IN THE

MATI’ER

OF:

PROPOSED

AMENDMENTS

TO

TIERED

APPROACH TO CORRECTIVE ACTION

35

ILL. ADM. CODE 742

)

)

)

)

RO0-19(C)

)

(Rulemaking-Land)

)

RECEiVED

CLERK’S

OF~V’)~

OCT

1

1

ZULU

•

STATE

OF ILLINOIS

PoJl~tto~

Control

Board

Dorothy M.

Gunn, Clerk

Illinois Pollution Control Board

James

R. Thompson Center

100

West Randolph

Street,

Suite 11-500

Chicago,

Illinois

60601

Robert

Lawley,

Chief Legal Counsel

Dept. ofNatural Resouróes

524 South Second Street

Springfield, Illinois

62701-1787

See Attached Service List

Matthew

J.

Dunn,

Chief

Environmental Bureau

Office of

the Attorney

General

188

W. Randolph,

20t1~

Floor

Chicago,

Illinois

60601

Amy

Jackson, Hearing Officer

Illinois Pollution Control Board

500 South Second

Street

Springfield,

Illinois

62706

NOTICE OF

FILiNG

PLEASE TAKE NOTICE that I have

filed

today with the Illinois Pollution Control Board the Supi~lernental

Comments

and Exhibits

ofthe

ENVIRONMENTAL

PROTECTION AGENCY,acopyofwliichis herewith

servethiport

you.

ENVIRONMENTAL

PROTECTION

AGENCY

OF

TE

STATE

OF

ILLiNOIS

Dated:

October

8, 2001

Illinois Environmental Protection

Agency

1021

N. Grand

Ave.

E.

P.O. Box

19276

Springfield, Illinois

62794-9276

(2

17/782-5544)

BEFORE

THE

POLLUTION CONTROL

BOARD

OF THE

STATE OF ILLINOIS

NOTICE OF

FILiNG

KimberijA.Geving

Assistant Counsel

Division

of

Legal Counsel

---

•

RECEIVED

CLERK’S OFF~’~

•

OC11~~Qüi

BEFORE THE ILLiNOIS POLLUTION

CONTROL BOAI~TEOF

ILLINOIS

•

•

•

•

•

Pollution control Board

iN THE MATTER OF:

•

)

)

PROPOSED AMENDMENTS TO TIERED

)

ROO-19(C)

APPROACH TO CORRECTIVE ACTION

)

(Rulemaking-Land)

OBJECTIVES (TACO)(MTBE):

)

35

ILL. ADM. CODE 742

SUPPLEMENTAL COMMENTS

The Illinois Environmental Protection Agency (“Illinois EPA”), by its attorney, Kimberly

Geving, and at the request ofthe Illinois Pollution Control Board (“Board”) in its September 6, 2001

First Notice Opinion and Order in the above-captioned matter, respectfully submits these Supplemental

Comments to the Board.

In its discussion ofthe Proposed Amendments regarding MTBE that were sent to First Notice via

the September

6th

Opinion and Order, the Board stated that “While the Agency has provided the Board

with information supporting the proposed standards, the record is lacking a detailed explanation ofthe

calculations employed by the Agency in reaching the proposed numbers.”

(Proposed Rule. First Notice

Opinion and Order dated September 6, 2001

at pages

4-5).

Page

5 ofthe Board’s Opinion and Order

specifically requested the illinois EPA to provide supplementation for its MTBE proposal during the first

notice period.

The Illinois EPA maintains that its proposal

was

technically substantiated on the record.

However, in the interest ofestablishing a more complete, technically sound record, the Illinois EPA

offers two attachments that we believe further explain how the objectives were established.

The first

attachment (Exhibit 1)’ provides a very detailed description ofthe Health Advisory that was proposed by

‘Exhibit

1

was

also submitted to the Boardthis

year

during

the

Part 620 regulatoryproceedings.

In

that

proceeding, it was labeledas Exhibit V to the Illinois EPA’s Statement ofReasons in R0l-14.

1

---

the illinois EPA in 1994 forMTBE and the scientificjustification forthe advisory.

The Health Advisory

served

as a base for determining remediation objectives for groundwater in this proceeding.

Exhibit

1

explains in detail how the numbers for MTBE were derived.

Additionally, Exhibit

1

includes Illinois

EPA responses to significant comments that were receivedregarding the health advisory proposal for

MTBE, further substantiating its scientific basis.

Exhibit 2 provides supplementation for how the illinois

EPA calculated the soil rernediation objectives for MTBE in

Part 742.

The Illinois EPA maintains that its proposed remediation objectives for MTBE in both soil and

groundwater have been scientificallyjustified.

The illinois EPAhopes that these Supplemental

Comments and attachments further clarify for the Board how the calculations were performed.

WhEREFORE, the Illinois EPA submits these Supplemental Comments to the Board for its

consideration

and

respectfully requests the Board to

adoptthe objectives proposed by the Illinois EPA in

theirentirety.

•

ILLiNOIS ENVIRONMENTAL

PROTECTION AGENCY

•

•

By~~th(L/4d4~Z

•

A5S~5~fltCOUlLl

if

•Dated:

October

5,

2001

1021 North Grand Ave. East

P.O. Box 19276

Springfield, illinois

62794-9276

•

(217)782-5544

•

THIS

FILING

SUBMITTED ON RECYCLED PAPER

2

---

Page

18/July,

1994

Eitviinnmen.za!

NOTICE OF

HEALTH ADVISORY FOR

•

METHYL

TERTIARY-B

UTYL ETHER

(MTBE)

_________

•

•

EXHIBIT

Prepared

by

•

•

Office

of

Chemical

Safety

•

-

Illinois EPA

June9,

1994

REASONFOR

ACTION

As

a

result

of

routine

monitoring

of

public

water

supply systems,

the

gasoline

additive

Methyl

Tertiaxy-Butyl

Ether

~MTBE) has been

detected

at

least in two

public

water supplies.

Therefore,

the

Illinois Environmental

Protection

Agency

(Agency) is announcing its intention to issue a health advisory, pursuant to 35

illinois Administrative Code

Part 620

Subpart

F:

Health Advisories,

for Methyl

Tertiary-Butyl

Ether.

According to Section

620.605 of Subpart F,

the

Agency shall

issue

a health advisory

for

a chemical

substance if all of

the following conditions are

met:

1)

A communny

water

supply well

is sampled

and a substance is

detected and confirmed

by

resampling;

2)

There

is no

standard under Section

620.410

for

such

chemical

substance;

and

3)

The

chemical

substance is

toxic or harmful to human health according

to the

procedures of Appendix

A,

B,

or

C.

-

The Agency

has determined

that

all

three

conditions have been

met,

prompting the

issuance of

this draft proposal

for

a

health advisory.

By

this issuance,

the

Agency is

opening

a

30-day

public

comment period, until Au~iist22.

1994,

regard-

ing

this health advisory draft.

Upon closing the

public

comment

period,

the

Agency

will

consider all

comments

received

and amend

the

health advisory if warranted.

The final health advisory will then

be published in the Environmental Register

(the

flhi.nois Pollution Control Board News)

with responses to

comments

received.

An abbreviated version of the final

health

advisory will also

be published in local newspapers which serve

communities in whose public

water

supply systems MTBE

has been detected.

