You are on page 1of 99


Technbl Report Series

No. 434



(CAS NO. lO&



Public He&h Semiqe
National Institutes of Health
The National Toxicology Program (NTP) is made up of four charter egenciesof the U.S. Department of
Health and Human Services@HI-IS): the-INational Guxer Inst@te .@CI), National Institutes of Health;
the National Institute of Environme,ntalHealth Sciences(NIEQS~, National Institutes of Health; the
National Center for Toxicological Research (NCTR), Food and, t a@ the National
Institute for Occupational Safety and He&& (NIC@H), Ccntera In July 1983, the
Cakinogenesis BioassayTesting Program, WI, wastrakfetred ,to the MI&IS. The NTP coordimttes the
relevant programa, staff, and resourceafrom these PublicIkalth Service agenciesrelating to basic and
applied research and to bioIogica1assaydetie;oprnentand valldatiori.

The NTP develops,evaluates,and disseminatesscientific ~o~~~~~ut potentially toxic and hazardous

chemicals. This knowledge is used for protecting the h&W of,fhe American People and for the primary
prevention of disease.

The studies descriibedin this Tcchmcal Report were performed under ,the dkection of the NIEHS and
were conducted in compliance with NTP laboratory .health.and safety requirements and must meet .or
exceedall applicable f&x&, state, and lo&health and safety regulations. Animal. care and use were in
accordancewith the Public-Health Service‘policy on Humanc,Carc amlUse. of Atiimals. The prechronic
and chronic studies were-conductedin compliance with Food and ~~Ad~~~ati~ (FDA) Good
Laboratory Practice Regul+ions, and ail aspectsof the chronic studies were subjected to retrospective
quality assuranceaudits before being presented for public revi@%

These studies are designedand conducted to characterizeand evaluate the toxicologic potential, including
carcinogenic activity, of selectedchemicalsm laboratory anima& (usWIly two species,rats and’mice).
Chemicals selected for NTP toxicology and carcinogen&s stud@ are enosenprimarily on the basesof
human exposure, level of production, and chemical structure. Selectionpi se.Jsnot an indicator of a
chemical’s carcinogenic potential,

These NTP Technical Reports are available for sale from t&e ~National,Te#nical Information Service,
U.S. Departmenf of Commerce,528s Port RoyalRoad, Springfield, VA 22lGI (703-487~). Single
copies of this Technical Report -are availaMte;:withkt charge %hiIe supplies la& @om the NTP Central
Data Management Office, NIEHS, P.O. Box -12233,MD-AIMIl, ResearchTriangle Park, NC 27709


ay 1993


_. ^
I,?)-Butadiene,NTP TR 434

National Toxicology Program In Lawa Systems

Evaluated and interpted results and reported fimfinp Preped @dip mwrmce audits

C.J. Alden, Ph.D. S.L. Sanith, J.D., principalInvestigator

GA Boorman, D.V.M., Ph.D.
D.A. Bridge, B.S.
S.L. Eustis, D.V.M., Ph.D.
T.J. Goehl, Ph.D.
R.A. Griesemer, D.V.M., Ph.D.
J.R. L&i&get, D.V.M.* Ph.D., Chair
J.R. Hailey, D.V.M. Patblogy Assotiates,Inc.
J.K. Haseman, Ph.D. M.R. Eluietl, D.V.M.; P&D.
R.L. Melnick, Ph.D. Nati?nal ToaicaiogyPioa;rap~
G.N. Rao, D.Y.M., Ph.D. S.L. Eustis, D.V*M*, FhXA
C.C. Shackelford, D.V.M;, MS., Ph.D. NationalTaxiwIqp Program
D.B. Walters, Ph.D. J.R. Ha&y, D.V.M.
K.L. Witt, M.S., Oak Ridge AssociatedUniversities NetionaLTaxic&gy Program
R.G Milier, D.V,M., PhD.
Battelle.P@fic Nort@est
Battelle Pacific Northwest ~~~t~~~s S.R. Qureshi, B.VSc., Ph.D.
Conducted stu&, evaluated pathoiogy findings
B.J. Chou, D.V.M., Ph.D., Principat bvestigator R.C. Sills, D.V.&I,,, Ph.D.
Natioqal Ibxi~logy Program
R.A. Miller, D.V.M., Ph.D. K. Yoshitmi, XXV&i., Ph.D.
~~rn~~l PatiologyLa+x-atories
Experimental Pathology Labor&tories, Inc,
Provided pathology quality asumnce

J.F. Hardisty, D.V.M., Principal Inves~igsrtor

H-R. Brown, D.V.M., M.S. D.D. ~mb~~~, Ph.D., priwipai investigator
B.F. Hamilton, D.V.M., Ph.D. G,F. Corley, D.V.M.
T.k King-Hunter, B.S.
W.D. Shzirp, B.A., B.S.


TECHNICAL REPORTS v SUBC .~...*.***,.b*...*~........~.,..*.. 12

SUMMARY OF TEiXNICAL SORTS- Ew suBcoM~ co e.*,....*.* 13

INTRODUCTION ..,.~,..,..~..~I~*..*~.~.......~...~...‘~..*.~*...~....~...... 15

MATERIALS AND METHODS ,~~,~.~~.,.,,1~‘1~.,.~~~*.*~....~~~~.*.~~~.~~~.~~~.~ 23

RESULTS ~...~~~,~~..~~~.~~...~.....~..~~~....~....,~..~....~~~.........~.,.. 31

DISCUSSION AND CONCLU$IONS ~*.......~.*t..,~.~...*~.~...*..~”........*~*~.,~ 81










Chemical Formula: C,H6 MolecularWeight: 54.09

Synonyms a,y-Butadiene;biinyl; diyinyl; etythrene;vinylethylene;biethylene;pyrrolylene


‘_, 1,3-Butadieneis producedin largevolumesforuse in each group were earning after 9 and 15 months of
the manufactureof synthetic rubber and of thermo- exposure.
i*, plastic resins. In previous inhalation studies
.’ conductedby the NTP (NTP, 19&Q there w& clear Survival ati Body W@gh{@ the &Year Studies: ‘TLvo-
evidenceof multiple organ &rcinogenicity in male year sur&al was decreasedfor males and females
and female mice expdsed to 625 or l,ZSOpcm exposed to ~n~~tra~~~ of 20 ppm or above,
1,3-butadienefor 60 or 61 weeks. To better charac- primarily due to the develo merit of chemical-related
terize exposure-response relationshipsfor neop3lasms maligrtant ~~~1~~. No femaie mice exposedto
and nonneoplastic lesions, toxicology and c&&o- 2oOor 625 ppm or males exposedto 625 ppm sur-
:’ genesisstudieswere conductedby exposinggroupsof vivedto. the end -of the stud@s(males: 35/S@39/50,
_- male and female B6C3FI x&e to air containing 24/50,‘22/50,4/50,0/7Qfemales: 37/SO,33/50,24lSO,
1,3-butadiene(greater than 99% pure) for up to llf50, O/SO,O/70). Mean body weights of exposed
2 years. An additional studyin male B@T3FImice, in male and fern& mice were similar to those of the
which exposureto 1,3-butadienewas stopped after controls.
limited exposureperiods (13, 26, 40, or 52 weeks),
wasperformedto assessthe effectsof varying concer&
tration and duration of exposureon the incidencesof Hematobg& Eflec@ in tdq .&YearS&&s: Hemato-
1,3-butadiene-induo+neoplasm& In V&Q genetic logic parameters were: evaluated after 9 and
toxicoIogy studies were conducted in SaZmoneUu 15 munths of ‘exposure. At 9 months, decreasesin
typhirnutim and mouse lymphoma c@s. In viva erythrocyte cuunts, hemog~~in concentration, and
genetic effects were assayed(in germ cells of male packedred cell volume were’observedin male mice
Drosophila melanogaster and in bone marrow and exposedto 62.5 ppm or above and in female mice
peripheral blood celis of B6C3Fj mice. exposed to 200 or -625 ppm. h&an erythrocyte
volume was increased in male mice exposed to
2-Year Sruc&s: Groups of 70 male and 70 female 625 ppm and in femalesexposedto 200 or 625 ppm,
mice were exposedto air containingO?6.25,20,62.5, At 15 months, decreases in erythrocyte counts,
or 200 ppm 1,3-butadieuefor 6 hours per day, 5 days hemoglol@ concentration, and packed red cell
per week for up to 2 years;.groups.of 90 male and volume and irtcreasesin mean erythrocyte volume
90 female mice were exposedto 625 ppm 1,3-buta- were observedin male and female mice exposedto
diene on the sameschedule. Up ta 10 anima%from 625 ppm.
6 1,3lutadiene,NTP TR 434

Neoplasms and Ntwaeoplastic Lesions in the ZY$ar exposweswere amplest, thesegroupswere placed
Studies: Exposureof mice to 1,3-butadieneinduced in control chambersfor the,remainderof the 2-year
benign and malignant neoplasmsat multiple sites. study. The totalt exposureOf 1,3-butadiene(concen-
Statisticallysignificant increases+t the incidence3of tration times duration of exposure)of the 13- and
neoplasmsat one or more sites were se& at concen- @ -week s&op~s~~ groups was approximately
trations of 20 ppm and higher in malesand 6.25 ppm S,OOO~ppm * weeks5while that of the 26- and 52”week
and higher in females. There was no exposurelevel stop-exp‘osure groups was approximately
in this study at which a significant carcinogenic 16,000ppm * weeks.
responsewas not observed. Statisticallysignificant
increasesoccurred in the incidencesof maligpant The survival of all stop-exposure groups was
lymphoma; histiocytic sarcoma; cardiac hemangio- markedly lower t&m that, of the controls. The
sarcoma;harderian gland adenoma;hepatocelfuiar incidenccs of l~p~~c. iymphoma, histiocytic
adenoma and carcinoma; alveolar/bronchiolarade- sarcoma, cardiac he~~~~a~~, alveolar/
noma and carcinoma; mammary gland carcinoma, bronchiolar adepoma and carcinoma, forestomach
adenoacanthoma, and malignant mixed tumor squamous piiloma and carcinoma, hepato-
(females only); benign and ,malignant ovarian cellular a barderi~~ gland adenoma and
granulosacell tumor; and forestomachsquamouscell adenoca&oma, and preputial gland carcinomawere
papilloma and carcinoma. s~gnif~~t~y i Neoplasmswere induced at
most of thesesitesafter on& f3 weeksof exposureto
Low incidences of uncommon neoplasms also X,3-butadiene.A~d~tio~a~ly~ low numbersof malig-
occurredin exposedmale and femalemice, including uant gliomas and’neurob~&~t~asof the bmin and
intestinalcarcinomasin males,renaltubule adenomas Zymbal’sgland carcinomasoccurred in one or more
in males and females,skin sarcomas(all types com- stop-exposuregraups.
bined) in females,and Zymbal’sgland adenomasand
carcinomasin females. At similar total exposures~the in~dence of lympho-
cytic l~ph~a wasgreaterwith exposureto a higher
Lymphocytic lymphomas appeared as early as concentration of 1,3-but iene for a short time
week 23 and were the principa! causeof death of comparedwith exposureto a lower concentrationfor
male and female mice exposed to 62.5‘ppm an extendedperiod (34% at 625 ppm for 13 weeks
1,3-butadiene.The early and extensivedevelopment versus 22% at, &.#I ppm for 40 weekq 60% at
of lethal lymphocyticlymphomasin .mice.exposeilto 625 ppm for 26 v&&s versus S % at 312 ppm for
625 ppm resultedin a reducednumberof miceat risk 52 weeks).
for neoplasms developing later at other sites.
Exposure-responserelationships for l,J-butadiene-
inducedneoplasmswere more clearlycharacterizedat Gettetic TQko!ogy: 1,3-3~tadienehas been tested
concentrationsbelow 625 ppm and after adjustment both + V&Q aad & vim’ for mutagenic activity.
for intercurrent mortality. In virro, positive results <were obtained in the
Salmonella ~Fh~u~rn gtzne mutation assaywith
Increased incidences of nonneoplastic lesions in strain TAl535; ~u~ge~ic act,fvitywas not observed
exposedmice included bone marrow atrophy; testic- in other S, ~h~~~~ strains (TAZOO,TM, and
ular atrophy; ovarian atrophy, angiectasis,germiaal TA98). 1,3-Butadiene was negative in the mouse
epithelial hyperplasia,and granulosacell hyperplasia; Iymphoma assay for induction of trifluorothymidine
uterine atrophy; cardiac endothelialh~rp~as~a and resistance in LSl7SY cells q&h and without S9.
mineralization; alveolar epithelial hyperplasia;fore-
stomachepithelial hyperplasia;and harderian gland In viva, 1,3-but~d~enedid not induce sex-linked
hyperplasia. recessivelethal ~u~~o~s in germ cells of male
@osopttsla ~~la~ag~~, heever, it did induce
Stop-Eqxxure &I@: The stop-exposurestudy con- significantincreasesin ch~-~o~rnal aberrationsand
sisted of groups of 50 male, mice exposed to sister chromatid ~xc~a~g~,ia bone marrow cells of
1,3-butadiene at concentrations of 2&l ppm for mice exposedfor2 weeksby inhalation. In addition,
40 weeks, 625 ppm for 13 weeks, 312 ppm, for significant increasesin micronucleatedcrythrocytes
52 weeks, or 625 ppm for 26 weeks. After the were observedin ~ri~~cra~ blood samplesobtained

scornmale and female mice exposedto 13.butadiene B6C3Ff ,mk.e b

Jr;rr2 or 33 weeks or 15 months by inhalation, plasms in the
~~~~~~iu~: The previous inhalation studies of brain, and kidney, There was &far evidence of
13wburrtdione in male and female B6C3FI mice ca~~~e~~~ of ~~-bu~~iene in female B6C3FI
provided clear edence of carcinqgerticitj! at, sure mice basedon i~~.~~.i~~~n~ of neoplastnsin
concentrations of 625 or 1;250 ppm. The present the hemato~je~~,s~tern* he&t, lung, forestomaeh,
~~~~al~rj~~ studies - 2-yeat exposures of 6.25, 20, liver, ovary, and mammary gland.
fr2,5,200, or 625 ppm or shorter duration exposures
of 200,312, or 625 ppm - provide a better charaoter- Low incidenoesof ~~~~a~.~~~o~ in male mice,
irrsttionof the eoncentrationdependent~respons&sfor Zymbal)s gland ~~~~orn~ In ma&and female mice,
1,3-butadiene-inducedneoplasmsand nonneop~astie and renal tubule -ad~om~ and shin sarcomas in
I~ions. The present studies confirmed the c&r female mice may.alsohave Sheenrelated to adminis-
avtince of carchwgenicity of 1;3-butadiene in maie tration of l~~bu~ie~e.

* Explanationof Level of Evictence of CarcinogenicActivity is on page 11. A sumary of&m twits cxmtnentsand the pubk
discussionon t&s TechnicalReport appearson pa& 1%
8 l,ZkButsdie~e,NTP TR 434

0,6.25,20; 62.5,200, or 2OfYppmfor 4-Oweeks, p 6.2!$20,62.5,200, or

625 ppm by inhalationfor 312 ppm for 52 we&s, 625 ppm by inhafation for
6 hours daily, 5 daysper week, 63. ppm for 13 we& or 6 hours daily, S daysper week,
for 103wqeks 625 ppm f&26 weeksby for 103 weeks
inhalation for 6 houssdaily,

Body weights z groupssimikarto group similar to J%Qosed groupssimilar to

coRtroJs j cunttols

Z-YearsurvivsJrrdes 35150,39ts0,2450, 22l50, 37&N,33fio. 24/50,11J50,

4lS0,or?0 o/so, on0

Bone marrow: atrophy {O/50, Heart: endotkiat iiyperplasia Bone marrowz atrophy (O/SO,
o/so, otso, 0/48,0149,23?73) (6!50,3/50,7/50,7/50); ot49, o/48, 0149,o/so, llf79)
mineralizetion(Ok% &SO,
Heart: endoth+W hyperpiasia 9/50,14/50) Heart: endotheli hyperplasia
(0150,l/49, o/!%,-2448,
wa, (O/M z/so, l/50, 4/49,5/90, \
5#3); mi~eralizatian(O/!SO, Alveolar epithelium: M); mineraliition (O/SO*
0149,o/50, 1/48,3/48,2W3) hypcrpla+a[W/St?,14f50, 2lx4 0/50,2I49,2/s0,11/80)
lOj50, 11150)
Alveolar epithelium: Alveolar epithelium:
hypepplasia@ ‘SO,g/SO,6/s@, Forestomachepithefium: hyperplasia(S/50,MO, 3/50,
n/49, 17150,Wi3) 9!5@,11/5o,lln8)

Forestokachepithelium: Forestomachepithelium:
hyperplasia(450, 3150,3/50, Ha&e&q g&u-k .hyperpt&ia hypetptasia(4/50,S/49,4147,
6/48,4/48,40172) (4/48,6/48,3f42,7/36) 7w, 14/50,47/79)

Harderiap gland: hyperplasia Testicle:/ atqhy (S/!&I,3Js0, Idverz kpatDce11ularfoci

(MO, 3149,4jso; q47,8147, 3/50, x50) (s/49,14/49,191sO,1200, s/50*
5/40) 4JW
Testicle: atrophy (MO, 3/50, Nardeean gland: hypcrplasia
4/50,2148,6~49,!#72> (l/SO, 5/49,9/48,4/49,4f49,
Ovary angiectasis(4iss, 6/49,
3/48, w50,14fm 17f79);
granuiosacell hyperplasia
(l/49, o/49, 2m,3fso, 4/50,
2/79); germinalepithelial
hyperpktsia(2/49,3/49, W48,
15150,W54 18/79);atrophy
(4/69, 19/49,32l48,42/50,

Uterusz atrophy (l/SO,O/49,

tadiene (continued)
M@ kW3FI Mice FemaleBK3F, Mice
(t-Year St&y)

Lymphoma (ail tymphoma@ Lymphoma(at! fymphomas) Lymphama (all lymphomas)

(4/50,2430,(r/50*6/50,2150, (8/50*wso, wgo, 3WO) (6150 12J50,11/50,7/50,9~50,
5lf73) ‘-1
Lymphocytictymphoma(2&O, LyntphwyticIymphoma(l/50,
0/50,2/50,4HO, wso, 49P3) 380, do, 3f5098/50,3liq
Hiitioeytic sateoma(XW,
Histiocytic sarcoma(O/SO, 7150,mo, 2750) Histioqtic sarcoma(3/W,
of54 4/50, ma, 7/50,4/n) use, 7/w, 4/50,7/50, a/so)
Heart: hemangiosarcoma k@art: hemangiosarcoma
(0150,0/49,l/SO, 5f4g, 20/48, (O/so O/50,O/50,l/49,21/50,
4f73) Lung: a~~~r~~o~r -1
Lung: aiveoJar/bronchiolar carcinoma(W50, Lung: aiveolar/bronchiolar
adenoma,adenocarciuoma,or 17/SO) ,adenoma;adenocareiuoma,or
carcinoma(21/50,U/SO*19/50, carcinoma(4k30,X+/SO,X9/50,
31/49,35/50,3n3) Foreatotnachz.~~rn~ cell 24/~0,2!mO,rn)
papillomaor squamouscell
Forestomach: squamousceh carcinoma@/%I,9/N, .7/5O, Foreatomaeh:squamousceil
papilioma or sqnamouscell 10/50) papillomaor squamouscell
carcinoma(160, o/50, om, carcinoma(ORJO, O/SO,3/SO,
l/50, 8/S034I73) 2m, 4/so, ic?m)
L&en hepatocellularadenoma Liver: hepatocellularadenoma
or carcinoma(2X/50,23/50, l-bad&m gland: adenomaor or carcinoma(15/49, x4/49,
~3oJSO 25/48,33l4g,Sf72) carcinoma(27/54 ~~,.~~, -U/SO, 19/50,16#0,2430)
Harderian gland: adenomaor Harderian glandz adenomaor
carcinoma(6/50,780,9@, Preputia!gland: carcinoma csrcinoma(WO, lO/sO,7#0,
20/50,31150,6in3) (l/SO, 4f50,4L50,3/sO) l!wO* 20/50,9&q
Preputial gland: carcinoma Kidney rettat tubule acWmma Ovaty: benign or malgnant
(O/SO,o/so, o/50, o/50, %%I, (4/48,3/49, l/SO, l/SO) granuloaacelt tumor (1149,
0173) OJ49,It4&,9/50, WSO,60’9);
Brain: malignantglioma (O/SO, adenomaor benign mbted
Kidnqr: renal tubule?adenoma -o/50,wm, My);. tumor (W49,4!49,1J48,4/50,
(O/SO,‘l/SO,om, 314-gl/49> neutoblaqoma(0/5O,O/so, 6m 2m)
W3) 260,0/50)
hiammaty gland: adeno-
mahgnantmixed tuntor (O/50,
2f50,4/50,12/50, WSO, 16k30)

IJnccrtaln fiRdiq& Smallintestine:xarcinoma Kidney renal tubule adenoma
(O/50,l/50, S/SO,l/SO, 280, pv4930149,ws, Of50,ztso,
onq Orso)
Skin, subcutsueoustissue:
neurotibrosarcomaor sarcoma
(1150,wso, 360, 5/50,3l!!O$
ZymWs.gian& adenomaor
o/so, Of50,2&0)

Level of widace of carciIJog$ntc activity Clear evidence Clear evidence

Gfsetic kudceiogy
Sabnonellarvphimuriumgenemutation: Pc#Jve in strain TAXi%
Neg@vein strainsTANlO, TM, qd TA9S
Mouse iymphomagene mutation: Negativewith and without $9
Sex-linkedrece&ve.lethal mutatiotis
Dmophila??x~m N&g?&iveby inhalation
Mouse bone marrow in viva: Positive
Sister chromatid exchanges
Mouse bone marrow in viw Positive
Mouse.peripheral blood erythrocyte&!
in viva: Positive
1’R 434 11


