Roles and responsibilities in Roles and responsibilities in traceability traceability implementation implementation Federica Braga Federica Braga University of Milan Medical School University of Milan Medical School 12th International 12th International Scientific Scientific Meeting Meeting STANDARDIZATION IN LABORATORY MEDICINE STANDARDIZATION IN LABORATORY MEDICINE AND PATIENT SAFETY AND PATIENT SAFETY
55
Embed
Roles and responsibilities in traceability implementation · Mario Plebani, Laura Sciacovelli, Daniela Bernardi, Ada Aita, Giorgia Antonelli, Andrea Padoan The role of external quality
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Roles and responsibilities in Roles and responsibilities in
traceabilitytraceability
implementationimplementation
Federica BragaFederica Braga
University of Milan Medical School University of Milan Medical School
12th International 12th International ScientificScientific Meeting Meeting
STANDARDIZATION IN LABORATORY MEDICINE STANDARDIZATION IN LABORATORY MEDICINE
•• Assays that claim to measure the same Assays that claim to measure the same analyteanalyte
should give equivalent measurement results (for should give equivalent measurement results (for
long term and within clinically meaningful limits)long term and within clinically meaningful limits)
•• Measurement results should be independent of:Measurement results should be independent of:
−− TimeTime
−− Location/laboratoryLocation/laboratory
−− Assay systemAssay system
Laboratory results should be Laboratory results should be
equivalent no matter where they equivalent no matter where they
are performedare performed
ISO/EN 17511 ISO/EN 17511 -- MeasurementMeasurement of of quantitiesquantities in in samplessamples of of biologicalbiological originorigin -- MetrologicalMetrological traceabilitytraceability of of
valuesvalues assignedassigned toto calibratorscalibrators and control and control materialsmaterials..
Objective of traceability Objective of traceability
implementation:implementation:
to enable the results obtained by the to enable the results obtained by the
calibrated routine procedure to be calibrated routine procedure to be
expressed in terms of the values obtained at expressed in terms of the values obtained at
the highest available level of the calibration the highest available level of the calibration
hierarchyhierarchy..
EU 98/79/ECEU 98/79/EC--IVD IVD DirectiveDirective
To become To become equivalent for long termequivalent for long term, ,
results must be traceable to higherresults must be traceable to higher--
order referencesorder references
requires manufacturers to ensure requires manufacturers to ensure
traceability of their analytical systems to traceability of their analytical systems to
recognized higher order referencesrecognized higher order references
REGULATION (EU) 2017/746REGULATION (EU) 2017/746 OF THE EUROPEAN PARLIAMENT OF THE EUROPEAN PARLIAMENT
AND OF THE COUNCIL of 5 April 2017 on AND OF THE COUNCIL of 5 April 2017 on in vitro in vitro diagnostic diagnostic
medical devices and medical devices and repealing Directive 98/79/ECrepealing Directive 98/79/EC and and
calibratorscalibrators, , controlscontrols) ) fulfillingfulfilling the the
aboveabove establishedestablished goalsgoals
PanteghiniPanteghini M,M,
ClinClin ChemChem Lab Lab MedMed
2010;48:72010;48:7
Roles and responsibilities in traceability implementationRoles and responsibilities in traceability implementation
3. End users (clinical laboratories):3. End users (clinical laboratories):
Survey assay and laboratory performance through:Survey assay and laboratory performance through:
-- IQC component I: testing system controls to confirm and verify IQC component I: testing system controls to confirm and verify manufacturermanufacturer’’s declared performance s declared performance
-- IQC component IIIQC component II: : estimation of the MEASUREMENT UNCERTAINTY due to the random effeestimation of the MEASUREMENT UNCERTAINTY due to the random effects and cts and
