Document by2ymYzk6DL0jBdzEwj3MeG73

REPORTS 171 matched control subjects from gested that part of the increase may be aBOnlrodgoadRnoiLsckehvlooelrfsinBoerfeRasetsidCuaenscer the not sdaemveelosptudcyanpceorp.ulRateisounltsw: hoMedaidn due to factors. unexplained environmental levels of DDE and PCBs were higher for breast cancer case patients than for control subjects, but paired dif ferences were statistically significant The possibility that organochlorine residues may be related to breast cancer arises from several observations. DDT (2.2-bis(/*-chlorophenyl)-1,1,1- MRToiavneriyroalo.S..NEWerioilclfDf,Wttb.PianLo*eleo. GM.arilyn only for DDE (P = .031). After adjustment for first-degree family history of breast cancer, lifetime trichloroethane). its metabolite DDE [ I. I-dichloro-2.2-bis(/>-chlorophenyl) ethylenel. similar pesticides, and re lactation, and age at first full-term lated chemicals (PCBs |polychlorinated pregnancy, conditional logistic re biphenyls] and PBBs Ipolybrominaled Background: Organochlorines such as DDT (2,2-bls0*-chlorophenyl)-1,1,1 - trichloroethane] and PCBs (poly chlorinated biphenyls), which have been used extensively as insecticides and as fluid insulators of electrical components, respectively, are known to be persistent environmental con taminants and animal carcinogens. These agents have been found in human tissue due to their inefficient metabolism and their solubility in lipids, which lead to lifelong se questration In adipose tissue. Their association with human cancer occur rence, however, has been explored ohnalvyin gm a2r0g inora llyfe, w ewrithc a smeso.st Pusrtupodsiees: This blinded study was designed to detecmine whether exposure to PCBs and to DDE [l,l-dichloro-2,2-bis(i- chlorophenyl) ethylene), the major metabolite of DDT, is associated with Mb reetha os tds: cancer risk .We analyzed in sera women. from the stored blood specimens of 14 290 participants enrolled between 1985 gression analysis showed a fourfold increase in relative risk of breast cancer for an elevation of serum DDE concentrations from 2.0 ng/mL (10th percentile) to 19.1 ng/mL (90th percentile). For PCBs, the relative risk for a change in serum levels from 3.9 ng/mL (10th percentile) to 10.6 ng/mL (90th percentile) was less than twofold, a nonsignificant asso csaiifoatnetir:onIandjttuhhsaisttmpeownptauslafotfirounrDthoDefErN. erwCeodYnucoclerudk City women, hreast cancer was tssietorrnousnm:glyTbuhteassensooctfiianwtdeidtinhgsPwCitshBusg.gDeismDt Epltihcaaint environmental chemical contamina tion with organochlorine residues may he an important etlologic factor In breast cancer. Given the wide spread dissemination of organo chlorine insecticides in the environ ment and the food chain, the implications are far-reaching for public health intervention worldwide. (J Natl Cancer Inst 85:648-652, 1993] cbiinpohgeennysls])(1.2a)re known animal car and suspected human carcinogens (/) and may compromise immune function (.?). Both DDT and apPcrCtoiBmvsiotytehras(v1e(.21..b42e))e. nanMdsohroteowonvhearvt,eo be tumor estrogenic they have become ubiquitous in the environment and in human tissues, due to their inefficient metabolism and their high solubility in lipids, which leads to lifelong sequestration in adipose tissue. Previous studies have been unable to establish a clear connection between exposure to these chemicals and breast c(1a.n0c)e. ralt(h5o).ughor soimndeeesdtudoieths e(r7-9c)anhcaevres reported possible associations with breast cancer. A large prospective cohort study designed to investigate the role of endogenous hormones and environmen tal factors in cancer among women offered us an opportunity to examine the association of