Document vVgE6XDBKaO6Z53vpGx37ke4Y

.' . <iMD DuPont Haskell Laboratory AR226-3372 January 9,1997 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Prepared by: L. B. Biegel, Ph.D. Senior Research lexicologist HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Table of Contents INTRODUCTION ................................................,...................................................,.........3 I. MAMMALIAN TOXICOLOGY ...................................................................................3 1.1. Acute Toxitity Studies............................................................................................. 3 I.l.a. Acute Oral Toxitity............................................................................................ 3 I.l.b. Acute Dermal Toxicity ..........--................................................................,,....... 5 I.l.c. Acute Ocular Toxicity........................................................................................ 5 I.l.d. Acute Inhalation Toxicity................................................................................ 6 I. I.e. Acute Injection Toxitity.................................................................................. 7 12. Subchronic Toxicity Studies.................................................................................... 7 I.2.a. Subchronic Oral Toxicity................................................................................... 7 L2.b. Subchronic Inhalation Toxicity........................................................................ 10 I.2.C. Subchronic Dermal Toxitity............................................................................ 10 13. Developmental Toxicity......................................................................................... 11 1.4. Reproductive Toxidty........................................................................................... 13 1.5. Mutagemcity...........................................................................................................l3 1.6. Chrome Toxicity and Oncogenitity....................................................................... 13 H. METABOLISM.......................................................................................................... 15 n.l. Animal Studies...................................................................................................... 15 11.2. Human Exposure................................................................................................... 19 in. MECHANISMS OF ACTION................................................................................ 20 ni.l. Investigation ofC8s* Effect on the Liver............................................................ 20 ffl.2. Investigation ofC8s' Effect on Testicular Leydig Cells..................................... 21 ffl.3. Investigation ofC8s' Effect on the Pancreas....................................................... 22 IV. CLINICAL REPORTS OF HUMAN EXPOSURE.............................................. 22 V. EPIDEMIOLOGY.................................................................................................... 23 VL DISCUSSION OF ENDPOINTS............................................................................... 23 VI.1. Discussion of Target Organs............................................................................... 23 VL2. Discussion of Differences in Species-Specific Sensitivities............................... 24 VL3. Tumors Associated with C8 in the Rat................................................................ 24 VL3.a. Significance ofC8-Induced Rodent Tumor to Human Risk......................... 24 VIL 25 SUMMARY............................................................................................................. REFERENCES................................................................................................................. 27 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRYNO. 3825-26-1 INTRODUCTION This document is a Hazard Characterization of C8 for human health. C8 is also known as ammonium perffuorooctanoate (APPO; CAS ft- 3825-26-1) and is the prfmaiy ingredient in PC-143 FLUORAD Brand Fluorochemical Surfactant. Within this document the chemical will be referred to as C8. However, it is acknowledged that many of die studies discussed actually tested (he product FC143, which is a mixture of several straight-chain perfluorocarboxylic acids containing approximately 93.0-97.0% C8. I. MAMMALIAN TOXICOLOGY LI. Acute Toxidty Studies 1.1.a. Acute Oral Toxidty Numerous acute oral toxieily studies, in several species (rats, mice, guinea pigs, dogs), have been conducted with C8 (see Table 1-1). The results of the various studies have been consistent in their results. Administration of a single dose of 12 mg/kg to 3 rats produced no clinical signs oftoxicity. Studies demonstrate that newborn and older adult rats appear to be more sensitive than weanlings and young adults. Additionally, while mice and rats appear to be equally sensitive to the acute toxicity ofC8, guinea pigs are more sensitive than mice or rats. m the rat, acute oral exposure generally results in enlarged livers, elevations of liver enzyme levels, gastrointestinal irritation, and weight loss. C8 is considered to have moderate acute oral toxidty. In addition to the numerous studies listed below, several other studies were conducted which investigated the effects ofC8 alone or on animals pre-exposed to other chemicals or drugs. Pre-treatment of rats with phenobarbital sodium or proadifen hydrochloride does not result in an alteration of the LDso of C8 (478 mg/kg). Pre or postdosing with Dowex 1-X2-C 1 Ion Exchange Resin at 1000 mg/kg reduced the mortality compared to rats dosed with C8 alone. A study was conducted to determine ifpretreatment with efhanol (a single dose of 60% or a 15% aqueous solution (v/v) in drinking water for 14 days) modifies the effects ofC8 on liver weight. This study determined mat pre-treatment wifh ethanol did not alter C8's effect on liver to body weight ratios. HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Table 1-1 SUMMARY OF ACUTE ORAL TOXICTTY STUDIES WITH C8 Study Type ALD Species Rat #/sex/dose 10 Males Dose (nig/kg) 200,480,670 Vehicle Corn oil Results mg/kg ~fs5' """ALD""""""Rat"" ""1-lMae''"""""""1Xi2:%:l;yoo73%74X^O:I680:"""""Wate7""'"" 1500,2250 """ALD""""""Rat"" -U5,o- "Rat" 'LD;,- "Rat" ' X&M aaa' filnctlon''' Rat"' of age -"-lMa1e-""-"-'"""l:0;r:<2-J;3:475T,26740766:7?^ 130,170.200,300,450,670,2250 "" i00; 2T5',46'4.T006.'2f56'' "2-0"0,"400;450:500:670:1Wtf 10 Weanling males 10 Weanling females 10 Young adull females 10 Mature adull males 10 Mature adull females 10 Newborn males 10 Newborn females 350,400,450.525,670,710 350,400,450,670 350.425,500,670 200,240,300,350,400,500,720 225,350,400,450.670 130.200,240,280,330,370 130,180,200,220,240,280,320 "Acetoiic(49%)' Corn oil (60%) ""ConioK"" Corn oil 'MalesFBBo^emaies "Mate:' 470"" Feiniifes 573 580 453 336 343 243 258 "Low castraiai vs" "Rat" intact adults 10 intact males 200,480,670 Corn oil 439 10 intact females 200,480,670 491 10 orchidectomizcd males 200,480,670 459 10 ovariectomized females 200,480,670 400 --LD-- Rat --L-Dffl--- """Rat" '400"50b"6'50" '"fioinoll" "478" "390- --U5;,--- Mice .................^.......................^.^.^.^.^^^^ ""Liver""" """Dog" function ----------3 Males-"-- " """"""45072b-0""""""""""""^oi;iatea""""Le%arit450-mg^-oyT8no Elevated OPT and GOT wh normalized within I week at LPM' " " " 'Qyineayig' _____--lo------- """"""I50^Do;25l);3W;4(fo7670""""""""C6mo;f""""""Malw:l76"Pcna a). Weanlings 21' days old b). Young adult 8-10 weeks old c). Mature adult = >10 weeks old d). Newborn = < 2 days old HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 LLb. Acute Dermal Toxidty Acute dermal toxidty and irritation studies in rats and rabbits have been conducted with C8. C8 is considered to be mild - moderately irritating to the skin and moderately toxic by the dermal route of exposure. Rat skin showed less irritation than rabbit and in general fhe effects were more pronounced in males than in females. In addition to dermal irritation several clinical signs of toxidty were observed in both. rats and rabbits in response to C8 exposure. These observations included body weight loss, wet and/or stained perineal area, cyanosis (rabbits only), diarrhea (rabbits only), lethargy (rabbits only), labored breathing (rabbits only), and chromodacryorrhea (rats at 7500 ing/kg) Table 1-2 SUMMARY OF ACUTE DERMAL TOXICITY/IRRTTATION STUDIES WITH C8 Study Species #/ses/dose Dose Type (ing/kg) LDso Rat 5 3000,5000. 