Obesity and breast cancer in premenopausal women: Current evidence and future perspectives

(1)

Full

length

article

Obesity

and

breast

cancer

in

premenopausal

women:

Current

evidence

and

future

perspectives

Daniela

Laudisio

1

,

Giovanna

Muscogiuri

1

,

Luigi

Barrea,

Silvia

Savastano,

Annamaria

Colao

*

DepartmentofClinicalMedicineandSurgery,University“FedericoII”,Naples,Italy

ARTICLE INFO Articlehistory: Received4January2018

Receivedinrevisedform17March2018 Accepted25March2018 Keywords: Breastcancer Premenopausal Obesity Insulin-resistance ABSTRACT

Thereisraisingevidencereportinganincreasedincidenceofbreastcanceroverthepastdecades.Every yearapproximately1.4millionnewcasesofbreastcancerarediagnosedworldwide,withamortalityrate ofapproximately450,000/year.Outofthesecases,6.6%arediagnosedinpremenopausalwomenwitha medianageatdiagnosisof40years:inpremenopausalwomenbreastcancerseemstobemoreaggressive thaninpost-menopausalwomen.Obesityhasbeenreportedtoincreasetheriskofdevelopingbreast cancerandtoworsentheprognosis.Thisseemstobeduetoseveralobesity-relatedmechanisms.Insulin resistancethatoftenoccurswithobesitymayresultsincompensatoryhyperinsulinemia.Insulin cross-bindsinsulin-likegrowthfactor-Ireceptorsexpressedonbreastcells,resultinginproliferativestimulion breastcancercells.Besides,hyperinsulinemiaup-regulatesthegrowthhormonereceptor(GHR)thus increasingGHRstimulationandresultinginanincreasedhepaticIGF-Isynthesis.Moreover,insulin decreasesthehepaticexpressionofbindingproteinsofIGF-I,suchasinsulin-likegrowthfactorbinding proteins(IGFBP)-1andIGFBP-2,thusleadingtohighcirculatingandbioavailablefreeIGF-I.Additionally, obesityisassociatedtochronic low-gradeinﬂammationthathasbeenreportedtobeanadditional stimulusfortumorgrowth.Thisreviewshowsthecurrentevidenceregardingtotheassociationof obesityandbreastcancerinpremenopausalstatefocusingonbothhumanandbasicstudies;showing thattheobesityisariskfactorforbreastcanceralsoamongpremenopausalwomen,especiallyforthe molecularsubtypeTripleNegativeBreastCancer.

Introduction

Breastcancer(BC)isagrowingpublichealthproblem.Although currenteffortsinpreventingBC,theincidenceisincreasinginmost developed and developing countries[1–4]. Eachyear about1.4 millionnewcasesofBCarediagnosedworldwide,withamortality rate of approximately 450,000/year [5]. Of these, 6.6% are

diagnosed in premenopausal women, with a median age at diagnosisof40years[6,7].InpremenopausalwomenBCseemsto be more aggressive than in post-menopausal women, showing higher histological grading, increased proliferation rate, higher ratesof vascularinvasion[8] anda higherproportionof triple-negativebreastcancer(TNBC)[9].Further,BCinpremenopausal womenhasbeenassociatedtoanincreasedriskofrecurrenceand mortalityratecomparedtopostmenopausalwomen[9,10].Several risk factors for BC have been identiﬁed, such as age, genetic mutations(BRCA1andBRCA2)[11],youngerageatmenarche,ﬁrst pregnancyaftertheage30,nulliparity,olderageatmenopause, densebreasttissue[12],hormonereplacementtherapy,useoforal contraceptives [13], personal and family history of BCor other breast diseases [14]. Recently, it hasbeenalsohighlighted that obesitycouldrepresentanimportantriskfactorofdevelopingBC [15,16].Thisseemstobeduetoseveralmetabolicderangements associatedtoobesity.First,obesityisoftenassociatedtoinsulin resistance thatcouldresultin secondaryhyperinsulinemiawith consequent cross-binding of insulin-like growthfactor-I (IGF-I)

Abbreviations:BC,breastcancer;BMI,bodymassindex;TNBC,TripleNegative BreastCancer;GHR,growthhormonereceptor;IGF-1,insulin-likegrowthfactor-I IGFBPFactorBindingProtein;ER,estrogenreceptor;IL,interleukin;NF-kB,nuclear factor-kappaB;MMP-9,metalloproteinase-9;MCP-1,monocytechemoattractant protein-1;HIF-1,hypoxiainduciblefactor-1;PI3K,phosphatidylinositol3-kinase; Akt,proteinkinaseB;MAPK,mitogen-activatedproteinkinase;VEGF,vascular endothelialgrowthfactor.

* Corresponding author at: Department of Clinical Medicine and Surgery, University“FedericoII”,ViaS.Pansini5,80131,Napoli,Italy.