nal ToxicologyProgramdescribesthe resttimof indivichnkiexperimentson agent and notes the strength of the
conclusionsregardinge&h study. Negativeresuhs,in which the study not havea greater hmidenceof
n control animals,do not necessarilymean that a chemicalis nor a masm aa the experimentsare
der a limited set of conditions.‘Poshive~resui,rs
demonstratethata -4tq+@tk for iakwetoly alI- unde.r
@ &~d~llona of the study and indicate that exposuretomthe chemicalhas the potentiaf~fdrhazardt~~~~~~. Other organ&ations,
&& &S~hcInternationalAgencyfor!Reaea%ch on Cancer,assigna strengthof op an exatninationof all
~~~~~~ Evidence,includinganimal studiessuch as thoseconductedby the WIP, hnabfsof aposrtre.
%ktrs,the actual determinationof risk to humansfrom chemicalsfound tobe carchmgett&in ~~~~.a~~~ requiresa wider
a~~~ that extendsbeyondthe purviewof thesestudies.

g’rvr;categoriesof evidenceof carcinogenicactivity are usedin the TechnicalReport set%%to summ?rhethe strength of the evidencx
added in eachexperiment:two categoriesfor positiveresults (clear evhksseeand.~~~~);~~~ categoryfor uncertain
(tqdvocaI twidtnce); one categoryfor no obselvable’effects(a0 tuiaoce); and~~t~o~ for experhnentsthat cannot be
becauseof major flaws (iubdequs& stMy). Thesec4@egWiea of iuterpremthfeWichtshmswere fhst adopted in June 1983
rrrd tbcn revisedin March 1986 for use in the TechnicafReport seriesto incorporatemore specificahythe conceptof actual weight of
~&once of carcinogenicactivity. For each,separateexpetiment(male rats, femak rats, malemice, fen%ate m&x), one of the fo&nving
f&c c?Htegotiesis selectedto describethe findings. Thesecategarie$refer to the strength of‘the experm&ntalevidenceand not to
potencyor mechanism.

l Clear evhknce of carcinogenicactivity is demonstratedby studiesthat are interpreted as sh+u&tga dose-t&ted

(i) increaseof malignantneoplas6ts,(ii) in& of a combiiination
of ~a~snt and benign neBpasms,or (iii) marked
increaseof benignneoplrrsm~if there is an indica&n from this or other studiesof the abili~ of such tumors to progre~ to
* Someevidenceof carcinogenicactivity is demonstratedby studiesthat are interpretedas .showhtga chemical-related
increasedincidenceof neoplasms(malignant,benign,or combined}in which the strength ofthe responseis lessthan
that required for clear evidence.
l Equiveeafcvklenceof carcinogenicactivity is demonstratedby studiesthat are fnterpreted0 showinga marginal
increaseof neoplasmsthat may be chemicalre&ted,
0 No tvidtnte of carcinogenic< activity is demonstt@?dby studiesthat are interpreted as s no chemicakehted
increase8in malignantor benign neopfasms. _
* Inadequatestudy of carcinogenic~activity is demonstratedby studiesthat, be@&seof ~j~.q~i~t~ or quantitative
limitations,cannot be interpreted as valid for sh&ng either the presenceor absenceof carcinogenicactivity.

When a conclusionstatementfor a particular t%perimentis selected,considerarionmust be given to..keyfactom that would extendthe
actual boundaryof an indiidual categoryof evidence. Stmhconsiderationshouldall- for incorpo@cm of scientific euperienceand
current understandingof long-term carcinogenesis
studiesin laboratory animals,eapecial$for th~levahmtiolt8 that may be on the
borderline betweentwo adjacentlevels. Theseconsiderationsshould include:

l adequacyof the txperimentai designand conduct;

l occurrenceof commonversusunccmmonneopi@ia;
l progression(or lack thereof) from benign to mahgnantneoplasia,a8well as from preneoph%%rie to neoplasth-lesk)t’ts;
l somebenignneoplasm?havethe capa& to regressbut others
is impossibleto identify the difference. Therehxe, where pro
courseis to assumethat bet&n neophW%s of these typeshave
* combiningbenignand malignanttumor incidenceknownor tho
or t&suet
l latencyin tumor induction;
* multiplifzityin site-specificneopIash3;
l metas-
* supportinginformationfrom proliferativelesions(hyperphrsia) in the samesiteof neqdasia
or nt other experiments(same
lesion in another sexor species);
* presenceor absenceof dose relationships;
* statisticalsignificanceof the observedtumor increase;
. concurrentcontrol tumor incidenceas well as the historicalcontrol rate and variab%tyfor a specificneOphWn;
l surviveil-adjusted analysesand false positiveor false negativeconcerns;
0 structure-activitycorrelations;and
* in somecases,genetictoxicObgy.
12 1,34Wadiene, NTP TR 434



The membets of the Technic+ Reports Review Subcommitteewho evaluated the dmFt NTP Technical Report on 1,~butadiene on
November 21,1991, are listed below. Panel membersserveas independents&ientists,not as representativesof any institution, company,
or governmentalagency. In this capacity,panel membershave five major responsibilitiesin reviewingNTP studiesz

l IO ascertainthat all relevant literature data have been adequateIycited and interpreted,
l to determine if the design and conditions of the NIP studieswere appropriate,
0 to ensure that the Technical Report presentsthe experimentalmsultwnd conclusionsf&y at-$ c&&y,
* to judge the significanceof the exparimentaIresults-byscientific criteria, and
l to assessthe evaiuation of the evidenceof c@inogenii activity and other observedtoxic responses.

Curtis D. Klaassen,Ph.D., Chair Jay I. Goodman, PkJ3., PrincipirlReviewer

Department of Pharmacologyand Tozdcology Department of -P~~a~io~ and Toxicology
University of KansasMedical Center Michigan State Umversity _
KansasCity, KS East Lansing;Mf

Paul T. Bailey, Ph.D. David W. Hayden, D4V.M., Ph.D.

ToxicologyDivision Department of V~~,~~.Path#bi~l~
Mobil Oil Corporation College of Veterinary Medicine ~
Princeton, NJ University of Minnesota
St. Paul, MN
Louis S. Beliczky, M.S., M.P.H.
Department of Industrial Hygiene Daniel S, ~n~~~~r~ MD.*
United Rubber Workers International Union Departmat of ~~tholo~
Akron, OH Dartmouth.Medical Schboi
Lebanon, NH
Gary P. Carlson, Ph.D.
Department of Pharmacologyand Toxicology Barbara ~~~~~~t,~ Ph.D.
Purdue University Depart&tentof $3iostatistics
W&t Lafayette, IN University df’Washin&ii
Kowetha A. Davidson, Ph.D.
Health and SafetyResearchDivision Ellen K SilbeqyAd; Ph.D.*
Oak Ridge National Laboratory University 0f’Mwyiand h&x&ii S@wot
Oak Ridge, TN Baltimore, MD

Harold Davis, D.V.M., Ph.D.* Matthkw J. van Zwiet~n, D.~M., Ph.D., Ptincipsl&viewer
SchooIof AerospaceMedicine Department of Safe* Assessment
Brooks Air Force Base,TX Merck Sharp & Dohme ReaeamhLaboratories
West Point, PA
Robert H. Barman, D.V.M.
Consultantsin Veterinary Pathology Lauren Z&se, PI@., ~rincipaI reviewer
Murtysville, PA California Deprtmqt of Hrr;tltb services/RCM
Berkeley, CA

*Did not attend


On November 21,1991, the draft ‘I’echnical Report analyze4 ~~~~1~ @t, in evaluating the e&et of
on the toxicology and careinogenes~sstudi& of l,%butadiene on an again the thinking was that all
1,3-butadienereceived public review by the National of the evidenceshould be blroughl to bear in drawing
Toxicology Program Board of Seientifie Counselors W~l~iQ~. Dr. ~~~~ t&ed that justification be
Technical Reports Review SuWmmittee. The given for the use,of ~~li~k~ recessive.lethal muta-
review meeting was held at the National Institute of tions in
Environmental Health Sciences,Rixearch‘‘I’Wngle ycleus t
Park, NC. the .a. meJ&ogu.&r assay” is ,extremeIypredictive for
~r~R~ge~~~ aa there i&e very few false positives
Dr. R.L. Melnick, NIEHS, in~oduced the to&ology and that the ~~~~~~~~ test is the only simple
and carcinogenesisstudiesof 1;3-butadienein 136C3F1 measureof somatic rnu~tiQ~ in v&o.
mice by discussing the uses of the ~hemi~i ‘-and
rationale for study,including the previous NTP study Dr. Z&e, the second~principalreviewer, agreedwith
in mice; describing the experimental design for.both _(I3ecausethe study was
standard and stop-exposure studies, tihick were at the issue of dose response,she
intended to assess the relationship of er&~~re eyelid analysis d the dose-
duration versus concentration on .carcinogeriici~~ responsedata should;‘beincluded, especiallypertain-
reporting on survival and toxicity, especially $0 the ing to the shape of Be curve at lower doses.
hematopoietic systemand gonads in both sexes;and Dr. M&tick &id some ,d&xrssion could be given
presenting data on neoplasms and nonneoplastic of the dose-responsecurve and the
lesions caused by 1,3-butadisne at multiple site%in to provide nedplasth rates adjusted
both sexes. Dr. Melnick reportecl on the &pression rno~~i~. Dr, JX Haseman,
of K-ras ommgenesin liver neoplasmsand said that cum?@ &at mathematicalmodel-
the K-ras is the most commonly detectedoncogenein ing of the data might to extrapoIat.ionand risk
human cancers. Dr. C.C. Sha#eford, NIEHS, assessmeit ~~~~a~# et&ities thai are normally
provided a morphologic descriptiort of hemangio-
sarcomasof the heart induced by &3-but&liene. The
proposed conclusions were that the present studies
provided a better characterization of the
concentration-dependentresponsesfor Id-butadiene-
induced neoplasms and nonneoplastic lesions, and
confirmed the clear evidenoe of c~~~e~i~i~ of
1,3-butadienein male and female B6C3F1mice. evaluations.

Dr. Goodman, a prin@pal reviewer, agreed with the Dr. van !&v&ten, the third principal reviewer,,agreed
proposed overall condusions in male and .feniale with>the con&usions. He thought there
mice but disagreed with the inch&on of brain and shouldtibe t.he4xmciusionsto the effect
kidney neoplasms in males and iivq neoplasms in that a carcinogenic,reqonse was induced at all expo-
femalesas support for rhe level of evidence.rHe said sure, lev& Also, a ment aboutduration of
the last sentenceshould read “equivocal evidence~of exposurenecessarybr @arciuogenic responsein the
carcinogenicity”instead of “low incidence of! and the stop-e$osure sttidy would be appropriate.
reference to Zymbal’s gland carcinomas in males Dr. M~lni~,.~id that statements would be brought
should be omitted. Dr. Melnick thought that for low forward to the Abstract.
numbers of rare neoplasms “low incidence” v&s
meaningful; however, another wording would be Mr. Belie&y reported that the data from these
considered. Dr. Goodman thought that the eonclu- studies had been recently used by NIOSH in con-
sions for the stop-exposurestudy shoild be presented ducting a risk ~~rn~~t and the results have been
separately from those for the 2-year studies. provided to the :De~~tment of Labor fur potential
Dr. Melnick noted that the results are presentedand regulatoiy action by QSHA on allowable exposure
14 l,?l-Hutadieue,
NTP TR 434

levels. Dr. Garman asked if separate e?ass~~~tions for the stop-exposurestudy wo,uldbe added: “There
of lymphomasreflect the current r~~e~d~t~n of
the NTP. Dr. SC. Rustis, .NIEHS, said average mice based on
distinctions betweentypes;were difficult to m?ke and
of little value. Rather, identifying whether the hematopo~eti~ system,I lung, forestomach, and
lymphomasoriginated in the thymusor elsewherewas hard~~a~~gland. Fi~a~~ ‘in the last sentence,“low
most useful. incidenees’ would be replaced with “marginal
increases.” ‘$be rn~tio~ was tabled for lack of a
Dr. Goodman moved that the Technical Report on seeend, Dr, Z&se muved that the Technical Report
1,3-butadienebe acceptedbut with the oxu$tsions on 1; a ted with the revisions
for the 2-year studies separated from those for the discussedand. tit ns as written for
stop-exposurestudyby inserting “chronic exposureto” male and femaIe clear evidenceof
in front of “l,3-butadiene”in the statementsfor male ~~~~e~~~. M r. ]Beliczkyseconded the motion,
and female mice. “Brain” and “kidney” would be and it was accepted by eight yes votes to one
deleted from the listing for male “mice and ‘liver” no vote (Dr. Mornay) with one abstention
from the listing for female mice. Then a c&%&on (Dr. Bailey).

cxs No. 106-99-a

ChemicalFormula:: C,H, MolecularWeight: 54.09

Synonyms o,y-Butadiene;bivinyl; divinyt; erythrene;vinylethylene;bietbylenq pyt=mIylane

emissions of 1,3-butadierzrefrom facilities that

produce or process~,3-b~~iene were approximately
10 million potmds peryeaq 70% of these emissions
1,3-Butadieneis a colorless, noncorrosive gas with a were attributed to equ#pment leeks and 30% to
boiling point of -4.4* C and a vapor pressure of processventing (Mul~i~, B90). lJ-Butadiene has
1,900mm Hg at 20” C (Kimhenbaum, 1978). The also been ide~~~~~in au~~obile exhaust,cigarette
conversion mctor for 1,3-butadiene at 25* C and smoke,a@ gasolineformulation; smaif amounts are
76OmmHgislppm = 2.21 mg/m’. l$-Butadiene is releasedby the;burning bf plastics or rubber (Miller,
a reactive material that can form the dimer Wnyl- 1978). Low Ieve& of l,$Lbutadiene (US to 10 ppb)
cyclohexeneand is flammable at atmospheric&u.?en- have b,eendetected in @bient air. in urban locations
trations of 2% or higher. l&Butadiene can ,form in the UnW States;however,levelsof l$butadiene
explosiveperoxidesin air, and therefore is shipped as in community air in:Port Neches,Texas,a town with
a liquified gas under pressure with a oxide a butadiene ‘pr~u~tio~ facility and’two styrene-
inhibitor. butadiene production platqs, were measured by the
TexasAir Control Board to be as high as 2 to 3 ppm
1,3-Butadiene is a coproduqt in steam cWl$ing of (I&-chin, 1990). Appr~~rn~t~y 52@0 workers are
petroleum fractions for the manufacture of ethylene. potentially, exposed to l&butadiene annually, as
The annual production volume of J,3-butadlene is estimated from data cqmpiIed from the National
approximately 12 billion pounds worldwfde and otxtap;BtionaI ure Survey (NIOSH, 1990). In-
3 billion pounds in the United States(Morrow, 1990, depth i@mtria~ hygiene sm-veyswere conducted by
USITC, 1990). The major usesof 1,3-butadieneare the National Institute for Occupational Safety and
in the manufacture of. synthetic rubber (such as Health at four monomer and five polymer manu-
styrene-butadienerubber or polybutadiene rubber) facturing plants (Fajen, et aL, 1990). Occupational
and of thermoplastic resins. Butadiene elastomers exposures to ~~-b’~~d~e~ in, most process areas
are used in the manufactureof rubber tires, footwear, were less than 10 ppm; however, maximum S-hour
sponges,hoseaand piking, l&gage, packaging,and a time-~e~ht~ average exposures were frequently
variety of other molded products, between19 and 150 ppm, and in one casethe average
exposurew&s as high’as 3’74ppm. These exposures
According to a 1984 survey by the United States ouqred‘ln o~~ti~hs involving decontaminatingand
Environmental Protection, Agency, atmospheric margining process equipment, sampling and
16 If-Butiadkme, NTP l-R 434

analyzingquality controt samples,and loading or un- ure fat 8 to 10 minutes induced deep
loading tank trucks or rail c&s. The odor threshold th due to respiratory paralysis oc-
or recognition concentration for ~~3~b~~die~e in air cymd after- a 25 to JS-minute exposure to this
is approximately 1 to 2 ppm (Amoore and Hautula, ~~~nt~~ti~~ of ~,3-~~~diene (Carpenter et aL,
1983). 1944).

The &hour, time-weighted, average workroom No tr~tme~t~r~~t~ grossor microscopicchangesor

permissible exposure limit for 1,3-bumdieneeatab- effects. on growth, swivel, hematologic or blood
lished by the United States Gccupa,tionalSafetyand biochemical parameters+urinary measurements,or
Health Administration (aSHA) is 1,000ppm (U.S. neurumusdularfunctions were observedin male or
Department of Labor, 1981). J&x&s of carcino- female ~prag~e~~a~l~ rats exposedto l,OW, 2,000,
genic&y studies of 1,3-butadienein rats and mice 4,ooO,or 8,000.ppm 1,3&tt@ene for 6 hours a day,
prompted the American Conferenceof Gove~men~l 5 daysa week, for 13 weeks (Crouch et cd, 1979).
Industrial Hygienists to loqer their recommend@
threshold limit value for 1,3-bumdienein, the’work N~~~~,~ti~ lesio$s associatedwith ,exposureof
environment from 1,000ppm to lO,pp,m (ACGIH, BflC?F, mice to 625 or 1,250ppm 1,3-butadienefor
1986). OSHA has proposed to lower the occupa- up to 6i wee& i~~l~~,~,ep~the~ial hyperpiasiaof the
tional exposure standard to ‘a perm&sible,exposure forestomach,~~d~theli~l hyperplasia of the heart,
limit of 2 ppm $ith a 15minute short-term exposure alveolar ep~tbeiial hyperplasia, hepatocellular
limit of 10 ppm (OSHA, 1990). A final decisionon necrosis, testicular atrophy, ovarian atrophy, and
the proposed change is pending. An ~ter~atio~a~ lesions in nasal tissu& including chronic in&m-
symposiumon the “Toxicolo&y,Carcinogeneais,and mation, fibrosis, osseousand cartilaginousmetaplasia,
Human Health Aspectsof 1,3-Butadienbwasheld at and-atrophyof the 01~~0~ epithelium (NW, 1384,
the National Institute of .En~ro~~en~l l-$ealth Melnick et al., 1988), The proliferative lesionsin the
Sciencesin 1988. The proceedingsof that symposium forestomach, heart, and lung may represent early
were publishedin EnvironmentalHealtb Perspectives preneoplastic changes in the development of neo-
(Melnick et aL, WOa). plasms induced by fi$butadiene. The nasal lesions
were seen only in male mice exposedto 1,250ppm
1,3-Butadienehas long been consider& to have a Erposure’of male B6C3F1.mice or NIH Swissmice to
low, noncumulative toxicity in animals and humans. 1,250ppm ~~3-~utad~e~e for 6 weekscauseddecreasea
For rats, the median lethal concentration (LG,) for in erythrocyte counts, ~~oglobin concentrations,
a 4-hour exposure was 285 m&n, equivalent to and bematocri$,and an increasein mean erythrocyte
129,000ppm or 12.9%; for -mice, the LC, for a volume (Irons et uL, W36a,b). Anemia due to
2-hour exposure was 27O’mg/L, equivalent to exposureto l~~~ntadie~e was not accompaniedby
123@0 ppm or 12.3% (Shugaev,,1969). Carpenter increasesin reticulocyte counts or in the frequencyof
et al (1944) exposed groups of 24 rats, 12,~inea nucleated.erytbrocytesin peripheral blood. These
pigs, 4 rabbits, and 1 dog to afmosphereacontaming changeswere con&dered to represent a maerocytic-
600,2,300,or 6,700 ppm 13-butadienefat 7.S,hours me~lob~~i~ anemia, becausethey were accompa-
a day, 6 days a week, for 8 months. The highest nied by mild m~galabl~s~~c changesin bone marrow
exposureconcentrationcausedslight growth retarda- cells. In relatedstudies, Tice et al. (1987) reported
tion and, in some animals,,‘aniild reversiibledegener- that exposureofmale B6aF1 mice to 1,3-butadiene
ation in the liver. This degenerationv$s reported as for 10 days causeddecry in the number and rate
light cloudy swelling. There were no reported of dividing cells in the borxemarrow. These findings
treatment-relatedeffects in hematologic para&eters establishedthe bone martow as a site of toxicity for
or blood or urine chemistries,,norwere therepatho- 1,5butadiene in mice. sure ‘of male B6C3F,
logic changesin the eye,adrenalgland,heart, kidney, mice to 1,250ppm 1,3dbutadienefor 6 hours a day,
skeletal muscle, pancreas,spleen, testis, or ovary: 5 daysa week, for 6 or 12 weeksdid not produceany
Exposure of rabbits to 250,000ppm (25%) persistent defects. in humoral or cell-mediated
1,3-butadienefor 2 minutes inducedlight anesthesia, immunity ~~r~ond et al, 1986).
Introduction 19

METABOLISMANJI Drsfosmo~ collected,it was cot .~ssible to determinethe actual

Malvoisin d UJ!(1979) identified 1,2-epoxy-3-butene percqttage of l,J-butadiene absorbed. Instead,only
as the first metabolite in l&butadiene metabolism; ttict ~r~~~ge of Italy ft4e]-butadieneequivalents
this intermediate is formed by an .indudblelrat liver that ,was retained at the end of the exposurewas
microsomal cytochrome p-450 monooxygenase reported, Aft&6 huum,of’exposure,the percentage
(Figure 1; Bolt et uL, 1983). 1,2-Epoxy-3-butene was of 14Cretained ~ang~~~rn 1.5% to 17% in rats and
also detected in the expired air of SpragueXWvley 4% to 2Q%in.mice;‘for each species,the percentage
rats (Rolt et aL, 1983; Filser and Bolt, 1984)and of retained decreasedas the exposureconcentrationof
B6C3F, mice (Kreiling et a&, 1987) e&posedto 1,3-butadieneincr yet the total amount of
l,Sbutadiene, indicating @at this epoxide inter- 1,3-bu~~~ene.i~~al~ and retained was increased.
mediate is systemicallyavailablein exposedanimals.
Further metabolic transformation of :1,2-qoxy- In a .follow~upstudy (D&hi et aL, 19911,the respira-
3-butene involves conjugation with ~u~t~one by tory data of Bond er a& (19S6)were combinedwith
glutathione-S-tians’ferase, oxidation to the met&wlic e~irn~nat~~rate data of Laib et al,
1,2:3&diepoxybutane, or hydrolysis by qoxide (19!3@to obtain valuesfor: the percentageof inhaled
hydrolase and further oxidation to 3$-epoxy: 1,3-butadlenethat was eliminated as metabolites.
1,2-butane&A (Malvoisin and Robe&&d, 1982). The ~lc~lat~ valu”~ were 15% for rats and 12% for
mice eqosed to either 110.or &Xl ppm 1,3-butadiene.
The,latter data&kct the constancyof uptake as the
In studiesby Laib et aL (1990), the metabolicelimi- expdsura~~~~~atio~,~f 1;3-butadieneis increased
nation of 1,3-butadieneor 1,2-epoxy-3-butene was in the rangeof ~~t~rd~r kinetics and also reflect the
evaluatedby measuringthe ,declinein conc@ttration striking sirni~~i~ +tw-&n rats and mice after adjust-
of thesechemicalsin the gas phaseOf desico&x jars ment for speciesdifferencesin breathing patterns.
containing Sprague-Dawleyrats or B6C3F~ mice.
Saturation of l&butadiene metllbolism ‘in each In rats and mice exposed to [14C]-1,3-butadiene,
specieswas reported at atmosphericconcentrations meqsuremeutsof tissuqconcentrationsof 14Cdid not
between 1,OQOand 2,000ppm. At comxntrations reveal any.apparent spec&sdifferences(Bond et aL,
below1,000pim, wherefirst-order kiietics apply,the 1987). 14C!was distribute to all tissuesexamined
metabolic clearancewas 2.6 times higher in mice without any n~ti~~l~ higher accumulationsin the
(7,300ML/kg per hour) than in rats (4,508mLlkg per target organs for carcinogenicityof either species.
hour) (Bolt et & 19sQ;Kreiling er sL, 1985’). The Thesestudiesdid not identify the metabolitesin the
slightly higher metabolic elimination rate in ‘miceis tissuesof expo&d rats ‘and mice.
probably due to the higher respiratory frequencyby
this strain and species. This ‘conclusionis basedon Spe&x differences in the metabolism of
the fact that the metabolicelimination rate uxtstants 1,3-butadienehave also been examined in in vim
of 7.6 hour-’ for mice (Kreiling et d, 1986) and stu&ies. Rates of met&al&m of 1,3-butadienewere
8.8 hour-’for rats (Bolt et a$, 1984)are ne&rlyequiv- slightly lower in microsomal fractions isolated from
alent, and the exhalation rate constants are also the liver or lung of Sprague-Qawleyrats than from
simiIar for these species (Rreiling et aL, 1986),, similar ~re~ar~ti~~ obtained from 36C3F, mice
whereas the rate constant for the uptake of (Bond .et aL, 1988). E!x&sure of rats or mice to
1,3-butadiene(Kreiling et aZ.,1986) and the minute 1,3-butadienefor 6 hours a day for 5 days neither
air volume @erbody weight (Bond d aL, 1986) are indu&d nor inhibited the microsomalmetabolismof
about 2 to 2.5 times higher in mice than in rats. this chemicalin~eitherspecies.The rate of formation
of ~,2~~~3~butene :from 1,3-butadienewas about
Bond et aL (1986) exposeclSprague-Hawleyrats and seven time higher in lung postmitochondrial frac-
l36C3Ftmice to various airborne concentrationsof tions obtained from mice than in similar fractions
1-[“Cl-1,3-butadiene and determined the uptakG* obtained from ~~ag~e-~awl~ rats (Schmidt and
distribution,’and elimination of “4C after specific Looser, 1985). No activity was detectedin a human
periodsof exposure.Respiratoryrn~uremen~ were lung sample. In liver postmitochondrialfractions,the
alsomadeto determinethe uptake of inhal@d1,3-bu- rate of formation of 1,2-epoxy-3-butenewas only
tadiene. However, because1,3butadien;%and its abptt 50% greater far mice than for rats or for a
metaboliteseliminatedduring the exposurewere not single human liver sample. Although this study

18 1,3-Butadiene, NTP TR 434




Epoxide NADPH
Hydrolase Cl, Mimsomes

CH, =CH- CHOH - CH,GH CH, -CH----CH-Cl-i,
\ o,l
3.Butene-1,2diol. Dkpwybutane

0, Miccmtnes
- CtiOH - CH,OH

Introductian 19

does not provide data on the variability of this laboratory rats, ~~~ri~g at a rate of about 0.2% to
activity in the human population,:itdoeaind&xte that 1.0% in untredtedmz$leSprague-Dawleyrats (Krinke
pathwaysfor 1,3-butadienemetabolismin the liver et aL, 1985;,Go~i~at~~1986).
may be qualitatively similar acrossspecies.
In l~~~terrn ~n~?lationstudies of 1,3-butadienein
%6C3F,mice.(PRY, :1 Huff et aL, 19S5),groups
of 50 male and SO e mice were exposedfor
The carcinogenicityof -l,$butadiene was~studiedby 6 hours a day,5 daysa week,to air containing0,625,
exposinggroups of’100 Sprague-Hawleyrats of each or 1,250,ppm l&butadiene. Thesestudies,designed
sex to 0, 1,000, or S,O$lO ppm ‘by inhalation for to last for 103 weew, ‘were terminated after 60 to
6 hours a day, 5 days a ;week,.for 2 ye&s (IISRP, 61 weeks because survival was decreasedat both
19Sla;Owen et aL, 19S7). 1,3-I&adiene V+ carcino- exposureconcentrationsdue to malignantneoplasms
genic at multiple organ sites in rats, as evidencedby
occurring in muiti~l~ organsof malesand females.
increased incidencesand dose-responsetrends for
several organ-specificcancers: pancreaticexocrine
neoplasmsand Leydig cell tumors of t&e testis in Malignant l~pho~s, hemangiosarcomasof the
males,and uterine stromal sarcomas,Zyntbal’sgland heart, and lung: neoplasmscmxrred with positive
carcinomas, mammary gland fibroadcnomas and trends in maleand fern&e mice,and the incidencesof
carcinomas,and thyroid follicular cell nixypl+smsin’ these n#~~srn~ at ,both exposure concentrations
females. Further, the averagenumber d mammary were-high&F than those,of controls (Table 1). The
gland fibroadenomasper, rat was incr in both high ineidences0%hemangiosarcomas of the heart
exposuregroups. The occurrenceof nine giial cell were particularly ~~~~1 ‘finding+ because these
neophxsmsof the brain in exposedmale.x&s (con- endothelialceil neoplasmsare uncommonin 136C31Tx
trols, l/loo; 1,000ppm, 4iloOYS,OWppm, 5fOO) may qice, occyrring $I nc[)m? of 573 untreated malesand
also have been related to expo&e to k,3-butadiene 558 untrtited ferna& in recent NTP studie, and
because neuroglial neoplasms are uncommon in they have rarely been induced in long-term studies,

Mafes &males
Neoplasm 0 pm -625 ppm 1,250 PPm 0 ppg 623 ppm 145Q ppm

M-t LynaplkonM o/so 23m 29lSO l/SO IO/49 Ml/49

Hemangicaraxna O/SO MI49 T/49 OJSO W48 lW49

Aiveolar/bronchiolarneoplasm 2/sa 14/49 IS/49 u49 n/48 2349

squamouscell neoplasm o/49 7r40 If44 o/49 S/42 w/49

AcInar 4x3 neoplasm o/so o&w am O/SO 2/49 w49

Granular cell tumor oi49 a45 12443

Hepatoalhtlar neoplasm slso 6149 a49 am a47 s/49

a Incidenceaare atprwxd as numberof n.eopiasmharin~anim&/numberof anitw& examinedmicroscopically.

20 &34hitadie~ NTP Tit 434

Irons and coworkers confirmed by ~o~~~~etric spermatids were adver&y altered by exposure to
analysis of ceil surface markers th&t the type of 1,3-bwadiege(~~~~ et al, 1990).
lymphomacausedby 1,3-butadienein B6C3F1mice is
a T-cell lymphoma (Irons et‘aL, 198%Irons, 1990). Exposureof pregnaurfemaleSprague-Dawleyrats to
Early induction and increasedincidencesof fore- 2OO,l$O?& or z,ooOppm;of 1,3-butadienefor 6 hours
stomachneoplasmswere alsotobserved in femalesand a day*from day 6 thr?~~~ dq 15 of gestation,caused
low-dose males, and increasedincidencesof neo- a dose-related%mase in, the incidence of major
plasmsof the mammarygland, ovary,and liver were skeiesaidefectsof pups @SRP, 1981b). “Wavy”ribs
observed in females. The gte of mahgnant was the most .commoabajor skeletal defect in the
lymphomaswas lower in femalesthan in malesand l@O ppm and &OOQppm exposuregroups. Other
the dose-responses for other neoplasms-werebetter major skeletal defects,~~i~arly in the high-dose
characterizedin females”These.studicsdemo~trat~ group, +ncluded ~~n~~liti~ of the skull, spfne,
that l$butadiene is a potent m~tiple-~~~ car- steiztium,ribs, a&d ilium. An exposure-relatedretar-
cinogenin mice. dation of maternal~,w~ight.~~ was not accom-
panied by’any gable effm on pregnancyinci-
Irons and coworkers compared the ~ducti~n of dence, ~rnpla~t~io~-loss, or gravid uterine weight.
thymic lymphomas and the expressionof murine M&n fernI weightsaitd,q&n-to-rump lengthswere
leukemia retrovirus in, BBCW, mice and NIH S@iss lower in the Mghdose group than in the controls;
mice exposed to 1,250p-pm 1,3lbumdiene for embryonic gro%h ~eta~~~i~n may have been a
52 weeks(Irons et a& 1987,f989; Irons, 199@ The consequenceof reduced maternal weight gain at the
NIH Swiss mouse was used because it not 8,v ppm exposurelevel.
express the ecotropic murine leukemia viruses
expressedin 136C3F1 mice and it has.a background In another ~~~~~~n~~ study, pregnant Sprague-
rate of nearlyzero for thymic lympboma.Theflnding Dawley rats and~Swiss CD-I micewere exposedto 40,
that exposure to l,3-butadiene cause&Ia 14% inci- 200, or l#OO pprn~~f1,3butadiene for 6 hours a day
dence of thymic lymphomas in NIH Swiss mice on gestation days 6 through 15 (Morrissey et al.,
clearly shows that l&butadiene induces this neo- 1990). There, was no ~Wnce of developmental
plasm independentlyof these activated retroviritses. toxicity in rats, although maternal body weight gain
Irons et al (1989) suggestedthat ecotropic viruses was decreased‘in ~the1,000ppm group, In mice,
were involved in the induction of l~ph~~~ by maternal body weiglrt gain was decreasedat the
1,3-butadienebasedon the higher incidenceof tbymic 200 pPm and. 1,000ppm exposure levels, whereas
lymphomain exposed36C3F1mice than in exposed body weightsof rnale~~t~ were reducedat concen-
N M Swissmice. m and above. Thus, the male fetus
is more susceptible than the dam to inhaled
1,3-butadlene,~ifo~~tio~ were not increasedin
rats or mice.

TesticuIar and ovarian atrophy occurred-in l%C3F,

mice exposedto 625 or 1,250ppm 1,3-butadicriefor
up to 61 weeks(NTP, 1984). Also, a concentration- 1,3-~u~die~e~a potent &z ,vivo ctastogen,is muta-
related increase in sperm head abno~alit~~ @as gen&-&.V&QHFh;en tqsted in the presenceof indud
observedin B6C3F,.mice exposedto 200, 1; , or liver S9 ~~i~ati~~ systqns. An overview of the
5,000ppm 1,3-butadienefor 6 hours a day fat’5 days rnu~~nici~ of ,1~3-b~~d~~neand its oxidative
(Morrissey et aL, 1990). Exposure of male Swiss intermediatesis presentedbelow.
CD-l mice to ZOO,l,ooO,or S$UOppm l$butadiene
for 6 hours a day for 5 daysfollowedby cohabitation l,%Butadiene was:, mutagenic in Salrmnella
with untreated female mice for 1 week did not affect qpliirqurium TA1530 and TA1535, strains that are
male fertility but did causean increasein fhe per- sensitiveto base-pairsubstitutions,in the presenceof
centage of female mice with twe or ~more,dead induced liver S9 fractions (de Meester et aL, 1980,
implantations. Becauseintrauterine deathswere not Arc+ et al., 1990). ~u~die~e monoxide (1,2-epoxy-
increasedat 3.weeks or more after exposure,it was 3-butene), dl-1,2:3,,4-diepoxybutane, and
concluded that the more mature spermatozoaand 1,~3,4-die~~b~~~e, c&dative intermediates of
rntFoductioll 21

l&butadiene biotransformation,wereaisomutagenic ~~parati~ g~~~to~~~ studies were performed

to base-pair substitution strains of S. ~~~~~~, with male 36C3FXmice and male Sprague-Dawiey
but without S9 (de ,Meesteret d, 1978; Simmon, rats (~~~~n~~~ et & 1986). In these studies,
1979;Wade et aL, 197%Dunkei et al, 1984;.Canter bone marrow ,ceils of mice exposed to 100 to
et al., 19@, Z&g& 1and Pagano, 1989). lO,OODppm ~~b~ta~i~~e for 6 hours a day for 2 days
1,2:3&Diepoxybutanehas been reported to induce ~how~,si~i~~t: inctleasesin micronucleiand sister
sex-linked recessive,lethat mutations,and. chromo- ~~matid ~~.~~g~~ no incre@ein either endpoint
somai translocations in Dro~upM~ ~~~~~~~t~ was seen in si&iiarly exposedrats. No induction of
(Watson, 1972; Shukia and Auerbach, l%Q G&en nns~fteduie(tDNA s~~b~~s wasnoted in hepatocytes
and Green, 1982,Sanksranarayanan et aL, 1983). isoiated from @ice &d rats exposedto lO,ooOppm
1,3~bu~dien~.for2 da* (Arc%et al., 1990).
In one study, 1,3-butadienewasnegativefor induction
of sister chromatid exchangesin human lpphocyte In ~qnclusion, l,%butadiene is mutagenic-in vine,
cultures in the presenceof rat, mouse,or noninduced inducing gene mutations in S. r)rphimur&n, and
human S9 (Arce et d, l&O);. however, in another ti V&Q,inducingsisterchromatidexchanges, chromo-
study, 1,3-butadienewas positive for the i~d~~tionof somai aberrations,and micronuciei in mice.
sister chromatid exchanges,with or without S9, ‘in
human lymphocytes(Sasiideka$ai!+,.1991),.Increases
in sister chromatid exchanges‘were reported in
Chinesehamster ovary cells fPerry and Eva&s,1975; Early tbxicx>iogystud&!&on l,J-butadiene indicated
Nishi et aL, 1984)and-humani~~h~~.~~d flbro- that this. chemical only causedirritation to mucous
blasts (Friedman et aL, 1982; Qbe et pi., .19&Z membranes,skin,.and.eyesor causednarcosisat high
Porfirio et all, lw, Sasiadeket d, 1991) treated ~~~nt~tio~s (Carpenter et aZ., 1944). Human
with the metaboiite 1,2:3,4-diepoqbutane. In addi- voiuntmrs exposed to 2@00, 4,OUO,or 8,000 ppm
tion, 1,2:3,4diepoxybutaqeim@ed chromosomal 1,3-b~tadie~~for 6 to 8 hours experiencedminor
aberrations in human lymphocytes and fibroblasts irritating to the eyesano difhculty in visual focusing.
obtained from patients with chrotiosome breakage
disorders (Auerbach and:Wolman, 197% Auerbach In, tank farm workers at a styrene-butadienesynthetic
et uL, 198%Marx d & 19w, ~Porfitioef et,, 1983). rubber piant, erythro&ytecounts,hemoglobinconcen-
trations,.and,packed‘r$ ceil volumes were siightiy,
1,3-Butadieneis a potent irt viva genotoxicagent to but not sig~i~~~t~y~lower, and mean erythrocyte
mousebone marrow celis. Exposureof m&e B6C3F, vo&mes were sl~gh~~~, but not significantly, higher
mice to 6.25, 62.5, or 625 ppm i$butadiene for than those of workers in other departments
6 hours a day for 1O:exposuredays prc$uc@ {~~~ko~y~a~d Withams, 1982). Thesechangesare
increasesin chromosomalqaberrations and-sisterchro- similar to those observed in mice exposed to
matid exchangesin bone marrow eelis and’micro- 1,3-butadim-s.
nuclei in erythrocytesobqined from peripheralblood
samples fTice et aE, 1987). The lowest effective Asso&tions between occupational exposure to
dosesfor each of these endpointswere 6s ppm for l$butadiene and i~cr~ dancer risk have been
sister chromatid exchanges,6.25 ppm for chromo- revaluated in retroactive mortality studies of
somal aberrations, and m2.5ppm for micronuclei. wo&e& employed at facilities which produce
EZxposure to 625 to 625 ppm 1,3-butadi~~e,fo~ 5 days lo-b~~diene .(Dow$sd uJ., 19w, Divine, 1990)and
a week for 13 weeks resulted in .significsantincreases at fadlities which produce styrenebutadienerubber
in micronucleated normochromatic .erythrocytes (~~i~~ardt et’&, 198% Matanoski and Schwartz,
isolated from peripheral blood of male and -female 1987;~a~~os~- et & 1990). Increasedmortalities
ERCJF, mice (Shelby, 199t$. The metaboiite atopoietic cancers were
butadiene monoxide was aiso shown to b+ a strong of occnpationallyexposed
inducer of sister chromatid ex&anges and udies. Mortality rates for
chromosomal aberrations in bone marrow ceils of l~ph~ar~~ and reticulum cell sarcoma were
male c57Bi/6 mice given a single hitraperitoneal increased”byas as S&fold amongworkers in a
injection of this chemicai (Sharief ef aL, 1986). 13.butadiene manufacturing plant (Downs et aL,
22 i,LJhtadtene,NTP TR 434

1987),while 5- to 6.6-fold increasesin rno~~~ from exposure, usethe first inhalation studies in
all lymphopoieticcancersa”d leukemiawere reported mice had inat&d early, additional studies
among black workers in production areasof styrene- were performed to better characterize ewsure-
butadiene-rubberplants (Matanoski @ aL, 1 reSpom+e~~~tio~h~~ for neoplasmsand nonneo-
a nested-casecontrol study compariagl~p~~~ietic plastic ~~~0~~~~~~ by ahischemical in mice, Five,
cancer cases to an internal population .of workers exposurelevels~~~~g from 625 ppm, corr&qonding
who did not have cancer, Matanoski et al (1989) to the lowest ~~~tra~~on used in the previous
found that the odds ratio was 9.4 for ,the,association inhalation studies, dowerto 6.25 ppm were included
of the leukemia caseswith l&butadiene exposure. ti, extend the ure- range over two orders of
magnitude. ~d~~o~~l studiesin which exposureto
l&butadieBe was st@pp@after limited periods of
STUDY RATIONALE time were~also inchtd&, to assessthe relationship
Because1,3-butadieneis an iriqortarrt chemi@ with between~~~~~at~~~ aRd.duration of exposureon
a large production volume, and a potential/ for the outcome of b~~die~e-i~du~ carcinogerricity.


l&Butadiene was periodicallyshipped,asa liquified red from the headspaceof
gas directly to the study laboratory, Battelle ,Pacific stainless steel tubing to a
Northwest-(Richland, WA), from Phillips Chemiml nd flow control system. Six
Company, Philtex Plant, (Barger,. TX). Although ~te~~g v@Gs and~@owmeters controlled the gas
seven lots (J-014, J-acts,J-038, J-050, Ll49, J-217, flow to ea;Chchamber(Figure G3). The gas entered
and J-375)were usedin the a-year.andstop-exposure the ~i~tr~rn near the top of the study chambers
studies, only one lot was used at any one time. (EIazleton.2ooQ,Lab Products, Inc.).
Battelle Pacific Northwest analyzed each lot of
1,3-butadieneand confirmed the ideutity by infrared The ~n~~~atio~ of 1,3-butadienein the chambers
spectroscopyand establishedthe purity a8-2QQ%by and room air was monitored with an automated
gaschromatography.The maximumallowabledimer sampling. &stem coupled to a gas chromatograph.
(4-vinyl-l-cyclohexene)content in the hea$spakeof’ The system~a~t~ma~i~~lycycled through all ports
any lot used in the studieswas setat 5tXXppm. The onceevery3Ffminutes. Calibration was performedby
dimer content increasedwith time and wasmonitored analyx#ng vo~u~etr~~~y prepared standards of
daily. Any cylinder yielding a dimer value greater 193-butadiene. At ~kast 93% of all concentration
than So0 ppm was not used. ~~~rern~~~ wire-stain 10% of the target concen-
trations. Monthly mean expostireconcentrationsare
According to NTP practice, a esmplete chemical presentedin Figures;C4 through GS. A summaryof
characterizationis performedon the chemicallo-tthat chamber~n~~trat~~us is presentedin Table 61.
will be usedin the toxicologystudy. The bulk chemi-
cal is then shippedto the,study laboratoe. Because
the dimer content of 1,3-butadiene-increased with
time, it was impractical to follow~thenokal proce-
dure. Therefore, to kllow for the i~~b~~~ of the
chemicaland to determinethe appropr~~~-~na~~~l U~~~i~ of ~~~~tration of 1,3-butadienein each
tests, a representativelot (F&SO) hrom the same exposurechamberwith animalspresent was checked
supplierwassubjectedto a full ,~ha~~e~i~tio~., The at,appro~~te~y,3”~~th intervals from 12 chamber
analyseswere performed by the analyticalchemistry positions by the samesystemused for daily concep-
laboratory, Midwesr Research Institute (Kansas tration monitoring. The chamberconcentrationwas
City, MO), and are.describedin +ppendaxG. The consideredto ‘be uniform if the variability ~8s less
studychemical,a clear,colorlessgas,wasidentified as than 5,% rel;ltive standard deviation. During tbe
1,3-butadieneby infrared and nuclear magnetic s@.tdies, the ~riabili~ did not exceed4.8%.
resonancespectroscopy.‘Lot F-850 was determined
by gaschromatographyto be greater than QQ%pure, The time (Tw9 f~ll~~~g the start of generation for
with no impurities with ,areas of O;l% or greater the w~~~tr~~on to ‘build to 90% of the final stable
relative to the major peak area. Approximstion of concentrationin the yearner and the times (TIOand
the concentrationof the inhibitor, c-butykatechol,in. Td fo~io~g cessationof generationfor the concen-
the liquid phaseindicatedapproximately4,-ppm. The tration to decayto, lQ% and 1% of the stableconcen-
level of 4-vinyl-l-cyclohexenewas $etermixe&by gas tration were ~t~rrni~~ with animals present. The
chromatographyto be 35 f 1 ppm for .the liquid T* T%*and T1 value+were 15 minutes, 12 minutes,
phaseand less than 1 ppm for the headspa+ and 35 minutes, respectively.

The major 1,3-butadiene degradation product duration of exposure.(52:weeks), the l&butadiene

expected in the studies was the. dime& 4+inyl- concentration times exposure duration was kept
1-cyclohexene, formed from the condensationof ,two constant. $~~ly, the total exposureat 200 ppm
moleculesof l$butadiene, Other potential dcgrada- for 40 weeks was near& equivalent to the total
tion productswere the ox@ationproducts3,4+&y; exposure at 625ppm for 13 weeks and equal to
l-butene and l$butadiene diepoxide. As discuss& a~~o~~a~~~~half, the t&a8 exposuregiven to the
previously, the dimer content increasedwith time. group exposedto 625 ppm for 26 weeks. Thus the
Once the headspaceconcentration in a cylinder total exposure to 1~3-b~~d~ene was approuimately
reached500 pptn, the cylinder was no longer used. 487,#f~ppm-hr for the tips of tiice exposedto
To monitor for the oxidation products,a fraction of 625 ppm for, 26 weeb or, to 312’ppm for 52 weeks,
the chamber atmosphere was hubby@ ‘through and a~~~o~a~,~iy%42,tXX+l ppm-hr for the groups of
dimethyl formamide; the dimethyl formamide was mice expos&ito 625 ppm for 13 weeksor to 200 ppm
then analyzedby ,gaschromatography. No evidence for 40 weeks.
of oxidativedegradationproductsat the 1% level was
ObselVed. Soulmemd S~~~~~~~ of Animals
Male and female SB6CZ3F;imice were obtained from
Frederick Cancer~Researcb Facility (Frederick, MD)
&YEAR !!&UDIEZS for use in the 2-year.chronic .and stop-exposure
Study Design studies. ,&f&ewere q~a~~ti~~ 13 or 15 days. Eve
Groups of 70 male and 70 femalemicewereadminis- male and five female‘mice were randomly selected
tered 0 (chambercontrol), 6.25,2&62.5, or 20 -ppm and killed for parasite evaluation and gross
observationof disease.Bfood sampleswere coilected
l$butadiene by inhalation for 6 hoursa day, $ days
a week, for up to 103 weeks,groups of 99 male and for viral screens,Male m&x were approximately6 to
8 weeks‘ohi and females were approximateiy7 to
90 female mice were a~mmistered’62Sppm 8 weeks aid when th& stuiji began,‘The health uf
1,3-butadieneon the samesch$duie. After 9 months the animalswas monitored +-@ ringthe courseof the
and again after 15 months of 1,3-butadiene.
‘studies ac&xding to the protocols of the NTP
tration, up to 10 male and 10 female m&e were
Sentinel Auimai Program,(Append& I).
randomly sehxted from each group for interim
Mice were housed ~i~~~iLy during the studies.
Water was available @ &E&u;, feed was avaiiable
~-YEW STQF-EXPCMJ~ -STUD%? ad hWurh e+qt during .~~ure periods. Cages
Study Design were rotated within the ’ ure chambersweekiy
Groups of 50 male mice were administered during the stu$ies, Further detaiis of animal mainte-
la-butadiene by inhalation at ~n~~trati?~ of nance are given in Table 2; Information on feed
200 ppm for 40 weeks, 312 ppm for 52 we&& or .~rn~it~~~ and ~~tarn~an~ is provided in
625 ppm for 13 or 26 weeks. After the exp$xures Apings W .
were stopped, animalswere placedin control cham-
bers and were evaluatedat 103 weeks. Mice were aad Pathology
exposedfor 6 hours a day, 5 d&u a week. twice daily and findings
rded ‘monthiy or as necessary. Animals
The stop-exposurestudy was designs to test the at the ~gi~~i~g of the .studies,weekly
hypothesisthat carcinogenicresponsesdue.~to‘&po- for 13 Cweeks,and moflthl~ thereafter.
sure to 1,3-butadienewere directly related to the
product of the exposure concentration times the Up to 10 mice from eachgrip in the 2-yearchronic
duration of exposure. flhe exposure-condition of studies w&e eval~~~~ at 9 months and after
625 ppm for 26 weeks was selectedbecausein the 15 months of l~-b~~die~e administration. Blood
previousstudy,lyntphomaswere inducedbythat time was.drawn from the su~~~orbi~ sinus for clinical
in male B6C3F1mice exposedto this concentrationof pathobgy ~val~tions. ne marrow was obtained
1,3-butadiene(NTP, 1984). By.reducingthe exposure from the right. femur of animah evaluated at
concentration by 50% (312 ppm) and doubling,the 15 months- for determination of bone marrow
Matminis and Methods 25

cellularity. The brain,heart, right kidney,h~er,,.lungs, This group examined the

spleen, right testis, snd thymus of each animal of dose groups or previ-
selectedfor the 9- and l%mottth interim ev&ations When the opinion of the
were weighed at necropsy. Further details of the of the laboratory pathoiogist,
interim evaluationsare presentedIn Tablb ,2. Thus, the final diagnoses
contractor pathologistsand
Necropsies were perfor&d on all an&hais. At review procedureshave
necropsy,all organs and ‘tissueswere examinedfor been ~~c~~ by aren’ t and Boorman (1982)
gross lesions,and all mdjor tissueswere fixed and and Boorman a hf. @&5). For subsequentanalysis
preserved in 10% neutral -buffered formalin, da$%, the diagnosed ‘lesions for each
processedand trimmed, embedded-tn.,paraffin,sec- tissue type are ~~~ua~~~ separately or combined
tioned, and stained with hemat&ylin and eosin for according to the guidelines of McConnell et al.
microscopicexamina$on~At the 9- and 15-month (1986).
interim evaluations, complete .~~pat~~io~ was
performedon all conrrol n&e, ali mice in.the highest
exposuregroup with survivalof at least $09, and sll
mice in groups with higher exposurecon~ntration~~ The pr~~a~~Ii~~of ,sblNivaiwas estimated by the
All animals that died or were killed rn@bynd, all pr~~~-li~~t procedureof Kaplan and Meier (1958)
[t-year core study mice, and all stop-exposuremice and is .presentedin the form of graphs. Animals
alsoreceiveda completehistop&~elogic were ce@xcd fro,m*be survivat analysesat the time
Tissuesexaminedare list&d in Tabie 2. they were.found dead of other than natural causesor
were&un$~to be aping; animalsdying Born natural
Upon completion of the microscopicevalluationby caus& were not cen$ored. ~Statisticalanalysesfor a
the study laboratory pathologist, the pathology data possible do~~~~~a~ effect on survival used the
were entered into the T9xicologyData management meth&t of Co% (1972) for testing two groups for
System. The microscopeslides,paraffin bIocks, and equality and Tarone?s(1975)life table test to identify
residual wet tissueswere sent to the Nl?? Archives do-relate trends.’ All reported P values for the
for.inventory,slide/blockmatch,aud wet-tissueaudit. survival an@sesare two sided.
The slides, individual animal data records, and
pathologytableswere sent to an tndependentpathol-
aey quality assessmentlaboratory. The heart, Iung, 73-k~~d~a~ of ~~p~~rns or nonneoplasticlesions
forestomach,liver, ovary, uterirs, harderian gland, is given as the- number of animals bearing such
Zymbal’sgland, kiduey,,small intestme, ;Eind~ skin of lesjonsat a-spe$fic$natomic site and the number of
male,and femalemice,the brain, preputial gland,and animals in which that site was examined. In most
testis of male mice, and the mammarygland, ovary, ins’tances,the de~orn~~to~ include only those
and uterus of fernAle mice were -reviewedmicro- ~~~rn~~ far whieb the site was examined histo-
scopicallyby the quality assessmentpatbolog~st” for lo&ally, Ho#everi when macroscopicexamination
neoplasmsor nonneoplasticRsions. wa;Srequired to detect lesions (e.g., shin, intestine,
mam~a~.g~and,or $arderianglandneoplasms)prior
Tote quality assessmentreport and slides were to histologic sampling,or when lesionshad multiple
submitted to the NTP PathologY~ Working Group .potential~sitesof o$eurrence(e.g., mononuclearcell
(PWG) chair, who reviewedthe sefectedtissuesand ~e~ke~~~,the ~~~~~ato~ consistof thenumber of
otty other tissuesfor which there was a disagreement animals on:which a,necropsywas performed.
In diagnosis between .the laboratory and quality
ment pathologists,Representativemmples of &l&$& ~~~e~~ Incidence
from the forestomach;liver, heart, harderian enlarge ~~~~ of exposedmice that died or were
and lung of male:and female mice,kidney,and killed moribund et@ y in these studies were consi-
bra~n’of male mice, ovary and mammarygland of dered to be due primar3y to lymphomaor to heman-
mice, and lesions of general m terest were gios&oma of the heart. Moreover, carcinomasof
ed by the chair to the PWC?for review. The the foresfomach, rung, preputial gland, mammary
PWO cons+d of the quality assessment pat~olog~t gland, and,Zymbsl’s @andand sarcomasof the skin
rmti other pathologists ~rienced ‘in rodent WfS considered to be lethal neoplasms.
26 l,;)-lgutadieue,
NTP.TR 434

Consequently,for theseparticular neopiasms,p&nary the NTP, modifiesthe ~h~nu~rni~ge~isher exact

emphasisin the analysisof neoplasmincidenc&was test by adjusting the denominatorsof the neoplasm
given to the life table test (Cox, 1972;Tarone, 19751, rates to take into accountstnvival differences. This
a survival-adjustedprocedureappropriate for rapidly_ adjustmentwas derivedby Portier and Bailer (1989),
lethal neoplasms. b& on their fittlng,a Weibull model to historical
control neoplasmdata for a variety of site-specific
For incidentalneoplasms(neoplasrnsdiscovere!las a neoplasms. The use-of the power of k=3 in the
result of death from an unrelatedcause),the primary poly-k adj~t~e~t: for iRt~r~~ent mortality in the
statistical method used in twe studies was logistic l$butadiene study is j~tifi~ by the following:
regression,which assumedthat the diagnosednee- (1) For liver and lung neoplasmsin historical control
plasmswere discuveredas the result of death from an animals, the obse@xdsu&ival adjusted value of k
unrelated causeand thus did not affect the, risk of that best fits the,data~,ap~ro~mately3 in both male
death. In this approach, neoplasm prevalencewas and femalemice@$tier et.&, 1986). For leukemia/
modeledas a logistic function of chemical exposure lymphoma ~rn~i~~, .a gristly higher value was
and time. Both linear and quadratic terms in, time observed(6 in females 5 in males). In male
were incorporated initially, and the quadrat& term mice, the value of k,for aI~~‘~~p?~~ combinedwas
was eliminated if it did not significantly enhancethe approximately3 and for femalesit was approximately
fit of the model. The exposedand control groups 4.5. (2) §irn~a~iu~ haveshown that it is better (in
were comparedon the basis of the likelihoodscore terms of preserving false positive error fate) to
test for the regression coefficient’of dose, This underestimate.tlte value.ofk than to overestimateit.
method of adjusting for intercurrent mortality is the Thus, for example,it is. better to use k=3 when in
prevalenceanalysis of Dim% and Lagakos (X983), truth k=6 thanit is to use k=6 when in truth k=3.
further described and illustrated by Dinse .and Becausenone of the valuesfor k estimatedby Portier
Haseman (1986). When neoplasmsare incidental, d al. (19861were aig~i~~~~y leas than 3, and few
this comparisonof the time-specificneoplasm.preva- were sig~i~~u~~y:greater’than 3, the choice of 3
lencesalso provideaa comparisonof the time-specific seems reasonable. The corresponding survival-
neoplasmincidence(M&night and Crowley~1984); adjustedneoplasmrates are also reported.

In addition to logistic regression,alternative methods ~~~,~~~~~ ~~~-~~~~ shape

of statistical analysiswere used, and the results of for 2~3-3~&~~
theseteatsare summarizedin the appendixes+T&se For those ~~~1~~ $&wing chemical-related
include the Fisher exact teat an6 ,the C&chran- effects; the shape of the dose-responsecurve was
Arm&age trend test (Arm&age, 1971; Gart et aL, estimated by fitting. the following modified Weibull
1979),proceduresbasedon the overall pro~rt~on of model (Portier et a& 2986) to the data:
P(do8e)= 1 _ e-~~t~~e~t=+scale . ctoseshaPe)
Teatsaf significanceincludedpainvisecomparisonsof
each exposedgroup with controls and a test for an where P(&we), is the prob~biIi~ of a neoplasm.prior
overall dose-responsetrend. ~~ti~~i~~~~t~ to study termination for animals administered dose
teatswere usedin the analysisof neoplasmincidence, doseof@but&$iene.~~e’pa~rnete~ htercep~scale,
and reported P valuesare one sided. The procedures and shapeare estimatedvia maximumlikelihood esti-
describedabove also were used to evaluate,~selectedmation using the likelihood
nonneoplasticlesions. For further discussionbfthese
statistical methods,refer to Hasemaa(1984). 5

Becausethe increasedmortality in the 1,3-butadiene-

exposed groups reduced the sensitivity of. logistic
regressionanalysesfor detectingcarcinogeniceff&%s, wherexi is the number of .animalswith neoplasmin
supplemental analyses were perfermed by the dose gi’oup.di?and ni is the poly-3 adjusted number
survival-adjusted “PolyO” quanta1 response test of animals at risk in dose group di, i=O,l,Z,..S. A
(Bailer and Portier, 19&8;Portier and Bailer, 19891, likelihood ratio, test is used to test the hypothesis
This procedure,which -iscurrently beingevaluatedby that the shapeparameterequals 1. The test statistic
Materials and Methods 27

is given as -2 times the differencesin the l6g’likeli- of~d&e+?spo%tse trends atul to determine whether a
hoods. A one-sidedteat was used so that the critic@ tren~-~~~t~e t&t (Willed’ or Shirley’s test) was
values are 2.706 for P=O.OSand S.410‘for P-U.01 more appr~p~te for p@wise comparisonsthan a
(these are the squaresof the eritic$ regions from test that doesnot assumea monotonic dose-response
standardnormal distribution). The sh~,pa~meter (Duunett’s or Dunn’s test).
was restricted to be leasthan or equal to 10. I

If the estimated shape parameter is greater than 1,

the resulting dose-responsehas more cutiature than Qua&+r~ CE h&THODS
a linear model and exhibits “threshold-like” The 2-y&r studies. were conducted in compliance
If the estimatedshapeparameteris iess than 1, then with Food and Drug ~min~tration Good Labora-’
the dose-response curve is very steep in thelaw-dose tory Prz3ctice Re~latio~s (21 CFR, Part 58). In
region. addition,.as study records were submitted to the NTP
Archives, they-were ‘audited retroq%ctively by an
independ& qua&y assurancecontractor. Separate
Although the concurrent control group is alwaysthe audits covering ~rnp~te~~ and accuracyof the,
first and most appropriate control group..used for patbo~o~.da~, pathologyspecimens,final pathoiom‘
evaluation,.there are certain instances in which tables, arid .preliminary review draft of the NTP
Tech@al,Report were conducted. Audit procedures
historical control data can be ltelp$tl in the overail
and findings are~presentedin the reports, which are
assessmentof neoplasm incidence. ~~s,~ue~tly~
on tile. at the NIl%i$. The audit findings were
control neoplasmincidenczr from the ~-h~tori~l
reviewedand assessedby NTP staff so all had been
control database(EBsemanet qL, 1984, 1985) are
resolved or were otherwise addressedduring the
includedin the NTP reports for neoplasmsappearing
preparation of this Te#mical Report.
to show compound-relatedeffects:

Amzlysis of ~~~~.~~~S
?Fto approacheswere employedto &se% the signifi-
cauce of pairwise comparisonsbetween dqsql and The genetic,toxicity oPl,$butadiene was assessedby
control groupsin the analysisof continuousvariables. testing the ability of the Ghemicalto induce muta-
organ and body weight data, wh&h haveapproxi- tions in var@usstrains of Salmonellatyphimurbf,
matelynormal distributions, were analyzedusirig t$te t~~~oruth~dine resistance in L5178Y mouse
parametric multiple comparison proeedufes of l~ph~ma cells, sister chromatid exchangesand
Williams (1971,1972)and Dunnett (1955). Clinical chro~oso~l a~rra~~~s in mouse bone marrow
chemistryand hematologydate, which ha%% typicshy cells, sex-links recessive lethal mutations in
skeweddistributions, were atialyxed using the non- ~sop~~# ~~~~~~~e~, and micronuclei in periph-
parametricmultiple comparison methods of Shirley eral blood e~~~~ of mic& The protocols for
(1977) aud lXmn (19&t), Jonckheqre’s test these studies and. mbular presentations of their
(Jonckheere,1954)was usedto assessthe significance findings arP;in Appendix D.

28 l,HWadiene, NTP TR 434

Z-Year Studies

Study Laboratory
BattellePacific NorthwestLaboratoria, Richland,WA BatteliePacific NorthwestLaboratories,Richland,WA

Strain and Species

B6C3Ft mice B6aF1 mice

Animal Source
FrederickCancerResearchFacility, Fredetick,MD FrederickCancerReaearchFacility, Frederick,hfD

Size of Study Groups

70 malesand 70 females. 90 malesand 90 fern&sin 50 males
625 ppm dosegroups

0, 6.25,29,62.5, 200,or 625 ppm in air z@, 312,or 625 air

Time Held Before Study

Males:15 days 15 days
Fernalex13 days

Average Age When Placed on Study

Males:6 to 8 weeks 6toSweeks
Females:7 to 8 weeks

Date of First JExposure

23 January1986 23 January19M

Duration of Exposure
6 hours daily, 5 daysa week,for up to 103weeks 6 hoursdaily, j; daysa week,for 13,26,40, or 52 weeks

Date of Last Expasu,re

g-monthinterims: 30 October1986 13-weekstop expccure: 23 April 1986
IS-monthinterims: 23 April 1987 26week stop expostlFe:23 July 1986
2-yearstudies: 13 January1988 #-week stop re: 29 October 1986
52-weekstopexposure: 21 J&uary 1987

Average Age When Killed

111 ta 113weeks 111to 113weeks

Method of Sacsiifce
70% carbondioxide 70% carbondioxide

Method af Auimal Distribution

Animals randomtyassignedto control and exposuregroups Animalsrandomtyssaiguedto control and exposuregroups
with body weightas the blockingvariable,usingthe X’YRJON with bodyweight‘a$the.blockingvariable,usingthe XYRION
PATHfPOX Syslem. PAMOX Sy.$tem.

Animals per Cage

1 1
Mat&ala and Methods 29

Experimsental.Design and Mattw&fs and ~~e~~s in the ~~~~~0~ Studies 1,343utadiene(matinued)

lieYear Studies Shop-Exposure


Metbod of Animal Ide&fCcation

Toe clip T&zdip
NHM7 open-formuladiet, pellets(w!gler Bras., Inc., NW07 open-formuladiet, pellets(Zeigler Bras., Inc.,
G$rdners,PA), availabterki &utn, exceptduring Gtiw,, PA), availableud E&if#i, exceptduring exposure
petid @d
Tap water (City of Rich@d water supply)via automatic Tap water (City of Richbnd water supply)via automatic
wateringsystem(SystemsEnglneerlng,Napa,CA), avallable ~aterbrgsystem(SystemsEngineering,Nap, CA), available
ad Iibiwn ad#TE&??~

Stainlesssteelw&e-bottomcage (HazletonSystems,Inc., Stainlesssteelwire-bottomcages(HazletonSystem%Inc.,

Aberdeen, MD), changed we&y Abcrdkn, MB), changedweekly

Stainless steefchambers(Lab Products,,Inc., Narford St&&q steel c@am+s&ab. Products,Inc., Warford

Division,Aberdeea,MD) Division; Aberdeen, MD>

Animal Chamber EovCronment

Averagetemperature:23.f” f 1.5”C Avmge tempeeat~~ C?+T 4 1.5” C
Relativehumidity: 55% 4 15% Relatjve h~rniclity 55% f’ 15%
Fiuorescentlight: 12 hourMay FluoIt?s4!mtlight: 12 llour@ay
Room air changes lWUhour Room air changes:17-21/hour

Type and Frequencyof Observation

Observedtwice daily; weighedl&ally, weeklyfor 13 weeka, Obsexved .twiy da@ welgYedinitially, weeklyfor 13 weeks,
monthlythereafter;clinical obaemitionsrecordedmontl?l$ faontblyth&eaftqq ctiW4 jabservatlonsrecordedmonthly

Nmopsy perfomed on all animals. The foll*ng organs Necropq p&formed on ail qnimais.
wereweighedat the 9- and #-month interim evaluati&s:
brain,heart, right.kidney,liver, lungs,spleen,right test& and

cliaiall PuthQtagy
etinical pa&ologystudieswere perlormedat’the 9- and None
f S-monthinterim evaluations.
NarumJogyrPackedted cell volume,hemqlobin,
erythroqtes, i-Well-Jolly bodies,meanerytliracytevolume,
meanerythrocytehemoglobin,meanetythroqte hemogIobin
~nW&n, platelets,,reticuloqrtes,and leuko$te,couf\t
and differential, total bone marrmvcelluiarity (15 months
a=&d*dckarriiaoytclplatlne klnaseand lactatedehydrogmase
30 l&Butadiene,NTP TR 434

ExperimWal Design and Materials and Methods in the InWation $Mdi~ of f&S

complete hiitopathologywas performedon all controls,200,
and 625 ppm mice at .theP-monthinterim evalnation;all
con&o& all txposedmales,and 625,200, and 625 ppm micreacopi~ grosslesionsand tiaauemasseswith regional
femalesat .the 15month interim evaluation;ah animalsdying lymph nod& adreiml‘gland, , epiciidymii,esophagus,
early or hilled moribund;and all 2-yearcore stndy mice. nie gallbladder,h&$er+n gland,he+, kidney,large intestine
following tissueswere routinely examinedmicrowopkahy (ceeum,eoloa,rectttm), larynx,hver, lung lymph node
gmsslesionsand tissuemasseswkh regional lymph nodes, (bronchfal,~~~~‘ .mqtdibukr, mesenteric),mammary
adrenalghmd,brain, epididymis,esophagus, gallbladder, gland,nose,ovary,pancreas,pwbththyroid gland,pharynx
harderiangland,heart, kidney,large,intestine(cecum,colan, pituitary gland,p&ate glirnd;s&vary gland,seminalvesicle,
rectum),kynx, liver, lung, lymph node (bronchial, smfitthit&tine (~~~urn~, jejnmtm, ileum), spleen,
mediastinal,mandibular,mesenteric),mammarygland,nose, sternebme{includingrtyrbvvh stomach(foreatomachand
ovary,pancreas,parathyroidglitnd, pharytm,pituitary gland, glanduiar},testis,thymus,thmid gland,and trachea.
prostategland,salivarygland,seminalvekcie, small intestine
(duodenum,jejunum, iieum),spleen,stemebrae(including
marrow),stomach(foreatomachand glandular),testis,thymus,
thyroid gland,trachea,and uterus


exposedto 625 ppm, and were due primarily to the
Survival induction. of fatal neoplasms and their associated
Estimatesof the probabilities of survival for m&e and lesions. Efo female mim exposedta 200 or 625 ppm
female mice exposed to 1,3-butadiene and control or male mice exposedto’625 ppm survived to the end
mice are shown in Table 3 and in the Kaplan-Meier of the studies. Su~Gvalwas decreasedfor males and
survival curvesin Figure 2. *sure-relat&l mortali- female-s to concentrations of 20 ppm or
ties first occurred during week 23, mainly in mice above. The decreasesia survival were dose related.

Survival of 1Mk.ein the &Year I~a~~t~~~ Studies of t,343utarIbne

0 PPt@ 6.25 ppm 20 Ppor 200 Pm 625 ppm

Animals iniliaUy in study 70 70 70 70 70 90
9-Month interim evatuation’ 10 10 10 10 10 10
SMonlh interim ev&uat’mna 10 10 10 PO 10 7
Accidentaldeathsa 0 0 0 0 0 1
Moribund 9 6 IS 15 23 33
Natural deaths 6 5 11 12 23 39
Missinga 0 0 0 1 0 0
Animals surviving until study termination 35 39 24 22 4b 0
Percentprobability of survival at end of study’ 70 78 49 46 8 0
Mean sunrival(days)d 597 611 57s 558. 502 280
Sunival analysise P<O*ool P-0.43off P-O.044 P=O.o22 Pc0.001 PcO.001

Animals initially in study 70 70 70 70 70 90
%Month interim evaluation’ 10 10 10 10 10 8
XS-Monthinterim evaluationa 10 10 10 10 10 2
Accidentaldeathsa 0 0 1 0 1 1
Moribund 10 10 14 31 37 46
Natural deaths 3 7 11 8 12 33
~~~~i~ surviviig until study termination 37 33 24 11 0 0
hwnl probability of sutvivai at end of study 74 66 50 23 0 0
n tunvival (days) 608 597 573 548 441 320