breast cancer with individual exposure to DDE and PCBs. as estimated by measurement of levels in blood serum. and 1991 in the New York University Women's Health Study, a prospective Environmental factors have long Subjects and Methods cohort study of hormones, diet, and cancer. Cohort members who devel oped breast cancer were Included as case patients in our nested case- been suspected to play a prominent role in breast cancer etiology. Both diet and hormones have been implicated, but specific agents have not yet been paSrtuicbijpeacntts. Siiundy ihsucbjeNctewwerYeorakuhL:neiivefr'iIi*y Women' Health Study, a cohort study of hormonal and environmental factors and cancer control study. DDE and PCBs were identified, and the recognized risks in women (10). Between 1985 and 1991. 14290 measured by gas chromatography in explain only a small proportion of the the sera of 58 women with a diag disease. Furthermore, breast cancer in nosis of breast cancer 1*6 months cidence has been steadily rising in the after they entered the cohort and in United States, and it has been sug *5*e " Noie" section following " Reference." 648 REPORTS Journal of the National Cancer Institute. Vol. 85. No (8. April 21. 1993 New York City women were enrolled while mbilteoenorddicndogrnadwaonlm.inagSmemrluoomgrhasappvheinycgimsce3rn0esen1wi1n1Legre oclfinvicenaonuds stored at -80 'C. Stud} subjects were aged 35-65 years (median age. 50.7 \earsj. and HO** were Caucasian Cohort members developing breast cancer were included as case patients in a case- control study nested within the cohort. Control csuobnjseicsttisngwoerfealslelceochteodrt amt ermanbdeorsmwfhroomwerirsekalsievtes and free of cancer at the lime of the cancer diagnosis in u case patient and who matched the case patient on menopausal status, age at entry into the studs (r0 .5 year), number and dates (rO.25 year) of blood donations, and, if premenopausal, day of menstrual cycle at the lime of first blood drawing. Two control subjects were selected for each postmenopausal case patient (n 34). and four were selected for each premenopausal case patient (n = 24). Case and control subjects did not differ with regard to socioeconomic status (as defined by years of school). DDE and PCB analyses were performed on specimens from the 58 breast cancer case DDSEtataisntdicaPlCBanalelyvseilss. iSntactaissteicaplatcioenmtspaarnisdon(sheoirf matched control subjects were made hy paired Student's / tests Relative risks. as estimated by c(hoendoidtidosnaraltimosul(tiOpKles)l,ogwisetriec dreegterremssiinoend whyithu%DeDoEf aufrnoeduqsuPeCnvcBayriaebxdlpiesosts.ruibrQeutuiaoisnntibloeotshf qwcuaeisrneetsilebsaasnaednddcocononntrtotihnles combined. Adjustment was made lor potential confounders (other than the matching criteria). Potential confounders included body mass index (pkregg/mna-ni.cya,gefirastt-dmeegnraerechfea.mailgye haitstfoirryst offullb-rteearmst cancer, history of benign breast disease, months of lactation, history of tobacco smoking and/or alcohol drinking, and race. There was no apparent modification of effect by age. Final models included only those covarlates that a(hlteerecdontthineuoruesgreesxspioonsurceoefvfaicriiaebnltess fobry ciatthelreaostf 15**. Covariatcs included in the final models (awyteersfeirosfrtirnsfotu-)ld,le-tlgeifrreemteimfpaermemgilnoyannthhciysstoo(rf<yl2aoc1ftaybtieroeanar,ss.tanc2da1n-a2cge9er, quintile, compared with the lowest, was 3.68 (95 % confidence interval = q1u.0i1n-ti1l3es.50w);asthe4.c4h4i-s(qPua=re for trend for .035). When serum DDE was examined as a contin uous risk factor, there was a 1.09 increase (995-) in the adjusted OR for breast cancer per unit increase in DDE (regression 