Document Em25Z151mBzq2MLNYpgvN3OYb
000075
>
oo
- r ti ~ e v \fW
METHOD VALIDATION REPORT
TITLE:
DATE: REPORT:
M ethod Validation for the Q uantitation o f Ammonium Perfluorooctanoate (APPO), Measured as the Perfluorooctanoate Anion (PFQA, 0.05 to 10 pg/mL C oncentration Range), in Human Serum by Turbo Ion Spray LC/MS/M S
12 July 2002
I
02120VDJAJDU.DOC
CBR.TIFIBD C O PY O F O IU Q W A l Date
JM L
HTw
AUTHORS:
PREPARED FOR:
David J. Anderson, M .S. Kimberly L. Norwood, B.A. Kathy Jo Palm er, B.S. Jennie L. Romney, M .P.S,
Advion BioSciences, Inc, 15 Catherwood Road Ithaca, NY 14850
DuPont Central Research and Development ^Experimental ation
NUMBER OF
PAGES:
54
COfftpsny 6jn)t]Z 8d. D oos n o tcontain T8CACBI
000076
AUTHORS: TITLE:
David J. Anderson, M.S. Kimberly L. Norwood, B.A. Kathy Jo Palmer, B.S. Jennie L. Romney, M.P.S.
Method Validation for the Quantitation of Ammonium Perfluorooctanoate (APFO), Measured as the PerfluorDOctanoate Anion (PFOA, 0.05 to 10 pg/mL Concentration Range), in Hitman Serum by Turbo Ion Spray LC/MS/MS
ABSTRACT
A sensitive, selective, accurate, and reproducible analytical method was developed by Advion BioSciences, Inc., Ithaca, New York to quantitate ammombm perfluorooctanoate (APFO), measured as the perfluorooctanoate anion (PFOA), in human serum samples, PFOA and its internal standard (IS) were isolated from human serum samples (25 pL) by a liquid-liquid extraction procedure. An aliquot of die organic layer from each sample was evaporated to dryness, reconstituted and analyzed by turbo ion spray liquid chromatography/tandem mass spectrometry (LC/MS/MS) in the negative ion mode. The assay demonstrated a lower limit of quantitation o f 0.05 pg/mL for PFOA. The calibration curve range was quadratic from 0.05 pg/mLto 10 pg/mL for PFOA. The coefficients o f determination (r2) of the calibration curves were >0.9950 for PFOA.
Precision and accuracy quality control (QC) samples were prepared at concentrations of 0.15, 3, and 8 pg/mL PFOA. The intra- and inter-assay precision (CV) results calculated from QC samples ranged from 2.07 to 7.70% for PFOA. The intra- and inter-assay accuracy (RE) calculated from QC samples ranged from -12.3 to 9.00% for PFOA.
The results for PFOA were reproducible upon reinjection of prepared samples after storage for up to 65 hours at room temperature. The extraction recovery of PFOA and the IS from human serum were 81.8 and 90.9%, respectively. The matrix effect
Page 2 o f54
02I20VDJA_DU.DOC
000077
on the ionization of PFOA and the IS was an enhancement o f 1.62 and 8.07%, respectively.
I
Page 3 o f54
02I20VDJAJJU.DOC
00007S
SIGNATURE PAGE
TITLE:
Method Validation for the Quantitation o f Ammonium Periluorooctanoate (APFO), Measured as the
Perfluorooctanoate Anion (PFOA, 0.Q5 to 10 pg/mL Concentration Range), in Human Serum by Turbo Ion Spray LC/MS/MS
Report Number:
02120VDJA DU.DOC
Reported by:
JM dL m , Date
Authorized for Release by:
Director o fBioanalytical Science
Sponsor Review:
T Ia m ' t 'atm w j Mary(A. Kaiser, PLD.
DuPont Corporate Center for
Analytical Sciences
b H/fy Elate '
Page4 of 54
0212QVDJA_DU.DOC
000079
QAU STATEMENT
Periodic audits of the method validation for the quantitation of APFO, measured as PFOA, in human serum were conducted by the Quality Assurance Unit o f Advion BioSciences, Inc. (Advion).
The study was inspected on the following dates:
Phase o f Study Inspected
Date o fInspection
Data Review Report
6, 7,11 June 2002 20 June 2002
Date Reported to Management
6, 7,11 June 2002 20 June 20Q2
Based on the inspections and the data reviewed, this report is a complete and accurate representation o f the data.
IS
Georgslc. Amedto, B.S. Quality Assurance Auditor
/ 4 t-tQ> o Dale
Page 5 o f 54
02120VDJA D U.DOC
ooooao
TABLE OF CONTENTS
ABSTRACT............ ...............................................
-.2
SIGNATURE PAGE........................
4
QAU STATEMENT....................... - ...........,................................ ........ .... ............. .......... 5
TABLE OF CONTENTS.... .................................. .............................................................. 6
LIST OF TABLES.......................................................... ............................... .......... ........... 8
LIST OF FIGURES.................... ...................................................................... ....... .......... 9
1. INTRODUCTION.......................................................................... !................ ......... 10
2. EXPERIMENTAL............ .....................
10
2.1. Chemicals and Materials.................................................................. .............10
2.2. LC/MS/MS IWSmUMENTATION..................
10
2.3. Sample Preparation and Extraction procedure....... ............. ................ 11
2.4. Human serum Validation d a ta ................................................. ....................n
2.5. assay EVALUATION..............................................................
12
2.5.1. Intra- and Mer-Assay Precision
......... ...................... ..........................12
2.5.2. Intra-and Inter-Assay Accuracy.... ............................. ............ ......................12
2.5.3. Sensitivity: Lower Limit of Quantitation (LLQ)...................... ....................13
2.5.4. Selectivity............................... -................................ ................................... 13
2.5.5. Chromatographic Carryover.,............................................ ........................... 13
2.5.6. Accuracy and Precision of Dilution........... .................. ................................ 14
2.6. stability of PFOA in human Serum............................................................... 14
2.7. Reproducibility of reinjecting prepared sa m ples....... ...........................14
2.8. Extraction Recovery.....................................
14
2.9. m atrix Effect on Ionization of PFOA and 9H-HDFNA in Human Serum..... .............
is
3. RESULTS AND DISCUSSION..... .......................................... .......................... .... 16
3.1. ASSAY EVALUATIONRESULTS..........................................................................17 3.L1. Intra- and Inter-Assay Precision......... ................................. ......................... 17 3.1.2. Intra- and Liter-Assay Accuracy........................................................... .........17
Page 6 o f54
02120VDJA_DU.DOC
000081
3.1.3. Calibration Curve Fit.......................................................... ........................... 17
3.1.4. Sensitivity: Lower Limit of Quantitation (LLQ) ......... .............................18
3.1.5. Selectivity.......... ...............................................................
.18
3.1.6. Chromatographic Carryover........................................
18
3.1.7. Accuracy and Precision of Dilution. ......................................................... 18
3.2. REPRODUCIBILITYOF REINJECTING EXTRACTEDSAMPLES ...... ................. . 19
3.3. Extraction Recovery........................
...19
3.4. Matrix Effect on Ionization of p f o a and 9H-HDFNA in Human Serum....... .......................
19
4. SOP DEVIATION.................................................
19
5. DATA RETRIEVAL....... ........................................ ................ ..................................20 I
6. CONCLUSIONS...................... .......................................... .......................................20
7. REFERENCE................................. ........................................................................... 20