kvival analysis P<O.oQl P=O.SlO P4~013 P<O.Wl P<O.OOl P<O.ool

l lZ4mwcd from sutvival analyses

lrtrtndcaone animal that died during the last week of the study
n-Meier determinations. Survivalrates adjustedfor interim evaluations,accidentaldeaths,and missinganimals.
5 of all de&is (uncensored,censored,terminal sacrifice)
c ult of the life table trend test (Tarone, 1975) is in the control column, and the results of t@ life table painvisecomparisons
ftka, 1972)with the controls are in the dosedcoiumns, A lower mortality in a dwe group is indirated by N.
32 &34Xutadiene, NTP TR 434

0.0 I I
5 I I ,‘+- I I &
0 15 30 b9 60 79 _ 90 105

i . . . . . ..*........ p... . . . . . . . . . . . . . . . . . . ..- )I.*. . ...* i.1 . ..*..*.* . . . . . . . . . . . . . . . . f . . . . 5’

; ..~. . . . . * ,...... g./zg&$...


Kaplan-Meier Survfval Cwrves for Mke Administt%t?d1,~0~~~di~~~by l[shaIakion for 2 Years

Body W e ights and cilin~ ~~~~~~ increases in. the percrsrttageof erythrocytes with
Mean body weights of exposedmale and female m&e Howell-Jo&y .body,W&ms. h~~eas~~in numbers
were similar to those of mntrols th~ug~out the of Howell-JoUy W ies can occur in regenerativeor
study (Tables 4 and 5 and Figure 3). No clinical rneg~~~b~ti~ anemias. Males exposedto 625 ppm
findings other than those associatedwith lesion l,~bu~d~~ne had a s~~~~~nt increase in mean
developmentand moribund&y were observed. platelet value, a finding -that correlated with the
accwmqe of neoplasms. &me marrow cell counts
of the h&o s~~~~g females in the 625 ppm group
At the g-month interim evaltiatiotqs~t~~ti~~~ysignif- were~greaterthan .$hos@ of the controls. No statis-
icant, dose-related decry in mean values for titil differen= wet% Roted in bone marrow cell
ezythrocytecounts, hemoglobin coi&entattin, aad counts titween cuntrol and exposedmales. M icro-
packed red cell volume occurred in male m ice scopic evaluationsof the marrow smearsrevealeda
exposedto 62.5 ppm or above (Table Fl). At the slight &t shifi~fn the qytbroid series of both sexes
625 ppm level, mean erythrqcyre~ volume was Sigxtifi- .to 200 or, ppm. There was also an
cantly increased, altkroughI there was no evident in erytIt+ar&s with some megaloblastic
reticulocytosisor excessive~~c~r~rnasia cc&pared characteristicsand an increasein necrobiotic erythro-
to the control males. The percentageof erytb&& blastsand (9 e~hrob~~~ with a muitilobed nucleus.
with Howell-Jolly body inciusions was also signifi- Two males in the, 625 ppm group had many hyper-
cantly increasedin males exposedto 625 pbm, and segments-~~troFhi~ prmnt, and one containeda
significant leukopenia and‘lymphopeniaoccurred in uniforti ~p~~atio~ of immature mononuclearc&is,
males exposed to 200 or ‘625 ppm. In ~femalles malignant lymphcima,
exposedto 200 or 62.5gpm, eqthrojd parameter (LDH) values were
including erythrocyte counts, hemoglobin1con&n-
tration, and packedred cell v’plymewere s~~~~~~ntly
reduced. Macrocytosis and significant increasesin
the percentage of erythro@tes %vith HoGel!-Jolly
bodiesand mean erythrocyte hemo~~bin, -without a the princip&I &enzyme ‘in ske;letalmuscle and liver,
significaqt increasein reticukqteq or excessivepoly- and the ‘biding of\& m&&ed increase in hepatocel-
chromasia,occurred in females exposedto 200 ar Mar &rosk may be ihe cause of- the increased
625 ppm. Leukopenia and lymphopenia did nor percent LDH-5; howev&, LDH is contained in all
occur in females in the 2 @ and 625 ppm goups, tittsu& in vajing a In no casewas there any
although neutrophils and lymphocyte$wqe lower in evidenceof ~~~~ tterns suggestiveof those
these groups than in the control females. These observed in humans with myocardial infarction or
findings are consistent with a m ild, to crate, rn~c~e~d~, lesions $bqt are primarily associated
poorly regenerative,macrocyticanemia. Theme&a- with ne&as&-landnot ~~~if~ratioa. There was no
nism for the anemia cannot be determined &qrn the ~~e~atia~;~~n the ta‘&l enzymeand isoenzyme
available data, but a m ild’ megaloblz&ic anemia value+on.aa ~~~~d~ basisand the histopathologic
resuliing from ineffective erythropoiesisin the bone lesion di~~~s~‘,
marrow cannot be exclude+‘There were no signifi-
cant differencesor exposur&related trends in total
serum enzyme activity of ra&tate de~~dro~~~~se or
creatine kinase of either the g-month ipterim This section describesthe statistically significant or
evaluation. bialogic&y noteworthy tihaztgesin .the incidencesof
neopiasmsor ao~n~pl~t~c lesions of the hemato-
At the 15month interim evaluation, theke was a poietic system, be&, Stig, stomach, liver, ovary,
statistically significant decqse in mean values for uterus, hard&a gf ’ mar-ygland, preputial
erythrocytecounts, hemoglobin coqq$tratians, and gland, kidney; skirt, gland, small intestine,
pack&~red ceil volumes,and a significant iic&ase in testis, aqd tu)se. Summariesof the incidencesof neo-
the mean erythrocytevolumes in. males and females plasms ‘and ~ona~p~~t~c lesions, the indiedua!
lexposedto 625 ppm (Table F2). M&B and females animal tumar-‘depose, the statistical analysesof
ia the 625 ppm groups had s~tisti~~~y scent primary ~~~~as~ that prred with an incidence
34 l,31Butadieoe, NTP TR 434

Mean Body Weights and Sw&al of Male Mb in the 2-Yew ~~~i~~~~$~udy of 1,3-Butadiene

wtoks tt DDIU 623 DDm 625 warn

A % wt. Na of Av. wit. ‘Wt. (Se’or No. of Av. Wt. Wt, (sb of Na of Av. Wt. Wt. (% ef No. of
szy 0 sluvtvors @ controlc?)‘
; Survtvors (e) c0&0&#) sprvlvom (g) controbi) survtvoro