calculation of uncertainty of laboratory measurements (calculation of uncertainty of laboratory measurements (UUcalcal + + UUrandomrandom))
-- EQA: true value in commutable materials for defining measuremenEQA: true value in commutable materials for defining measurement error of laboratoryt error of laboratory
calibratorscalibrators, , controlscontrols) ) fulfillingfulfilling the the
aboveabove establishedestablished goalsgoals
PanteghiniPanteghini M,M,
ClinClin ChemChem Lab Lab MedMed
2010;48:72010;48:7
Roles and responsibilities in traceability implementationRoles and responsibilities in traceability implementation
3. End users (clinical laboratories):3. End users (clinical laboratories):
Survey assay and laboratory performance through:Survey assay and laboratory performance through:
-- IQC component I: testing system controls to confirm and verify IQC component I: testing system controls to confirm and verify manufacturermanufacturer’’s declared performance s declared performance
-- IQC component IIIQC component II: : estimation of the MEASUREMENT UNCERTAINTY due to the random effeestimation of the MEASUREMENT UNCERTAINTY due to the random effects and cts and
calculation of uncertainty of laboratory measurements (calculation of uncertainty of laboratory measurements (UUcalcal + + UUrandomrandom))
-- EQA: true value in commutable materials for defining measuremenEQA: true value in commutable materials for defining measurement error of laboratoryt error of laboratory
a) IVD industry (a) IVD industry (to assist them in following the EU Directive on compliance and to assist them in following the EU Directive on compliance and
traceability of commercial systemstraceability of commercial systems))
b) Regulators (to verify that results produced by b) Regulators (to verify that results produced by IVDsIVDs are traceable to)are traceable to)
REFERENCE MEASUREMENT SYSTEMREFERENCE MEASUREMENT SYSTEMREFERENCE MEASUREMENT SYSTEM
ANALYTICAL PERFORMANCE
SPECIFICATIONS
ANALYTICAL PERFORMANCE ANALYTICAL PERFORMANCE
SPECIFICATIONSSPECIFICATIONS
Some models are better suited for certain Some models are better suited for certain
measurandsmeasurands than for others. The attention is than for others. The attention is
therefore primarily directed towards the therefore primarily directed towards the
measurandmeasurand and its biological and clinical and its biological and clinical
characteristics.characteristics.
•• Model 1: Based on the effect of analytical Model 1: Based on the effect of analytical
performance on clinical outcomeperformance on clinical outcome
•• Model 2: Based on components of biological Model 2: Based on components of biological
variation of the variation of the measurandmeasurand
•• Model 3: Based on state of the art of the Model 3: Based on state of the art of the
measurement (i.e., the highest level of measurement (i.e., the highest level of
homeostatichomeostatic controlcontrolThe The measurandmeasurand has a central has a central
role in diagnosis and role in diagnosis and
monitoring of a specific monitoring of a specific
disease disease
a. Done by direct outcome studies – investigating the impact of analytical performance
of the test on clinical outcome.
b. Done by indirect outcome studies – investigating the impact of analytical
performance of the test on clinical classifications or decisions and thereby on the
probability of patient outcome, e.g., by simulation analysis.
a.a. Done by direct outcome studies Done by direct outcome studies –– investigating the impact of analytical performance investigating the impact of analytical performance
of the test on clinical outcome.of the test on clinical outcome.
b. Done by indirect outcome studies b. Done by indirect outcome studies –– investigating the impact of analytical investigating the impact of analytical
performance of the test on clinical classifications or decisionsperformance of the test on clinical classifications or decisions and thereby on the and thereby on the
probability of patient outcome, e.g., by simulation analysis.probability of patient outcome, e.g., by simulation analysis.
