tce non-cancer and development risk: issues and … · archivo pdfconsultation august 2013 ......
TRANSCRIPT
Problem Formulation in Practical
Application Two major problem areas
1. The RfC was set at a very low environmental
concentration.
◦ This creates a significant change in how remedial decisions are
made in 10-5 states and how to address communication of risk
above the RfC
2. Understanding short-term developmental action
levels and how, when and where to implement action.
Problem Foundation
In 2011 EPA (NCEA) developed a Chronic Non-cancer
inhalation toxicity value termed the Reference
Concentration (RfC) at 2.0 ug/m3
The Chronic RfC was developed using two principal
toxicological studies with support from a 3rd
One of the studies was a highly controversial 22 day
short term “oral study” that evidenced developmental
toxicity including fetal heart malformations
Because the RfC is chronic and not a developmental
RfC or RfCdt there has been confusion regarding the
application
Developmental Risk
USEPA Region 09 stepped into the void and made use
of a 2008 Revised Removal Action Level EPA policy
memo that basically has a 3 times multiplier on the
screening level (RfC) that sets 6 ug/m3 as a target for
Removal Action (urgent action).
Many states and Regions are still uncertain and have
opted not to use the 3 times multiplier, taking urgent
action at the RfC or the Commercial screening level
Regions and many others requested EPA Headquarters
address the issue.
Headquarters
August 27, 2014 EPA Headquarters
Responded to the Developmental Risk
“Compilation of Information Relating to
Early/Interim Actions at Superfund Sties
and the TCE IRIS Assessment”
Blank Check Permission without position
Historically
EPA received considerable push back on the use of the study (Johnson et al 2003) as a part of the RfC derivation
There is good reason for the push back, the Johnson work has many confounding, toxicological and questionable science problems
What most fail to realize, unless you digest the volumes of work on TCE, is that there is a thread of connectivity all throughout a broad range of human and animal studies that evidence developmental effects.
Main EPA and State Regulatory
Result The RfC became the screening level driver at VI sites
in States where 10-5 is used
Short-term residential exposure concerns for
developmental risk became the major driver at many
TCE VI sites
Residential 10-6 10-5
TCE Cancer Risk 0.48 ug/m3 4.8 ug/m3
TCE Non-cancer
Risk
2.1 ug/m3 2.1 ug/m3
Residential Chronic Risk Short-Term <30
days Risk
Cancer 0.48 ug/m3 NA
Non-Cancer 2.1 ug/m3 2-6 ug/m3
Residential and Commercial
Screening Level Exposure
Parameter Differences
Residential
◦ 350 days/year
◦ 24hrs/day
◦ 30 years
Commercial
◦ 250-260 days/yr
◦ 8-12 hrs/day
◦ 25 years
Roughly a 3-5 times Residential to Commercial
multiplier depending on adult/child toxicity and
exposure, etc.
Commercial The NC screening level becomes the driver at
commercial sites in states using 10-5 risk
Short-term or urgent commercial exposure
concerns for developmental risk becomes the major
driver at many TCE VI sites
Commercial 10-6 10-5
TCE Cancer Risk 3.0 ug/m3 30 ug/m3
TCE Non-cancer
Risk
8.8 ug/m3 8.4-8.8 ug/m3
Commercial Chronic Risk Urgent or Short-
Term <30 days
Risk
Cancer 0.48 ug/m3 NA
Non-Cancer 8.8 ug/m3 8.4-24 ug/m3
General Concerns
EPA RAL is a numerical guideline with no
toxicity consideration
◦ Does not consider how steep or flat dose-
response curve above the RfC
States and Federal reactions appear to be
a response to public protection fears
where decisions are not based on sound
science but “safe” positions
What is needed?
Risk Managers perspective
◦ Because Fetal Heart Malformations are such a
serious consequence of exposure:
What short term exposure level should be used for
prompt action, how to measure, when take action?
◦ Need to understand risk management options
using the non-cancer endpoint
Balancing is no longer an option at 10-5 States or
anywhere
What is needed
◦ Need more clarity on risks and exposures,
cannot communicate what numbers mean
◦ Commercial/Industrial/Residential exposures
and liability
◦ Low exposure levels 0.48-2 ug/m3 are well
within or at the upper percentiles of indoor
air background.
Background Indoor Air Concentrations of Volatile Organic Compounds in North American Residences (1990–
2005): A Compilation of Statistics for Assessing Vapor Intrusion EPA 530-R-10-001 June 2011
Investigate Good Science and EPA
Policy Because not common to use NC in Risk
Management decisions:
◦ What does the non-cancer endpoint mean
and how should it be used in Risk
Management decisions?