1 23.0 70 23.1 100 70 22.9 loo 70 22x9 loo 70

2 70 24.2 99 70 24.1 99 70 24.1 99 70
3 2: 70 256 99 70 25.3 98 70 24.9 97 69
27.1 70 26.6 98 70 26.5 98 70 26.3 97 69
5 27.8 70 27.4 99 70 27.7 100 70 27.S 99 69
6 28.7 70 28.5 99 70 286 loo 70 28.3 99 69
7 29.2 70 28.9 99 70 29.2 loo 70 29.1 100 69
8 30.1 70 29s 98 70 30.0 100 70 29.7 99 69
9 30.7 70 29.9 97 70 3.6 1OQ 70 30.1 9s 69
10 31.1 70 30.7 99 70 31.2 loo io 30.9 99 69
11 32.0 70 31.4 98 70 32.3 101 70 32.0 100 69
12 32.7 70 31.7 97 70 32.3 99 70 32.1 98 69
13 33.0 70 31.9 97 70 32.7 99 70 32.2 98 69
14 33.7 70 32.4 96 70 33.4 99 70 32.9 98 69
18 35.2 70 34.7 99 70 36.0 102 70 35.6 101 69
22 36.6 70 36.8 101 70 37,9 70 99 69
26 38.9 70 37.6 97 70 39.5 :z 70, Z 96 69
30 40.8 70 40.0 98 70 42.0 103 70 40.7 100 6s
34 41.7 70 41.0 98 70 429 103 70 41.9 101 68
38 43.2 70 42.s 98 70 44.5 103 70 42.2 98 68
42” 43.3 60 42.8 99 60 44.2 102 Ml‘ 43.8 101 58
46 44.1 60 43.6 99 60 45.1 102 59 45.0 102 58
49 45.5 60 44,3 97 60 45.9 101 59, 45.1 99 5s
54 46.0 60 44.8 97 46‘6 101 59 45.7 99 57
58 45.7 59 4s.i 99 z 46.1 101 58 46.6 102 56
62 46.1 59 45.9 100 a- 46.9 102 58 46.5 101 56
66a 45.7 49 45.4 99 50 46.1 301 48 46.5 102 4s
70 45.8 48 46.4 101 4s 4645.6 102 47 47.0 103 43
74 45.3 46 46.0 102 48 46.5 103 47 46.8 103 43
78 45.1 46 45.4 101 46.5 103 43 46.7 104 41
82 44.8 44 44.9 100 ii 45.5 102 42 46.5 104 3.3
86 44.3 43 45.5 103 47 44.9 101 38 46.4 105 36
90 44.0 41 45.8 104 4s 44.8 102 39 45.2 103 34
94 43.1 39 44.5 103 44 43.7 101 32 43.9 102 31
% 42.9 39 44.6 104 42 43.0 loo 30 43.6 102 29
97 43.0 37 44.6 104 41 41.6 97 41.6 97 29
99 42.3 37 44.0 104 41 41.9 99 2 41.4 9s 28
101 41.6 36 43.4 104 41 42.0 X01 24 40.2 97 2s
103 415 3s 43.7 105 40 41.6 “MO 24 403 97 22

Termhalsac.~ 3s 39 24 22

Meon for weeks

1-13 28.9 28.4 98 28.7 99 28.5 99
14-52 40.3 39.6 98 41.1 102 40.1 100
53-103 44.2 45.0 102 44.6 101 44.7 101


ResUltS 35

W&S 0 DDDI i&OD~m. 62sDUrn

OU Av. Wt. No.of Av. Wt wt. (Skof Na at Av. Wt. wt. (% ot No, at
study (s) sunivors -(g) 4%x&0$) SWVlVO@S (9) controla) survtvora

1 23.0 70 22.5 98 70 22.4 97 90

2 24.4 70 23.9 98 70 23.6 97 90
3 25.8 70 25.2 98 70 25.0 97 90
4 27.1 70 97 70 26.1 96 90
5 27.8 70 it: 98 70 26.8 96 90
6 28.7 70 28.1 98 70 90
7 29.2 70 28.9 99 70 it2 ii 90
8 30‘1 70 29.9 99 70 29.2 97 90
9 30.7 70 30.2 98 70 29.5 % 90
10 31.1 70 30.5 99 70 302 97 90
11 32.0 70 31.6 99’ 70 31.0 97 90
12 32.7 70 32.4 99 70 31.8 97 90
13 33.0 70 33.i 100 70 32.1 97 90
14 33.7 70 33.3 99 70 32.4 % 90
18 35.2 70 35.9 102 70 34;Q 99 89
22 36.6 70 37.9 104 70 37.2 102 89
26 38.9 70 40.1 103 69 39.2 101 85
30 40.8 70 42.2 103 69 41.11 101 78
34 41.7 70 a3.i 103 68 41.5 loo 68
38 43.2 70 45.1 104 67 42.4 98 49
42a 43.3 60 46.4 107 56 42.9 99 29
46 44.1 60 46.3 105 56 43.8 99 18
49 45.5 60 47.2 104 56 44.0 97 16
54 46.0 60 47.2 103 54 45.9 100 13
St? 45.7 59 47.4 104 53 45.7 100 11
62 46.1 59 47.6 103 53 4S.4 99 8
66a 4%7 49 47.5 104 43
70 45.8 48 47.0 103 39
74 45.3 46 46.7 103 37
78 45.1 46 47.2 105 33
82 44.8 44 44.7 100 33
86 44.3 43 43.7 99 23
90 44.0 41 42.2 96 15
94 43.1 39 415’ % 10
96 429 39 408 9s 8
97 43.0 37 93 6
99 423 37 53 91 5
101 41.6 36 39as 95 4
103 41.5 3s 39.0 94 4

TermId sacrifice 35 4 0

Mean tor weeks

1-13 28.9 28.4 98 28.0 !87
14-52 40.3 41.8 1oQ 39.9 99
53-103 44.2 43.8 99

’ interim e-valuationsoccun-edduring weeks41 and 66.

W&?&S QD~UI 6.25 ,M)m 20 t&n 62.5 .Bum
Av. Wt. N0.d : Av. WI. Wt. j$ al No. of Av. Wt. Wt #(of N&of Av. Wt Wt. (% df No.or
Iii0 moors tQj controls) survhtors (Ei) eontre’hp) ‘iJwv-1pors (8) txarols) survivors

18.8 70 18.6. 99 70 .18.6 99 70 18.5 98 70

21.0 70 21.0 loo 69 21.1 ,101 70 20.6 98 70
3 22.4 70 218 97 69 21.9 70 21.8 97 70
4 22.9 70 22.1 97 69 22.3 it! 70 22.6 99 70
5 23.7 70 23.2 98 69 233 70 98 70
6 24.5 70 241 98 69 z 70 iti; 9% 70
7 25.0 70 24.6 98 69 ii; 99 70 97 70
8 25.4 70 3.2 99 69 25.0 98 70 2: 97 70
9 26*3 70 257 98 69 25.8 9% 70 96 70
10 26.9 70 25;8 96 69 25.9 96 70 iii: 95 70
11 27-l 70 26.4 97 69 26.7 99 70 26.4 97 70
12 27.2 70 26.7 98 69 26.5 97 70 26.1 % 70
13 27.9 70 27.0 97 69 27.1 97 70 26.2 94 70
14 28.0 70 27.4 98 69 27.4 98 70 27.0 % 70
18 30.6 70 29.8 97 69 29.9 98 70 28.5 93 70
22 31.6 70 31.4 99 69 31.3 39 70 30.5 97 70
26 32.9 70 322 98 69 32.6 99 70 324 99 70
30 34.4 70 34.7 101 69 35.5 103 69 34.6 101 70
34 36.1 70 35.7 99 69 35.0 97 69 35.9 99 70
37.9 70 37.4 99 69 37.8 1oQ 69 34.9 97 70
2 38.4 60 39% 103 59 39.% I04 S8 39.5 103 60
46 40.2 60 40.0 106 59 40.5 101 58 40.6 101 60
50 41.5 60 41.0 9!3 59 42.5 to2 33 a.9 103 60
54 42.6 60 42.8 101 59 43.6 x0-2 57 44.6 105 60
58 43.4 60 421. 99 59 43.9 201 57 46.2 107 58
62 44.0 59 45.1 103 59 46.1 105 57 47.9 109 56
66* 45.4 49 46.1 102 49 47.7 go.5 47 48.7 107 44
70 46.7 49 47.4 102 48 49.0 PO5 47 50.1 107 44
74 48.1 48 48.1 lob 48 50.0 tm 4s 51.5 107 42
78 48.x 48 43.6 101 ‘45 50,7 iOS 43 50-4 105 39
82 46.6 46 48.5 104 45 so.3 l&J 42 ‘a7 105 37
86 47.4 46 49.0 303 44 48.7 103 40 47.7 101 32
90 46.9 4s 48.7 104 44 48.6 104 38 45.7 97 29
94 46.0 43 46.6 101 41 48‘6 106 34 42.5 92 23
96 46.1 42 46.3 loo 40 49.3 107 31 43.6 9.5 18
98 44.8 41 45.7 902 39 48.0 107 29 43.7 9% 17
100 44.8 41 46.0 103 39 47.9 107 2s 445 99 16
102 44.1 40 455 103 37 4$.Q 104 24 42.8 97 13
104 44.2 37 45.6 103 37 46>4 105 24 43.1 98 13

Teminatsacrilicc 37 33 24 11

Mean for weeks

l-13 24.5 24.1 98 23.8 97
14-52 35.2 35.2 loo 34.9 99
53-m 45.6 47.8 IQ5 46.4 102


Mean Body Weights and Survivsl Oc @km&e the ~-VW IaW@on Study ol l,t,lButadbne


AK wt. No.of Av.WC wt (% ir No.or Av. WL wt. (%d No.of
szy @ SWVkO~ 0 ConWok) SWdVOrO (I# controk) SUWiVOCS

1 18.8 70 18.2 97 70 18.3 97 90

2 21.0 70 al.9 la0 &9 20.6. 98 90
3 22.4 70 22.3 100 69 21.7 97 89
4 22.9 70 22.8 100 69 224 98 89
5 23.7 70 23.4 99 69 23.2 98 89
6 245 70 24.3 99 69 24.2 99 89
7 25.0 70 24.7 99 69 24.7 99 89
8 25.4 70 25.7 101 69 ~ 25.4 loo 89
9 26.3 70 26.1 99 69 25.6 97 89
10 26.9 70 26.1 97 69 25.8 % 89
I1 27.1 70 26.6 98 69 98 89
12 27.2 70 26.7 98 69 z; 97 89
13 27.9 70 27.2 98 69 26.9 % 89
14 28.0 70 27.8 99 69 27.4 98 89
18 30.6 70 29.3 96 69 % 89
31.6 70 30.8 98 69 iz 100 89
iti 32.9 70 327 99 69 33.2 101 89
30 34.4 70 35.1 102 69 35.0 102 86
34 36.1 70 37.3 103 36.2 100 81
38 37.9 70 39.7 1@5 ,i!i 37.6 99 75
42 38.4 60 42.4 110 57 39.8 104 56
46 40.2 60 43.2 108 57 40.4 101 45
50 41.5 60 45.1 109 54 41.8 101 32
54 426 60 46.0 108 51 41.4 97 22
58 43.4 60 47.1 109 48 41.7 96 11
62 44.0 59 49.3 112 45 43.9 100 5
66a 45.4 49 49.2 log 31
70 46.7 49 48.8 105 23
74 48.1 48 51.1 106 16
78 48.1 48 50.8 106 13
82 46.6 46 45.s 98 12
86 47.4 46 47.4 100 7
90 46.9 45 53.3 114 4
94 46.0 43 46.6 101 3
% 42 48.6 185 2
98 2 41 495 111 2
loo 44.8 41 46.6 104 2
102 44.1 40
IO4 44.2 37

37 0 0

Mean for weeks

1.13 24.5 242 99 24.0 98
14.52 35.2 36.3 103 35.3 100
a-i04 45.6 425 93

* Lterim evaluations occurred during weeks 41 and 66.


y CONmOl.
.............. j ................... . ............ * . ..... j*. ................. . ................... i .. 0 6,26PPM
I t A 204'W
R 62sIrw
20 ................... . ............... *...‘......... ............................ ..!. ............... ...? ..
0 io5Pw
: 8 626PFU
1 .