7500 "Skm'ASsoiptiM" "' Rat"'" 'TFemaies"'" "5006"and"7560 LDso Re;suits MaleLD5o==6959mg/kg Female LDsos=>7500mg/kg 'LDs^SOO'mg/kg" Mild skin irritation Reference HL-659-79 (Kennedy, 1985) "HL/.682;80" 'Skm'Abs6iptiM""'Rat'"'"TFeaffles'""'50Wand7'500""""" LDso Mild skin irritation LDso Rabbit 5 Males (2 at 7500) 1500,3000, 5000,7500 LDjo=4278mgAcg HQ59-79 (Kennedy, 1985) Rabbit 4 100,1000,2000 Skin Irritation Rabbit 6 500 mg on intact and abraded sites Lefbal 4/4 at 2000 3/4 at 1000 0/4 at 100 Non-irritating Rite Laboratories Report No. 09790AB0485 (Griffith and Long, 1980) 500 mg Mild-moderate iiritation at 24 hours HL-636-79 Slight-moderate irritation at 48 horns_____ I. I.e. Acute Ocular Toxicity Eye irritation studies in rabbits have been conducted with C8. C8 is considered to be moderately irritating to fhe eye. Instillation of solid C8 into rabbit eyes producer moderate comeal opacity, iritis, and conjunctivitis. These ocular effects gradually receded over time. Prompt washing of the eye reduced fhe effects and provided a more rapid recovery. In addition to the eye irritation studies that have been conducted, rats HAZARD CHARACTERIZATION FOR-HUMAN HEALTH CS EXPOSURE CAS REGISTRY NO. 3825-26-1 exposed to C8 during a 4-hour inhalation period exhibited comeal opadty and ulceration, which were microscopically evident 42 days post-exposure. Table 1-3 SUMMARY OF EYE IRRITATION STUDIES WITH C8 Species#/sex/doseDoseResultsReference _R_a_bb_it________2_______(m38.g3 )__Ih_w_a_sh_e_de_y_e ___________________H_L-_635-79 (1 unwashed. Moderate-severe comeal opacity 1 washed) Moderate iritis Moderate conjunctivitis At 21-28 days Comeal opacity Mild vascularization Washed eye Slight-moderate comeal opacity Slight-moderate conjunctivitis At 7 days Mild conjimctival redness At 14 days 'Rabbit"""" 6 unwashed""" 6 washed Normal ' 100 Unwashed eye ""'"" " ------ Moderate irritation Conjunctivitis """Biosearch.'inc'.'Report' No. T1395 Iritis Washed eye Conjunctivitis (Griffifli and Long, 1980) At 7 days 4/6 eyes were free of irritation 1.1 .d. Acute Inhalation Toxicity Acute inhalation toxicity studies in rats have been conducted with C8. Acute exposure to C8 by inhalation is considered to be highly toxic, with a 4-hour approximate I.I lethal concentration (ALC) in rats of 0.8 mg/L. At concentrations of mg/L C8 and higher, all rats died within 48 hours of exposure. At concentrations between 0.38 and 0.83 mg/L C8> rats experienced an initial weight loss following exposure and an increased liver-to-body weight ratio which returned to the high end of the normal range 42 days post-exposure. Additionally, all rats exposed to 0.81 mg/L C8 and higher showed comeal opacity and corrosion. HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Table 1-4 SUMMARY OP ACUTE INHALATION STUDIES WITH C8 Species#/sex/doseConcentrationResultsReference __R_at_______6_M_al_es_________4-_ho(umr egxp^osLur)e_to_: _____4-_ho_u_r A_L_C_= _0_.8_m_g_/L_____H_L_-1_6_0-_69_ 038,0.81,0.83,2.2,4.8,5.7 LCa, = 0.98 mg/L (Kennedy, et aL. 1986) -g^-------- """"^'"""""""'f-lioiir'eaqJosi^to-.iQ'mg/L " Nodeaibs Eye and respiratory irritation (Chif!fthaiid'Loiig7 1980) I. I.e. Acute Injection Toxidty Acute toxicity ofC8 when administered by mtraperitoneal injectfon was assessed in nrice (3M, 1979). The LDso by intraperitoneal injection in mice is 192 mg/kg. 1.2. Subchronic Toxicity Studies I.2.a. Subchronic Oral Toxicity Numerous subchronic oral toxicity studies in several species (rats, mice, and monkeys), have been conducted with C8 (see Table 1-5). The results of the various studies have been quite consistent in their results. Administration ofC8 in the diet or by daily gastric intubation produced death at concentrations of 1000 ppm and higher for rats and mice and at 30 mg/kg/day for monkeys. The primary target organ for toxic responses in all species studied is the liver. C8 produces increased liver weights, increased liver enzyme activity, hepatocellular hypertrophy, and hepatic peroxisome proliferation. In addition to the numerous studies listed below, other studies were conducted which investigated the mechanism of action ofC8. These studies are summarized in Section DI. Mechanisms of Action. HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Study Type 14-day feeding Species Mice 14-day feeding Mice 21-day feeding Mice 14-day feeding Mice 9-dose gavage Mice Rat l4^dayfeeding Rat with a 2-week recovery period 14-day feeding Rat With a 56 day recovery period Table!-; SSUMMARY OF SUBCHRONIC ORAL TOXICITY STUDIES WITH C8 #/sex/dose 5 5 5 Concentration (ppm unless specified) 10,30,100,300,1000,3000,10000 Results 100% mortality at 23000; deaths at 1000; increase weight/body weight ratio at MO 30,300,3000 6,0.01,0.03,0.1,6.3,1,3,16,30 100% mortality aOOOO; deaths at 300; weight los S 300 increased liver weight/body weight ratios a Significantlyincreased liver weight at 30" 5 5 5 6 Males 5 Males 30 6.1, l.O.iOmg/kg 0.1,1.0,10mg/kg 25% Teflon with C8 as the dispersing agent 36,300 Increased liver weight; when C8 was combined with an equal amount of nonadecafluorodecanoic acid, a similar effect was produced Weight loss, death in 10 ing/kg females, increased weight at I and lOmg/kg Weight loss, increased liver weight in 10 nig/kg m Slightlyincreased liver weights tbilowmgthe reco period Decreased bodyweiglits at 300; increased liver we the end of the feeding period at 30 and 300 and on recovery days 7 and 28 (300 ppm only; elevated b fluoride levels out to recovery day 7 (final day tes HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE ! CAS REGISTRY NO. 3825-26-1 Study Type 28-day feeding Species Rat Mice 90-day feeding Rat 90-day gavage Monkey " " '90-day'feeding'" with an 8-week recovery period "Rat' ; Table 1-5 (Con't) SUMMARY OF SUBCHRONIC ORAL TOXICITY STUDIES WITH C8 Sex (ff/sex/dose) 5 Concentration (ppm unless specified) 30,100,300,1000,3000,10000, 30000 30,100,300,1000,3000,10000, 30000 Results 100% mortality at ^ 3000; decreased body weigh 1000 and 3000 for females; increased liver weig weight ratios at 2 30 for males and 2 300 for fema 100% mortality at ^ 1000; deaths at S 30; decreas weights at ^ 30; cyanosis and muscle weakness a 3000; increased liver weight/body weight ratios a panlobular diffuse hypertrophy ofhepatocytes 55 Males 10,30.100,300,1000 3,10,30,100 nig/kg/day '"""""T"ro"3b"i6o"'"""" Decreased body weights at;; 300; increased liver at >, 300; panlobular diffuse hypertrophy ofhepato at ^ 1000 with males more affected than females; fluoride concentration increased 75 to 226 fold w higher concentrations observed in males 100% mortality at >. \ 00; deaths at ^ 30 (females decreased body weights at ^ 30; no signs oftoxic mg/kg/day; dose dependent increases in serum an fluorine levels (no apparent sex difference) "Reduced body weigh at' 100;"mcre^dpaimitoylC oxidase activity at >, 30 and transient increases at palmitoyl CoA oxidase activity returned to norma the 8 weeks of recovery; increased liver weights a hepatocellular hypertrophy at ^ 10 which was rev following the recovery period. Serum estradiol, testosterone and lutienizing hormone levels were affected by dietary exposure to C8. while estradio were slightly elevated at 100 ppm at week 5. The NOAEL = 100 ppm; the NOEL = 1 ppm HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 I.2.b. Subcfaronic Inhalation Toxidty Similar to the oral toxidty studies, inhalation exposure to C8 produces reduced body weight, increased liver weights, increases in plasma enzymes indicative of liver injury, and pathological lesions in (he liver. Measurement of the blood fluoride levels (indicative of the presence ofC8) determined that the blood half-life ofC8 in the rat is 5- 7 days following inhalation exposure. Table 