that the Test Strip was functioning properly. As the concentrations of the various microcystin
variants increased, the line generated in the test area region was observed to change as expected.
The lowest concentration samples generated dark lines, the mid-level concentrations generated
lighter lines, and the highest concentrations were either very faint lines or generated no line at
all. The line colors were consistent within the three replicates at each concentration and were
consistent with the results from the reference method with one exception. The 7.0 ppb RR PT
sample did not generate a line, which indicated a concentration of greater than 10 ppb. The
results of the Strip Test kit identified solutions with > 10 ppb microcystin concentration.
49
-------
Table 31. Abraxis Strip Test Kit Microcystin-LR DI Water Sample Results
Sample
Description
O.lOppbLR
0.50 ppb LR
l.OppbLR
2.0 ppb LR
4.0 ppb LR
15 ppb LR
Control line?
(Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
Dark Line
Dark Line
Dark Line
Line
Line
Line
Line
Line
Line
Line
Line
Line
Faint line
Faint line
Faint line
No line
No line
No line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
Reference
Concentration (ppb)
0.10
0.42
0.83
1.9
3.7
15
50
-------
Table 32. Abraxis Strip Test Kit Microcystin-LA DI Water Sample Results
Sample
Description
0.50 ppb LA
l.OppbLA
2.0 ppb LA
4.0 ppb LA
7.0 ppb LA
15 ppb LA
Control line?
(Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Line
Line
Line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
No line
No line
No line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
Reference
Concentration (ppb)
0.40
0.70
1.7
3.0
4.7
12
51
-------
Table 33. Abraxis Strip Test Kit Microcystin-RR DI Water Sample Results
Sample
Description
0.50 ppb RR
l.OppbRR
2.0 ppb RR
4.0 ppb RR
7.0 ppb RR
15 ppb RR
Control line?
(Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
Dark Line
Dark Line
Dark Line
Line
Line
Line
Line
Line
Line
Faint line
Faint line
Faint line
No line
No line
No line
No line
No line
No line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
Reference
Concentration (ppb)
0.38
0.54
1.6
3.2
4.5
15
8.3.2 Matrix Interference Samples
Matrix interference Table 34 presents the RW matrix interference sample results for the Strip
Test kit and Table 35 presents the chlorophyll-a interference sample results. The triplicate
analyses of the samples all agreed and the interpretation of the results is consistent with the
spiked amount of microcystins in the samples. Consequently, there was no indication that the
different matrices affected the test kit performance.
52
-------
Table 34. RW Matrix Interferent Sample Results for the Strip Test Kit
Variant
LR
LA
RR
Sample Description
4.0ppbLRinDI
4.0 ppb LR in lOx
dilution of RW Matrix
4.0ppbLRinRW
Matrix
4.0 ppb LA in DI
4.0 ppb LA in lOx
dilution of RW Matrix
4.0 ppb LA in RW
Matrix
4.0ppbRRinDI
4.0 ppb RR in lOx
dilution of RW Matrix
4.0ppbRRinRW
Matrix
Control line?
(Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
53
-------
Table 35. Matrix Interferent Sample Results for the Abraxis Strip Test Kit
Variant
LR
LA
RR
Sample Description
4.0ppbLRinDI
4.0 ppb LR in 1 mg/L
Chlorophyll-a DI
4.0 ppb LR in 10 mg/L
Chlorophyll-a DI
4.0 ppb LA in DI
4.0 ppb LA in 1 mg/L
Chlorophyll-a DI
4.0 ppb LA in 10 mg/L
Chlorophyll-a DI
4.0ppbRRinDI
4.0 ppb RR in 1 mg/L
Chlorophyll-a DI
4.0 ppb RR in 10 mg/L
Chlorophyll-a DI
Control line?
(Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Faint line
Interpretation
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
54
-------
8.4 RW Samples
Table 36 presents the RW results for the Strip Test Kit and the reference method results. The
microcystin concentrations in the samples that were determined by the reference method
represent three of the approximately 80 variants that are naturally occurring in RWs. The total
measured microcystin result may have other variants present that cannot be detected by the
reference method. To assess the lysing procedure, aliquots of RW 6, RW 7, and RW 8 were
analyzed before and after the sample went through the freeze-thaw lysing procedure. The
observations and interpretations of these RW samples were not different with respect to lysing
procedures. With the exception of RW 5, the triplicate measurements of the samples generated
lines that were very similar in intensity. In two replicates, RW 5 generated faint lines indicating
a 0 to 10 ppb concentration and one replicate that had no line, indicating a concentration of
greater than 10 ppb. Again, except for one exception for RW 5, the sample results were
consistent with the reference laboratory results of the RW samples. That is, with the one RW 5
exception, when the reference concentrations were greater than 10 ppb, there was no line
generated; when the reference concentration was between 2.0 and 4.0 ppb, the lines were
medium dark or faint, and when the reference concentrations were lower than 2.0, the lines
generated were dark.