a 6.2SPPu ,.*..
lj 62.5PPM
0 02swN
Results 39

of at least 5% in at least one animal group, and reticulum .cell burns type A, or malignant
historical incidences for the biologi&ly significant lymphoma, histio$tio but currently are believed
neoplasmsmentioned in this section are presented in to be derived from lymphoid, mononuclear
Appendix A for male mice and Append&r B for phagoeytic cells. ~~tol~g~ly, the sarcoma cells
female mice. were large and ~on~rno~p~~ with dark basophilic
nuclei and relatively abundant eosinophilic cytoplasm,
HemutopozM’c System: At the 9.month interim evalu- ~ul~~uci~t~ giant. &i@were sometimes seen.
ation, the absolute and relative thymus
females receiving 625 ppm were lower *t Bone marrow atrophy w@ ~observedin both sexes
controls (Table El). The relative thymus weight of only at the ltighest exposure concentration, 625 ppm
200 ppm femaleswas also lower than that of controls. (Tables A5 and IS). ,Atr~~~y varied in severity from
In addition, the relative spleen weight of females that mild depletion- u~‘~ernato~~etic cells in mice evalu-
received 62.5 ppm or more and the absolute spleen ated at 9 ~oaths to marked depletion in mice that
weight of females that received 200 ppm were lower died or were killed before 15 months. Incidences of
than those of- controls. At the 15-month interim bone marrow hyperplasia-in female mice exposed to
evaluation, the absolute and relative spleen 62.5, 200; or 6s ppm were greater than the inci-
of males and females receiving 625 ppm were dences in tire control group (Table B5). The hyper-
than those of controls (Table E2). The absolute plasia was primarily an increase in granuiocyte
spleen weight of the 200 ppm females w& also precursor cells, a response that correlated with the
greater than that of the controls. presence of, large neoplasms in the skin, mammary
gland,. or other ,oFgans,..Increased severity or inci-
Exposure of mice to l+butadiene was associated dence of hematopoiesis in the spieen, liver, and lung
with the development of malignant lymphoma,and to was also 0bserve4~in femal& at the three highest
a lesser extent, histiocytic sarcoma (‘Table 6) The exposure con&zttrations. The hematopoiesis in these
incidence of malignant lymphomas, particularly the organs paralleled~the finding of bone marrow hyper-
tymphocytic type, was significantly greater in males plasia.
exposedto 625 ppm and in females exposedto 200 or
625 ppm than in the controls. The At the g-month interim evaluations, absolute thymus
tymphomas occurred as,early as week 23 weights were decreasedin, males and females exposed
before the 15.month interim evaluations. to 625 ppm (Table~E.1);.the decrease in females was
tymphomas of the mixed and un~fferenti~~~ cell statistically~ sig~~~~~, IThese decreased weights
types occurred at incidences more typical of the correlated with the presence of thy&c necrosis
xpontaneous lesions seen in BbCJF, mice. The (atrophy). The incidence of;thymic necrosis was also
tncidence of histiocytic sarcoma was signiffcantly. increased in female .mice exposed to 62.5, 200, or
greater in males and females exposed to 2Q@ppm 625 ppm.
tbrzn,in the controls. MoreoveFSthe incider&s in
m& mice exposed to 20, 62.5, or 625 ppm and in Hear%- The absolute heart weights of females
l~mates exposed to 625 ppm were marginally higher receiving 625 ppm were greater than those of the
~httn in the controls. controls at both the 9- and Smonth interim
evaluations (Tables El and E2). The absolute heart
~tthough many organs, particularly the spleen, &mph weight of 200 ppm ferns@ was also greater than that
tivcr, lung, and. kidney, were aff=ted in mice of controls at the l~-mo~~h.~~t~rim evaluation.
mphocytic lymphoma, the thymus was involved
1n test mice, and in some, the thymus was the pre- The incidences ef hem~~~r~rna of the heart in
inant organ affected. The lymphocytic l$tnpho- males exposed Ilrf?inor greater and in males
consisted of uniform populations of small- to and. females to 200 or 625 ppm were
~~~~~~-~i~ lymphocytes, whereas t&e mixed and sign~~~tly higher than .in the. controls (Table 7).
krentiated lymphomas generally consisted of Endothelial h~r~~ia w&-also observed in exposed
hetcrogenous populations of lymphocytes with mice. The ~rdi~~‘herna~gio~r~m~ occurred in all
lfutrir pleamorphism and atypia. The histiocytic ventricular locations, but~were more frequent in the
F~~YY~~S have previously been referred to as left ventricular wall. They were occasionally
40 &34Wadieoe, NTP TR 434

Malignant Lymphma. and Hi~tiocytk §a In Mice in t&e Z-Year

@Pm 6.2sPPm MPpotl 6=Ppm 200 PPm 625 ppm

9.Month f~twim Evaluation
LymphocyticW@nant Lymphoma on0 (0%) 000 (0%) on0 (0%) WI0 (0%) oflo (0%) ulo(10%)
lS-Mouth Interim Evuluath
Lymphwytic Malignant Lymphoma oflo (0%) o/lo (0%) on0 (o%} .OflQ(0%) oflo (0%)

Z-Year study
mwy$fwP- LYwb-
x50 (4%) OfSO(0%) wo (4%) ~~~%) ;zMo (4%) 49l73 (67%)
Adjwed lateb 4.7% 4.2% 4.0% 95.1%
Terminal cad OB5 (yxJ> ~~(o?+) Q@l (fm ~y$@w 014 (0%) 010
First incidence(days) 511 456 In 161
Life table testd P c0.001 P=O$27N P=O,671 P=O.?+53 P=O.S29 P ~0.001

z z- (Mixed
orU~5~~4~)~~ Tnn)
2150 (4%) 2?so (4%) ofso (0%) Of50(0%) 13 (1%)
Mm Lymphomp (Hia(locytk; LyInp&ka@h% MI*ed, NOS, wum6m*M ccuTypc)f
overall rate 4/50 (8%) z/so (4%) s/50 (8%) 6fso (12%) 2&o {4%) 51173 (70%)
Adjusted rate 9.ft% 5.1% 1223% 17.7% 4.0% 95.4%
Terminal rate 1s (3%) 2439VW w (8W mm (5%) Of4 (osa) O/o
Fiit incidence (days) 511 733 (T) 456 200 171 161
Life table test P<O,OOl P=O.3@2N P=O.S28 P-o.a? P=O.627 P<O.oOl

Hlsnocytic Gsreoplsg
overall rats ofso (0%) of50 (W $g%) s/so (1V) 7150 (14%) w3 (5%)
Adjusted rate tp% 14.3% 31.9% 10.8%
Terminal rate ~~(0%) o/Js Pw ;i (@w la? (5%) 014 (0%) O/o
First incidence(days) 398 288 120
Life table test P<O”oo1 - P=O.OSl P=O.o21 P<O.OOl P*O.O43

WS?SUltS 41

Maliit Lymphoma and Hkbcytk Sarcorkiain M ice &I ~~,2-Y~~,~uha~~~ Stud&s of 1,3-Butadiene

OPpat 6.25ppm BPpot 62sppin 200PPm 625 PP~

Y-Month Interim Evaluation

I .ymphocyticMalignant Lymphoma wlo (U%) um (0%) oil0 (0%) on0 (0%) o/lo (w6) l/8 (13%)
15Montt~ Interim Evaluation
I .ymphoqtic Malignant Lymphorpa 1110(10%) o/lo (0%) o/lo p%) o/lo (0%) l/l0 (10%) on (@w
Z-Year Study
I.ymphocyUc Ma-t Lymphoma
Overall rate l/50 (2%) W ? W56) 6150(52%) 3/S0(6%) S/J8 (16%) 31180(39%)
Adjusted rate 22% 8.5% 19%6% 13.5% 37&% 70.5%
Terminal rate o/37 (0%) 2Jj3 (6%) 3m (13%) ml (9%) 010 010
First incidence(days) 623 69s 533 407 245 203
Life table test P~O.OOl P=O.278 P=O.O26 P=O.l6u P<0.001 P<O.OOl
Mdgnanl Lymphma (Mbred or Undi@erenliekd C& ‘J$pe)
Overall rate s/so (10%) 9-150
(is%) 5150(10%) 3iMo(6%) mo (2%) @8uww
M+pant Lppkima (Lymphocytk, Mixed, NW, or Und#errnlfal.ed CelIType)k
overall rate 6/50 (12%) a/so (24%) ll/SO (22%) 760 (14%) 9/so (18%) 3z/Sfl ww
Adjustedrate 14.6% 34.0% 38.7% 39*7% 70.8%
Terminal rate 3f3i@ % ) lon3(39%) 8/24(33%-y ~~~%) o/o W
First incidence(days) 623 695 533 2u5 189
I Xe table teat P<WOl P=&*O6g P-O.029 ELns P<0.#1 P<O.o01
rilsIlocyuc 8areom18i
Civerail rate 3/50 (6%) 2/50 (4%) ?/50 (14%) 4150@%) 7150(14%) 4/M (5%)
Adjustedrate 6.9% 4.5% 2&O% 17.7% 28.1% 10.3%
‘krminal rate O/37(0%) on3 (W m-4 (4%) oni (0sj O/B
1:int incjdence(days) 485 467 492 524 z 300
I .ife table test P‘cO.901 P=Q.SlSN P=O.w7 P=&l95 P=+MO2 pro.038

I ‘I’)‘l’cnninafsacrifice
b Incidencea are givenas numberof neoplasm-bearing anhqlslnumbq of animalsn~pskd.
fficplan-Meierestimatedneoplasmincidenceat the end of the study after adjustmentfor tntemurtentmortality
’ ( )bsetvedincidenceat terminal kill
” Hcneaththe control incidenceare the P valuesassoctatedwith the trend test. Beneaththe douedgroup incidencesate the P values
wrresponding.topairwisecomparisonsbetweenthe controlsand-that dosed,group. The hfe table analysisregardsneoplasmsin
animalsdying prior to terminal kill as being(directly or indirectly) the causeof death. A lower ktcideace in a dosegroup is
indicatedby N.
’ Not appkabtq no neopiasmsin animal group,
’ L‘Year historicalincidencefor untreatedcontrol groupsin, hiW inhalationstudies(mean $Zstandarddeviation): 46/S7S
(W% r 4.5%); range2%-16%
s L-Yearhistoricalincidence: S/S75(0.9% * 1,4%); rangeu % 4 %
Ir 2 Year historicalincidence: 153/561:(27.3% i 15.1%);range8%44%
’ L.Ycar historical incidence: lo/S61 (1.3% -C 3.3%> range0%-S%
42 X,3-Butadieae, NTP TR 434


9-Month Interim Evaluation

Endothelial Hyperplasii o/la (0%) ,b on0 (0%) l/-l0 (10%)

U-Month I&erim Rvaivatb

Endothelial Hyperplasia o/lo (0%) o/l (0%) o/m (0%) a/la (b%) l/l0 (la%) ‘w7G-J)
Hemangiosarccma o/-lo(0%) 00 (0%) o/lo (0%) wlo (a%) mo (10%) 3r7 (43%)

f-Year Study
Endo#enaI Hspcrplasip
Overall rate o/so(0%) l/49 (2%) o/so(0%) 2448(4%) 4t48(8%) s/73 (7%)
Logiitic regressiontest’ P &.OOl P=O.516 -d P-O.143 P=O.o6S P=&O2S

“Gwe o/so (a%) o/49 (0%) ma (2%) s/48 (10%) W48 (42%) 4R3 (5%)
Adjusted ratef 0.0% 0.0% 3.4% 19.4% 933% 44.6%
Terminal rateg om (0%) OF38(0%) O/w(G%) 3t22 (14%) 3/4 (75%) o/o
Flnt incidence(days) 682 649 519 289
Life table test P CO.001 P=O.4Sl P=O.Ol1 P<O.Ool P40.001


9.Month Interim Evaluation

Endothelial Hyperplasia o/lo (0%) -b on0 (0%) l/8 (13%)

15.Month Iuterim Evaluation

Endothelial Hyperptasia on0 (0%) pm (0%) 4/lo (40%). o/2 (0%)
Hemangicaarcmna o/lo (0%) ,wo fyq I/l0 (10%) m (lOO%)*

2.Year Study
Eadotbtllal Hyprplneia
Oveeallrate O/SO(0%) 2lso (4%) l/SO (2%) 4/49 (8%) 5150(10%) 8/m (10%)
Logistic regressiontest P=iMMS P=O,26l P=OS21 P-O.148 P=O.1@4 P=O.o07

Oveeallrate o/so (0%) o/so (0%) o/so (a%) 1149(2%) 21/SO(42%) 23BO (29%)
Adjusted rate 0.0% 0.0% 0.0% 4.0% 100.0% 100.0%
Terminal rate am (0%) o/33 (rn) MzJ+em on1 p?q O/o OhI
First incidence(days) 649 343 307
Life table test P~O,Oo1 P=O.392 P~0.001 PCO.001

* Significantlydifferent (P<Q.OS),from the Contfotgrcrupby the Fisher ewct teat

a Incidenceaare given as number‘of lesion-bearing,animaIs/numberof animaismicrosccpicalfyexamined.
b Not examined
c Beneaththe control incidenceare the P valuesassociatedwith the trend teat. Beneat&the dosr;dgroup incidenceare the P vajum
correapondiigto painvisecomparisonsbetweenthe.ccntrols and that dosedgroup. The life.tabie anaiysii regardsneopk+,wsin
animalsdying prior to terminal ,till as being (direct@’
br indirectly) the causeof death.
d Not appiiibie; no neoptasmsin aaima~group
e 2-Year hk&rical incidencefor untreatedCbltrot gr6ups in NTP inhalation studies(mean + standarddeviation): O/S73(0.0%);
upper 95% confidencelimit = 0.5%
f &plan-Meier estimatedneoplasm~incidence at the end of the sttudyafter adjustmentfor intercurrent mortality
g O&twd incidenceat terminal ,kill
h Z-Year &tori& incidence: O/S58(0.0%): upper 94% confidencelimit = O.S%
Results 43

multifocal and frequently coexisted with foci of collcretiollsraking i ml myofibers. The

endothelial hyperplasia distant and separateBorn the pathogeriesisof this lesion was undetermined,
main neoplasm. A few hetnahgiosarconutsinvaded
through the epicardium and form& grossly visible, I&n& At the 15~~0~~ interim evaluation, the
dark neoplastic growths. Typical h~~~~~~~~ absoluteand relative lung weights of males receiving
had solid foci of anaplastic,pkomorphic spindie cells 625 ppm and the absolute lung weight of females
at the center of the masswith a loose a~ng~~t b receiving 200 pPm were greater than those of the
the periphery (Plate 1). Mitotic figures were controls (Table E2).
common. Variably sired clefts and spaces,frequently
containing erythrocytes,lined by neoplastic ccIls with ure of mice to l$butadiene was .associated
large vesicular nuclei were scattered thfougbrout th& ~~~ of focal hyperplasia and
masses(Plate 2). The endorhelial ~~tio~-~p with . Focal alveolar
the myofihers was most apparent at the petiphery of with a significant
the neoplasms, where the interfiber spaces were ce 1n malt?&exposedto
widened by capillary spacesllled by neoplastic cells. 62.5,2tJO,.or625 ppm w&r ~~~~~ntly higher than in
The foci of endothclial hyperplasiawere characteri@ the ~ntroIs by pairwise comparisons. Although the
by variable ,separation,of m$ibem, provably by incidence df ~~~l~r~~~c~o~~ adenoma in males
distended capillaries, and increased~Ilu~~~ of the Iy in the group exposed
capillary endothelium (Plates 3 and 4). En~thel~l incidences of alveolar/
hyperplasia was considered a :preneoplasticd&ion by or caftinoma were
the Pathology Working Group, and some members e logistic regre&un and
considered them to represent incipient ~e~n~o- ttxpowi to 20,625, or
sarcomas. 200 ppm. In female nr&c, the incidence of focal greeter than that of the
When hemangiosarcomas occurred in multiple cantrql group in the 625 ppm group only, while the
organs,the cardiac neoplasmswere usually d&gnated incidences of a~~~~~~~~o~r adenoma and of
as primary, because the incidence of he~~g~~ aden~~n~~ or carcim~ in all exposuregroups
sarca~ was highrestin the heart, the earliest lesions were signi&csntIy higher than in the control group.
occurred in the heart, neoplastic emboii were seen in
some animals, and liver and lung lesions were:&ome* Iasia was considered the
times multifocaL It couid not he determined defini- adenoma and
tlvely if the hemangiosarcom+sin the other organs f a focal increase
were me&stases or primary neoplasms. T&e sub- EU~Iwith retention
cutaneous, splenic, and hepatic ~~rnan~~,~~
that were present at low incidencesin mice without
cardiac hemangiosarcomasI may reflect the
spontaneous subcutaneousvascular neoplasms that lined by relativ-ly uniform cuboidal
occur ‘in B6C3F, nrice. The apparent low in~den~ or wlmmlar cells, T&e alveoiaribronchiolar
(unadjusted for survival) of cardiac hemangio- carcitiomas,yvere sir&&~ but consisted of hetero-
sarcomasin male miceexposed to 625 ppm may be with varying degrees of
explained by the high rate of ~elop~~~t of and atypia (Plate 5). The
lymphocytic lymphoma early fn the study. Presum- larger, highly aitaplastic
ably, the lymphomas caused deaths before cardiac neoplasms,often ~~~~~g areas of hemorrhage or
hemangiosarcomasdeveloped. llectosIs.

Myocardial mineralization was also observe4l fre- .%r~& Focal hyena of the forestomach
quently in mice exposed to 625 ppm, but was not epithelium. occurred ~~~~~~ in males exposedto
observed in the controls (Tables A8 and BS).. The 625 ppm, and the incirfence of squamous cell
mineralization consistedof scatuxed,small basophilic papisluma in the xw) group was significantly
&Sh&adieae, NTP TR 434

9-Mouth Interim Evalltatiou
Alveolar Epitheliai Hypeqbsii on0 (0%) ~cw4
AIwdar/bmn~Ade~ on0 @%) lfl (uIo%)
Alveolar/bl carciaonla on (Qm
AJveolar~mncbiokr Adenoma
or cardnoma l/IO (10%) 1n (m%)

1SMontb lnterim Evoluat$on

Alveolar EpitheW Hypep&sia on0 (0%) m (w@-q
AkeoWbmnchioiar Meaorna ‘OflO (0%) on (0%)
Alvcolarbmnchll Adenocarcinoma
or Carcinoma on0 p%) on (046)
AlveolaribmncbWar Adenoma,
Menocarcinoma, or Carcinoxw on0 (0%) on WQ

t-Year Study
ovaall rate :a50 (4%) 9&O (18%) 660 (32%) X3/49 (27%) l?SI (34%) IW73 (16%)
Logistic regrrasiora tcptb P<O.oOl P=BB40 PQ=O.o99 P<O.oOl P<Ornl P=O.w4

Ahw~maeblolor Meaomo
ovetall rate 18150 (36%) aom ly%) lw5o (oxt%) 25149 (53%) Z~rss (42%) 3/73 (4%)
Mjaatal rat& 46.9% 47.3% 28.2% 74.2% ~~.~ 59.4%
Terminal rated 15135(43%) 17139(44%) 3t.a (13%). 1@2(&396) 4/4(ltwq t-Mb
Fmt incidence (days) 587 517 434 351 251
Lagiatlc regmsaioa tart ;~0.200 PAM7 P~O,O8ON P&MX36 P4+&061 P -0.492

Ahwolor/broac~Ade -a¶eorculclnolap
Overall mte s/50 (10%) 6fso (12%)
Adjusted rate 14.3% lS.4%
Terminal rate x35 (14%) 6f.39 {X5%)
First incidence (days) =m 7sm
Life table test PCO.001 hO.577
Logistic regrwioa test P<O&Oi P-0377

Asvco~~MoiiuklaoInml.#aecare iaoma, orcktrdne~e

Ovemll rate 2160 (42%) 2300 (46%) 19/so (38%) 3l/49 (63%) .3&&O (70%] 3/f5 (4%)
Adjaated rate 34.9% 542% s3.696 87.9% lCI&@% 59.4%
Tetmiaal rate r8m (51%) 2-o/39(51%) 9114 (38%) l&z2 (82%) 4f4 (1
F&at incidence (days) sn 587 517 434 351 251
Life table teat PCO.001 P=OS2N P-0276 P *tI.QkX Pa.003 P<O.OOl
Lagiatll~teat P+O.OM P-=0.5 P=@J5sN P=o;oos P~~.~~ P=O.42z
9-Month Interian Evaiuatiom
Alveolar Epitbelial Hype&a& on0 (0%) -f o/Lo(0%) VT0 (10%) 1/a (13%)
AiveolarMwxhioiar Adenoma OflO - on0 (@%I 2m (2oas) l/8 (13%)
15-Month Intetim Evahatlon
Ah&r Epitbelial Hyperpfasia on0 (0%) on0 (0%) s/lo (so%)* u? (50%)
Alveoiar/bconchiolarAdenonm w w+) Qflo WJJ) z+WWV IL! (50%)
Akofar/bwwhiofar Carcinoma on0 p%) on0 (0%) 1MO(0%) of2(0%)
Alwoiar/bronchiofar Adenoma
or Carcinoma on0(osa) on0(0%) 3no l?w m (50%)
t-Year Sbdy
Axvdar Ezpil&upr Hyporplosia
Overaif rate S/50(Xc%) s/so (10%) 3fSO(6%) F/S@(lg%) l-US0 (22%) Itl/78 (14%)
Logistic regressionlest P=0.092 PGO.558 P=O.SlON P=O.IS3 PaO.162 P ao.037
Alveolarfbronchiolor Adonoma
Overall rate 450 (8%) 1150 (22%) 12I50 (24%) i7HQ (34%) 14/SO(28%) 17rI8 (22%)
Adjusted rate lO.S% 30.9% 40.7% 64.a% MOB% 100.O%
. Terminal rate 3l37 (8%) 9;t3j (27%) sfa c=w Mtl(36%) OEO ON
Fust incidence(days) 714 524 492 443 408 275
Logistic regressiontest P50.002 P*o.q9 P=0.013 P<0.001 P=0.002 P=O.OlO
AhoIarjbronchio&r Me- ma or carclonma
Overall rate Of50(0%) SfSO(10%) ll/S0 (2296) 9.80 (18%) 19/so (38%) 8n8 (10%)
Adjusted rate 0.0% 13.3% 429% 40.8% 1000% 100.0%
Terminal rate 3133Pw lOf24 (42%) 3/x1(23%) O/Q
Fust incidence(days) Yf7 tosa) 539 so7 532 iti3 307
Life table teat PC0.W~ P =0.029 P<O.#l PCO.OO1 P<O.OOl PuO.O01
Logistic regrtxsion teat P==0.001 P=O.Oti P~O.oal F4.002 PK0.001 P=O.O12

Ahno~roacMelprMenoma?Adenocare Irromp, orepd*h

OvemUrate 4fSO(8%) x%50 (UVSQ) 19150(38%) 2450 ($3%) 2950(50%) 22/78(28%)
Adjusted rate 10.5% 3&S% 63.7% 785% 1oao% 100.096
Terminal rate 3L37(8%) w33 (33%) 14/D (93%) 6n1; {SS%) Of0 cm
Fimt incidence (days) 714 519 492 443 373 275
Life tapfe test PC0001 P==O.OW Pc0,001 P<O.001 P<O.cml P<O00l
Logistic lE?gmsbn teat P<O.O01 p‘dMos PC0001 P~Wl?l PeO.OOl P<O.001

* Signifkantly different (PcO.05) from the contrd group by tbe Fiiher exact test
.; ** P<O.Ol
t (T)Terminai saaifice
c l lncidencesare given as number ot iesion/bea~ anhalsfnumber of animal, t&roseopiaby examined.
Fteneaththe control incideuce are the P valulasass&& with the trend.tem. qincidmceare tbePvabtes
correspondingto p&wise compar$ons betweenthe controls and that daseo life table ~uaiysiaregardsneopfasmsin
animalsdying prior to terminal bilEas being (d&&y or lndimctly) tbe causeof death; The tic regres&n test regardstbeae
lesionsas nonfatal. A kwer incidencein a dose groupis indicatedby N,
“, fCap&n-Meier&mated neoplasmincidenceat the em$f tbe study after adjusrment for iotas mortality
Observedincidenceat terminal bill
’ ‘L-Yearhistorical incidence for untreated coned groupsin NTP inhalation studies (mean f standarddeviation): 118/573
(20.6% f 7.6%); rsnge 10%-30%
f Not examined
g Not applicable;no neopfasmsin animal group
h ‘t-Year historical incideacez47/S60(8.4% f 3.3%); rangeO%-12%
46 I,HWadiew, NTP TR 434

higher than that of the controf group (Table 9). i~phom~‘ ~e~a~~llu~r adenomasand carcinomas
Squamousc&Z carcinomasoccurred in twlo males in are common ~~,pl~~ in B6aFI mice, occurring in
the 625 ppm group and in one mate in the 200 ppm 1% of 572 control ma (34.3%) and 87 of 558
group. In female mice, the inqidencesof hyprsrplasia, control bmaiea (15.6 the NTP historical
squamous c&i papilloma, and squamous eel1 carci- database(Tables A4e
noma were significantly higher in the 625 ppm group
than in the control group. The iower,rates of these Basophiiic fti, were characterizedby cells with base-
forestomach lesions in males than in females in the phi& cytophtsm, while eosiuophilic foci were com-
625 ppm~groupare likely related to the lower survival posed of c&r with cyt@asni that stained more
of the males. Squamouscell papillomas and carcino- intensely with eosin than did normal hepatocytes.
mas of the forestomach are relatively uncommon in ErliKedcell foci cunssistet!of tiixtures of cells with
EWBF, mice, occurring with a combined incidenceof eosinophilic copes and cells with clear cytoplasm.
4 of 575 historical control males (0.7%) and ,9 of 561Hepatocellular a~~~rn~ were discrete, expansile
historical control females (1.6%) (Tables /+4d and massesthat were larger than hepatic lobules and
B4Q compressedthe adjacent parenchyma. Hepatic plates
within the adenomaswere not organized in a normat
Forestomach epithelial hyperplasia was typically a lobular pattern and oftea intersected at near-right
focal lesion consisting of thickened epithelium angles with pl&s in the adjacent normal liver;,
forming blunt rugose folds of varying length. Shallow Hepatoceiluiar carcinomas were larger than the
ulcers were sometimes present in the center of the adenoinas and ~~~t~ of markedly disorganized
hyperpiastic lesion with submucosai infiltrates of hepatocytes that formed soiid clusters, glafiduiar
inflammatory cells. The papillomas were generally structures, or’ broad trabecttiae several ceil layers
more complex, with a short stalk and branching thick (Plate 8). ~~p~~ic hepatocytes genera
papillae cot&sting of well-differentiated stt&fied showed rn~era~e to marked pieomorphism and
squamousepitheiium overlying a fibrovascuhr stroma atypia.
(Plate 6). The squamous cell carcinomas exhibited
invasion of the forestomach mucosa by cords and The incidence of liver necrosis was increasedat the
clusters of anaplastic c&is (Plate 7). higher dose 1~~~ in le and female mice
(Tables A;5 and B5). This l&onconsisted of patchy
Liver: The absolute liver weight of females rece&ing areas of nervy that had no particular lobular
62.5 ppm or more was greater than that of the distribution, and occurred frequently in animals with
controls at the g-month interim evaluation malignant ~~~h~~ or h~rna~g~~~~. Centri-
(Table El). At the l%month interim evafuation, the lobular hepat~~~uiar ne.&osis also occurred with
absolute and relative iiver weights of m .@s exposed incrcascd i~~de~~ in ~~~~~~ male and female
to 625 ppm and the absolute liver weight of females mice (Tab&s A5 and BS], and occurred frequently in
exposedto 2olBppm were greater than those of the animals describea as anemic or with atria1 thrambi
controls (Table F2).
iAmy.- A variqy of ovarian neoplasms were ob-
The incidenceof hepatoproiiferative Iesionsincluding se~ai, but only ina~igu&t and I>enigngRWioSa c&i
hepatocelhdar adenomas and carcinomas was tumors were,,de~n~tive~y@ated to exposure. In-
significantly increasedin female mice exposedto 62.5 creased ~~~d~~ of granu&a cell tumors were
or 200 ppm 1,3-butadiene(Table 10). Hepa~~~~~ar observe& iu females to 62.5 or 200 ppm
foci including basophilic, clear cell, mixed c&l, or (Table II). Granulosa c&f rumors varied from small
eosinophilic foci occurred with increased incidencea benign apex with granuiosa ceils aligned on a
in female mice exposedto 2CJ62.5,or 200 ppm. The scant s$romain a discreteiy packeted tubular pattern
incident of hepatoceiluiar neoplasmswere sigulfi= to large Gysticneoptasms with thick trabecuiae and
cantly increasedin male mice exposedto 200 ppm’for spaces fiWd with blood or clear fluid (Plate 9).
2 years. The low incidences of: liver neop&sms in Malignant and benign ~~~~~ cell tumors each
male and female mice exposedto 625 ppm prob;lbly have an NTP historicai control incidence of 1 of 548
reflect the occurrenceof early deaths from malignant (0.2%) (Table B4f).

P-Mod Interim Evaluation

Epithdial Hypaplaria on0(0%) ,b 7no (7o%y
squamoll8 CkIi Papilloma oflo(0%) WV (10%)

lS&ionth lfotecim Evnhation

Epithdial Hypmplaaia ona(0%) m (43%)
!Squamom CW Papillama on0(0%) l/7 (14%)
squaxnowceiiclarcinolna on0m@ ~rnl
Squamous CM Papiilom
or !Squamour Cell Carcinoma o/lo(0%) 3/F (43%)

2-Year stuily
Ezpitholi8l li~l8etll
overall rate 4isQ
@%) wo (6%) am (56%)
Logistic regrasion ted PCO.001 P=OS79N P=V.o89

sq8u8mone cm P8pmoaul
ovetall rate l/50 (2%) o/50 (0%) 7/So (14%) mz (3%)
Adjusted rated 23% 51.7% 4o.o%J
Terminal rat8 o/Js (@m ij!&o%) 114 (25%) OP
Fit in- (daya) 652 530 39S
L.ogwc l%zgmkn test P<O.o@l Pd.535N P-O.012 PmO.446

overal rate oL+Q@w l/SO (2%) 373473w
MjnaY3 aate O*V% 5.9%
Terminal rate o/35. (0%) 014 (0%) OP
Fiit ie (day) 620 302
Life table teat P<O.OVl P=O.325 P=O.OlS
Lmgiatic l.cgE&n lest P=@.184 P=@O.S03 P=O.675

overall rate US0 (2%) om @+q ~W m (5-W
Adjusted rate 2.5% 51.8%
T- rate 0/3s(o%) ~~(~) l/4 (25%) o/Q
Firstincidawr(days) 652 - 530 302
Life tabk tm P<@Iol P=O&UN P<O.OOl P<O.Ool
Lqistic regaskm test P<O*oOl P=O.S3SN P=O.O06 P-O.199

48 I,$Butadke, NTP TR 434


9.Month MerJm Evaluation

Epitheiiil Hyperpbb on0 (0%) ,b
1sMonth Interim Evaltt8tiotl
Epithelial Hypwpbia on0(0%) wx(f=w on0p%)
sqn8mous cell Papiuoma on0(o%] OflO’
squamous cell cardwma on0(0%) 2 ~~~ on0(0%)
squamott8cell Papuloma
or SquamousCXl Carcinoma omfo%) finww
t-Year study
EpnbeI.iaI Hyperpbda
OvernUrate 5/411(!0%) 4l47p%)
Logistic regresb test P mo.477 P=O.sS
squ8m8ua,cetl P8piuolm
oven311rate o/50
(0%) o/50 (0%) 260 (4%)
Adjusted rate 0.0% 0.0% 8.3%
- (@% a33 (0%) 2124(8%)
Terminal rate
First incidcwe (days) mm
L.qistic rege&x~ test PCO.UOl P=O.t49

overau rate o/50(0%) olw(o%$ l/50 (2%)
Adjusted note ‘0.0% 0.0% 4.2%
Temtinal rate olx (a) om (0%) v24 C4%1
Fits incidence (daya) 7330
Life table test P<O,ool P=0.414
Logistic regrrssioa tests ‘P=0.009 P-=0.414

Ovetxdlrate otio (0%) o/50 (0%) 3m (6%1
Adjusted rate 0.0% 0.0% 225%
Terminal rate 0137(0%) of33 @w 3l-24(13%)
Fuat incidence (days) 73303
Life tit& tat P<O.Wl 1 PMM56
Logistic rtqpmsh teat P<O.ool - P=O.O%

l * Signifkarttly dif&rent (PcOLJOl) from the control grwp by tjre Fisher exact test

; Incidenccaare giw!a as number of lesion-bearing,titna~~~ of animals neaopricd.
Not examined

ksions are ttot&aL A lower incidettcein a dwe group is @k&ted by N.

d Kaplaut-Mcim neoplasm incidence at the end of rho study after adjustpaottt for into
cstinmted istxwiEty
i Observedinci&ncs at terudnal kill
Not applicable; w aeoplasmsin animal group
g 2-Year histti 8n&fewe for untrrated controi grwjs in NTP W&ion stud@ (mtaa f- @ndwd d&&n): 4/!i75
(0.7% *3.496); range w&4%
’ ZYear historical iaddence: 9/M (1.6% + 21%); range O%&%

9-Mont4 interim Evaloath
Mued cell Focu$ o/lo (0%) mo (10%) on0 @%) Qno PJ) on0 VW on0 (0%)
f iepatoeeuular Adeaoola 4m p-m on0 (0%) l/l0 flo%) on0 (0%) l/l0 (10%) Ino (10%)

1544ontb Interim Evduation

f+i (l.hqhiic, EooinophUiG
or Ml cell) 1no (1096) on0 W Q on0 (0%)
f fepatooellular Adcnr?alo 200 tyw on0 (0%) 1 ~~ww ~~;
Wepfmcelfufar carcinoma on0 (0%) Id0 (10%) o/lo (6%) an0 (0%) x/lo (So%) 3n (43%)
f-iepatoctllufar A6enolna
or carcinonla 2430 (20%) 1110(10%) 4m (40%) 3m (30%) 4no (40%) sn (71%)

ZYear Study
hd @asop- Clear Ce& Eohoph&k+ or Wxed Cd)
Overall rate 9/50 (18%) lOIS (26%) 6/56 (12%) 11/%8(23%) 6648 (13%) l/72 0%)
Logistic fcgre&on testb P-O.418 PM.602N b0.414N P=fU3O P!=O*221 P =0.800

f ttpoloEclllalur A&noma
chwau me 13/5a (26%) 13150(26%) WSO (38%) 16/48 (33%) 24/48 (50%) SIT2 (7%)
Adjusted tiMd 32.1% 31.3% 52.1% 57,0% 92.2% 100.0%
Termhal rated 90s (26%) 11/39 (28%) 8f24 (33%) t1m (So%] 3/4 (75%) 089
First Incidence (days) 379 48% 397 434 35351 310
Logistic ttqrehon test P=O.W P=wS2 P =0.X58 P=O.261. P=O.oOa P90.253

clverau rate 11150 (22%) H/SO (32%) 16/SO~f32%) 17/48 (35%) 2&48 (54%) 1m (1%)
Adjusted rate 26.0% 3X$6% 44b8% Slk3% 1008% !%a%
Tern&al rate St35 (14%) at39 (31%) x24 (29%) wz-(SO%) 4/4 (100%) OID
Rmt iaeidena (llays) 540 517 434 351 422
Logistii tefpmdm teat PpO.036 PsQ.142” PdO.389 P=O*o88 P<O.oa P=O.347

l@dodluliu Adeoalue or earrlaw*ac

overaflrate * y&(42%) ~~(~%) ~~%(~~ .~~%(52%~ ;gw$Qw sn2 (7%)
Adjusted ate 100.0%
Terminal rate 13b5 (37%) l?& (49%) 12h po%] 16h2 (73%] %/4 ‘(100%) O/If
Fitst ina’dena: (dap) 379 484 397 434 WI 310
Logistktt?gr&otlteat P mo.067 Pdk37S P-O.@78 P=O.m5 P=O.O30 P*O.rlSO

ltepntoeelluhr Fool,Atkaoum, or Cprrlooma

overall rate 2WSO (54%) 29/i% (S8%) 3l!!W (62%) 30/4@(63%) 34148 (71%) ~2 @%I
Adjusted rote 62.1% 67.1% 726% 87.s% lf?W% 100.0%
Tuminai rate 19ns (S4%) 25139 (64%) 13fz4 (S4%) 1#22 (82%) 414 (EOOg+
Fii-st inddence (days) 379 397 494 351 z3
Lo&die l%.gresion tat P=o.m EL.434 P=O.291 P*OA85 P=O.O07 P-O.564
0 PPm

9-Month Merim Evaluation
Basophilic Focus o/lo (a%) alla (a%) v/lo (v%) l/N (lW6) MO (;a%) ofi ww
lS-Month Interim Evalwtioa
Foci (Basophik, Clear CM,
Eosinophilic, or Mixed CM) on0 (09b) 2nfJ a@4 l/lo (la%) 2110@o%) aR0(m%) on (0%)
Hepatoceblar Adenoma ylo (fQ%) 10 (10%) on0 tom l& ~~~) 3rro (~1 lrz; (Sv%)
Hepatocellular Carcinoma on0 (a%) o/lo (a%) o/lo (a%) me (l=q of2 PI
or Carcinoma ,l/lO (10%) 1no (io%) oil0 (a%) 10 (la%) MO Pw l/2 (50%)
t-Year Study
Foci (Basophlllc, Char Cd, &mbepht&, or Mkd ceu)
Over&l Ate s/49 (16%) 14/w(2Q%) 19/50(38%) %) SBOp%) 4m (5%)
Logistic regressiontest ,j+o.OO3 P=O.u87 P=O*VOl P*O.O4S P=O.618
Overall rate llf49 (22%) 10/4!2(20%) S/50 (18%) 14fio (243%) @#o (24%) 1180(1%)
Adjusted rate 29;7% 27.8% 30.3% 65.8% 8p.o% lOW%
Terminal rate 11137(w?q 8Kq (24%) 6i-24ww Sfll(45%) ff# 010
First incidence(days) 733 (T) 519 365 48s 370 450
L.ogiaticregressiontest P=OSQQN P=O.S31N P=O,Sl9N F~O.a25 P=O.OOQ P=O.SOS

Hepareceualw careinom48
Overall rate <4/49(8%) 6/49 (12%) o/so (16%) 9/sa (18%) s/so (16%) l/80 (1%)
Adjust+ rate 10.3% 14.5% 82.7% 12.5%
Terminal rate 3i37 (8%) X33 (6%) E!&3%) :~~o~) 6% 010
Fimt incidence (days) 677 a7 ala 5% 333 418
-tic regressiontest P=O.178 P-O.381 Ps0.141 P=O.wi P=V;oo6 P=0.910

HepIIroeeuularAdenomo OPCPnJnolnof
ovwall rate lS/49 (31%) l4/49 (29%) wso (3@%) 19/q (38%] 16/s (32%) 2&O (3%)
Adjusted rate 39.3% 34.3% 453% 74.8% 9i.75 100.0%
Terminal rate ,14137(38%) 803 (24%) S/t4 (33%) tyll (SS%) ON O/Q
Fit incidence(days) 677 467 365 48s 3s3 418
Logistic .qreardon test ‘P-O.497 P=b.SWN PoO.441 P10.027 P4wa8 P=O.302

Hepntocellular &cl, Menoma, or CarcLqom~

Overall rate ‘18149437%) 25149(51%) 27lsO(S4%) 2iS/$O$W-&) WSO (36%) a/80 (8%)