8. TABLES....... .... .............
21
9. FIGURES..................................
,.33
10. APPENDIX A: QUANTITATION OF AMMONIUM PERFLUOROOCTANOATE (APFO), MEASURED AS THE
.V * PERFLUOROOCTANOATE ANION (PFOA, 0.05 TO 10 pg/mL ) CONCENTRATION RANGE), IN HUMAN SERUM BY TURBO ION SPRAY U C M S /m ......................................................................................... 41
Page 7 of 54
32120VDJAJDU.DOC
000032
LIST OF TABLES
Table l: Validation Run Descriptions....... ......................................... .............................21
Table 2: Accuracy and Precision of the PFOA Assay in Human Serum for Quality Control Samples from Three Validation Runs........ ..........................................22
Table 3: Accuracy and Precision o f the PFOA Assay in Human Serum for Calibration Standards from Three Validation Runs......... ................................ 23
Table 4: Calibration Curve Parameters for PFOA in Human Serum from Three Validation Runs......... ,............................... .......,............................................. 24
Table 5: Lower Limit of Quantitation of PFOA in Human Serum..................................25 Table 6: Lower Limit o f Quantitation o f PFOA and Selectivity in Sixj Separate Lots
o f Human Serum........................... ..................................... ................... ........... 26 Table 7: Accuracy and Precision of the PFOA Assay in Human Serum Dilution QC
Samples from One Validation Run,....................................................... .......... 27 Table 8: Reinjection Reproducibility of QC Samples Containing PFOA After
Storage in Reconstitution Solution for 65 Hours at Room Temperature.,...... 28 Table 9: Extraction Recovery of PFOA from Human Seram ...................................... 29 Table 10: Extraction Recovery of the Internal Standard, 9H-HDFNA, from Human
Seram..... ...................... ..................................................... ............................. 30 Table 11: Matrix Effect on Ionization of PFOA in Human Serum............... ................. ,,31 Table 12: Matrix Effect on Ionization of 9H-HDFN in Human Serum ...................... ,,32
Page 8 of 54
02120VDJA_DU.DOC
00083
LIST OF FIGURES
Figure 1: Full-Scan Turbo Ion Spray Mass Spectrum of PFOA........ ..............................33 Figure 2: Full-Scan Product Ion Mass Spectrum of PFOA........ ......................................34 Figure 3: Full-Scan Turbo Ion Spray Mass Spectrum of 9H-HDFNA......... ....................35 Figure 4: Full-Scan Product Ion Mass Spectrum of 9H-HDFNA....... .............................36 Figure 5: Selected Reaction Monitoring Chromatograms o f PFOA and the Internal
Standard in Control Blank Human Serum Extract...........................................,37 Figure 6: Selected Reaction Monitoring Chromatograms o f PFOA and the Internal
Standard in a Human Serum Extract Containing Internal Standard Only (Zero Sample)........... .......................................... ,................p.............. ...........38 Figure 7: Selected Reaction Monitoring Chromatograms o f PFOA and the Internal Standard from a Standard l Human Serum Extract (0.05 pg/mL)..... ........... ,,39 Figure 8; Selected Reaction Monitoring Chromatograms o f PFOA and the Internal Standard from a Standard 9 Human Serum Extract (10 pg/mL)......................40
i
Page 9 of 54
02I20VDJAJDU.DOC
000084
1. INTRODUCTION
The purpose of this report is to describe the method validation for the quantitative determination o f ammonium perfluorooetanoate (APFO), measured as the perfluorooctanoate anion (PFOA), in human serum samples. The objectives of the method validation were to validate a sample preparation procedure, determine the lower limit o f quantitation (LLQ) for routine analysis, demonstrate accuracy of dilution, determine recovery and matrix effect, and to provide selective, accurate, and reproducible quantitative results by turbo ion spray liquid chromatography/tandem mass spectrometry (LC/MS/MS). This work was conducted in accordance with Advion BioSciences, Tnc. (Advion) Standard Operating Procedures (SpP).
2. EXPERIMENTAL
2 .1 . CHEM ICALS AND MATERIALS
Pentadecafluorooctanoic acid ammonium salt (Lot# 421207/1) was obtained from Sigma-Aldrich, St. Louis, MO. The internal standard (IS), 9H-hexadecaflu0rononanoic acid (9H-HDFNA, No Lot# available), was obtained from Sigma-Aldrich, St. Louis, MO. A detailed list o f chemicals and materials is found in Appendix A.
Stock and working solutions used to prepare the calibration curves and quality control (QC) samples for PFOA were prepared as described in Appendix A.
2 .2 . LC/MS/MS INSTRUMENTATION
The liquid chromatography/mass spectrometry system consisted of two LC-10AD vp pumps (Shimadzu, Columbia, MD), an SCL-10A vp pump controller (Shimadzu, Columbia, MD), a Perkin-Elmer Series 200 autosampler (Perkin-Elmer Corp., Norwalk, CT), a Betasil Cta column (2 x 50 mm, 5 pm particle sizes, ThermoHypersil-Keystone, Bellefonte, PA), a model 7970 column heater (Jones Chromatography USA Inc., Lakewood, CO), and a SCIEX API 3000 mass
Page 10 o f54
02120VDJA_DU.DOC
0000S5
spectrometer (Applied BioSystems, Concord, Ontario). A detailed list o f the instrumentation and instrument conditions is found in Appendix A.
2.3. Sam ple Preparation and E xtraction Procedure
For each analytical run, duplicate 25-pL aliquots of tire calibration curve samples were prepared as described in Appendix A, The calibration curve concentrations were 0.05,0.1, 0.25,0.5, 1,2,5,7.5, and 10 pg/mL for PFOA.
Human serum QC samples were prepared in advance o f the validation study at concentrations of 0.15,3, and 8 pg/mL PFOA for QC1, QC2, and QC3, respectively, as detailed in Appendix A. QC4 (50-pg/mL dilution QC) was prepared at a concentration exceeding the calibration curve range, and was assayed using a 10-fold dilution.
2 .4 . HUM AN SERUM VALIDATION DATA
t
The data were collected using selected reaction monitoring (SRM) turbo ion spray
LC/MS/MS in the negative ion mode. Peak areas were integrated by the SCIEX
program MacQuan, version 1.6, residing on a Macintosh computer. Background
subtraction was perfonmed using Excel 98 to correct for the presence of PFOA in the
control ston lots as detailed in Appendix A (A. 14.0). Following peak area
integration and background subtraction, the results tables from MacQuan were saved
as text files and uploaded to the Advion BioSciences, Inc. (Advion) file server where
a weighted ( 1/x2) quadratic regression was performed using the software package
Watson v 6.1.0.03 (InnaPhase, Inc., Philadelphia, PA). All concentration data were
downloaded from Watson to Excel 97 with three decimal places for this report, with
the exception of QC LLQ and Std 1 replicates, which were downloaded with four
decimal places. Means, CV values, and RE values were calculated in Excel prior to
rounding. Concentration data, concentration means, CV values, and RE values were
rounded to three significant figures in Excel prior to reporting. The data for the
human serum validation are stored in Advion Notebook 02120.
Page It o f $4
02120VDJA_DU.DOC
000086
All calculations were based on the peak area ratio of PFOA to the internal standard. Concentrations of PFOA in QC samples were determined by inverse prediction from the calibration curve.
2.5. a ssay evaluation
2.5.1. Intra- and Inter-Assay Precision
The intra- and inter-assay precision of the method was assessed by determining the
coefficient of variation (CV) observed in the analysis of QC samples. The CV was i
determined by dividing the standard deviation by the mean concentration for each
QC level and expressing the result as a percentage. The intra-assay precision was
expressed as the CV o f the replicate analyses o f a QC level during a single validation
run. The inter-assay precision was expressed as the CV o f the overall mean o f the
individual replicate concentrations of the QC samples for three validation runs.
QC1, QC2, and QC3 were assayed in replicates of six. In addition, the inter-assay
CV was also reported for calibration standards at all levels.