0.0301 (iS Ec|o; efPficifeonrt = trend 0.0823 i = .0037). This increase corresponded to a four fold risk for an elevation of serum DDE from 2.0 ng/mL (10th percentile) to 19.1 ng/mL (90th percentile; OR = 4.08; 95% confidence interval = 1.49-11.20). The effect of adjusting for lactation history was to strengthen the associa tion between DDE and breast cancer. (Regression coefficient for DDE as a continuous variable = 0.0509 0.0218 patients who were diagnosed within 6 months after entry into the study (prevalent cases) and (mhaeticrhe1d71semts,atocnheedexctorantcroonl trsoulbhjeacdtsb. ee(Fnosrelseecvteedn in the parent project, and all were used in the pbinrlgiensdesentudi dsywtuitsdhuyb.r)jeesTcpthese.ct latoboarnaytoriynfoinrmveasttiiognatocrosncweernre ThDe emteertmhoindatfioorn doeftelremvienlsatioofn DoDf EtheandlevPelCsBosf. DDE and PCBs. including the quality-control protocols, was the same as (hat described >29, nulliparous). Similar results were obtained in models that in'! Jed all of (he potential con ftraoeigunrneeddsesrifsor.nomCcootnehffeifdicesinteacnnetdsairndteervrraolsrs (o9f5`*thte wpeerretinoenbt Results The mean levels of both DDE and PCBs were higher among the breast [iS E l without adjustment for lifetime duration of lactation compared with 0.0823 i 0.0301 in the model de scribed above.) However, the effect of lactation itself wus protective. When we examined months of lactation as a continuous variable, the decrease in OR for breast cuncer per additional month of lactation cient = -0.1 was 308 0.808.0(5r5eg6re[ssSioEn]\coPefffoir elsewhere (//). with minor modifications to obtain satisfactory results with the use of 1-mL aliquots of serum. The cleanup step used a 9-mL fraction of lr* methanol in hexane eluted from W1.2. VgaF)l.orainsdil g(Uas.SchrSoimlicaatoCgroa..phByerwkealseypeSrfporrimngesd. cancer case patients than among their matched control subjects. The paired differences were statistically significant only for DDE (Table I). DDE was approximately 35% (or 2.7 ng/mL) trend = .0069 in a conditional logistic regression model adjusting for DDE. PCBs. first-degree family history of breast cancer, and age at first full-term pregnancy). with electron capture detection on a 0.53 pirn X t3e0mpmera5t*u*rempertohgyrlapmhenmylodseiliwcoitnhe dciorelucmt ninjienctitohne o(aPrfee0rkr5einp-poE.rLtleindoenr aasCn onArapnu..otogsNryaosmlrewsmalpkge.rasCmcohinlrlnoil.mi)t.eartRog(epsrauaplnthss higher in patients than in control subjects; PCBs were 15% higher (by 1.0 ng/mL). In conditional logistic regression analyses, alter adjustment for first- For PCBs. the adjusted ORs for breast cancer rose from 5.18 in the second quintile to 7.02 in the third, but they decreased thereafter to 4.10 and 4.35 in the two upper quintiles. The per billion |ppb|) of /./'-DDE and of the higher PCB congeners, (hose with retention time lunger othnan(htehatpeoafk DpDeErc.enTthaegescalrceuploatritoedns fworereArboacsleodr ct1ho2em6s0epobnpyeenaWktssebwbwitaehprejoj uMrrecsmoColavdlelidfi(e.1di2n).tgoaSsintwccheorocmoerratatoitnhgrreoaef phs technique, the method of calculation was degree family history of breast cancer, lifetime lactation, and age at first fullterm pregnancy, the ORs for breast cancer increased markedly with in creasing quintile of DDE concentration (Fig. I). The adjusted OR for the upper chi-square tiles was for 1.99 t(rPen=d of the PCB .16). When quin serum PCBs were examined as a continuous variable, the increase in OR per unit c=ha0n.0g7e95was 1.08 (regression 0.0587 (SE1; Pcofeofrfictrieenndt revised by use of the data of Mullin et al. as reported by Burse et al. (13). The Webb and aMgceCs!alwl erpeea2k0s3 ((1429)**(.an3d3*t*heinir orredveirseodf rpeteerncteinont Table 1. Serum concentrations of DDEmaatnchdedPCcBonstrionl bsruebajsetctcsancer case patients and in individually time). 