55
-------
Table 36. Recreational Water Sample Results for the Abraxis Strip Test Kit
Sample Description
RW1
RW2
RW3
RW4
RW5
RW6
RW6Notlysedby
freeze-thaw method
RW7
RWVNotlysedby
freeze-thaw method
RW8
RWSNotlysedby
freeze-thaw method
RW 9 (RW Matrix)
Control
line? (Y/N)
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Observation
No line
No line
No line
No line
No line
No line
No line
No line
No line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
Very Faint line
No line
Line
Line
Line
Line
Line
Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Dark Line
Interpretation
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
Reference
Concentration
(ppb)
30
16
10
3.5
3.6
2.0
-------
8.5 Operational Factors
During testing activities, the technical operators were instructed to fill out an Ease of Use
Questionnaire that is an appendix in the TQAP1 for this verification test. This section
summarizes these observations, as well as other considerations regarding the technology.
8.5.1 Ease of Use
The test kit operator reported that the Strip Test Kit was very easy to use and needs no technical
skills to operate (although the Battelle operator was a trained technician). The brochure and flow
charts with illustrations were clear and easy to follow. There was no solution or sample
preparation needed. The entire procedure is approximately 40 minutes long, including the
QuikLyse™ procedure and the microcystins analysis. The QuikLyse™ process uses 1 mL of
sample through 2 x 8 minute incubation periods. Then the sample is transferred into the
microcystins reagent conical tube. The sample is incubated for 10 minutes and then the test strip
is added to the conical tube. The test strip is interpreted according to the figure in the brochure
after 5 minutes of exposure to the sample.
8.5.2 Cost and Consumables
There were no consumables required for this technology. The test strips were disposed in the
regular trash after use, producing no hazardous waste.
The listed price for the Strip Test kit at the time of the verification test was $480 for a 20 strip kit
and $150 for a five strip kit. Each kit has a 12-month shelf life as received and should be stored
at room temperature.
57
-------
Chapter 9
Performance Summary for the Abraxis ADDA, DM, and Strip Test
9.1 Performance Summary for the ADDA ELISA Test Kit
The verification of the Abraxis ADDA ELISA Test Kit is summarized by the parameters
described in Table 37.
Table 37. Abraxis ADDA Test Kit Performance Summary
Verification Parameters
LR
LA
RR
Accuracy (range of % difference from reference method; low %Ds indicate increased accuracy)
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
Precision (range of %RSD)
Precision (RW samples)
Linearity (y=)
Method Detection Limit (ppb)
-13% to 2%
-9% to 40%
19% to 58%
-45% to 3 9%
-6% and 3%
5% to 45% (7 of 9
samples < 16%)
27% to 109%
77% to 133%
69% to 113%
105% and 113%
3% to 25%
42% to 64%
98% to 123%
23% to 69%
11% to 50%
4% to 16%
3% to 47%, all except 2 RSDs were < 12%
0.933x + 0.087
r2=0.906
0.137
2.66x - 0.220
r2=0.990
0.218
1.18x + 0.168
r2=0.961
0.047
Inter-kit lot reproducibility. Calibration standards from two different lots were measured and
the RPD of the resulting optical densities ranged from 0% to 13% with eight of the 12 being less
than 6%.
Matrix Interference. Matrix interference effects were assessed by using a t-test to compare
results from samples made by spiking undiluted and diluted interference matrices with the PT
sample results at 2.0 ppb spiked concentration. For chlorophyll-a and RW matrices, only three
of the 18 comparisons resulted in statistically significant differences: 1) 2.0 ppb LA spike into
DI water was significantly different from the 2.0 ppb LA spike into both 1 mg/L (p=0.005); 2) 10
mg/L (p=0.006) chlorophyll-a; and 3) The other statistically significant difference was between
the RR spikes into undiluted and diluted RW (p=0.01). Due to the limited number of replicates
and low statistical power of this study, additional testing would be required to provide a better
understanding as to whether there is a matrix interference due to chlorophyll-a.
Recreational Water (RW). Because the reference method did not measure all possible
microcystin variants, no quantitative comparison was made between the ADDA Test Kit and the
reference method results. The reference data were converted into LR-equivalents according to
the ADDA Test Kit cross reactivity for the variants. In general, the samples that were
determined to have higher total concentrations by the ADDA Test Kit had higher total
concentrations as determined by the reference method. All of the ADDA test kit total
microcystin results were greater than the reference method results, which was consistent with the
likelihood that all of the microcystins were not being measured by the reference method.
58
-------
Operational Factors. The test kit operator reported that the ADDA Test Kit was easy to use.
Solution or sample preparation was minimal, mostly involving diluting the samples that were
initially above the quantification range. The procedure included three incubation periods that
total 2.5 hours. Previous knowledge or training on the use of micro-pipettes and or multi-
channel pipettes with 96-well plates is recommended for consistent readings. A
spectrophotometer plate reader is necessary for obtaining the spectrophotometric readings that
are then analyzed using any commercial ELISA evaluation program (four-parameter is
recommended by the vendor). Once the analysis was complete, the remaining solutions were
disposed of in the trash in accordance with local regulations.