1-6 SUMMARY OF SUBCHRONIC INHALATION TOXICnY STUDIES WITH C8 IN THE RAT Study Type Sex Concentration. Results Reference exposure2011,83 _1_0 ______(#/scx/dose)___m__g_/m__'D_o_s_e__re_la_te_d _de_cr_ea_s_e (inmbogd/ykwge)ig_h_t, _s_u_p_p_r_e_s_s_io_n__H_L_2_5_3-_79_ with a 42-day for 6 hours/day of body weight maintained during the 42-day recovery recovery period at 83; increased plasma enzymes indicative of liver injury present up to 28 days following the last exposure; granular degeneration ofhepatocytes; increased liver weights, no ocular effects were observed. The liver effects were not observed after 14,32, or 42 days of recovery. "lOexposure" with an 84day .recovery ____ ~ ~" 24'" 1,8','84 mg/m3 ' Deaths at 84; increased Itmg, liver and testes for 6 hours/day weights, no ocular effects observed; increased plasma enzymes indicative of liver injury; increased Kver weights at >: 8 mg/m3;panlobular and centrilobular hspatocelhilar hypertrophy and ___ necrosis. The liver effects were reversible following a 28day recovery period. Dose related presence ofC8 in the blood, which decreased with time during me recovery period but was still detectable after 84 days of recovery. NOAEL = 1 mg/m3', although 13 ppm organofluoridc was detected immediately following to 1 ____________exposure mg/m3________________ -""HL 205:81""'" (Kennedy, Hall etal., 1986) I.2.c. Subchronic Dermal Toxicity The subchronic dermal toxidty ofC8 has been studied in the rat and rabbit (see Table 1-7). Similar to the oral toxidty studies, dermal exposure to C8 produces reduced body weight, increased liver weights, increases in plasma enzymes indicative of liver injury and lesions in the liver. Measurement of blood fluoride levels (indicative of the presence ofC8) determined that me blood half-Hfe ofC8 in the rat is 5-7 days follawing dermal exposure. A comparison of the dermal exposure studies to the feeding studies leads to the conclusion mat the rates ofabsorption ofC8 by these two routes are not significantly different 10 HAZARD CHARACTERIZATION TOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Table 1-6 SUMMARY OF SUBCHRONIC INHALATION TOXICITy STUDIES WITH C8 IN THE RAT Study Type 10 dose with an 84 day recovery Species (#/sex/dose) 15 Male Rats Concentration (nig/kg) 20,200,2000 for 6 houra/day,5 days/week Results (nig/kg) Skin irritation, at >, 200; reversible reduction in body weight at >, 200; increased plasma enzymes indicative of liver injury; increased liver weights at >: 20; hepatocelhilar hypertrophy and necrosis at 20; no ocular effects observed The liver effects were generally reversible following a 42-day recovery period at $ 200. Dose related presence of C8 in the blood, which decreased with time during the recovery period but was still detectable after 42 days of recovery. Range-finder 4 Rabbits (sex not specified) 100,1000,2000 Lethal to 4 of 4 at 2000,3 of 4 at IOOO, 0 of 4 at 100 10 exposure wi1hal4 day recovery 10 Rabbits 100 for 6 hours/day, 5 days/week Reversible reduction in body weight; Blood fluorine levels were 5.4,6.8,4.6 ppm for males and 10.1,12.1, and 3.5 for females at 7, 14, and 28 days of the study, respectively. Reference HL 589-80 (Kennedy, 1985) Riker Laboratories, Report 09790AB0485, March 15.1981 1.3. Developmental Toxicity Developmental toxidty studies have been conducted in rats and rabbits (See Table 1-7). The original developmental toxidty study in rats indicated that C8 might be a teratogen in rats. However, because the results were questionable, additional studies were conducted to clarify the result. The additional studies did not confirm the original result Overall, C8 is not considered to be uniquely hazardous to the conceptus. The two areas of question were apparent lens abnormalities and skeletal alterations. In the original study, the lens alterations consisted of the following: large leas cleft, dark streak running % to % of the way through me lens; or disorganized lens fibers. In the subsequent studies, the lens alterations were determined to be an artifact created in the lens during freehand sectioning. Processing Bouin's-fixed fetal heads that were trimmed on either side of the orbit, instead of through the center of the eye, essentially eliminated this artifact. Examination of me eyes of offspring using focal illumination, indirect ophthatmoseopy, and sliuamp microscopy were also used and did not detect any CS-related eye alterations. The skeletal alterations included ossification sites on the first lumbar vertebrae in rats and 13 ribs in rabbits. Bom of these alterations are considered to represent stress-related changes indirectly related to OS-administration. 11 HA2ABD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Table 1-7 SUMMARY OF DEVELOPMENTAL/REPRODUCTIVE TOXICITY STUDIES WITH C8" Species (#/dose) Concentration Results mg/kg Rats25,50,75,100,150 mg/kgReduced maternal body weight gain and (# not specified) bygavage clinical signs of toxicity at 150; eye abnormalities at 25 and 150 Reference 3M Report M-601 (1981) "25"Ray""T60"mgA^by'gavage' 'i&ffdeaifis,'decrcasedmateinai'body"""""']^l^^^^ weight gain; no developmental toxicity or abnormalities observed. (Staples, et aL, 1984) 12 Rats" 100 mg/kg bygavage Rats 'o.o3ri3,To7i50 mg/kg' (# not specified) bygavage Maternal deaths, decreased maternal body weight gain; no alterations in postpartum viability, growth rate, or development No ocular effects observed. Maternal deaths at 150; C8 was not embryotoxic, no abnormal gross findings, no malformations'1. Fetal lens findings were observed in all groups. Determined to be a processing artifact No effect on ovaries, reproductive tract, male/female ratio, implantation sites, corpora lutea, or fetal weights. 3M Report 068 rmOHO, 1981 Rats ' "O.T4^L2,'9^72img/m1"' "Maternal deams'at 21; overt mateinal" by inhalation toxicity at 9.9; No teratogenic effects were observed in any of the exposed groups; embryo-fetal toxicity was observed at 21; processing artifacts were observed on lens'. '"HL88T81"" (Staples, etaL, 1984) 18 Rabbits 1.5,5,50 mg/kg bygavage Reduced maternal body weight gain at 50, C8 was not embiyotoxic or teratogenic' 3M Product Toxicity Sheet, May 24.1996 a. Pregnant rats were dosed by gavagc on days 6-15 of pregnancy. Pregnant rabbits were dosed by gavage on days 6-18 of pregnancy. b. Sacrificed on Day 21 of gestation. c. Pups sacrificed on day 35 postpartum. d. A significantly higher incidence of me skeletal finding "one stemabrae missing", occurred in the high- dose group. This was a minor skeletal aberration and was not considered a malformation in this study. Furthermore, me incidence of this finding did not differ from the control group or the 3 lower-level treatment groups. The incidences of skeletal findings associated with delayed ossification and rib aberrations were not different among file treatment groups and controls. e. There was a statistically significant increase in the incidence of 13 ribs in the high dose group and 13 ribs spurred in me mid-dose group. While me findings are significantly greater in the treated animals than in the controls, they are not considered to be teratogenic changes or malformations, rather they are considered to represent stress-related changes to compound administration. 12 HAZARD CHABACIBRIZATION FOR.HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 1.4. Reproductive Toxidty No information is available on the reproductive toxidfy of C8 L5. Mutagenicily K has been demonstrated that C8 is not mutagenic in a variety ofnmtagemcify tests (See Table 1-9). Study Type Table 1-9 SUMMARY OF MUTAGENICTTY STUDIES WITH C8 IN TEE RAT Study DescriptionResults Reference Mutagenicity assay Assayed in S. 'fyphimuriumCrA1535, TA1537, TA1538, and TA100) and S. cerevisfae D4 yeast, with and without metabolic activation. Negative Litton Bionedcs; LBI Project 20838, Feb. 1,1978 (Griffith and Long, 1980) In vivo mouse micromicleus 3 mice/sex were dosed with 200,400, 600,800, and 1000 mg/kg and bone man-ow was evaluated at 24,48 and 72 hours after dosing. Negative 'Chroniosomaf """Assayed for abullyto induce aberration chromosomal aberrations in CHO cells with and without metabolic activation. Negative Mammalian, cell transformation assay Assayed for ceiltransfonnauonpoientiai and cytotoxicity in C3H 1OT1/2 colony cells. LD = 50 g/niL; low cytotoxicity No evidence of cell transformation Corning Harieton, 17388-0-455, May 16,1996 Coining Hazleton, 17388-0-437, April 25,1996 University of Minnesota Environ. PamLab,T2942, April 9.1981 1.6. Chrome Toxidty and Oncogenicity The chronic toxidty and oncogemcity ofC8 has been investigated in two 2-year feeding studies in rats (see Table 1-10). 