Rese

ElevWoPap

Troy JAna CRolan

Abst

Intro

Cerextensvirusetionsyearsthe whas band p

AuthorInternaFrederiGenetiBranch4ProyecInvestigBiologíRica; an

Note: Smiology

Cance1954

Cancer

Epidemiology,

iomarkersrevention

arch Article

ated Systemic Levels of Inflammatory Cytokines in Oldermen with Persistent Cervical Human

B& P

illomavirus Infection

. Kemp1, Allan Hildesheim2, Alfonso García-Piñeres1, Marcus C. Williams1, Gene M. Shearer3,

ecilia Rodriguez4, Mark Schiffman2, Robert Burk6, Enrique Freer5, Jose Bonilla5, do Herrero4, and Ligia A. Pinto1

ractBac

a persMe

periphlymphability

Resversusfactor-ulocytand ILdue toLengthand Mwith csupern

Conan inc

een descostulated

s' Affiliationtional Corpck, Fredericcs, Nationa, CCR, Nato Epidemioación enEsa Celular yd 6Albert E

upplement, Biomarke

r Epidemio

kground: Defects in lymphoproliferative responses to mitogens/antigens in women >45 years old withistent type-specific human papillomavirus (HPV) infection have been reported.thods: To determine whether these defects were associated with altered cytokine profiles, plasma anderal blood mononuclear cell (PBMC) culture supernatants from 50 cases (oversampled for their reducedoproliferative ability) and 50 uninfected controls (oversampled for their robust lymphoproliferative) were examined for 24 cytokines using multiplexed bead–based immunoassays and ELISA.ults: The following plasma cytokines were significantly increased in cases relative to controls (casescontrols; median pg/mL): interleukin (IL)-6, 393.1 versus 14.5; IL-8, 1,128.5 versus 43.9; tumor necrosisα (TNF-α), 164.1 versus 9.2; macrophage inflammatory protein-1α (MIP-1α), 1,368.9 versus 25.5; gran-e macrophage colony-stimulating factor (GM-CSF), 13.8 versus 7.3; IL-1β, 8.3 versus 1.6 (all P < 0.0001);-1α, 218.2 versus 169.5 (P = 0.02). We focused our analysis on the cytokines IL-6, IL-8, TNF-α, and MIP-1αtheir high fold change (>10) and highly statistically significant difference between cases and controls.of persistence or type of infection (high risk and low risk) did not affect these differences. IL-6, TNF-α,IP-1α levels were also increased in unstimulated PBMC culture supernatants from cases comparedontrols (P < 0.05), however, the cytokine levels from phytohemagglutinin-stimulated PBMC cultureatants were significantly lower in the cases (P < 0.0001).clusions: Persistent HPV infection in older women with evidence of immune deficit is associated withrease in systemic inflammatory cytokines.act: Future studies are needed to determine whether the inflammatory profile is age dependent and to

Imp

examine the role that inflammatory cytokines play in HPV-induced progression from infection to cervicalcancer. Cancer Epidemiol Biomarkers Prev; 19(8); 1954–9. ©2010 AACR.

inducoccurWe

and rpersispersisperiphSev

duction

vical cancer and its precursor lesions have beenively shown to be caused by human papilloma-s (HPV; refs. 1, 2). The prevalence of HPV infec-in women peaks shortly after sexual debut (20-25old) and declines thereafter. In many regions oforld, a second wave of increased HPV prevalence

ribed among women older than 55 (3)to be the result of factors such as age-

munesentia

s: 1HPV Immunology Laboratory, Science Applicationsoration-Frederick, Inc., National Cancer Institute-k, Maryland; 2Division of Cancer Epidemiology andl Cancer Institute, NIH; 3Experimental Immunologytional Cancer Institute, NIH, Bethesda, Maryland;lógico Guanacaste, Fundación INCIENSA; 5Centro detructurasMicroscópicas andCentro de Investigación enMolecular, University of Costa Rica, San José, Costainstein College of Medicine, Bronx, New York

ary data for this article are available at Cancer Epide-rs & Prevention Online (http://cebp.aacrjournals.org/).