P
2.5.2. Intra- and Inter-Assay Accuracy
The intra- and inter-assay accuracy o f the method was assessed by determining the relative error (RE) observed in the analysis o f QC samples. The RE was determined by subtracting the nominal concentration from the mean concentration for each QC level, dividing the result by the nominal value, and then expressing the result as a percentage. The intra-assay accuracy was expressed as the RB o f the replicate analyses o f QC samples during a single validation run. Inter-assay accuracy was expressed as the RE of the overall mean of the individual replicate concentrations of the QC samples for three validation runs. QC1, QC2, and QC3 were assayed in replicates of six. In addition, the inter-assay EE was also reported for calibration standards at all levels.
Page 12 o f 54
0212DVDJA_DU.DOC
000087
2.5.3, Sensitivity: Lower Limit of Quantitation (LLQ) A human serum QC was prepared at a concentration of 0.05 pg/mL PFOA and assayed in replicates o f six. The LLQ was determined by obtaining the backcalculated concentrations from these QC samples. The mean concentration, CV, and RE were also calculated.
2.5.4. Selectivity The selectivity of the assay was determined by LC/MS/MS, To monitor for interference from the biological matrix within runs, unfortified control blank human serum samples were assayed with all experiments except recovery and matrix effect.
In addition, six replicate zero samples (control blank human serum samples fortified with IS) were extracted in each of the two different lots o f control serum used for the standard curves and QC samples. These zero samples were used for the background subtraction due to the presence o f PFOA in the control serum lots (Section A. 14.0).
I
To monitor for lot-specific matrix interference, six separate lots o f human serum were spiked with PFOA at the LLQ and upper limit o f quantitation (ULQ) concentration levels and were analyzed along with a control blank and zero sample prepared from each lot.
2.5.5. Chromatographic Carryover The carryover of analyte from the injection of one sample to another was assessed in each nun by placing control blanks after each high standard (STD9) and monitoring for analyte and IS response.
Page D of 54
02120VDM. DU.DOC
000088
2.5.6. Accuracy and Precision of Dilution The accuracy and precision of dilution in the analysis of PFOA in human serum was assessed. QC4 samples (50 pg/mL) were analyzed at a 10-fold dilution in replicates of six. The mean concentration, CV, and RE were calculated.
2 .6 . STABILITY O F PFO A IN HUM AN SERUM The stability of PFOA in human serum during sample storage, preparation and analysis will be assessed and reported in the 1 to 100 ng/mL-concentration range human serum method validation.1
2.7, Reproducibility of Reinjecting Prepared Samples The reproducibility o f reinjecting extracted and reconstituted serum samples was investigated by reinjecting a set o f previously-assayed standards and QC samples that had been stored after injection at room temperature for 65 hours. The CV and RE of the Q samples were used to assess processed sample stability in reconstitution solution. All QC samples were assayed in replicates o f six.
2 .8 . EXTRACTION RECOVERY
The determination o f extraction recovery was initially performed in Runs 6 and 7, but the data obtained from those experiments are believed to be inaccurate, mainly due to the higher than expected absolute recoveries seen (100% maximum expected from procedure after correction for 1/10 aliquot removed from organic layer during extraction). It is suspected that because of the large differences (10-fbld) in concentration between the samples spiked before extraction (pre-extract) and the samples spiked after extraction (post-extract) in those experiments, coupled with the non-linearity seen in detector responses, the predicted recoveries were biased. The experiment was repeated in Run 8 in which the post-extract samples were diluted 10-fold to compensate for the 1/10 aliquot Accordingly, the concentrations of the
Page 14 o f 54
02I20VMA_DU.DOC
000089
pre- and post-extract samples would now be much closer to each other and thus be less biased by the non-linearity.
The extraction recovery of PFOA from human serum was determined in MacQuan by performing separate weighted (1/x2) regressions for the pre-extract and post-extract samples which were prepared at the nominal concentrations 0.15,3, and 8 pg/raL in replicates of five. The extraction recovery was calculated as the slope ratio of the pre-extract curve over the post-extract curve. The IS was spiked post-extraction for PFOA recovery determination.
The extraction recovery of the IS from human serum was determined jby comparing
the peak area ratios (PAR) of pre-extract and post-extract samples (5 pg/mL
9H-HDFNA) assayed in replicates of five. In this experiment only, the PFOA was
used as internal standard and was spiked post-extraction for the 9H-HDFNA
recovery determination. The extraction recovery (% Recovery) was determined by
dividing the pre-extract PAR by the post-extract PAR and expressing the result as a
percentage.
i
2.9. Matrix Effect on Ionization of PFOA and 9H-HDFNA m Human Serum
The matrix effect on the ionization o f PFOA in human serum extract was determined by comparing the peak area of samples (0.15,3, and 8 pg/mL for PFOA) spiked in blank human serum exlract (post-extract) with the peak area o f an unextracted solution (neat spike). The matrix effect was determined by subtracting the neat spike peak area from the post-extract peak area, dividing the result by the neat peak spike peak area, and expressing the result as a percentage. Five replicates were used at each concentration.
The matrix effect on the ionization of 9H-HDFNA in human serum extract was similarly determined by comparing the peak area of blank human serum extract spiked with 5 pg/mL 9H-HDFNA after extraction with the peak area of an
Page IS o f 54
02i2QVDJA_DU.DOC
000090
unextracted solution (neat spike). The post-extract and neat solution samples were prepared in replicates o f five.
3. RESULTS AND DISCUSSION
A quantitative analytical procedure using turbo ion spray LC/M8/MS in the negative ion mode was developed to meet the high sensitivity, selectivity, and reproducibility requirements for the determination of PFOA in human serum. A list o f the validation runs is presented in Table l.
The full-scan turbo ion spray mass spectrum o f PFOA showed the b a ^ peak ion at m/z ~ 413 which is attributed to the deprotonated molecule (Figure 1). The full-scan product ion spectrum of PFOA showed the base peak product ion at 169 (Figure 2).
The full-scan turbo ion spray mass spectrum of 9H-HDFNA showed the base peak ion at m/z = 445 which is attributed to the deprotonated molecule (Figure 3), The lull-scan product ion spectrum of 9H-HDFNA showed the base peak product ion at 381 (Figure 4).
The following respective [M-H]' ions were used in data acquisition for PFOA and the IS in human serum (nominal):
PFOA 9H-HDFNA
m/z = 413 --> 169 m/z = 445 ->381
The SRM chromatograms from a representative control blank human serum extract are shown in Figure 5. SRM chromatograms from a representative control human serum extract containing only the internal standard (zero sample), are shown in Figure 6. Both Figures 5 and 6 show the presence o f PFOA in the control human serum.
Page 1.6of 54
02120VDJA_DU.DOC
000091
Figures 7 and 8 are SRM chromatograms from representative extracts from calibration standards 1 (LLQ) and 9 (ULQ), respectively.
3.1. Assay Evaluation Results
3 4 4 . Intra- and Inter-Assay Precision Table 2 presents the precision data for the PFOA QC samples from the three validation runs. The intra- and inter-assay precision (CV) was <7.70% for PFOA at QC1, QC2. and QC3 concentration levels. Table 3 presents the precision data for calibration standards. The CV for calibration standards was <10.0% lpr PFOA. These data indicate acceptable intra- and inter-assay precision for the determination of PFQA in human serum.
Only five replicates o f QC1 were successfully prepared in Run 3 due to a broken test tube. Also, results for only five replicates of QC3 in Run 5 were calculated due to a probable preparation error. This is in compliance with Advion SOPs.
34,2. Intra- and Inter-Assay Accuracy Table 2 presents the accuracy data for the PFOA QC samples from the three validation runs. The intra- and inter-assay accuracy (RE) ranged from -12.3 to 9.00% for PFOA. Table 3 presents the accuracy data for calibration standards. The RE for calibration standards ranged from -3.99 to 6.92% for PFOA in human serum. These data indicate acceptable intra- and inter-assay accuracy for the determinalion o f PFOA in human serum.
