DSHW-2014-006136 - 0901a06880428823

ATK Launch Systems
P.O. Box 707
Brigham City, UT 84302
www.atk.com
April 15,2014
8200-FY15-005
Mr. Scott T. Anderson, Director
Utah Department of Environmental Quality
Division of Solid and Hazardous Waste
195 North 1950 West
P.O. Box 144880
Salt Lake City, Utah 84114-4880
Re: ATK Launch Systems Inc., EPA ID #UTD009081357, Recommendation on Selection of
Toxicity Values for Risk Assessments
Solid and Hazardous Waste
APR 1 6 2014
Dear Mr. Anderson:
ATK is in the process of performing corrective actions that include risk assessments and cleanup
actions. This regulatory process requires using toxicity values to determine constituents that may
pose an unacceptable risk.
ATK is currently investigating soil sites with potential chromium (VI) contamination. The
recommended cleanup value ATK received from the Division is a screening level found in the
EPA Regional Screening Levels (RSLs) table. RSL table chromium (VI) values are based on
toxicity data developed in New Jersey and are different from values currently found in the EPA
Integrated Risk Information System) (IRIS) database. ATK questions if it is appropriate to use
RSL values for this application, and recommends following UAC R315-101.5.3(a)(6), which
specifies, in order of preference, first, use of the IRIS database. Similarly, EPA guidance1
specifies the recommended hierarchy of toxicological information sources, which regional risk
assessors and managers should initially consider for site-specific risk assessments:
• The first source is the toxicity values retrieved from the U.S. EPA's IRIS database (Tier I).
IRIS contains toxicity values that have been reviewed by different divisions within the
EPA, nationally recognized scientists within the National Science Foundation and the
National Academy of Sciences, and university academics conducting research on the
chemical of interest.
• If there is no data available in IRIS, then Tier II sources can be used such as Provisional
Peer Reviewed Toxicity Values (PPRTV).
• The last option is to use Tier III sources, such as those developed by state agencies.
The RSL table takes into account toxicity values from many sources, including Tier III, and then
typically lists the most conservative value. Specifically, the chromium (VI) ingestion cancer slope
factor used in the RSL tables is adopted from the State of New Jersey. The basis for this cancer
slope factor has been reviewed by EPA's science advisory panel, and was found to be unsupported
by data for the purpose of revising the IRIS file.2 Since the New Jersey value is significantly
lower, its use also results in significantly more costs to analyze and cleanup chromium related
sites. ATK recommends following the regulatory hierarchy process as stated in Utah rules, and in
the referenced EPA guidance.
ATK believes it is important to assure that the best science is used, and to establish a consistent
approach that avoids confusion, unnecessary effort and costs for the regulated community. We
understand that the lower chromium (VI) values have been used at other sites in the State. For
future activities, ATK recommends following regulatory guidance especially since revising the
dose-response factors selection process would require a compound-by-compound confirmation
process with the Division. Attached is a report that provides more details of the chromium (VI)
information summarized above.
We appreciate your consideration of our request. Please refer question on this recommendation to
Paul Hancock at [phone] or me at [phone].
George E. Gooch
Manager, Environmental Services
ATK Launch Systems
1. Human Health Toxicity Values in Superfund Risk Assessments, Memorandum from Michael; B Cook, Office of Superfund
Remediation and Technology Innovation, to Superfund National Policy Managers, Regions 1 - 10, December 2004
2. Toxicological Review of Hexavalent Chromium, US Environmental Protection Agency, EPA/63 5/R-107004C,
April 2010. Peer Review Workshop for EPA's Draft Toxicological Review of Hexavalent Chromium, Reviewer
Post-Meeting Comments, July, 2011.