The listed price for the ADDA Test Kit at the time of the verification test was $440. The kit has
a 12-month shelf life as received and should be stored at 4 to 8 °C. Of the 96-wells on one plate,
16 wells are needed for calibration and control samples. The remaining 80 wells are for sample
analyses that are performed in duplicate. Other consumables not included in the kit are pipettes,
pipette tips, and distilled or DI water can be provided by the vendor.
9.2 Performance Summary for the DM ELISA Test Kit
The verification of the Abraxis DM ELISA Test Kit is summarized by the parameters described
in Table 38.
Table 38. Abraxis DM Test Kit Performance Summary
Verification Parameters
LR
LA
RR
Accuracy (range of % difference from reference method; low %Ds indicate increased accuracy)
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
Precision (range of %RSD)
Precision (RW samples)
Linearity (y=)
Method Detection Limit (ppb)
50% to 94%
47% to 94%
61% to 71%
28% to 45%
24% to 29%
2% to 11%
For LA, 147% to 229%, and for RR, 81% to
260%. For both variants, the data suggested that
the uncorrected results in LR equivalents would
provide concentrations that were more similar to
the reference method concentrations.
I%to9%
I%tol2%
2% to 9%
1.23x + 0.187
r2 = 0.993
0.105
2.57x + 0.135
r2 = 0.997
0.078
2.16x + 0.207
r2 = 0.985
0.099
Inter-kit lot reproducibility. Calibration standards from two different lots were measured and
the RPD of the resulting optical densities ranged from 1% to 10% with seven of the 12 being less
than 5%.
Matrix Interference. Matrix interference effects were assessed by using a t-test to compare
results from samples made by spiking undiluted and diluted interference matrices with the PT
sample results at 2.0 ppb spiked concentration. Both the RW matrix results for LA and the lOx
RW sample were significantly different from the DI water results and the diluted and undiluted
RW were also significantly different from one another. The chlorophyll-a results for LR, LA,
and RR were all statistically different when compared to the DI results except the 1 mg/L
chlorophyll-a solutions spiked with RR (p = 0.169). There was no difference determined when
comparing the two levels of chlorophyll-a solution results for all three variants. All of the
undiluted and diluted RW samples were significantly different from one another for all three
variants. The spiked undiluted RW samples each exhibited a higher microcystin concentration
than did the diluted RW sample even after the samples were corrected for any background
59
-------
microcystin present. Due to the limited number of replicates and low statistical power of this
study, additional testing would be required to provide a better understanding as to whether there
is a matrix interference due to chlorophyll-a.
Recreational Water (RW). Because the reference method did not measure all possible
microcystin variants, no quantitative comparison was made between the DM Test Kit and the
reference method results. The reference data were converted into LR-equivalents according to
the DM Test Kit cross reactivity for the variants. As for the ADDA Test Kit, the samples that
were determined to have higher total concentrations by the DM Test Kit had higher total
concentrations as determined by the reference method. All of the DM Test Kit total microcystin
results were greater than the reference method results, which was consistent with the likelihood
that all of the microcystins were not being measured by the reference method.
Operational Factors. The test kit operator reported that the DM test kit was easy to use.
Solution or sample preparation was minimal, mostly involving diluting samples that were above
the quantification range. The procedure included two incubation periods that total 2 hours. The
solutions in the kit produce a color change in the wells, confirming that those wells contain the
solution. This feature was extremely helpful as technicians can become confused about what
wells have had the solution added and which ones have not when analyzing 96-well plates.
Previous knowledge or training on the use of micro-pipettes and or multi-channel pipettes with
96-well plates is recommended for consistent readings. A spectrophotometer plate reader is
necessary for obtaining the spectrophotometric readings that are then analyzed using any
commercial ELISA evaluation program (four-parameter is recommended by the vendor). Once
the analysis was complete, the remaining solutions were disposed of in the trash in accordance
with local regulations.
The listed price for the DM test kit at the time of the verification test was $400. The kit has a 12-
month shelf life as received and should be stored at 4 to 8 °C. Of the 96-wells on one plate, 16
wells were needed for calibrators and controls. The remaining 80 wells are for sample analyses
that are performed in duplicate. Other consumables not included in the kit are pipettes, pipette
tips, and distilled or DI water can be provided by the vendor.
60
-------
9.3 Performance Summary for the Strip Test Kit
The verification of the Abraxis Strip Test is summarized in Table 39 and by the parameters
described below.
Table 39. Abraxis Strip Test Kit Performance Summary
Sample
Description
Spiked Cone.