calibratorscalibrators, , controlscontrols) ) fulfillingfulfilling the the
aboveabove establishedestablished goalsgoals
PanteghiniPanteghini M,M,
ClinClin ChemChem Lab Lab MedMed
2010;48:72010;48:7
Roles and responsibilities in traceability implementationRoles and responsibilities in traceability implementation
3. End users (clinical laboratories):3. End users (clinical laboratories):
Survey assay and laboratory performance through:Survey assay and laboratory performance through:
-- IQC component I: testing system controls to confirm and verify IQC component I: testing system controls to confirm and verify manufacturermanufacturer’’s declared performance s declared performance
-- IQC component IIIQC component II: : estimation of the MEASUREMENT UNCERTAINTY due to the random effeestimation of the MEASUREMENT UNCERTAINTY due to the random effects and cts and
calculation of uncertainty of laboratory measurements (calculation of uncertainty of laboratory measurements (UUcalcal + + UUrandomrandom))
-- EQA: true value in commutable materials for defining measuremenEQA: true value in commutable materials for defining measurement error of laboratoryt error of laboratory
3. ESTIMATION OF MEASUREMENT 3. ESTIMATION OF MEASUREMENT
UNCERTAINTY OF CALIBRATORSUNCERTAINTY OF CALIBRATORS
4. VERIFICATION THAT UNCERTAINTY 4. VERIFICATION THAT UNCERTAINTY
GOALS ARE FULLFILLED GOALS ARE FULLFILLED
GUMGUM
Eurachem/
CITAC Guide
Eurachem/
CITAC Guide
CLSI document EP29-ACLSI document EP29-A
PARADIGM SHIFT IN THE THINKINGPARADIGM SHIFT IN THE THINKING
•• IfIf the the manufacturermanufacturer assumesassumes total total responsibilityresponsibility forfor supplyingsupplying productsproducts of of acceptableacceptable qualityquality in in termsterms of of traceabilitytraceability and and uncertaintyuncertainty of the system (of the system (““CE CE markedmarked””), ), itit isisno no longerlonger possiblepossible toto considerconsider separatelyseparately the the componentscomponents of of eacheach measuringmeasuring system (i.e., system (i.e., platformplatform, , reagentsreagents, , calibratorscalibrators and control and control materialsmaterials), ), whichwhich in in termsterms of performance can of performance can onlyonly bebe guaranteedguaranteed and and certifiedcertified byby the the manufacturermanufacturer asas a a wholewhole..
•• AnyAny changechange introducedintroduced byby usersusers or or thirdthirdpartiesparties (e.g., the (e.g., the useuse of of reagentsreagents, , calibratorscalibrators or or control control materialsmaterials fromfrom otherother supplierssuppliers) ) maymaysignificantlysignificantly alter the alter the qualityquality of the of the measuringmeasuringsystem performance, system performance, removingremoving anyanyresponsibilityresponsibility fromfrom the the manufacturermanufacturer and and deprivingdepriving the system (and, the system (and, consequentlyconsequently, the , the producedproduced resultsresults) of the ) of the certificationcertification originallyoriginallyprovidedprovided through CE through CE markingmarking..
calibratorscalibrators, , controlscontrols) ) fulfillingfulfilling the the
aboveabove establishedestablished goalsgoals
PanteghiniPanteghini M,M,
ClinClin ChemChem Lab Lab MedMed
2010;48:72010;48:7
Roles and responsibilities in traceability implementationRoles and responsibilities in traceability implementation
3. End users (clinical laboratories):3. End users (clinical laboratories):
Survey assay and laboratory performance through:Survey assay and laboratory performance through:
-- IQC component I: testing system controls to confirm and verify IQC component I: testing system controls to confirm and verify manufacturermanufacturer’’s declared performance s declared performance
-- IQC component IIIQC component II: : estimation of the MEASUREMENT UNCERTAINTY due to the random effeestimation of the MEASUREMENT UNCERTAINTY due to the random effects and cts and
calculation of uncertainty of laboratory measurements (calculation of uncertainty of laboratory measurements (UUcalcal + + UUrandomrandom))
-- EQA: true value in commutable materials for defining measuremenEQA: true value in commutable materials for defining measurement error of laboratoryt error of laboratory
Types of metrological chains that can be used to implement the tTypes of metrological chains that can be used to implement the traceability of blood glucose results*raceability of blood glucose results*
Braga F & Panteghini M, Clin Chim Acta 2014;432:55Braga F & Panteghini M, Clin Chim Acta 2014;432:55
*all*all JCTLM JCTLM recognizedrecognized
Chain A = 1.45% vs. Chain C = 3.26%
Are the Are the measuringmeasuring systemssystems commerciallycommercially availableavailable forfor glucoseglucose determinationdetermination ableable
toto achieveachieve the the desirabledesirable limitlimit forfor combinedcombined uncertaintyuncertainty in a in a clinicalclinical settingsetting??
uurefref
(u(u22refref ++ uu22
calcal))½½
((uu22ref ref + + uu22
calcal + + uu22randomrandom))½½
System imprecision
System calibration
uncertainty
Individual lab
performance
(IQC safety margin)
Measurement
uncertainty
budget
Uncertainty of
references
Measurand definition
Patient result
5.4%5.4% desirabledesirable
Braga F & Panteghini M, Clin Chim Acta 2014;432:55Braga F & Panteghini M, Clin Chim Acta 2014;432:55
Pasqualetti S, Braga F, Panteghini M, Pasqualetti S, Braga F, Panteghini M, Clin Biochem 2017; 50:587Clin Biochem 2017; 50:587--594.594.