Evaluate the Integrated Risk Information
Systems (IRIS) statement:
Reference Concentration as defined
by the Integrated Risk Information
Systems(IRIS)
“The RfC is defined as an estimate (with
uncertainty spanning perhaps an
order of magnitude) of a daily
exposure to the human population
(including sensitive subgroups) that is
likely to be without an appreciable risk of
deleterious effects during a lifetime.”
15
Risk Assessment and Risk
Management using NC
Excess Lifetime Cancer Risk (ELCR)
Range: 10-6 to 10-4
Given the IRIS definition, is a similar
evaluation possible with respect to the
non-cancer endpoint?
16
Common Past Risk Assessor
Approach Remedial Objectives
Cancer risk rules
◦ Cancer Screening, Remedial Objectives and
Health Effects Level are all established using
a risk range of 10-6 to 10-4
10-6 Screening
Departure
10-4 Effects Level
Remedial
Objective
Purpose of 10-4 to 10-6 Cancer Risk Range
Provides risk managers flexibility ◦ Screening level and closure (RSLTs)
◦ Majority are small sites not Superfund
Balance acceptable exposure levels with property transaction needs: Small sites ◦ Technical feasibility
◦ Implementability
◦ Timeliness
◦ Economic considerations
◦ Equity, consistency, administrative simplicity,…
Non-Cancer Risk
When considering a range NC has historically
viewed Non-Cancer RfC as a strict threshold
This is consistent with how we “used” to
calculate the RfC, commonly using NOAEL from
animal studies and uncertainty factors to establish
“human” safe dose
𝑯𝒖𝒎𝒂𝒏 𝑵𝒐𝒏 − 𝑪𝒂𝒏𝒄𝒆𝒓 𝑬𝒙𝒑𝒐𝒔𝒖𝒓𝒆 𝑳𝒆𝒗𝒆𝒍 = 𝑵𝑶𝑨𝑬𝑳
𝑼𝒏𝒄𝒆𝒓𝒕𝒂𝒊𝒏𝒕𝒚 𝑭𝒂𝒄𝒕𝒐𝒓𝒔
Historical Risk Assessor Non-Cancer
Understanding NOAEL implies that any exposure level
above this value, must have possibility of
an adverse effect
Allowed to exist because no real past
impact
Strict Yes/No threshold at RfC overly
simplistic understanding
ARA
Given the IRIS definition and changes to
RfC derivation process, seemed plausible
to explore the science and policy basis of
a risk range concept for NC chemicals
Petitioned the Alliance for Risk
Assessment to investigate this concept
ARA non-profit part of Toxicology
Excellence for Risk Assessment
Addressed the Question to ARA
Formed workgroup of state, federal and
private members
Evaluated all literature and guidance on
non-cancer risk
Historically the order of magnitude was
intended to address uncertainty similar to
how cancer risk was addressed
Developed range guidance entirely
consistent with EPA Guidance
Results and Reasoning
That is, the RfC/RfD is expected to be below the actual threshold for adverse effect in a sensitive subgroup.
The RfC is therefore the bottom or the “floor” and the “spanning perhaps an order of magnitude” must be above this value
Evaluating the individual studies used to derive the RfC determined that a range was reasonable
Hazard Range
Is a judgment that meshes four
considerations:
oCollective magnitude of the margin of
safety(UFs)
oSteepness of the hazard slope describing
exposures above the RfC/RfD
oThe confidence in the selection of the critical
effect
oThe confidence in the animal dose used to
extrapolate to humans (POD)
24
TCE Non-Cancer Risk Range
RfC defined as: an estimate with uncertainty
“spanning perhaps up to an order of
magnitude” of a continuous…
2.1 ug/m3
Non Cancer
Screening
9 ug/m3
Remedial
Objectives
Level
21 ug/m3
Health
Effects
Level
Solves
Risk Manager does have some flexibility
to make risk based decisions (range)
Communicate meaning of exposures
above the RfC/RfD (range placement)
Guidance on prompt action, immediate
concern levels at the “effects level”
considered the “ceiling”
Complete Range and Risk
Management Analysis
Alliance for Risk Assessment:
Guidance for Contaminated Sites:
Trichloroethylene (TCE) Risk
Assessment at:
http://www.allianceforrisk.org/index.htm
Is 21 really safe to use as a short
term number? Are there other ways to evaluate the 21
ug/m3 effects level
How can we vet or increase confidence in the 21 ug/m3
◦ How have other national health agencies responded?
◦ What is the evidence linking TCE and developmental risk?
◦ Was there a wide margin of safety used in the determination?
