Document XOGxwEXdkXge7wYBqmZkREgvy

NORTHERN KENTUCKY QFF,CE SUITE HO 'TIT OIXIE HIGHWAY COVINGTON KENTUCKY 4 '0 " -*T0A 5!U0 S-.MJ3ii..2;Si3Ma FAX S IJ- M I- M U ROBERTA. BlLOTT (513) 357-9638 bilottQtaftlaw.com -. AR Z36 TAFT, STETTINIUS & HOLLISTER LLP 1800 FIRSTAR TOWER 425 WALNUT STREET CINCINNATI, OHIO 45202-3957 513 -331 -25!"'S FAX 5i3-3S-:-r..?05 www!afttaw coin June 10. 2002 TELECOPY Chris Negley, Esq. West Virginia Department O f Environmental Protection Office Of Legal Services 1356 Hansford Street Charleston, WV 25301 Dr. Dee Ann Staats West Virginia Department o f Environmental Protection 1356 Hansford Street Charleston, WV 25301 1182 cr.isCLEVE-anC Z*<C p :0 0 Pu b i C $CLA E Cl EVELANO 0 *-"0 *a - , * - 23C2 -ax ; 5242*933 C CL-M 0L5 3 - `C ? c ; 2 : 24 EAS' 3 * * ' t S' 55~ c c l jm sl s 2!33 - a x ;:c- rcC*rZoZvt (yzm, CJTTO -'Or 'i __-1 ~"Qo cn -H f"'i 1 2 ! i ~Jr',1 O '-J Pam Nixon Environmental Advocate West Virginia Department o f Environmental Protection 1356 Hansford Street Charleston. WV 25301 Jan R. Taylor, Ph.D. National Institute for Chemical Studies 2300 MacCorkle Avenue, S.E. Charleston, WV 25304 James Becker, M.D. Marshall University Center for Rural and Environmental Health 1600 Medical Center Drive Suite 1500 Huntington, WV 25701 Joan Dollarhide, Ph.D. Michael Dourson. Ph.D. Toxicology Excellence for Risk Assessment 1757 Chase Avenue Cincinnati, OH 45223 William Toomey Manager o f Source Water Assessment Program West Virginia Department o f Health and Human Resources Bureau for Public Health 815 Quarrier Street, Suite 418 Charleston, WV 25301 Barbara Taylor Director, Office o f Environmental Health Services West Virginia Department o f Health and Human Resources - Bureau for Public Health 815 Quarrier Street, Suite 418 Charleston, WV 25301 CONTAIN NO CB 000123 f June 10. 2002 Page 2 % Jennifer Seed U.S. Environmental Protection Agency 410 M Street. S.W. Washington. DC 20460 Samuel Rotenberg, Ph.D. (3W C 11) U.S. Environmental Protection Agency 1650 Arch Street Philadelphia. PA 19103-2029 Roger Reinhart U.S. Environmental Protection Agency Region III 1650 Arch Street Philadelphia, PA 19103-2029 John Wheeler, Ph.D. (E-29) ATSDR 1600 Clifton Road, N.E. Atlanta, GA 30333 Christopher Jones. Esq. Director Ohio Environmental Protection Agency 122 South Front Street Columbus. OH 43215 Lillian Pinzon, Esq. (C-14J) U.S. Environmental Protection Agency Region V 77 West Jackson Blvd. Chicago, IL 60604-3590 Janet E. Sharke. Esq. (3EC00) U.S. Environmental Protection Agency Region III Office Of Enforcement. Compliance and Environmental Justice 1650 .Arch Street Philadelphia, PA 19103-2029 Garth Connor U.S. Environmental Protection Agency Region III 1650 Arch Street Philadelphia, PA 19103-2029 John Cicmanec. DVM U.S. Environmental Protection Agency 26 West Martin Luther King Drive Cincinnati, OH 45220 Laura Werner (3HS00) ATSDR U.S. Environmental Protection Agency Region III 1650 Arch Street Philadelphia. PA 19103-2029 Greg Smith. Esq. Office o f Legal Services Ohio Environmental Protection Agency 122 South Front Street P O. Box 1049 Columbus. OH 43216-1049 Mary Dominiak FOR INCLUSION IN AR-226 United States Environmental Protection Agency Office o f Pollution. Prevention and Toxics Chemical Control Division 1201 Constitution Avenue, N.W. Mail Code 7405M Washington. DC 20004-001 000124 r . une 10. 2002 'age 3 Re: C-8 Assessment O f Toxicity Team Work Under November 14. 2001 Consent Order Between DuPont .And State O f West Virginia (Order No. GRW-20QI -fl!Q> adies and Gentlemen: As indicated in our prior correspondence to the State and Federal agencies, our law firm id co-counsel in Charleston. West Virginia, have been certified by a State Court in West irginia to serve as counsel for a class o f all persons whose drinking water is or has been mtaminated with ammonium perfluorooctanoate ("C-8") attributable to releases from E.l. iPont de Nemours and Company's ("DuPont's") Washington Works in Wood County. West irginia. As class counsel, we seek clarification from the agencies with respect to a situation volving a threat to the health o f the members o f the class we represent. More specifically, we ek clarification o f the manner in which the C-8 Assessment o f Toxicity Team ("CAT Team i tablished under the referenced Consent Order selected 150 parts per billion (ppb) as a creening level" for C-8 in drinking water in West Virginia, which members o f the class are inking on a daily basis. As explained below, the information that has been made available to to date indicates that the CAT Team's analysis is fundamentally inconsistent with the facts d agency guidance for interpreting those facts with respect to the calculation o f "screening /els" or "lifetime drinking water heath advisories" ("DWHAs"). Based on the press releases d public meetings sponsored by the CAT Team regarding the 150 ppb number, the class imbers are being led to believe that drinking water with up to 150 