L M H
•• Measured in triplicate in 3 Measured in triplicate in 3
consecutive daysconsecutive days byby participatingparticipating
laboratorieslaboratories
59
.9 U
/L5
9.9
U/L
18
6.9
U/L
18
6.9
U/L
40
1.5
U/L
40
1.5
U/L
•• ALP target values ALP target values
assigned with assigned with IFCC RMPIFCC RMP
The importance of the postThe importance of the post--marketing surveillancemarketing surveillance
EQA experimentEQA experiment
(performed 5(performed 5 years later years later the availability of the IFCC RMPthe availability of the IFCC RMP) )
to evaluate the level of ALP standardizationto evaluate the level of ALP standardization
AdviaAdvia pp--NPP/AMPNPP/AMP Chemistry calibrator control 1Chemistry calibrator control 1 3.70%3.70% IFCC reference method (2011)IFCC reference method (2011)
Chemistry calibrator control 2Chemistry calibrator control 2 1.00%1.00% IFCC reference method (2011)IFCC reference method (2011)
pp--NPP/DEANPP/DEA Chemistry calibrator control 1Chemistry calibrator control 1 1.40%1.40%cc Molar extinction coefficientMolar extinction coefficient
Chemistry calibrator control 2Chemistry calibrator control 2 1.30%1.30%cc Molar extinction coefficientMolar extinction coefficient
• The availability of an internationally agreed reference
measurement system does not automatically mean that
the traceability to it is implemented!
• If a post-marketing surveillance is lacking, alternatives
that do not comply with the EU Directive can remain
undisturbed on the market.
• The availability of an internationally agreed reference The availability of an internationally agreed reference
measurement system does not automatically mean that measurement system does not automatically mean that
the traceability to it is implemented!the traceability to it is implemented!
•• If a postIf a post--marketing surveillance is lacking, alternatives marketing surveillance is lacking, alternatives
that do not comply with the EU Directive can remain that do not comply with the EU Directive can remain
undisturbed on the market.undisturbed on the market.
Roche recalibrated Roche recalibrated
its analytical its analytical
methods for the methods for the
ALP measurement ALP measurement
according to 2011 according to 2011
IFCC RMP!IFCC RMP!
Trueness evaluation and verification of interTrueness evaluation and verification of inter--assay agreement of 11 serum assay agreement of 11 serum
immunoglobulin A measuring systems: implications for medical decimmunoglobulin A measuring systems: implications for medical decisionsisions
Braga F, Infusino I, Frusciante E, Ceriotti F, Panteghini M [under review]
Evaluating the status of standardization of available Evaluating the status of standardization of available IgAIgA measuring systems to measuring systems to
investigate if a common lower reference limit can be implementedinvestigate if a common lower reference limit can be implemented to identify to identify
an an IgAIgA deficiencydeficiency, , especially for pediatric populationespecially for pediatric population
[stored at -80°C until the distribution to centres]
AimAim
Laboratories were requested to perform analysis within two days Laboratories were requested to perform analysis within two days of delivery, in a single of delivery, in a single
analytical run, by applying the manufactureranalytical run, by applying the manufacturer’’s recommended instructions and strictly s recommended instructions and strictly
following a specific analysis sequencefollowing a specific analysis sequence
ResultsResults
• Commutability assessment of ERM-DA470k/IFCC
1. CLSI 1. CLSI guidelineguideline EP30EP30--AA
[based on the pair-wise comparison of sets of IgA results on patient pools by the 11 MS]
DemingDeming regressionregression
95% 95% predictionprediction
intervalsintervals
2. IFCC 2. IFCC recommendationsrecommendations
ReferenceReference
materialmaterial
Results confirmed the acceptable commutability of Results confirmed the acceptable commutability of
ERMERM--DA470k/IFCC for all MS we evaluated, DA470k/IFCC for all MS we evaluated,
suggesting that this material was appropriate for suggesting that this material was appropriate for
estimating the trueness of the procedures.estimating the trueness of the procedures.