E-mailaddress:[email protected](A.Colao).

1

Theseauthorsequallycontributedtothismanuscript. https://doi.org/10.1016/j.ejogrb.2018.03.050

ContentslistsavailableatScienceDirect

European

Journal

of

Obstetrics

&

Gynecology

and

Reproductive

Biology

(2)

receptors, thus exerting a mitogen effect on both normal and neoplasticbreast epithelial cells [17]. Furthermore,obesity has beenassociatedtoasubclinical,chronic,low-gradeinﬂammation thatrepresentsanadditionalpredisposingfactorforcancer[18]. Thus,theaimofthismanuscriptistoprovideageneraloverviewof the current evidence on the association of BC and obesity in premenopausalstate,reviewingbothclinicalandbasicstudies. ObesityandBreastCancer

Clinicalevidence

Several studies investigatedtheassociation betweenobesity and BC [19–21], providing conflicting results [22,23]. In a longitudinalstudyMichelsetal.reportedaninverseassociation betweenbodymassindex(BMI)andtheriskofBC(p<0.001)in premenopausalwomen [22]. Thewomen includedinthestudy aged25–42yearsandwereaskedtoreplytoaself-administered questionnaireabouttheirmedicalhistoryandlifestyle.Theywere followedfor 14years and were askedtoupdate questionnaires information on demographic variables, lifestyle factors, and medicaleventseverytwoyears.Thisstudyreportedinteresting resultbuttheself-administrationofthequestionnairerepresents animportantbias[22].SameresultswerereportedbyCecchini etal.inachemopreventionstudyfindingthatBMIhadasignificant relationwithincreasedriskofBCamongpremenopausalwomen >35years of age, although this result was not confirmed in postmenopausal women [23]. This study used data from the NationalSurgicalAdjuvantBreastandBowelProject(NSABP)P-1 (BreastCancerPreventionTrial)andSTAR(StudyofTamoxifenand Raloxifene)thatwere2-arms,double-blinded,randomizedclinical trialsinvestigatingtheuseofchemopreventionforBC.Thewomen includedinthestudywereaged>35years,didnothavehistoryof invasiveBCbutpresentedahighriskofdevelopingBCevaluatedby Gailscore(Gail score1.66).Increased risk ofinvasive BC was significantlyassociatedwithhighBMI(p=0.01)inpremenopausal women [23]. Furthermore, the authors found a statistically significantassociationbetweenBMIandestrogenreceptor (ER)-positiveBC,withhazardratiosforthe2uppercategoriesofBMIof 1.41and1.78(p=0.04)[23].Thistrendwasnotconfirmedfor ER-negativeBCandthiscouldbeduetothesmallnumberofBCevents inthisgroup.Inthisstudy,however,theassessmentofBMIwas performedonlyatthestudybeginningsopreventingtodrawfirm conclusionsonthelong-termeffectofhighBMI[23].Theeffectof obesityonpremenopausalBCriskdiffersacrossdiseasesubtypes. Several studies have reported a positive association between obesityandtheriskofdevelopingTNBCinpremenopausalwomen [24–26]. TNBCsubtypeisparticularly aggressiveand frequently occurs in premenopausalwomen [24,26,27]. In a recent retro-spectivestudy,Sahinetal.evaluatedtheassociationbetweenBMI andimmunohistochemicalsubtypesinBCforpremenopausaland postmenopausalwomen [25]. Inthestudy populationBMI was stratifiedinto3groupsaccordingtoBMIasnormal-weighted(BMI <25kg/m2_),_{over-weighted}_(BMI₂₅

–29.9kg/m2_),_and_obese_(BMI

30kg/m2_),_{immunohistochemical}_classi_ﬁcation_of_the_tumors_was

categorizedinto4groups:luminal-like,HER2/luminal-like, HER2-like,andTNBCaccordingtotheER/PRandHER2status.Thisstudy demonstrated that there was a high prevalence of obesity at diagnosis of BC in both premenopausal and postmenopausal women(p<0.001andp<0.001,respectively)TNBCsubtypewas signiﬁcantly more common in obese premenopausal women comparedtowomenwithBMI<30kg/m2_(p₌_0.007)._In_addition,

obesepremenopausalpatientsexperiencedlesscommon luminal-like subtype (p=0.033), and more frequently presented with higher tumor stage (p=0.012) and tumor grade (p=0.004) at diagnosiswhencomparedtopatientswithBMI<25kg/m2_[₂₅_].