PROPOSED

GUIDANCE

LEVELS

Section

620.605 of Subpart

F prescribes the

methods for developing health advisories for carcinogens and no~ncarcino-

gens.

Since

the

Agency

has

determined

that

there. is

insufficient

evidence of

the

carcinogenicity

of MTBE

at

this

time

(discussed

in

the

attachment

to

this

notice),

the

method

for

developing

a

health advisory

for

noncarcinogens

was

used.

Briefly, this method

specifies that

the USEPA’s maximum contaminant level

goal (MCL.G)

is the guidance level, ifavailable,

or

the

human

threshold

toxica.nt advisory

concentration (W~TAC)must

be determined

using the procedures

contained

in

Appendix

A of Section 620.

USEPA has not published an

MCLG for

MTBE, therefore

the

Agency

used

the

Appendix

A

procedures to calculate

the HTTAC.

Appendix A specifies in

prescribed order

the

toxicological

data

to be used in developing the 1-ITTAC,

ranging from a

verified

Reference

Dose developed

by

USEPA to a laboratory

animal

study of subchronic

duration in which only

a lowest

observable

adverse

effect level

(LOAEL)

has

been

determined.

This preferred

order reflects increasing

uncertainty in the

toxicological

database

regarding

a

chemical’s

potential

to

cause

adverse

health

effects

in

humans,

and

is

manifested

in

increasingly large safety factors which are applied to the data to calculate the HTTAC (maximum

10,000-fold safety factor).

In

the

case

of MTBE,

the

Agency

has selected

the only study available

in

which

the

test animals

were exposed by

the

oral

route of exposure

as the

basis

for the HTTAC.

Among other findings, this 90-day

subchrooic study reported increases

-..3

..._,-;~...._.

~

;,.

—*1

~

~

:.,..~,,4:....,

,L

1,,.,,,,..

1.,~..

...C

irv~

_.,t~....l,l

~

,

~.,,I.

r~

using this subchronic study in

which only

a LOAEL

was

determined,

the

language

of Subpart

F

specifies the

application of

safety

factors totalling to

10,000

to

the

animal

data,

resulting

in the HTI~ACguidance

level

of 0.07

mgI!,

or

70

parts per

billion

(ppb).

The details of

the derivation of the HTTAC

are

presented in

the attachment

to

this

notice.

At

this point

it

is

necessary

to discuss an aspect of the

evolving science of risk

assessment

which

has a bearing

on

this

notice.

The

Agency

has

been

informed

verbally,

by

USEPA personnel

that

in

most

cases

USEPA

no

longer

favors

the

---

Environme!Lta!

Register No.

484

July,

1994/Page 19

calculation of acceptable

exposure

values

for

humans

by

using laboratory a.aimal

data divided

by

uncertainty

factors

totalling

~

10,000.

This

preference

will

be

included

in

a

chapter

in

the

book

Essential

Elements

(in

press;

ILSI

Press,

1994).

Instead,

USEPA

now

prefers

to

utilize

uncertainty

factors

totalling

to

no

more

than

3,000.

The

Agency

agrees

with

this

approach

in

general,

except

in

cases where

the

overall toxicity database for

a

chemical

is very weak.

In

the

case of MTBE,

the database

contains enough

laboratory

animal

data

to determine

that

there

are not major

toxicity

gaps which

would

warrant

the

use

of

a

10,000-fold uncertainty

factor.

The Agency

is

therefore

also using

an overall

uncertainty

factor,

of

3,000

to

calcu.late

a guid.ance

level

for

MTBE.

Use of a

3,000-fold safety factor with the same

laboratory animal data described above

results

in

a

HTTAC

21.Iidance

level

of

0.23

mg/f,

or

230

ppb.

The

details

of

the

derivation

of

this

HTTAC

are

also

presented

in the attachment to

this notice.

Since

there is no

provision in the

language

of

Subpart F

for the

use

of

a

3,000-fold uncertainty

factor in

the

derivation

of

the

WrTAC,

the Agency

is proposing toutilize

HTTACS derived by

both

a

3,000-fold and

a

10,000-fold

uncertainty factor

in

the

health advisory

for

MTBE.

It

is

proposed

that the

HTTAC derived using

the

10,000-folduncertainty factor

(70

ppb)

be

a

precautionary

health advisory concentration and

the H1TAC

derived

using

the

3,000.-fold uncertainty

factor (230 ppb)

be the

final health

advisory

concentration.

The precautionary health advisory

would be

a level

in a public water

supply

below

which

no

action

would

be

necessary

and

above

which caution

should

be

exercised

by

the

public

water

supply

(such

as

increased sampling of the water

and identification of the potential

source(s)), while the

final health advisory would

be a

level

above which the

public water

supply should begin

actions to decrease

the concentration or utilize

an

alternate water supply.

The Agency

is

requesting comment. on the

use of this approach when

a total

uncertainty factor of

10,000-fold

is utilized

to

calculate a health advisory.

SUPPLEMENTARY INFORMATION

Section

620.605 also

specifies

that

the

health

advisory

must contain

a

general

description of

the

characteristics

of

the

chemical

substance and

its potential

adverse health

effects.

General D~criotion

of

MTBE

MTBE (Chemical Abstracts Service Number 1634-04-4), also known as

2-methoxy-2-methylpropane, is a colorless liquid

with a disagreeable taste and odor.

Its

taste in water

can be recognized

at

approximately 0.7

rug/f (700

ppb) (Connecticut

DEP), although

recent research

suggests

that some

pecple

may

be able

to detect its presence

in the

range of 0.25 mg/f and

possibly as

low

as 0.04

mgI!

(API,

1993).

It

has

a high solubility in water,

approximately 48,000 mg/! (von

Burg,

1992).

Because

of

this high

solubility,

it has

a high propensity

to move

through

soil ‘with

infiltrating rainwater

and snowmelt and

to

potentially reach

groundwater.

•

Its

main

use

is as an octane

booster in unleaded

gasoline;

it also has minor uses

as

an intermediate in the

production of

other chemicals,

especially

isobutene,

and’

as

a

treatment to dissolve gallstones.

Its

use has been increasing

recently

due

to

requirements under the Clean Air Act Amendments of

1990

for metropolitan areas which are not

in compliance with carbon

monoxide

standards to

increase the

percentage of oxygenated

fuel in

gasolines, especially

in the wintertime.

As

a result,

it has been estimated that approximately 20

of the gasoline sold in the United States contains MTBE, at

levels ranging from

2

to

15

in the gasolines.(Costantini,

1993).

Potential

Adverse Health

Effects of

MTBE

•

Relatively

few

reports of adverse

effects

of MTBE

on

humans

exist,

and

testing

for the

full

range

of possible health

effects

in

laboratory

animals

has not yet

been completed.

Summaries

of

the

acute,

reproductive and

developmental,

arid

chronic

toxicity data for MTBE

are presented.

Acute

Toxicity

-

Other than

a single

report

in

the

medical

literature of acute

kidney

failure

due to

leakage of MTBE

during

gallstone

treatment

(Ponchon.