13 HAZARD CHARACTERIZATION FOR%iMAN HEALTH C& EXPOSURE CAS BEGISTRYNO. 3825-26-1 Table 1-10 SUMMARY OF CHRONIC TOXICITY AND ONCOGENICITY STUDIES WTm C8 IN RATS #/sex/dose 50 156 Males Concentration (mean daily intake, mg/kg/day) 0,30.300 ppm (0.1.5, and 15 m&'kg/day) 0,0-pair-fed, 300 ppm Reference Decreased body weight gain and food consumption, increased ataxia. Decreased RBC counts, hemoglobin, and hematocrit values. Increased liver weights, liver cell hypertrophy, degeneration and necrosis. Not considered to be carcinogenic. Riter Laboratory 0281CR001 (April mi-May 1983) Decreased body weight gain and food consumption. Increased estradiol levels. Increased incidence of liver, Leydig cell and pancreatic acinar cell adenomas. '~(CwS!^'e~aL,'i994)' DuPont MR-5686 In the original study, in-life findings consisted of a dose dependent decrease in mean body weight gain and increase in food consumption in males, and a slight treatment-related increase in me incidence ofataxia in females. No increase in mortality was observed. CS-related hematologic alteration included decreased red blood cell counts, hemoglobin and hematocrit values observed at various times throughout the 2year test period. However, me decreases in erythrocyte counts were observed early in the study and did not progress into generalized anemia. Histopamologically, CS-associated alterations were observed in me liver. These changes were characterized by increased liver weights, hypertrophy, hepatocellular degeneration, and necrosis. As with the eryfhrocyte counts, the hepatic alterations were observed early in the study and showed little progression over the remainder of the 2-year study. The incidence of tumors was relatively low, and the types of neoplasms found were not different from the tumor profiles commonly observed in geriatric rats. Hepatocellular tumors were slightly increased in the 300 ppm males, however, not to the extent that would be expected considering the morphological evidence of hepatocellular stimulation observed at the 1year necropsy. The incidence oftesticular Leydig cell adenomas (0/50,3/50, and 7/50 at 0,30, and 300 ppm, respectively) was suggestive of a compound-related effect. However, because me incidence was within the historical control range, it was not considered to be a compound-related effect. Based on the tumor incidence, types of tumors, time of tumor appearance, and the survival rate at the 2 year time point, the overall conclusion was that C8 was not carcinogenic in me rat (Riker Laboratory, 0281CR0012). However, in this original study, some of the pathological findings were equivocal (liver and Leydig cell tumors), even when evaluated by an outside laboratory, and therefore a second 2-year study was conducted to clarify some of these findings. The second study included many mechanistic endpoints to help determine the mechanism of tumor formation (DuPont MR-5686, Cook, et al., 1994). In addition to me ad Ubitunt control, a second control was pair-fed to the C8 group. Peroxisome proliferation (B-oxidation activity) and cell proliferation (BrdU, 6-day osmotic pumps) 14 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 were measured in the liver and testis. Serum hormone levels (testosterone, estradiol, lutienizmg hormone (LH), follicle stimulating hormone (FSH) and prolactin) were also measured. Interim sacrifices were performed at 3-month intervals as well as at 1 month. Increased relative liver weights were observed in me CS-treated rats. Hepatic 6-oxidation activity was also increased in the CS-treated rats at all time points. In contrast, hepatic cell proliferation was not significantly increased in the CS-treated group. C8 did not significantly alter me rate ofLeydig cell B-oxidation or Leydig cell proliferation. Moreover, me rate ofB-oxidation in Leydig cells was approximately 20-times less than the rate of hepatic B-oxidanon, irrespective of treatment Serum testosterone, FSH, prolactin, and LH levels were unchanged in the CS-treated rate when compared to me controls. There were, however, significant increases in serum estradiol levels in the C8- treated rats at 1,3,6,9,12,15,18 and 21 months. Histopathological evaluation revealed compound-related increases in liver, Leydig cell, and pancreatic acinar cell tumors in C8" treated rats. Based on the data, the Leydig cell tumors appear to be due to the combination of elevated estradiol levels and reduced prolactin levels. The pancreatic acinar cell tumors are related to an increase in serum cholecystokimn (CCK) levels. H. METABOLISM Numerous studies have been conducted investigating the excretion and disposition of C8 in various species. Additionally, studies have been conducted with exposed workers at a manufacturing plant which produces C8. Sex and species differences have been noted, whereas reproductive status in females did not have an effect on excretion or disposition in rats. Rabbits (both sexes), female rats, and male hamsters rapidly excrete C8, while male rats and female hamsters excrete C8 more slowly. Mice (bom sexes) excrete C8 even more slowly. C8 also has a long '/^-life in humans. Measurement ofC8 blood levels in an exposed worker showed that the Vs-Ufe in men is greater than 1.5 years. 11.1. Animal Studies The excretion and disposition ofC8 has been investigated in rats, mice, hamsters and rabbits. Studies have also investigated the influence of route of exposure. These studies are summarized below. 11.1.a. Male and female rats were administered radiolabeled C8 by intravenous injection. Females excreted essentially 100% of me administered dose by 24 hours, while males had excreted only 20% of the administered dose. Radioactive tissue residues were not detectable after 17 days in the females, while at 36 days males had 2.8% of me ^C in the liver, 1.1% in the plasma and lower but detectable levels in other organs (Riker Laboratory drug Metabolism Report 1-20 (1980)). n.l .b. Sex differences in me excretion and disposition of radiolabeled C8 were observed in a study of rate, mice, hamsters, and rabbits. Male and female animals @f each species were dosed by gavage with 10 mg/kg C8, and urine and feces were collected at 24,48,72,96, and 120 hours post-dosing. Animals were then sacrificed, and blood and 15 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8EXPOSURE CAS REGISTRY NO. 3825-26-1 tissues were analyzed. The urine and feces of rabbits was also collected at 144 and 168 hours post-dosing, and rabbits were sacrificed at 168 hours. The female rat and male hamster had excreted over 99% of the administered dose at the time of sacrifice. The male rat and the female hamster had excreted 39 and 60% of the administered dose, respectively, at me time of sacrifice. Bom sexes of rabbits excreted the C8 as rapidly and completely as me female rat and male hamster. The male and female mice retained substantial amounts of the total administered radioactivity in their tissues at the time of sacrifice, only excreting 21 % of the administered dose at 120 hours post-dosing (HL 62-82) BLI.c. Cholestyramine, a non-absorbable anion-exchange resin, was demonstrated to protect rats from the acute lethal effect ofC8 when administered within 2 hours ofC8 dosing (HL 828-81). A second study was conducted to investigate the effect ofeholestyramine on the elimination of ^C-CS (10 mg/kg by gavage) from rats and mice (HL 405-82). Adult male rats and mice were given cholestyramine (1000 mg/kg by gavage) 24 hours after ^G^. dosing with The cholestyramine did not enhance the elimination ofC8 via the feces, urine, or exhaled air. Similarly, Dowex Ion Exchange Resin was also able to reduce the acute lethal effect ofC8. When rats and mice were given Dowex resin 