ppb C-8 presents qo risk ot v kind to their long-term health. We do not believe that is correct. Available facts and idance do not support either a "screening level" or DWHA above even l ppb, let alone 0 ppb. Although we requested an opportunity to designate a representative o f the class mem be'' sit on the CAT Team for purposes o f determining an appropriate screening level for C-8 m nking water, that request was refused. We did not. therefore, have a representative present ring the CAT Team's meeting on May 6-7, 2002. nor have we received any minutes or w n n eorts clarifying the details o f the analysis used by the CAT Team in selecting the C-8 screening el during that meeting. Although we understand that a formal report summarizing the CAT un's screening level analysis most likely will not be available for another several months, the ^rmation that has been made available to date by the CAT Team through its public meetings Vest Virginia and Ohio on May 15, and May 16. 2002. and during our deposition last irsday and Friday o f the CAT Team's Leader. Dr. Dee Ann Staats, is sufficient to raise sern . cem s with respect to the manner in which the CAT Team performed its analysis. Because . public health threat involved, we are compelled to raise these concerns with you now for mpt resolution. Based upon the information made available to date, we understand that the CAT Team erstood that its purpose was to derive a "screening level" for C-8 in water o f the nature often i as simply the threshold for triggering a cleanup or remediation o f a Superfund or other 000125 r June 10, 2002 Page 4 hazardous waste site, as opposed to a DWHA designed specifically to conservatively protect human health from exposure to C-8 in drinking water. We further understand that the CAT Team calculated its "screening level for C-8 in drinking water through use of the following basic five-step process: (1) Selection o f "key" toxicity studies on C-8; (2) Identification o f the "crmcui effect" from C-8 exposure indicated in each o f the "key" studies; (3) Identification o f the C-8 dosing level at which the "critical effect" occurred in each o f the "key" studies (the "critical effect level"); (4) Calculation o f a reference dose or "RfD/RfC" for each "key" study by dividing the "critical effect level" by the product o f the Intraspecies Extrapolation Uncertainty Factor (I F the Interspecies Extrapolation Uncertainty Factor (UFA), the Subchronic-to-Chronic Exposure Extrapolation Factor (UFS), the LOAEL-to-NOAEL Extrapolation Uncertainty Factor (UF >. Database Quality and Completeness Uncertainty Factor (UF0); and (5) Multiplication of the RfD/RfC derived for each o f the "key" studies by "USEPA Region IX Screening Levels" assumptions regarding adult body weight and drinking water consumption level to derive the proposed drinking water screening level. Even assuming solely for purposes o f argument that :he CAT Team identified the appropriate "key studies." the appropriate "critical effects" for each of the "key" studies, and the appropriate "critical effect levels" for each o f the "key" studies, available information and guidance do not support the CAT Team's calculations o f corresponding RfD/RfCs or the ultimate screening level selected for C-8 in human drinking water, let alone use o f the 150 ppb number as a DHWA for C-8. The bases for our concerns are summarized below. I. The CAT Team Did Not Assign Appropriate Uncertainty Factor Values. In calculating the RfDs/RfCs for each o f the studies identified as the "key studies" b> CAT Team. CAT Team representatives have stated that each Uncertainty Factor is typically assigned a "default" value o f 10, which can only be reduced to a less conservative value of 3 >r even less conservative value o f 1, if there is sufficient information or data to confirm that the basis for the standard default value o f 10 is not appropriate. Although the calculations pro^ iJeu by the CAT Team indicate that the standard, conservative default value o f 10 was used by the CAT Team for the UFAand UFHfactors in calculating RfDs/RfCs for each o f the "key" studies appears that the CAT Team generally selected the least conservative values for each o f the remaining Uncertainty Factors. As explained below, it is not clear how the CAT Team can justify the use o f those less conservative Uncertainty Factors, given available information on y > and agency guidance. (See, e.g., Exhibit C (USEPA's 1/17/92 "Background Document 3: o f Drinking Water Health Advisories").) A. The CAT Team Did Not Apply The Correct "LOAEL-To-NOAEL Extrapolation Uncertainty^actor'' (LX k------------------------------- Available data and guidance does not support the LOAEL-to-NOAEL Extrapolation Uncertainty Factor (UFL) values used by the CAT Team in developing its screening level for L * in drinking water. Available guidance confirms that the UFLfactor, which addresses the 00012$ f June 10. 