Sincerely,
TERRA MENTIS
Inter-office Memorandum
To: Paul Hancock, Project Manager, ATK
From: Stephen Foster, Terra Mentis
Date: February 19,2014
Subject: Clarification on the Selection of Dose-response Information by the Utah
Division of Solid and Hazardous Waste
The purpose of the memorandum is to seek clarification on the process by which the Utah
Division of Solid and Hazardous Waste selects dose-response factors for use in risk
assessments in Utah. This memorandum illustrates the question by providing two relevant
examples: trichloroethene and hexavalent chromium.
SELECTION PROCESS
The current process for selecting Dose-Response Factors (DRF) for any chemical at a
hazardous waste site regulated under Superfund, RCRA and other waste site regulations is
based on a three-tiered process identified in a US Environmental Protection Agency 2004
memorandum (EPA, 2004), and states "This memorandum specifies the recommended
hierarchy of toxicological information sources, which Regional risk assessors and managers
should initially consider for site-specific risk assessments." First, the DRF is retrieved from the
US EPA's Integrated Risk Information System (IRIS) database (Tier I). IRIS contains DRFs
that have been reviewed by different divisions within EPA, nationally recognized scientists
with the National Science Foundation and the National Academy of Sciences, and university
academics conducting research on the chemical of interest. This memorandum and the
supporting documentation in IRIS indicate that IRIS is the best source of DRF in the US.
EPA 2004, states that, "IRIS normally represents the official Agency scientific position
regarding the toxicity of the chemicals based on the data available at the time of the review."
1
TERRA MENTIS
Second, if no information is present in IRIS, then the Provisional Peer Reviewed
Toxicity Values (PPRTVs) (Tier II) are used to provide the DRF.
Finally, if the relevant data is still not provided in either of these sources, the next
source of dose-response should be other State Agencies and EPA Agencies; EPA 2004
names the California EPA (Cal EPA), the Agency for Toxic Substances and Disease Registry
and other credible sources (Tier III). EPA provides information for the selection of DRF
nationwide to help standardize procedures for risk assessors. EPA also states, "Consultation
with the STSC or headquarters program office is recommended regarding the use of the Tier 3
values for Superfund response decisions when the contaminant appears to be a risk driver for the
site."
Hexavalent Chromium
Hexavalent chromium (Cr(VI)) is listed in IRIS (EPA, 2014) where it states in section
"II.A.l. Weight-of-Evidence Characterization, under the current guidelines (EPA, 1986),
Cr(VI) is classified as Group A - known human carcinogen by the inhalation route of
exposure. Carcinogenicity by the oral route of exposure cannot be determined and is
classified as Group D."
Cr(VI) was listed as carcinogenic by the inhalation route in 1998. Subsequent to this
classification, a drinking water study was conducted under the National Toxicology Program
(NTP, 2008). It was used by the New Jersey Department of Environmental Protection
(NJDEP, 2009) to derive a cancer DRF for application in New Jersey (NJDEP, 2009).
Authored by Dr. Alan Stern, it considered Cr(VI) carcinogenic by ingestion, and derived an
oral cancer slope factor (SF0). In the case of chromium, the NJDEP identified an experiment
from the literature with a different DRF from that ofthe IRIS DRF. Dr. Stern's DRF was
picked up by the group within the EPA that lists DRF on the Regional Screening Level
2
TERRA MENTIS
(RSL) tables, and they listed Cr(VI) as a carcinogen by ingestion. The RSL group does not
peer review values before they are entered. In this case, Tier III data is used over Tier I data
against EPA's toxicological hierarchy (EPA, 2004).