Adjusted rate 47.2% 60.2% 803% 9s34% 100.0%
Terminal rate 17/37 (46%) 17/33(S2%) 18/24 (7@%;) IO/i1 (Ql?& On,
Fit incidence (days) 677 467 36s 48s 315
Logistic regressiontest ,P-0384 P+WQ7 P=O.OlO lk0:lW P*O.136

(T)Teminai sac&ice
’ lncidenoesarc given as number of k&n-Wq animai&umba of anifnab n a;i,
b Beneath the control incidenceare the .P vaiws mwciat@ i&h th tread tes& Beneaththe dr#cd pup in&em are the P vainer
cortes~diag to paim co* hfsween the contm@and thar daaedgnwp. The logistic regr+ii teat regardslesiom in
a&nab, dying prior to terminal kill as :&oat&al. A negativetrend or lwur ,inciaencr:in 8:d&e group io itidicated by N.
i Kaplan-Meier estimatedneoplrrsmincidetwe at the ehd of the study after adjustxn~ fur,~t~~nt Wrtality
Ohaerveditienw at terminal kill
’ 2-Year histotical incidencefor untreated ooatrol groupsin NTP inhaiation studies (mean L startdarddeuiatl~tt): M/S72
(34.3% f 124%); range 12%.56%
f Z-Year historical incidencez 87/558(15.6% f 8.4%): rangc3%-27%
HMUltS 53


%Month Interim Evaluation

hlrophy O~Q (v4 o/14(0%) on0 (0%) on0 (0%) 9/lo (9Q%p 8f%(MO%)-
(icnuinal Epithelium Hyperptasia oflo (0%) oflo(0%) o/lo (0%) on4 (0%) o/lo (0%) l/8 (13%)
itcnign Gtanulow Cell Tumor on0 (0%) Q/SO(O%] 4m (0%) W I0 (0%) l/IO (10%) Q/s (Q%)
I S-Month Interhn Evaluation
Alrophy O/IOq%) l/-IO (14%) wlo(9o%p7tlolp%y a2(NM%)
Angicctasis I/l0 (10%) Q/IO (4%) W QW Q au t=s or2(0%)
Gcnsinai Epitheiium Hyperplasia oflo (Q%] O/IQ (0%) onQ(o%)3/x0 (3m) In (54%)
Granulceaceil Hjperpwa on0 (0%) on0 (o%> oflo (046) t/lo (10%) l/2(50%)
knign Granuba Ceil Tumor on0 (0%) o/lo (0%~ t/lo (lo%) 3/10 (30%) l/2(50%)
Malignant GranulosaCell Tumor
llcnign or Malignant Granulosa
o/w pm on0 (0%) O/m(Q%) 1no (10%) Q f2
CetI Tumor o/lo (0%) o/lo (0%) o/lo (Q%) liQ0 (10%) 4/&o (4Q%)* l/2 (50%)
Adenomaor Benign Mied Tumor on0 (0%) o/lo (0%) o/lo (0%) l/10 (10%) l/IO (10%) on (0%)
Z-Year Study
Overall rate 4/49 (8%) 19/49 (39%) 32/48 (67%) 42is0 (8+%) 43/50 (86%) 69#9 (87%)
Logistic regressiontestb P<O.oQi PeO.001 P~O.001 P<O.oOl PdMOl P<0.001

CJveraU rate 4149(8%) 6149 (12%) 3148 (6%) 13/s0 (26%) 14fSO (2fs%) 17n9 (22%)
Logistic regressiontest P60.259 PdU66 P=+O.6@6 P=O.o_17 P -0.021 P==O.425

Gennhld Epl(hlium Hyperpias~

Overall rate 2f49 (4%) 3149 (6%) s/48 (17%) IS/SO (30%,3 wso (30%) l&v9 (B%)
Logistic regressiontest %0.001 Pdbl60 P=OM7 P~O.Om ‘P=O.OlO P*0.001
Granulo6a Cell Hyperpiassp
Overall rate l/49 (2%) o/49 (0%) 248:(4%) 3154(6%) 3/so (4%) 209 (3%)
Logistic regressiontest P=0.370N P=-@.517N P-O.360 P==O.588 P=0.23S P=O.887

BeuignGranuloaaCdl Tumor
overall rate l/49 (2%) or49 (0%) II48 (2%) 6150(12%) 6/50 (12%) 6P9 (8%)
Adjust* rate’ 28% Q.o% 3.2% 28.5% 100.0% 27.1%
Terminal rated l/36 (3%) O/33 (0%) om (O%j l/l1 (P%J o/v OiQ
First incidence(days) ‘Srn -3 677 608 393 311
Logistic regressiontest P==O.Mo P=OSl7N P=O.735 P-O.026 P=O.Q20 P-O.303

Maligaant GranpuosPCell Tumor

Ovemll rate 0149(0%) o/f%9(0%) Q/48(43%) 3/50 (6%) 2450 (4%) O/79 (Q%)
Adjusted rate 0% 0.0% 0.0% 19.3% 54.2% 0.4%
Terminal rate ON (0%) O/33 (0%) o/zLI(0%) l/l1 (9%) O/Q O/o
First incidence (days) 64 369
Life table test P<O.O#l P~&O18 F=O.O03 -
Logistic regrewion test P=O.O68 P--O.046 P=O.Q37 -


t-Ytar study (cwltinual)

lkrrlga or hla@mM Grsmdam Celt Tatao~
overauratc 1149(2%) O/49(0%) 1Mf4Gw t@o (rw @7pwf9
Adjustedtate 2.8% o.t.pfi 1aom Z?S%
TCDlliMinuC l/z6 (3%) OK+3WYI ~(~) 0s
Fuat incidearx (days) 7330 - 677 393 iti
Lifetablctua Pe?,ool P=&!ifRI F--o*6t# P~Q.WS F<0*001 P<0*001
Logkiticngarionteet P=@.OM F=WWN F=&35 P=M!Ol P-0.001 P-o.303
ovemll rate! U49(4%) 449 (a%) l/s8 (2%) m(3%)
Adjusted rate 5.6% 12.1% 42% 4.1%
Tetminai rate 2% (6%) 4c33(12%) l/24(4%) am
Ew iJt&kna (days) 733 (13 733cr) '733(T) 664 46a 257
Logistic reg?ceeiontest Pd.244 P-o.296 P=0.64oN F=El,lll P=O.OlS P=O.9BlN

Sigttificantdydifferent (PcO.05) from the controt gruup by the Fishcr exau tat
** P’CO.01
$l’)Tcnninai sxrifkx
lncidenm are given as au@er of lesi&xaring an~~~~rn~ of w&nuis tqicmsqic+i cd.
b Beneaththe aratrd incidenceam the P valuesrwwci&&witk t&e tratd test. .Huwqth t ipup incidenaMc the P wlucd
awresponding to pahwise comparisonsbewcen the -contmb and that s3o~~’ The iifc.tahle Ualysis regardk ncop&nm in
animab dyittg prior to teqinai :W aa be&g (&&tty or inditwt~) the causq tlY. ‘&?I ieregseasiantatrrtpnirthere
bions as MI&&L A wgativc trend or bwcr incideaa: in a d&q! gmup t indkat~ by N.
fi Kaplan-Mekr estimated ncopism incidenceat the end of the study after adjustmqt for inter@mmt mottaiity
Observediaoidcace at tctminai.kitl
f Not appiicttjk; no ntwpbsms in animsi group
Z-Year historicif incidma for untrwted controt.~rorrpain NT’P inbaiatioa studi& (mtwo 2 stu@wd deviation): 4648
(0.7% * 0.9%); rangeO%~2%

The incidence of g&mix@ epitheliaf ~~~~~ia ;waS ~~~iarly maI- is :tukusnai. These malignant
markedly &reased in the ovarim of-mice exposedto xmpltism are rare In oriczil matrols (males:
20 ppm or above. This fusion consistedof prominent 35% OF i)&-$ fetiies: or 0.5%) (Tables A4f
downgrowth of the surface mewttrelium into the rice of harderian gland
parenchymaof the ovary, forming tubular aztdgland- hat were exposedto
like structures. Ovarian atrophy was an-exposure- deaths due to Iym-
related.eff&ct and the incid&xe in phocytid Lipton whkhprecluded thedevelopment
Characteristically,affected fen&s h The incidence of
oocytea,folliclq or corpora lutqa. also increased in
ar 200 ppm I$-bWwIiene.
Uterus= Uterine atrophy (and, therefore, 1
hyperplasia)was a prominent exposure-reIa incidences of mammary
at 625 ppm and was also observed in the 200 ppm e mice eztpmedto 625,200,
group (Table B5). to l$-butadiene qure.
neopiasms were mammary
Hard&an Gkuk Male-and female mice in the $25‘ carcinomas crlrpe B) (Plate 12); the resf were,adeno-
and 200 ppm groups had @xased inciderxes of a~~tbo~ or rnaii~~a~t mixed tumors (Table 13).
harderian gland adenomas (Table 12; Plates 10 and The aden~~~~orn~ are considered here to be
11). The occnrremx of carcinomas in variants of the ~~~o~ that have prominent
E ‘&gker mqpifk+ion of PBte 1 showing the variably sized clefts snd
spac8w(ftvqelIt~ containing blood &IS) IinaJ by plump c&Is with kg~
veaictdrrrnudei. ~Fcbde ,W3F rwtlse e to 625 ppm l,Sb~tadlene
in the 2.year isthalafion study. A&E x300
CSAIKs lBr;rsne6
I WI Rlveolar~ronchiolar cminoma with cellular atypia and perip@xai Foreaomach: fjrpmow projects from the mucosaisurf&
NWWI into the remaining normal pulmonary pareachytpa.~F~II&!c B6C3Ft into the Iumw of the Note the and folded
uuv aposed to 200 ppm l&butadiene in the Z-year inhalation study. LerathliZed e~i~~iu~. F1 mow to 625 ppa
ftrl -1.50 l,fbwadi&e in thg 2-year ~~~ study, H&E xl6

KAtu 7 PrA3%8
h*wc*uach: Squamouscell carcinoma. C&s and clusters of neopktstic Liva: l-hptbtoccllulat karcinoma~witba trabecular pattern. h-&deWC3Ft
wI*trtic4 squamousepithelial &is with foci of Eeratinizatiou have in&&d mouse exposed to 260 ppm l&tine in the 2-year inhaiation study.
W tmina propria. F&male ESC3F mouse, expgaed to 62S$pm H&E xl50
1,s Mulicne in the 2-year inhalation stA. H&E xl50

15MontIs Interiim Evaluation

HyperplaSii3 an0(0%) onww
Adenoma ona,(0%) 3/7 (43%)

t-Year Study
Ovfmll rate l/so (2%) 3i49 (6%) 4/m (8%) 6/47 (13%) &r47 (17%) 5140 (13%)
Logistic cegressioa testb P=O.o69 P&.335 P-O*154 P=O.Q32 P=0.016 P=O.740

overall rate 6150 (12%) mp (14%) S/SOQ6%)
Adjusted rate’ 148% 17.3% 25.8%
Terminal rated UJS @ W 6J39 (15%) 4ta (17%)
Fmt incidence (days) 543 652 536 464 382
Logistic regressioa test P<O.Wl P-O.497 Pa0.395 P<O.tnn PCO.001

Overall mite o/50 (0%) l/50 (2%) l/50 (2%) 3150 (6%) Xi0 (4%)
Adjusted rate 0.0% 2&i% 4.2% 11.7% 6.3%
Terminal rate y (0%) Ii39 (3%) l/24 (4%) If22 @% 014 (0%)
Fmt incidence (days) 733 (IT) 733 (T) 680 495
Logistic tegmsia test P=O.?20 P wo.522 pro.425 PdLtB6 P=O.352

AdeIMJnla or CuciDolaoC
ovemil rate 660 (12%) 7mt (14%) Q/SO(18%) 20@0 (4@%) 3lJ50 (62%) 6I73 (8%)
Adjusted t-ate 14&% 173% 29.5% 64.9% 9%s% ltW.O%
Tetminaj rate 283s (6%) B/39 (lS%) 5124(21%) Kz.32 (SS%) 3f4 (75%) OIQ
Fmt incidence {days) 543 a2 536 464 382 209
Logistic rem test P<O.Wl PmO.497 P~O.217 Pcoao1 P&X#l l-O.002



15Month Intwim Evaluation

Adeuoma l/l (100%) l/f (mm)

t-bar Study
Overall rate l/50 (2%) 5/49 (lQ%) 9J48 (19%) 4J49 (8%) 7J66 (11%)
Logistic regression test P=O.213 bdM97 P=O.OlZ PdG?71 P=O.5#

Overall rate s/50. (16%) lo/q3 (m%) fiJs0 (22%) 15Js (30%) 20150 (40%) 9/w (11%)
Adjusted rate 20.8% $4246 61.0% 893% 45.2%
Terminal rate 7J37 (19%) 9!33 (27%) iK&%) 4Jll(36%) Q/o OIO
First incidence (day) 658 722 569 S32 370 307
bglstic regression tet P-O.046 P mtI.356 P-OXlN P*Wl6 P4J*QOl P-O.176

Overall rate
Adjusted rate
l/5@ (2%}
l/50(2%) QJ50
pm l/so (2%)
Q&Q (Qw
Terminal rate Q/33 (@%) z&%) oJll(Q%) o/Q or0
First in~dence (dqa) -= 722 541 695
LQgistic t-e* test P=O.S73N PmO.493 P=M31 Pd).o&3 -

A&BorM or c.arciaorrmh
overall rate WO (16%) lO,kk?(20%) 7150 (24%)
Adjusted rate 2W% 29.2% 2223%
Terminal rate 7J37 (19%) 9133 (n%) 424 (17%)
First incidence (days) 658 722 541
Logistic regrwsioa test P=O.O61 P&356 P=W7SN

cr)TenninaI saui&e
* Sipificautty diit (P<O.OS) fawn the control pup ty the Fisher wct test
** PCO.01
a lttddencea are ghwo as aumher of legion-bearing anime+&umber of animals ~~~~ in@d 05 numbes of aldnuds
b incldencc are the P vslue8
k&MS in

i observed ltlci~ at temlinal ldu

f Not appliibk; no neoplasms in animal group
2-Y- hkt&cai i&&am for untreated contr!ol groups in NTP inbabtion studies (UleM ti Standani dcvistion): 25J575
(4.3% A 3.8%); range O%-12%
g Not dried
’ ‘t-Year historical incidence: 13JS41(2.3% k 2.5%); m 0%-8%
Results 5!4

Mammary Gland Lesions in Female M&s in the &Year Inhaiath~ S

!J-Month Interim Evalwtioln


H-Mouth ieterinm Evaluatiou

Hyperplasia o/lo(0%) l/l@(10%) QQ(0469
Menocarcinoma QAtlPw zm Pm tn (50%)
L-Year Study
C%rall rate a50 (4%) 4J50wo 7BO (14%) 2@Q(3%)
Logistic regreaaiottteatc PmO.626 P*O.132 P=O*OOS PwO.914

Overall rate 6.&o(12%) 4150(8%) QmQcm
Adjusted rated 325% 13.6% 0.0%
Terminal ratee 2Al(18%) o/Q OKI
first incidence (days) 579 ,268
Logistic regressiontest 5=0*005 P*O.331 -

overall rate o/so(0%) wo (4%) 6w (12%) llC50 (22%) 12Bo (15%)
Adjustedrate 0.0% 5.7% L6.2% 392% 100.0%
Terminal rate om (0%) Q&4 cosa) oil1 (0%) ox)
first incidence(days) 645 392 370 iti
Logistic regrcaaionteat P=O.OsO PpO.2s7 P Ml.076 P=alo9 pro.004

Maiipmbt Mixed Twmor

Overall rate o/so(0%) Q/50(0%) _o/so(0%) 4/w (5%)
Adjustedrate 0.0% 0.0% 0.0% 29.4%
Terminal.rate 0137(0%) on1 (0%) on, O/o
First incidence(days) - 335
Logistic regressiontest P=O.O41 PmO.276

M+ooaca~tbomq C’A~CIQOIQW,
or MalIgna@ Wxed Tuqr
Overallrate Q/JQ@?J9 2450(4%) 4/50 (8%)
Adjwkd rate 0.0% 5.8% 13.0%
Tcnninal rate 007 (Q%) l/33 (3%) l/24 (4%)
Iht incidence(days) 732 64s
Logistic reps&m test P=OB26 P ~0.228 P *0.056’

* lncidencesare given as numb of lesion-b@arittgani

h Not examined
’ lkneath the control incidencewe the P values w&a&d with the trend Itit. Beneath the dosed up incidencearc the P vaiwca
wrrqmnding to paiwiac compwiwns betweenthe @ontrolsand that dosed group. on test regard0lesion3in
animalsdying prior to terminal kIl1IB nonfat& A negat&et&d or lower &ide~+~ in,+ dose grotSpiarindicated by N.
’ Kqlan-Me&x estimatedneoplasmiucidetice twthe end of the study afteradj~t~~t for intercur+p?nt
’ c~twwcd incidenceat terminal Lill
’ Not applicable;no neoplastnsin animal group
56 &343utadiene, NTP ‘I32 434

squamous differentiation. The malignant mixed they were probably related to exposure to
tumors consisted of epithelial components arranged 1,3-bumdiene in males ,and were possibly related to
in glandlike structures and anaplastic. spindle-cell exposure in females.
components. Mammary gland carcinomas and adeno-
acanthomas are uncommon in female BKSF, mice, Skin: The incidencea of neuroflbrosarcoma or
with carcinomas occurring ins 21 of 561 control sarcoma of the s~~~n~~ tissue in female mice
females (3.7%) and adenoacanthomasoccurring in 1 : to 62.5; 200, or’625 Ppm were significantly
of 561 control females (0.2%) in the NTP historical t but not by the logistic
database (Table B4i). Mammary gland hypet@~ia d B3). Nevertheless,
occurred at a slightly. increasedincidence in female (all types) are urrcom-
in the 62.5 and 200 ppm groups (Table 13). and have occurred in
historic& control females
PreputiaJ Gi&n& In the 200 ppm group, We male (Table B4k). Thus, these Napier may be exposure
mice had preputial carcinomas (Table A3). ,Preputial related.
gland carcinomas are rare in B6C3F1 mice; none were
reported in one NTP survey of historical control data.
preputial ,gland carcinomas in the present study were Zymbal’s G&$+ Z~~al’s,~d neoplasms are rare
considered to be-related to exposure to l&butadiene. spontaneous lesions in mice, occurring in no
Some preputial carcinomas were composed of large historical extol anim adenoma was seen in
eosinophilic epithelial ceils that were we11differenti- a control male mouse (Ttible Al). In females, no
ated toward squamous cells with keratin pearls, or Zymbai’s gland in controls, and
toward sebaceous-like cells. More &equently, the one adenoma a, re diagnosed in
carcinomas had necrotic cores and a thin layer of very the 625 ppm group (Tabie Bl). These neoplasms
anapiastie pleomorphic basophilic epithelial cells that may be related to ~~erni~~ eqosure.
aggressively invaded surrounding &sue and blood
Vesseis. Sr~~UJntesljltezC.qrc@ma# of the small intestine are
uncommon in the mouse, with none oceur-
Kidrtey: The absolute kidney weights of females ring in h~to~~~ t&e from recent NTF+
receiving 625 ppm at the 9month interim evaluation inhalation studies; t presence of carcino-
and females receiving 62.5 and 2Of.-ppm at the masin 6.25 ppm and in one
U-month interim evaluation were greater than‘shose female is of interest, One
of the controls (Tables El and’E2). adenoma was present in a female eqosed to
6.25 ppm (Table’Bl), noted in one
Renal tubule adenomas were not $iagnosed in, the male in each of the 6. PPm groups;
control mice and are considered to be rare carcinomas were present two ammals exposed to
spontaneous neopiasms,with an incidence in control 200 .ppm (Tab& Al and A$). It is difficult to
mice inNlT studies of 0.2% for males and 0.0% for determine the relatio,nship of these. neoplasms to
females (Tables A4g and B4j). Two femaie mice exposure to ~~-b~#dien~ however, no proliferative
exposed to 200 ppm had renal tubnlo: adenomas intestinal lesions were no In controls.
(‘Table 14). Male mice had a higher ittcid@me of
renal tubule adenomas than females, with thre& in Testis: At the 9-month iuterim evaluation, the
the 625 ppm group, one in the ,200 ppm group, and absolute testis weight of es receiving 62.5 ppm or
one in the 625 ppm group. H~t~l~~~lly, the renal more and the relative testif weight of males receiving
tubule adenomas contained multiple dilated tubules 200 and 625.ppm were lower than those of the
separated by thin connective tissue septa. Epitlteiiai At. the Smonth interim
cells in these neoplasms causeil a papillaty ive testis weights of
ance, and some cytopiasmlc and nuclear pleomor- pm were lower than
phism was present. Renal tubule hyperplasia was those of controls (‘Table E2). The decreases in
characterized by one or two dilated renal tubules, testicular weights seen at the 9-- and 15month
usually in the outer cortex, lined by enlarged epithe- interim eviction spare to be dose related, and
lial cells showing some degree of nuclear enlargement correlated we11 with the diagnosis of testicular
and piling up. Considering the rarity of these lesions, atrophy.

l!LMontb Interim Evalugtioa

Renal Tubule: Adewma ona(wq o/3@w Oh (0%) on0 (o%} 1R (14%)
Z-Year Study
RellaiTubllk2 liprph&
overall rate
Logistic regIms& ted
cm f?w
1M (1%)
RmaiTubukz kLewmsc
overall rate ol50 (0%) l&I (2%) IMP (2%) $7&W@
Adjusted rated 0.0% 2.6% 3.7%
Terminal tatee tJp5 -w@ l/39 (3%) w p%) Oh
First incideikce(days) 733 (T) 573
Logistic l‘egreM* test P=O.63U P=AkS2 P=J0.580


Z-Year Study
Rend Tub&z Hyperpiasia
Overall rate 0149@%) 0149(0%) o/50 (0%) oB@(os6)
RcnalTubuk &
overail rate 2450(4%) y&f?%)
Adjusted rate 352%
Terminal rate OAt O/Q
Fmt incidence (day) 418
bgistic regRssioa teat P=O.276

~Tetminal sacrifk
z Incidenceaare given as number of lesion-@wing ~irna~~~ bf animais mictxSopkrrlty examined.
Beneath the coat& incideace are the P values associatedwith the @end teat. &n&tit t he P values
coneqading to pa&wise eom$ari&ns donedgwp: )The naoplasms
in animals dying prior to termiaal til riuwerincidencei N.
’ 2-Year historical incidmce CK linka ation ~studkv(moan t atandatd d&t&@: U57L
(0.2% zt,0.3%% range 0%-l%
d Number of neopksm-bearing animals/effectivenumber qf animals, i.c, numkr & a@ais alive at first accurrena: of this n
F CWetved incidenwat twminaf kill
Not apptieabk; no neoplasm8ia anhnal group
g Z-Year histori iacicEenae:O/$59(0.0%); upper 95% cfxdidence timit = OS%

0 PP*

2-Year Study
overall rate 3150
(6%) on3(0%)
Adjusted r8tec 11.7% o.u%
Terminal rated o/4 (0%) O#
F’imt incidence (day+) 530
Lie t8Me test’ P~O*Q51 -

U-Month Iotersm Evaluation
Neurofibrosarcoma m (50%)
2-Year study
Over811rate o/so(0%) Of50(0%) 1/m (2%) Iis0(2%) o/so(0%) 0~ (0%

Htz*b l/SO (2%) 2Jso (4%} 2456(4%)j 2/!w (4%) Ft3(3%”

Adjusted rate 27% 5.2% 13.2% 21.4%
Terminal rate lR3 (3%) 103 (3%) onl(o96) o/o On,
Fit incidence (days) 733 (I) 572 665 s69 328
Life table test PCO.uul PmO.471 P=4w3 P=O.O14 PmO.153
chm8ll rate l/50 (2%) 260 (4%) 315046%) Slso(losb) 3/50 (6%) 3/80 (4%)
Adjusted rate 2.4% 5.3% 8&I% m3% 43.0% 19.3%
Terminal rate o/37 p%) lLs3 (3%) w-24(W .onr (cy&> w OED
First incidence (days) 677 642 569 597 524 316
Life table test PcO.001 P=O.476 P-O.238 PMkO17 PMMK! PmO.013

(TJTehninal sac&iee
t Incidence43 8m given as nuder of neopksm-bearinganimals/numberof an+” necK@ed.
2-Year histofid kidcnw for untreated“amttvl gwps in NTP inhalation drudges(wan + deviation): Oi575
(0.0% 4 o.o%> mge o%a
’ Number of neop&m-bearing snimak/cffceti number of animals, i.e., uumbcr of an@na+afive at first occurrenceof this neoplarun
d Observedi&knee at terminal kill
’ Beneaththe control ine&etwe are the P v&is ass&&ted with the trend test. Beneath the pup incidenecate the P vaIues
correep&ding to p&w&c compwisons be$wecntbc’$unt& and that d@cd group The life ta@c anaiysii rcgasls ncopkms in
animals dying p&r to k&al kill 89 being (dire&y or indirectly) the causeof death.
f Not applica~ no neo& ia anhnal group
i a-Year hktorkal incidence: O/561
2-Y-r historicai ineidcmx l/561 (0.2% ,t 0.6%); rangeO%-2%
i 2-Year historical incidcnk WS61(0.4% c_ OS%}; rangea%-2%
ttesutts 39

Testicular atrophy was observed to be a prominent with single lay- of coiumnar, cuboidat, squamtxts,
exposure-relatedresponse ,only in:~mide or reapirataty ~p~~~~~i cells covering the defti.
625 ppm (Table AS). Testicular atrophy The atrophy-~,~~l~y rni~~~ to mild in sever&y
terized by a uniform minimal to : mild decreasein and usually affrectedthe ~~~~0~ epithehurn at the
cellularity of the seminiferous tubules. dorsal meat- af .~,~te~~r nasal secticxt. These
nasai l&o* were similaf to those seen in previous
Nose: Olfktq epitbelial atrophy was observedin M M ), but no osseousor
seven female mice exposed to 625 ppm and is1one observed in the present
control (Table BS). Atrophy was ass&iated with an
inflamed tooth. in one control mouse and was uni-
lateral in the other contrc31mouse. In. female tic43
exposedto 625 ppm, the lesion was pr-cibabiyr&&?d
to exposure to l$butadiene; lesions were bilateral
and were present in the middle and paster&x nasal
sections. Olfactory epirhelial attipby @xurkd in
male mice exposedto 20 ppm or abovq the inkence to 625 ppm tn the present studies. CWactory
in males exposed to 625 ppm was lower than the epithehal ksiarts observed in animals exposed to
incidencein females,but was possibly also Mated to lower do++ in the Pratt studies were unilateral or
exposure (Table AS). Atrophy not of the same chara~&~ as those observed at
terized by focal loss of olfactory sensory neurons, the highest dose, but were still diagnosedas atrophy.
Estimates of survival probabilities for male micx are
shown in Table 16 and in the K&an+kier curve8in
Figure 4. Survivalof all groupspf
to Wbutadiene was signi&antly
the controls due to the develop
neoplasms, particularly mali*a to 625 ppm for 26 weeks was sig-
hemangiosarcomasof the heart.1 nificantly- liar than that of the group exposed to
of male mice exposedto 625 ppm for 13 or 26 #x&s 312 ppm for 32 weeksi
became apparent between man
study, whereasthat of male mice
for 52 we&s and 200 ppm for tSOf and control male mice
apparent between months 12 an were sin&r (Tab& 17 an& Egure 5).


Q PPea

Survival ana@isd P<O.W P<O.W PCO.OQl PCO.001

‘, cenrwnd from survival awIyxs

Kaplan-Meier d&tminations. Sutvivalrates adjufltedfor interim evaluationsa~,~~~ d-t&s.
i Mean of all deaths(unw&, cenxkd, tkminal tiC@@)
Tbe result of the life table trend test (‘I’%-, 1973) is in &hecontrol column, and the rest&6 af the life table pairwipeannparisaar
(Cox, 1972)with the controls are in the dosedcolumns.
0.0 .............. ..$ ..............
G;;j.. .............i”“” ..
0.1) ....... ........+................ ....b.

Cl ii?6 C P M SEtY : :
,_.....*..& .. . .I...,.. ...i .*...... ‘...I.

0.0 0
1 23.0 70 22.6 98 HI 99 50
2 70 25.6 105 50 103 50
3 2: 70 26.7 103 so 102 50
4 27.1 70 27.9 103 50 iit: 100 50
5 27.8 70 28.8 104 50 27.9 loo 50
6 28.7 70 29.4 X02 so 282 98 50
7 w.2 70 30.3 X04 50 29.0 99 50
8 30.1 70 31.2 104 50 29.6 98 50
9 30.7 70 32.0 104 50 .30.1 98 so
10 31.1 70 325 to5 50 3&Q 99 50
11 32.0 70 33.4 X05 JO 3L2 98 50
12 327 70 33.7 103 SO 31.0 95 50
13 33.0 70 34.0 103 50 31.5 % 50
17 35.2 70 36.7 104 50 34s 98 50
21 36.6 70 39.7 109 50 36.2 99 50
25 38.9 70 41.7 107 so 98 50
w 40.8 70 43.6 107 50 iz 99 50
33 41.7 70 ‘43.8 105 49 41.4 99 49
37 43.2 70 45.1 104 48 43.1 100 49
41b 43.3 60 46.3 107 47 44.1 102 49
45 44.1 60 46.3 105 47 45s 103 47
49 45.5 60 ‘47.4 104 46 46.1 101 45
53 46.0 60 47.2 103 46 4.8 100 41
57 45.7 59 4741 104 46 46.8 102 39
61 46.1 59 48.x 104 44 47.1 102 36
65b 45.7 49 46.7 102 43 475 104 W
69 45.3 48 46.0 100 40 47.9 105 27
73 45.3 46 45.0 99 37 47.0 104 22
77 45.1 46 44.6 99 32 44.9 100 21
81 44.8 44 43.9 98 k7 104 14
85 443 43 43.9 99 it: 46.9 106 12
89 44.0 41 421 % 24 427 97 9
93 421 39 394 91 20 37.6 4
95 s 429 39 40.0 93 15 35.8 z 3
97 43.0 37 40.4 94 13 38.3 89 2
99 423 37 38.6 91 13 36.1 85 2
101 41.6 36 37.8 91 9 34.6 83 2
103 41.5 35 37.1 89 9 31.8 77 1

TermhdstcriNce 35 9 1

Meon for weeks

l-13 z&9 29.9 103 28s 99
14-52 41.0 43.4 106 45.0 100
53-103 44.2 43.0 -97 42.3 96

Itesults 63

Meaa Body WeLgbts and Svcvfvol of &fide Mice in tbet y of I,%Bta-

Wak 8 DDUk
Av. Wt .Ntt&er of Av. %%%a wL(% of Nosr@ of As. wt. -wfi (% d Numbor of
Sltrvivors w aentrols) swvm e, colttro~) SOL

1 a0 70 223 97 so 227 99 50
2 24.4 70 25.1 l# so 1w 50
3 25.8 70 26.4 102 so -z 104 50
4 27.1 70 100 101 so
5 27.8 70 $2: 101 ifi 273
28.0 101 so
6 2a7 70 28.9 101 so 101
7 29.2 70 29.4 10% 50 z IO1 zii
8 30.1 70 305 101 3101 so
9 30.7 70 31.0 101 2 30.4
31.3 102 so
10 31.1 70 31s 101 SO 31.6 102 so
11 320 70 322 101 50 32.4 101 SO
12 327 70 32.7 100 50 33.0 101 50
13 33.0 70 33.0 loo 50 33.2 101 SO
17 35.2 70 36.2 103 so 363 103 50
21 36.6 70 385 10s 50 303 10s so
25 36.9 70 40.4 104 48 406 10s 49
29 40.8 70 43.7 107 47 43.2 106 46
41.7 70 43.4 104 4s 424 102 42
z 43.2 70 443 103 42 44s 103 34
41b 433 60 44s 303 38 46.0 106
4.5 44.1 60 45.9 104 3s 484 110 z
455 60 46.4 102 34 40.7 107 24
zi 4&O 60 47.4 103 31 49.3 107 19
57 45.7 59 47.5 104 31 4024 106 19
61 46.1 59 4&O 104 30 a4 10s 17
tsb 45.7 49 47.7 104 30 48.6 106 14
69 458 48 $7.2 I03 2!a 43.7 106 13
73 45.3 46 47s 10s 46.4 102 9
77 45.1 46 45.5 101 n 41.9 93 8
81 448 44 44.0 96
85 443 43 43.0 ‘97 zi
89 44.0 41 40.6 92 16
.= 43.1 39 403 94 11
95 429 39 38.2 89 11
97 43.0 37 37.6 87 9
99 42.3 37. 35.6 84 8
101 41.6 36 33.7 01 7
103 41.5 35 34.9 84 6

35 S

Moau for weeka

l-13 28.9 29J 101 293 101
M-52 41.0 42.6 104 43.2 10s
53.103 44.2 424 96 47.4 107

’ Week on study of stopnpolhurc group. For 17 thmugb 95, Kmrot Dlasar wtn WA one week after atoparp~~lr

b Interim Nalwtiotu in control ttatka oecumd during mcks 41 and 66.

64 1,3-Blltsdieae, NTP TR 434

male and frail mice in the 2-yeasstudies,

was also seen in We mice exposed to 312 ppm for
This section describes the statistically s~~~~~t or 52 wee& or 625.ppm for 13 or 26 weeks (TabIe C!!).
biologically noteworthy changesin the incidenceaof
neoplasms or nonneopiastic lesions of the hemato- Lung: The i~~~en~‘of hyperplasia of the alveotar
poietic system, heart, lung, liver, stomach, harderian ep~the~urn~ ~iv~lar~r~n~h~olar adenoma, and
gland, kidney; Zymbal’s gland, brain, and fireputitil a~v~lar~~n~h~~ar’ ad~n~r~noma or carcinoma
gland Summaries of the incident of n~~l~~ were si~i~~~t~y greaterin eachof the stop-exposure
and nonneoplastic lesions, the individual animal groups than in the controls (Table 20).
tumor diagnoses,the statistical analysesof primary
neoplasmsthat occucfedwith an incidenceoflat least Liver: The incident of hepatocellulat adenoma of
5% in at least one group, and the historical the liver were si~i~~~t~y greater in the 200, 312,
incidences for the biulogicajy signifi-caatneopiasms and 625 pp~.~~3-w~~) stop+?xposuregroups than in
mentioned in this section are presented in the controls, while the ‘irkidence of hepatocellutar
Appendix C carcinomas Gas not incr in any of the stop-
exposure groups (Table 21). Because the cause of
Exposure of male mice to ,200ppm for 4 weeks, death of mice with he~atocellular adenoma or carci-
625 ppm for 13 weeks, 312 ppm for 52 weeks, or noma was generally attributed to other malignant
625 ppm for 26 weeksinduced neoph&ns at the same neopiasms,the f-ogisticregressiontest was considered
sites observed in the 2.year studies. the most a~pr~~~~te analysis.

Hematopoietic System: The incidencesof mice with Stomach: The increasedincidence of hyperplasia of
malignant lymphoma in groups exposedto 625 ppm the foreatomatih e~iEhelium in each of the stop-
for 13 or 26 weeks were markedly greaserthan the exposure groups was not significantly greater than
incidence in the controls, while the incidencea in that in the control g 22). Squamousceil
groups exposedto 200.ppm for 40 weeks or 312 ppm papiflomas ooxtrred enceain each of the
for 52 weeks were only marginally increased grdups, and the ~inctdkes were not s@tifkantly
(Table 18). The majority of the I)mphomas tiere the greater than the control incidence by the logistic
lymphocytic type apparentlyarising frondthe thymus, regression test, thk moat appropriate analysis for
During the first 9 months of the study, a$1&rly these no~f~~~ lesions. Squamous cell carcinomas
deathsexcept one were related tothe d~el~~ment of oqwred i‘n groups of mice exposed to 312 or
malignant lymphomas; the causeof death of one male 625 ppm (13 or 26 weeks), but not in mice exposed
exposedto 2QOppm was not determined. The it&‘- to 2fXI ppm nor in the camrol group. The incidences
dence of histiocytic sarcomaswas also sigmficantly of carcinoma in tbe.~t~~~u~e groupswere signi-
increased in each of the stop-exposuregroups, but ficantly greater than the incidence in the control
they occurred more frequently irrmice exposedto 200 group by the life table test, the most appropriate
or 3i2 ppm analysis for thesefetal ~~~1~~.

Hearr: The incidencesof endothelial hyperp Hard&&B i&r& The in+dences of adenoma of the
hemangiwma in most stop-exposure harderian gland yere significantly greater in each of
significantly greater than in the controls ure groups than in the controls
In contraat to the increased‘kcurrence of malignant ~r~narn~ ockurred at low incidencesin
lymphamasin groupsexposedto 625 ppm, hknangio- seti to 200 ppm for 41)weeks, 312 ppm for
sarcomasoccurred more frequetttly in miqe,exposed ,, or 625 ppm for 13 weeks; none were
to 200 or 312 ppm. Herna~~o~~ were obsem in the controls or in males exposed to
observedas early as 9 months in each of the200,.312, Focal hyperplasia of the
and 625 ppm (26-week) stop-exposure groups. with low frequency in each
Myocardial mineralization, a lesion that occurred in
66 1,3-Butadkw, NTP TR 434

Mslignatit Lymphoma end His
d 1,3&utadiene

LymphocytiC Maligoant Lympbqma

overall rate* a50 (12%)
2&J (4%)
om (@w
lb (w%)
35.B% ifZitml
FQst iaeideaa (days) 511 208 169 159
Life table testd P4.033 Pa.001 P<O.ool

All oqputst Lympbcmla (Nixed or NOS)

overall rate 2/m (4%) x50 (4%) 4150(8%) 3150(6%)
Adjusted rate 5.3% 7.8% 5a.B% 43.3%
Terminal rate l/35 (3%) oil) (0%) o/l CQW 010
Fit incideaa (days) 666 514 217 251
Life table test P=OSG! P~O.WS P=0.002

Mnligmnt Lympitoma o;y~~Piwytic, Mixed, or NW)

Overall rate 450 (8%) a/50 (16%) 22f50 (44%) IHO (16%) 3380 (66%)
Adjustedrate 9.B% 324% 5&2% loQ.o% 89.5%
Tetmiaai rate l/x (3%) 119(11%) 18%(rn) m @w) 010
Fit iacideace(days) 511 208 169 217 159
Life table test P4.023 P <O.ool P<

Histiocytic Sarcoma
clverall rate s/so (10%) 2J-50(4%)
Adjusted rate 21.3% 15.6%
Termiaai rate OfJtom ON
mst IncMeaee(days) 576 364
Life table test P=O.w6 P4M36

i Number of neoplwahmiag tiaimals/iaaabet’,ofsaiawis aecropsied

Kaplan-Meierestimatedae~pbwa incidW at the end of the study after t fov latercurreai &xwt3dity
i Obsecvd incideaa at teradael kill
Beneaththe dosedgroup iacidencc a~. +e P valuescwreqxmdiag tb pairyise uxn the coat&s aad .that dosed
group. Tbe life t&e analysisrqgardsneopiams in animals dying prkw to4ezMaaLkill 8s Wig (@e&Q or iadkectQ) the causeof
e Not applicable;ao aeopkms in aabnal group