232 (4.5*. 3.3**. 2.2**). and 372 (2.1**. nft1hrgo9r/emi'en)LtthimfLeoiermisoDiwttsDheesEotfsaqtdnauednatdel2iactrytdni-ogcn/domenwvLtrieoarflteoiorsanePpruCpomrfBosxthpibmeoaoasrlteeedsoluyvoletnsrI Mean r SD. ng/mL Ca(sne *p5at8i)e*nts Con(ntro=l 1s7u1b)jects dinfMfge/mreeaLnnc*e, ft P the course of the analyses (n = 18). Statistical cSocmhopoultaotifonMs edpiecrifnoermaesd paalrt tohfe thMeoulanbtoraStionrayi DPCDBE 181..00 ir 49..11 67..77 ii 62..89 21..70 21..2924 ..003518 aUnnailvyesressity uotiflizNeedw (YheorkfaCcoilmitipeustinogf Ctehneter City +*APasirethde Sdtiuffdeernetn'scei bleetswt,eetwnoa-tacilaesde patient and the mean of her matched control subjects. inumai of the National Cancer Institute. Voi 85. No. 8. April 21. 1993 REPORTS 649 ccFfPrreaoCieggseqBer.fuefisIepcq.snaiiuectoRiinienneetntsUitcl(siefoiosveaSrefnfwEdDirce)iDicsreekowEnnta(atqsors(oudl0idfnSo.ss0tlEuil8lrob)e2awjst3eiwsocw:)tasesQorcef00lo..bm0a0r=s37eb09aif41ns5o-tel3ld(oc8.Paw, nOs0Qc=:f.e02rQt5.h0Rbl=e0y73(571lq8o()2up;w-ic3n)ca3nths.i*elie-.Qs1s6pq36-auo3)t;aif9=er.cneDhtQ1sDif0,-2oEs-or2q*n9uatl,anry6erdQ-en34d4P49fC,of=bBroQar3td1cre2Docnn-nDo3dcm0Ele.bnfp-otlaq2rrenua7tdeit.PnioCtQdQniBla4se5tsaiqcn=fu*o8isnr1e-45t1rtiu5h-4le2me.4s9.a.c(naoQPAdnu)d1=Qii.nt9F5it9o.io0lnr(e3h(as5PilDg)a.Dhlraoe)EBgbi*).s1a(t6cFsi1c)oeo.3dnr-rt2eio6Pgnn.CureoBCtsuhsaesis(ocenfvo-racenaroqtininaunabtelurylnooescl)uey,ssf.rdtehviCqseatuarriseirabeenbg-uclcrtieioee)ossn,nstirftoohoonerlf = .16). corresponding lo less than a twofold risk for an elevation of levels of serum PCBs from 3.9 ng/mE (10th percentile) to 10.6 ng/mL (90th percen patients for some risk factors and to statistically adjust for others. andA bstrreoanstg acassnocceiratiownasbeftowuenedn, DwDitEh DDE. since inclusion of both in the mPCodBesl. Tgrheeatlryiskredcaulcceudlattihoens efsfheocwt inogf twofold to fourfold increases across the tile; OR = 1.70; 9.V# confidence higher DDE levels among women who range of serum concentrations arc interval = 0.79-3.68). When both DDE developed breast cancer than among comparable to those calculated from a and PCBs were included in the model their matched control subjects. The smaller study (7) and are also similar as continuous variables. DDE remained relative risk rose in an approximately to cancer rate estimates obtained from essentially unchanged, but the coeffi cient for PCBs was greatly reduced in log-linear fashion with increasing levels of DDE (Fig. 1. A). With PCBs. the rainskimalasdseastasm(e2n.1t 4).com putations using magnitude (by 539?-) and in statistical trend for breast cancer risk with Organochlorine residues may be sDigDnEific=an0c.3e60(;chPi-s=qu.a5r5e). adjusted for increasing levels was not statistically significant. Moreover, even though cer found throughout the body and in concentrations proportional to the lipid tain individual ORs were significantly content of a tissue. The primary site of Discussion elevated within quintiles of PCBs. the organochlorine storage in the body is pattern of increase was characterized adipose tissue, which has been obtained Previous studies (5.7-9) have raised by a flattened dose-response curve from autopsy and biopsy specimens to the question of organochlorine residues along the upper quintiles (Fig. 1. B). measure organochlorine residues in and their possible association with This trend, which may have arisen several studies (5.7-9). However, blood breast cancer, but the results of these from an imbalance in the numbers of serum concentration is a more