These data confirm the results obtained bya case–casestudy, whichhasreportedthepositiveassociationofriskforTNBCwith both weight and BMI in premenopausal women (p=0.012 and 0.004,respectively)witha5%increaseinriskobservedfor5kg increaseinweightanda16%increaseinriskper5kg/m2increasein BMI[26].The Cancerand Steroid Hormone(CASH) population-based, case-controlstudy, foundan increased risk for TNBC in premenopausal women reporting a strong positive association betweenBMIandpremenopausalTNBCrisk(OR(oddsratio)=1.67, 95%CI(confidenceinterval)1.22–2.28;p=0.026)[28].Conversely Yangetal.reportedthatobesityinyoungerwomenwasassociated withan increasedrisk ofdeveloping ER-positiveor PR-positive tumorsratherthanTNBCsuggestingthatBMIseemstobemore associatedwithhormonereceptorpositivetumors[29].However, theresultsofthisstudywerelimitedbythefactthatdatawere pooledfromdifferentstudiesthatwerenottotallycomparablein terms of enrolled subjects and experimental design [29]. The associationbetweenBCandobesityseemstobemediatedby intra-abdominalvisceralfat thatis themain responsiblefor obesity-relatedmetabolicandhormonalchangespromotingthe develop-ment of BC [30]. Another prospective study, evaluated the associationbetween abdominalfat, measured as waistand hip circumferencesandbythewaist-to-hipratio,andtheriskofBCin premenopausalwomen[31]:116.430womenwereincludedand duringa12-yearsfollowup630casesofpremenopausalinvasive BC were recorded. Abdominal adiposity was not significantly associatedwithoverallincidenceofpremenopausalBC:however, butadipositywasassociatedwithriskofdevelopingofER-negative BCmorestronglythanwiththeriskofER–positiveBC.Although these results were promising, they shouldbe interpreted with caution due tothe fact that theassessment of anthropometric measurements was self measured by the patients. Further, the measurementofwaistcircumferenceisanindirectmeasurement of visceraladipose tissue, notproviding an accurate evaluation [31]. Further research with more detailed measurements of abdominal fat with methods not investigator-dependent is requiredtoelucidatetheexactroleplayedbytotalandabdominal fatmassindeterminingBCrisksoidentifyingpotentialtargetsfor treatmentintervention.

Pathogenesis

Severalmechanismsareinvolvedintheassociationofobesity and increased risk of developing BC. In lean premenopausal women,estrogensaremainlyproducedbyovariangranulosacells and only a little part is produced by other peripheral tissues especially the adipose one [32] while in obese premenopausal women, most of estrogens derive from the conversion of androgens into estrogens by aromatase in the adipose tissue [32].Highlevelsofestrogensreleasedintothecirculationfromthe adiposetissueactivatethenegativefeedbackinthehypothalamus pituitary axisleading toreduced gonadotrophin secretion[33], thusresultinginamenorrheaandreducedovarianactivitywitha markedly reduced synthesis of progesterone [34]. Initially, this mechanismhasbeenconsideredasprotectiveagainsttheriskofBC inobesepremenopausalwomen,becauseitwashypothesizedthat progesterone wouldincrease proliferation ofbreast cells in the luteal phase when progesterone levels are usually high [35]. However, this theory has not been supported by subsequent studiesthathavehypothesizedthatprogesteroneisneutraloreven protectiveforBC[36].Infactithasbeenhypothesizedtoeither decreaseBCrisk,bymitigatingtheestrogen-inducedproliferation in breast epithelial cells [37]. Cohort studies have examined circulating progesterone levels and BC risk in premenopausal womenandhavereportedinverseassociationbetweencirculating progesterone levels and risk of BC. The European Prospective

(3)

Investigation on Cancer and Nutrition (EPIC), investigated the associationbetweensexsteroidhormonesandBCoccurrencein premenopausalwomen,reportingastatisticallysignificantinverse relationship between serum levels of progesterone and BC incidenceOR for highest versuslowest quartile (OR=0.61,95% CI=0.38–0.98;p=0.06)[38].Inanotherprospectivenested case-controlstudySchernhammeretal.haveinvestigatedthe associa-tionbetweencirculatingsexsteroidsandBCriskinpremenopausal womenfindingthatendogenousprogesteronewasnotstatistically associatedwithBC(OR,1.16;95%CI,0.60–2.27;p=0.75)[39].In additiontohormonalmechanisms,metabolicfactorshavebeen hypothesizedtobeinvolvedinthelinkbetweenobesityandBC. Several studies have confirmed that insulin resistance and compensatory hyperinsulinemia have been associated with increasedriskofBCandwithaworseprognosticoutcomesboth in premenopausal and postmenopausal women [40–43]. IGF-I receptorsarethemainmediatorsofproliferativeeffectsofinsulin. Thiscouldbeduetothehighhomology(morethan50%)ofthe insulinreceptor(IR)andIGF-Ireceptor(IGF-IR)thatiseven84%at the