[CLSI. Document EP30-A Wayne (PA): CLSI; 2010]
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0 1 2 3 4 5
Mean conc both (x+y)/2 (g/L)
Bia
s ln
(AU
48
0)-
ln(A
rch
ite
ct
c4
00
0)_
ReferenceReference materialmaterial
Minimum Minimum
quality level quality level
for bias from for bias from
BVBV
[Nilsson G, et al Clin Chem 2018;64:455–64]
ResultsResults
• Trueness evaluation of IgA MS
Significantly
overestimated
Significantly
underestimated
ExcludedExcluded
fromfrom the the
followingfollowing
data data
elaborationelaboration
ResultsResults
• Verification of agreement among
IgA MS
PassingPassing--BablokBablok regression analysis and average percentage bias of measured regression analysis and average percentage bias of measured
IgAIgA values for patient pools vs. median values for patient pools vs. median IgAIgA concentrations for MS. concentrations for MS.
In In generalgeneral, the , the situationsituation clearlyclearly worsenedworsened whenwhen the the
analysisanalysis focusedfocused on the on the lowlow IgAIgA concentrationsconcentrations
typicaltypical of of childrenchildren
ResultsResults
• Measurement uncertainty of IgA MS on patient pools
For suitable clinical application of For suitable clinical application of IgAIgA measurements, the expanded measurement measurements, the expanded measurement
uncertainty at the patient sample level should remain within uncertainty at the patient sample level should remain within ±±2.7%, 2.7%, ±±5.4%, and 5.4%, and ±±8.1%8.1%
for the optimum, desirable, and minimum quality levels, respectifor the optimum, desirable, and minimum quality levels, respectivelyvely..
ConclusionConclusion
Although the harmonization among marketed Although the harmonization among marketed IgAIgA MS MS
is substantially good, is substantially good, the implementation of the implementation of
traceability to higher order references is still far from traceability to higher order references is still far from
perfectperfect. Most of the MS evaluated in this study . Most of the MS evaluated in this study
showed unacceptable standardization of showed unacceptable standardization of IgAIgA
measurements, especially at concentrations measurements, especially at concentrations ≤≤0.4 g/L. 0.4 g/L.
The definition of traceable reference intervals in The definition of traceable reference intervals in
pediatric population should consider this analytical pediatric population should consider this analytical
background.background.
Daily surveillance of IVD system traceability Daily surveillance of IVD system traceability
1.1. Verification of the consistency of declared performance Verification of the consistency of declared performance
during routine operations performed in accordance with the during routine operations performed in accordance with the
AnalyticalAnalytical QualityQuality Control in the Traceability EraControl in the Traceability Era
PanteghiniPanteghini M, M, ClinClin ChemChem Lab Lab MedMed 2010;48:72010;48:7
Braga F & Panteghini M, Clin Chim Acta 2014;432:55Braga F & Panteghini M, Clin Chim Acta 2014;432:55
10-M
ay
25-M
ay
28-J
un
13-Jul
28-J
ul
11-A
ug
27-Aug
11-Sep
26-Sep
13-O
ct
24
+3SD
+2SD
+1SD
19
29
34
Monitoring the reliability of the measuring system through IQC: Monitoring the reliability of the measuring system through IQC:
Component I. Check alignment (Component I. Check alignment (““system traceabilitysystem traceability””))A
cce
pta
ble
Acc
ep
tab
lera
ng
era
ng
eo
f co
ntr
ol
ma
teri
al
of
con
tro
l m
ate
ria
l
This program checks whether in the course of an analytical run tThis program checks whether in the course of an analytical run the performance of a he performance of a
measuring system complies with the set goals, represented by themeasuring system complies with the set goals, represented by the acceptable acceptable
ranges of control materials. ranges of control materials.
Clinical laboratories must verify the consistency of declared peClinical laboratories must verify the consistency of declared performance during routine operations rformance during routine operations
performed in accordance with the manufacturerperformed in accordance with the manufacturer’’s instructions, by s instructions, by checking that values of control checking that values of control
materials provided by the manufacturer as component of the measumaterials provided by the manufacturer as component of the measuring system are in the ring system are in the
established control rangeestablished control range, with no clinically significant changes in the assumed traceabl, with no clinically significant changes in the assumed traceable results.e results.