The US EPA was also aware of the study used by Dr. Stern, and issued a revised
toxicological profile in support of a proposed revision to IRIS (EPA, 2009); the EPA
prepared and issued a profile for public comment. The review identified a number of
problems with the toxicological profile, and more specifically, identified problems in the
research, and its application to environmentally relevant doses required under Superfund
(EPA, 2010). When a revision to IRIS is considered, new research is reviewed; IRIS holds a
public forum for review, including scientific, inter-agency and peer review. Upon review, the
research will either be incorporated or excluded from an updated IRIS report. The revised file
is placed under review once again before publication for use on the IRIS database. For
hexavalent chromium by ingestion comments were critical of the profile, and indicated that
Cr(VI) acts via a threshold mechanism, it is not mutagenic, and questions the conclusions of
the IRIS toxicological review. In 2011, (EPA, 2011) the EPA reviewed these comments, and
others, and current research, including the 2009 NJEPA report and decided to archive the
toxicological profile, and the associated listing in the IRIS database, and no updated
document DRF has been added to IRIS.
In order for new information to be included into the IRIS database, it must be
approved by the US EPA and then will be incorporated in an updated release of the IRIS
entry. In the case of Cr(VI), the first IRIS release was in 1998. The original document was
based on meta-analysis of experiments on mice and rats as well as experiments focusing on
the post exposure results on humans. The NJEPA released a memorandum in 2009 with the
3
TERRA MENTIS
results showing a lower DRF then the IRIS database (Stern, 2009). However, subsequent to
the NJDEP report, the EPA deemed the data inadequate to establish a cancer slope factor for
Cr(VI), a factor which should have been taken into account when the RSL group was listing
their value. It should be noted that the use of the RSL tables is for screening, not establishing
a site-specific DRF. EPA, 2004, states that, "Consultation with the STSC or headquarters
program office is recommended regarding the use of the Tier III values for Superfund response
decisions when the contaminant appears to be a risk driver for the site." Terra Mentis contacted
EPA's Region 8 Superfund toxicologist, Dr. Robert Benson, to determine how EPA Region deals
with this issue, and Dr. Benson responded that they wait until IRIS is officially revised before
acting (EPA, 2014).
Based on this review of the literature for Cr(VI), and on Dr. Benson's comments, we
believe it is important to follow the three-tier approach identified by the EPA for selecting the
DRF, as these primary sources have receive the most peer review. The primary source of the
DRF is the EPA's IRIS database, followed by Tier II sources. We believe Tier III DRF
sources should be used primarily if there are no DRF available in Tier I or II.
Where toxicological studies are conducted after a DRF is issued in IRIS, the study and
associated DRF may be appropriate for inclusion. In the case of Cr(VI), the EPA's review of the
2008 dose-response data concluded that there was no sufficient data to support a "carcinogenic
by ingestion" designation, and the non-linearity in the dose-response did not allow for a cancer
slope factor to be developed. The EPA archived the toxicological review.
Trichloroethene
Trichloroethene (TCE) is a volatile organic compound that has been the subject of many
investigation at hazardous waste sites because it was commonly used by industry as a degreasing
4
TERRA MENTIS
solvent. In 2001, EPA issued an External Draft Risk Characterization for TCE, that identified
TCE as "highly likely to produce cancer in humans" based on studies of adults (EPA, 2001).
This was followed by a Review draft of the document for TCE: the "Review of Draft
Trichloroethylene Health Risk Assessment: Synthesis and Characterization: An EPA Science
Advisory Board Report (EPA, 2002) that provided a SF0 and an Inhalation Unit Risk (IUR)
factor for TCE. These DRFs were never uploaded to IRIS because the peer review process
identified problems with the studies and methodology used for Cancer Slope Factor
development. The draft DRFs were adopted by the RSL tables, and some other States.
However, other states agencies, including Indiana and Cal EPA agreed with the criticism of
the 2002 draft risk characterization, and developed their own DRFs. Some state agencies,
including Utah DEQ, used the pre-existing Slope Factors until the US EPA resolved the
problems with the risk characterization. The Cal EPA DRFs were subsequently incorporated
into the RSL tables, and the US EPA rejected the 2001 study before developing another
toxicology review, and associated DRFs. In 2011, following the EPA's review process, the
revised toxicology profile was accepted by EPA and the scientific review committee, the
profile was loaded into the IRIS database (EPA, 201 lb).