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
7.0 ppb
15 ppb
LR Results
(Observation and
Interpretation)
Dark Line
Dark Line
Dark Line
Line
Line
Line
Line
Line
Line
Line
Line
Line
Faint line
Faint line
Faint line
NA
NA
NA
No line
No line
No line
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
NA
NA
NA
> 10 ppb
> 10 ppb
> 10 ppb
LA Results
(Observation and
Interpretation)
Dark Line
Dark Line
Dark Line
Line
Line
Line
Line
Line
Line
Line
Line
Line
Faint line
Faint line
Faint line
Very Faint line
Very Faint line
Very Faint line
No line
No line
No line
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
RR Results
(Observation and
Interpretation)
Dark Line
Dark Line
Dark Line
Line
Line
Line
Line
Line
Line
Line
Line
Line
Faint line
Faint line
Faint line
No line
No line
No line
No line
No line
No line
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
0-10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
> 10 ppb
NA = not applicable; 7.0 ppb level was not performed for LR samples.
Accuracy. The DI samples (concentrations between 0.10 and 15 ppb) for the three variants of
microcystin were used during this verification test. All of the samples were analyzed in triplicate
and each of them produced a control line on the Strip Test to indicate that the Strip Test was
functioning properly. As the concentrations of the various microcystin variants increased, the
line generated in the test area region was observed to change as expected. The lowest
concentration samples generated dark lines, the mid-level concentrations generated lighter lines,
and the highest concentrations were either very faint lines or generated no line at all. The Strip
Test kit correctly detected each of the 15 ppb samples as being greater than 10 ppb.
Precision. The line colors were consistent within the three replicates at each concentration and
were consistent with the results from the reference method results with one exception. The 7.0
ppb RR DI water sample did not generate a line, indicating a concentration of greater than 10
ppb. With the exception of RW 5, the triplicate measurements of the samples generated lines
that were very similar in intensity. In two replicates, RW 5 generated faint lines indicating a 0 to
10 ppb concentration and one replicate that had no line, indicating a concentration of greater than
10 ppb.
Matrix Interference. The RW matrix interferent and chlorophyll-a interferent sample results for
the Strip Test Kit all agreed and the interpretation of the results is consistent with the spiked
61
-------
amount of microcystins in the samples. There was no indication that different matrices affected
the test kit performance.
Recreational Water (RW). RW sample results for the Strip Test Kit were also consistent with
the reference method results. While the total measured microcystin result may have other
variants present that cannot be detected by the reference method, with one exception for RW 5,
the sample results were all consistent with the reference laboratory results of the RW samples.
That is, when the reference concentrations were greater than 10 ppb, there was no line generated,
when the reference concentration was between 2.0 and 4.0 ppb, the lines were normal or faint,
and when the reference concentrations were lower than 2.0, the lines generated were dark.
Operational Factors. The test kit operator reported that the Strip Test kit was very easy to use
and needs no technical skills to operate. The brochure and flowcharts with illustrations were
clear and easy to follow. There was no solution or sample preparation needed. The entire
procedure is approximately 40 minutes long, including the QuikLyse™ procedure and the
microcystins analysis. The QuikLyse™ process uses 1 mL of sample through 2 x 8 minute
incubation periods. Then the sample is transferred into the microcystins reagent conical tube.
The sample is incubated for 10 minutes and then the test strip is added to the conical tube. The
test strip is interpreted according to the figure in the brochure after 5 minutes of exposure to the
sample.
There were no consumables required for this technology. The test strips were disposed in the
regular trash after use, producing no hazardous waste.
The listed price for the Strip Test Kit at the time of the verification test was $480 for a 20 strip
kit and $150 for a five strip kit. The kit has a 12-month shelf life as received and should be
stored at room temperature.
62
-------
Chapter 10
References
1. Test/Quality Assurance Plan for Verification ofMicrocystin Test KitsTest/Quality
Assurance Plan for Verification ofMicrocystin Test Kits. U.S. Environmental
Technology Verification Program, Battelle, July 2010.
2. Quality Management Plan for the ETVAdvancedMonitoring Systems Center, Version 7.
U.S. Environmental Technology Verification Program, Battelle, November 2008.
3. Cong, L.H., B.; Chen, Q.; Lu, B.; Zhang, J.; Ren, Y., Determination of trace amount of
microcystins in water samples using liquid chromatography coupled with triple
quadrupole mass spectrometry. Anal. Chim. Acta, 2006. 569 (1-2): p. 157-168.
4. Cassada, D., Standard Operating Procedure (SOP) Determination ofalgaltoxin residues
in water extracts by solid phase extraction (SPE), liquid chromatography (LC)-
atmosphericpressure electrospray ionization tandem mass spectrometry (MS/MS). July,
2010, Water Sciences Laboratory, University of Nebraska.
5. "Guidelines Establishing Test Procedures for the Analysis of Pollutants. ", USEP A,
Editor. 2000, U.S. Code of Federal Regulations.
6. SOP# SWS-2320.1A: Microcystin Analysis Using the Abraxis ELISA (Enzyme-Linked
Immuno-Sorbent Assay) Method. Nebraska Department of Environmental Quality.