1988),

there

is

no

information

regarding

the

effects

of short-term,

high

level

eApUSLLfe

U)

Mt

~n

Ui

numaa~.

toe

u.ai.a from

Laocr~Loryanirual

stuc1e~uiuicate

Lnat

uus

encmlcai

~

not very

tox1~.

during

briefexposures, with lethal doses

in

the

range

of 3,000-4,000 ppm by

oral

exposure

(about one pint for

an adult

human)

and 24,000-40,000 ppm (in air) by

inhalation exposure (this would

be within the

explosive range in air) (Reese

and

Kimbrough,

1993; von

Burg,

1992;

USEPA,

1993).

The

toxic

effect

in

both exposure

types was

central

nervous

system depression.

MTBE does ~ot

appear

to

cause

skin

irritation except

in cases of previously

damaged skin,

and eye

irritation

and opacity of

the cornea has been

reported

(von Burg,

1992).

---

Fag!

20/filly,

1994

Ein~ronmenWJ

Register

No.484

~r~duc~e

and

Developme!ital Toxicity

-

The

reproductive

effects

of

MTBE

have

been

~ported

in

three

studies,

and

reproductive

and

developmental toxicity

has

been

assessed

in

a

fourth,

using

rats,

mice,

and/or

rabbits.

No

significant

effects

were

reported

in

two of

the

reproductive studies (Biles

~

~.,

1987;

Conaway

~

i!.,

1985),

and

the

third reported effects

on

offspring (reduced body

weight and

reduced weight

gain

in rat

pUps,

and slightly

reduced

pU~

survival)

only at doses

which were also

toxic

to the

parents

(Neeper-Bradley,

1991).

Similarly,

the

reproductive and

developmental

study also reported offspring effects (reduced numbers of viable implantatio~sandior live births,

reduced

body

weight,

decreased

ossification, and

increased

incidence

of cleft

palate

in mouse

pups)

only

at

doses

toxic

to

the

adults (Tyl

and

Neeper-Bradley,

1989).

This makes

it di~cultto say whether

the

effects

on

reproductive performance

were tnily

an effect of MTBE on

the offspring,

or whether

these effects resulted from

the

toxicity to the

parents.

Since

the

doses

which showed these toxic

effects were

high (3,000-4,000 ppm), the potentialfor

human

reproductive

effects

at

the

much lower anticipated environmental

exposure

levels

is extremely small.

Chr~rnic

Tpxicity

-

There are no

studies of the effects

on hurnanc

exposed

to

MTBE for longperiods, although anecdotal

reports of

increased

complaints of headache,

nausea,

vomiting, eye

irritation,

and respiratory

problems

have

surfaced

recently

in certain

areas

in

conjunction

with wintertime

MTBE increases in gasoline.

These complaints are

the

subject

of on-going

research.

There is

only

one

90-day subchronic

study in

laboratory

animals exposed

by

the oral

route,

which

was

the study

Finally

selected

to

derive

the

WITAC

by

the Agency after following

the procedures

of

Appendix

A.

This

study

is evaluated

in

depth

in

the

attachment

to

this

notice.

There

are

several

animal

subchroàic

and

chronic

studies

using

the

inhalation

route

of

exposure,

primarily

evaluating

the

neurotoxic

effects

of

MTBE.

In

one

study

(Greenough

~

.~J.,

1980)

in

which

the

maximum dose tested

was

1,000 ppm for 6

hrs/day,

5

days/wk, for

13

weeks, no

significant

effects

(other

than

anesthesia

following dosing

at high concentrationS) were reported.

In another study (Dodd and Kintigh,

1989), in which

the

ma.ximum

dose

tested was

8,000 ppm (same

dosing regimen),

slight

changes

in blood chemistry,

increased

serum

cortisone

levels

in

both

sexes,

reduced weight

gain, increased kidney,

liver, and adrenal

gland weights, and sporadic

neurotoxic

effects

were

at

doses of

4,000

andlor

8,000 ppm.

There

is also

a recently

completed

lifetime

cancer

bioassay

in

mice

and

rats

(Burleigh-Flayer ~

~j., unpublished; Chun ~

~.,

unpublished), the

details

of which are

evaluated

in

the

attachment to

this

notice.

FOR

FrJRTRER

INFORMATION.

COMMENTS

Persons

who

wish

to

receive

further information

about this

notice

or who wish

to provide

comment

on

its content.~

are.

requested

to

contact:

illinois Environmental Protection

Agency

Office

of

Chemical

Safety

P. 0.

Box

19276

2200

Churchill Road

Springfield,

Illinois

62794-9276

2171785-0830

---

Environiflen~al

Register

No.

484

-___________________

july,

1994”Page2l

ATTACHMENT

TO

NOTICE

OF

HEALTH

ADVISORY FOR

METHYL

TERTLA..RY-BUTYL ETHER

(MTBE)

OVER

VIEW

OF THE

KEY STUDIES

In

the

only

oral

study (Robinson

et al.,

1990),

rats were

given 0,

100,

300,

900,

or

1,200

mg/kg

(ppm)

by

gavage.

Rats

given

1,200

ppm

exhibited profound anesthesia

after

dosing throughout

the study,

but

recovered

after

the

dbse

within

two

hours

and

suffered

no

aftereffects.

Body

weight

decreased

with

increasing

dose,

with

the difference

between

treated

and

control

rats being statistically

significant

at

1,200 ppm.

Other

measurements

showing statistical significance

included:

decreased

blood

urea

nitrogen

(BUN)

and

serum

creatinine

(measures

of

kidney

function) a~t

all

doses;

increased

serum

cholesterol

at

all

doses; increased

kidney weight

at

300

ppm and above;

increases in several other organ weights at

900

ppm

and above;

and changes

in

blood parameters

at

1,200 ppm.

Microscopic examinations

revealed

effects

only

at

1,200

ppm,

where degenerative

changes

in the

kidneys of the

male rats were noted.

Finally,

loose stools and diarrhea

were

seen

at

all

doses

throughout

the study.

Viewing

the

results of

this

study,

it would

appear that the

kidney

is the target

organ of

MTBE.

However,

these results

must be

interpreted

carefully.

The decreases

in BUN

and

serum

cre.atinine probably have

no

adverse effect on

the

animals

(decreased

kidney

function is

often signaled

by

increases in these parameters),

and

may

even indicate an

increase

in kidney

function.

The

increased

kidney

weights

seen

at

300

ppm

and

above

are

not

in

themselves

an

adverse

effect,

only

an

indication

of a

possible

adverse effect

at even higher doses or longer

exposure

times.

Finally,

the

microscopic

changes seen

at

1,200

ppm

in

males

are

often

seen

in

male

rats

(and

only

male

rats)

exposed

to

certain

organic

chemicals,

due

to

overproduction

of a

unique

protein

in

the

male

rat

kidney.

Thus,

it

is not

clear

at

this

time whether MTBE

is toxic

to

the

kidney.

It would appear

that

a

no

observed

adverse effect

level (NOAEL)

has

not been

determined by

this

study,

since increased

serum cholesterol

and

diarrhea

were

observed

at

all

doses.

Thus,

the

100

ppm

dose

would

be

considered

to

be

the

lowest

observable

adverse

effect

level (LOAEL)

for MTBE.

The

procedure

for calculating

a

health advisory

for drinking

water

in

the

groundwater

quality standards

(35

Ill. Adm.

Code

620,

Subpart

F)

gives

preference

to

oral

studies

which

determine

a

NOAEL

or

LOAEL,

and

this

study

may

be considered

to

develop the

health

advisory

for MTBE.