CurrentJosé, C

CorresNationaCorpor21702.mail.nih

doi: 10

©2010

l Biomarkers Prev; 19(8) August 2010

ed immune suppression or cervical changes thatwith menopause.have previously observed a decrease in mitogenecall antigen lymphoproliferation among womentently infected with HPV (4), suggesting that HPVtence in the cervix is associated with a decrease ineral immune fitness.eral factors can lead to a decrease in peripheral im-

esponsiveness. Cytokines and chemokines are es-for a multitude of immune-related activities such

ddress for A. García-Piñeres: University of Costa Rica, Sansta Rica.

onding Author: Ligia A. Pinto, HPV Immunology Laboratory,Cancer Institute-Frederick/Science Applications Internationalion-Frederick, Inc., Building 469, Room 205, Frederick, MDhone: 301-846-1766; Fax: 1-301-846-6954. E-mail: pintol@

rl

ao

platP

.gov

.1158/1055-9965.EPI-10-0184

American Association for Cancer Research.

as recexpanand re(5-7).cytokTo

olderwe evolderwithinfect

Mate

Partion obasedneopltails othe nferatinegatBrieflstudythe Cof woHPVof HPbutionclearhistorFor

screenwith rwithweighas parbasedwithfromyearsrollmphyto(n = 5follow(7-9 ytive reisticsSupplformeU.S. NethicaHP

ing fofinalporteconseand d

(10).is deinfecthistor[follo(7-9 ywere(n =ing advisit.termmonthsistentype-sfollow(7-9 ytors hfollow(7-9 ywomethey45, 51low-r6, 11,69, 70womeanalysistenAll

the fiHPVment)and cIL-10phagesis fa(MCPIFN-inlocyteand IMillipmeasua singtantsPHAtionedminima valuThe

Wilcoanalysidere

Resu

The

HPV and Inflammatory Cytokines

www.a

ruiting phagocytes to areas of insult, stimulating thesion of T cells and B cells on antigen recognition,gulating the activation state of the immune systemPrevious studies have noted associations betweenine alterations and various infection-related states.determine biomarkers of immune suppression inwomen who are persistently infected with HPV,aluated cytokine profiles in a subpopulation ofwomen with evidence of persistent infectionHPV and similarly aged women without HPVion.

rials and Methods

ticipants in this study were selected from a popula-f 10,049 women who participated in a population-natural history cohort study of HPV and cervicalasia in the province of Guanacaste, Costa Rica. De-n the design and methods of the main cohort andested case-control study evaluating lymphoproli-ve responses among HPV-positive and HPV-ive older women have been published (4, 8, 9).y, we previously conducted a nested case-controlto examine lymphoproliferative responses withinosta Rican natural history cohort where a groupmen (n = 284) older than 45 who were infected withwere compared with a similarly sized control groupV-negative women (n = 291) of the same age distri-. Testing was done on peripheral blood mononu-cells (PBMC) collected at the final visit of naturaly cohort study (7-9 years after enrollment).the present evaluation, we were interested ining for possible immune biomarkers associatededuced lymphoproliferative ability among womenpersistent HPV infection. We therefore selected ated subpopulation of women previously evaluatedt of the older women nested study described aboveon the following criteria: cases (n = 50), women

type-specific persistent HPV infection as denotedPCR results obtained at enrollment, follow-up (5-6after enrollment), and final visit (∼9th year after en-ent) and weak lymphoproliferative responses tohemagglutinin (PHA) or HPV16 L1 VLP; controls0), women with no evidence of HPV infection at-up (5-6 years after enrollment) and final visitears after enrollment) and strong lymphoprolifera-sponses to PHA and HPV16 L1 VLP. The character-of the selected cases and controls are described inementary Table S1. All participants provided in-d consent, and this study was approved by theational Cancer Institute and Costa Rica INCIENSAl committees.V DNA testing was evaluated at enrollment, dur-llow-up (5-6 years after enrollment), and at thestudy visit (∼9th year after enrollment) as re-d (4), PCR testing was done with the MY09/11

nsus primers and AmpliTaq Gold polymerase,ot blot hybridization was used for genotyping