7. Loftin, K.A., et al., Comparison of Two Cell Lysis Procedures for Recovery of
Microcystins in Water Samples from Silver Lake in Dover, Delaware, with Microcystin
Producing Cyanobacterial Accumulations, in USGS Open-File Report 2008 -1341. 2008.
p. 9.
8. Fischer, W.; Garthwaite, I; Miles, C.; Ross, K.; Aggen, J.; Chamberlin, A.; Towers, N.;
Dietrich, D. Congener-Independent Immunoassay for Microcystins and Nodular ins.,
Environ. Sci. Techno!., 2001, 35, 4849-4856.
63
-------
APPENDIX A
Reference Laboratory Method Detection Limit Memo
July 14,2010
To: Anne Gregg and Ryan James, Battelle Laboratories
From: Daniel Snow and David Cassada, UNL Water Sciences Laboratory
Re: Summary of Microcystin SPE method validation - July 13-14, 2010
Microcystins LA, LR and RR were spiked into water and extracted using solid phase extraction (SPE) to
evaluate method accuracy and precision, and method detection limits. The method described in Cong et
al. 2006 was modified to allow for extraction of a larger sample by using higher capacity polymeric
(Waters Oasis, HLB) SPE cartridges. Briefly, 400-milliliter (mL) of purified reagent water was fortified
with 1500 uL of a diluted mixed stock (0.1 ng/uL) obtained from Battelle to produce 0.375 ug/L of each
analyte. Nodularin (1600uLofa0.1 ng/uL solution) was also added to produce a concentration of 0.40
ug/L . Eight 50 mL portions of this fortified water were weighed into 125 mL amber glass bottles and
each portion separately spiked with 100 uL of the enkephalin-Leu internal standard (IS) solution (0.1
ng/uL) to give a concentration of 2.0 ug/L. A single method blank was prepared by spiking with IS and
surrogate only.
After capping and shaking each solution to equilibrate, samples were drawn under vacuum through pre-
conditioned 200 mg Oasis HLB SPE cartridges at a rate of approximately 10 mL/min. When the sample
had completely passed through the cartridge, it was allowed to air-dry under vacuum, removed from the
extraction apparatus and prepared for elution. Ten (10) milliliters of high purity methanol (Fisher Optima
Grade) were used to elute analyte, IS and surrogate compounds from the cartridges. The methanol was
evaporated under nitrogen to approximately 0.4 mL and the extracts transferred to low volume inserts for
analysis on the LCQ ion trap tandem LC/MS system. Calibration solutions (5, 10, 30, 60 and 75 ng/mL)
were prepared in water from the same mixed stock as the spiking solutions. A table summarizing the
results of the validation is copied below (Table A-l.).
A second 10-mL aliquot of methanol was passed through 4 of SPE cartridges and collected separately to
check for completeness of analyte elution. These second aliquots were blown down to the same 0.4 mL
volume as the MDLs eluants and analyzed. The resulting absolute areas of the analyte, surrogate, and
internal standard peaks obtained were approximately 1% of the areas obtained in the first portion. This
suggests that lower elution volumes can result in decreased analyte recovery.
References
Cong, L.; Huang, B.; Chen, Q.; Lu, B.; Zhang, J.; Ren, Y. (2006) Determination of trace amount of
microcystins in water samples using liquid chromatography coupled with triple quadrupole mass
spectrometry. Anal. Chim. Acta, 569 (1-2), 157-168.
64
-------
Table A-l. Average, standard deviation, method detection limits (MDL = S x tN-1) and recoveries
of microcystins obtained from extraction and analysis of 8 fortified reagent water (0.375 jig/L)
samples.
50 mL
sample
Aliquot
MDL1
MDL 2
MDL 3
MDL 4
MDL 5
MDL 6
MDL 7
MDL 8
AVG
STD DEV
MDL
%REC
Expected
value
Amount obtained (ng)
Nodularin
23.994
24.647
22.716
23.157
26.361
19.618
20.254
19.889
22.580
2.460
7.371
112.9
20.0
MC-RR
19.679
19.661
17.660
19.715
19.731
18.214
14.533
15.247
18.055
2.113
6.333
96.3
18.75
MC-LR
18.084
21.985
20.524
21.022
20.462
18.322
20.046
17.518
19.745
1.586
4.753
105.3
18.75
MC-LA
19.913
21.752
18.404
20.304
21.182
18.393
21.490
14.614
19.507
2.360
7.072
104.0
18.75
Concentration (jig/L)
Nodularin
0.480
0.493
0.454
0.463
0.527
0.392
0.405
0.398
0.452
0.049
0.147
112.9
0.4
MC-RR
0.394
0.393
0.353
0.394
0.395
0.364
0.291
0.305
0.361
0.042
0.127
96.3
0.375
MC-LR
0.362
0.440
0.410
0.420
0.409
0.366
0.401
0.350
0.395
0.032
0.095
105.3
0.375
MC-LA
0.398
0.435
0.368
0.406
0.424
0.368
0.430
0.292
0.390
0.047
0.141
104.0
0.375
65
-------
APPENDIX B
Abraxis Test Kit Raw Data
Table B-l. Abraxis ADDA ELISA Test Kit Raw Data
Sample Description
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Positive Control 1
Positive Control 2
Positive Control 2a
Positive Control 3
Positive control 4
Positive Control 5
Positive Control 5a
Positive Control 5b
Positive Control 6
StdOdifflot
StdO.lSdifflot
Std0.4difflot
Stdl.Odifflot
Std2.0difflot
StdS.Odifflot
0. LR
0. LR
0. LR
0. LR
0. LR
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
Variant
RB
RB
RB
RB
RB
RB
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
Mean Cone, (ppb)
Range?