A

lifetime

inhalation

cancer

bioassay

has

recently

been

completed

with

mice

and

rats,

but

the

results

have

not

been

published

(Burleigh-Flayer ~

~j.; Chun ~

~j.).

The Agency

has been

given summaries

of

the

studies

submitted

to

USEPA

by

the

USEPA

contact

for MTBE.

These

results

are

briefly

summarized,

but since

the studies

are

still

undergoing

review

it

must

berealized

that

this information

is

preliminary.

Both

species

were

exposed toO,

4.00,

3,000,

or

8,000

ppm

in air.

As

in the

oral

study above,

the

male rats

experienced

an

increased

incidence of kidney degeneration.

This became

the

leading cause

of

deathjn

male

rats,

and resulted in early

termination

of the

3,000 and

8,000

ppm

male

groups.

The other main cause of

death

in

male rats

was

leukemia,

seen

in

both

the

control

and

400

ppm

group.

(In fact,

the

incidence

in

the

control group

was

higher, 33/50,

than

in

the

400

ppm

group, 22/50.)

Non-cancer effects of

MTBE

included symptoms

of

central

nervous system depression

in

both sexes of

rats

at

3,000 and

8,000

ppm, but

not at

400

ppm,

and an

increased

incidence of kidney

degeneration

in

male

rats

at

400

ppm.

The only

tumors which were related to

MTBE

exposure

were

rumors

in

the kidneys of male

rats

in

the

3,000 and

8,000 ppm

groups.

These

tumor types are

also

thought

to

be

related to

the

overproduction of the

male

rat

protein,

and

the

significance

of

these

results

for

humans

is

questionable.

In

the

mouse

study, symptoms of central

nervous

system depression

similar

to

those seen

in rats

were

observed-at 3,000

and 8,000

ppm.

Increases

in

liver

and

kidney weights

were also seen

at

these doses, and

an

increase in

the

number of liver

cells

(nonc.ancerou.s),

an indication of toxic effects

on

the liver,

was

reported at 8,000 ppm.

The only

tumors found

in excess

of controls were

liver

tumors

in females

in

the

8,000

ppm

group.

However,

the significance

of

this

finding

for

humans

is

also

questionable,

since

this

tumor type

is

common

in

the

strain

of

mouse

used

in

this

study,

and

is

known

to occur

in

~

I. ~&4U

~Ly

U.~L&

In reviewing

the

results

of these studies,

it

is difficult to

say

whether MTBE

presents a carcinogenic

hazard

to

humans.

However,

the

noncancer

effects

may

be

relevant for

determining a health

advisory level

for MTBE.

In

this

regard,

the

rat

study

has

produced

a

LOAEL of

400

ppm

based

on

kidney

effects

in

male

rats

(this

dose

may

be

a

NOAEL

given

the

questionable

significance

of

this effect

for

humans),

while the

mouse

study

has

produced

a

NOAEL of

400

ppm.

The

mouse

---

Page 22/ July,

1994

Environmental_Reg&er

No.

484

portion of

this study

may

be

considered

to

develop

the

health

advisorj

for

MTBE,

once

it

has

F.nis~~USEPA’s

review

procesS.

-

DERIVATIO~’(OF THE HEALTH

ADVISORY

FOR MTBE

The first step

in the derivation of a heslth advisory is to determine

whether the

chemidai

pre~entsacarcinogenic

hazard

to

humans.

To

date,

there

have

been

no

investigations

whether

there

is

an

increased

incidence

of

cancer

in

humans

associated

with exposure

to

MTBE.

As

discu~àedabove,

there

is some evidence

that

MTBE causes

tumors

in

laboratory

animals,

but

the

types

of

tumors

found

in

the

rat

and

mouse

cancer

bioassays

may

not

provide

good

evidence

of

a

carcinogenic

hazard

to

hunianc since these tumors may

be species-specific responses with little or

no

relevance

to humans.

Furthermore,

these studies

are

still undergoing

review by

USEPA

and a

final determination of

the

usability of

the

results

for determining the

carcinogenic

hazard

to humans

has not been made.

Therefore,

the

Agency

has determined at

this time

that

the

derivation

of

the

health

advisory

for

MTBE

will

be

based

on

the

non-cancer

effects

of

this

chemical.

This

derivation may be changed

in the future,

depending

on

the USEPA’s ~determinations,

once the cancer

bioassay data

have been published and

the

weight-of-evidence for human

carcinogenic potential

has

been

determined.

In deriving a health advisory

to

protect against a health effect for which there is a threshold dose below which

no

damage

occurs

(i.e., noncarcinogenic

effects),

Section

620.605 specifies that

USEPA’s maximum cont~rnin~nt

level

goal

(MCLG).

if

available,

is

the

health advisory concentration.

USEPA has not publisheda MCLO for MTBE, therefore,

the

Agency

must

calculate

the

human

threshold

toxicant advisory

concentration

(WITAC)

as

the

health advisory

concentration, using

the

procedures specified

in

Appendix

A

of Section 620.

Appendix A specifies in subsection

(a)

that

the

ifITAC

is calculated

a~sfollows:

RTL4C=RS~~~~

w

Where:

HTTAC

=

Human threshold toxicant

advisory

concentration in

milligrams

per liter

(mg/I);

RSC

=

Relative

source

contribution,

the

relative

contribution of

the

amount

of

the

exposure

to

a

chemical

via drinking

water when compared

to the

total

exposure

to

that

chemical

from

all

sources.

Valid chemical-specific data

shall

be

used if available.

If valid

chemical-specific

data

are not available,

a

value of 20

(=0.20)

must be

used;

ADE

=

Acceptable

daily exposure of

substance

in milligrams

per

day (mg/d)

as

determined

pursuant

to

subsection

(b);

and

W

=

Per

capita daily

water consumption equal

to

2

liters

per

day

(Lid).

Subsection

(b) of

Appendix

A

specifies

that

the

ADE

be calculated using,

in specified

order~

USEPA’s

Verified

Oral

Reference

Dose (an

estimate

of

a

daily

exposure to

a

chemical

which is

expected to

be without

adverse

effect

for

humans,

including

sensitive

subgroups,

for

a

lifetime

of

exposure);

a

NOAEL

which

has

been

identified

as

a

result

of

human

exposures;

a

LOAEL which

has

been identified as

a

result of human

exposures;

a

NOAEL. which

has

been determined

from

studies with laboratory

animals; and

a LOAEL

which has been

determined from studies

with laboratory

animals.

There is no

Verified

Reference Dose currently available from USEPA.

As mentioned

above, there

is a

paucityof studies

on

the

adverse

effects

in

humans exposed

to

MTBE.

Thus,

the Agency has

determined that

a

NOAEL

or

LOA.EL

based

uQ

QLW)AL1

CA~’.J~U~~ 1~OUL

4VU~C

d~. i.W~ UU~.

L

~1C~U1C,

U~G

i~#i.:. L~aU.~L

tiC

U~Cte

i~tJh.h~

I

d~JI..JtJ

~

~

the

studies reviewed

by the

Agency,

the

90-day

rat

subchronic

study

and

the

cancer

bioassay (noocarcinogenic

effects) are

the most

appropriate animal

studies

for calculation of the

ADE.

It

is then

necessary

to

determine

which

study

is

the

most

valid

for

purposes

of calculating the

ADE.