34.3. Calibration Curve Fit The calibration curves were fit by a weighted (1/x2) quadratic regression. Coefficients of determination (r2) were 2:0.9950 for PFOA in human serum. The calibration curve statistics are shown in Table 4,
Page 17 o f54
02120VDiA._DU.DOC
000092
3.1.4. Sensitivity: Lower Limit of Quantitation (LLQ) Tables 5 shows the results from the QC at the LLQ for FFOA. The experiment demonstrated that 0.05 jig/mL is an acceptable LLQ level for PFOA. The CV was 2.95%, and the RE was 9.73% for the LLQ QC.
3.1.5. Selectivity The assay was highly selective for PFOA. No chromatographic interferences were observed at the retention time o f PFOA or the IS in the control serum^samples analyzed (Figure 5). The control blanks did show the presence o f PFOA in the control human serum.
The selectivity between matrix lots produced acceptable results after background subtraction (Table 6). The precision (CV) for the PFOA samples was 7.18 and 5.21% for the LLQ and ULQ samples, respectively. The corresponding accuracy was 0.133 and 11.8% for the LLQ and ULQ samples.
3.1.6, Chromatographic Carryover A chromatographie peak was observed at the retention time for PFOA in the human serum control blanks following high standards. The responses were comparable to those o f the control blanks injected before the standard curves. This indicates that carryover was negligible for PFOA. Carryover for the internal standard was insignificant.
3.1.7. Accuracy and Precision of Dilution The results from QC4 samples diluted 10-fold are shown in Table 7. The CV was 4.25% and the RE value was -10.6% for PFOA These data demonstrate acceptable results for the dilution o f samples.
Page 18 o f54
02120VDJA_Dl ! DOC
)
000093
3.2. R eproducibility of r ein jec tin g Extracted Sam ples
The results of reinjecting extracted and reconstituted samples containing PFOA after storage for 65 hours at room temperature in reconstitution solution are shown in Table 8. The CV values for samples injected initially and those reinjected after 65 hours in reconstitution solution ranged from 2.68 to 8.12%, and the RE values range from --12.3 to -1.41% for PFOA. These data indicate that reinjecting samples after storage for 65 hours in reconstitution solution at room temperature is acceptable.
3.3. EXTRACTION RECOVERY
The extraction recoveries o f PFOA and the IS, 9H-HPFNA, from human serum are shown in Tables 9 and 10, respectively. The mean recovery values were 81.8% for PFOA and 90,9% for 9H-HDFNA.
3.4. Matrix Effect on Ionization of PFOA and 9H-HDFNA in human
Serum
i
The results o f the matrix effect ofhuman serum extract on the ionization o f PFOA and the IS are shown in Tables 11 and 12, respectively. The overall mean matrix effect was 1.62% for PFOA and 8.07% for 9H-HDFNA, mdicating slight enhancement ofionization.
4. SOP DEVIATION According to Advion SOP POCU0015.02, the volume of non-matrix solvent used for spiking QC samples must be 5% of the total volume. QC4 (dilution QC) was prepared by spiking a volume of stock solution that was 10% of the total volume so that new stock solution did not have to be prepared. This deviation did not affect the quality o f the data.
There were no other known SOP deviations during the validation.
Page 19o f54
02120VDJA_DU.DOC
000094
5. DATA RETRIEVAL
The data for the human serum validation are stored in Advion Notebook 02120 and in the Advion archives.
6. CONCLUSIONS
The turbo ion spray LC/MS/MS assay procedure for PFOA in human serum has
proven to be sensitive, selective, accurate, and reproducible. Its high sensitivity
allows reliable and reproducible quantitation o f PFOA down to a level of 0.05 pg/mL
in human serum based on 25-pL samples.
*
7. REFERENCE
1. Advion Report No. 02086VDJA DU.DOC, "Method Validation for the Quantitation o f Ammonium Perfluorooctanoate (APFO), Measured as the Perfluorooetanoate Anion (PFOA, 1 to 100 ng/mL Concentration Range) in Human Serum by Turbo Ion Spray LC/MS/MS " to be issued.
Page 20 o f 54
02 !20VDJA_DU.DOC
00009
8. TABLES
Table 1: Vaiidadon Run Descriptions
Run No. 1 2 3 4 5 6 7
Assay Date 6-May-02 7-May-02 8-May-02 9-May-02 10-May-02 22-May-02 23-May-02
8 5-June-02
Run Description
Status
Accuracy and Precision; Dilution
Accepted
Accuracy and Precision; LLQ
Rejected
Accuracy and Precision; LLQ
Accepted
Reinjection of Run 1
Accepted
Accuracy and Precision; Selectivity * Accepted
Recovery and Matrix Effect
Rejected
Recovery and Matrix Effect Recovery
Accepted (recovery rejected)
Accepted
(
Page 21 of 54
02120VDM_DU.DOC
000096
Table 2
Accuracy and Precision of the PFOA Assay in Human Serum for Quality Control Samples from Three Validation Runs
PFOA Concentration (pg/tnL)
QC1 QC2 QC3
Run Number
0.15
3
8
1
0.147
2.73
8.47
0.138
2.55
8.35
0.139
2.56
7.54
0.156 0.144 0.125
2.59 2.67 2.68
7.45 7.37 7.57
Mean
0-142
2.63
7.79
CV(%) RE(%)
7.33 2.68 6.26 -5.67 -12.3 -2.61
3 0.164 2.82 7.62
Mean CV (%) RE (%)
0.159 2.78 7.97
0,158
2.74
7.48
0.152
2.85
7.33
0.153
2.77
7.77
NS 2.69 7,83
0.157
2.77
7.67
3.10 2.07 3.06
4.80 -7.56 -4.16
5
0.169
2.96
7.83
0.166
2.88
8.17
0.157
2.78
8.06
0.162 2.88 EQB
0.165
2.81
7.75
0.162
2.92
8.65
Mean CV(%) RE (%} Overall Mean Overall CV (%) Overall RE (%)
0.164 2.53 9.00 0.154 7.70 2.59
2.87 2.32 -4.24 2.76 4.33 .8.04
8.09 4.38 1.12 7.84 5.02 -2.06
CV = (SD/Mean)*100 RE = [(Mean-Nominal)/Nominal]* l 00
NS = No sample. Only 5 replicates were prepared for QC1 in Run 3. EQB = Exceeds quadratic bounds
