Document mbv8J60qGkO6paJaQLr9KLKk0

DownloadRandom document
FINAL REPORT Epidemiology, 220-3W-05 Medical Department 3M Company St. Paul, MN 55144 Date: June 25,2001 Title: Identification of Fluorochemicals in Human Tissue BACK TO MAIN 3M Company EPI-00 18 Page 1 of 29 Study Start Date: March 22, 1999 Estimated Date of Final Report: June 15,2001 Protocol Number EPI-0018 IRB Approval Exempt Expedited X IRB Approval Date: 3M IRB March 10, 1999 IIAM Approval March 22, 1999 Principal Investigator: Co-investiga.tors: Geary W. Olsen, DVM, Ph.D.' Kristen J. Hansen, Ph.D. Lisa A. Clemen, M S ~ Jean M. Burris, MPH, RN' Jeffrey H. Mandel, MD, MPH Study Director: Jeffrey H. Mandel, MD, MPH 1. 3M Company, Medical Department, Building 220-3W-05, St. Paul, MN 55144 2. 3M Company, Environmental Laboratory, Building 220-2E-09, St. Paul, MN 55133 BACK TO MAIN ABSTRACT 3M Company EPI-00 18 Page 2 of 29 Analytical methods have recently been developed, using high performance liquid chromatographyJelectrospray tandem mass spectrometry (HPLC-ESMSMS), that can be used to measure specific perfluoroinated chemicals in serum and liver samples. A recently completed six-month feeding study of perfluorooctanesulfonate (PFOS, C ~ F I ~ S Oto~c-y)nomolgus monkeys reported liver to serum ratios at end of study that were approximately 1:l in the low- (0.03 mg/kg/day) and mid- (0.15 mg/kg/day) dose groups. A non-linear response with a plateau in serum levels was reported for the high dose group (0.75 mg/kg/day). These primate data suggested that serum PFOS levels may serve as an adequate marker of liver burden at levels less than 100 ppm (parts per million, ug/mL). The purpose of this investigation was to determine whether the 1:1 liver to serum ratio reported in this primate study was mirrored in PFOS levels measured in non- occupationally exposed humans. Average serum levels (30 ng/mL) in humans have been measured at levels slightly less than three orders of magnitude lower than the average low-dose group serum PFOS level (20 ug/mL) in the primate study. For the present study, all human sera and liver donor tissue were obtained through the International Institute for the Advancement of Medicine (IIAM). IIAM is a non- profit organization whose purpose is to facilitate the placement of non-transplantable human organs and tissues for biomedical research and education. Upon acceptable donor qualifications, IIAM obtained 5 ml of blood and 10 grams of liver and then froze the samples until shipment to the 3M Medical Department. A total of 31 donor samples were obtained over an 18 month time period. BACK TO MAIN All samples were analyzed for quantitative determination of four 3M Company EPI-0018 Page 3 of 29 perfluorochemicals: PFOS; perfluorosulfonamide (PFOSA; C8F17S02NH2); peffluorooctanoate (PFOA; C7F15C02'); and perfluorohexanesulfonate (PFHS; C ~ F I ~ S O ~Se-r)a. and liver samples were extracted using an ion-pairing extraction procedure. The extracts were quantitatively assayed using HPLC-ESMSMS and evaluated versus an unextracted curve. Extensive matrix spike studies were performed to evaluate the precision and accuracy of the analysis. These matrix spike studies indicated that the data can be considered to be accurate to within one standard deviation of the average fortified sample recovery. For example, the average fortified sample recovery of PFOS from human sera was 89% (SD 21%). The average fortified sample recovery of PFOS from human liver was 78% (SD 24%). A total of 31 donor samples (16 male, 15 female) were obtained over an 18 month time period. Average age of the male donors was 50 years (SD 15.6, range 5-69) and 45 years (SD 18.5, range 13-74) for the female donors. Causes of death were intracranial hemorrhage (n = 16,52%), motor vehicle accident (n = 7,23%), head trauma (n = 4, 13%),brain tumor (n = 2,6%), drug overdose (N = 1,3%) and respiratory arrest (n = 1, 3%). Serum POSranged