Subsection

(c)

of

Appendix

A

specifies

criteria

for

establishing

the

validity

of

data

from

animal

studies,

leading

to

determinations

of high,

medium,

or

low validity.

High validity

studies

are

those

using the

oral

route

of exposure and which

---

Environmertta!

Register

No.

484

July,

l994/Page

23

meet

specified

criteria

depending

on

the

type

of

study,

and

are

to

be

used

preferentially

if availabie.The

rat

90-da’~

subch.roaic study was conducted

using

the

oral

route,

while

the

cancer

bioassay was

an

inhalation

study.

Therefore,

only

the

subchronic study could

be

a

high validity study.

However,

the

requirements

for a high validity

subchronic

study

include,

among

other

things,

a

study

using

two

species

and determining

a well-de5ned NOAEL.

The

90-day

rat

subchronjc study

used only

one

species

and

only determined a LOAEL,

as discussed

above.

Having

no high validity

study, the

Agency

must

determine which

of

the

two

studies is most

appropriate

for calculating

the

AIDE.

Subsection (c)

goes

on

to

specify

that

in

order for

a

subchrocic

study in which

a

LOAEL

is

determined

to

be

deemed

a

medium validity study,

the

study

must satisfy

all

other standards

for a high validity

study.

This is not

the case

for

the

90-

day

rat subchronic study, since there was only

one

species tested.

Similarly,

in order

for a study

other

than

an

oral exposure

study

to

be deemed

a

medium

validity

study,

the

sttidy

must

satisfy

all other

standards

for

a

high validity

study

arid

use

appropriate

correction factors

for conversion to the

oral route.

However,

the

requirements

for a high validity

cancer bioassay

include,

among other things,

at

least

2.5

survival

at

18

months

in

mice

and

24

months

in rats.

This

was

not

the

case

in

the cancer

bioassay,

since

the male

rats

in

the

3,000 and 8,000

ppm groups were

terminated

early due to

excessive

mortality.

Thus,

both

candidate

studies

are

defined

as low

validity studies,

and

the

90-day

rat

subchronic

study

is

selected

because

exposure

was

by

the

oral

route.

The

determination

of

the

ADE

from

the

subchronic

study is made

using

the

language

of subsections

(b)(5)

and

(b)(6).

Subsection (b)(6)

specifies

that

for

substances

for

which

a NOAEL

is not

available,

onà-tenth of

the

LOA.EL

is

substituted

for the NOAEL

in subsection

(b)(5).

Subsection

(b)(5)

specifies that

if

studies

of low validity

must

be used,

the

ADE

must

be

calculated

using

1/1000

of the

NOAEL.

The

overall

result

of

the procedures

in

these two subsections

is

that

the

ADE

is

1/10,000

of the

LOAEL,

times the

average

weight

of an

adult

human,

70kg:

ADE= lOOmg/k1g/dxlQkg

=0 7m

Id

10,000k2Jd

S

At

this point,

the calculation of the

TTTAC

would

proceed according

to

the formula

listed above.

However, the

Agency

has been

informed by

USEPA personnel

that in

most

cases

USEPA now prefers

to

calculate acceptable exposure

values

for

humans

by

using laboratory

animal

data divided by

no

more than

a 3,000-fold uncertainty factor;

a

10,000-fold uncertainty

factor would

be

used

only

where

the

overall

toxicity

database

is

very weak

for a

chemical.

The Agency

agrees

with

this

emerging USEPA approach.

Since

the MTBE

database

contains

enough

laboratory

animal

research

to

indicate that

there

are

not

major

toxicity

data

gaps

which would

warrant

the

use

of

a

10,000-fold

uncertainty

factor,

the

Agency

is

also

calculating

the

ADE

using a 3,000-fold uncertainty

factor:

ADE= 10C~ng/kg/dx70kg 2.3n~gJd

3,000

-

Finally,

the

determination of the

HTTAC

is straight-forward, since

there are no

chemical-specific data

available

for the

RSC

term:

H7TAC= 0i0x0.7mgJd007~

2.0~/d

Or:

0’0x23m

Id

HITAC=

—

~‘

=023ing/~

‘~

(~3th3

The

final

step

in

detei-rninirig

the

health advisory

is

to

compare

the

HTTAC

value

calculated

from

the

Appendix

A

procedures

to

the

chemical’s

Practical

Qu.antit.ation

Limit

(PQL).

In

the

case

of

MTBE,

no

USEPA

SW-846

analytical

method

specifies

a

PQL

for this chemical.

However,

the

Agency’s Division of

Laboratories

has

determined that a detection

limit of

0.005

mg/f

is

appropriate

for water

samples.

Therefore,

the

HTTAC

value

is above

the

detection

Limit.

---

Page 24/July, 1994

Envfrvnme~aa1R.!gt~arr

No.

484

The Agency has decided to

issue

a

two-part

health

advisory.

The precautionary

health ad~isory~ôncentration

for

Methyl Tertiary-Butyl

Ether

(MTBE)

is

0.07 mg/I

or

70

parts per billion

in drinking

water.

People

can

be

exposed

to

this concefltration of MTBE

in drinking

water

over a

70

year

lifetime.

Above thi~concentration, appropriate caution

should be exercised

by

the

Public

Water Supply,

such

as increased

frequency of sampling and

identification of the

MTBE

source(s).

The final

health advisory

concentration

is 0.23 rag/I or 230 parts per billion

in drinking water.

Above this

concentratiofl,

the Public

Water Supply should begin actions

to

decrease

the

amount

of MTBE

in

the

system.

REFERENCES

API, American Petroleum Institute.

1993.

Odor Threshold

Studies Performed

with Gasoline

and

Gasoline

Combined with

MTBE,

ETBE,

and

TAME.

API

publication

Number

4592.

Biles,

R.

W.,

Schroeder,

R.

E.,

and

Hold.sworth,

C.

E.

1987.

Methyl

Tertiary

Butyl

Ether

Inhalation in

Rats:

A

Single

Generation

Reproductive Study.

Toxicol.

md.

Health

3:

519-534.

Burleigh-Flayer,

H.

D.,

Chun, I.

S.,

and

Kintigh,

W.

J.

(unpublished).

Methyl Tertiary

Butyl

Ether:

Vapor

Inhalation

Oncogenicity

Study

in CD-i Mice.

Submitted

to

USEPA,

Docket

No.:

OPTS-42098.

Chun,

J.

S.,

Burleigh-Flayer,

H.

D.,

and

Kintigh,

W.

I.

(unpublished).

Methyl

Tertiary

Buryl

Ether:

Vapor

Inhalation

Oncogenicity

Study

in

Fischer

344

Rats.

Submitted

to USEPA,

Docket

No:

OPTS-42098.

Con.away,

C.

C., Schroeder,

R. E.,

and

Snyder,

N.

K.

1985.

Teratology Evaluation of Methyl Tertiary-butyl Ether in

Rats

and

Mice.

3.

Toxicol.

Environ, Health

16:

797-809.

Connecticut

Dept.

of

Environmental

Protection.

(undated).

Action

Level

for

Methyl

Tertiary

Butyl

Ether

(MTBE)

in

Drinking

Water.

Prepared

by

H.

V.

Ran, C.

3.

Dupuy,

and

D.

R.

Brown,

Connecticut

Dept. of Health Services.