2002 Page 5 uncertainty inherent in extrapolating'data from.a- Lowest Observable Adverse Effect Level ( ''LOAEL"). should be kept at the default value of 10. unless a No Observable Adverse Effect Level C'NOAEL") was actually identified for the study in question. USEPA has taken the position in several other risk assessments, such as USEPA's risk assessment for 1. 3-butadiene, that a benchmark dose level ("BMDL"). although not defined as being the equivalent of either me NOAEL or LOAEL, should be viewed as the equivalent o f a LOAEL. If the BMDL is close to a LOAEL. it is particularly appropriate to render it the equivalent o f the LOAEL, especially if a NOAEL is not identified for a study, since effects may then occur much below the LOAEL or BMDL. For the C-8 2-generation reproductive rat study, no NOAEL was identified by the L \ r Team and the BMDL (0.42 mg/kg/day) identified by the CAT Team is close to the LOAEL t i mg/kg/day). In addition, the CAT Team incorrectly viewed the 1.6 BMDL referenced as a critic-, effect level for the 3M 1983 two-year rat study as a NOAEL, even though 1.6 was identified b> the CAT Team as a LOAEL for the study making it impossible to dispute that the BMDL calculated by the CAT Team was the same as the LOAEL. The CAT Team also erred in labeling the 0.47 critical effect level identified by the CAT Team for the 1993 Palazzolo 90-dav studv as a NOAEL, given that USEPA had just confirmed in its draft health assessment for C-8 that the 0.47 effect level was a LOAEL - not a NOAEL. (See Draft Hazard Assessment o f Perfluorooctanoate Acid and its Salts (USEPA. OPPT (2/20/02) (as amended 4/15/02) 0`USEP \ Report"), at 47.) The existence o f a 90-day LOAEL essentially at the chronic BMDL (0.47 versus 0.42 mg/kg/day) indicates that effects probably occur below the BMDL and a UFLof !" .~ appropriate. Nevertheless, it appears that the CAT Team inappropriately selected far less conservative UFLvalues o f 1 and 3 for studies in which LOAELS or BMDLs were used, inste.;., of NOAELs. Even if there were a dispute as to whether critical effect levels based on anything other than NOAELs justify use o f a default value lower than 10 for the UFL, the CAT Team has not explained why it did not choose to err on the side o f being as conservative and as protective oi human health as possible, which would have required selection o f the default value o f 10 as ;hin appropriate UFLvalue for any o f the studies where the critical effect level was not based on the existence o f an NOAEL. In other words, if there was any room for argument as to which UF value can be justified for any o f the "key" studies, why did the CAT Team not err on the side v being the most protective o f human health - not the least protective? Moreover, it seems strar - , that the CAT Team (including members o f governmental regulatory agencies charged w ith protecting human health and the environment) opted to select far less conservative (and thereby less protective) values for the UFLfactor than even the manufacturer o f C-8 believed was appropriate. More specifically, in January o f 2002. the manufacturer o f C-8, the 3M Company ("3M"), submitted to USEPA a Lifetime Drinking Water Health Advisory for C-8 (the "3M Report") in which 3M stated that the UFLfactor for the 26-week monkey study reviewed b> tb.e CAT Team should be assigned a value o f 6. (See 3M Report, at 8.) Yet, the CAT Team selec-.. : an even less conservative, less protective UFUvalue o f 3 for the exact same monkey study. Thi simply does not make sense, particularly when 3M's lead toxicologist for C-8 was at the C A Team meeting. 000127 r June 10. 2002 Page 6 B. The CAT Team Did Not Apply The Correct "Subchronic-To-Chronic Exposure Extrapolation Factor" (UFc).___________ ______________________________ Available data and guidance also does not support the Subchronic-to-Chronic Exposure Extrapolation Uncertainty Factor (UFS) values selected by the CAT Team for the various kev studies" reviewed by the Team. According to CAT Team representatives, the UFS factor is assigned a default value o f 10, unless the study at issue is a "chronic"(lifetime) study. In a January 2002, report prepared by DuPont's consultant, ENVIRON (the "ENVIRON Report k which was submitted by DuPont's attorneys to the WVDEP Leader o f the CAT Team. Dr. Dee Ann Staats, ENVIRON stated that there had been only 2 chronic (lifetime) studies o f C-8 -- the 2-year rat study conducted on behalf o f 3M from 1983 (the "Sibinsky study") and a 2-year rat study from 2001 referenced as the "Biegel Study.'-' (ENVIRON Report, at 6 and 8.) E N V IR O N also stated that the other studies that had been conducted with C-8 (including the 90-dav rat stud\ by Palazzolo from 1993 and the 26-week monkey study by Thomford from 2001) were all "subchronic" studies, requiring use o f the default value o f 10 for the UFS value when calculating an RfD/RfC from such studies. (See ENVIRON Report, at 5 (referencing the 1993 Palazzolo study as a "subchronic feeding study") and at 7, 26 (referencing the 26-week monkey study co sponsored by DuPont and 3M as a "subchronic toxicity study")). 