Page 22 o f 54
02120VDJAJDU.DOC
7
000098
Table 3:
Accuracy and Precision of the PFOA Assay in Human Serum for Calibration Standards from Three Validation Runs
STD1 STD2 STD3
Run Number 0.05 0.1 0.25
1 0.0534 0.103 0.255
0.0463 0.0967 0.246
3 0.0513 0.112 0.243
0.0457 0.102 0.231
5 0.0562 0.106 0.252
0.0436 0.0950 0.244
Mean ev(% ) RE{%)
0.0494 10.0 -1.17
0.102 6.06 2.45
0.245 3.42 -1.93
CV = (SD/Mean)*100
RE = [(Mean-Nomiial)/Noininal]*iOO
PFOA Concentration (ng/rnL)
STD4 STD5 STD6
0.5 1 2
0.574
1.01
2.06
0.469 0.932
1.95
0.539 1.02 2.02
0.492
1.01
1.90
0.505 1.05 2.11
0.488 0.996
1.95
0.511 1.00 2.00
7.55 3.82 3.88
2.23 0.233 -0.125
STD7 5
4.80 4.83 4.90 4.66 4.95 4.66 4.80 2.49 -3.99
STD8 7.5 7.75 6.97 7.34 7.20 7.26 7.26 7.30 3.49 -2.71
STD9 10 9.84 11.6 10.5 11.0 11.0 10.1 10.7
6.08 6.92
Page 23 o f 54
02120VDM_DU.DOC
S
Table 4:
Calibration Curve Parameters for PFOA in Human Serum from Three Validation Runs
Run Date Run
A
Number
6-May-02 1 -0.011789
8-May-02 3 -0.009249
10-May-02 5 -0.008130
Mean
-0.009723
y = Ax2 + Bx + C, weighted 1/x2
B
0.360080 0.317953 0.289836 0.322623
C Coefficient of
Determination (r2)
0.000634
0.9950
-0.000595
0.9955
-0.000240
0.9951
-0.000067
0.9952
I
I
Page 24 o f 54
02120VDJA_DU.DOC
000099
Table 5: Lower Limit of Quantitation of PFOA in Human Serum
PFOA Cone. (jfg/mL)
Run 3
QCLLQ 0.05
0.0562
0.0537
0.0574
0.0540
0.0547
0.0532
Mean CV(%)
0.0549 2.95
RE (%)
9.73
C V --(SD/Mean)*L00
RE ~ [(Mean-Noniinai)/Nominal]*100
Page 25 o f 54
02120VDJA_DU.DOC
O O O IO O
Table 6:
Lower Limit of Quantitation of PFGA and Selectivity in Six Separate Lots of Human Serum
PFOA Concentration (pg/inL)
Run 5
LLQ
ULQ
Serum lot
0.05
10
22-48635
0.0544
10.6
55-03231 55-03232 55-03681 55-03624 55-03707
0.0467 0.0508 0.0531 0.0503 0.0451
10.6 10.8 11.4 11.6 12.0
Mean V (% ) RE (%)
0.0501 7.18 0.133
11.2 5.21 11.8
CV = (SD/Mean)*100
RE = [(Mean-Nominal)/Norninal]*100
1
Page 26 o f54
02120VDJA_DU.DOC
000101
Table?:
Accuracy and Precision of the PFGA Assay in Human Serum Dilution QC Samples from One Validation Run
PFOA Cone. (pg/mL)
QC4 (1 in 10 dilution)
Run 1
50
47.5
42.7
45.3
43.9
42.9
46.0
Mean
44.7
CV(%)
4.25
RE(%)
- 10.6
CV = (SD/Meaji)* 100
RE = [(Mean-NommalJ/Noaunal]*!^
Page 27 of 54
02120VDJA_DU.DOC
000102
Table 8:
Reinjection Reproducibility of QC Samples Containing PFOA After Storage in Reconstitution Solution for 65 Hours at Room T em p eratu re
PFOA Concentration (pg/mL)
QC1 QC2 QC3 QC4 (1 in 10 dilution)
0.15 3
8
50
Run 1 t=0
0.147 0.138 0.139 0.156
2.73 2.55 2.56 2.59
8.47 8-35 7.54 7.45
47.5 42.7 45.3 43.9 (
0.144
2.67
7.37
42.9
0.125
2.68
7.57
46.0
Mean
0.142
2.63
7.79
44.7
C V (% ) R E (% )
7.33 -5.67
2.68 -12.3
6.26 -2.61
4.25 - 10.6
Run 4
0.150
2.78
8.10
47,.6
t=65 hours
Mean CV(%) R E (% )
0.146 0.140 0.145 0.141 0.118 0.140 8,12 -6.67
2.88 2.62 2.79 2.65 2.57 2.72 4.43 -9.48
8.81 7.76 7,52 7.63 7.50 7.89 6.37 -1.41
45.2 47.4 45.8 46.0 46.3 46.4 2.05 -7.27
CV - (SD/Mean)* 100
RE - t(Mean-Nominal)/NominaI]*100
Page 28 o f 54
02I20VDJA_DU.DOC
00103
Table 9: Extraction Recovery of PFOA from Human Serum
Run 8 QC1 0.15 pg/mL
Mean RE (%)
QC2 3 pg/inL
Mean R E (% )
QC3 8 pg/mL
Mean RE (%)
PFOA Concentration (pg/niL)
Pre-Extract Post-Extract
0.162 0.141 0.155 0.145 0.146 0.150 -0.133
0.155 0.152 0.147 0.149 0.144 0.149 -0.400
3.21 3.40 3.07 3.27
2.95 3.35 3.05 3.36 3.50 3.20 3.16 3-32 5.22 10.5
8.15 7.40 7.51 7.53 7.39
7.60 -5.05
7.25 7.15 7.12 7.18 7.22 7.18 - 10.2
Regression
Pre-Extract Post-Extract Recovery (%)
Coefficient of Determination (r2)
Intercept Slope
0.9930 0.01125 0.23325
0.9878 -0.01220 0.28522
81.8
Pre-Extract = Matrix spiked with PI7OA prior to extraction Post-Extract= Matrix spiked with PFOA after extraction RE ~ [(Mean-NominaiyNominai]* 100
%RecOVeiy --(Slopepre<xnac/S1^6post-exiract)*100
Page 29 of 54
02120VDJA_DjCr.DOC
C00104
Table 10: Extraction Recovery o f the Internal Standard, 9H-HDFNA, from Human Serum
Run 8 IS
5 pg/mL
Mean CV(%)
Peak Area Ratio (IS/PFOA)
Pre-Extract Post-Extract
0.43745
0.49204
0.45424 0.44736
0.49670 0.49028
0.46195
0.50136
0.44684
0.49258
0.44957
0.49459
2.03 0.901
Recovery (%)
90.91
Pre-Extract = Matrix spiked with 9H-HDFNA prior to extraction Post-Extract = Matrix spiked with 9H-HDFNA after extraction CV = (SD/Mean)*100 %Recovery= (Meanpre-extraa/Meanpos^^,)*100
i
Page 30 of54
02120VDJA_DU.DOC
000105
Table 11: Matrix Effect on Ionization of PFGA in Human Serum
PFOA Area
Matrix Effect.
Run? QCl
Post-Extract 19775
Neat Spike 19573
(%)
0.15 pg/mL
19835
18856
20826
20522
19880
18616
20597
18333
M ean CV(% )
20183 2.43
19180 4.59
5.23
QC2 3 pg/mL
426092 418407 408036
399353 408972 419660
425636
400556
430791
401508
M ean CV (% )
421792 2.10
406010 2.10
3.89
QC3 8 pg/mL
772284 747001
772628 794112
757322 769807
805450 817268
776010
803311
M ean CV(% )
764485 1.57
798554 2.09
-4.27
Overall Mean
1.62
Post-Extract-M atrix spiked vrith PFOA after extraction Neat Spike - Solution of PFOA without matrix
C V - (SD/Mean)* 100 Matrix Effect = [(MeanPostExtract- MeanNouspiteJ/M ean^, spike]*100, where a negative number indicates ion suppression and a positive number indicates enhancement
Page 3 1of 54
02120VDM_DU.DOC
000106
Table 12: Matrix Effect 00 Ionization of 9H-HDFNA in Human Serum
9H-HDFNA Area
Matrix Effect
Run 7 IS
Post-Extract 349304
Neat Spike 306352
(%)
5 pg/mL
336593
330146
351612
310681
335236
322237
335737
311522
M ean CV (%)
341696 236
316188 3.08
8.?