◦ Are we regulating as a threshold response?
Developmental Risk Perspective
Other Agencies Acceptable short term exposure levels
◦ Considerable disagreement between ATSDR,
EPA-Superfund, EPA-TSCA
Regulatory Comparison of Acceptable TCE Exposure Levels
All units in ug/m3
Regulatory Body Residential Short-
term Action Levels
Commercial
Short-term
Action Levels
ARA Range and
EPA Health
Effects Level
(HEC99)
USEPA Region 091 (2014) 6 24
USEPA Region 10 (2012) 2 8.4
New Hampshire (2013) 2 8.8
ATSDR (resident al , 2013)
Short-term action levels 21 ~60-88! 21
USEPA AEGLs (8 hr) (2013) 413,816
ACGIH TWA (8 hr) (2007) 53,742
NIOSH 10 hr TWA (2011) 134,356
OSHA (100 ppm) 537,423
.
88 ug/m3 after ATSDR and approximating a 40 hr work week, and consistent
with Region 09 reasoning, 60 is provide based on 24 hour event.
OSHA/ACGIH/NIOSH
EPA screening levels are typically 1-4
orders of magnitude lower than OSHA
OSHA PELs are legally binding but
established many years ago using ACGIH
values
Commonly ACGIH/NIOSH values are
used as workplace guidelines
Industrial/Commercial
OSHA/ACGIH
◦ Risk 10-3 to 10-4
◦ Some acceptable impact to sensitive populations
◦ Relies more on human studies no extrapolation
◦ 40 year ED
◦ Greater exposure awareness, protective training
EPA
◦ Risk 10-4 to 10-6
◦ Protects sensitive population
◦ Widely uses animal data
◦ Extrapolation from animal to human very conservative
◦ 25 year ED
ATSDR Millsboro, DE TCE Health
Consultation August 2013
“EPA’s reference dose and reference concentration are both intend[ed] for comparison to chronic or longer duration exposure scenarios”
“Of note, a suitable comparison value does not yet exist for the intermediate duration (one year or less) of exposure that was experienced in Millsboro. Therefore, ATSDR compared the estimated exposure doses with effect levels from available studies.”
ATSDR used effects levels: human equivalent dose (HED99) for ingestion and the human equivalent concentration (HEC99) for inhalation during showering.”
ATSDR Millsboro, DE Health
Consultation August 2013
Exposure 24 hour or longer?
◦ ATSDR used one time daily shower event,
assumed no exposure the rest of the day and
averaged the one event across a 24 hour time
period concluding:
◦ “The HEC99 of 0.021 mg/m3 was exceeded by
all age groups ……. This suggests that there
may be an increased likelihood of adverse fetal
cardiac effects.”
What is HEC99
The HEC99 is the human exposure
concentration for which there is a 99%
likelihood that a randomly selected individual
will have an internal dose less than or equal to
the 1% response rate in the animal (BMDL01).
What does this mean in the real world?
I Internal Dose
Model
Internal Dose
Model -Animal
Dose-
Response
Model
Internal Dose
Model-Human
Use model to
predict internal
concentration as a
function of applied
or exposed dose
Now, RfC generally modeled value
Exposure dose in
humans that
insures internal
dose in humans
below the
selected
“protective level”
WOE Highlights
There are no data evidencing FHM as a
critical effect by inhalation, despite several
well conducted studies
The only animal studies evidencing FHM
results were conducted by a single
laboratory, they were oral studies
extrapolated to inhalation by modeling
◦ Direct attempts to duplicate the FHM
study (Johnson et al 2003) results using
both a subsequent oral and inhalation
study were negative
◦ The dose response, controls and
assessment relationships from Johnson et
al. (2003) were particularly confounded. Intermediate dose had lower response than low dose
High dose 4400 times conc., but only 1.5 times increase
in response
Sampling Considerations
Animals exposed for 22 days—entire
pregnancy
In Humans the fetal heart development
occurs across a 24 day time period
Sampling Period /Averaging Times
Region X has taken the position: “average
exposures over any 21-day period of time not
to exceed the [recommended] concentrations
in air…..
Region 09 has stated: “we did not feel the
science to date supports taking action on 24
hour exposures for TCE”
Sampling Period /Averaging Times
◦ EPA developmental guidance states that even a
single exposure may be sufficient to cause a
developmental effect
◦ ATSDR makes the point that they are averaging
residential exposure across a 24 hour period and
treat the 24 hour unit as the complete exposure
period (do not adjust for time away from
structure, etc.)