As noted above, the Utah DEQ used the older slope factor rather than quickly adopting a
value that was later withdrawn (ATK, 2014).
While using a lower slope factor for a chemical may be more conservative, it can lead
to the installation of remedies that are unnecessary or require a risk level that is not
appropriate. In some cases attaining a lower goal may require special remedial measures or
options that would otherwise not be required.
5
TERRA MENTIS
CONCLUSION
EPA (2004) identifies a hierarchy for selecting DRFs. In some cases the RSL group
selects a DRF that is not Tier I, but is a Tier III level source. Many of these Tier III DRFs
are developed to support a particular State Agency goal or for a particular site within a State.
These idiosyncratic DRFs may be developed by a particular individual and may not have
received the professional or scientific peer review that is appropriate for regulatory goal
setting, and that would require significant financial commitment to achieve. Cr(VI) is a good
example because the cost of achieving 0.29 mg/kg (EPA, 2013-cancer based goal) is
significant compared with the RSL non-cancer soil goal of 230 mg/kg (EPA, 2013-non-
cancer based goal). Cal EPA values are often developed because of California Proposition
65, which states that any chemical that has the potential to cause cancer should reach a risk
goal of lxlO"5. This Proposition 65 approach may not be appropriate for Utah.
For future projects that will be reviewed by the Utah DSHW, it is important that risk
assessment guidelines are clear, especially if Tier III values will take precedent over Tier I
values.
6
TERRA MENTIS
WORKS CITED
ATK, 2014 Personal Communication from Paul Hancock concerning the use of TCE at
waste sites in Utah
EPA, 1998 Toxicological Review of Hexavalent Chromium, USEPA
EPA, 2001 Trichloroethylene Health Risk Assessment: Synthesis and Characterization
(External Review Draft), U.S. Environmental Protection Agency, Office of Research
and Development, EPA/600/P-01/002A, August
EPA, 2002 Review of Draft Trichloroethylene Health Risk Assessment: Synthesis and
Characterization: An EPA Science Advisory Board Report, U.S. Environmental
Protection Agency, Office of Research and Development, EPA-SAB-EHC-03-002,
December
EPA, 2004 Human Health Toxicity Values in Superfund Risk Assessments,
Memorandum from Michael; B Cook, Office of Superfund Remediation and
Technology Innovation, to Superfund National Policy Managers, Regions 1-10,
December
EPA, 2010 Toxicological Review of Hexavalent Chromium, US Environmental
Protection Agency, EPA/63 5/R-10/004C, April
EPA, 2011 Peer Review Workshop for EPA's Draft Toxicological Review of Hexavalent
Chromium, Reviewer Post-Meeting Comments, July
EPA, 201 lb Toxicological Review of Trichloroethylene, U.S. Environmental Protection
Agency, Office of Research and Development, EPA/635/R-09/01 IF, September
EPA, 2013 Regional Screening Levels Tables, USEPA on-line spreadsheets, November
EPA, 2014 Personal communication for Dr. Robert Benson, US EPA Region 8
NJDEP, 2009 Derivation of Ingestion-Based Soil Remediation Criterion for Cr+6 Based on
the NTP Chronic Bioassay Data for Sodium Dichromate Dihydrate, Alan Stern,
Division of Science, Research and Technology New Jersey Department of
Environmental Protection, for the Risk Assessment Subgroup of the NJDEP
Chromium Workgroup, April
NTP, 2008 National Toxicology Program Technical Report on the Toxicology and
7
TERRA MENTIS
Carcinogenesis studies of Sodium Dichromate Dihydrate (CAS No. 7789-12-0) in
F344/N Rats and B6C3Fi Mice, Drinking Water Study, NTP Publication 08-5887,
July
EPA, 2014 Hexavalent Chromium Integrated Risk Information System, on-line
database, Accessed 2014, carcinogenicity listed in file in 1998.
8