24 hours after dosing with C8 no signs of enhanced elimination ofC8, via the feces, urine or exhaled air, were observed (HL 405-82). To further investigate the use ofeholestyramine to enhance C8 elimination, a third study was conducted in rats. In this study, rats were dosed with ^C-CS (13.3 mg/kg, by iv.) and men were fed diets containing 4% cholestyramine for 14 days. The cholestyramine increased the elimination ofC8 via the feces by 9.8 fold and decreased the concentration ofC8 found in the liver, plasma, and red blood cells (Johnson, et al., 1984). n.l.d. A series of experiments was conducted to evaluate the uptake and clearance ofC8 from me blood of male and female (pregnant and non-pregnant) rats following oral exposure, and inhalation exposure. The uptake and clearance of C8 from the blood of female rats following a single oral dose was rapid, with peak reached 1-2 hours post-treatment and virtual total clearance by 24 hours. A dose-response was demonstrated with no apparent changes in blood C8 levels following multiple oral dosing. The slower clearance rate in male rats was demonstrated following a single oral dose. The same general statements apply following inhalation exposure. A single 6-hour inhalation exposure resulted in: peak blood levels within 1 hour after cessation of exposure; the material rapidly cleared from me blood; the number of exposures did not affect blood levels; and male rats cleared the compound much more slowly. Pregnant and non-pregnant rats showed similar C8 blood 16 HAZARD CHARACTERIZATION FORHUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 levels following dther oral or inhalation exposure (HL 593-91). Specifics of the experiments are summarized below. H. 1.e. 1. Oral administration-Blood levels of C8 as a function of time post-dosing (female rats) C8 levels of 14 ppm were seen 15 minutes following administration ofC8. These levels peaked at 30 ppm at 1-2 hows, dropped to 26 ppm by 8 hours, and to 0.7 and 0.045 ppm at 24 and 168 hows, respectively. C8 is absorbed and rapidly cleared from the blood of female rate given a single oral dose. n.l.e.2. Oral administration-Blood levels ofC8asa function of dose (female rats) C8 levels 30 minutes following administration of 2.5 -150 ing/kg ranged from 3 --162 ppm. The same dose response was observed at 24 hours with blood levels ranging from 0.12 - 18 ppm. The response was linear. The level ofC8 in the blood is directly related to the amount ofC8 administered. II.l.e.3. Oral administration-Blood levels ofC8 as a function of number of doses (female rats) Blood levels in female rats given 1 versus 11 doses ofC8 were not considerably different Concentrations at 15 minutes following administration were 14 and 17 ppm for 1 and 11 doses, respectively. At 30 minutes C8 concentratioas were 16 and 25 ppm; at 8 hours 26 and 13 ppm; at 24 hours, 0.7 and 0.8 ppm; and at 168 hours, 0.045 and 0.10 ppm for 1 and 11 doses, respectively. C8 does not appear to accumulate in the blood of female rats following repeated oral administration. The number of treatments does not appear to influence tine C8 blood level. II.l.e.4. Oral administration-Blood levels ofC8 following a single 25 nig/kg dose (male and female rats) Time following single oral doseBlood Levels ofC8 (ppm) ___ (hours)____ A Male Flats___Female Rats 23 16 8 63 26 24 50 Q.7 ________168___________23_______0.045 C8 is retained in the blood of male rats to a greater extent than female rats. n.l.e.5. Inhalation exposure-Blood levels ofC8 as a function of time post-exposure (female rats) C8 levels of 96 ppm were observed 15 minutes following a single 6-hour exposure to 10 mg C8/m . The level was maintained through 1 hour, fell to approximately 70 ppm at 8 hours, 52 ppm at 24 hours, and dropped to 0.39 ppm at 168 17 HAZARD CHARACTERIZATION FOB-HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 hours post-exposure. This same general pattern was observed in rats exposed to either 0.1 or 1 mg/m3. The lag phase seen following oral exposure was not observed here due to blood sampling following a 6-hour inhalation exposure (rather than a single dose at a given time). C8 is absorbed and rapidly cleared from the blood of female rats following a single inhalation exposure. n.l.e.6. Inhalation exposure -Blood levels ofCSasa function of dose (female rats) singleBlood Time following inhalation exposure dose Levels ofC8 (ppm) Following exposure to C8 at (horns)______'''''"''0"l'm^m3"'''''''''Tingfe3''''''''''''^'0''mg/m>' Vi 2 7 109 2 2 17 69 8 0.85 4 71 24 0.14 056 52 The response is linear at 30 minutes. C8 blood levels are directly related to the amount ofC8 inhaled. At the high concentration used, the clearance rate is somewhat slower man observed at me lower levels. This suggests massive overloads in the clearance system. BLl.e.7. Inhalation exposure -Blood levels ofC8 following a single 6 hour exposure to 10 mg/m3(male and female rats) doseBlood Time following single oral Levels ofC8 (ppm)~ _____(hour^s--)_--_--__--_--_--_--__--M--a--te^yRja--ts--_--__--__--_--Fe--m--a^le--R--ate-- 2 157 69 8 182 71 24 147 52 C8 is retained in me'blood of male rats to a greater extent man female rats following inhalation exposure. n.l.e.8. Oral administration-Blood levels ofC8 following a single 25 mg/kg dose (pregnant and non-pregnant female rats) Time following single oral dose Blood Levels of C8 (ppm) _______(hours)________Pregnant Rats i6 16 Non-pregnant Rate____ 10 2 33 39 8 26 31 C8 clearance following oral dosing is similar in pregnant and non-pregnant female rats. 18 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-T " ILle.9. Oral and Inhalation exposure -Blood levels ofC8 as a function ofnumber of exposure concentration (pregnant female rats) Time following single of25mg/iq? oral doseBlood Levels ofC8 Following (ppm) " '(horns)""" """""{ exposure" 6 exposures''""TOexposinres '/4 18 12 12 2 39 37 15 8 31 nd" 11 24 2 nd 1 Time following single inhalation exposure to 10 nig/in3 Blood Levels ofC8 (ppm) Followhg _______(ho%urs)___________1 exposure________10 77 exposures 53 2 90 nd a. nd == not done When comparing the C8 levels seen at 2 and 8 hours, following 10 consecutive oral doses, there appears to be a lowering of the C8 blood levels. Blood levels following 1 or 10 consecutive inhalation exposures (6 hours/day) were not different. C8 does not appear to accumulate in the blood of pregnant rats following repeated oral or inhalation exposures. IILf. The ability of 14C-C8 to transfer through the placenta was investigated in rats (HL 61-82). A single dose of 10 mg/kg "C-C8 was administered to pregnant rats on the 19thday of pregnancy. Maternal blood and placental levels of ^C-CS increased between 2 and 4 hours post-dosing, and decreased between 4 and 8 hours post-dosing. Time following single oral dose Levels of C8 of 10 mg/kg .-..-.---..--....-------.-.-....-.-- .-...-^^^--- --......-.-- -.--.^^-- -.----.. ^ ^ ^ equivalents/mLblood) ______(hoars)_______(u.g 4 20 8 12 (pg equivalents/mL tissue) 3 3 n.2. Human Exposure Determinations of organic fluorine blood levels in workers exposed to C8 in an industrial environment were performed. Approximately 90% of the organic fluorine was composed of the C8 anion. The highest levels were found in workers with fhe longest work history in ffuorochemical production. The majority of the values remained an,, approximately me same level throughout the 2 Vi year monitoring period. Monitoring of C8 blood levels of a worker who was removed from the fluorochemical production site due to high C8 blood levels (70 ppm) suggests that fluorochemicals are very slowly 19 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 eliminated. From {his limited data it is hypothesized that the 'y^-life ofC8 is 1.5 years in men(Ubel,etaL, 1980). Group Number analyzed Blood Organic Fluorine Levels _N_o_rm_a_l _hu_m_an_s_er_a_____f_ro_m_p_ub_li_sh_ed_li_te_ra_tu_re___________0(.P01P-0."13)______ Industrial controls 4 0.01-0.08 Laboratory pei-sonnel 8 0-04-2.00 (>20 years exposure) Plant workers 49 1.00-71.00 III. MECHANISMS OP ACTION C8 is not metabolized in rats. C8 produces hepatomegaly, induces hepatic peroxisomes in mice and rate, and has been shown to