a

-subunitofthetyrosinekinasedomain[44].Further,insulin andIGF-Ishare40–50%homology.Becauseofthis,insulinand IGF-IcaninteracteitherwithIRorwithIGF-IR[45],andbothofthese receptorshavebeenassociatedwithtumordevelopment[46].It hasalsobeenreportedanoverexpressionofIGF-IRinBCthatis functionalforestablishinganenhancedanabolicstatenecessary for cell proliferation, differentiation, and anti-apoptosis via deregulating or over-activating multiple downstream signaling pathways, including the phosphatidylinositol 3-kinase (PI3K)/ protein kinase B (Akt) and mitogen-activated protein kinase (MAPK)pathways.Thisoverexpressionseemstobestimulatedby highIGF-Iandinsulinlevels[47–49].Instead,theindirecteffectsof hyperinsulinemiaoncarcinogenesisareattributabletotheaction of insulin on circulating endogenous growth factors and their bindingproteins.Inhyperinsulinemicstates,thegrowthhormone receptor(GHR)isupregulatedbytheincreaseofinsulin concen-trationsinportalcirculationthustheincreaseofGHRstimulation

result in an increased hepatic IGF-I synthesis [50]. Moreover, insulin decreases thehepatic expressionof binding proteins of IGF-I,suchasinsulin-likegrowthfactorbindingproteins(IGFBP)-1 and IGFBP-2 [51], thus leading to high plasma levels and bioavailabilityoffreeIGF-I[52].Inthiscontext,IGFBP-1inhibits thegrowthofBCcellsinmicewithMCF-7BCxenografts,aswellas hepatocellular cancergrowth inmiceover-expressingIGF-I and IGF-II[53,54].Additionally,obesityisaccompaniedbyincreased circulating levelsofpro-inﬂammatory mediators,suchas inter-leukin(IL)-6,tumornecrosisfactoralpha(TNF

a

), matrix metal-loproteinase (MMP)-9 and IL-1

b

, that have been reported to promote tumorigenesis [55]. Additionally obesity enhancesthe secretion ofmonocytechemo-attractantprotein(MCP)-1,which stimulates the recruitment of macrophages to adipose tissue, including the breast one. These tumor-associated macrophages likely contribute to tumor growth by increasing local and/or systemic inflammatory and angiogenic factors and generating reactiveoxygenspecies[56].Thus,theincreaseofpro-in flamma-torycytokinessecretioncreatesafavorableenvironmentinducing tumorcellstoacquireaphenotypewithmajorinvasivenessand aggressiveness [57–60] (Fig. 1). Furthermore, the progressive hypertrophyofwhiteadipocytesandtheexpansionofthewhite adipose tissue make the same adipocytesmoreaway fromthe vascular network, withthereduction ofthe oxygenavailability [61]. Inhypoxia statetheadiposetissue generatesanincreased state of oxidative stress, able to make the white adipocytes hypertrophic/hypoxicandtosecreteinflammatoryproteinsandto inducebrownadipocytesdysfunction[62].Hypoxia,furthermore, induces a status of insulin resistance [63,64], ischemia and necrosis of white adipocytes [65]; production and release of inflammatorycytokinesandangiogenicfactors[66].Thedecrease oxygentensionactivatesthetranscriptionfactorhypoxiainducible factor(HIF)-1generatingaseriesofnegativeeffectsasinhibitionof productionofadiponectinbywhiteadipocytesandincreasedlevel ofleptin[67].Infactadiponectinappearstohavearegulatoryrole in insulin resistance and to exert antineoplastic activities, a

Fig.1.PotentialMechanismsofobesityleadingtocancerdevelopment.Schematicrepresentationofaspectsofobesitythatshowshowinsulinresistanceandinﬂammation maypromote,directlyandindirectly,breastcancergrowth.IGF-1:insulin-likegrowthfactor-1;IR-A:insulinreceptortypeA;IR-B:insulinreceptortypeB;SHBG:sexbinding protein;IGFBP1/2:insulin-likegrowthfactorbindingprotein1/2;IL-1b:interleukin-1beta;IL-6:interleukin-6;TNF-a:tumornecrosisfactor-alpha.

(4)

reductionintheproliferationofadipocytecells,endothelialcells andtumorcells[68].Inaddition,adiponectinhasbeenreportedto block angiogenesis by decreasing the expression of vascular endothelialgrowth factor(VEGF)and Bcl-2(anti-apoptotic)and increasingtheactivityof p53,Baxand caspase (pro-apoptotic), with resulting in apoptosis of endothelial cells. Likewise, adiponectinwasshowntoreduceTNF-

a

inducedeffectsoncell proliferation and migration [69,70]. Conversely low levels of adiponectinexertpro-inﬂammatoryeffectsrisingthesecretionof several proinﬂammatory cytokines including TNF-

a

and IL-6, leadingtotheonsetofatumorigenicmicroenvironment promot-ingtumor development[71]. Leptin,closely related to adipose tissuemassandvolumeofadipocytes[72],hasinsteadbeenshown tohavecarcinogenicproperties,increasingtheexpressionof anti-apoptoticproteins, inﬂammatory markers (TNF-

a

, IL-6), angio-genic factors (VEGF), and also the hypoxia-inducible factor-1a (HIF-1

a

) by promoting cancer cell survival, proliferation and migration[73].(Fig.2).