from ~ 6 . (1limit of quantitation, LOQ) to 58.3 ng/mL with a mean of 17.7 ng/mL (95% CI 13.0 - 22.5) for the average of the 24 donor samples analyzed. In the calculation of arithmetic means, any serum PFOS values that were determined to be <LOQ were assigned values midpoint between zero and the <LOQ. Liver PFOS ranged from ~ 3 . (7LOQ) to 57.0 ng/g with a mean of 18.8 ng/g (95% CI 14.1 - 23.5) for the average of the 30 donor liver samples analyzed; likewise, any liver PFOS BACK TO MAIN 3M Company EPI-00 18 Page 4 of 29 values that were determined to be cLOQ were assigned values midpoint between zero and the <LOQ. Fifteen (50%) of the 30 donors had liver analyses cLOQ. A ranked distribution of the average PFOS serum and liver data for each of the 23 paired samples showed good correlation (Spearman's Rho = 0.41, p c .OS). In these 23 pairs, the mean liver to serum ratio was 1.3:1 (95% CI 0.9:1 - 1.7:1). Assuming the variation of the analytical error is one standard deviation, the mean 1iver:serumratio ranged from a minimum of 0.8:l (95% CI 0.5- 1.0)to a maximum of 2.1:l (95% CI 1.5 2.8). Although the study data were limited by the number of donors, it should be noted that the average serum PFOS levels determined in the present study were comparable to those reported elsewhere in nonoccupationally exposed human populations. The levels of PFOA, PFHS, and PFOSA were determined in the samples but we chose not to estimate 1iver:serumratios for these analytes as 90% of the liver samples were determined to be cLOQ. In conclusion, our data, as well as the data from the six month primate study, would suggest that a 1iver:serumratio that approximates 1:1 for the non-occupationallyexposed human population may be a reasonable assumption to use in a risk characterization assessment of PFOS. Assuming the variation of the analytical error is one standard deviation, the mean 1iver:serumratio in our study would range from a minimum of 0.8:l(95% CI 0.5- 1.0)to a maximum of 2.1:l(95% CI 1.5 - 2.8). We chose not to estimate mean 1iver:serum ratios for PFOSA, PFOA and PFHS because 90% of these liver sample analyses were cLOQ. BACK TO MAIN INTRODUCTION 3M Company EPI-00 18 Page 5 of 29 Evidence. that organic fluorine can be characterized as a component of human blood has existed for more than 30 years, although compound-specific characterization of the organic fluorine was not complete (Guy et al, 1974). Hansen et a1 (2001) have recently published an analytical method utilizing high performance liquid chromatography/electrospray tandem mass spectrometry (HPLCESMSMS) to measure specific perfluoroinated chemicals in human serum and animal serum and liver. Among 65 human serum samples purchased from biological supply companies, Hansen et a1 reported perfluorooctanesulfonate (PFOS) may be the primary contributor to the total organic fluorine levels in human sera. The average reported serum PFOS level was 28.4 ng/mL (SD 13.6, range 6.7-81.5 ng/mL). Lesser amounts of perfluorooctanoate (PFOA, mean = 6.4 ng/mL, SD = 4.8), perfluorohexanesulfonate (PFHS, mean = 6.6 ng/mL, SD = 5.1) and peffluorooctanesulfonamide (PFOSA, mean < 1.6 ng/mL) were also determined. Hansen et a1 calculated that the combined organic fluorine content of these four fluorochemicals approximated the total organic fluorine measured by Guy et a1 in the 1970's. Additional analyses of human serum from nonoccupationally exposed population samples (3M Company, 1999) has collaborated the serum PFOS findings reported by Hansen et al. During their serum method development, Hansen et a1 (2001) also developed comparable methods for the an'alysis of fluorochemical levels in animal liver. PFOS has been demonstrated to primarily distribute into the liver (Johnson et al, 1979) and enter the enterohepatic circulation. The scope of work described by Hansen et a1 did not include the analysis of fluorochemicals in any human livers. BACK