conven studies have been inconsistent. All of case patients and control subjects ient means of estimating body burden, these studies had methodological limi tations. None adequately adjusted for known breast cancer risk factors, and only one study (9) had more than 20 within the lowest quintile, was less consistent with the stepwise incremen tal function typical of a biological dose-response. The nominal association since it is easier to obtain from humans and to process in the laboratory. Blood serum was used in our study because archival sera, having been collected case patients. In our study, it was of PCBs with breast cancer may have over a period of several years, were possible to individually match case been due mainly to its relationship with already available for a large cohort of 650 REPORTS Journal of ihc Naiional Cancer Institute, Vol. 85. No 8. April 21. 1993 (w1o5m,1e6n). A number have shown of previous studies that, if detectable, concentrations in serum accurately re from statistical analyses. Although lac tation has not been found consistently to be associated with risk for breast status, lactation, and diet. Statistical power was not sufficient to study diet in this investigation of 58 case patients. flect the levels of organochlorines in cancer, one report (26) suggested that It would also be desirable to have fat. Thus, the partition between the adipose and serum compartments is women with lactation failure were at greater risk than women who were able knowledge of blood or adipose levels at a point long before diagnosis of approximately 200 to 1. the rutio of their respective lipid contents (90% and to breast-feed for longer periods. The interrelationship of curtailed lactation cancer, in case organochlorine con centrations change near the time of 9.4%) (/7). Because fluctuations in serum tions (l1ip8i)d, s can it is introduce minor preferable to varia adjust serum concentrations for serum lipid content. We plan to make this adjust ment when serum lipid data become jvailable from other stages of the and DDE or DDT residues might bear further investigation in terms of their relationship to breast cancer and estrogenic effects. Environmental factors have been in voked to explain rates of breast cancer in developed countries, and nutrition cancer onset because of variations in lipids or body weight. Confirmation of our findings among incident cases (those diagnosed >12 months after blood was drawn) would be important in this regard. Moreover, since the risk associated with organochlorines may be investigation. !n terms of mechanism, the possible elationship between organochlorines especially has received much attention as a potential determinant of this disease (27). That breast cancer is sex merely a surrogate for other agents in the diet, the question of fat or chemical carcinogens is also an important one to md breast cancer is based on studies of hormone dependent is also widely pursue. inirnal carcinogenesis as well as on a number of experimental observations hat are consistent with breast cancer tiology. Induction of the cytochrome acknowledged. However, although the relationship among diet, sex hormones, and risk of breast cancer has been examined in many studies (27-29), a Breast cancer incidence has been steadily rising during the past two or three decades, a trend characterized by increasing rates among estrogen- ?