Conclusion

Epidemiologicalstudiesindicateprogressivelyincreasednumber of cases with BC in most developed and developingcountries. Premenopausal state at diagnosis is highly associated with a signiﬁcantlyincreasedriskofrecurrenceandhighermortalityrate. ObesityhasbeenreportedtobeariskfactorforBC,especiallyforthe molecularsubtypeTNBC.Theeffectsofobesityontheriskofbreast cancerinpremenopausalaremediatedbymolecularmechanismsas compensatory hyperinsulinemiatoinsulin resistanceand high levels ofinsulin andIGF-1,increasedpro-inﬂammatorycytokinessecretion (TNF-

a

and IL-6), changes in the leptin and the hypoxia. This informationisimportanteitherforOncologistandNutritioniststo increases theknowledge on the possible relationships between obesityandbreastcancerinpremenopausalwomen,withtheaim notonlytoreducetheriskofbreastcancerinobesepatientbutalso fordelineatingtheriskproﬁleformetabolicriskfactorsthatmay resultfromobesity.Beingobesityamultifactorialdiseaseandan importantrisk factorforBC,furtherstudyshould befocusedto implementindividual lifestyle recommendations based onboth preventionofcancerriskandweightmanagement.

References

[1]EcclesS.A.,AboagyeEO,AliS,AndersonAS,ArmesJ,BerditchevskiF,etal. Criticalresearchgapsandtranslationalprioritiesforthesuccessfulprevention andtreatmentofbreastcancer.BreastCancerRes2013;15:R92.

[2]ArnoldM,Karim-KosHE,CoeberghJW,ByrnesG,AntillaA,FerlayJ,etal. Recenttrendsinincidenceofﬁvecommoncancersin26Europeancountries since1988:analysisoftheEuropeanCancerObservatory.EurJCancer2015;51 (9):1164–87.

[3]RahibL,SmithBD,AizenbergR,RosenzweigAB,FleshmanJM,MatrisianLM. Projectingcancerincidenceanddeathsto2030:theunexpectedburdenof thyroid, liver, and pancreas cancers in the United States. Cancer Res 2014;74:2913–21.

[4]ColditzGA,BohlkeK.Prioritiesfortheprimarypreventionofbreastcancer.CA CancerJClin2014;64:186–94.

[5]BotrelTE,PaladiniL,ClarkOA.Lapatinibpluschemotherapyorendocrine therapy(CET)versusCETaloneinthetreatmentofHER2 overexpressinglo-callyadvancedormetastaticbreastcancer:systematicreviewand meta–analy-sis.CoreEvid2013;8:69–78.

[6]FerlayJ, ShinHR,BrayF,FormanD,MathersC, ParkinDM.Estimates of worldwideburdenofcancerin2008:GLOBOCAN2008.IntJCancer201012:. [7]NarodSA.Breastcancerinyoungwomen.NatRevClinOncol2012;9:460–70. [8]AzimJrHA,PartridgeAH.Biologyofbreastcancerinyoungwomen.Breast

CancerRes2014;16(4):427.

[9]CollinsLC,MarottiJD,GelberS.Pathologicfeaturesandmolecularphenotype bypatientageinalargecohortofyoungwomenwithbreastcancer.Breast CancerResTreat2012;131(3):1061–6.

[10]VoogdAC,NielsenM,PeterseJL.Differencesinriskfactorsforlocalanddistant recurrenceafterbreast-conservingtherapyormastectomyforstageIandII breastcancer:pooledresultsoftwolargeEuropeanrandomizedtrials.JClin Oncol2001;19(6):1688–97.

[11]MetcalfeK,LubinskiJ,LynchHT,GhadirianP,FoulkesWD,Kim-SingC,etal. FamilyhistoryofcancerandcancerrisksinwomenwithBRCA1orBRCA2 mutations.JNatlCancerInst2010;102(24):1874–8.

[12]Collaborative Group on Hormonal Factors in Breast Cancer. Menarche, menopause, andbreast cancer risk:individualparticipant meta-analysis, including 118964 women with breast cancer from 117 epidemiological studies.LancetOncol2012;13(11):1141–51.

[13]GierischJM,CoeytauxRR,UrrutiaRP,HavrileskyLJ,MoormanPG,LoweryWJ, etal. Oral contraceptive useand riskof breast,cervical,colorectal,and endometrialcancers:asystematicreview.CancerEpidemiolBiomarkPrev 2013;22:1931–43.

[14]Ritte R, Tikk K, Lukanova A, Tjonneland A, Olsen A, Overvad K, et al. Reproductivefactorsand riskofhormonereceptorpositive andnegative breastcancer:acohortstudy.BMCCancer2013;13:584.