Range?
0.015
Range?
Range?
Range?
0.653
0.826
0.808
0.566
0.625
0.717
0.656
0.769
0.903
0.072
0.195
0.467
1.086
2.034
13.829
0.096
0.022
0.087
0.110
0.102
0.667
0.914
0.718
0.969
1.046
0.927
0.842
0.905
0.740
0.750
0.634
0.588
0.584
0.590
0.553
0.477
0.538
Standard Deviation (ppb)
Range?
Range?
0
Range?
Range?
Range?
0.054
0.035
0.050
0.101
0.156
0.014
0.011
0.015
0.147
0.016
0.027
0.004
0.025
0.430
11.59
0.003
0.005
0.010
0.021
0.011
0.157
0.050
0.009
0.035
0.047
0.052
0.030
0.020
0.004
0.094
0.014
0.052
0.014
0.013
0.058
0.071
0.019
cv%
Range?
Range?
0.5
Range?
Range?
Range?
8.2
4.2
6.2
17.8
24.9
2
1.7
1.9
16.3
21.8
13.7
0.8
2.3
21.2
83.8
3.5
21.4
11.2
18.8
11
23.5
5.5
1.2
3.6
4.5
5.6
3.6
2.2
0.6
12.5
2.1
8.9
2.5
2.3
10.5
14.8
3.6
66
-------
Table B-l. Abraxis ADDA ELISA Test Kit Raw Data Continued
Sample Description
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
.OLA
.OLA
.OLA
.OLR
.OLR
.OLR
.ORR
.ORR
.ORR
2.0 LA
2.0 LA (2x dil)
2.0 LA (2x dil)
2.0 LR
2.0 LR
2.0 LR
2.0 LR
2.0 RR
2.0 RR
2.0 RR
4.0 LA (4x dil)
4.0 LA (4x dil)
4.0 LR
4.0 LR
4.0 RR
4.0 RR
4.0 RR
4.0 RR
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
52671-36-09 2x
52671-36-09 2x
Variant
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
LR
RR
RR
RR
LA
LA
LR
LR
RR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LA
LA
LA
Mean Cone, (ppb)
0.550
0.441
0.381
0.434
0.490
0.544
0.552
0.568
0.567
0.518
0.514
0.549
0.563
0.552
2.037
1.781
1.545
1.311
1.003
0.991
1.095
0.973
0.972
3.598
2.160
2.264
1.873
2.632
1.101
1.045
1.796
2.461
2.200
1.924
1.997
3.473
3.819
4.359
4.074
3.241
3.550
0.576
0.617
0.548
0.533
0.669
0.581
0.425
3.014
2.173
Standard Deviation (ppb)
0.054
0.083
0.027
0.007
0.007
0.003
0.018
0.096
0.051
0.004
0.060
0.035
0.050
0.026
0.173
0.074
0.071
0.265
0.236
0.018
0.104
0.039
0.015
0
0.007
0.038
0.465
0.348
0.061
0.182
0.022
0.603
0.232
0.36
0.011
0
0
0.234
0.562
0.594
0.125
0.119
0.052
0.106
0.041
0.106
0.019
0.042
0.247
0.284
cv%
9.8
18.7
7
1.6
1.3
0.5
3.3
17
8.9
0.8
11.7
6.3
8.9
4.7
8.5
4.1
4.6
20.2
23.5
1.8
9.5
4
1.6
0
0.3
1.7
24.8
13.2
5.6
17.4
1.2
24.5
10.5
18.7
0.5
0
0
5.4
13.8
18.3
3.5
20.7
8.4
19.4
7.7
15.8
3.2
9.9
8.2
13.1
67
-------
Table B-l. Abraxis ADDA ELISA Test Kit Raw Data Continued
Sample Description
52671-36-09 2x
2.0 LA Matrix lOx
2.0 Matrix lOx LA (2x
dil)
52671-36-12 2x
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Matrix
2.0 LR Matrix
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix lOx
2.0 RR Matrix lOx
2.0 RR Matrix lOx
RW1 (20x dil)
RW1 (20x dil)
RW2 (lOx dil)
RW2(10xdil)
RW2(10xdil)
RW3 (lOxdil)
RW3 (lOx dil)
RW3 (20x dil)
RW3 (20x dil)
RW3 (20x dil)
RW 4 (4x dil)
RW 4 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW6 (2xdil)
RW6 (2xdil)
RW6 (2xdil)
RW6 (2xdil)
Variant
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone, (ppb)
1.894
3.861
3.030
2.655
0.488
0.532
0.493
0.391
0.432
0.486
4.050
2.598
1.703
1.903
1.658
2.238
1.769