3M also agreed that the default value o f 10 is the appropriate UFS factor for use in calculating an RfD/RfC from the 26-week monkey study, based on recognition that the study was "significantly less than chronic." (3M Report, at 7.) USEPA also recognized that the 1993 90-day rat study by Palazzolo was a "subchronic" study in the draft Hazard Assessment for C-8 that the agency released to the public in March o f this year. (See USEPA Report, at 4, 41.) This is not surprising, given that USEP A defines chronic as lifetime or not substantially different from lifetime. (USEPA Office o f Drinking Water Health Advisories Web site; USEPA Risk Assessment Guidelines, Part A (1989)). It is, therefore, not clear how the CAT Team, with representatives o f DuPont. 3M. and USEPA present, assigned values less than 10, including the least conservative value o f 1. to the exact same Palazzolo 90-day rat study and the Thomford 26-week monkey study that E N V IR O N 3M, and USEPA all previously agreed were subchronic. It also is not clear how the CAT Team viewed the 2-generation rat study by York from 2002 as a "chronic'' study, when the period from the first dosing o f the parental generation to the sacrifice o f the second generation o f rat pups was only approximately 8 months - considerable shorter than 2 years. In fact, each generation o f the rats was dosed for only a period o f approximately 4 months, which is only approximately 10% o f the 2-3 year lifespan o f a rat. USEPA specifically defines such exposures o f approximately 10% o f the experimental animal's lifespan as "subchronic." (See, e.g., Exh. C.) It is not clear, therefore, how the CAT Team justified selection o f the least conservative value o f l for the UFS value, which is justified onI> when data from a chronic (lifetime) study is used. It is not clear at this point whether the Cat Team even reviewed the Biegel Study. 0001&8 r June 10, 2002 Page 7 C. The CAT Team Did Not Apply The Correct "Database Quality And Com pleteness Uncertainty Factor" (U FP)._______________ _________ Currently-available data and agency guidance also do not support the CAT Team's selection o f the least conservative value o f l for the Database Quality and Completeness Uncertainty Factor (UFD) in calculating an RfD/RfC. According to statements by CAT Team members, the UFDfactor is the value used to account for potential uncertainty arising from a less than complete database for the particular chemical at issue. DuPont's own consultant, ENVIRON, stated in its January, 2002, report submitted to the CAT Team's Leader. Dr. Staats. that a database is not considered "complete" for purposes o f deviating from the standard default value o f 10 for the UF0, unless all o f the following data exists: "(1) chronic toxicity studies m 2 species (1 non-rodent), (2) a multi-generation reproduction study, and (3) developmental toxicitv studies in 2 species." (ENVIRON Report, at 25.) According to ENVIRON, the database for C-8 cannot be viewed as complete, even though "the available database for [C-8] includes developmental toxicity studies in 2 species (rat and rabbit)" and "[p]reliminary results o f a 2generation reproduction study are now available," because the available C-8 database "lacks at this time a non-rodent chronic study." {Id.) Thus, according to DuPont's own consultant, ENVIRON, the C-8 database, even with the existence o f the 2-generation rat study, cannot be viewed as "complete." Consequently, it is, again, not clear how the CAT Team can justify use of the least conservative and least protective value o f 1 for all o f the available C-8 studies as the appropriate UFDvalue, when even DuPont's consultant believes that the database for C-8 is not complete and could not justify a UF0 value o f 1. The "completeness" o f the available C-8 database was discussed during a public meetir.o sponsored by WVDEP on May 15, 2002, in which the CAT Team representatives were not abie to identify what actually caused the death o f any o f the monkeys at any o f the dose levels during the 26-week monkey study, and seemed to suggest that some o f the effects observed in the rat studies may not correlate with effects observed in primates or humans. Such apparent uncertainty would appear to raise serious concerns with respect to the "completeness" o f the database on C-8 with respect to primate and human health. Such uncertainty would, in fact, appear to support multiplication o f the sum o f the five Uncertainty Factors by an additional "Modifying Factor (MF)" that, according to DuPont's consultant, ENVIRON, is justified for "expressing residual uncertainty associated with the study and database not explicitly treated b> the standard UFS (e . g completeness o f the overall database or the number o f species examined)." (ENVIRON Report, at 21) "As with the UFS, the value o f the MF may be up to 10." (Id.) It would appear that use o f a Modifying Factor would be particularly appropriate in evaluating C-8 studies, given indications from the recent 2-generation rat study that C-8 causes developmental effects. Yet, despite comments made by the CAT Team representatives themselves during the public meeting expressing uncertainty with extrapolation o f the existing database to human health and again during the deposition o f the CAT Team's leader, Dr. Staats it does not appear that the CAT Team considered use o f any Modifying Factor o f any value in calculating its RfDs/RfCs. 000129 f June 10. 