owrall rate 15m (30%) 7m (14%) 13m (26%)
Adjusted rate’ 76.2% 61.8% lwt.O%
Terminal rated w (54%) otv
First lncidencc(days) 330 306
Life table teatb P<O,uQl P< P<OJm

i Number of lesion-beatinganiqaldnumber oEan@& qemopskxi

Beneaththe dosed group incidcna are t&e P valuesmqqxmding,to paitwkw c& b@vtm $e amtmls and that dosed
group. The life tabk ana&& regards n~piwms in anitiais dybg prior td terminal kiil as %ng (dJmt@ or indinctiy) the causeof
death. Tbc logistic t-epedon”tcptq regard tlpe lasiaagas nonfat&
f, Kaplan-Meier estimated ntopiasm ,yzM&x at the.en& of the study after ardjustmnt for int elvsmwlt mortality
obl3crwd hlcldul~ at tclminal kill
e Not appiicablq no neoplaatnsin animaI group
68 1,3Jhtadiene, NTP TR 434

Lung Lesions in Male M&e in the St0 St&y of 1,3=Butadhe

Aivdar Epitheiial Hyperpbia

Overall rad 14/50.@8%) ll/SO (22%)
Logistic regressiontesCb P<O.ool P<O.Wl

Alveolar/bronchioIar Adenoma
Overall rate x8/50 (36%j 24/50 (48%) 2&B (52%) U/so (24%)
Adjusted rateC 46.9% 94.3% 1po.m 100.0%
Terminal rated lV3.5 (43%) 8@ (89% xn (100%) OJJ a
First incidence (days) 572 39 344 358
Logistic regressiontest P*O.W P=Q&Ql P<O.ool

AiveoJar/bronchiolar Aden~rcinoma or ~~~rn~

OveraUrate 5/5Q(10%) 22J50(44%) 16/50,(32%) ll/SO (22%)
Adjusted rate 14.3% 89.5% w?.o% 100*0%
Terminal rate St35 (14%;) 7/9 (78%) 1n @Qo%j W
First incidence(days) 729(-q . 481 392 241
Life table teatb PC0.W P<O.M)l P<O.ool
Logistic regressionteat PCOAHZ P<O.Wl P<O.ool

Ahwlar/bronchiolar Adeuoma, Adiqw@notia, or Carciiwma I

Overall fate 21/ro (42%) 34250 (72%) 32/w (64%) &/&/O$@W
Adjusted rate 54.9% 100.0~ 100.0%
Terminal rate 18135(51%6) 9!9 (100%) on (lwq Olr)’
First incidence(days) 572 399 241
Life table test PCO.001 PcO.001 PUO.001
Logistic regressiontest P <O.Wl Pt0&01 P<0*0@1

(T)Terainal sacrifice
t Number of l&ion-bearing aainnals/uuruberof aninaalsaecrq@ed
Beneaththe dosedgroup~incidenceare the P valuescorrespondingto paiwise cotuparkanrrbe- the controls and that dosed
group. The life table analysisregards.neoplasuwinqkuals dying prior to terininal kill as being cditwtly or indirectly) the causeof
death. The logistic -iion teats regard theaeWons as nonfatal.
i Kaplan-Meier estimated neoplasmincidence at the cud &the study after adjustiwkntfor iuterc~ukeutmortality
Observedincidenceat terminal kill
Ueaulk? 69

Liver Neqhsms in M&e,_ Mice in StopExpogure 1~~~~~ etude of L~~~~die~~

0 PPW 200 PI= 92s pjml 312 ppm 625 ppm

(40 weeks) (63 weelm) (S2 weeks) (26 weeka)

Hepatocellular Adenoma
Overall ratea 13150(26%) 27149(SS%) 19/49 (39%) WSO (38%) IllSO (22%)
Adjusted rateb 32.1% 91.1% 91.0% ‘100.0% 100.0%
Terminal rate’ 9135 Pm. 7/9
3~ (78%) 4/s (80%) ‘l/l (EOO%) O/o
First incidence(days) 379 471 326 313
L.ogisticregressiontestd P<tj.ool P-m42 P =OMS P=OL!M

Hepatacehlar Carcinoma
Overall rate n/so (22%) 14149 (29%) 14/49 (29%) lo/50 (20%) 4/50 (8%)
Adjusted rate 26.0% 50.3% 90.9% 74.6% 50.5%
Terminal rate s/z%(14%) l/9 (11%) 41s po%> on (0%) O/o
First incidence(days) 540 407 520 382 483
Logistic regressionteat P-O.S3ON P=O.142 P=0.453N P=0.393N

Hepatocellder Adenoma-or Carclaoma

Overall rate zwso (42%} 3349 (67%) 24149-(49%) 2450 (48%) 13/.50 (26%)
Adjusted rate 47.9% 93.4% 94,4%‘. 100.0% 100.0%
Terminal rate *x3/35 (37%) T/9 (78%} 4iis @o%) 1n (100%) OP
First incidence(days) 379 399 471 326 313
Logistic regressionteat P =0.004 P=O.o63 P=O&SQ P=O.SW

Hepatoblastoma, Hepatocellular &h~%~a, or Carcinoma

Overall rate 2itSO (42%) 33/49 (67%) 2449 f49%) zs/so {So%) 13/so (26%)
Adjusted rate 47.9% 93,4% 94.4% 100.0% 100.0%
Terminal rate 13L35 (37%) 7/9 (78%) 4/5 (80%) ld (109%) O/u,
First incidence(days) 379 399 471 326 313
Logistic regressiontest P=O.OtM P==‘o.O63 P-O.103 P=O.S61

E Number of neopIasm-bearinganimais/numberof d$mals micrpscepic$EIyqmined

Kaplan-Meier estimatedneoplasmincidenceat the end of takestuq after ~justment for~lntercurrentmortality
i Observedincidenceat terminal kill
Beneaththe dosed group incidenceare the P yalttesoarreapondingto pajrwise comparisqnsbetweenthe controls and that dosed
group. The logistic regressiontests regard neoptasmsIn animals dying prior to terminal kill as nonfatal. A lower incidence in a
dose group is indicated by N.
70 1,.Butadiene, NTP TR 434

Q PPm 312 ppm 625 pm

(6% weeks) (al weeks)

Epithelial Elyperplasia
Cwerallratea 4/so (8%) lQl48 (21%) ’ 201’48(42%) lY50 (30%)
Logistic regressionteatb P=O.l63 ‘P=O.oM P=O.S66

Squamous Cell Papilloma

Overdf rate l/SO (2%) MO (6%) 4/50 (8%) 4iso (8%)
Adjusted rate’ 2.5% 21.4% 100.0% 20.1%
Terminal rated 1A (11%) l/l (100%) QM,
First incidence(days) 584 491 359
Logistic regressionteat P~O.195 PmO.181 P-O.301

Squamous Cell Carcinoma

bverall rate 4fio (8%) 550 (10%) 6/50 (12%)
Adjusted rate 51.6% 33.1% 40.9%
Terminal rate 26 PQw o/l tQm Q@
First incidence(days) 370 422 288
Life table testb PCO.@lx P<O.oOl Pc0.001
Logistic regressionteat P =0.013 PW.017 P=OAMl

Squamous Cell Papilloma orI Squamous Ceil Gxwcinoma

Overall rate : wo%(2%) 3/SO(6%) 9BQ (18%) 10/50(2m)
Adjusted rate 21.4% 100.0% 52*8%
Terminal rate 0845{Q%j l/9 (11%) In (100%) OM
First incidence(days) 652 584 401 288
Life table test P*O.o65 P<O.QQl P<O.OOl
Logistic regressiontest P=O.195 P~O.004 P=O313

t Number of lesion-bearinganimafs/tnic~pically *mined or number 6f a~irn?~~n~~~

Beneaththe dosedgroup incidenceare the P vahte4correspondingto pakwisecompatisensbetweenthe controls and that dosed
group. The life table anafysisregardsneopiasmsin pnimais dying ,prior to tqminal~~killas being(directly or indirectly) the causeof
death. The logistic regressiontestsregard~ihesef&ions as.nonfatal. A lower incident in a desegroup is indicated by N.
i Kaplan-Meierestimatedneoplasmincidenceat the end of the study after adjustmentfor iatereurrentmortality
Obsetvedincidenceat,terminal kill
e Not applicable;no neoptasmsin animal group
RWUltS 71

SWd$ aP l,L%Butadime

Overall ratea l/so (2%) 4/48-(W) 3/42 (7%) 6148(13%) 7m (19%)
Logistic regressionteatb P+0.174 P‘=0.179 P+I323 P=O.O02

Overall rate 26m (S2%) 20150,(40%) 13/so (26%)
Adjusted rate’ 879% 94.3% izc?f”“’ 100.0%
Terminal rated 6rE’(67%) boo l/l &tO%)
First incidence(days) 440 344 iti
Logistic regressionteat P<&oOl P<t?.Wl P<O.#l P=O.O#

Overall rate o/so(0%) 2/50 (4%) 4f50 (8%) 2/.50(4%) o/50 (0%)
Adjusted rate 0.0% S‘6% 38.8% 51.5% 0.0%
Terminal rate ms (0%) o/9 (0%) l/S (zo%) ofi (@%I O/o
First incidence(days) . ..e 510 , 567 441
Life table testb PmO.182 ikO.001 P =0.028
Logistic regressiontest PpO.397 P=O.W P =0.190

Adenoma or Carcinoma
overall rate 6150(12%) 27,50 (54%) 2360 (46%) ,30/50 (60%) 13/so (26%)
Adjusted rate 14.8% @L3% 100.0% 100.0% 100.0%
Terminal rate zf35 (6%:) 519(67%) 515(100%) l/-i (200%) O/Q
First incidence(days) 543 440 410 344 306
Life table teat PCO.001 P<O,aOf P<O.Ool P<0.001
Logistic regressiontest P<O.OOl P-zzcOi001 P<O.ool P==O.O46

t Number of lesion-bow&g animals/ttumber.ofar&na!s microsayricalIy examinedor number of @mais necropaied

Beneath the dosed group incidence are the P kalues.wrreapondiogto paimise kaparisorts ,beWeenthe controls and that dosed
group. The life table analysisregardsneaplasmsin animals dying prior to terminal kill as being (directly or indirectly) the causeof
death. ‘Ihe logistic regressiontes%regard these lesionsas nonfataL
i Kaplan-Meier estimated neoplasm h&e&e at the end of the study after adjustmentfor inter@rrent mortahty
Obswved incidenceat terminal hill
e Not applicable;no neoplasmsin ammat group
72 1,3-Butadiene, NTP TR 434

Kidney: Renal tubule epithelium focal hyperplasiaor greater than the incidence in the controls by the life
adenoma occurred at low incidencesin each of the table test (Table C3a).
stop-exposure groups (Table 24). Although the
incidencesof adenoma in the exposedgroups were Brain: Malignant glionqs occurred in the brain of
not significantly greater than the incidence in the two male mice exposedto 625 ppm for 13 weeks and
controls by the logistic regression analysis, renal in one male mouseexposedto 625 ppm for 26 weeks
tubule adenomasare rare spontaneousneopIasmsin (Ta,ble Cl). In addition, malignant neuroblastomas
untreated male mice. The incidence of renal tubule were seen in two males exposed to 625 ppm for
neoplasms in NTP historical control mal,e mice 13 weeks (Table Cl). 1 of the neoplasmsoccurred
is l/571 (Table C4). The small numbers of renal in the anterior or olfactory lobe of the brain.
tubule adenomas in male mice exposed to Gliomas and ~euroblastom~ are rare spontaneous
1,3-butadiene are considered to be reiated to neoplasms;none havebeen observedin 574 NTP his-
chemical administration because of their rare torical c$trof male m&: For this reason,both the
Occurrencein historical controls. gliomas and the neuroblastomas are considered
related to.chemical administration.
Zymbai’s Gland: Carcinomasof the Zymbal’s gland PreputialGland: Preputial gland carcinomasoccurred
were seen in one male exposed to 200 ppm for at low incidencesin eachof the stop-exposuregroups
40 weeks, two males exposed to 625 ppm for but none we$eseen in the controls {Table 25). An
13 weeks, and two males exposed to 625 ‘ppm for adenoma was seen in one male mouse exposed to
26 weeks;an adenomawas seen in one control male 625 ppm for’13 weeks. ,‘I%e combined incidencesof
(Table Cl). The incidence of ZymbaI’s gland neo- preputial gland adenomaor carcinoma were signifi-
plasms (adenomas or carcinomas) in the group cantly increasedin the groups exposed to 312 ppm
exposed to 625 ppm for 26 weeks was signifkantly for 52 weeks or 625 ppm for 13 or 26 weeks.

Kidney Lesions in Male Mice in. the Stop-

d PPm 2m wm 625 ppm 312ppm 625 ppm

(40 week) (13 weeks) (SZ weeks) (26 weeks)

Renal Tubule: Hyperplasia

Overall ratea 2m (4%) 4/48 (8%) l/50 (2%) t/49 (2%) z/50 (4%)
Logistic regressiontfxi.tb P-Q.163 P-0.678. P==Q.O13N P=0.429N

Renal Tubule: Menama

Overall rate 4/48 (8%) r/so (2%) J/49 (6%) l/SO (2%)
Adjusted rateC 17.4% E4.3% 27.8% 6.3%
Teminal rated o/Q (0%) w @w wl (0%) O/o
First incidence(days) 516 707 539 440
Logistic regressiontest F==0.073 Pd.273 P=O,O75 P=O.731

a Number of lesion-bearinganimals/nutiberof animalsmicroswpically examined

b Beneaththe dosedgroupincidenceare the P values&resp&ding to pairwisecomparisonsbetweenthe controls and that dosedgroup.
The logistic regressiontestsregard lesionsin anima%dying prior to terminal kill as nonfatal. A. lower incidencein a dose group is
indicatedby N.
“, Kaplan-Meierestimatedneoplasmincidenceat the end of the sttidy after adjustmentfor intercurrentmortality
Observedincidenceat terminal kill
e Not applicable;no neoptasmsin animal group
74 1,5;Butadhae, NTP TR 434


All Organs: Lymptweytic Maiignaat L~pb~~

Overall ratd 6isO112%) 17/B (34%)
Adjusted rateb 26.7% 35.8%
Terminal ratec l/9 (11%) o/s (0%)
Fimt incidence(days) 208 169
Life table tmtd P=O.ooS

Ail Organs: Malignant Lymphoma {Mixed or NOS)

Overal rate 2450(4%) s/50 (lQ%)
Adjusted rate 7.8% 34.8%
Terminal rate Qf9 (0%) l/s (20%)
First incidence(daya) 514 251
Life table teat P==0.117

All Organs Malignaet Lymphoma (Lymphocytk, Mixed, or NOS)

Overall rate 8/50 (16%) nlso (44%)
Adjpatedrate 32.4% 58.2%
Terminal rate l/9 (11%) 16 ml
Elrst incidence(days) 208 169
Life table test P=O.Qol

Stomach (Forestomacb): Squamoirs Cell Cwcbma

OkraIl rate 4/M (8%)
Adjusted rate 51.6%
Terminal rate 215(4095)
Eiirstincidence(days) 370
Life table teat P=0.019
Logiitic regressionteatd P&.031

Stomseh (Fotwtomach): Squamous Ce& Papitloma or Squ~mow CA{ ~~~i~~~

Overall rate 3/SQ (6%) 7/50 (14%)
Adjusted Me 21.4% 56‘6%
Terminal rate l/9 (11%) 26 (Jo%)
Pimt incidence(days} 584 327
Life table test P=Q.o4S
Logistic regremionteat P=O.099

B Number of neoplasm-bearing animals/numberof at$malsnecropsied

Kaplan-hfeierestimatedn,r?oplasm incidence:at the end of the study aftor adjustmomfor intercurrentmortality
’ Observedincidenceat termina\ hih
d Beneaththe dosedgroup incidenceare the P value correspondingto p&+&e comparisonsbetweenthe controk and that dosed
group. The life table analysisregardsneoplasm<in anim& dying prior to terminal ldlk as bcinfr:(directlyor indirectly) the causeof
death. The logistic regre&on yzstsregrkd theselegionsas nonfatal.
e Not applicable;no neoplasmsin animal group
Resuits 75

312 ppm 625 PPol

(52 wl5eka) (26 weeks)

AU Organs: Lynphocytic &fd@~ar L$rqhoma

overall rate” MO (8%) 3ofSO@ I%)
Adjusted IWeb 100.0% 81;5%
Terminal rate’ In (laoa) a/o
First incidence(days) 239 159
Life table testd P4.001

Al1 Organs: Malignant Lymphoma @&ced or N?S)

Overall rate 4/S@(8%) 3/M (6%)
Adjustedrate S8‘@ % 43.3%
Terminal rate ;gPW on,
First incidence(days) 251
Life table test P=0.244

All Organs: Malignant Lyanphoma (Lpmphorgtic, Mixed, or NOS)

Over& rate 8/50 (16%) 33/B (66%)
Adjustedrate 100.0% 89.5%
Terminal me l/l (l@@ql Ofi
First incidence(days) 217 159
Life table test P<O.Wl

Number of neoplasm-bearing anit@s/nuWer of @mais aramiQ,ed.Derrom~naf~& num~r of animalsexaminedmicroscopically

for lung and preputial gland; for otk t&w%, dmotiiaa~or is ~ymber d &+I& wcropSe&
Kaplan-Meierestimatedneoplasmincidenceat the end of the study after adJustm!Znt for interwrrent mortality
Observedincidenceat terminal kill
Beneaththe dosedgroup incidenceare ibe P vahteacorrespondingto pairWe mmp@sbnsbe-n the controls and that dosed
group. ‘I%elife table analysisregards-hdopasmsln‘animalsdying~prkxto terolinai $iil as being (dkec~lyor indirectly) the causeof

by the nPoly-3wquanta1 response test @ iler and neoplasm rates are given as the shape parameter
Portier, 1988, Portier and Baibr, 1989). The values in Table “;5$,,Far appro~rn~~~y half of the
survival-adjustedrates for those Naples showing neoplasms ~~i~t~~ the dose-responsetrend is
chemical-related increases are summarized in consistent with a’ linear model (Le., shape
Tables 28 and 29. The effect of this, additional parameter = 1). In most of the instancesin which a
analysiswas to detect as signifkant certain neoplastic departure from. &ear&y was evident, the shape
responsesthat were not detected by th9 logistic parameter was s~~~~~nt~y less than a value of one
regrmion analysis in mice expsas& $0 GSppm., The (liver ~~~~~rns,,~ males,mammary gland neoplasms
results of the “Foiy-3* quanta1responsetest do not in- female&+andharderian gland neoplasmsand lung
change the overall interpretation of these studies. neoplasmsin males and females) implying that the
Mortality-adjusted dose-responsecurves for neo- dose-r~~~ curve is very steep in the low-dose
plastic lesionsindud by‘l,%butadiene are shown in region. Qkly for ma%gnantiymphoma in male mice
Figures6 (males) and 7 (females). and h~~a~~~a~~a of the heart in female mice
was there evidens&of a “threshold-like”dose-response
The results of fitting a modified Weibull mddel curve, i.e., a curve in which the shape parameter is
(Portier et aL, 1986) to the Poly-3 surviv&l-adjusted signifkantly greater &an one.
76 l,SBUsdiene, NTP TR 434

Survival-Adjusted Nehplasm Rates for 36423F,Mice in the %%mr ~~~~~~ Btudks of 1,3-Butadler+

0 ppm 625 ppm MPpm 623 ppm. 200 @pm 625 ppm Parameter


Malignant lymphoma 9.0 4.4 15,o x5.3 7.a 97.3** 10.aooAA
Hiitiocytic sarcoma ‘0.0 0.0 10.0 ra9* 24.P 50.7“ 0.716
Iiemangicaareema 0.0 0.0 2.6 13.5" 64.3** 52.9*' 1.033
Adenomaor carcinoma 47.5 49.0. 44.9 74.2: 87.8** 45.1 0.457-
Adenomaor carcinoma 13.5 15.2 22.4 50P' 8Q.6" 64*2**b 0.64700
Papilloma or carcinoma 2.3 0.0 0.0 2.7 z&7* * 53.5**b 1.413
Adenomaor carcinoma 44.6 48.2 65.2*b 61.6 85,9"* 61.2 0.37400
Preputial gland
Carcinoma (0.0 O*O 0.0 0.0 189Q' 0.0 1.207


Malignant lymphoma 13.1 27.2 27.5 20.2 4&i* t35,s* 1.690
Histioeytic sarcoma 6.5 4;4 17.2 11.8 34.0"" 35‘s* 0.583
Hemangiosarcoma 0.0 0.0 0.0 3.1 719**
I 83.4+* 1.293r
Adenomaor carcinoma 8.8 33.0. 469' 6&l'* 815** 82.4'. 0.374011
Adenomaor carcinoma 17.5 22.7 17.4 4x.2* 309** 58.0eeb 0.5720
Papilloma or carcinoma 0.0 0,o 7.8 6.1 22s**b 82.6" 1.182
Menoma or carcinoma 33.3 3Q.3 36.4 51.4 649 21.7 0.315
Granukxa ceil tumor,
benignor malignant 2.3 0.0 2.4 26.3- 41.1** 46.Pb 0.777
Mammary gland
adenoacanthoma 0.0 4.5 10.2*b 32.6" 56.4** 66.8.. 0.645OQ

* Significantlydifferent (P ~0.05) from the control group by the Poly-3 quanta1responseteSt(Pwlier and Bailer, 1989)
** PeO.01
A Shapeis significantlygreaterthan 1, P<O.U5by iikeiihood ratio teat
a Shapeis significantly less than 1, P<OA%by likelfhood ratio test
00 P-CO.01
a Neoplasmrateadeterminedby PO&3 quarttatresponsemethod
b Not significant by the k&tic regressiontests
MatignantIymphoma 9.0 24.1 56.1**
Hiitiocytic sarcoma 0.0 16.3’ 9.4
Hemangiosarcoma 0.0 47-I” 30.9”
Adenomaor carcinoma 47.5 88&’ 89.5*+ 88.0** 87.2**
Adenomaor carcinoma 13.5 72.1*+ 82.W’ S&6** 76.5**
Papillomaor carcinoma 10.2 z&.7** 39.2** 60.7**b
Adenomaor carcinoma
Adenomaor carcinoma

* SignifkantIy different (PcO.05)from the wntrokgroup Iy the PoIy-3quanta1responsetest (Po@icrand Bailer, 19S9)
” P<O.Ol
l Neoplasmratesdeterminedby J’oty-3qua@aIresponsemethod
Not significantby the Iogistic regressiontests
Mortality Adjusted Tumor Rate (X)
M ;5
0 0 % t3 i3 0

Positive results were ab~n~’ with 1,3-butadienein

1,3-Butadiene was mutagenic in $z&me~la tphi- cytogeqdc twits with. ~a.m~iian cells In vivo (Tice
mu& strain TA1535 when tested Bs a gas in a et a& t937). ~gnifi~~t: increasesin the frequencyof
sealeddesiccatorchamber,with and without Aroclor chromosomal a~~t~o~s (‘TableD4) and sister
1254-inducedmale Sprague-Dawley rat or ~Syrian chromatid ,~~~g~, frabfc DS) were observed in
hamster liver S9 (Table Dl). The positive response bone marrow ceQsof-ma@mice exposedfor 2 weeks
observedwith TAX535 in the absertccof exogenous to l~~b~~e~’ (6.25 to 625 ppm in air). For both
metabolic activation was unexpectedand may,ja,fact, tests, the trend ana@& were significant; both the
be an artifact of the exposureprotocol, To eff&iently mid- andb~~~~-~~irn~~ showedincreases-insister
conduct these desiccator exposures,ali plates to be ChFcmytid ex&ange+ wh$le,only the high-dosemice
exposed to a particular con&ntration of l$-buta- had.&evatod lev@ sof ~h~~rno~~l aberrations. In
diene, with or without S9+wefi: housed together in a addition, ceil cycle time @assignificantly,lengthened
single desiccator and treat& ‘~~rn~~~~~ly. as dose+of ~~~bu~d~~~ ‘wereincreased,as indicated
Previous investigations demonstrated that such an by the average g~~~ti~~ time measurements
arrangementproducedpositive responsesin cmtures (Table D5). Petipherat‘Mood smearsprepared from
that did not contain $9 activationenzymes,whereas these same; animals.:(eqmsed to l$butadiene by
removingthe S9-containingplatesfrom the desiccator iirhalation for 2,weeks) r&Wed significant increases
and treating only those ct@uresthat did -not contain in rn~c~~uci~t~ ~~yehr~~t~c erythrocytes and
S9 resulted in no increasein mutatiotis (de hi&ester no~~h~Q~t~~e~~r~~ (Table W ). Elevations
et at, 1980). The induction of mutations in the in the ~~~en~ of: micronucleated polychromatic
cultures that did not con&n S9 ‘was believql to be erythrocytes (a -me&ure of acute exposure) were
caused by the formation of a voE&e mutagenic obsmkil iethe mgd- and high-dosemice, whiie only
intermediate in the S9-containing plate& that the high-dose mice show a statistically significant
migrated to the plates without S!3. No m tfxatgenic increasei~,rni~~~*cl~ normochromatic erythro-
activity was detected for &3-butadiene in- strain of ceils, the trend analyseswere
TAloO, TA97, or TA9S under the same conditions. of erythropoiesiswas increased
in expo@ mice; ~~~c~l~~ly at the 625 ppm level, as .the increasein the percentageof poly-
No mutagenic activity was observed in the. mouse chromatic erythrocytes in the total erythrocyte
lymphoma L5178Y cell assay, with or without ~pu~~t~o~ in the ~ti~h~l blood (Table D6). ThS,
Aroclor $254.induced male Fischer rat !&er S9 along with the increasein average generation time
(Table D2; McGregor et.&,’ 199b). The m$imum (Table DS), indicates@&ular (bone marrow) toxicity
dosewas 30% in air (v/v). One possiblefactor in the induced by ~,3~~u,~~~e~~The frequenciesof micro-
lack of mutagenic activity is the low soiubihty bf nucleated ~~~hro~atic erythrocytes and normo-
l&butadiene in the cell culture me&m, which niay chromatic erythrocytes&vcrealsoscored in -peripheral
havepreventedadequateexposure. 1,3-Butad@nedid blood samplesof male and female mice exposedfor
not induce a significant increase in the, number of 13 weeks (Table D7j and 15 months (Table DS) to
sex-linkedreozssivelethal mutations in germ cells of 6.25 to- 625 ppm l&butWene; both exposureregi-
male Drosophila melanogasterexposedby $&Wation mens produced positive results in both sexes. Also,
to 360,OQO ppm in air (Table D3). These negative the percentage of polychromatic erythrocytes in
resultswith 1,3-butadienewere somewhatsurprismg, scd for I$ months to 1,3-butadiene
given its demonstrated activity in mammahan cells was elevatedat the two highestconcentrationstested,
in vivo. which pro&n@ a positive trend (Table D8).

1,3-Butadieneis produced in. large volumes fog use presentedin this Technical Report, were conducted
mainly in the manufqcture of synthetic rubber and at ~n~~t~tio~~ rangins from 6.25 to 625 ppm
thermoplastic resins. Previous long-term inhalation l&butadiene, - The exposure level of 625 ppm
studieshaveshown that l,&bytadiene is.carcinogenic wrrespon~ to-the low-exposurelevel in the previous
at m&pie organ sites in’ Sprague-Diiwley.rats inhalatidn. studies ik mice, and 6.25 ppm is two
(IISRP, 1981a;Owen et d, 1987) and B6C!3F1mice orders of ma~l~~de lowei. A preliminary accountof
(NTP, 1984,Huff et aL, 1985): The 2:year st@ i& in the results of these $&dies has been reported
rats, sponsored by the Int~rnaticqal Institute of (Melnick et al, 1990b,c),
SyntheticRubber Producers(JISRP),were coqducted
at exposureconcentrationsof 1,000 and 8,~~ppm. Expo-sureto X,3-butadierrefor up to 2 years had no
The highest exposurelevel was lim&ed by &e safety apparentadv&rsqeffect on body weight gains for male
requirement of being below SO.%of the explosive or female mice; however,’ survival was reducedin all
limit of l$butadiene in air, while the ~~~.pprn groups exposed to con&ntrations of 20 ppm or
concentrationwas selectedbecauseit representedthe hiiher. As in the,previous studies, lymphomas that
occupational exposure stan&rd for this chemical. occurred early, prior to 15 months, were the major
The NTP usually conducts lon@ tgm studies in causeof death for male &ad female mice exposedto
F344/N rats and B6C3F, mice; however,becausethe 625 ppm 1,3ibutadiene. .T-cell lymphoma is caused
by exposureto 1,3-bu~die?e(Irons et aL, 198%Irons,
IISRP studies in rats were in the time of
chemical selection, the NTP studies were limited to 1990). In the present studies, most butadiene-
long-term evaluations of 1,3?butadieneexposureiA induced lymphomas were ‘well differentiated and
mice. ~pho~ic~ andappi=are&l to originate in the thymus.
After month 19, there was a marginal but statistically
The exposure concentrations selected for the NTP significar‘ltincreasein l&t$ocytic sarcomas.Addition-
studies,625 and 1,250ppm, tiere basedon iricreased ‘ally+other histological typesof lymphoma (malignant
mortplity and decreasedbody weight gains in mice mixed and ~iignant undifferentiated) commonly
exposedto concentrationsof &Sod ppm or higher for associatedw&-hthe spontaneous-lymphoma of aging
14 weeks(NTP, 1984). The carcinogen&&ystudiesin B6C3Fi mice were obseqed in all remaining groups.
mice,designedto last for 103 weeks,were terminated Lymphocy$ic l~mp~omas were analyzed separately
after 61 weeb becauseof reduced survival-due to from W tiocytic sarcomasrand all lymphomas,because
malignant neoplasms inv@ ing ~uitiple organs at they provide a clearer response of 1,3-butadiene-
both exposureconcentmtions,Malignant lympbomas, induce&hematopoietic cancers.
which appeared-tooriginate in the thymiis and were
observedas early as week 20; were considered:tobe The incidenceof be~a~~~osar~masof the heart was
the major cause of e&rly d&h, ,while hem~~gi~- iricreased in ‘male mice; exposed to 62.5, 200, or
sarcomasof the heart, an uncommon nec#t~m in 625 ppm an& in female mice exposed to 200 or
untreatedBdCJF, mice (none.occurre&iri573control 625 ppm. ,In qidition, one male exposedto 20 ppm
malesor 558 control females‘in recent NTP stGdies), and one ~erna~e~exp~~to 62.5 ppm were observed
were the second major cause of peath. Th& inci- to have this uncommon,endotheiial cell neoplasm;
dcncesof primary neoplasmscausedby exposureto the occurrenceof these Tare sarcomasat the lower
1,3-bu~dieneare shown in Table 1, W w these concentra-tionswas also Xkely due to 1,3-butadiene
studieshad been terminated early and ,dose-response exposure. increasedincidemzesof endothelial hyper-
relationships for various lesions were sometimes plasia in the h&art at all exposure concentrations
u nclrar (e.g.,hemangiosarco&s of the heart ‘ia male probably iepresent ~re~eoplastic changescausedby
mice), a secondset of long-term inhalation studiesof 1,3-butadiene.
1,3-butadiene in mice was, performed to better
characterize the carcinogen@y of this important enceof beman~iosarwmasof the heart was
industrial chemical. The latter studies, wbi@ are greater 19 male mice osed to 200 ppm than in
82 I&Butadieae, NTP TR 434

male mice exposedto 625 ppm. The lower incidence likely that exposureconcentrations below 6.25 ppm
in males receiving625 ppm was probably due to the would also cqse cancers in mice. The reduced
early and extensive induction of lym@xzytic incidence of lung neoplasms in mice exposed to
lymphoma, which resulted in a significantly reduced 625 ppm compared with the incidence in mice
number of mice at risk for the ~ter~velopi~~ heart exposedto ZOOppm is attributed to the high rate of
hemangiosarcomas.The me4lian survival time wqs early deaths,due~tocont?eting risks of lymphocytic
about 40 weeks for male mice exposedto 625 ppm arbor in -female mice exposed to 625 ppm
and about 70 weeks for male mice exposed to {Figure 8). The time-to-neoplasm detection of
200 ppm. The effect of competing risks of early aIv~l~r~ron~h~~lar ’neoplasmswas slightly shorter
occurring lethal thymic lymphomas on the develop- for ‘anirrudsexposedto 62S ppm than for animals
ment of hemangiosarcorrnts of the heart is ~evident exposed to 200 ppm; however, because ail female
from the plots of the cumulative d~th-~th-~pl~rn mice exposed to 62S ppm 1,3-butadiene died by
rates of thae neoplasmsagainstthe number of weeks 65.weeks,the:final ~n~id~~~ of this later developing
on study for male mice exposedto 2QOor.625 ppm and rarely l&ha1 neoplasm was lower than that for
1,3-butadiene(Figure 8). In: the 62s ppm~grtiup, the femafe mice exposed to. XK, ppm. Increased inci-
incidence of early lymphoc$ic lymphoma was very dencesof alveolar ep~t~el~l hyperplasia in exposed
high (67%) and the indigenceof he~n~~~~~m~ male and female mice Frobably represent pre-
of the heart was low (5%); however,in the 2m ppm neoplastic changescaused by 1,3-butadienein the
group, the incidenceof early.l~~h~c lym~homa lung
waslow (4%) and the incidenceof he~~gi~~r~rnas
of the heart (42%) was much higher than that,in the IncrF& incidencesof neoplasmsof the forestomach
625 ppm group. Furthermore, in’male mice that died (sqtmmousceli papillom~sor carcinomas),mammary
early after exposureto 200 or 625 ppm 1,3-butadiene,gland ~~~i~omas, adenoaoanthomas, and malignant
the incidenees of hern~~~~r~mas of the heart mixed tours), ‘ovary(benign or malignant granulosa
were nearly equivalent for the first 6S weeksIof thecell tumors), liver (hepatocellular adenomas or
‘carcinomas),and other ‘organ sites identified in the
study. After that time, a high incidenceof hernangio-
sarcomas of the heart (approximately Sd%) was first studieswere again observedin mice exposedto
observedin the 200 ppm group, whereasthere were l&bMadIene, Additionally, the harderian gland and
no surviving animals in the; 62S ppm group, Thus, preputial gland were identified as sites of
for male mice exposed to 1,3-bWadiene 1,3-butadiene-indu~ neoplasia. Increased inci-
concentrationsbelow 625 p@m,the doseresponsefor dertcesof proliferative nonneoplasticlesions in these
hemangiosarcomasof the ’heart is more clearly organs, including epithelial hyperplasia of the fore-
demonstrated. The impact of early mortality .on the stomach, mammal gland hyperplasia, germinal
expressionof later-developing neopl3smsis largely epithelium and granulosa cell hyperplasia of the
accounted for in the mortWy-adjusted neoplasm ovary, and by~e~~l~i~ og the harderian gland, proba-
rates shown in Tables-30and 31 for each ~~plasm bly represent~~~~pl~tic changesat thesesites. In
induced by exposureto 1,3-hutadiena control female,mice, no neoplasms,or only benign
neoplasms~were observed in the harderian gland,
The incidence of alveolar/bronGhiolWtmqlasms in forestomach, and ovary; however, in female mice
male mice was increased at concentrations of 62.5 exposedto l&butadien&, malignant neoplasmswere
and 200 ppm compared,to that of the controls. In obs at ~eaehof these&es. The greater tendency
female mice, the incidence:of alv~~r~ron~io~r to malignant neopla$a in mice exposed to
neoplasmswas significantly increasedin all exposure l$butadiene further demonstratesthe strong carcin-
groups comparedto that of the controls. Thus, even ogenic potencyof this *emical.
at a concentration of 6.25 ppm, 1,3-butadbne is
carcinogenic to B6C3F1 mice. Furthermore, in The coxtdusion that the marginally increased
control female mice, all of Xhe alv~lar~ro~~hi~lar incidenti of he~,~t~~~lar neoplasmsin male and