[15]RiceMS,EliassenAH,HankinsonSE,LenartEB,WillettWC,TamimiRM.Breast cancerresearchinthenurses'HealthStudies:exposuresAcrossthelifecourse. AmJPublicHealth2016;106(9):1592–8.

[16]CastelloA,MartínM,RuizA.LowerBreastcancerriskamongwomenfollowing theworldcancerresearchfundandamericaninstituteforcancerresearch lifestylerecommendations:epiGEICAMcase-controlstudy.PLoSOne2015;10 (5):e0126096.

Fig.2.Potentialmechanismsofobesityleadingtocancerdevelopment.Schematicrepresentationofaspectsofobesitythatshowshowthechangesinadiposetissueduring weightgain.Duringobeseadiposetissueexpansion,preadipocytedifferentiationisimpaired,andhypoxiaactivateshypoxia-induciblefactor1(HIF-1)withconsequently decreaseadiponectinexpressionandupregulatleptin,secretionofinﬂammatorycytokines,includingtumornecrosisfactora(TNF-a),interleukin-6(IL-6),inductionastatus ofinsulinresistanceandicreasedofoxidativestress.

(5)

[17]Vona-DavisL,RoseDP.Type2diabetesandobesitymetabolicinteractions: commonfactorsforbreastcancerriskandnovelapproachestopreventionand therapy.CurrDiabetesRev2012;8:116–30.

[18]LashingerLM,FordNA,HurstingSD.Interactinginﬂammatoryandgrowth signalsunderlietheobesity-cancerlink.JNutr2014;144:109–13.

[19]WhiteAJ,NicholsHB,BradshawPT,SandlerDP.Overallandcentraladiposity andbreastcancerriskintheSisterStudy.Cancer2015;121:3700–8. [20]NeuhouserML,AragakiAK,PrenticeRL,MansonJE,ChlebowskiR,CartyCL,

etal.Overweight,obesity,andpostmenopausalinvasivebreastcancerrisk:a secondaryanalysisoftheWomen’sHealthInitiativerandomizedclinicaltrials. JAMAOncol2015;1:611–21.

[21] SebastianiF,CortesiL,SantM,LucariniV,CirilliC,DeMatteisE,etal.Increased incidenceofbreastcancerinpostmenopausalwomenwithhighbodymass indexattheModenaScreeningProgram.JBreastCancer2016;19:283–91. [22]MichelsKB,TerryKL,WillettWC.Longitudinalstudyontheroleofbodysizein

premenopausalbreastcancer.ArchInternMed2006;166:2395–402. [23]CecchiniRS,CostantinoJP,CauleyJA,CroninWM,WickerhamDL,LandSR,

etal.Bodymassindexandtheriskfordevelopinginvasivebreastcancer amonghigh-riskwomeninNSABPP-1andSTARbreastcancerprevention trials.CancerPrevRes(Phila)2012;5:583–92.

[24]KawaiM,MaloneKE,TangMT,LiCI.Height,bodymassindex(BMI),BMI change,andtheriskofestrogenreceptorpositive,HER2-positive,and triple-negative breast cancer among women ages 20 to 44 years. Cancer 2014;120:1548–56.

[25]SahinS,Erdem GU, KaratasF,Aytekin A,Sever AR,Ozisik Y,et al. The associationbetweenbodymassindexandimmunohistochemicalsubtypesin breastcancer.Breast2017;32:227–36.

[26]ChenL,CookLS,TangMT,PorterPL,HillDA,WigginsCL,etal.Bodymassindex andriskofluminal,HER2-overexpressing,andtriplenegativebreastcancer. BreastCancerResTreat2016;157(3):545–54.

[27] DolleJM,DalingJR,WhiteE,etal.Riskfactorsfortriple-negativebreastcancer inwomen under the age of 45 years.Cancer EpidemiolBiomark Prev 2009;18:1157–66.

[28]GaudetMM,PressMF,HaileRW,LynchCF,GlaserSL,SchildkrautJ,etal.Risk factorsbymolecularsubtypesofbreastcanceracrossapopulation-based study of women 56 years or younger. Breast Cancer Res Treat 2011;130:587597.

[29]YangXR,Chang-ClaudeJ,GoodeEL,CouchFJ,NevanlinnaH,MilneRL,etal. Associationsofbreast cancerriskfactorswithtumorsubtypes: apooled analysisfromtheBreastCancerAssociationConsortiumstudies.JNatlCancer Inst2011;103(3):250e63.

[30]VanGemertWA,MonninkhofEM,MayAM,EliasSG,vanderPalenJ,Veldhuis W,etal.Associationbetweenchangesinfatdistributionandbiomarkersfor breastcancer.EndocrRelatCancer2017;24(6):297–305.