2002 Page 8 II. The CAT Team Ignored Potential C-8 Exposures From Non-Drinking W ater Sources._________________________________________ It is not clear from the information made available to date by the CAT Team how the CAT Team converted its RfDs/RfCs into its 150 ppb number for C-8 in drinking water, or exactly what the purpose o f such a "screening level" is in this case. The CAT Team's Leader. Dr. Staats, has explained that the RfDs/RfCs were convened into a 150 ppb screening levei b\ using "USEPA Region IX Screening Level" calculations. Although exactly what those calculations were has not yet been explained, it appears that the calculations incorporated the same standard USEPA default values for calculating a drinking water exposure limit ("DWEL" for use in calculating a DWHA, which assume a standard human body weight o f 70 kg and standard human drinking water rate o f 2 liters per day: DWEL = (RfD/RfC x 70) / 2. After ill. incorporation o f the CAT Team's 0.0042 RfD/RfC into this same standard DWEL formula %'.eiu147 ppb. which we understand was then rounded up by the CAT Team to 150 ppb (0.0042 x "> : = 147 ppb). The CAT Team apparently used this DWEL as its "screening level" under the referenced Consent Order, instead o f following the standard agency guidance for converting the 150 ppb DWT-L into a DWHA. In fact, we understand that the CAT Team's leader, Dr. Staats did not view the task o f the CAT Team as deriving a lifetime DWHA (as opposed to derivation o f the type o f "screening level" typically used to trigger cleanup obligations), and therefore did not consider the available agency guidance for deriving lifetime DHWAs. If this is the "screening level" approach actually used by the CAT Team, the CAT Team ignored the relatr. _ source contribution factor ("RSC") required under agency guidance for converting a DWEL a lifetime DWHA, which both 3M and DuPont's consultant, ENVIRON, stated should be useu calculating any C-8 drinking water standard to account for the potential that community mem tv could be exposed to C-8 from sources other than their drinking water. As required under agency guidance for calculating lifetime DWHAs, Both 3M and ENVIRON have stated that a DWEL should be multiplied by the standard RSC default value 0.2 (20%) to account for the "relative contribution o f various potential pathways to total [C-8' exposure" such as "food and air." (3M Report, at 4. See also ENVIRON Report, at 26.) In otrv- words, the more conservative DWHA process for identifying a "safe" level o f a chemical in drinking water, requires use o f the RSC to take into account the fact that community member' typically may be exposed to a chemical from non-drinking water sources, such as air pollution containing the chemical, soils contaminated with the chemical, food contaminated with the chemical, etc., which, if ignored when determining an exposure level for drinking water. coulJ result in too high o f a total daily dose o f the chemical from all sources, combined. Conversion the CAT Team's 150 ppb number into a more protective DWHA by using the standard RSC factor advocated by both 3M and DuPont's consultant. ENVIRON, would require the CAT Tea.' to multiply its 150 ppb number by 0.2 (a reduction o f the 150 ppb number by 80 %) to take tnn> account the potential that members o f the community may be exposed to C-8 from sources othe' than drinking water, such as air pollution. That step alone would reduce the CAT Team's number from 150 ppb to 30 ppb. Use o f the DWHA RSC would appear to be particularly 00013 f June 10, 2002 Page 9 appropriate in this situation when DuPont is discharging C-8 directly into the air from stacks at its Washington Works plant. In addition, when asked specifically why the CAT Team did not use the 0.2 (20%) RSC factor, the CAT Team members stated during the recent public meetings that the 0.2 (20% i RSC was "not applicable," because the CAT Team members allegedly had "unanimously" agreed that C-8 is not a "volatile" chemical and, therefore, could not possibly escape from the drinkins uatcr to create any inhalation risk from the drinking water itself. This seems strange, given L'SEPAC statements in its recent draft Hazard Assessment for C-8 that "quantitative conclusions regarding rates o f volatilization from water or