neaplasmswere adenomas,whereas in female mice female mice were related to chemical administration
exposed to l,Sbutadiene, inclu@ tg the Ei.25ppm is strengthetiedby-the ‘detection of activated K-ras
exposurelevel, alveolar/bronchiohr carcinomaswere oncogenes with a s~e~i~~ codon 13 mutation
observed. Becausethere was no exposure level at in liver neoplasms obtained from mice exposed
which a carcinogenicresponsewas not induced, it is to 1,3-b~tadie~e(Go&row et al., 1990). Activated
Discussion and Concl~ons 83




84 1,3dhta@ene;~NTP TR:434

Thus, for the imjuctioon of thymic

; the cx+XWation of l$;butadiene is a
ter ~n~rib~tj~g faktor than is the length of

.~e~~‘t~b~e cell adfenoma$

were observedin 9 of the
2~,-~e mice: it? the stop&sur& groups; ,the

trciatexfEW3FI tii& (his-

than 0.2% in recent NTP
ion of late-developing renal
,the stop;;expoSuregroitps is
the increased‘&viva1 of t&se qni-
to ,&kie groups’of male mice *at
ta, similar cpncehttitkns of ’
ro$h?ut their lifetime&w for-up to

nd extensive development of ,i
reqxsnse. to 625 pp” -1
the‘number ‘of
Discussion aad conchlsions 85

in.males and at levelsof 200 or 625 ppm in females. w~~nfr~tio~ as low as 6.25 ppm, the lowest
These changeswere not accompa S~~j~~NC ~n~~~ion studied. :
increasesin retic&cyCe counts or ~ucncy of
polychromatic erythrocytes in periI$eral blood; The ~~h~n~ of b~~d~~indu~ carcinogenic&y
however,there was a statistically significant incree is not khowiu hcwever, oxidative intermediates of
in the percentageof erythrocyteswith HowelLlolly 1,3-butadiene~~t~~~~~~tion, 1,2-epoxy-3-butene,
body inclusions, Other hemstologic ~h~~~-~~ tame, or’ a combination of these
by exposure to 625 ppm Id-butadieue were an berfroid, 1982),are likely involved.
increasein mean erythrocytevolume and,au in;nease These metawlites .are . direct-acting mutagens in
in mean erythrocyte hemoglobin. ~it~nally~ e Meester et al;, 197%
changesat other organ dtes, bono Wade et al, 1979j,w the elicitation of in v&o
and increasesin splenic lrnd hepat rnu~~eai~~ of l~-~~~ene appears to require
hematopoiesis were observed in metabolic activa+n. (de Meester et al;, 1980).
62.5ppm 1,3-butadiene. These fi ~rt~~~ore, tlrese ep&idea have been shown to
partial or poorly regenerative,ma induce local (a~pl’~~c~n site). neopiasms when
The mechauiirmof the anemia cannot be determined applied to the sk& of Swi& mice or -when adminis-
from the data available from these studio, however, tered to Swiss mice or Sprague-Dawley rats by
a mild megaloblascicanemia resulting &orn ineffec- subcutan~us injection (Van Duuren et aZ., l%R
tive erythropoiesis in the bone marrow cannot be 1966).
excluded.Tice et al. (1987j reported that expo!ure of
male B6C3F1 mice to 1,3+butadienefor 2 weeks The ~r~i~o~~ici~ stud&s of 1,3-butadiene in
causeda deereasein the number and rate of cfividing’, Spra~e~D~wIey ‘rats (IISEP, 1981a; Owen et aL,
cells in the bone marrow. Thus, in mice edtu 1987) and..l36GJF, mioe (NTP, 19% Huff et aL,
1,3&utadiene,hematopoiesiain Chebone marrow iS 19&f), ~cludiug the current studies, demonstrate a
suppressed,and younger, htrger ceils are probably speciesdifference ih ChksiC= of neoplasm induction
released into the blood from e~~rn~ull~~ sites.. and the magRiC~de of the dose-dependentresponses.
Consistent with this explanation, Thurmond er UL In add&ion, -‘in rir Viva genotoxicity studies,
(1986) observed extramedullary hemqtopoiesis in lo-buCadie~~ induced chromosomal aberrations,
spleensof male B6C?3F,mice exposedto l,W , p @ m sister chromatid excbauges,and micronuclei in mice
l&butadiene for approximately6. months. {~u~~i~~~rn et a&, 19$X$Tice d a& 1987), but not
in rats (~uu~i~~arn et,uL, 1986). Biochemical and
Testicular atrophy was induced in male ,B6QE1mice pharmacokinetio stud& have been performed to-
exposedto 1,3-butadieneconc&ttratIonsof 625 ppm d~term~e the ‘m~banism of neoplasm induction by
or above in the current studies and in previous 13:bu$diene and,tom an explanation for the
studies (NTP, 1984). In female mice to different Coxicand ~~inoge~c responsesbetween
1,3-butadiene for 9 months, ovarian atrophy of rats andmice. The possibility that the induction of
moderate severity was 0bseCve.din t nd thymic leghorns in B6c3F1 mice was a conse-
625 ppm groups; the ovaries of mice to quence of the expression of a murine leukemia
625 ppm for 9 months appeared normal, The rerroiiws has been considered (Irons et aL, 1987,
atrophic ovarieshad no identifiab @ , 1;98S;irons, 3990) Hoyever, the finding that Chymic
or corpora lutea. AfCer 15 mon t0 ~~~orn~, were induced by l&butadiene in NIH
1,3-butadiene, ovarian atrophy was observed at t%$ss mi& a strain that does not express the
exposurelevelsof 20 ppm and above. In female mice ecotropic murine leukemia viruses expressed in
exposed to 1,3-butadiene for up to =IJ, mice, de~o~$trat~ that these neoplasms
incidence of ovarian atrophy was,i wcrc producedinde~~de~tly of theseactivatedretro-
exposureconcentrations(625 to 625 ~~rn),~~ar~ vitih.
with controls. Even though ovarian atro.phyin the
6.25 ppm group was not observeduntil late in the 1~ V&J alkylation of #v@r DNA was equivalent in
study, when reproductive senescencewas Jprobably l&X3F1 mice and W istar rats exposed to
occurring,the dose-response data clearlyestablishthe &3-butadiene (Kreiling el al;, 1986); however,
ovary as a target organ of 1~.~u~diene toxicity at expect&l reaction products betsveenguanine and
84 l&MWadkne, IWP~TR 434

l,&epoxy-3-buteneor diepoxybutanewere detectedin

liver @MA fi-om’exposedm@%,but~not,.from exposed
rats (Jditto et al;, E&9). ~h~.s~~d~~~a~~-n~~
onthe :dose-responses for DNA”addjuct
the majda larger. ~oqganspf:- 1,3;3-b%%ad
wcimgetiici4y in rats and$cean~ on the nature of
the butadienedetived material bound to. rat liver
Discussion an& cogcluslons 87

additional concern for the ~r~i~ogeni~9 of 625 ppm - pro~~e a b&&x characterization of the
1,3-butadieneto humans, particularly becausethese ~n~ntrat~on-de~ndent responsesfor l,3-butadiene-
results correspond to increased incidents of induced neuplasmsand mmneoplastic ksions. The
lymphomaobservedin mice exposedto 1,3+butadiene. present studies ~~~~~ the clear evidence of
The detection of K-ras oncogenes in neoplasms carcMger&ty of l,3-butadiene in male I%C3F, mice
induced by ll$butadiene adds further relevance to based on increasedincidenczs of neoplasms in the
the potential carcinogeniciry of 1,3-butadiene in ~e~~o~~eti~ system,heart, hmg, forestomach,liver,
humans, because IS-ras is the most commonly harderia~ giaud, preputial gland, brain, and kidney.
detectedoncogenein human cancers(Bos, 1 There was dear m&hce of carcinogerticity of
1,3-butadiene in female B6C3F, mice based on
incrwsed incident of neopliasms in the
h~ato~i~ti~ system,heart, Bung,forestomach,liver,
CONCLUSIONS harderian gland, ovary, and mammary gland.
The previous inhalation studies of 1,3-butadieuein
male and female B6C3F, mice providedcieur et4dHzce Low incidencesof intestinal.carcinomasin male mice,
of carcinogenk$+ at exposureconcentrationsof 625 Zymbat’s gland carcinomasin male and female mice,
or 1,250ppm. The present inhalation studies - and renal tubulle adenom~ and skin sarcomas in
2-year exposuresof 6.25, 20, 62.5, 2OQ or 625 ppm female mice may also have been related to
or shorter duration exposures of 200, 312, or administration of 1,3-butadiene.

l Explanation of Levels of Evidepce of Carcinogenit: Activity is an page 11. A summaxy of r review comments end the public
discussion on thii Technic+ Report appew on page 13.
l&Butadiene, NTP TR 434

American Conference of Governmental lCndustria1 Bolt, H.M., Filset, J-G., and Stormer, F. (1984).
Hygienists (ACGIH) (1986). 7&~&o&, Limit .J%lues Inhalation ~~a~~~~t~~ based on gas uptake
and Bidogkd Exposure Indices ftir 1P&$-1987, studies. V. ~rn~~a~~ve pharmacokinetics of
pp, 68-70. Cincinnati, OH. ethyl&t: and 1;3-butadienein rats. Arch. ToxkoL 55,
Amoore, J.E., and Hantula, E. (1983). Odor as an
aid to chemical safety: Odor thresholds compared Bond, J.A., Dahl, A%, Henderson, R.F.,
with threshold limit values and voladlilies fur 214 Dutcher, JS., ~~~der~~, J.L., and Birnbaum, LS.
industrial chemicals in air and water dihttion. (19%). Species differences in the disposition of
J. AppL Toxicok 3,272-290. inhaled butadiene. Toticol, Apple PhannacoL 84,
Arce, G.T., Vincent, D.R., Cunningham, &J.,
Choy, W.N., and Sarrif, A.M. (1990). In ‘ti&@ and Bond, J.A., &hi, AR., Henderson, R-F., and
in viva genotoxicity of 1,3-butadieneand mem,bolites. Birnbaum, I.&X (19g7). Species differences in the
Environ. Health Perspect. 86,75-78; d&rib&x& of inhaled butadiene in tissues. Am Iki

Armitage, P. (1971). St@i.dcai Methods $a Medical Bond, J.A., Martin, Q.&:Birnbaum, L.S., Dahl, A.-R.,
Research, pp. 362-m. John Wiley and Sons, Melnick, R.l,., and Henderson, R.F. (1988).
New York Metabolism of l,~bu~di~ne by luttg and liver
microsomes of rats and mice repeatedly exposedby
Auerbach, A.D., and Wolman, S.R. (1979). inhalation to l~-bu~~i~~e. Toxicol, Lett. 44,
Carcinogen-induced chromosome breakage in 143-m.
chromosome instabihty syndromes. Cartrz@’
Gener. Boorman, G.k, MontgQme~, C.k, Jr., Eustis, S.L.,
Cyrogenet.1,21-28. Wolfe, M& M~~~~e~l, EE., and Hardisty, J.F.
(1985). Quality assumtiee in pathology for rodent
Auerbach, kD., Weiner, M.A., Warburtoq D., ~~n~~en~~i~ studies; In Handbook of Carcinogen
Ycboa, K., Lu, L., and Brortmeyer, HE. (1982)~ Tesr&g @A. Miiman ;md E.K. Weisburger, Eds.),
Acute myel.oid leukemia as the first hematologic pp. 345-357. Noyes P~~li~tio~, Park Ridge. NJ.
manifestation of Fanconi anemia. Am. J, &imatd
12, 289-300. Bos, J.L. (T989), Ras oncogenesin human cancer: A
review. Cam&r Res. 49, ‘46824689.
Bailer, A.J., and Portier, C.J. (19&3). Effects of
treatment-induced mortality and/ tumor-induced Canter, D.A., Haworth, S., Mortelmans, K,
mortality on tests for carcinoge&&y in small Zeiger, E., Lawior, T;, and Speck, W. (1986).
samples. Biometrics 44, 417-431. Comparative rn~tage~~~~~of aliphatic epoxides in
Sa~#ne~~, M&tat. Res. 172, 105138.

MI, H.M., Schmiedel, G., Filser; J;G., CUpenter, C.P., Shaffer, C.B., Weil, C.S., and
Kolxhiuser, H.P., Lfeser, K, Wistuba, D, and Smjrh, HE, Jr; (X%4). Studies on the inhalation of
Schurig, ‘V. (1983). Biological activation of l:Ibu,tadiene; with a comparison of its narcotic effect
1,3-hutadkne to vinyl oxirane by fat liver microsomes with benzol, toluol, and styrene, and a note on the
ml expiration of the reactive metaholite by exposed elimination of styrene by the human. J. In.& Hyg.
rats. Cancer Res. Clin Oncol. 166, 112-116. * Ttxiccd 26, 69-78.

Checkoway, H., and Williams T.M. (1982). A J.K (1986). Logistic

hematologystuvey of workers at a sine-b~~diene ntal-tumor data from
synthetic rubber manufacturing plant. h I+ E&g ems. Fundam AppL
Assoc..t 43,166169. T&oL 6,444X

Choy, W.-N., Vlachos, D.A.., ~~~~~~~, NJ., Dinse, GE., and kos, SW. (1983). Regression
Arce, G.T., and Sarrif, A.M. (1986). Genotoz&ity of analyiii Of tumot~r p~~~~ data. AppL Statist. 32,
1,3-butadiene.Induction of bone Marrow micronuclei 23644%
in B6C3Ft mice and Sprague-Dawleyrats in V&O.
Enviro& Muta~ 8 (SuppL 6), 18.

Code of Federal Regulations (CFR) 21, Fart 58.

Cox, DR. (1972). Regressionmodels and life tables. lhwqs, T,D., Crane, and Rim, K.W. (1987).
L R Stat. Sot. B34,187-228. Mortality among,~o~ a butadiene &i&y. Am.
J. fnd Afed ‘12,311429-
Crouch,C.N., Pullinger, D.H., and Gaunt, I.F. (1979).
Inhalation toxicity studies with 1,3-butadieue- 2.3 rusick, D,, Mixoy, E,
Month toxicity studies in rats. Am. In& Hyg.‘Assoc. IL, Rosenkranz, I-IS.,
J. 4Q,796-802. Reproducibility of
1. Tests with
Cunningham, MJ., Choy, W=N., Axe, G.T.,
Rickard,.L.B, Vlachos, D.A., Kinney, L.A+ and
Sam* AM (1986). In vivo~ sister chiomatid (Suppl. 2), l-251.
exchangeand micronucleus induction studies with
1,3-butadienein B6C3F1 mice and ~p~~~-~awl~ le comparisons using rank
rats. Mutagemk 1,449~452.

Dahl, AR., Sun, J.D., Birnbaum, LS, Bond, J.A., Dunnett, C.W. (4955). A muitiple -comparison
Griffith, WC, Jr., Mauderly, J.L., M procedure-for ~rn~~~ seweral treatments with a
Sabourin,, PJ.,. and Henderson, RF. (@91). control. J, Am. S&s&,Y&+x~So, 1095-1121..
Toxicokineti~ of inhaled l$butadiene in monkeys:
Comparison to toxicokinetics in rats and r&e. Durchin, J, (1990). pi&g and ~anqlysisof the
ToxicoL AppL Piiarmacd llQ, g-19. ambient a& in the ,ar&- of .Port Neches, .Jefferson
cbBnw,,TB: Tina& Sampling and Analysla
de Meester, C, Poncelet, E, Roberfroid, M., and Divisioq ~o~it~~ng onrs,Texas Air ContmI
Mercier, M. (1978). Mutagen&@ of butadiene and. Board, Austin, TX
butadiene monoxide- ~Bidwm. 3&p&w Res.
Common. So, 298-30!5.
Fajen; J&f., ,Roberg D.R., Ungers, LJ., and
de+Meester, C, . Poncelet, F., Roberfroid, I$ and Krishnan+ RR (1990). Occupational exposureof
Mercier, M. (1989). The mutagenicity of butadiene workers to l~~~~~e~ Envtin. HaaWr Pe~pact;,
towards Sal;iaDnella..typhimtim. T&&x& &E 6, 86, 11-18.
Filsetg- J.G., and Bolt . (2984). Inhalation
Dkmschmann, S.,, and Laib, RJ. (1989). pha~~~neti~ on. gas uptake studies.
Concentration-dependentdepletion-of non-protein VI, colmparative align of ethylene oxide and
sulfhydryl (NPSH) content in lung, heart and liver butadiene horde as exhaled reaclive metabolites
tissue of rats and mice after a;cute i~~tion of ethylene,and ~~-b~m~ie~e in rats. hk. TO&U&
exposureto btuadiene, T&oL Letz. 6s, 175;%83. S5,2%9-223.
References 91

Friedman, E., Qrnright, K., and Llpkln, M, (1982). I~ter~tionaL Institute of stoic Rubber Producers
Differential responseof familial polyposis fibroblasts (IISRP) @@lb). ~~-~~~diene: Inhalation
to two bifunctional alkylating agents. Careinogeit&.s te~t~ge~i~~ study in the rat. Report No.
3, 1481-1485. 2788-522/S Houston, TX.

Gart, J.J., Chu, KC, and Tarone,, R.E. (1979). Irons, R.D. ‘(1990). Studios on the mechanism of
Statistical issue ln interpretation of chronic bioassay lo-bu~diene-i~du~ leuk~ogen~~: The potential
tests for carcinogenic&y. J. N& Cancer Z”c 62, role of e~do~~no~ murine bukemia virus. Environ
957-974. H&h Fe9s-ct. 86,49*55.

Goodrow, T., Reynolds, S., Maronpot, R., and Irons, R.D., Smith C.N., Stillman, W.S., Shah, R.S.,
Anderson, M. (1990). Activation of Kara by codon Steinhagen, W.H., and L.J. (1986a).
13 mutations in CS7BUb x C3H F, mouse tumors ~~~i~rne~lub~tic male 136C3F1
induced by exposure to l,$butadiene. Cancer Z&s. mice foI~o~~g chronic sure to 1,3-butadiene.
50,481~4823. ~oxicd ApPr P~~c~~

Gopinath, C (1986). Spontaneous brsin turnours in Irons, R.D., Smith, CN., Stillman, W.S., Shah, R-S.,
Sprague-Dawleyrats. Fd Chem. Toxicol. 24,113-120. Steinhapn, W.H., and ~i~erman, L.J. (1986b).
~cr~i~rn~~~o~l~tic anemia in male NIH Swlss
Goto, K, Akematsa, T., Shit&u, H., and mice following rep*eated tare to 1,3-butadiene.
Sugiyama,T. (1982). Simple differential Gicmsa TuxicoL AppE ‘PhmnaiwL
staining of sister chromatids after treatment with
photosensitizing dyes and exposure to light and the Irons, R.D., Stillman, W.S., and Cloyd, M. W. (1987).
mechanismsof staining. Ch9omosoma 53,22ZLwO. Selective activation ~of endogenous ecotropic
retrovirus in hematopoietic, tissues of B6C3F1 mice
Haseman,J.K (1984). Statistical issuesin the design, d&ing the ~releukemic base of l$butadiene
analysisand interpretation of animsl cqrcinogenicity exposure. Vw~logy i61,457-462
studies. Envirorz Health Per$pecct.SS, 385-392.
Irons, JLD., Cathro, .P., Stillman, W.S.,
Haseman,J.K, Huff, J., and Boorman, GA (1984). Steinhagen, ~W.W., and Shah, R.S. (1989).
Use of historical control data in carcinogenic&y Susceptibillity to 1,3-butadiene-induced
studies in rodents TWoL Pathot 12,X&5-135. leukemogenesiscorrelateswith endogenousecotropic
retrovirdl background in the: mouse. To&oL AppL
Phamacd 161, 170-176,
Haseman, J.K., Huff, J.E., Rao, G.N., Arnold, SE.,
Boorman, G.A., and M&onnell, EE (1 Jelitto, B., Vangala, R.R.; and Laib, R.J. (1989).
Neoplasmsobserved in untreated and corn oil gavage Species differences in DNA damage by butadiene:
control groups of F344/N rats and (C57l3I&l?, x Rule of die~~b~~ue~ &$t. TQ%-icol. (Suppl. 13),
C3H/HeN)F, (B6UF,) mice. JNCI 75,97X%4. 246-249.

Huff, J.E., Melnick, RL, Solleveld, H.k, Haseman, Jonckheere, AR. (1954). A distribution-free
J.K, Powers, MY.,and Miller, R.A. (1985). Muhlple k-sampletest againstordered alternatives. Biornt%ri&z~
organ carcinogenic@ of l,$butadiene in I%QF, 41, m-145.
mice after 60 weeks of inhalation exposure. Sc&zce
227,~~549. Kaplan, Ef, and Meier, F. (19S8), Nonparametric
estimation from i~~rnple~ observations. J. Am.
International Institute of Synthetic Rubber Produ&rs S&z&Assoc, 53,4,57-M%
(IISRP) (1981a). The toxici@ and carciaogenicityof
butadiene gas administered to rats by inhalation for Kate, H. (1974}. Spontaneous sister chromatid
approximately 24 months. Report No. 2@X%&‘2. excltanges dot&ted by a ‘BudR-labeling method.
Houston, TX. Nature 2s-1, 70-72.
92 1,33-Butadiene, NTP TR 434

Kirshenbaum,1. (1978), Butadjene. In KrkdMmer Ma~g~~~, I&H.+ ~Collings, B.J., and Mason, J&I.
Encyclopediaof Chemical Technology,Vol. 4,.3rd ed,, (1983). Statistical analysis and sample-size
pp. 313-337. John W iley and Suns, New York. determi~ti~~ for rn~~ge~ici~ experiments with
binomid reSponse& Environ. M trtngen 5,X%-716.
Kreiling, R., L&b, R.J., Filser, J.G., and Bolt, H.M.
(1986). Speciesdifferencesin butadiene rn~~~~rn Ma~gol~, ~IU-I., Rest&k, M .A., Rimo, J.V.,
between mice and rats evaluate6 by ~~iatlo~ Archer, P,, Gall~w~y~S.M Bloom, AD., and Geiger;
pharmacokinetics.Ar& ToxicoL S&23 E. (1986)~‘S~t~~~l.~I~~ for irt v&0 cytogenetic
assaysusing Chinese hamster ovary cells. Entiufi
Kreiling, R., Laib, R.J., Filser, .J.G., and Bolt, H.M.
(1987). Inhalation phtirmacokinctics of Maronpot, R.R., and Boorman, G.A. (1982).
l&epoxybutene-3 reveal speciesdifferencesbetween Interpretation of rodent hepatocellular proliferative
rats and mice sensirive to bhtadiene-induced alterations .and bepat~~lular tumors in chemicat
carcinogenesis.Arch. Toxicol. 61, 7-11. safety assessment.Toxicor!&th& 10,71-80.
Kreiling, R,, Laib, R.J., and Bolt, H.h(I1 (1988). Marx, M ,P.,, Smith, S., Heyns, A. du P., and van
Depletion of hepatic non-protein ~ul~y~~ ‘content Tender, I2 (1983),, Fanconi’s anemia: A
during exposure of rats and mice to butadiene. cytogenetic study on lymphocyte and bone marrow
ToxicoL Lett. 41,209-214. cultures utilizing 1,2:3,4diepox@utane. Cancer
Geuret,c$togenet. 9,%60.
Krinke, G,, Naylor, D.C., Schmid, s., Froblich, E.,
and Schnider, K. (1985). The incidence of Matanoski; G.M., and 8chwartz,L. (1987). Mortality
naturally-occurring primary brain turnout% in the of worke,rsin ,~~~~e~&u~die~e polymer production.
laboratory rat. J. Comp. Path& 95, 175-192. J. &cup. Med. 29,67%680.
Ma~~~~~ G&f., Santos-Burgoa,C., Zeger, S.L, and
Laib, R.J., Filser, J.G., Kreiling, R., Vangala, RR., Schwa&, L (1989). epidemiologic data related to
and Bolt, H.M. (1990). Inhalation p~ar~a~~~efi~ he&h effects. of 1,3-butadiene. In Assessment of
of 1,3-butadieneand l,Z-epoxybutene-3in rats and Inhaler f$zzar& (U. Mohr, D.V. l3ates, D.L
m ice. Environ. Health Perspect. 86,57-63. Deports P.N. Lee, R.G. McCellan, and F.J.C.
Roe,%&.), pp. 201-2 SpiingerVerlag, New York.
MacGregor, J.T., Wehr, CM., and Langlcis, ,R*G.
(1983). A simple flourescent staining,procedurefor Matanc&i, GM., San * C., and Schwartz,L.
micronucleiand RNA in erythrocytesusing Hoeschst (1990). Mortality of a cohort of workers in the
33258and pyronin Y. Mutat. Res. 120,269-275. styrene-butadiene pal r manufacturing industry
(1943-1982). Ipw~on. ~IeaIth Pei-spect.86, 107-117.
Malvoisin, E, and Roberfroid, M . (1982). Hepatic McC$tnell,E.E., Solleveld,H.k, Swenberg,J.A-, and
microsomalmetabolismdf l$butadiene. Xenobiotica Boorman, G.k (1986). Guidelines for combining
K&137-144. neoplasms for evaluaticn of rodent carcinogenesis
studies. JIK’f ?6,283?289.
Malvoisin, E., Lhoest, G., Poncelet, F,,
Roberfroid, M ., and Mercier, M . (1979). McFee, A& Lowe, ECLW.,and San Sabastian,J.R.
Identification and quantitation -of I&epoxybutene-3 (19832 Improved s&ter-chFomatid differentiation
as the primary ntetabolite of 1,3-butadiene. using ~ra~~~~at~ ,bro~od~~ridine tablets in
J. Chromatogr. 178,419;423. mice. B&at. tiles* llgj 83-88.

McGregor, D.B., Brown, A., Cattanach, P., Morrissey, I&E, S&wet+, B.A., Hackett, P.L,
Edwards,I., McBride, D., R&h, C., and Caspary, sikov, MB., Hardin, r&D., cClanahan,B.J., Decker,
W.J. (1988). Responses of the L5178Y tk+/tk- J.R., and Mast, ‘I?.&(l&X& Overview of reproductive
mouse lymphoma c&I forward mutationassay~ III. 72 and dev~i~prn~nt~t&i&y studies of 1,3-butadienein
coded chemicals. Environ. Md Mutagm. 12,85-154. rodents. Emkm. Mea&h Perspect. $6, 79-84.

McGregor, D.B., Brown, AC., Howgate, S., McBride, Morrow, N.L (1990). ~~i~d~t~l production and
D., Riach, C, and Caspaty, W.J. (1992). Responses use of 1,3-butadiene. &avtron Health Perspect. 86,
of the L5178Y mouse iymphoma ceil fatiard 7-8.
mutation assay. V. 27 coded chemicals. Eirtiu~
Moi Mutagen 17, 196-219. Muilins, 3.k (1990). Industrial emissions of
l$-butadiene. ‘&wimn. &x&h Pemyxct. 86, 9-20.
M&night, B., and Crowley, J. (19841. Tests for Mybr, B,, Bowers, L, and Caspary, W.J. (198s).
differences in tumor incidence based on animal Assays for the ~~uctioa Of gene mutations at the
carcinogen&s experiments. J, Am, Stat. Rrscx, 79, thymidine l&n%@iocus in I.$f78Y mouse lymphoma .
639-648. cells in culture. Prq. Nutatt Rex 5,555~568.

Meinhardt, T.J., Lemen, R.k, Cr&tdail, M.S., and National Cancer Insti&ute(NCI) (1976). Guidelines
Young, R.J. (1982). Environmental ~epid~~iogic for ~~~oge~~~o~y in Small Rodents. Technical
investigationof the styrene-butadienerubber industry. Report Series -No. 1’. NIH Publication No. 76&K
Scand. J. Work Envirm. Health 8,250~2591 U.S. Department of Health, Education, and Welfare,
Public Health Service,National Institutes of He&h,
Melnick, R.L? Huff, J.E., -Haseman, J.K, and Bethesda,MD;
McConnell, EE. (1988). Chronic toxicity results and
ongoing studies of 1, the National National Institutes of Health (NE-I) (1978). Open
Toxicology Program. Ann. MI! A& Sci 534, Formula Rat and.Meuse.Ration (NIH-07). Specifi-
~ti&~ NIH 11-13X5 US. Department of Health,
Education, and WaIfare, Public Health Service,
National Institutes ef Heaith, Bethesda, MD.
Melnick, R.L, Huff, J.E., Bird,“.M.G.,and Acquavella,
J.F. (199Oa). Symposiuin overview: Toxi~logy, National hstItute for ~~~~o*aI Safetyand Heaith
carcinogenesis, and human health asp-& of (NIOSH) (11)90). Na Occupational Exposure
1,3-butadiene. Envirm Health Persptxt. 8fi;, 3-5. Survey(NOES) (19&l- unpublished provisional
data as of July 1,199O.
Melnick, R.L, Huff, J.E., Roycroft, J,H., Chou, B.J.,
and Miller, R.A. (1990b). Inhalation to&o&y and National Toxieolagy Program (NTP) (1984).
carcinogenic@ of l&butadiene in ,B6c=3ft, mice Toxicology and Careiaogenesis Studies of
following 65 weeks of exposure. Env&on. He&a 1,343utadiene.(CAS No; 106-99-O)in B6C3F1 Mice
Perspect 86, 27-S. ~I~~tio~ stud&Q. Technic& Report Series
No. 288, NIH Pubii~~~~ No. 84-2544, U.S.
Department of IIealth and Human Services,Public
Melnick, R.L, Huff, J., Chou, B-J,, and Milleq R.A Health Service, Natiermi Institutes of Heaith,
(199Oc).Careinogenicityof 13-butadbne in C57BU6 ResearchTriangle Park, NC.
x C3H F1 mice at few exposure conc&ttrations.
Cancer.Res.50,6X&6599. Nishi, Y., Hasegawa, M.M., Taketomi, M.,
Ohkawa, Y, and. lrmi, N. (1984). Comparison of
Miller, LM. (1978), Investigations of selected ~thiog~~ine-~~ism~t mu.tation and sister chromatid
potential environmental contaminants: Butadiene exchangesin Chinese hamster V79 cells with forty
and its oligomers. In: EPA-S60/2-78-008. U.S. chemical and ,phys&at-agents. Cancer Res. 44,
Environmental Protection Agency, W~hi~gt~n, DC. 3270-3279.
94 1,3-3ut8diene,NTFTR434

Obe, G., Kalweit, S., Nowak, C., and Ali-Osnan, F. ~~m~~~,n~ XI, rro, W., and Zijktra, J.A
(1982). Liquid holding experiments with human (1,983). St&es- on mutagen-sensitive strains of
peripherai lymphocytes. 1. Effects of liquid holding Dro~~ph~ ~~~~~~i~~. III. A comparison of the
on sister chrornatid -exchange-s
induced by ‘trenimon, m~~g~nic s~~i~i~~~ of the ebuny (UV and x-ray
diepoxybutane, bleomycin, and x-rays. Biol. Zbf 101, sensitive) atid Cu~~~~-~.,~~ld-~~ strains to MMS,
97-113. ENU, DEB, DEN and 2&X1,-PDMT. Mutat. Res.
Occupationai Safety and Health ~dm~~tratio~
(OSHA) (1990). Occupational exposure to Sasiadek, M, Norppa, H., and Sorsa, M. (1991).
l$butadiene; proposed ruie and notice of hearing.
1,3-B~~d~ene and its epoxides induce sister-
Fed. Reg. .55,32,736-32,826. chromatid exchangesin uman lymphocytes in vitro.
Olsen, O.-A, and Green, M.M. (1982). The Mirta& Res. 261, 117-121.
mutagenic effects of diepoxybutane in wild-type and
mutagen-sensitive mutants of Dros4phiin Schtnidt, U., .and Loeser, E. (1985). SpeCit!S
meianogaster. h&tat. Res, 92,107-115. differences in the fosmatjon of butadiene monoxide
from 1,3-butadienc. Arch. Toxic~l. 57, 222-225.
Gwen, P.E., Glaister, J.R., Gaunt, I.F,, and Pullinger,
D.H. (1987). Inhalation tox&ity studies with Sharief, Y., Brown, AM., Backer, LX., Campbell,
1,3-butadiene. 3. Two year to~c,i~~~rcinugeni~i~ J.A, W~~r~k-~~~ins, B., Stead, AC., and Allen,
study in rats, &. Imi HE A.sxw~ J. 48, 407;413. Sister chromatid exchange and
chromosome aberration slrnalyses
in mice after in viva
Perry, P., and Evans, H.J. (1975). Cytological exposure to acryionitriie, styrece, or butadiene
detection of mutagen-carcinogen exposure by sister monoxide., EnvsrcJn,,Mutagtm.8,439-448.
chromatid exchange. Nature 258, 121-125.

Porfirio, B., Daliapiceoia, B., Makini, v., Shelby, M.D. (1990). Results of NTP-sponsored
mouse cytogeneticstudieson 1,3-butadiene, isoprene,
Alimena, G., and Gandini, E. (1983). Failure of
and chloroprene. E&ro& Health Perspect. 86,71-73.
diepoxybutane to enhance sister chrorixatidexchange
levelsin Fanconi’s anemia patients and heterozygotes.
Hum. Genet. 63,117”120. Shirley, E (1977’). A non-parametric equivaient of
Williams test for ~~tr~~~~g increasing dose levelsof
Portier, C.J., and Bailer,, AJ. (1989). Testing for a treatment. Bti~etics 33,386389.
increased carcinogenic@ using a survival-adjusted
quantai response test. Flu&am. Apple To&c& 12, Shugaev, B.B. (1969) @oncentrations of
731-737. hydrocarbons in tissues as a meaSure of toxicity.
Arc&. E~vkw~ &a& l&878882.
Pot-tier, C.J., Hedges, J.C., and Hoei, D.G. (1986).
Age-specific models of mortality and tumor onset for Shukla, F,T,, and A~e~ba~~,C. (1980). Genetic tests
historical control animalsin the NatioszaiToxicology for the detection of chemically induced small
Program’s carcinogenicity experiments. Cancer Res. deletions in ~u~u~~i~~ chromosome. Mutat. Res.
46,4372-487X& 75 232-243.
Reynolds, S-H., Stowers, S.J., Patterson, R.M.,
Maronpot, RR., Aaronson, S.A, and Anderson, Simmon, V.F.. (1979). ?&Ivitro mutagenicity assaysof
M.W. (1987). Activated oucogenesin B6C3F,fmouse chernicai ~ci~~~~ and reiated compounds with
liver tumors: Impiiatiom for risk wsment, ~~~~il~ ~~~u~u~* S. NaaL C~CW Inst. 62,
Science237, X309-1316. *

Sadtier Standard Spectra. IR No. 893. Sadtler Tarorte, RE. C1975). Tests for trend in life table
Research Laboratories, Philadelphia. analysis. Biom&ika 62, 679-682.
References . 9s

Thurmond, L.M., Lauer, L.D., House, RV., Wade, MJ,, Moyer, J.W., an$ Hine, CH. (1979).
Stillman, W .S., Irons, RD., Steinhagea, W .H., and, Mutagenic action of a seriesof epotides. Mutat. Res.
Dean, J.H. (1986). Effect of short-terra ~halati~~ x%,367-371.
exposure to I$-butadiene on mu&e immune
functions. ToxicoLApple Phamacol. 86, 170-179. Watson, W ,kF. (1972). Studieson a recombination-
de&i&t mutant. of~~~~~~ responseto x-raysand
Tice, RR., Boucher, R., Luke, CA, and aIkyIating agents. Mutat. Res. i4, 299-307.
Shelby,M .D. (1987). Comparstive ~ogeneti~
analysis of bone marrow damage :induoed in male W ithams, D.A. (1971). A test for differencesbetween
B6C3F1 mice by multipie exposures to gaseous treatment “mearis when several dose levels are
1,3-butadiene.Em&m. Mutagen. 9, 235-250. twrq.&ed with ‘i zero dose control. Borne&s 27,
103-l 17.
United StatesDepartment of Labor (1981). OSEtGL W iIiiams, D.d (1972). The ~mpar~o~ of several
Safety and Health Standards. 2%X% 1910.1ooO. dose Ieveb with a zero dose c&roi. &me&ks 28,
Table Z-1. 519-531.

United States International Trade Commission Geiger, E. (1990). ~u~ge~ic~~ of 42 chemicals in

(USITC) (1990). Synthetic Organic Chemikals; Salmonella.. EBV&DB.~Md. ~~~a~~ I.6 (Suppl. 18),
United States Production and Sales, 1989. Publ. 32-54.
2338. USITC, Washington, DC. Geiger,E+,and Pagsno,D.r3, (1989). Mutagenicity of
the human careiqogen tr~~~~~n in Salmonella.
Van Duuren, H.L., Nelson, ,N., Orris, L, Bnvkon. MOL‘Mutagen. 13,343~346.
Palmes, E.D., and Schmitt, F.L. (1963).
Carcinogenic&yof epoxides, lactones, and peroxy Zeiggr* ET,Anderson, I%, Hawqt&, S.,Lawlor, T., and
compounds. J. NatL Caitcer Inst. 3X,41-55. ’ Mortelmans, K.. (1992). sairnotie&r mutagenicity
teats. V. I?es&s~from the testing of 311 chemicals.
Etaviro~.Mot ~utug~ 19 (Sup@.21), 2-141.
Van Duuren, EL., Langseth, L., Orris, L,
T&or, G., Nelson, N., and Kusehner, M ; (1966). Zimmeringj S,, Mason, JM., Valencia, R., and
Carcinogenicity of epoxides, lactones, and permy ‘Woodruff, RC. (I% @ ). ~~i~Imu~gen~is testing
compounds.IV. Tumor response in epithelial and in Lki7sophiI~. II. Results of 20 coded compounds
connectivetissue in mice and rats. J. NatL. t%x&?r test& for the N~tio~aI Toxic01 Program. Environ.
Inst. 37, 825-838. Mutagen 7,87-io0.
% 1,3-Madiene, NTP TR 434