[31] HollyRH,WalterCW,KathrynLT,KarinBM.Bodyfatdistributionandriskof premenopausalBreastcancerinthenurses’healthstudyII.JNatlCancerInst 2011;103(3):273–8.

[32]Simpson ER. Aromatase:biologic relevance of tissue-speciﬁc expression. SeminReprodMed2004;22(1):11–23.

[33]PolotskyAJ,HailpernSM,SkurnickJH,LoJC,SternfeldB,SantoroN.Association ofadolescentobesityandlifetimenulliparity-thestudyofwomen’shealth acrossthenation(SWAN).FertilSteril2010;93:2004–11.

[34]JainA,PolotskyAJ,RochesterD,BergaSL,LoucksT,ZeitlianG,etal.Pulsatile luteinizinghormoneamplitudeandprogesteronemetaboliteexcretionare reducedinobesewomen.JClinEndocrinolMetab2007;92:2468–73. [35]Lanari C, Molinolo AA. Progesterone receptors–animal models and cell

signallinginbreastcancer.Diverseactivationpathwaysfortheprogesterone receptor:possibleimplicationsforbreastbiologyandcancer.BreastCancer Res2002;4(6):240–3.

[36]CampagnoliC,AbbaC,AmbroggioS,PerisC.Pregnancy,progesteroneand progestinsin relationto breast cancer risk.JSteroid BiochemMol Biol 2005;97:441–450et.

[37] Doisneau-SixouSF,SergioCM,CarrollJS,HuiR,MusgroveEA,SutherlandRL. Estrogenandantiestrogenregulationofcellcycleprogressioninbreastcancer cells.EndocrRelatCancer2003;10(2):179–86.

[38]KaaksR,BerrinoF,KeyT,RinaldiS,DossusL,BiessyC,etal.Serumsexsteroids in premenopausal women and breast cancer risk within the European PropsectiveInvestigationintoCancerandNutrition(EPIC).JNatlCancerInst 2005;97:755–65.

[39]SchernhammerES, SperatiF,RazaviP, etal. Endogenoussexsteroidsin premenopausalwomenandriskofbreastcancer:theORDETcohort.Breast CancerRes2013;15:R46.

[40]MutiP,QuattrinT,GrantBJ,KroghV,MicheliA,SchünemannHJ,etal.Fasting glucoseisariskfactorforbreastcancer:aprospectivestudy.CancerEpidemiol BiomarkPrev2002;11:1361–8.

[41]SchairerC, Hill D, SturgeonSR, FearsT, PollakM,Mies C,et al. Serum concentrationsofIGF-I,IGFBP-3andc-peptideandriskofhyperplasiaand cancerofthebreastinpostmenopausalwomen.IntJCancer2004;108:773–9. [42]MichelsKB,SolomonCG,HuFB,RosnerBA,HankinsonSE,ColditzGA,etal. Type2diabetesandsubsequentincidenceofbreastcancerintheNurses’ HealthStudy.DiabetesCare2003;26:1752–8.

[43]GoodwinPJ,EnnisM,PritchardKI,TrudeauME,KooJ,MadarnasY,etal.Fasting insulinandoutcomeinearly-stagebreastcancer:resultsofaprospective cohortstudy.JClinOncol2002;20:42–51.

[44]GunterMJ,HooverDR,YuH,Wassertheil-SmollerS,RohanTE,MansonJE,etal. Insulin, insulin-like growth factor-I, and risk of breast cancer in post-menopausalwomen.JNatlCancerInst2009;101(1):48–60.

[45]GunterMJ,HooverDR,YuH,Wassertheil-SmollerS,MansonJE,LiJ,etal.A prospective evaluationofinsulin and insulin-likegrowth factor-I asrisk factors for endometrial cancer.Cancer EpidemiolBiomark Prev 2008;17 (4):921–9.

[46]HumpertPM,DjuricZ,ZeugeU,OikonomouD,SereginY,LaineK,etal.Insulin stimulatestheclonogenicpotentialofangiogenicendothelialprogenitorcells byIGF-1receptor-dependentsignaling.MolMed2008;14:301–8.

[47]ParekhN,GuffantiG,LinY,Ochs-BalcomHM,MakaremN,HayesR.Insulin receptorvariantsandobesity-relatedcancersintheFraminghamheartstudy. CancerCausesControl2015;26(8):1189–95.

[48]ArgilesJM,López-SorianoFJ.Insulinandcancer(review).IntJOncol2001;18 (4):683–7.

[49]ArcidiaconoB,IiritanoS,NoceraA,PossidenteK,NevoloMT,VenturaV,etal. Insulin resistance and cancer risk: an overview of the pathogenetic mechanisms.ExpDiabetRes2012;2012:789174.

[50]GallagherEJ,LeRoithD. Theproliferatingroleof insulinand insulin-like growthfactorsincancer.TrendsEndocrinolMetab2010;21(10)610–853. [51]RenehanAG,FrystykJ,FlyvbjergA.Obesityandcancerrisk:theroleofthe

insulin-IGFaxis.TrendsEndocrinolMetab2006;17(8):328–36.