Henry's Law constant are not possible. However. APFO ar.J PFOS are capable o f transport out o f water." (USEPA Report, at 11-12.) Moreover, even it' the existing data proved that it is not possible for any C-8 to escape from drinking water to pose am inhalation risk, the CAT Team's rejection o f the RSC based on the inhalation risk from the drinking water itself completely ignores the purpose for which the RSC factor is used - name l\ . to take into account potential exposure to C-8 from sources other than drinking water, such as C8 in air emissions from the DuPont's Washington Works. It is not, therefore, clear how the CA T Team can justify not using the RSC factor for purposes o f calculating a drinking water standard for C-8, particularly when both 3M and DuPont's own consultant, ENVIRON, agree that the RSC should be used, and it is not clear that C-8 will not escape from drinking water. III. The CAT Team Did Not Include Cancer Data In Its Screening Level Calculations. It is not clear why the CAT Team did not perform a dose-response analysis for C-8 usir cancer endpoints. Recognizing that C-8 exposure has been confirmed to cause Leydig cell tumors, hepatocellular tumors, and pancreatic acinar cell tumors in rodents, the most common animal model used to identify probable human carcinogens, DuPont's own consultant, ENVIRON, included a dose-response analysis for cancer endpoints in the original version of :.% report it submitted to the CAT Team's Leader, Dr. Staats, in January o f 2002, that advocated a 10 ppb level for C-8 in drinking water. (See ENVIRON Report, at 26-30.) In addition, the Consent Order between DuPont and the State o f West Virginia requires the CAT Team to perform a "determination o f the potential carcinogenicity o f C-8." (Consent Order, at C -4.) Nc* when asked how the CAT Team incorporated the C-8 cancer data into its calculations for a C-8 screening level for drinking water, CAT Team representatives have stated that the triad o f cancers attributable to C-8 exposure had somehow been determined by the CAT Team to be totally "irrelevant" to humans. Any such conclusion would, however, be completely inconsistent with the conclusion recently reached by USEPA in its draft Hazard Assessment for C-8 released just 2 months ago, in which USEPA stated that "as the mechanisms o f carcinogenic action o f [C-8] have not been fully elucidated, it is assumed that the tumors induced in rats are relevant v humans." (USEPA Report, at 9 (emphasis added) (Report also refers to 1997 Clegg panel findings that Leydeg cell tumors in rodents are a legitimate end point for cancer risk assessment when the mechanisms o f action are not completely understood.)) It is, therefore, not clear how, the CAT Team, with USEPA participants, could support a totally contradictory conclusion. 000131 r June 10. 2002 Page 10 The inherent inconsistencies between the CAT Team's analysis and available facts and guidance generate stunning differences between the "screening level" 'cleanup numbers generated using the CAT Team analysis and the levels that would be generated using an analysis that is more consistent with available data and agency guidance interpreting that data in the consen. ati\ e manner required to ensure adequate protection o f human health from exposure to chemicals in drinking water. The nature o f some o f those differences is highlighted in the attached chans (Exhibits A and B) showing the numbers allegedly used by the CAT Team and the numbers that would appear to be more supported by available data and agency guidance, assuming for purposes o f argument only that one does not dispute the "key" studies, "critical effects." and "critical effect levels" selected by the CAT Team and that no Modifying Factor is warranted indicated in those attached charts, use o f available data and agency guidance would not suppon use o f an exposure level for C-8 in drinking water above even 1 ppb, which also happens to be the "safe" level for C-8 in drinking water that DuPont has used internally since at least 1991 Thus, given the fact that available data and guidance does not support the CAT Team's 150 ppb level for C-8 in drinking water, we request that the State and Federal agencies take immediate steps to clarify how and why the CAT Team came up with a number that is more than 10-15 times higher than the numbers calculated just several months ago by both 3M and DuPont's consultant, and more than 150 times higher than any number that would be possible when viewing the available facts and available agency guidance in the light most protective o f human health, particularly when the CAT Team did not review the results o f any studies that had not also been reviewed by DuPont and 3M. We also request prompt clarification to the public that the "screening level" numbers being developed by the CAT Team are not the same thing as DWHAs. We also request that the State and Federal agencies immediately take those