1.574
2.012
2.204
1.975
1.437
1.611
1.573
2.784
2.378
2.505
1.368
1.387
1.286
2.132
1.843
2.235
2.298
2.087
1.742
3.497
1.019
1.202
1.101
2.258
2.667
.574
.587
.280
.388
.405
1.710
1.207
Standard Deviation (ppb)
0.127
0.123
0.103
0.362
0.015
0.064
0.094
0.048
0.019
0.067
0.500
0.236
0.144
0.082
0.226
0.070
0.187
0.317
0.099
0.413
0.072
0.255
0.106
0.114
0.330
0.115
0.204
0.131
0.211
0.044
0.141
0.023
0.272
0.159
0.022
0.168
0.521
0.112
0.089
0.032
0.314
0.018
0.267
0.113
0.061
0.050
0.203
0.268
0.01
cv%
6.7
3.2
3.4
13.6
3.1
12.1
19.2
12.2
4.4
13.7
12.3
9.1
8.5
4.3
13.6
3.1
10.6
20.1
4.9
18.7
3.7
17.8
6.6
7.2
11.8
4.8
8.2
9.6
15.2
3.4
6.6
1.2
12.2
6.9
1.1
9.7
14.9
10.9
7.4
2.9
13.9
0.7
17
7.1
4.7
3.6
14.5
15.7
0.8
68
-------
Table B-l. Abraxis ADDA ELISA Test Kit Raw Data Continued
Sample Description
RW6 (2xdil)
RW6
RW7
RW7
RW7
RW8
RW8
RW8
RW 9 Matrix
RW 9 Matrix
RW 9 Matrix
Variant
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone, (ppb)
0.931
4.588
0.012
0.015
0.050
0.885
0.807
0.857
0.639
0.600
0.621
Standard Deviation (ppb)
0.024
0.931
0.004
0.008
0.015
0.147
0.021
0.026
0.038
0.087
0.119
cv%
2.6
20.3
33
55.1
28.9
16.6
2.6
3.1
5.9
14.4
19.1
69
-------
Table B-2. Abraxis DM ELISA Test Kit Raw Data
Sample Description
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Positive Control 1
Positive Control 2a
Positive Control 2b
Positive Control 3
Positive control 4
Positive Control 5a
Positive Control 5b
StdOdifflot
StdO.lSdifflot
Std0.4difflot
Stdl.Odifflot
Std2.0difflot
StdS.Odifflot
0.1 LR
0.1 LR
0.1 LR
0.1 LR
0.1 LR
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
Variant
RB
RB
RB
RB
RB
RB
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
Mean Cone, (ppb)
0.038
0.004
0.044
0.061
0.061
0.060
0.836
0.746
0.740
0.603
0.756
0.616
0.505
0.018
0.169
0.415
1.094
2.164
4.632
0.176
0.157
0.150
0.183
0.194
0.559
0.575
0.631
0.506
0.536
0.474
0.516
0.534
0.523
0.536
0.701
0.729
0.683
0.705
0.734
0.619
0.680
0.777
0.816
0.776
0.501
0.482
0.471
0.506
0.411
Standard Deviation (ppb)
0
0
0
0.007
0.010
0.004
0.075
0.068
0.035
0.001
0.020
0.039
0.007
0
0.003
0.061
0.009
0.080
0.149
0.017
0.033
0.001
0.003
0.025
0.012
0.063
0.028
0.037
0.036
0.016
0.013
0.015
0.067
0.022
0.014
0.021
0.032
0.118
0.083
0.022
0.019
0.011
0.042
0.059
0.064
0.024
0.026
0.020
0.016
cv%
0
0
0
12
17.3
7
9
9.1
4.7
0.1
2.7
6.4
1.3
0
1.6
14.7
0.8
3.7
3.2
9.5
20.7
0.5
1.6
13.1
2.2
11
4.4
7.2
6.7
3.4
2.5
2.8
12.8
4.2
2
2.9
4.7
16.7
11.4
3.5
2.8
1.4
5.1
7.6
12.8
4.9
5.5
3.9
4
70
-------
Table B-2. Abraxis DM ELISA Test Kit Raw Data Continued
Sample Description
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
LOLA
LOLA
LOLA
1.0 LR
1.0 LR
1.0 LR
1.0 RR
1.0 RR
1.0 RR
2.0 LA
2.0 LA
2.0 LA
2.0 LR
2.0 LR
2.0 LR
2.0 RR
2.0 RR
2.0 RR
4.0 LA
4.0 LA
4.0 LA
4.0 LR
4.0 LR
4.0 LR
4.0 RR
4.0 RR
4.0 RR
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Matrix
2.0 LA Matrix
2.0 LA Matrix
2.0 LA Matrix lOx
2.0 LA Matrix lOx