PlatformPlatform
CalibratorsCalibratorsReagentsReagents
Control material(s)Control material(s)
PlatformPlatform
CalibratorsCalibratorsReagentsReagents
PlatformPlatform
CalibratorsCalibratorsReagentsReagents
Control material(s)Control material(s)
Acceptance/rejection of
the analytical run in
“real time”
Acceptance/rejection of Acceptance/rejection of
the analytical run in the analytical run in
““real timereal time””
Internal Quality Control
(Component I)
Internal Quality Control Internal Quality Control
(Component I)(Component I)
Testing alignmentTesting alignment
[[““system traceabilitysystem traceability””]]
Any “out of control” signal must be made available with
sufficient time to allow immediate corrective actions
to bring again the situation under control (virtually
“unbiased”) and before reports related to the samples
analyzed in the affected analytical run are issued.
Any Any ““out of controlout of control”” signal must be made available with signal must be made available with
sufficient time to allow immediate corrective actions sufficient time to allow immediate corrective actions
to bring again the situation under control (virtually to bring again the situation under control (virtually
““unbiasedunbiased””) and before reports related to the samples ) and before reports related to the samples
analyzed in the affected analytical run are issued.analyzed in the affected analytical run are issued.
Daily surveillance of IVD system traceability Daily surveillance of IVD system traceability
1.1. Verification of the consistency of declared performance Verification of the consistency of declared performance
during routine operations performed in accordance with the during routine operations performed in accordance with the
To verify the suitability of laboratory To verify the suitability of laboratory
measurements in clinical settingmeasurements in clinical setting
PanteghiniPanteghini M, M, ClinClin ChemChem Lab Lab MedMed 2010;48:7 2010;48:7
InfusinoInfusino I I etet al., al., ClinClin ChemChem Lab Lab MedMed 2010;48:3012010;48:301
Braga F & Braga F & PanteghiniPanteghini M. M. ClinClin ChemChem Lab Lab MedMed 2013;51:17192013;51:1719
Braga F & Panteghini M, Clin Chim Acta 2014;432:55Braga F & Panteghini M, Clin Chim Acta 2014;432:55
[Ferraro S, Braga F, Panteghini M. Clin Chem Lab Med 2016;54:523]
Analytical quality
of measurement
Check alignment Imprecision
qualify
Reliability of the analytical system
AnalyticalAnalytical QualityQuality Control in the Traceability EraControl in the Traceability Era
PanteghiniPanteghini M, M, ClinClin ChemChem Lab Lab MedMed 2010;48:72010;48:7
Braga F & Panteghini M, Clin Chim Acta 2014;432:55Braga F & Panteghini M, Clin Chim Acta 2014;432:55
Internal Quality Control
(Component II)
Internal Quality Control Internal Quality Control
(Component II)(Component II)
Testing the uncertainty Testing the uncertainty
due to the random effectsdue to the random effects
((““imprecisionimprecision””))
System stability at
medium/long term
System stability at System stability at
medium/long termmedium/long term
This program provides, through
mechanisms of retrospective
evaluation, data useful to the
knowledge of variability of
measuring system and of its use by
the individual laboratory.
This program provides, through This program provides, through
mechanisms of retrospective mechanisms of retrospective
evaluation, data useful to the evaluation, data useful to the
knowledge of variability of knowledge of variability of
measuring system and of its use by measuring system and of its use by
the individual laboratory.the individual laboratory.
System
Reagent lots
Laboratory
Monitoring the reliability of the measuring system through IQC: Monitoring the reliability of the measuring system through IQC:
Component II. Evaluate the system + individual lab imprecisionComponent II. Evaluate the system + individual lab imprecision
System imprecision
System calibration
(combined) uncertainty
Individual lab
performance
(IQC safety margin)
Measurement
uncertainty
budget
CardiacCardiac troponintroponin T high sensitiveT high sensitiveMonthlyMonthly monitoringmonitoring of of imprecisionimprecision byby IQC materialIQC material
0
1
2
3
4
5
6
7
8
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Imp
recis
ion
, C
V%
Requirements for IQC material
(Component II)
RequirementsRequirements forfor IQC material IQC material
((ComponentComponent II)II)
RequirementRequirement CommentComment
MatrixedMatrixed material from a thirdmaterial from a third--
party independent source party independent source
should be used (e.g., freshshould be used (e.g., fresh--
frozen pool)frozen pool)
Material must be different from the Material must be different from the
system control material used for system control material used for