Post-Extract - Matrix spiked with 9H-HDFNA after extraction
Neat Spike = Solution o f 9H-HDFNA without matrix CV = (SD/Mean)*lOO
Matrix Effeet = [(MeanPostEw t- MeanNeilSpike)/MeanNEa, Spike]*100, where a negative number indicates ion suppression and a positive number indicates enhancement
Page 32 of 54
02120VDJA_DU.DOC
000107
9. FIGURES Figure 1: Full-Scan Turbo Ion Spray Mass Spectrum of PFOA
Page 33 of54
Q2120VDJA_DU.DOC
000108
Figure 2: Full-Scan Product Ion Mass Spectrum of PFO
Spectrum from PFOA MS/MSPROD -10 OR
1,47e6 cps
Relative Ion Abundance (%)
Page 34 of 54
02120VDJA_DU.DOC
000109
Figure 3 : Full-Scan Turbo Ion Spray Mass Spectrum of 9H-HDFNA
Spectrum from 9H-HDFNA Q1 -10 OR
1.58e6 cps
Relative Ian Abundance (%)
Page 35 o f 54
02120VDJA_DU.DOC
O O Xlo
Figure 4 : FuII-Scau Product Ion Mass Spectrum of 9H-HDFNA
Relative Ion Abundance (%)
Spectrum from 9H-HDFNA MS/MS PROD -10 OR 96-
88-
381
80-
72-
64'
5648 -
40-
3 2 ' 119
24-
16-
8-
63. -- r 100
169
23,1 --r~ 200
-- j_ 331
300
mlz
i
400
9.91e5 cps I
~I----------------1
500
Page 36 o f 54
02120VD1A.PU.DOC
O O O IH
Figure 5:
Selected Reaction M onitoring C hrom atogram s of PFGA and the Internal Standard in C ontrol B lank Hum an Serum E xtract
Relative Ioa Abundance (%)
Control Blank, Run l, Injection 7
Pagc37of54
0212QVDJA_DU.DOC
O O O ll2
Figure 6:
Selected Reaction M onitoring C hrom atogram s of PFOA and the Internal Standard in a Human Serum E xtract Containing In tern al Standard Only (Zero Sam ple)
m /z = 413.0 --> 169-0, PFOA
3.85e2 cps
Relative Ion Abundance (%)
Zero Sample, Run 1, Injection I
Page 38 o f 54
'J2120VDJA_DU.DOC
000113
Figure 7:
Selected Reaction M onitoring Chromatogram!; of PFOA and the Internal S tandard from a S tandard 1 Hum an Serum E xtract (0.05 ng/mL)
m /z = 4 l3 .0 - > 169.0, PFOA
8.0Qe2 cps
Relative Ion Abundance (%)
Standard 1, Run 1, Injection 8
Page 39 o f 54
02120VDJA_DU.DOC
000114
Figure 8:
Selected Reaction M onitoring Chrom atogram s of PFOA and the Internal Standard from a Standard 9 Human Serum Extract (10 (tg/raL)
m /z 5413.0 -> 169.0, PFOA
5.74e4 cps
Standard 9, Run 1, Injection 16
Page 40 o f 54
02UOVDJA_DU.DOC
10. APPENDIX A: QUANTITATION OF AMMONIUM PERFLUOROOCTANOATE (APFO), MEASURED AS THE PERFLUOROOCTANOATE ANION (PFOA, 0.05 TO 10 pg/mL CONCENTRATION RANGE), IN HUMAN SERUM BY TURBO ION SPRAY LC/MS/MS
AUTHORS:
David J, Anderson, .MLS. Kimberly L . Norwood, B.A. Kathy Jo Palm er, B.S. Jennie L. Romney, M.P.S.
TABLE OF CONTENTS
I
A.1.0 Chemical Structure(s)............ .............................. ............................... ...............42
A 2.0 Specim en^).............. ..... >.......... ......................................................................43
A.3,0 Assay Principle................ ............................ .................. .................................. 43
A.4.0 Compounds............. ............................. ............................ 1............... ............. .43
A.5.0 Chem icals ...................................... ..... ......................................................... 43
A.6.0 M aterials and Equipment................. ....................................... ......................... 44
A.7.0 Preparation o f Calibration Standards........ .............................
46
A.8.0 Preparation of Internal Standard Solutions................
......47
A.9.0 preparation o f Quality Control Samples........................
...47
A.10.0 Preparation of Other Solutions............................
48
A.11.0 Liquid-liquid Extraction Procedure................................................................. 50
A.12.0 Typical Instrument Conditions............ ............................................................. 51
A.13.0 System Suitability.......................................................... ................................. .53
A. 14.0 Calculations..................................................................................................... 53
Page 41 o f54
02120VDIA DUDOC
A 4 .0 CHEM ICAL STRUCTURER)
m 4+
o
. II
o -- c -- c 7f 15
Ammonium Perfluorooctanoate (APFO) C 8H4F 1SN 02 Monoisotopic M ass = 431.0
O C --- G7F 15 Perfluorooctanoate Anion (PFOA) C 8Fig 0 2 Monoisotopic M ass = 413.0
i
H O - C -- C F 2(CF2)6C F2H 9H-Hexadecafluorononanoic Acid (9H-HDFNA) C9H2F 160 2 Monoisotopic M ass = 446.0
Page 42 of 54
02120VDJA DU.DOC
000117
A.2.0 SPECIM EN S) This assay uses 25-\iL aliquots o f human serum. Human serum samples should be stored at -2 0 C.
A.3.0 ASSAY PRINCIPLE Perfluorooctanoate anion (PFOA) and its IS, 9H-hexadecafluorononanoic acid (9H-HDFNA), were extracted from human serum samples (25 pL) using a liquidliquid extraction procedure. An aliquot o f the organic layer o f each sample was evaporated to dryness, reconstituted, and then analyzed by turbo ion spray liquid chromatography/tandem mass spectrometry (LC/MS/MS) in the negative ion mode.
A.4.0 COMPOUNDS
Pentadecafluorooctanoic acid ammonium salt (also known as ammonium
perfluorooctanoate, APFO): Lot# 421207/1,100% purity, 0.9582 salt factor,
Sigma-Aldrich, St. Louis, MO.
i
O FrsCh _ 413.064 Salt Factor:
CsH 4FisN 02 431.10258 9H-HDFNA: No Lot# available, 100% purity assumed, Sigma-Aldrich, St. Louis,
MO.
A.5.0 CHEMICALS All chenricals may be substituted with that o f an equivalent manufacturer and grade o f chemical.
Tetrabutylammonium Hydrogen Sulfate
Sodium Carbonate, Annhydrous
HPLC Grade, Cat# 1360-07, J.T. Baker, Phillipsburg, NJ
HPLC Grade, Cat# 32503-27-8, Aldrich, Milwaukee, Wl
Cat# 3604-01, J.T. Baker, Phillipsburg, NJ
/S"*-
Page 43 of 54
02120VDJA_DU.DOC
Sodium Bicarbonate Ethyl Acetate Ammonium Acetate Ammonium Hydroxide W ater Methanol Acetonitrile Human Serum
Cat# 3509-01, J.T. Baker, Pbillipsburg, NJ
Cat# 100-4, Burdick & Jackson, Muskegon, MI
Cat# 24,019-2, Aldrich, Milwaukee, WI
Cat# AX1303-13, EM Science, Gibbstown, NJ
Cat# 365-4, Burdick & Jackson, Muskegon, MI
Cat# 230-4, Burdick & Jackson, Muskegon, MI
Cat# BJ015-4, Burdick & Japkson, Muskegon, MI
Biological Specialty Corporation, Colmar, PA
A.6.0 M ATERIALS AND EQUIPMENT
Mass Spectrometer
Data Acquisition System
HPLC Pumps LC Pump Controller Autosampler HPLC Column
SCDEX API 3000 atmospheric pressure ionization tandem triple quadrupole mass spectrometer equipped with TurboIonSprayTM interface, Applied BioSystems, Concord, Ontario
API Standard Software, MacQuan v 1.6, LC2Tune v 1.4, MacDAD v 1.4.1, Bundler v 1.4, Multiview v 1.4, and Sample Control v 1.4, on a Power Macintosh G3/450, Applied BioSystems, Concord, Ontario
Shimadzu LC-10AD vp pumps, Shimadzu Co., Columbia, MD
Shimadzu SCL-10A vp, Shimadzu Co., Columbia, MD
Perkin-Blmer Series 200, Perkin-Elmer Corp., Norwalk, CT
Betasil CiSi 5 pm particle size, 2 x 50 mm, Cat# 852055-701, ThermoHypersil-Keystone, Bellefonte, PA
Page44of54
02120VDJA_DU.DOC
000119
Column Heater
Harvard Syringe Pump 11
Polypropylene 96-Well Blocks
Screw-capped Polypropylene Vials
Screw-capped Polypropylene Vials
Siliconized Graduates M icrotubes
Polypropylene Centrifuge Tubes
Polypropylene Bottles
Polypropylene Bottles
TurboVap96 Evaporator
Beckman GS6KR Refrigerated Centrifuge
M ulti-tube Vortexer
Mechanical Shalcer
Model 7970, Jones Chromatography USA Inc., Lakewood, CO Harvard Apparatus Inc., South Natick, MA
Cat# 267006, Beckman Instruments, Inc., Fullerton, CA
16.5 x 101 mxn, Cat# 60.541, Sarstedt, Inc., Newton, SC 28658 16.5 x 57 mm, Cat# 60.542, Sarstedt, Inc., Newton, SC
1.5 mL, Cat# MH-815S, Phenix Research Products, Hayward, CA 94545
50 mL with screw caps. Cat# 21008-178, VWR Scientific, Bridgeport, NJ 175 mL Nalgene, Cat# 16120-953, VWR Scientific, Bridgeport, NJ 1 L Nalgene, Cat# 16120-962, VWR Scientific, Bridgeport, NJ Cat# 71000, Zymark Instruments, Hopkinton, MA 01748 Beckman, Palo Alto, CA 94304
Cat# 58816-115, VWR Scientific, West Chester, PA Cal# 6000, Eberbach Corp., Ann Arbor, MI
W atson
Laboratory Information Management System v 6.1.0.03, InnaPhase Coip., Philadelphia, PA
Other general laboratory glassware and supplies were used,
Page 45 o f54
02120VDJA_DU.DOC
000120
kk
A.7.0 PREPARATION OF CALIBRATION STANDARDS
Mix all solutions well. Unless otherwise noted, store solutions at 4 C and bring to ambient temperature before use. Unless otherwise noted, prepare fresh solutions every three months or as needed.