steps necessary to abate and remediate this on-going threat to public health. Thank you. Very truly yours, Robert A. Bilott RAB/mdm Attachments cc: R. Edison Hill, Esq. (w/o attachments) Larry A. Winter, Esq. (w/o attachments) Gerald J. Rapien, Esq. (w/o attachments) 000132 I Mimi I A <Al II \MHI <.M>Nl\ \< RI I N|N(. I I \ I l AN\l AS|S A lte rn a tiv e D e riva tion * o f (h r RfT) in d R f f ' V * |m fo r <`8 Rflrm icf O ral SlodiM < ritirai Effect f r i t real Effect Level" HE, UF, Palaj/oln cl al ( 199 1)' 90-day ral study Increased relative liver weight with hislnpalholopy in male rats York cl al (2002) Two Generation rat. study Increased liver weiphl in male rats, supported by hislnpalholopy at higher doses 047 (NOAF.I, in males) 0 72 (BMI)L) 1 (LOAEI, in males) 10 10 1 10 10 3 1 1 3M(I9R1) 7 wo-year ral study Increased liver weight in male rals. supported hy hislnpalholopy al higher loses (hislnpalholopy was not examined al the lowest lose, hill incidence of hypertrophy was 100*/. at next highest dose) 1uhiilar hyperplasia of the ovarian stroma and clinical signs (alaxia) in female rats 0 42<nMI)l. in males)'1 1 6(1 OAI.I. in females) 1 57 (HM I)I) III 10 10 10 1 1 1 1 Ihomfnrd el at ( 2 0 0 1 ) '2 6 -week cynomnlgiis monkey study Ilcpntic mcgalocylnsi* in male i;K 1lecrcased thyroid hormone levels in male cynomnlgiis monkeys, and supported hy a Nl >AI I, al the same dose for clinical signs of toxicity in the co critical rhesus monkey study (Goldcnthal et a l , 1978) 0 73(HMI)I. in males) 3 - 10 (LOAN, in males) 10 10 10 10 1 3 1 3 Il 1 1 1 1 1 1 Composite IIF* RfO/RfC 100 0 005 0 0072 too X 100 0 004 100 0 0157 100 00073 1000 0 003 - 001 Screening Uvei X X |S |lp ,,|. X X X 000133 IX111II11 H I II FUMI l)l(|NMN(, WAIIK ANAI.VMS Reference C ritical Effect O ral Sladtes Palar^olo et al (1991)' 90-day rat study Increased relative liver weigh! with hislopathology in male rats York rt al (2002) Two-General ion raf study Increased liver weight in male rats, supported by hislopathology al higher doses. Increased liver weight in male rats, supported by hislopathology at higher doses (hislopathology was not examined al the lowest dose, but incidence of hypertrophy was 100% at next highest dose). 3M (1981) Two-year rat study Tubular hyperplasia of the ovarian stroma and clinical signs (ataxia) in female rats. Hepatic megalocytosts in male rats Thomfordet al (2001 )*26-week cynnmolgus monkey study h 1decreased thyroid hormone levels in male cynomolgus monkeys, and supported by a NOAH, al the same dose for clinical signs of toxicity in the Co-critical rhesus monkey study ((ioldenthal et al.. 1978) Alternative Derivations of the RfD and RfC Values for 0 8 Critical Effect LeveP UF* UF,, IF, UF, UF,, 047 (LOAF.L in males) 0 72(flMt>l.) 1 (LOAF.L in males) 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 0 42 (BMI)I. in males)'1 10 10 10 10 10 1 6 (LOAF.L in females) 1 57 (RMI)I.) 0.7.1 (l)MDL in males) 3 (LOAF.L in males) 10 10 10 1 10 10 10 10 1 10 10 10 10 1 10 10 10 10 10 10 Composite UF* 100.000 100.000 100.000 100.000 10.000 10.000 10,000 100,000 RfD/RIC DWEI, L___________ 00000047 0 0000007 1 65 ppb 0 00001 0 15 ppb 0 0000042 1 is >t h 000016 000016 000007 0 00003 5.6 ppb 5.5 ppb 2 6 ppb 1 ppb 2- - DtVHA 0.3 ppb 0 Ii7 p: b It 'll p; |- 1 ppb i ppb 0 5 ppb 0 2 ppb 000134 T m EPA s IRIS Site - Drinking Water Health Advisories Page l or' A'nrted State* Envrronr.vental Protection Agency V"5* tt-f'jrin/fiun 01/17/92 BACKGROUND DOCUMENT 3 OFFICE OF DRINKING WATER HEALTH ADVISORIES The LTS EPA's Office o f Drinking W'ater (ODW) develops Health Advisories i HAsi for individual contaminants representing less-than-lifetime exposures o f One-dav. Tenday, Longer-term (an exposure o f several months up to 7 years), and for Lifetime exposures The HAs are developed from data describing noncarcinogemc endpoints of toxicity HAs developed for Lifetime exposures are based on the chemical's oral Reference Dose (RfD) (described in Background Document 1A l) HAs serve as informal technical guidance levels to assist public health officials when emergenev spills or contamination situations occur They are not legally enforceable and are subject to change as new information becomes available The HAs represent guidance levels for drinking water exposures The values for the One-day, Ten-day and Longer-term exposure periods do not consider other sources of exposure such as food or air. For each, the resulting value, in mg/L, assumes that 100 : o f an individual's exposure comes from drinking water The lifetime HA, calculated only for an adult from a chronic study, does take into consideration other sources of exposure by applying a relative source contribution (RSC) In the absence o f chemica,specific data, an RSC o f 10 is used for inorganic contaminants and an RSC o f 20 is used for organic contaminants (NAS, 1977) The HAs are derived employing an approach similar to that used to derive RfDs m that a NOAEL (or