Costantini,

M. 0.

1993.

Health

Effects

of

Oxygenated

Fuels.

Environ.

Health

Peispectives

Suppl.

101

(Suppl. 6):

151:.

160.

Dodd,

D.,E.

and Kintigh, W.

J.

1989.

MethyL Tertiary ButyL

Ether (MTBE):

Repeated

(13-Week) Vapor Inhalation Study

in

Rats

with

Neurotoxicity

Evaluation

(unpublished

study).

Union

Carbide,

Bushy

Run

Research

Center

for

MTBE

Committee.

TSCATS

403189.

EPAJOTS

#

FY1-OTS-0889-0689.

Cited

in

Revised

and

Updated

Drinking

Water

Quantification

ofToxicological

Effects

for Methyl Tert-Butyl Ether

(MTBE), Final

Draft.

USEPA;

ECAO-CIN-D023, July,

1993.

Essential

Elements

(in press).

ILSI

Press,

Washington,

D.C.

1994.

-

Greenough,

R.

3.,

McDonald,

P.,

Robinson,

P.,

et

al.

1980.

Methyl

Tertiary-Butyl

Ether

(Driveron)

Three

Month

Inhalation Toxicity in

Rats.

Project No.

413038.

Unpublished

report submitted

to

Cbemische

Werke Hôls AG,

Marl,

West

Germany.

230

p.

Cited

in

Revised

and Updated

Drinking Water

Quantification of Toxicological

Effects

for Methyl

Tert-

Butyl

Ether (MTBE),

Final

Draft.

USEPA;

ECAO-CIN-D023,

July,

1993.

Neeper-Bradley,

T.

L.

1991.

Two-Generation Reproduction Study

of

Lnhaled

Methyl

tert-Butyt

Ether

in

CD

Sprague-

Dawley

Rats

(unpublished

study).

Union

Carbide,

Bushy

Run Research

Center.

Cited

in

Revised

and Updated Drinking

Water

Quantification of

Toxicological Effects

for

Methyl Tert-Butyl

Ether

(MTBE),

Final

Draft.

USEPA;

ECAO-CIN-

D023,

July,

1993.

Ponchon, T.,

Baroud, 3.,

Pujol, B.,

Valette,

P. 3.,

and

Perrot,

D.

1988.

Renal

Failure

during Dissolution of Gallstone by

......Z..~.

...

.....j.

~

............Z

—.

—.

Reese,

E.,

and

Kimbrough,

R.

D.

1993.

Acute Toxicity of Gasoline and

Some

Additives.

Environ.

Health

Perspectives

Suppl.

101

(Suppl.

6):

115-131.

Robinson,

M., Bruner,

R.H.,

and

Olson,

G.R.

1990.

Fourteen

and

Ninety-Day Oral

Toxicity

Studies of Methyl Tertiary-

Butyl

Ether

in Sprague-Dawley Rats.

Journal

of

the

American College of Toxicology

9(5):

525-540.

---

Enviroflinefl1~12IRegister

No.

484

July,

1994/Page

25

Tyl,

R.

W..

and

Neeper-Bradley,

1.

L.

1989.

Developmental

Toxicity

Study of

Inhaled

Methyl.

Tertiar)T

Butyl

Ether

in

~D-l

Mice.

Project

Report

52-526.

Prepared

by

Bushy

Run

Research

Center,

Union Carbide

Corporation

for the Methyl

fertiary

Butyl

Ether

Committee,

Washington.

D.C.

Microfiche

No.

0TS0000689..1.

Cited

in

Revised

and

Updated

Drinking

Water

Quantification

of

Toxicological Effects

for Methyl Tert-Butyl

Ether (MTBE),

Final Draft.

USEPA;

ECAQ-

-

CIN-D023,

July’,

1993.

USEPA.

1993.

Revised

and Updated

Drinking Water

Quantification of Toxicological

Effects

for Methyl TertButyl

Ether

(MTBE),

Final

Draft.

USEPA;

ECAO-CIN-D023, July,

1993.

von Burg,

R.

1992,

Toxicology

Update.

Methyl Tert-Butyl

Ether.

I.

AppI.

Toxicol.

12:

73-74.

---

•

State ofillinois

--__ENVIRONMENTAL

PROTECTION AGENCY

~vfaryA.

Gade, Director

2200

Churchill

Road, Springfield, IL 62794-9276

217/785-0830

-

November 4,

1994

G.A. Van Gelder,

DVM,

Ph.D., ABVT

Manager, Toxicology

Health, Safety and Environment

Shell

Oil Company

One Shell

Plaza

P.O.

Box 4320

3ouston,

TX

77210

•Dear Dr. Van Gelder:

•

This letter confirms

the meeting to evaluate comments received regarding the

Illinois Environmental

Protection Agency’s proposed Health Advisory for MTBE

which we discussed over the telephone.

The meeting is scheduled for November

14,

1994,

beginning

at

12:30.

The room

is available until

5:00 PM,

if

necessary.

The meeting will

be held

in Room 031

on Floor

8, James

R. Thompson

Center,

100 W. Randolph,

Chicago,

Illinois,

60601.

I

have enclosed an agenda for the meeting,

a copy of the Health Advisory

Section of the Illinois Groundwater Quality Standards,

and a summary of the

Agency’s opinions on two key issues which have emerged from the comments.

I’m looking forward to

a productive meeting.

Please call

(217/785-0830)

if

you have any further comments or questions.

Sincerely,

T—

Thomas

C. Hornshaw,

Ph.

D.

Manager, Toxicity Assessment Unit

Office of Chemical

Safety

f:\psf\epa8566lmtbe.mtg

Attachment

Printid

on

R!qvlad

Paper

---

MTBE Meeting Agenda

12:30

-

12:45

12:45

-

1:45

1:45

-

2:00

2:00

-

3:15

3:15

-

3:30

Introductions and Background

Key Issues (LOAEL vs.

NOAEL,

RSC)

Break

Other

Issues (Tase/Odor Threshold,

Uncertainty

Factors, 2-Tier Vs. Single Advisory, Edits

Wrap-up

•

---

RESPONSES

COM1VIENTS

REGARDING

PROPOSAL FOR

HEALTH

ADVISORY

-

FOR

METHYL

TERTIARY-BUTYL

ETHER

The Illinois

Environmental Protection Agency

(Agency) has received three

cowments in response

to the

Notice of Health

Advisory

for

Methyl

Tertiary-Butyl

Ether

(MTBE),

published

in

the

Illinois

Environmental

Register No.

484,

July,

1994.

The

comments

were received

from

the

American

Petroleum Institute

(API),

the Methyl Tertiary Butyl Ether Task Force

(Task Force),

and

Shell

Oil

Company

(Shell).

The

comments

cover

several

technical

and

typographical

subjects,

the

most

significant

of

which

address

the

Agency’s

determination

of

a

Lowest

Observable Adverse

Effect

Level

(LOAEL)

versus

a

No

Observable

Adverse

Effect

Level

(NOAEL)

and

the uncertainty

factors

which

result from

this

determination,

and

the Agency’s

use of the default value of 20

as

the Relative Source

Contribution

(RSC)

term

versus the use

of an

RSC

derived

from chemical-specific

data in the calculation of the Health Advisory.