A.7.1 Preparation of Analytical S tandard Solutions
S tandard Stock Solution, PFQA (500 pg/mL): Weigh approximately 2 mg pentadecafluoroactanoic acid ammonium salt and transfer to a 16.5 X57 polypropylene vial. Dilute with appropriate volume o f methanol to yield a 500pgfrnL solution after correction for salt factor. Vortex for one minute.^
A.7.2 Preparation o f Standard Curve
Prepare Standards 1 through 9 by dilution o f Standard Standard Stock Solution or th appropriate standard into control serum according to the following table. Standard 9 is prepared in a 5-mL volumetric flask. Standard 1 through Standard 8 are prepared in 1.5-mL siliconized polypropylene microtubes.
A nalytical S tandard Dilution Table:
Standard Final Cone. No. (pg/mL)
Solution to Use
9 10 Std Stock
8 7.5 75 62 51 4 0.5 3 0.25
2 0.1 1 0.05
STD9 STD9 STD9 STD9 STD7 STD7 STD5 STD5
pL of Solution
100
375 250 100 50 50 25 50 25
pL C ontrol Semim
Add serum to 5-mL made
125 250 400 450 450 475 450 475
nptL T o ta l V olum e
5
0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
Page 45 of 54
02I20VDJA_DLf.DOC
000121
A.8.0 PREPARATION OF INTERNAL STANDARD SOLUTIONS Mix all solutions well. Unless otherwise noted, store solutions at 4 C and bring to ambient temperature before use. Unless otherwise noted, prepare fresh solutions every three months or as needed.
Internal Standard Stock Solution, 9H-HDFNA (500 pg/m L): Weigh approximately 2 mg 9H-HDFNA and transfer to a 16.5 x 57 polypropylene vial. Dilute with appropriate volume of methanol to yield a 500-pg/mL solution. Vortex for one minute,
Internal Standard W orking Solution, 9H-HDFNA (5 pg/mL): Transfer 250 pL of Internal Standard Stock Solution into a 25-mL class "A" volumetrie flask containing approximately 20 mL o fwater. Dilute to volume with water. Vortex and transfer to a 50-mL polypropylene centrifuge tube.
A.9.0 PREPARATION OF QUALITY CONTROL SAMPLES
A.9.1 Preparation o f Serum QC Samples
Prepare a 500-pgfrnL QC Stock Solution in the same manner as the Standard Stock
Solution in A.7.1. Prepare QC samples in a 5-mL volumetric flask by dilution o f the
QC Stock Solution, QC3, or QC2 into approximately 3 mL o f control human serum
(from a different lot than is used to prepare the standard curve) and diluting to
volume according to the following table:
QC#
A nalyte C one. (pg/m L)
Solution to U se
pLof Solution
D ilute to F inal V olum e (m L)
QC4
50
QC Stock
500
QC3
8
QC Stock
80
QC2
3
QC3
1875
5 5 5
QCI
0.15
QC3
93.75
5
QCLLQ
0.05
QC2
83.1
5
Page 47 o f 54
0212QVDJAJDU.DOC
000122
Transfer 0.2-mL aliquots o f each QC into microcentrifuge tubes and store frozen at -20 C. Note that the QC LLQ was a one-time preparation and was not stored.
A.10.0 PREPARATION OF OTHER SOLUTIONS HPLC-grade bottled water or equivalent should be used wherever water is called for. Mix all solutions well. Different volumes o f the solutions in this section may be prepared if necessary. Unless otherwise noted, store at room temperature. Prepare fresh solutions every three months or as needed, unless otherwise noted.
0.5 M Tetrabutylam naonium Hydrogen Sulfate, pH 10: Dissolve approximately 17 g o f tetrabutylaornmonium hydrogen sulfate in 40 mL water in a 175-mL polypropylene bottle. Adjust the pH to 10 with 10 M and 1 M sodium hydroxide. Transfer to a 100-mL class "A" volumetric flask and dilute to the mark with water. Transfer to a 175-mL polypropylene bottle.
10 M Sodium Hydroxide: Dilute approximately 40 g sodium hydroxide to the mark with water in a 100-mL class "A" volumetric flask. Mix and transfer to a 175-mL polypropylene bottle.
1 M Sodium Hydroxide: Dilute 10 mL 10 M sodium hydroxide to fee mark with water in a 100-mL class "A" volumetric flask. Mix and transfer to a 175-mL polypropylene bottle.
0.25 M Sodium Garbonate:0.25 M Sodium B icarbonate: Dissolve 26.5 g sodium carbonate and 21 g sodium bicarbonate in approximately 500 m l. water. Transfer to a 1-L polypropylene graduated cylinder and dilute to the mark with water. Mix and transfer to a 1-L polypropylene bottle.
1 M Ammonium Acetate: Dissolve 7.7 g ammonium acetate to the mark with water in a 100-mL class "A" volumetric flask. Mix and transfer to a 175-mL polypropylene bottle.
Page 48 o f 54
0Zl20VDtAJDU.DOC
000123
2 mM Ammonium Acetate: Add 2 mL o f l M ammonium acetate to approximately 900 mL water in a 1000-mL polypropylene graduated cylinder and dilute to volume with water. Mix and transfer to a 1-L polypropylene bottle.
10:90 M ethanol:2 mM Ammonium Acetate: Combine 100 mL methanol and 900 mL 2 mM ammonium acetate in a 1000-mL polypropylene graduated cylinder. Transfer to a 1-L polypropylene bottle.
90:10 M ethanol:2 mM Ammonium Acetate: Combine 900 mL methanol and 100 mL 2 mM ammonium acetate in a 1000-mL polypropylene graduated cylinder. Transfer to a 1-L polypropylene bottle.
1% Ammonium Hydroxide: Dilute 3.6 mL ammonium hydroxide to the marie with water in a 100-mL class "A" volumetric flask. Mix and transfer to a 100-mL glass reagent bottle.
0.01% Ammonium Hydroxide: Dilute 2.5 mL 1% ammonium hydroxide to the mark with water in a 250-mL class "A" volumetric flask. Mix and transfer to a 250-mL glass reagent bottle.