TO MAIN 3M Company EPI-0018 Page 6 of 29 In a recently completed six-month feeding study of PFOS to cynomolgus monkeys, liver to serum ratios at end of study were reported to be approximately 1:1 in the low- (0.03 mg/kg/day) and mid- (0.15 mg/kg/day) dose groups (Seacat et al, 2001). A linear increase in serum PFOS was observed throughout the course of study for the low- and mid-dose groups. In the primate study, PFOS was measured with comparable analytical techniques to those described by Hansen et a1 (2001). End-of-study low- and mid-dose groups' serum PFOS levels averaged 15 and 75 ppm, respectively, and liver PFOS levels averaged 20 and 60 ppm, respectively. No toxicologically significant effects were observed in the low- and mid-dose groups. A non-linear response with a plateau in serum levels was reported for the high dose group (0.75 mg/kg/day); decreased cholesterol was the most sensitive clinical response, occurring only at serum PFOS levels > 100 ppm. End-of-study serum and liver levels for the high-dose group averaged 172 pprn and 334 ppm, respectively. These studies suggested that serum PFOS levels may serve as an adequate marker of liver burden at levels less than 100 ppm . The purpose of this investigation was to determine whether the 1:1 liver to serum ratio observed in the primate study in the low- and mid-dose groups also occurrs in individual, non-occupationally exposed humans. Previous measurements by Hansen et a1 (2001) and others (3M Company 1999), have indicated that the average serum PFOS levels in nonoccupationally exposed populations may be approximately three orders of magnitude lower than the average low-dose serum PFOS level (20 ppm) measured in the six month primate study by Seacat et al. BACK TO MAIN METHODS 3M Company EPI-00 18 Page 7 of 29 Procurement of Human Tissue All human sera and liver donor tissue were obtained through the International Institute for the Advancement of Medicine (IIAM). IIAM is a non-profit organization whose purpose is to facilitate the placement of non-transplantable human organs and tissues for biomedical research and education (Bode 1997). IIAM is a division of the Pennsylvania Regional Tissue and Transplant Bank. This is an inspected and accredited program which meets or exceeds all of the Food and Drug Administration (FDA) and America Association of Tissue Banks (AATB) standards. IIAM works with a network of organ procurement organizations and tissue banks throughout the United States from which tissues are received from postmortem donors. IIAM abides by U.S. legislation governing the consent process for organ and tissue donation in accordance with the Uniform Anatomical Gift Act and Title I11 of the National Organ Transplant Act. IIAM ensures and guarantees donor confidentiality per Title 45, Part 46 of the U.S. Code of Federal Regulations. All donors were tested for HIV-1 and HIV-2 antibodies, hepatitis B surface antigen, hepatitis C antibody, HTLV-1 antibody, syphilis, and hepatitis B core antibody (IgG plus IgM). In addition, one or more of the following tests were sometimes done: CMV antibody, hepatitis B surface antibody, hepatitis B core IgM and HIV p24 antigen. Results that suggested the donor was infectious precluded the distribution of the tissue. The Medical Department at the 3M Company completed an agreement with IIAM for human biological material for research use which included a letter of approval from the 3M Institutional Review Board. Upon acceptable donor qualifications, IIAM BACK TO MAIN 3M Company EPI-0018 Page 8 of 29 obtained 5 ml of blood and 10 grams of liver and then froze the samples until shipment to the 3M Medical Department. Analysis of Samples All samples were analyzed for quantitative determination of four perfluorochemicals: PFOS (CgFl7SO3-),PFOSA (CgFl7S02NH2), PFOA ( C ~ F I ~ C Oa~nd. ) PFHS ( C ~ F I ~ S O ~S-e)ra. and liver samples were extracted using an ion-pairing extraction procedure, as detailed by Hansen et a1 (2001). The extracts were quantitatively assayed using WLC-ESMSMS and evaluated versus an unextracted curve. The difficulties presented by background levels of fluorochemical in samples of "blank" test matrix were circumvented by utilizing rabbit sera as a surrogate matrix for extraction blanks. The linear range was determined by analyzing duplicate curves over a wide range (approx. 