-450 mixed-function oxidase enzymes, consistent picture has yet to emerge. responsive tumors, by continuing in aanipmhaelnsom(2e.n3o)n, widely investigated has been observed in in (h1u9m.a2n0s). exposed to Through such this chemicals pathway, aMreorleoowve(r1, .2t-h1e.5)reilnatisvtuedireissk(s28o.2b9se)rtvheadt find an effect for dietary fat or caloric intake. <- creases among older women, and by growing numbers in both developed and developing countries (29). Between 1973 and 1980, the incidence of breast organochlorines may alter the metabo Overall, the epidemiologic data sug cancer in the United Slutes increased a lism of xenobiotics and steroid hor mones. In metabolites, adedsitpioenc.ia llDyDTtheando ,pi't-s isomers, exert estrogenic effects that are epitomized by eggshell thinning and reproductive failure in birds (2/) gest that estrogens exert a cancer- promoting effect (56>) and that a diet rich in animal products and fat may increase cancer (2a8). woman's risk of breast Our data suggest that organochlorine residues and. in particu modest 8.0% among women under 50 years of age. while it rose 32.1% among years women or older in(31th).e age group 50 Although this increase enhanced casncrebeeninegxpl(a3i2n)e,d in the part by upward and that are also shown by the ability tttoeuimnion(rt2se1ra(.22ct23),2w4ain)t.hd estrogen-binding pro to promote estrogenic Also, an inverse asso ciation has been reported in humans lar. DDE are strongly associated with breast cancer risk. These observations are important in light of the fact that DDE is a widespread contaminant of animal food products and that human shift is also consistent with the histor ical pattern of accumulation of organo chlorine residues in the environment; i.e., older women who had the greatest potential cumulative exposure to DDT between levels of DDE in breast milk absorption is related to the ingestion of between 1945 and 1972 may now at birth and subsequent duration of lactation (25), which offers additional animal fat. Once absorbed. DDE may affect breast cells through binding to experience a higher risk of breast cancer than women much older or indirect evidence for an effect of some organochlorines on the endocrine balance. Lactation exerted a strong effect on the association between DDE and breast cancer in our data. Without adjustment for a woman's lifetime breast-feeding history, the coefficient either cytosol or nuclear receptors (thereby modifying cell regulation), through the mediation of the cytochrome P-450 oxidative enzymes that are intimately involved in steroid hormone metabolism, or by interference with sex hormone metabolism at the hypothalamic level. younger who were not similarly ex posed. Likewise, diminishing rates of breast cancer in Israel have paralleled a precipitous decline in environmental contamination with DDT and benzene hexachloride (55). The existence of a limited exposure window would pro vide an explanation for an earlier for DDE would have been reduced by 38%. Thus, among women who had breast-fed, DDT exposures prior to While these findings are novel and suggest a number of potential follow up studies, the conclusions require negative study (5) and would suggest that the risk of cancer can be expected to decline as the levels of organo nursing may have been elevated, but lactation would have reduced their serum DDE levels (25). and an effect of DDE could have been overlooked if confirmation, ideally in an expanded study design that will allow examina tion of confounding factors. Of particu lar merit for further study would be chhulmorainnetirsessuiedsue(s34c)o.ntinue to decline in Environmental accumulation of DDT and PCB residues has been documented information on lactation were omitted country of origin, estrogen receptor for at least three decades. Levels in the Journal of the National Cancer Institute. Vol. 85. No 8. April 21. 