produce hepatic, Leydig cell, and pancreatic acinar tumors in a 2-year feeding study in rats. The male rat is more susceptible to the toxic effects ofC8 than the female rat, presumably due to the longer 2- life in males. Short-term studies have been conducted investigating me mechanisms of action responsible for me various effects. ffl. 1. Investigation of C8s' Effect on the Liver. m. 1.a. Because C8 had been shown to induce a striking hepatomegaly in rats, a study was conducted to investigate the hepatic biochemical and morphological changes associated with C8-induced hepatomegaly in rats (Pastoor, et aL, 1987). In this study male rats were dosed daily for 1,3, or 7 days with 50 mg CSYkgbody weight by intragastric intubation. The total cytochrome P450 content and activity ofbenzphetamine A'-demethylase was increased in me livers ofC8-treated rats, indicating me proliferation of smooth endoplasmic reticulum. m contrast, me soluble, cytoplasmic enzymes, . glutaifaione 5'-transferase and UDP-glucuronyltransferase, were unaffected. Camitme acetyltransferase activity was disproportionately increased relative to camitme palmitoyl transferase activity, confirming the predominant proliferation ofperoxisomes versus mitochondria. Electron microscopy confirmed me proliferative response of the endoplasmic reticulum, peroxisomes, and microsomes in the livers of the C8-treated rats. This study also demonstrated mat C8 does not possess hypolipidemic activity. in.l.b. C8 increased serum estradiol concentrations in 2-week gavage studies, and feeding studies at various time points up to 2 years. This was accompanied by increases in liver weights, and hepatic p-oxidation activity (Cook, et aL, 1992; Cook, et al., 1994). Since peroxisomeproliferators induce both P-oxidation activity and cytochrome P450 en2ymes, an investigation was conducted to determine ifC8 increases serum estradiol levels by stimulating aromatase activity (Liu, et al., 1996a). Fourteen days of treatment with up to 40 mg C8/kg/day produced dose-dependent increases la liver weights, serum estradiol, and hepatic aromatase activity. A significant linear correlation was established between estradiol and hepatic aromatase activity. In vitro experiments using cultured hepatocytes suggest that the increase in serum estradiol is at least partly 20 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 due to a direct effect on the liver to increase synthesis ofestradiol through induction of aromatase cytochrome P450 in the endoplasmic reticulum. ni.2. Investigation ofCSs' Effect on Testicular Leydig Cells Because C8 produced an increased incidence oftesticular Leydig cell tumors in a 2-year feeding study in rats, and because C8 was negative in short-term tests for genotoxidty, a non-genotoxic (hormonal-mediated) mechanism for tumor formation was investigated. The studies summarized below support a hormonally-mediated mechanism of Leydig cell tumorigenesis: C8 produces an increase in hepatic aromatase activity, which elevates serum estradiol concentrations, which in turn modulates growth factors in the testis, which results in tumor formation. IIL2.a. Fourteen days of treatment with up to 50 mg CS/kg/day produced dosedependent increases in hepatic p-oxidation activity, and serum concentrations of estradiol, and decreases in serum testosterone concentrations, body weights, and relative accessory sex organ weights in male rats (Cook, et al., 1992). Challenge experiments, using human chorionic gonadotropin (hCG), gondaotropin-releasing hormone (GnrH), or naloxone challenges, suggest that the decrease in testosterone may he due to a lesion at me level of the testes, due to a decrease in the conversion of ITa-hydroxyprogesterone to androstenedione. ni.2.b. Using in vitro, in vivo and ex vivo studies, C8 was examined for its ability to directly affect Leydig cells in vitro using isolated Leydig cells from untreated rats, and ex vivo using Leydig cells isolated from CS-treated rats. Additionally, the ability ofCS to afiEect testicular interstitial fluid hormone levels and induce aromatase activity was investigated (Biegel, et al., 1995). The in vitro studies demonstrated that C8 directly inhibits testosterone production, while me ex vivo studies demonstrated mat this inhibition is reversible. In the in vivo study, serum and testicular interstitial fluid estradiol were increased and testicular interstitial fluid transforming growth factor a, were increased. Additionally, hepatic aromatase activity was increased while aromatase activity levels ware not affected in the testis, muscle, or fat; These data suggest that the increases in estradiol levels are primarily due to increases in aromatase activity. IIL2.C. Previous studies with C8 showed a direct effect on Leydig cells to alter steriodogenesis. It was therefore proposed that peroxisome proliferators, in general, may directly affect Leydig cell function to produce Leydig cell tumors. A study investigating whether several peroxisome proliferators (including C8), directly affect Leydig cell function in vitro was conducted. This study showed that peroxisome proliferators, as a class of compounds, directly modify the steroidogenic function of Leydig cells in vitro. This also suggests that compounds which directly affect Leydig cell function in vitro may also induce Leydig cell tumors in vivo (Liu, et al., 1996b). 21 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 in.3. Investigation ofC8s' Effect on the Pancreas Several peroxisome proliferators have been shown to produce pancreatic acinar cell hyperplasia/adenocarcinomasin 2-year feeding studies, including C8. Therefore, in vitro and in vivo investigations of C8's {in vitro only) and Wyeth-14,643's (a model peroxisome proliferator) mechanism oftumorigenesis in the pancreas were conducted. These mechanisms include cholecystoldnin receptor agonism (CCKA ) trypsin inhibition, alterations in gut fat content, cholestasis and altered bile flow/composition. All offhese mechanisms enhance pancreatic growth dfher by binding to the CCKA receptor or by increasing plasma CCK levels. C8 did not bind directly to the CCKA receptor and it tailed to inhibit trypsin, a common mechanism for increasing plasma CCK levels. In vivo studies with Wyefh-14,643 suggest that these peroxisome proliferators produce pancreatic tumors by cholestasis, which may be responsible for fhe decrease in bile add output which contributes to the increase in plasma CCK levels. Therefore, for Wyeth-14, 643 (and perhaps C8), the pancreatic tumors may be secondary to hepatic cholestasis (Oboum,etal.,1997). IV. CLINICAL REPORTS OF HUMAN EXPOSURE IV. 1.a. Health screening examinations were offered to employees of a 3M plant that produced C8, as well as other fluorochemicals. No health problems related to exposure were encountered among those examined. Additionally, no relationship was observed between deviations from normal laboratory test results and blood levels of organic fluorine (the liver enzyme SGGT was the most frequently encountered test result exceeding me normal range. C8 exposure levels ranged from 0,03 to 7.6 mg/m3(Ubel, et aL, 1980). IV. 1.b. A study was made of Washington Works employees potentially exposed to C8. Results of blood chemistry testing (SGOT, LDH, AP, andbilirubin) indicated no conclusive evidence of an occupationally related health problem among workers exposed to C8 (Fayerweather, 1981). IV.l .c. Although C8 is me major organofluorine compound found in humans, little information is available concerning human responses to C8 exposure. Therefore, a study was conducted among 115 workers exposed to C8 occupationally (serum fluorine levels varied between 0 and 26 ppm, wilfa a mean of 3.3). In an examination of me cross- sectional associations between C8 and hepatic enzymes, lipoproteins, and cholesterol, there was no significant clinical hepatic toxidty of the C8 levels observed in this study (Gilliland and Mandel, 1996). Serum C8 levels were positively associated with estradiol and negatively assodated with free testosterone and not assodated with luteinizing hoimone. The negative assodation between testosterone and C8 was stronger in older men. Thyroid stimulating