0.005-1.OO ug/mL). Each sample was extracted and analyzed in duplicate. Extensive matrix spike studies were performed to evaluate the precision and accuracy of the analysis. Thirty one matrix spikes were prepared in human sera and 41 spikes were prepared in human liver. Spikes were prepared to approximate the levels of PFOS determined in the samples. It is not possible to verify true recovery of endogenous analyte from tissues without radiolabeled reference material. The only measurement of accuracy available, matrix spike studies, indicated that the data can be considered to be accurate to within one standard deviation of the average fortified sample recovery. For PFOS, the average fortified sample recovery in human sera was 89% (SD 21%). The average fortified sample recovery of PFOS from human liver was 78% (SD 24%). For the remaining analytes, PFOSA, PFHS and PFOA the average fortified sample recovery BACK TO MAIN 3M Company EPI-0018 Page 9 of 29 in human sera were 86% (SD 30%), 55% (SD 26%) and 89% (SD 33%), respectively. Likewise, the average fortified sample recovery in human liver for these three analytes were 83% (SD 20%), 42% (SD 30%) and 67% (SD 30%), respectively. Results for sera are reported as ng/mL and for liver as ng/g. RESULTS Serum and/or liver samples were obtained from 3 1 donors over an 18 month time period (Table 1). Both serum and liver tissue were harvested from 23 donors; 7 donors contributed liver tissue only and 1 donor contributed serum only. Average age was 50 years (SD 15.6, range 5-69) for male donors (n = 16) and 45 years (SD 18.5, range 13-74) for female donors (n = 15). Causes of death as provided by IIAM were intracranial hemorrhage (n = 16,52%), motor vehicle accident (n = 7,23%), head trauma (n = 4, 13%), brain tumor (n = 2,6%), drug overdose (N = 1,3%) and respiratory arrest (n = 1, 3%). Serum and liver results for PFOS are provided in Table 2. Serum PFOS levels determined to be less than the limit of quantitation (<LOQ) were assigned a value midpoint between zero and the LOQ. Mean serum PFOS level was 17.7 ng/mL (95% CI 13.0-22.5;range ~ 6 . -158.3 ng/mL) for the average of the 24 serum donors' samples analyzed. The geometric mean for serum PFOS was 14.7 ng/mL (95% CI 11.1 - 19.4). Similarly, PFOS liver samples that were determined to be <LOQ were assigned a value midpoint between zero and the LOQ. Fifteen (50%) of the 30 liver donors had liver PFOS results at c:LOQ. The mean liver PFOS was 18.8 ng/g (95% CI 14.1 - 23.5; range ~ 3 . -757.0 ng/g) for the average of the 30 liver donors' samples analyzed. The geometric BACK TO MAIN 3M Company EPI-00 18 Page 10 of 29 mean for liver PEOS was 15.2 ng/g (95% CI 11.9 - 19.6). Mean PFOS levels between male and female donors for serum (male = 18.2 ng/mL; female = 17.2 ng/mL) or liver (male = 19.2 ng/g; female = 18.4 ng/g) were similar. No associations were observed between measured PFOS levels and age (data not shown). A ranked distribution of the average PFOS serum and liver data for each of the 23 paired samples (sera and liver from the same individual) showed good correlation (Spearman's Rho = 0.41, p c .OS). The mean liver to serum ratio was 1.3:l (95% CI 0.91.7; range 0.2 - 3.7) for these 23 pairs. There was no significant difference between genders. Of the 13 male donors with paired samples, the mean liverserum ratio was 1.3:1 (95% CI 0.8:1 - 1.9:1 ). Their mean serum and liver levels were 18.2 ng/mL (95% CI 10.2 - 26.2) and 20.8 ng/g (95% CI 11.4 - 