80:20 A cetonitrile:0.01% Ammonium Hydroxide: Combine 800 mL acetonitrile and 200 mL 0.01% ammonium hydroxide in a 1-L glass reagent bottle.
,,J
Page 49 of 54
O2i20VDJA_DU.DOC
000124
A.11.0 LIQUID-LIQUID EXTRACTION PROCEDURE
1. Thaw samples and QCs. 2. Prepare the standards as outlined in Section A.7.2. 3. Label 16.5 x 101 mm screw-capped polypropylene tubes for each sample,
standard, QC, control blank, and zero sample. Standards, carryover control blanks, and selectivity control blanks are analyzed in duplicate. QCs are analyzed in replicates o f six. Six replicate zero samples are prepared in each of the two different lots o f control serum used for the human serum standard curves and human serum QC samples. 4. Aliquot 25 pL o f each standard, QC, zero sample, control blank, and sample into the appropriate tube. 5. Add 25 pL o f Internal Standard Working Solution (5 pg/mL 9H-HDFNA) to all tubes except for the control blank samples. Add 25 pL o f water to each control blank. 6- Add I mL o f 0.5 M tetrabutylammomum hydrogen sulfate, pH 10 to each tube. 7. Add 2 mL o f 0.25 M sodium carbonaie:0.25 M sodium bicarbonate to each tube. 8. Add 3 mL ethyl acetate to each tube. 9. Shake on reciprocal shaker for 20 minutes at medium speed10. Centrifuge at 3200 rpm for 20 minutes 11. Transfer a 30Q-pL aliquot o f the organic layer to a 96-well block. I 12. Evaporate the eluate to dryness under nitrogen at 35 C for approximately 30 minutes at a maximum pressure o f 50 psi. 13. Reconstitue fee dried detracts in two steps: Add 400 pL acetonitrile and vortex for 60 seconds. Add 400 pL purified water and vortex for 60 seconds. 14. Apply a cover film to the 96-well block.
Page50 of54
0 2 120VDJA_DU.DOC
000125
A.12.0 TYPICAL INSTRUMENT CONDITIONS
HPLC Conditions: Mobile Phase (gradient):
A: 10:90 methanoi:2 mM ammonium acetate
Mobile Phase Gradient Conditions:
B: 90:10 methanol:2 mM ammonium acetate
Time (min) 0.0 0.5 2.3 3.0 3.5 5.00
F unction/V alue 45% B 45% B 100% B 100% B 45% B STOP
Flow Rate Autosampler
300 pl/m in PE Series 200 (with 50-pL stainless steel
Autosampler Temperature Autosampler Needle Wash
Injection Volume HPLC Column
Typical Initial Column Pressure Autosampler Run Time
Ambient 80:20 acetonitrile:0.01% ammonium hydroxide 10 to 15 pL Betasil-C|g, 5 pm particle size, 2 mm x 50 mm 73 bar (variable) 4.3 minutes
Analyte PFOA* 9H-HDFNA
Representative Retention Time /m inutest 2.5 2.3
*Note: The PFOA chromatographic peat may show a small peak on its leading edge, which is most likely an isomeric component in the PFOA chemical standard.
Page 51 of 54
02I20VDJA_DU.DOC
000126
Mass Spectrometer Conditions:
Curtain Gas Nebulizer Gas Collision Gas TurboIonSprayTM Temperature Auxiliary Gas
UHP Nitrogen UKP Nitrogen UHP Nitrogen 350 C UHP Nitrogen at 8 L/min
Selected Reaction Monitoring:
Analyte PFOA 9H-HDFNA
Transition Monitored (0.21* Dwell Time
ra/z = 413 -> 169 m/z = 445 - 381
200 ms 200 ms
Ionization Mode Ion Spray Voltage: Declustering Potential Collision Energy MS Acquisition Time Pause Time
Negative Ion 3000 V 10V 25.5 eV 3.0 minutes 2 ms
Calibrate the mass axis o f the instrument by infusion o f Rhodapex mass calibration solution at a flow rate o f 10 pL/rnin. Optimize the sensitivity o f the instrument sing an infusion o f a 04 pg/mL mixed solution o f PFOA and 9H-HDFNA at 10 p l/min into a flow of 190 pL/min o f mobile phase using the initial gradient condition (45% eluent B). Mass spectrum peak widths should be approximately 0.6 amu at half-height in both single MS and MS/MS modes.
One set o f calibration standards was injected at the beginning and one set o f calibration standards was injected at the end o f each analytical run. Additional control blank and zero samples may be used to monitor possible carryover or
Page 52 o f 54
02120VDJA_DU.DOC
000X27
contamination. Three replicates o f each background subtraction zero sample are placed before each standard curve.
All polytetrafluoroethylene (PTFE)-containing materials are removed from the LC pumps and replaced with either stainless steel or PEEK. This is done to remove these items as potential sources o fPFOA.
A.13.0 SYSTEM SUITABILITY
System suitability criteria were established during the validation based upon the response o f the 0.05 pg/mL PFOA calibration standards in human serum. The system suitability samples for sample analysis are prepared for each analytical ran, and injections are made into die system prior to sample analysis and subject to the following criteria. The peak shape and retention times for the analyte and internal standard should be comparable to that observed during the validation. The signal (peak height) for the 0.05-pg/mL system suitability samples should be a minimum of 400 cps in order to proceed with sample analysis.
A.14.0 CALCULATIONS Calculated concentrations are based on peak area ratios o f analyte to IS. The peak area for the analyte transition is divided by the peak area for the corresponding internal standard transition. The calibration curves are fit using a 1/x2quadratic regression.
Background Subtraction procedure: 1. Background subtraction zero samples prepared with the control human serum lot
used to prepare the standard curves are identified as ZERO_HSTD, and those prepared with the control human serum lot used to prepare the QC samples are identified as ZEROJIQC. Three replicates are placed before each standard curve. 2. After analysis, the data are processed in MacQuan, and the mean; area ratio (area PFQA/area IS) from the six zero samples in each o f the two control serum lots is
Page S3 o f54
02I20VDJA_DU.DOC
000128
calculated in Excel from the data in the MaeQuan results table (.DAT file). This yields two mean arearatios. If the coefficient o f variation (CV) o f a mean ratio is greater than 20%, then outlier values are removed and the mean is re-calculated. The Excel data are saved electronically (_ZERO file), and a hardcopy o f the spreadsheet, showing the mean area ratios, is included in the analytical run documentation. 3. The applicable mean area ratios ate subtracted from the raw area ratios for the standards and QC samples. Background subtraction calculations are performed in Excel using four significant figures for the mean area ratios, and again using the data in the MaeQuan results table (.DAT file). The Excel data are saved electronically (_SUB file), and a hardcopy Ofthe spreadsheet, shoying the original and the background-subtracted data, is included in the analytical run documentation. 4. The background-subtracted data are copied to the original MaeQuan results table (.DAT file) in Excel with the background-subtracted area ratios in the "Area" column o f the .DAT file. The Excel feature o f "paste values" is used. A value of 1 is placed in die "IS Area" column in the .DAT file. This will provide the correct background-subtracted area ratios for the Watson LIMS regression. The label "n/a" is entered in the analyte area column for control blanks. The Excel data are saved electronically (_SUB.DAT file) but not printed. 5. The background-subtracted data (JSUBJDAT file) are uploaded to W atson LEWS. These data will appear in the hardcopy printout from Watson LIMS (Import Information: Peak Areas). 6. The electronic files are placed into die appropriate folder for the analytical run on the file server in the LIMFILES directory.
(v, 1.4) and MaeQuan (y. 1.6). The data are formatted in the Advion laboratory information management system (Watson, v. 6.1.0.03).
PFQA concentrations can be expressed as APFO by multiplying the PFOA values by 1/salt factor.
Page 54 o f 54
02120VDJA_DU.DOC
000129