hormone and C8 were positively associated. Prolactin and C8 were positively assodated in moderate drinkers. The effect of adiposity on serum glutamyi oxaloacetic acid and glutamyi pyruvic transaminase decreased as C8 increased. The induction of gamma glutamyi transferase by alcohol was decreased as C8 increased. The effect of alcohol on HDL was reduced as C8 increased. A positive association 22 HAZARD CHARACTERIZATION FOR HUMAN HEALTH CSEXPOSURE CAS REGISTRY NO. 3825-26-1 between hemoglobin, mean cellular volume, and leukocyte counts with C8 was observed. These results suggest that C8 affects male reproductive hormones and that the liver is not a significant site oftoxidty in humans at the C8 levels observed in this study. However, C8 appears to modify hepatic and immune responses to xenobiotics (Gilliland and Mandel, 1993). V. EPIDEMIOLOGY V. 1.a. A retrospective cohort mortality study was made of employees at a 3M plant where C8 and other fluorocompounds are manufactured. Records on 4218 employees were reviewed. Only those who worked for 6 months or more (3688 workers) were included in the mortality follow-up. Of the 180 known deaths, 177 death certificates were obtained. Overall the number of deaths was significantly less than expected. The observed-to-expected ratio for cancer deaths was 1.0 (Ubel, et al., 1980). V. 1.b. m a retrospective cohort mortality study, a relationship between mortality and employment at a plant where C8 and other fluorocompounds are manufactured were investigated (Gilliland and Mandel, 1993). The cohort consisted of 2788 male and 749 female workers employed between 1947 and 1983. The all-causes standardized mortality rate (SMR) was 0.75 for males and 0.77 for females. There was no significantly increased cause-specific SMR for men or women. The SMRs for prostate cancer were 2.03 in the exposed group and 0.58 in the not-exposed group. In the exposed group there were 4 observed and 2 expected deaths from prostate cancer. Among men, 10 years of employment in C8 production was associated with a significant 3-fold increase in prostate cancer mortality when compared to no employment in production. Given the small number of prostate cancer deaths and the natural history of the disease, the association between production work and prostate cancer must be viewed as hypothesis generating and not over interpreted. If the prostate cancer mortality excess is related to C8, the results of this study and other clinical studies suggest mat C8 may increase prostate cancer mortality through endocrine alterations. VI. DISCUSSION OF ENDPOINTS VI. I. Discussion of Target Organs The primary target organ for C8-induced toxicity is the liver in mice, rats, and dogs, regardless of route of exposure. The hepatotoxicity manifests as increased liver weights, hepatocellular hypertrophy, liver degeneration, increases in liver enzymes, necrosis of the liver, and induction ofperoxisomes (rats and mice only). Many of these effects were demonstrated to be reversible when animals were provided with a recovery period. Evidence of hepatotoxicity was not evident in studies in monkeys or humans. m. contrast with the rodent, the target organs in me monkey were the 23 HAZARD CHARACTERIZATIONFOR HUMAN HEALTH C& EXPOSURE CAS KEGISTRYNO. 3825-26-1 gastrointestmal tract and the reticuloendothelial system (Griffith and Long, 1980). While the liver does not appear to be a primary target organ in humans, exposure to C8 appears to modify the hepatic and immune response to xenobiotics (Gilliland and Mandel, 1996). VI.2. Discussion of Differences in Species-Specific Sensitivities The induction ofperoxisome proliferation by xenobiotics is generally determined as an increase in fhe activities of certain peroxisome-specific enzymes, or as an increase in fhe numerical or volume density ofperoxisomes in fhe affected organ. Peroxisome proliferation is associated with: increases in number and volume ofperoxisomes; an increase in DNA synthesis and liver growth; and liver, Leydig cell, and pancreatic acinar cell tumors. The phenomenon ofperoxisome proliferation is not uniform across all species. While rats and mice are particularly sensitive to tins phenomenon, guinea pigs, cats, dogs and primates (including man), are predominantly non-responsive. VI.3. Tumors Associated with C8in the Rat C8 has been demonstrated to he a peroxisome proliferator in the rat C8 exposure in fhe rat was found to be associated with tumors in fhe liver, Leydig cell, and pancreatic acinar cell. Peroxisome proliferators, in general, were initially recognized to be associated with hepatocarcinogenesis in rats. However, more recently peroxisome proliferators have been associated with fhe induction of a triad of tumors in rats: liver, Leydig cell, and pancreatic acinar cell. Hyperplasia of these cell types is typically observed prior to, and along with, fhe occurrence of neoplasia. Several known peroxisome proliferators (clofibrate, HCFC-123, memylclofenapate, and Wyefh-14,643) are reported to induce this triad of tumors in rats. Hence, this tumor profile appears to be common phenomenon for at least a subset of compounds that are peroxisome proliferators. VI.3 .a. Significance ofCS-Induced Rodent Tumor to Human Risk VL3.a.l. Liver Tumors The abundance of data indicates that there is a hepatocarcinogemc hazard of peroxisome proliferators to responsive species (rats and mice) in chronic studies, whereas the carcinogenic hazard to non-responding species, such as humans, is clearly questionable. The epidemiology data, albeit limited, strongly support that fhe relevance of the hepatoearcinogenic effects of C8 and other peroxisome proliferators for human hazard assessment should be considered negligible. VL3.a.2. Leydig Cell Tumors Leydig cell hyperplasia and adenomas are commonly observed in laboratoly rats. The incidence of spontaneous Leydig cell adenomas in Cri:CDBR rats ranges from approximately 0-12% by 2 years of age, and ranges from approximately 64 -100 % in F344 rats. In contrast, fhe rate in humans has been. reported to be approximately 0.4 per 24 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 million (0.00004%). Although a direct comparison is somewhat tenuous, the data suggest a substantial difference in the susceptibility of rodents and humans to Leydig cell tumodgenesis. This is supported by epidemiology data from compounds that clearly produce Leydig cell tumors in rodent studies but are commonly ingested by humans and are not associated with Leydig cell tumorigenesis in humans. C8 and other peroxisome proliferators do not produce increases in peroxisomes in Leydig cells and are hypothesizedto produce these tumors via a different mechanism than me liver tumors. The mechanism of tumorigenesis is not completely understood, and therefore relevance to humans can not be completely ruled out. However, it is known that non-genotoxic compounds (such as C8) produce Leydig cell tumors by altering the endocrine system. Therefore, a threshold for tumorigenesis is expected. If this is the case, use of a margin of safety approach is appropriate for the quantitative dose-response assessment. It is important to consider the slope of (he dose-response at the low end of the observed range in determining an acceptable margin of safety. VI.3.a.3. Pancreatic Acinar Cell Tumors C8 and other peroxisome proliferators do not produce increases in peroxisomes in the pancreas and are hypothesized to produce these tumors via a different mechanism than the liver tumors. The mechanism of tumorigenesis is not understood, and therefore relevance to humans can not be completely ruled out However mere is a growing weight of evidence that the pancreatic acinar cell tumors are honnonally mediated, therefore they should be treated similarly to peroxisome-proliferator-induced Leydig cell tumors. Vn. SUMMARY C8 has moderate acute oral toxicity with LDso's ranging from 178 mg/kg in male guinea pigs to 680mg/kg in adult male rats. An aqueous paste ofC8 produced mild to moderate dermal irritation in rabbits and clinical signs of toxicity were observed at doses as low as 1000 mg/kg. Instillation of solid C8 into me rabbit eye produced moderate comeal opacity, iritis, and