30.l), respectively. Of the 10 female donors with paired samples, their mean 1iver:serumratio was 1.3:1 (95% CI 0.6:1 2.0:l). Their mean serum and liver levels were 16.9 ng/mL (95% CI 10.0 - 23.7) and 16.3 ng/g (95% CI 9.3 - 23.2), respectively. In order to study the maximum liverserum ratio for PFOS, the average values were adjusted to accommodate the upper (liver) and lower (sera) quantitative limits described by the analytical accuracy (Table 3). As stated previously, the stated accuracy for PFOS determination in sera was 89+/-21% and 78+/-28% in liver. Assuming a one standard deviation increase in liver and one standard deviation decrease in serum, the 1iver:serum ratio became 2.1:l (95% CI 1.5:l - 2.8:l). In order to examine the lower limits of the 1iver:serum ratio the average liver PFOS levels were decreased by one standard deviation deviation and the sera PFOS levels were increased by one standard deviation which resulted in a 1iver:serum ratio of 0.8: 1 (95% CI 0.5: 1 - 1.O:l). [Note: BACK TO MAIN 3M Company EPI-00 18 Page 11 of 29 Assuming a two standard deviation increase in liver PFOS levels and a two standard deviation decrease in sera PFOS levels resulted in a maximum 1iver:serumratio of 3.4:l (95% CI 2.3:1 - 4.4:1). Likewise, a two standard deviation decrease in liver PFOS levels and a two standard deviation increase in sera PFOS levels resulted in a minimum 1iver:serumratio of 0.5:1 (95% CI 0.3- 0.6). Provided in Tables 4 through 6 are the serum and liver values for the 31 donors for PFOSA, PFOA and PFHS, respectively. Unlike the results of PFOS determination where liver samples from 15 of the donors were determined to contain PFOS above the LOQ in at least one of the duplicate measurements, values greater than the LOQ for PFOSA, PFOA, and PFHS were sparse. A larger number of values less than the LOQ for PFOSA, PFOA, and PFHS were also determined in the analysis of sera samples. Serum values ranged from cLOQ (4.3)to 22.1 ng/mL for PFOSA ,< LOQ (c3.0)to 7.0ng/mL for PFOA and c LOQ (c1.2)to 5.9ng/mL for PFHS. Again, assuming the midpoint value between zero and the LOQ serum value for cLOQ samples, the mean serum PFOSA level was 4.5ng/mL (95% CI 2.6 - 6.5). The geometric mean for serum PFOSA was 3.0 ng/mL (95% CI 2.0 - 4.4). The mean serum PFOA level was 3.1 ng/mL (95% CI 1.94.3)with a geometric mean of 2.5ng/mL (95% CI 1.9- 3.2). The mean serum PFHS level was 2.4ng/mL (95% CI 1.7- 3.0)with a geometric mean of 1.8ng/mL (95% CI 1.3 - 2.6). We did not provide any liver/serum ratios in Tables 3 through 5 because more than 90% of the individual liver samples were determined to be cLOQ. BACK TO MAIN DISCUSS ION 3M Company EPI-00 18 Page 12 of 29 The purpose of this study was to estimate a mean 1iver:serum ratio in human donors from a nonoccupationally exposed population. Our data indicated that at non- occupationally exposed population serum levels (approximately 30 ng/mL), the mean 1iver:serum ratio for PFOS was 1.3:1 (95% CI 0.9 - 1.7) which is similar to the 1:1 ratio reported by Seacat et a1 (2001) in the low- and mid- dose groups in their six month primate study. Serum PFOS levels measured at the end of this primate study were considerably greater for the low- and mid-dose groups (average 20 and 75 ug/mL, respectively) than what has been determined in sera samples from nonoccupationally exposed human populations. Further refinement of this ratio for humans is unlikely without more precise analytical techniques that can consistently measure PFOS at lower levels as well as the other fluorochemical analytes that were measured this study. Although the study data were limited by the number of donors (raising the question about the representiveness of the sample set), it should be noted that the average serum PFOS levels determined in the present study were comparable to those reported elsewhere (Hansen et al, 2001; 3M Company, 1999). In summary, our study