2.0 LA Matrix lOx
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Chloro lOx
Variant
RR
RR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
Mean Cone, (ppb)
0.417
0.399
0.364
0.532
0.468
0.998
0.975
.045
.422
.356
.334
.007
.029
.025
2.064
2.294
2.122
2.439
2.720
2.756
2.066
1.954
1.948
3.715
3.823
3.750
4.605
4.606
4.773
3.838
3.723
3.668
0.293
0.323
0.352
0.350
0.343
0.328
1.939
2.117
2.003
1.695
1.758
1.672
0.492
0.477
0.460
0.482
0.503
0.413
Standard Deviation (ppb)
0.052
0.025
0.010
0.012
0.013
0.194
0.102
0.083
0.003
0.077
0.219
0.122
0.079
0.010
0.037
0.076
0.140
0.002
0.117
0.293
0.018
0.024
0.028
0.101
0.067
0.036
0.167
0.080
0.060
0.151
0.090
0.267
0.002
0.066
0.019
0.050
0.014
0.010
0.079
0.229
0.023
0.083
0.074
0.076
0.013
0.028
0.053
0.010
0.009
0.052
CV%
12.5
6.2
2.6
2.3
2.8
19.4
10.4
8
0.2
5.7
16.4
12.1
7.7
1
1.8
3.3
6.6
0.1
4.3
10.6
0.9
1.2
1.4
2.7
1.7
1
3.6
1.7
1.3
3.9
2.4
7.3
0.6
20.4
5.5
14.4
4.1
3.1
4.1
10.8
1.2
4.9
4.2
4.5
2.6
5.8
11.6
2
1.8
12.5
71
-------
Table B-2. Abraxis DM ELISA Test Kit Raw Data Continued
Sample Description
2.0 LR Matrix
2.0 LR Matrix
2.0 LR Matrix
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix lOx
2.0 RR Matrix lOx
2.0 RR Matrix lOx
RW1 (lOx dil)
RW1 (lOx dil)
RW1 (lOxdil)
RW1 (20x dil)
RW1 (20x dil)
RW2(10xdil)
RW2 (lOx dil)
RW2 (20x dil)
RW2 (20x dil)
RW2 (20x dil)
RW3 (lOx dil)
RW3 (lOx dil)
RW3 (lOx dil)
RW3 (20x dil)
RW3 (20x dil)
RW3 (20x dil)
RW4
RW4
RW4
RW 4 (4x dil)
RW 4 (4x dil)
RW 4 (4x dil)
RW5
RW5
RW5
RW 5 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW6
RW6
RW6
RW 6 (2x dil)
Variant
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone, (ppb)
3.484
3.102
3.208
2.345
2.405
2.560
2.143
1.993
2.035
1.819
1.920
1.835
2.139
2.219
2.129
1.813
1.837
1.759
2.417
2.194
.168
.157
.226
.410
.298
0.721
0.668
0.735
1.427
1.362
1.337
0.737
0.698
0.690
3.914
4.039
4.543
1.079
0.974
1.003
4.568
4.608
4.410
.128
.221
.154
.682
.881
2.009
0.904
Standard Deviation (ppb)
0.082
0.033
0.127
0.071
0.025
0.109
0.064
0.057
0.091
0.050
0.060
0.084
0.056
0.045
0.019
0.004
0.043
0.048
0.110
0.126
0.019
0.064
0.027
0.175
0.038
0.055
0.094
0.068
0.170
0.016
0.098
0.028
0.065
0.046
0.170
0.074
0.201
0.084
0.007
0.077
0.346
0.417
0.192
0.036
0.087
0.197
0.203
0.055
0.122
0.015
CV%
2.3
1.1
4
3
1
4.3
o
5
2.9
4.5
2.7
3.1
4.6
2.6
2
0.9
0.2
2.3
2.7
4.6
5.7
1.6
5.5
2.2
12.4
2.9
7.7
14
9.3
11.9
1.2
7.3
3.8
9.3
6.7
4.3
1.8
4.4
7.8
0.7
7.7
7.6
9
4.4
3.2
7.1
17.1
12.1
2.9
6.1
1.6
72
-------
Table B-2. Abraxis DM ELISA Test Kit Raw Data Continued
Sample Description
RW 6 (2x dil)
RW 6 (2x dil)
RW7
RW7
RW7
RW8
RW8
RW8
RW 9 Matrix
RW 9 Matrix
RW 9 Matrix
Variant
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone, (ppb)
0.934
0.924
0.054
0.057
Range?
0.633
0.715
0.667
0.441
0.489
0.475
Standard Deviation (ppb)
0.015
0.018
0
0
Range?
0.022
0.074
0.008
0
0.013
0.006
CV%
1.6
2
0
0
Range?
3.4
10.4
1.2
0.1
2.6
1.2
73
------- |