LOAEL) is divided by an uncertainty factor (LT) This value is adjuster for the body weight o f the protected individual and assumed daily water consumption The study selected for deriving a HA ideally employs an oral route o f exposure Therefore, a study where the chemical exposure is by drinking water is ideal Studies using dietary or gavage exposure are also acceptable. Inhalation data are only used instances where oral data are not available and by applying specific assumptions usea in route-to-route extrapolation The data used for HA derivations are generally from a study o f comparable duration the HA time period being calculated For a One-day H A the study time period shouu ideally be from a single exposure However, longer study durations may be acceptable if the data base is limited. The Ten-day HA may be calculated from a study o f less tha" or equal to 30 days. Developmental studies involving maternal exposure during pan . the gestational period have also been used Longer-term HAs are derived from subchronic studies where animals are exposed for approximately 10% o f their lifetime http //www epa gov/iriswebp/iris/dwater.html 05 3! : 000135 EPA s IRIS Sue - Drmwng water rteaun Advisories c- (for example, 90 days for rodents) The NOAELs or LOAELs are identified from studies that provide information on the target organ affected A study that demonstrates a dose-response relationship is preferred over a study where only a single dose has been tested Lethality studies are not considered for HA derivation HAs are derived to protect sensitive members o f the population. For the One-dav and Ten-day HAs, the protected individual is assumed to be a child. The child is assumed to weigh 10 kg unless otherwise noted It is also assumed that the child consumes i L o f water/day For a Longer-term exposure, HAs are calculated for both a child and an adult. It is assumed that an adult weighs 70 kg and consumes 2 L o f water/dav For a Lifetime exposure, the HA is only calculated for an adult, since the child will not be m this exposure category for a lifetime HAs are calculated according to the equation: HA = (NOAEL or LOAEL) (BW) / (UF) ( _ L/day) = _ mg/L where: NOAEL = No-Observed-Adverse-Effect Level (the exposure in mg/kg bw/day) or LOAEL = Lowest-Observed-Adverse Effect Level (the exposure dose in mg/kg bw/day) BW = assumed body weight o f protected individual (10 kg child or 70 kg adult) LT = uncertainty factors, based on quality and nature o f data __L/day = assumed water consumption (1 L/day for child or 2 L/day for adult) The uncertainty factor accounts for the inherent variability within the human population and between the animal species An uncertainty factor o f 10 is used when good acute or chronic human data that identify a NOAEL are the basis for the HA 1: the human data are o f good quality, but identify a LOAEL rather than a NOAEL. me an uncertainty factor o f 100 is used .An uncer- tainty factor o f 100 is also used when a NOAEL from a well-conducted animal study serves as the basis o f the HA. If an animal study that identifies a LOAEL, or an animal study o f limited quality is the basio f the HA, then an uncertainty factor o f 1000 is employed Use o f the uncertainty factor is largely judgmental, and may take into account the quality o f the toxicological data base, the significance o f the adverse effect or the counterbalancing o f possible beneficial effects. Lifetime HAs (noncarcinogens only) are calculated from the Drinking Water htp //www epa gov/iriswebp/iris/dwater html 05 3 1 I 00013 I EPA s LRib sue - Uriruung water heaitn aumow, Equivalent Level (DWEL) which, in turn, is based on the unrounded RfD DWELs are calculated from the equation: DWEL = (RfD) (BW) / ( _ L/dav) T, where RfD is the unrounded oral RfD derived in Section 1 o f the chemical file. and. BW an d __L/day are the adult reference values for body weight and water consumption, respectively Lifetime HAs assume that only a given percentage of the total compound intake is by drinking water and are derived by the equation Lifetime HA = DWEL x RSC where RSC = relative source contribution; the assumed exposure from drinking water Lifetime HAs are not derived for compounds potentially carcinogenic for humans because o f the difference in assumptions concerning toxic thresholds for carcinogenic and noncarcinogemc effects In situations where data are not available to calculate Longer-term HAs, the DUEL may be recommended for the Longer-term HA. For the child Longer-term HA, the DWEL may be recalculated using the child's weight and water consumption. For more information about Health Advisories, call the Office o f Drinking Water at (202)382-7571 or FTS 382-7571. Reference: NAS (National Academy o f Sciences) 1977 Drinking Water and Health Vol 1 Washington, DC te n d C annent] V C ** OrO H om e I'jq n H ome Home Piv;? Last updated 20 February 1998 URL. http /www epa htm http //www epa gov/iriswebp/iris/dwater html 000137 t