data, as well as the results from the recently completed six month primate study (Seacat et al, 2001), suggest that a liverserum ratio of 1:1 for the nonoccupationally exposed human population may be a reasonable approximation to use in a risk characterization process for PFOS. Assuming the variation of the analytical error is one standard deviation, the mean 1iver:serumratio in our study would range from a minimum of 0.8:I (95% CI 0.5 - 1.O) to a maximum of 2.1: 1 (95% CI 1.5 - 2.8). We BACK TO MAIN 3M Company EPI-OO 18 Page 13 of 29 chose not to estimate mean 1iver:serumratios for PFOSA, PFOA and PFHS because 90% of these liver sample analyses were <LOQ. BACK TO MAIN 3M Company EPI-00 18 Page 14 of 29 REFERENCES 3M Company (1999). Perfluorooctane Sulfonate: Current Summary of Human Sera, Health and Toxicology Data. February 5, 1999 (unpublished report). Bode DC (1997). Organization and activities of human tissue banks in the U.S. with special reference to the International Institute for the Advancement of Medicine. UNOS Bulletin. January/February 1997. Guy WS, Taves DR, Brey WS. Organic fluorocompounds in human plasma: prevalence and characterization. Biochemistry Involving Carbon-Fluorine Bonds; ACS Symposium, 1976; pp 117-134. Hansen KJ, Clemen LA, Ellefson ME, Johnson HO (2001). Compound-specific, quantitative characterization of organic fluorochemicals in biological matrices. Environ Sci Techno1 35:766-770. Johnson JD, Gibson SJ, Ober RE (1979). Extent and route of excretion and tissue distribution of total carbon-14 in rats after a single intravenous dose of FC-95-14C. St. Pau1:Riker Laboratories. Seacat AM, Thornford PJ, Hansen KJ, Olsen GW, Case MT, Butenhoff JL (2001). Investigation of the no observeable effect level for perfluorooctanesulfonic acid potassium salt in cynomolgus monkeys after twenty-six weeks of oral dosing and one year of recovery. Toxicol Sciences (submitted). Individual M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 Table 1 BACK TO MAIN 3M Company EPI-00 1 8 Page 15 of 29 Demographic and Cause of Death Distribution of 31Human Donors Gender Race Tissue Available Age Cause of Death Serum Liver Male White 5 MVA No Yes Male White 34 Brain tumor Yes Yes Male Hispanic 38 MVA Yes Yes Male Black 41 Head trauma Yes Yes Male White 43 MVA** Yes Yes Male Hispanic 46 ICH Yes Yes Male White 52 Head trauma Yes Yes Male White 55 ICH Yes Yes Male White 56 MVA Yes Yes Male White 56 Head trauma Yes Yes Male mspanic 57 ICH Yes Yes Male White 58 ICH No Yes Male Asian 60 ICH No Yes Male White 64 ICH* Yes Yes Male White 64 ICH Yes Yes BACK TO MAIN Table 1 (continued) 3M Company EPI-0018 Page 16 of 29 M16 Male White 69 ICH Yes Yes Fl Female White 13 Brain tumor No Yes F2 Female White 18 MVA No Yes F3 Female White 26 Drug overdose Yes Yes F4 Female White 30 Respiratory arrest Yes Yes F5 Female White 37 ICH Yes Yes F6 Female White 38 MVA No Yes F7 Female White 42 ICH Yes No F8 Female Unknown 44 ICH Yes Yes F9 Female White 47 ICH Yes Yes FlO Female White 51 MVA Yes Yes Fll Female White 58 ICH Yes Yes F12 Female White 60 ICH Yes Yes F13 Female White 66 ICH Yes Yes F14 Female White 70 Head trauma Yes Yes F15 Female White 74 ICH No Yes *ICH = Intracranial hemorrhage **MVA = Motor vehicle accident Individual M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 BACK TO MAIN Table 2 PFOS Serum, Liver and Liver Serum Ratios for 3 1 Human Donors Serum (ng/ml) Analysis Mean Liver (ng/g) Analysis Mean No serum - <LOQ (18.4) <LOQ (18.4) <LOQ (18.4) 13.9 13.6 <LOQ (4.5) <4.5 13.2 4.5 9.8 10.3 <LOQ (18.4) <LOQ (18.4) 10.8 <LOQ (18.4) 15.6 16.0 16.4 13.4 15.7 18.0 15.5 16.0 16.5 36.4 40.0 43.6 20.4 19.5 <LOQ (18.4) <LOQ (18.4) 18.6 <LOQ (18.4) 9.4 9.5 <LOQ (36.8) <LOQ (36.8 9.5 <LOQ (36.8) 7.4 7.6 7.7 22.7 21.7 20.7 25.3 24.9 24.5 14.6 15.4 16.2 14.5 14.2 13.9 12.4 11.4 10.4 3M Company EPI-00 18 Page 17 of 29 Liver/Serum Ratio - <0.3: 1 <1.8: 1 1.0:1 2.5: 1 <1.0:1 <3.9: 1 2.9: 1 0.6: 1 0.8: 1 M11 M12 M13 M14 M15 M16 F1 F2 F3 F4 F5 F6 11.9 13.9 No serum No serum 58.3 55.7 29.4 25.9 8.1 6.9 No serum No serum 14.2 14.3 31.2 28.5 19.3 20.1 No serum BACK TO MAIN Table 2 (continued) 12.9 <LOQ (18.4) <LOQ (18.4) <LOQ (18.4) - 18.9 18.7 18.5 - <LOQ (18.4) <LOQ (18.4) <LOQ (18.4) 57.0 47.0 44.2 41.4 27.6 57.0 53.8 50.6 7.5 <LOQ (36.8) <LOQ (36.8) <LOQ(36.8) - <LOQ (36.8) <LOQ (36.8) <LOQ (36.8) - 17.2 17.1 16.9 14.2 <LOQ (32.6) <LOQ (32.6) <LOQ (32.6) 29.8 <LOQ (32.6) <LOQ (32.6) <LOQ (32.6) 19.7 <LOQ (7.3) <LOQ (7.3) <LOQ (7.3) - 23.6 <LOQ (18.4) 3M Company EPI-0018 Page 18 of 29 <1.4: 1 - <0.8: 1 1.9:1 <4.9: 1 - <2.3: 1 <1.1:1 <0.4: 1 BACK TO MAIN Table 2 (continued) 3M Company EPI-0018 Page 19 of 29 F7 20.5 20.6 No liver - 20.7 F8 22.9 24.7 26.5 26.5 23.7 <1 .o: 1 20.9 F9 3.0 2.7 <LOQ (18.4) <LOQ (18.4) <6.8:1 2.4 <LOQ (7.3) F10 <LOQ (6.1) <6.9 6.9 <LOQ (36.8) <LOQ(36.8) <LOQ (36.8) <5.3:1 F11 26.0 21.3 16.6 25.2 27.0 1.3:l 28.9 F12 8.2 7.4 <LOQ (18.4) <LOQ (18.4) <2.5:1 6.5 <LOQ (18.4) F13 15.8 15.6 <LOQ (16.3) <LOQ (16.3) <1.1:1 15.3 <LOQ (16.3) F14 24.5 28.4 32.2 F15 No serum - 39.3 33.5 <1.2:1 27.8 42.2 42.5 42.7 Limit Described Average Lower Limit I Upper Limit I Lower Limit I1 Upper Limit I1 Table 3 BACK TO MAIN 3M Company EPI-0018 Page 20 of 29 Liver [PFOS]:Serum [PFOS] Ratio Analysis Alterations Liver [PFOS]:Serum [PFOS] Ratio 95% C.I. of Ratio None 1.3:l 0.9:1 - 1.711 Liver PFOS decreased 1 SD* Serum PFOS increased 1 SD** 0.8: 1 0.5:1 - 1.O:l Liver PFOS increased 1 SD 2.1:l Serum PFOS decreased 1 SD 1.5:1 - 2.8:1 Liver PFOS decreased 2 SD Serum PFOS increased 2 SD 0.5: 1 0.3~1- 0.6~1 Liver PFOS increased 2 SD 3.4:l Serum PFOS decreased 2 SD 2.3: 1 - 4.4:1 *Liver [PFOS] SD = 24% **Serum [PFOS] SD = 21% Individual MI M2 M3 M4 M5 M6 M7 M8 M9 M10 BACK TO MAIN Table 4 PFOSA Serum and Liver Analyses for 3 1 Human Donors Serum (nglml) Analysis Mean Liver (ng/g) Analysis Mean No serum - cLOQ (18.8) cLOQ (18.8) cLOQ (18.8) cLOQ (1.3) cLOQ (1.3) CLOQ (1.3) cLOQ (7.5) cLOQ (7.5) cLOQ (7.5) 2.0 2.7 cLOQ (18.8) cLOQ (18.8) 3.4 cLOQ (18.8) 9.9 9.6 cLOQ (7.6) cLOQ(7.6) 9.3 cLOQ (7.6) cLOQ (1.3) cLOQ (1.3) cLOQ (1.3) cLOQ (7.5) cLOQ (7.5) cLOQ (7.5) 6.4 6.1 cLOQ (18.8) cLOQ (18.8) 5.9 cLOQ (18.8) cLOQ (3.1) cLOQ (3.1) cLOQ (3.1) cLOQ (18.8) cLOQ (18.8) cLOQ (18.8) 4.4 6.8 <LOQ (18.8) CLOQ (18.8) 9.2 cLOQ (18.8) cLOQ (3.3) cLOQ (3.3) CLOQ(3.3) cLOQ (7.9) cLOQ(7.9) cLOQ (7.9) 12.8 12.9 cLOQ (7.5) cLOQ (17.5) 13.0 cLOQ (7.5) 3M Company EPI-00 18 Page 21 of 29 MI 1 M12 M13 5.1 5.1 No serum No serum Table 4 (continued) 5.1 - BACK TO MAIN <LOQ (18.8) eLOQ (18.8) <LOQ (19.6) <LOQ (19.6) <LOQ (18.8) <LOQ (18.8) <LOQ (18.8) <LOQ (19.6) <LOQ (18.8) 3M Company EPI-00 18 Page 22 of 29 M14 3.6 3.7 <LOQ (7.5) <LOQ (7.5) 3.8 <LOQ (7.5) M15 2.9 3.2 <LOQ (19.6) <LOQ (19.6) 3.5 <LOQ (19.6) M16 <LOQ (3.1) <LOQ (3.1) <LOQ (18.8) <LOQ (18.8) <LOQ (3.1) <LOQ (18.8) Fl No serum - <LOQ (18.8) <LOQ (18.8) <LOQ (18.8) F2 No serum - <LOQ (19.6) <LOQ (19.6) <LOQ (19.6) F3 <LOQ (1.3) d.3 <LOQ (7.9) eLOQ (7.9) 1.3 <LOQ(7.9) F4 2.2 2.1 eLOQ(7.9) <LOQ (7.9) 2.0 <LOQ (7.9) F5 <LOQ (1.3) d O Q (1.3) <LOQ (7.5) <LOQ (7.5) <LOQ (1.3) dOQ(7.5) F6 No serum eLOQ (18.8) <LOQ (18.8) <LOQ (18.8) Table 4 (continued) BACK TO MAIN 3M Company EPI-00 18 Page 23 of 29 F7 2.5 2.8 No liver 3.1 F8 7 -9 10.0 <LOQ (18.8) <LOQ (18.8) 12.1 <LOQ (18.8) F9 1.6 1.5 6.3 <LOQ (7.5) 1.5 <LOQ (7.5) FIO (LOQ (3.1) <LOQ (3.1) <LOQ (18.8) <LOQ (18.8) <LOQ (3.1) <LOQ (18.8) Fll 7.5 6.5 <LOQ (19.6) <LOQ (19.6) 5.5 <LOQ (19.6) F12 5.0 4.5 (LOQ (18.8) <LOQ (18.8) 4.0 <LOQ (18.8) F13 <LOQ (3.3) <LOQ (3.3) dOQ(7.9) <LOQ (7.9) <LOQ (3.3) <LOQ (7.9) F14 17.8 20.0 <LOQ (18.8) <LOQ (18.8) 22.1 <LOQ (18.8) F15 No serum <LOQ (19.6) <LOQ (19.6) <LOQ (19.6) Individual M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 BACK TO MAIN Table 5 PFOA Serum and Liver Analyses for 31 Human Donors Serum (ng/ml) Analysis