The

Agency’s

responses

to these key

issues

are

LOAEL

vs. NOAEL

API

and

Shell disagree

with the Agency’s

characterization of the

diarrhea

and

elevated

serum

cholesterol

reported at

the

100

mg/kg

dose

in

the Robinson

et al.

(1990)

study

as

a

LOAEL.

In

reviewing

the

results

of

this

study,

the

Agency determined

that the

authors’

reports

that

“treated

rats

in

all

dose

groups

also

displayed

diarrhea

throughout

the exposure

period”

and

their

findings

that

“females

exposed to

all

dose levels exhibited

significant increases

in

serum

cholesterol”

indicated

that

the

study

had

not

identified

a

No

Observed

Adverse

Effect

Level.

This

determination

is

an

outcome

of

the

evaluation

of the

validity

of

the

candidate

studies

required

by

the

Groundwater

Quality

Standards

regulation

when

animal

studies

must be

used

to

develop

a

Health

Advisory.

This

evaluation

was

discussed

briefly

in

the

July,

1994

Notice,

and will

be

expanded for

explanation of the Agency’s

rationale.

Section

620.

Appendix

A(c)(1)(A)(iii),

which

identifies

the

elements

necessary

for

High

Validity

Studies,

requires:

Data

from

animal

subchronic

studies

with

a

minimum

of

3

dose

levels

and

control,

2

species,

both

sexes,

4

animals

per dose per

sex for non-rodent

species

or

10

animals

per dose per sex for rodent species,

a

duration of at least

5

ofthe

test species’

lifespan,

and

a well-defined NOAEL

(emphasis

added).

The Agency determined that the reports of diarrhea in all animals

and

elevated serum cholesterol

in

females

in

all dose

groups

could

not

be

called

a

“well-defined

NOAEL”

for

purposes

of

establishing

High

Validity

for

this

study.

Thus,

the

lowest

dose

tested,

100

mg/kg,

was

API

and

Shell

have

commented

that the

results

of the

study

should

not

be

interpreted

in

this

manner.

Both claim

that the occurrence of diarrhea

in treated

animals

is

not well-documented

---

or

described

in

the

Robinson

study,

that

diarrhea is

a

common

observation

in

rats

dosed

with

-

corn

oil,

and

it

is

a questionable

endpoint for extrapolation

to

low-dose lifetime

health

effects.

Both

also

claim

that

the

modest

increases

in

serum

cholesterol

in

the

female

rats

are

not

indicative of a meaningful health effect,

arguing

that the authors’ statistical evaluation incorrectly

attributes

a significant

difference

for the

300

mg/kg

dose,

that

there

is no

compelling evidence

for

a

dose

response,

that

only

the

900

mg/kg

dose

in

males

achieved

values

significantly

different from

controls,

and

that the increases are near

the range of

normal

variability.

Finally,

API argues

that the diarrhea and

elevated serum cholesterol

are

not significant results,

citing

the

authors’ conclusions

that the

study indicated that dose levels below those which induce anesthesia

(1200 mg/kg)

do

not

result in

significant pathophysiological

changes.

The

Agency remains

unconvinced

that

the Robinson

et al.

study

has

identified

a

well-defined

NOAEL.

Regarding

the occurrence of diarrhea,

we have

interpreted

the

authors’

reports of

diarrhea

in

“treated

rats

in

all dose groups”

to

mean

all

groups

receiving doses of MTBE,

but

not

those

receiving

the vehicle

control

(corn

oil).

Thus,

we believe that the diarrhea

is

likely

to be treatment -related,

at least

in females;

this belief is supported

by the findings of the 14-day

study

also

reported

in

this

paper,

in

which

“by

the

third

day

of

dosing,

all

treated

animals

displayed loose stools

which continued throughout

the remainder of the exposure period.

‘

We

have

reviewed

the

National

Toxicology

Program’s

report

on

the

lifetime cancer bioassays

of

gavage

vehicles in

male

Fisher

rats,

which

included corn

oil,

and

find

no

mention of diarrhea

as

an effect of corn oil (NTP,

1994).

Finally,

we have

relied on

the experience of one

of the

Agency’s

Office

of

Chemical

Safety

toxicologists,

who

reports

that,

in

over

8

1/2

years

of

experience

in

an

industrial

toxicology

laboratory,

the

occurrence

of

diarrhea

in

rats

in

conjunction

with corn oil vehicles was very infrequent (Morrow,

1994).

While we cannot rule

out the possibility that

the diarrhea reported by Robinson et al.

was vehicle-related,

we continue

to

believe

that

this

effect was a

result of the MTBE exposure.

Regarding the elevated serum cholesterol findings,

the Agency acknowledges that the statistical

significance

of

the

300

mg/kg

dose

in

female

rats

is

questionable

and

possibly

incorrectly

reported,

and

that

there

is

no

obvious

dose-response

relationship

among

the female treatment

groups even though all but the 300 mg/kg group is significantly greater than controls.

However,

we

maintain

that these results

are

potentially indicative of a

real

effect in

the rats;

it

is possible

(although

unlikely)

that

the effect may

plateau

relatively quickly,

such

that

the

dose-response

relationship

is defined at doses

below those tested in

this

study.

Further,

we again note that the

results

of

the

14-day

study

reported

in

this

paper

also

include

elevated

serum cholesterol

in

females

of most treatment groups.

Regarding the biological significance of the diarrhea and elevated serum cholesterol and whether

these endpoints

are

relevant for

extrapolating to human health risks, the Agency

maintains

that

•

such

effects

are

relevant for

use in

developing the Health Advisory.

While

neither endpoint is

relatively

serious,

diarrhea

can

be• deleterious

to

the

organism

over

time

by

contributing

to

dehydration, electrolyte imbalance, and/orpoor nutritional status,

and

elevated cholesterol, while

not

in useil

a

olologicany serious

ei1~eL,

is

a

CauLioti

IOr

inure

serious

CilectS

over

Li-file.

Willie

the

authors’

concluded

that

dose levels

below

those

which

induce

anesthesia

do

not

result

in

significant pathophysiological changes,

the Agency would be very

uncomfortable

using

a dose

which does

not induce

anesthesia as

the basis

for developing

a Health

Advisory.

We

continue

---

or

described

in

the

Robinson study,

that

diarrhea

is

a

common

observation

in

rats

dosed

with

corn

oil,

and

it is

a

questionable endpoint

for extrapolation to

low-dose

lifetime health effects.

Both

also

claim

that

the

modest

increases

in

serum

cholesterol

in

the

female

rats

are

not

indicative of a meaningful

health effect,

arguing that the authors’

statistical evaluation

incor~.ectly

attributes

a

significant difference for the

300

mg/kg dose,

that there is no

compelling

evidence

for

a

dose

response,

that

only

the

900

mg/kg

dose

in

males

achieved

values

significantly

different from

controls,

and that the increases are

near the

range

of

normal variability.

Finally,

API argues

that the

diarrhea and

elevated

serum cholesterol

are

not

significant

results,

citing

the

-

authors’

conclusions that the study indicated

that dose levels below

(1200 mg/kg)

do

not result in

significant pathophysiological

changes.

The

Agency remains

unconvinced

that the

Robinson

et

al.

study

has

identified

a

well-defined

NOAEL.