◦ The critical effect study also used 24 hour
averages
VI Sampling Concerns Residential
Practical Application Single 24 hour results 6-21 ug/m3 take
prompt action (up to a few weeks)
If above 21 ug/m3 take immediate action
(days)
If State and RP feels it is warranted to
take a 21 day average then:
Three (3) choices
Deploy passive samplers for 24 or more days assume average is representative of episodes
Take 8-10 summa canister samples
◦ Compare peak 24 hr levels to 21 ug/m3
◦ Also compute UCL and compare to 21 ug/m3
These methods do not help you to understand peaks or episodes; however,
Use of a continuous recording instrument for 21-28 days.
◦ See all episodic events
◦ Still Compute average and UCL of 24 hour exposures
This and the following three slides contain references for
Regulatory Comparison of Acceptable TCE Exposure Levels Table on slide 10
USEPA Region 091: EPA Region 9 Interim Action Levels and Response
Recommendations to Address Potential Developmental Hazards Arising from
Inhalation Exposures to TCE in Indoor Air from Subsurface Vapor Intrusion June 30,
2014 as provided in: EPA Region 9 Response Action Levels and Recommendations
to Address Near-Term Inhalation Exposures to TCE in Air from Subsurface Vapor
Intrusion from Enrique Manzanilla, July 9, 2014
USEPA Region 10 values taken from Dec 12, 2012 Memorandum: OEA
Recommendations Regarding Trichloroethylene Toxicity in Human Health Risk
Assessments
TSCA: Data taken from TSCA Workplan Chemical Risk Assessment for
Trichloroethylene: Degreaser and Arts/Crafts Uses CASRN: 79-01-6 Ethene, 1,1,2-
trichloro. This is considered a “light commercial exposure.”
Here TSCA used a margin of safety approach to make risk management
recommendations. Similar to ATSDR, TSCA used the HEC99 as the effects level.
However, TSCA used only data from the inhalation studies. TSCA did not use the
USEPA 2011 TCE oral to inhalation modeling extrapolations in their risk assessment.
Health effects levels were divided by the Exposure levels and then compared to a
Margin of Exposure (MOE) of 30 to determine acceptable risk. MOE based on a
factor of 10 for intraspecies variability times an uncertainty and a factor of 3 for the
pharmacodynamic portion of the interspecies extrapolation factor; the latter being
reduced based on the kinetic modeling performed to arrive at an HEC (at page 61 of
TSCA Risk Assessment)
The use of the MOE is conceptually similar to standard USEPA practice of using
uncertainty factors in the derivation of the RfC. Using the following equation:
MOE acute or chronic = Hazard value (POD)/Exposure value (pg 60)
The acceptable exposure screening level is determined by dividing the POD by the
MOE (POD/MOE = Acceptable Exposure Level).
Acute levels taken from the lowest HEC99 from acute studies listed in TSCA
conceptually consistent with the ATSDR Millsboro approach (see slide 10).
AEGL-1 information from
http://www.epa.gov/oppt/aegl/pubs/define.htm is the airborne
concentration, expressed as parts per million or milligrams per cubic
meter (ppm or mg/m3) of a substance above which it is predicted that
the general population, including susceptible individuals, could
experience notable discomfort, irritation, or certain asymptomatic
nonsensory effects. However, the effects are not disabling and are
transient and reversible upon cessation of exposure. TCE has Interim
AEGLs 77 ppm is most sensitive of all the AEGL categories. Note that
“AEGLs are intended to describe the risk to humans resulting from
once-in-a-lifetime, or rare, exposure to airborne chemicals”
ACGIH STEL is short term exposure limit from 2010 ACGIH published
values; ACGIH TWA is from same reference.
Agrees with or less than NIOSH 10 hr. TWA see:
http://www.cdc.gov/niosh/npg/nengapdxc.html
New Hampshire values taken from February 7, 2013 Waste Management
Division Update RE: Revised Vapor Intrusion Screening Levels and TCE
Update
ATSDR Health Consultation Millsboro TCE Millsboro Delaware.
February 13, 2013 U.S. Department of Health and Human Services
Agency for Toxic Substances and Disease Registry Division of
Community Health Investigations Atlanta, Georgia 30333
At page 20: “Of note, a suitable comparison value does not yet exist for
the intermediate duration of exposure that was experienced in
Millsboro. Therefore, ATSDR must compare its estimated 24-hour
concentrations with effect levels from available studies.
Also at page 20: …..“to obtain a 99th percentile HEC99 of 0.021
mg/m3” and at page 21: “ATSDR compared the preceding HEC99 with
the estimated 24-hour average concentrations for men,
women, and children at the Millsboro site to evaluate the potential for
adverse health effects resulting from past [intermediate duration]
exposure while showering “