Mean No serum - Liver (ng/g) Analysis Mean <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) 2.1 2.1 <LOQ (5.4) <LOQ (5.4) 2.0 2.5 <LOQ (3.0) <LOQ (3.0) <LOQ (3.0) <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) 15.6 14.7 13.8 50.0 46.9 43.8 <LOQ (6.0) <LOQ (6.0) <LOQ (6.0) <LOQ (35.9) <LOQ (35.9) <LOQ(35.9) 6.7 6.1 <LOQ (17.9) <LOQ (17.9) 5.5 <LOQ (17.9) <LOQ (3.0) <LOQ (3.0) <LOQ (3.0) <LOQ (17.9) CLOQ (17.9) <LOQ (17.9) <LOQ (3.0) <LOQ (3.0) <LOQ (3.0) <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) 3.3 3.2 <LOQ (18.7) CLOQ (18.7) 3.1 <LOQ (18.7) <LOQ (3.0) <LOQ (3.0) <LOQ (3.0) <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) 3M Company EPI-00 18 Page 24 of 29 MI 1 <LOQ (3.0) (LOQ (3.0) Table 5 (continued) <LOQ (3.0) BACK TO MAIN <LOQ (17.9) d O Q (17.9) <LOQ (17.9) 3M Company EPIr0018 Page 25 of 29 M12 No serum <LOQ (18.7) (LOQ (18.7) <LOQ (18.7) M13 No serum <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) M14 7 .O <7.0 cLOQ(35.9) <LOQ(35.9) <LOQ (6.0) <LOQ(35.9) M15 5.6 5.3 (LOQ (18.7) <LOQ (18.7) 4.9 <LOQ (18.7) M16 cLOQ (3.0) <LOQ (3.0) <LOQ (17.9) <LOQ (17.9) <LOQ (3.0) (LOQ (17.9) F1 No serum <LOQ (17.9) (LOQ (17.9) (LOQ (17.9) F2 No serum <LOQ (18.7) <LOQ (18.7) <LOQ (18.7) F3 <LOQ (3.1) <LOQ (3.1) <LOQ (18.7) <LOQ (18.7) <LOQ (3.1) <LOQ (18.7) F4 <LOQ (3.1) <LOQ (3.1) dOQ(18.7) <LOQ (18.7) <LOQ (3.1) <LOQ (18.7) F5 3.0 <3 .O <LOQ (17.9) <LOQ (17.9) 4 O Q (3.0) (LOQ (17.9) F6 No serum <LOQ (17.9) <LOQ (17.9) <LOQ (17.9) BACK TO MAIN Table 5 (continued) 3M Company EPI-00 18 Page 26 of 29 F7 3.4 c3.4 No liver - cLOQ (3.0) F8 cLOQ (3.0) <LOQ (3.0) cLOQ (17.9) cLOQ (17.9) <LOQ (3.0) cLOQ (17.9) F9 <LOQ (3.0) cLOQ (3.0) <LOQ(35.9) (LOQ (35.9) 4 O Q (3.0) <LOQ (17.8) FlO cLOQ (3.0) cLOQ (3.0) <LOQ (17.9) cLOQ (17.9) <LOQ (3.0) <LOQ (17.9) Fll 3.8 3.1 cLOQ (18.7) cLOQ (18.7) 2.5 (LOQ (18.7) F12 cLOQ (3.0) cLOQ (3.0) <LOQ (17.9) cLOQ (17.9) <LOQ (3.0) <LOQ (17.9) F13 3.5 3.7 <LOQ (18.7) cLOQ (18.7) 4.0 <LOQ (18.7) F14 5.1 5.5 <LOQ (17.9) cLOQ (17.9) 5.9 cLOQ (17.9) F15 No serum - <LOQ (18.7) cLOQ (18.7) (LOQ (18.7) Individual M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 BACK TO MAIN Table 6 PFHS Serum and Liver Analyses for 31Human Donors Serum (ng/ml) Analysis Mean Liver (ng/g) Analysis Mean No serum - cLOQ (3.7) cLOQ (3.7) cLOQ(3.7) 4.1 4.1 cLOQ (3.4) cLOQ(3.4) 4.0 1.4 2.5 3 .O cLOQ (5.7) <LOQ (5.7) 2.8 <LOQ (5.7) cLOQ (3.7) <LOQ (3.7) Not analyzed cLOQ(3.7) - 3.1 2.9 <LOQ (3.4) cLOQ(3.4) 2.6 cLOQ (3.4) 0.8 0.7 cLOQ(3.7) cLOQ (3.7) 0.6 <LOQ (3.7) 4.9 4.8 iLOQ (18.5) cLOQ (18.5) 4.8 cLOQ (18.5) 0.4 0.4 cLOQ (3.7) cLOQ (3.7) 0.4 cLOQ (3.7) 3.2 3.3 <LOQ (3.7) cLOQ (3.7) 3.4 cLOQ (3.7) cLOQ (1.2) cLOQ (1.2) cLOQ (1.2) cLOQ(7.4) <LOQ (7.4) cLOQ (7.4) 3M Company EPI-00 18 Page 27 of 29 MI 1 M12 M13 M14 M15 M16 F1 F2 F3 F4 F5 F6 3.7 3.6 No serum No serum 3.7 3.3 5.9 5.3 d O Q (3.1) <LOQ (3.1) No serum No serum 1.4 1.5 2.0 1.8 <LOQ (1.2) <LOQ (1.2) No serum Table 6 (continued) 3.7 3.5 5.6 <LOQ (3.1) 1.5 1.9 4 O Q (1.2) BACK TO MAIN <LOQ (3.7) dOQ(3.7) <LOQ (3.7) cLOQ (3.7) <LOQ (3.7) <LOQ (3.7) <LOQ (3.4) cLOQ (3.4) 7.5 8.9 <LOQ (18.5) 4 O Q (18.5) cLOQ (18.5) <LOQ (18.5) 4 O Q (3.7) <LOQ (3.7) <LOQ (7.4) d O Q (7.4) <LOQ (7.4) <LOQ (7.4) <LOQ (7.4) <LOQ (7.4) dOQ(3.7) <LOQ (3.7) <LOQ (3.7) cLOQ(3.7) <LOQ(3.7) <LOQ (3.4) 8.2 <LOQ (18.5) (LOQ (18.5) <LOQ (3.7) dOQ(7.4) <LOQ (7.4) <LOQ (7.4) <LOQ (3.7) 3M Company EPI-00 18 Page 28 of 29 BACK TO MAIN Table 6 (continued) 3M Company EPI-00 18 Page 29 of 29 F7 1.1 1.2 No liver - 1.3 F8 3 .O 3.1 <LOQ (3.7) <LOQ(3.7) 3.2 <LOQ (3.7) F9 <LOQ (1.2) <LOQ (1-2) 5.6 <LOQ (7.4) <LUQ (1.2) <LOQ (7.4) FlO 4.7 4.7 <LOQ (18.5) d O Q (18.5) 4.7 <LOQ (18.5) F11 2.0 1.6 <LOQ (3.7) <LOQ(3.7) 1.3 <LOQ (3.7) F12 0.5 0.5 <LOQ (3.7) <LOQ(3.7) 0.4 <LOQ (3.7) F13 2.1 2.3 dOQ(3.7) dOQ(3.7) 2.4 <LOQ (3.7) F14 2.3 2.7 <LOQ(3.7) dOQ(3.7) 3 .O <LOQ (3.7) F15 No serum - <LOQ (3.4) dOQ(3.4) <LOQ (3.4)