[52]Héron-MilhavetL,LeRoithD.Insulin-likegrowthfactorIinduces MDM2-dependentdegradationofp53viathep38MAPKpathwayinresponsetoDNA damage.JBiolChem2002;277(18):15600–6.

[53]ZhangX,YeeD.Insulin-likegrowthfactorbindingprotein-1(IGFBP-1)inhibits Breastcancercellmotility.CancerRes2002;62(15):4369–75.

[54]LuS,ArcherMC.Insulin-likegrowthfactorbindingprotein-1over-expression in transgenic mice inhibits hepatic preneoplasia. Mol Carcinog2003;36 (3):142–6.

[55]DivellaR,DeLucaR,AbbateI,NaglieriE,DanieleA.Obesityandcancer:therole ofadiposetissueandadipo-cytokines-inducedchronicinﬂammation.JCancer 2016;7(15):2346–59.

[56]LashingerLM,FordNA,HurstingSD.Interactinginﬂammatoryandgrowth signalsunderlietheobesity-cancerlink.JNutr2014;144:109–13.

[57]ZhangY,NowickaA,SolleyTN,WeiC,ParikhA,CourtL,etal.Stromalcells derivedfromvisceralandobeseadiposetissuepromotegrowthofovarian cancers.PLoSOne2015;10(8):e0136361.

[58]JungYY,LeeYK,KooJS.Expressionofcancer-associatedﬁbroblast-related proteinsinadiposestromaofbreastcancer.TumourBiol2015;36(11):8685– 95.

[59]Booth A, Magnuson A, Fouts J, Foster M. Adipose tissue, obesity and adipokines:roleincancerpromotion.HormMolBiolClinInvestig2015;21 (1):57–74.

[60]ParkEJ,LeeJH,YuGY,HeG,AliSR,HolzerRG,etal.Dietaryandgeneticobesity promoteliverinﬂammationandtumorgenesisbyenhancingIL-6andTNF expression.Cell2010;140(2):197–208.

[61]EnginA.Adiposetissuehypoxiainobesityanditsimpactonpreadipocytesand macrophages:hypoxiahypothesis.AdvExpMedBiol2017;960:305–26. [62]WoodIS,deHerediaFP,WangB,TrayhurnP.Cellularhypoxiaandadipose

tissuedysfunctioninobesity.ProcNutrSoc2009;68(4):370–7.

[63]HalbergN,KhanT,TrujilloME,etal.Hypoxia-induciblefactor1ainduces ﬁbrosis and insulin resistance in white adipose tissue. Mol Cell Biol 2009;29:4467–83.

[64]CaoY.Angiogenesisandvascularfunctionsinmodulationofobesity,adipose metabolism,andinsulinsensitivity.CellMetab2013;18(4):478–89. [65]TrayhurnP, AlomarSY.Oxygendeprivationandthe cellularresponse to

hypoxiainadipocytes–perspectivesonwhiteandbrownadiposetissuesin obesity.FrontEndocrinol(Lausanne)2015;6:19.

[66]ClosaD,Folch-PuyE.Oxygenfreeradicalsandthesystemicinﬂammatory response.IUBMBLife2004;56(4):185–91.

[67]TrayhurnP.Hypoxiaandadipocytephysiology:implicationsforadiposetissue dysfunctioninobesity.AnnuRevNutr2014;34:207–36.

[68]KhanS,ShuklaS,SinhaS,MeeranSM.Roleofadipokinesandcytokinesin obesity-associatedbreastcancer:therapeutictargets.CytokineGrowthFactor Rev2013;24(6):503–13.

[69]NakayamaS,MiyoshiY,IshiharaH,NoguchiS.Growth-inhibitoryeffectof adiponectinviaadiponectinreceptor1onhumanbreastcancercellsthrough inhibitionofS-phaseentrywithoutinducingapoptosis.BreastCancerRes Treat2008;112(3):405–10.

[70]KörnerA,Pazaitou-PanayiotouK,KelesidisT,KelesidisI,WilliamsCJ,Kaprara A,etal.Totalandhigh-molecular-weightadiponectininbreastcancer:invitro andinvivostudies.JClinEndocrinolMetab2007;92(3):1041–8.

[71]HebbardL,RanschtB.Multifacetedrolesofadiponectinincancer.BestPract ResClinEndocrinolMetab2014;28(1):59–69.

[72]MünzbergH,MorrisonCD. Structure,productionandsignaling ofleptin. Metabolism2015;64(1):13–23.

[73]Booth A, Magnuson A, Fouts J, Foster M. Adipose tissue, obesity and adipokines:roleincancerpromotion.HormMolBiolClinInvestig2015;21 (1):57–74.