conjunctivitis. These ocular effects gradually receded. C8 has high acute inhalation toxicity with a 4-hour ALC of 0.8 mg/L in the rat Subchronic inhalation exposure to C8 produced reversible liver effects at concentrations as low as 8 mg/m3 (measured as 7.6 mg/m3). Oral and skin absorption subchronic studies confirmed the hepatotoxidty ofC8 in the rat. In chronic feeding studies in rats, C8 produced an increased incidence of tumors in me liver, pancreas, and testis. C8 was found not to be a developmental toxic or mutageaic in several tests for mutagenidty. The relevance to human health of tumors induced by peroxisome proliferators in rodents has been me focus of several investigators. Regarding me liver, there is a strong association and probable link between peroxisome-proliferator-induced liver growth and the subsequent development of rodent liver tumors. A combination ofm vivo and i^ vitro studies as well as epidemiology data, has led several investigators to conclude mat humans appear to be insensitive or unresponsive to peroxisome-proHferator-mdwed hepatic effects, and therefore these nongenotoxic agents pose little or no 25 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 hepatocarcmogenic hazard to humans. Evidence is also accumulating that fhe initiating events, which lead to the development ofLeydig cell and pancreatic acinar cell tumors are from changes in me liver. These hepatic changes appear to alter the hormonal control of fhe testis and pancreas. Although these relationships need to be confirmed, it is likely that these extrahepatic tumors pose little or no carcinogenic hazard to humans. Additionally, programs monitoring fhe health of CS-exposed workers and retrospective cohort studies of workers exposed to C8 provide no evidence of an association between C8 exposure and adverse human health effects. Of primary concern in humans is the slow clearance of C8 from human blood, the opportunity for exposure in the work place, and the moderate-high acute toxidty, regardless of route of exposure. 26 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 REFERENCES DuPont Co.. Haskell Laboratory Report NmnbersflHU: 55-61, Acute Oral Test.1961, H. Shennan. 56-61, Subacute Feeding Study.1961, H. Shennan. 123-65, Effects ofFluorocarbon Disbursing Agents on the Livers of Rats and Dogs.1965, J. R. Bames and H. Sherman. 124-65, Two-Week Feeding Study.1965, H. Shennan. 128-68, Acute Oral Test.1963, S. B. Fretz. 160-69, Acute Inhalation Dust Toxidty. 1969, F. 0. Tayfun and R. S. Waritz. 253-79, Inhalation Subacute. 1979, G. T. Hall. 635-79, Eye Irritation Test in Rabbits.1979, D. F. Edwards. 636-79, Skin Irritation Test on Rabbits.1979, D. P. Edwards. 659-79, Skin Absorption LD5o - Rabbit and Rat.1979, D. F. Edwards. 589-80, Rat Skin Absorption Subacute Study wifh FC-143.1980, L. S. Silber. 682-80, Skin Absorption LDso-Female Rats.1980, J. E. Henry. 205-81, Subacute Inhalation Toxicity ofPentadecafluorooctanoic Acid, Ammonium Salt.1981, S. D. Nash and M. A. BritelU. 291-81, Oral LDso Test in Guinea Pigs.1981, J. A. Hall. 295-81, Oral LDso Test in Rats.1981, J. A. Hall. 329-81, Oral LDso Test in Mice.1981, J. A. HalL 560-81, Mouse Feeding Study- 14Day.l981,L. S. Ford. 565-81, Acute Oral Toxidty Comparison Test in Rats.1981, L. Hinckle. 567-81, Oral LDso Test in Rats.1981, J. E. Henry. 593-81, Ammonium Perfluorooctanoate Blood Levels in me Female Rat 1981, Kennedy, G.L. 27 HAZARD CHARACTERIZATION FOR HUMAN HEALTH- OS-EXPOSURE CAS REGISTRY NO. 3825-26-1 600-81, Liver Weight Comparison in Castrated and Ovariectomized Rats vs. Normal Rats.l981,L.Hinclde. 537-82, Fourteen-Day Feeding Study in Male and Female Mice.1982, J. E. Henry. 788-82, Acute Oral Toxidty ofFC-143 as a Function ofAge.1983, L. Hindde. 138-83, C-8 Liver Weights in Rats and Mice: Oral Administration to Males and Females.1983, C. N. Wyle. DuPont MR 5686, Investigation ofHormonally-Mediated Mechanisms for Xenobiotic- Induced Leydig Cell Tumors in Male Rats. (1989). 3M Co. Reports: Personal communication, J. Long to PJ. Gort, January 8,1979. 3M Report M-601 (1981). 3M Report 0681TR0110,1981. 3M Product Toxicity Sheet, May 24,1996. Biosearch, Inc. Report No. T1395, March 4,1976. Coming Hazleton, 17388-0-455, May 16,1996. Coming Hazleton, 17388-0-437, April 25,1996. Hazleton Laboratory America, Inc. 2-6-87 (Submitted to EPA, TSCA 8eCAP Submission, Fiche 536984). Litton Bionetics; LBI Project 20838, February 1,1978. Riker Laboratory Drug Metabolism Report 1-20 (1980). Riker Laboratories, Report 09790AB0485, March 15,1981. Riker Laboratory 0281CR0012, April 1981-May 1983. University of Minnesota Environ. Path Lab, T2942, April 9,1981. Published References: Biegel, L. B., R. C. M. Liu, M. E. Hurtt, and J. C. Cook (1995). Effects ofammomum perfluorooctanoate on Leydig cell function: In vitro^ in vivo, and ex vivo studies. Toricol. Appl. Pharmacol. 134:18-25. 28 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Cook, J. C., M. E. Hurtt, S. R. Frame, and L. B. Biegel (1994). Mechanisms of extrahepatic tumor induction by peroxisome proliferators in Cri:CDBR (CD) rats. Toxicologist (abstract^ 14. Cook, J. C., S. M. Murray, S. R. Frame, and M. E. Hurtt (1992). Induction ofLeydig cell adenomas by ammonium perfluorooctanoate: A possible endocrine-related mechanism. Toxicol. Appl. Pharmacol. 113:209-217. Fayerweather, W. E., Liver study of Washington Works employees exposed to C8: Results ofblood biochemistry testing. DuPont, Epidemiology Division of Corporate Medical, January 15,1981. Gilliland, F. D. andj. S. Mandel (1993). Mortality among employees of a perfluorooctanoic add production plant J. Occup. Med. 35:950-954. Gilliland, P. D. and J. S. Mandel (1996). Serum perfluorooctanoic add and hepatic enzymes, Upoproteins, and cholesterol: A study ofoccupationally exposed men. American Journal of Industrial Medicine 29: 560-568. Griffith, F. D. and J. E. Long (1980). Animal toxicity studies with ammonium perfluorooctanoate. Am. fad. Hvg. Assoc. J. 41:576-583. Johnson, J. D., S. J. Gibson, and R. E. Ober (1984). Cholestyramine-enhanced fecal elimination of carbon-14 in rats after administration of ammonium [14C]perfluorooctanoate or potassium [ 14C]perfluorooctanesulfonate. Fundam. Appl. Toxicol. 4:972-976. Kennedy, Jr., G.L. (1985). Dermal toxicity of ammonium perfluorooctanoate. Toxicol. Appl. Pharmacol. 81:348-355. Kennedy, Jr., G.L., G. T. Hall, M. R. Brittem, J. R. Bames, and H. C. Chen (1986). Inhalation Toxidty of Ammonium Perfluorooctanoate. Fd. Chem. Toxic. 24(12); 1325- 1329. Kennedy Jr., G. L. (1987). Increase in mouse liver wdght following feeding of ammonium perfluorooctanoate and related fluorochemicals. Toxicol. Letters 39:295300. Liu, R. C. M., C. Hahn, and M. E. Hurtt (1996a). The direct effect of hepatic peroxisome proliferators on rat Leydig cell function in vitro. Fundam. Appl. Toxicol. 30:102-108. Liu, R. C. M., M. E. Hurtt, J. C. Cook, and L. B. Biegel (1996b). Effect of the peroxisome proliferator, ammonium perfluorooctanoate (C8), on hepatic aromatase activity in adult male Crl:CDBR (CD) rats. Fundam. Anm. Toxicol. 30: 220-228. ' 29 HAZARD CHARACTERIZATION FOR HUMAN HEALTH C8 EXPOSURE CAS REGISTRY NO. 3825-26-1 Oboum, J. D., S. R. Frame, R. H. Bell Jr., D. S. Longnecker, G. S. EUiott, and J. C. Cook (1997). Mechanisms for me pancreatic oncogenie effects of me peroxisome proliferator Wyefh-14,643. Toxicology and Applied Pharmacology 145:425-436. Pastoor, T. P., K. P. Lee, M. A. Perri, and P. J. Gillies (1987). Biochemical and morphological studies of ammonium perfluorooctanoate-induced hepatomegaly and peroxisome proliferation. Experimental and Molecular Pathology 47:98-109. Perldns, R. G. (1992). Investigation of ammonium perfluorooctanoate effect on hormone levels and peroxisomal proliferation in the rat. The Toxicologist 12(0:52. Staples, R. E., B. A. Burgess, and W. D. Kerns (1984). The embryo-fetal toxidty and teratogenic potential of ammonium perfluorooctanoate (APFO) in the rat. Fundamental and Applied Toxicology 4:429-440. Ubel, F. A., S. D. Sorenson, and D. E. Roach (1980). Health status of plant workers exposed to fluorochemicals-a preliminary report. Am. fad. Hyg. Assoc. J. 41(8); 584589. 30