1993 REPORTS 651  4' sregptlUiioisexoooshnnvpkvsneigeuoestdrerrsaeondmusllmfariaineStsyee;cbtdnnaeoscetcttdtoecityhlhysnar,leeedtrigaceenbhlnuofsauuadol,mteasrerdtemEife,oc1bounoon9errnn,c7eoa.a0snipsbneslSeuqociaotsenunhslrfgecoah,neawvutptiem,rwmeaelraidseitthtnmuaihiecmkalaiutfienilntlioxnteaatghsfrr cnrccscfooiphahennnnwOllrdoomeccuirreaeniiererrdnngnnveest.iosadibnlrdesenhieiscsncarwiosdvevneouaetmtmterahiesomeeilninnamitsta.noietmanintpgtGoiehvrondoieinvrlivooaeirwntndiweosme-kiflteethhvniaomeneotf,oldrprgbogwetarrahnetinfnaedvaaoonsesrie--t-t rineatecrhvienngtioimn pwlicoarltdiownisdef.or public health ( / /) WtioonlflifmMitsS.ofRiovregranaocMh,lorBinaekepresDtiBci:deDsetaencd related compounds in Mood serum. Bull i/2) EWnfvrirronRGC.onMtacmC Toxicol all AC; 4Q7u:4a9n9ti-t5a0ti3v.e P19C9B1 standards for electron capture gas chro matography. J Chromalogr Sci 11:366-373, (/J) UB19us7er3soef VreWfe.reGnrceucpfo. oDlsF.inKcoormvkparreMtPh.eeqtuaalii.: tative and quantitative determination of poly chlorinated biphenyls hy packed and capillary gas chromatography with electron capture detection. Pan I. Serum. Analyst 115:243- (J4) c2Fo5on1ks.aunm19pJ9tA0io.nCaidnviMso.riCeshoanxdkjpnroDje:ctSedpocnanfcisehr risks in the Great Lakes basin. Am J Puhl (J5) JaHPLdaei.atpliotKthstres7aut9iins:.:s3u2CDe2G-ol3ne2vfJ5leak.lts.io1oNn9f8e92eb,de3th.w7a.e8me-tneLirsLae,rcuhPmloirrokadnii.de-: ben/.n-p-dioxin in 50 persons from Missouri. Arch Environ Contain Toxicol 17:139-143. (J6) NI9eRe8dham LL. Burse VW. Head SL. ft ai.: Adipose (issue/serum partitioning of chlori nated hydrocarbon pesticides in humans. Combined analysis of 12 case-control studies. (29) JHalaN:rarBtlirseCaJasRtn.cceLarnipcInpemrsta(8np2a:Md5 6E1I.)-.5V6Nf9.r. oEn1n9fg,,9sl0i U. ft J Med (30) K32e7y:3T19J.-3P2i8k.e 1992 MC: The role of oestrogen* apnrdevepnrtoiognestnafgebnrseaisnt the epidemiology and cancer. Eur J Cancer (J f ) CCRlaiiknnsceOLrnAcSGota,l tiH2s4tai:cn2sk9h-R3y4ev.BiFe1w.98M819.f7.r3-8B8A, ;DeHt EaWl: PubI No. (NIH)91-2789. Washington. DC: (32) WoUfShtihGteeoviEntc. rPLerafi.snret COYff,. K1r9i9s1tai. AR: Evaluation in breast cancer incidence in relation to mammography use. J Natl Cancer (33) WcInasnftjcte8tri2n:1aJ5nB4o,6m-R1a5liyc5.h2t, eAr1n9n9E0: NTheYIsraAecliadbreSacsti (.M. ) 6OK0ri9ng:va2.n6o9cF-h2Wl7o.r9i,nWe19o9rp0mesticPidHe.s RaonidtimpooklykchlDorPi: nated biphenyls in human adipose tissue. Rev Environ Contam Toxicol 120:1-82, 1991 Notes References (J) CInatnecrenra:tiOonccAut.paAticoinkanlcvexfpoorsuRree*seainrcihn on sec ticide application, and tome pesticide*. IARC Monogr Eval Carcinog Risk Chein Man 53:179-249. 1991 (2) SLlWLB:ERCHaOrcRiNnoEgeMni.cGityt.AoHfERpTolyHhPa.loRgOenBaFt.RedTSOhNi- (J) p2Eh0xe:o4nn4y0ls-:J4H9P.6C.BK1*f9r9a0nkdvuPeBtB v Crii Rev Toxicol NI. Talcott PA: Immunotoxiciiy of carcinogenic pesticides and related chemicals. Environ Sci Health. Pan C: Environmental Carcinogenesis Rev (4) CSPao5lf:y7ec3hS-loI.2rif0nj.ai:t1e9dE8n7vBiriopnhmenenytlsal Toxin Series I. (PCBs): Mam malian and Environmental Toxicology, New til YAUton.:rgke:OrSrgpMarn.inoKgcehiralo-fVrrienreHla.gc.Bomu1c9p8ho7eurntd-sTciniit hMum. afnt breast fat from deceased with and without breast cancer and in a biopsy material from newly diagnosed patients undergoing breast (6) sNuircgheroyu. dEnnviWroJn. RSfejspm34a:n24-H2.8,Sf1.i9.i8k4off IJ: Mortality experience of workers exposed to polychlorinated biphenyls dunng manufacture of electrical capacitors. Report to the Indus trial Disease Standard* Panel. Ontario Mini*- (7) tFrayi.cokf LFa'Jbnor. .R1ic9c8i7 A Jr, Wolff MS. ft al: Pesticides and polychlorinated biphenyl resi dues in human breast lipids and their relation (8) 4Wto7a:1sb4sr3ee-ra1sm4ta6n,cna1n9Mc9e2.r. Arch Environ Health Nogueira DP, Tomatis (/7) CBhreomwnnspJhFerJer. 