Regarding

the occurrence of diarrhea,

we have

interpreted

the authors’

reports

of

diarrhea

in

“treated rats

in

all

dose groups”

to

mean

all

groups

receiving doses of MTBE,

but

not

those

receiving

the vehicle

control

(corn

oil).

Thus,

we

believe that the

diarrhea

is

likely

to be treatment

-related, at least in females; this beliefis

supported

by the findings of the 14-day

study

also

reported

in

this

paper,

in

which

“by

the

third

day

of

dosing,

all

treated animals

displayed loose stools

which continued throughout

the remainder of the exposure period.”

We

have

reviewed the National

Toxicology

Program’s

report

on

the

lifetime

cancer bioassays of

gavage vehicles

in

male

Fisher

rats,

which

included corn

oil,

and

find

no

mention of diarrhea

as

an effect of corn

oil (NTP,

1994).

Finally,

we have relied

on

the experience of one of the

Agency’s

Office

of Chemical

Safety

toxicologists,

who

reports

that,

in

over

8

1/2

years

of

experience

in

an

industrial

toxicology

laboratory,

the

occurrence

of

diarrhea

in

rats

in

conjunction with corn oil vehicles was very

infrequent

(Morrow, 1994).

While

we cannot rule

out the possibility

that the diarrhea reported by Robinson et al.

was vehicle-related, we continue

to believe that

this

effect

was

a result of the MTBE

exposure.

Regarding the elevated serum cholesterol findings,

the Agency acknowledges that the statistical

significance

of the

300

mg/kg

dose

in

female

rats

is

questionable

and

possibly

incorrectly

reported,

and

that

there

is

no

obvious

dose-response

relationship

among

the female

treatment

groups even though all but the 300 mg/kg group is significantly

greater than controls.

However,

we

maintain

that these results

are

potentially

indicative of a

real

effect in the

rats; it is possible

(although

unlikely)

that the

effect may

plateau

relatively

quickly,

such

that

the dose-response

relationship

is defined at doses below

those tested in this study.

Further, we again

note that the

results of the

14-thy

study

reported

in

this

paper

also

include

elevated

serum

cholesterol

in

females of most treatment

groups.

Regarding the biological significance of the

diarrhea

and

elevated serum cholesterol

and

whether

these endpoints

are

relevant for extrapolating to human health risks,

the Agency

maintains

that

such

effects are

relevant for use

in developing the Health Advisory.

While neither endpoint is

relatively

serious,

diarrhea

can

be

deleterious

overtime

to

the

organism

by

contributing

to

dehydration, electrolyte imbalance,

and/or poor nutritional status,

and elevated cholesterol, while

not

in

itsett a

oloioglcany serious ertect,

is

a cauuon ror more senous

ertects over

time.

While

the

authors’

concluded

that

dose levels

below

those

which

induce

anesthesia

do

not

result

in

significant

pathophysiological changes,

the Agency

would

be

very uncomfortable

using

a

dose

which does

not

induce

anesthesia

as the basis for

developing

a Health Advisory.

We

continue

---

to

believe that

the

100

mg/kg

dose,

as

a

LOA.EL,

is

the

most

relevant

value

to

use

in

the

-

development of the Health Advisory.

This

reasoning

plus the relative paucity of data regarding

the ingestion of MTBE,

argues for

the continued

use of the 3000-fold

uncertainty

factor as the

most

appropriate

value

for

the final Health

Advisory.

•

•

•

—

MTBE

RELATIVE

SOURCE

CONTRIBUTION

TERM

The

comments

of

both

API

and

the

Task

Force

addressed

the

Agency’s

use of

the default

value

of 20

as

the Relative

Source

Contribution

(RSC)

term,

which

is

specified

in

Section

620.

Appendix

A(a).

(This

is

also

a

standard

TJSEPA

default

assumption,

used

in

risk

assessments

to

account

for

all

other

exposures

to

a chemical

other

than

direct

ingestion

in

drinking

water,

such

as through the

diet,

ambient air,

the workplace,

and

volatilization

from

the household water supply).

Both

comments

cite

a

tJSEPA

study

(tJSEPA,

1993)

which estimates

the

amount

of

MTBE

exposure

experienced by

the general

public during

activities

other

than

drinking

water,

such

as

working,

outdoor

exercise,

refueling,

driving,

etc.

This

study

is

proposed

to

be

used

as

chemical-specific

data

instead

of the

default

value

to

account

for

exposures to

MTBE other

than

via

direct ingestion of drinking water.

If this

study

is

used

to

define

the

RSC term,

the

range

of

weighted

annual

MTBE

ambient

air

concentrations

of

0.04

-

0.07

mg/rn3

would

result

in

a

RSC

term of approximately

45

-

70

for

drinking

water

exposures.

Depending

on

the final determination of the

RSC

term,

the Health Advisory

(HA)

for

MTBE

would

then

be

in

the

range

of

0.52

-

0.80

rng/l,

instead of the proposed

0.23

mg/l

using

a

20

RSC

term.

While

the

Agency

agrees

with

the

data

presented

in

the

USEPA

study,

it

cannot

agree

that

these

data

fully

account

for

all

other

sources

of MTBE

contributing

to

a

person’s

daily

exposure.

Use

of

only

this

study

to

account

for

inhalation

exposures

does

not

consider

inhalation

exposures which

will

Occur in the home as a result of volatilization of MT.BE

from

the household

water supply during uses

of the supply for purposes

other than

drinking.

Since

the

Agency

is

not aware

of studies

evaluating

such

exposures,

an

evaluation of the

indoor

inhalation pathway

was

undertaken using

data reported for trichloroethylene

(TCE).

• The

transfer

of

volatile

organic

chemicals

(VOCs),

including

TCE,

from

water

to

air

has

been

studied

by

several

investigators

(Andelman,

1985;

McKone,

1987;

McKone

and

Knezovich,

1991).

Of particular

interest

for

this

analysis

are

studies

which

measure

the

transfer of VOCs

during

showering

since this

activity is

likely

to

be

the

greatest

contributor

to

indoor

VOC

exposure due

to

the

temperature,

amount

of water

used,

turbulent

flow,

and

the relatively

small

volume of

air

in

the

bathroom.

Therefore,

the McKone

and

Knezovich

study,

which

measures

the

evolution

of

TCE

into

a

bathroom’s

air

during operation

of

the

~

~

~

~4~l~r.,’t

r’f

ti,’

~

,.-,‘.~

-..e

~c’rt,i~

~

‘.1.,.-.

showering.

This

study

evaluated

the

effects

of

shower

temperature

and

duration

on

the

transfer

efficiency

of TCE from

water

to

air,

concluding

that

the

transfer

efficiency

is

61 ±

9

and

that

inhalation

exposures

in

the shower

could

be

equivalent

to

an

ingestion

exposure

---

of from 1~4liters per day.

Assuming

that

the

transfer, efficiency of any

VOC

for

which

transfer

efficiency

has

not

been

measured is directly proportional

to that of another VOC having a measured transfer effici~ncy,

the

transfer

efficiency

of MTBE

from

water

to

air

can

be

estimated

from

the

TCE

data

by

comparing

the overall

mass

transfer coefficients

from

water

to

air (K~)for

both

chemicals.

McKone

(1987)

has

shown that K~

can

be

approximated by:

K~

r

2.5

+

RT

~1-’,

where

LD~L

2/3

~3A

2/3

J

DBL

=

diffusion

coefficient