2L0a:9w7t5o-9n8 0. 1990 RW: Polychlorinated biphenyl (PCB) partitioning between adipose tissue and sennn. Bull Environ Contain (fS) TPhoxilicloipls3D3L:2.7P7i-k2k8l0h. J1L9.84Burse VW. ft ai.: Chlorinated hydrocarbon levels in human serum: Effects of fasting and feeding. Arch (J9) PEonvliaronnd CAo. nStammithToDx,icKoul n1t8z:m49a5n-5R0.0.ft19a8l9: Effect of intensive occupational exposure to DDT on phenylbutazone and cortisol metabo lism in human beings. Clin Pharmacol Ther (20) kA1t1l:v7aa2ir4.:k-7sA32Alt.ePr,1a9tFi7oi0sncshbineindrAug. Anderson KE, metabolism in workers exposed to polychlorinated bi phenyls. Clin Pharmacol Thcr 22:140-146. (2/1 oB1f9u7iD7.(1DeTr aWnaHlo. gKs.uApfmkrJ Dln:dEMstreodge4n:i1c63a-c1t7io3n, (22) M19a83min RR. Snti't.iF CiJ: Interaction of ,/)'- DDT with the estrogen-binding protein (EBP) of DMBA-induced rat mammary tumors. Res Commun Chem Pathol Pharmacol 33:119- (23) St1ico2n8r.iro1nf9f8rm1 aJmDm. aMryotgtlkani dNtKu:moDrDs Tindauccceedlerain the m ale Spraguc-D aw ley rat by (24) 22R:-da1rc2ie3st5da-nm12iAd39oK.p.h1Sc9in8ra!rnatshkrucnDc.A Carcinogenesis . Stancfi. GM: DDT supports the growth of an estrogen- responsive tumor. Toxicol Leu 27:109-114. (25) Ra19lo8:g5aPnolyWchJ,loGrilnaadteedn BC. McKinney JD, f.t biphenyls (PCBs) and Auihnr affiliations; M. S. Wolff, M. Rivera, DMievdisiicoinne. oMfounEtnvSiirnoanimSenchtaolol aonfdMeOdcicciunpea.tiNoneawl York. N.Y. P. G. Tnnioio, N. Dubin, Institute of Environ Cmaennctaerl CMeendteirc.ineNewandYoKrkaplUanniveCrosmitypreMheendsicivael Center. New York. N.Y. E. W. Lee. Institute of Environmental Medicine. NeCworYreosrpkonUdneinvceersittyo: MMedaircyal SC.entWero. lff. Ph.D.. Division of Environmental and Occupational Medicine. Mount Sinai School o f Medicine, I G10u0s2ta9v. e L. Levy Place. Box 1057, New York. NY Supported hy Public Health Service (PHS) grams ES-05638. ES-0092K, and ES-00260 from the National Institute of Environmental Health Sciences. National Institutes of Health (NIH). Department of Health and Human Service* (CDAH-1H6S0)8;7andfrohmy PtHhSe cNenatteiorngarlanCtsaCncAer-13l3n4s3iiluanled, NIH. DHHS. We thank Dr. Bernard Pasternack for permis sion to use data and biological specimens from the New York University Women's Health Study. We also thank Ms. Frances Mastrnta for technical aaFssrssaiinssttzaabnnlaccuee faoonfrd tthhDeeirrs.GsuuPpt.tpmoJar.nt aLnBadnredargsiutgaidnaDniacaegn.ndoTstAhice. Institute and or the Institute's Medical Director Emeritus. Dr. Philip Strax. in all phase* of data collection is acknowledged. Manuscript received October 5. 1992: revised January 12. 1993; accepted January 14. 1993. L, irr At.: Organochlorine compounds in neo dichlorodiphcnyl dichlorneihane (DDE) in plastic and adjacent apparently normal breast (9) 4tMPi8sysU4yu.SesS.a1Al9BL7ou6OllH-RE,anfCvtiHroAAnMt.:CArOAmclcauHmr.rTenoHcxeaic.SooAlfN10E5-Nh:4ex7aE8*-, chlorncydohexane in breast cancer patients. (JO) TfCtoannicameir.l:o6E6P:nG2d1.o2gP4e-an2so1tu2es8r.nhao1cr9mk90onBeS*, Shore; RE. and breast cCaanncceerr: RAe*pTror*epatcct)i8vfe.sucpophlo)r:t23s-2tu6d, y.199B1reast human milk: Effects on growth, morbidity, and duration of lactation. Am J Pub) Health (26) B7 7y.e1r2s9 4T-1, 2 9G7r. ah19a8m7 Lactation and breast cSan. ceRrz.fA.pkma J T. et ai.: Epidemiol 121:664-674. 1985 (27) Wti.i>nr W: The search for the causes of breast and colon cancer. Nature 338:389-394. (28) HD19ioe8wt9afr.yGRfa.ctHorisroahnadtarisTk. Hofistb.reasGt, chatncaerl:: FoRf EGEoverCnmAeTntABLoOoksG Send fo r your copy today1 Free Catalog WashingtoBnn*D3C7002(0X)JJ-7iXV m m m am m m m m m m m m m 652 REPORTS Journal of the National Cancer Institute, Vol. 85. No. 8. April 21, 1993