of thementa

acrjournals.org

The control group (HPV DNA negative women)fined as not having a type-specific persistention at two essential time points within the naturaly cohort study of HPV and cervical neoplasiaw-up (5-6 years after enrollment) and final visitears after enrollment)]. Although the controlsHPV DNA negative at these time points, some6) had transitory HPV infections as detected dur-ditional clinic visits between enrollment and finalThe cases were divided into two groups: long-persistors (n = 21; median persistence, 108s) and short-term persistors (n = 29; median per-ce, 19 months). The long-term persistors have apecific HPV DNA positive result at enrollment,-up (5-6 years after enrollment), and final visitears after enrollment), and the short-term persis-ave a type-specific HPV DNA positive result at-up (5-6 years after enrollment) and final visitears after enrollment). Furthermore, we classifiedn as being infected with a high-risk HPV type ifwere positive for HPV type 16, 18, 31, 33, 35, 39,, 52, 56, 58, 59, or 68 and as being infected with aisk HPV type if they were positive for HPV type26, 32, 34, 40, 42, 53, 54, 55, 57, 61, 62, 64, 66, 67,, 71, 72, 73, 74, 81, 82, 83, 84, 85, or 89. Fourn were excluded from the high-risk and low-riskses because they had high-risk and low-risk per-t infections.heparinized plasma specimens were collected atnal visit of the natural history cohort study ofand cervical neoplasia (∼9 years following enroll-. The plasma was tested blindly for 22 cytokineshemokines [interleukin (IL)-2, IL-4, IL-5, IL-6, IL-8,, IL-13, IL-17, IL-1α, IFN-γ, granulocyte macro-colony-stimulating factor (GM-CSF), tumor necro-

ctor-α (TNF-α), monocyte chemotactic protein-1-1), macrophage inflammatory protein-1α (MIP-1α),ducible protein 10 (IP-10), RANTES, eotaxin, granu-colony-stimulating factor (G-CSF), IL-12, IL-15, IL-7,L-1β] using the Linco-plex assay (Linco Research,ore). Transforming growth factor-β1 (TGF-β1) wasred by ELISA (Biosource). IFN-α was measured asle analyte in a bead array (Biosource). Cell superna-from PBMCs incubated with AIM V medium orfor 48 hours were also assayed with the aforemen-cytokine assays. Specimens that were below theum detectable limit for any cytokine were assignede of 1/2 the lowest detectable limit.cytokine data were not normally distributed. The

xon rank-sum nonparametric test was used for allses comparing cases and controls. P < 0.05 was con-d significant.

lts

mean, median, and range of responses for each

24 different cytokines tested are shown in Supple-ry Table S2. The following plasma cytokines were

Cancer Epidemiol Biomarkers Prev; 19(8) August 2010 1955

signif(medifor ILP < 01α: 1,0.000218.2,analy

due tand cdifferof perisk;predoled ufromthe lealleletoryPBMCwithcontrTNF-P < 0as shotion oPBMCnatanalonein supculturcases,for ILMIP-1α: 3,372.3, 9,449.3; P < 0.0001 for all analytes;Fig. 2

Discu

Table sma velswomen with persistent H

IL-6n 5 5Res 5 9Mea 206 71MedRanP‡ <0.

IL-8n 5 5Res 9 1Mea 392 1,5MedRanP <0.

IL-1αn 5Res 10 1Mea 244 32MedRanP 0

GM-Cn 5Res 10 1Mea 14 3MedRanP <0.

TNF-αn 5Res 9 1Mea 61 18MedRanP <0.

MIP-1n 5Res 5 1Mea 576 2,1MedRanP <0.0001

Table sma velswomen with persistent H (Co

IL-1βn 5 5Res 4 8Mea 6 2MRP

*HPfolloenro†HPfolloafte‡Thtestcontrols.

Kemp et al.

Cance1956

icantly increased from cases relative to controlsan among cases, median among controls, P value-6: 393.1, 14.5, P < 0.0001; for IL-8: 1,128.5, 43.9,.0001; for TNF-α: 164.1, 9.2, P < 0.0001; for MIP-368.9, 25.5, P < 0.0001; for GM-CSF: 13.8, 7.3, P <1; for IL-1β: 8.3, 1.6, P < 0.0001; and for IL-1α:

169.5, P = 0.02; Table 1). Next, we focused our

sis on the cytokines IL-6, IL-8, TNF-α, and MIP-1αWe

who w

r Epidemiol Biomarkers Prev; 19(8) August 2010

o their high fold change (>10) between controlsases and having a highly statistically significantence between controls and cases. The lengthrsistence or type of infection (high versus lowFig. 1A-D) did not affect these differences. Theminance of inflammatory cytokines in the plasmas to examine the transient secretion of cytokines48-hour cultured PBMCs to determine whethervels of these cytokines produced by PBMCs par-d those observed in plasma. The same inflamma-cytokines were also increased in unstimulatedculture supernatant fluid from cases compared

controls (median among cases, median amongols, P value for IL-6: 471.4, 13.8, P < 0.0001; forα: 52.3, 11.8, P = 0.01; and for MIP-1α: 503.3, 8,.0001) except for IL-8 (6,360.9, 4,663.2, P = 0.09)wn in Fig. 2A. Finally, we measured the concentra-f cytokines in supernatants from PHA-stimulatedcultures. In contrast to the results from the super-

ts from the 48-hour culture of PBMCs with medium, the levels IL-6, IL-8, MIP-1α, and TNF-α measuredernatants collected from PHA-stimulated PBMCes were significantly lower in cases (median amongmedian among controls, for IL-6: 1,287.2, 3,122.8;-8: 14,059.8, 45,160.3; for TNF-α: 993.8, 3,312.1; for

B) compared with controls.

ssion

1. Elevated pla

cytokine lePV infection

from

HPV−* HPV+†

0

0 ponders (%) 0 6 n .1 5.5 ian .5 3.1 14 39ge 8.0-3,202.5 8.0-3,338.0

0001

0

0 ponders (%) 6 00 n .5 45.7 ian .9 28.5 43 1,1ge 1.6-3,615.8 17.8-7,535.2

0001

0

50 ponders (%) 0 00 n .8 8.5 ian .5 8.2 169 21ge 52.4-2,714.8 57.6-2,113.9

.02

SF

0

50 ponders (%) 0 00 n .5 3.6 ian .3 3.8 7 1ge 2.1-293.5 4.4-886.4

0001

0

50 ponders (%) 6 00 n .4 9.1 ian .2 4.1 9 16ge 0.8-833.6 12.3-608.1

0001

α

0

50 ponders (%) 4 00 n .0 64.9 ian .5 68.9 25 1,3ge 8.0-10,097.2 17.9-10,097.2

he

C

1. Elevated pla

ave previously obsere persistently infected

ancer Epidemiology,

cytokine lePV infection

rved that olderwith HPV had a

Biomarkers & Pr

fromnt'd)

HPV−* HPV+†

0

0 ponders (%) 8 4 n .6 3.6 ian .6 .3 ed 1 8

ange 1.6-73.9 1.6-536.5<0.0001

V− women (controls) were negative for any HPV type atw-up (5-6 y after enrollment) and final visit (9 y afterllment).V+ women (cases) tested positive on at least thew-up visit (5-6 y after enrollment) and final visit (9 yr enrollment) for a type-specific HPV infection.e P value was calculated using the Wilcoxon rank-sumto compare cytokine values between cases and

womenmarked

evention

decreFLU,derlyia weigand emightobservOur rcreasethe silevelsripherthis gof immThe

tory cinfectnovelevideactionto remthat ptolerathe ceperiphavechanipressSecondifferflamm

flammto indas HPthat whad a1α whthat tencedAlthohave(14), wnormOu

fromthe pery isphopBecauest prlikelyconduyearsliferaHPVOn

werestudiflammor ind

Figure(B) IL-8levels inhigh-ristype–spHR andwith theThere wdifferengroupsAll comand HRsignificWilcoxobars re

HPV and Inflammatory Cytokines

www.a

ase in lymphoproliferation in response to PHA,and HPV16 L1 VLP (4). To investigate possible un-ng mechanisms for this phenomenon, we evaluatedhted subset of participants from our previous studyxplored whether alterations in their cytokine profileaccount for the increased risk of HPV persistenceed among women with low proliferative response.esults suggest that the cases have markedly in-d levels of inflammatory cytokines as indicated bygnificant increase in IL-6, IL-8, TNF-α, and MIP-1αas compared with controls. These were the only pe-al markers we have noted to be markedly altered inroup of persistently infected women with evidenceune dysfunction in vitro.finding of elevated systemic levels of inflamma-ytokines in women with persistent cervical HPVions and marked decrease in immune function isand somewhat unexpected because there is littlence for an HPV-induced systemic inflammatory re-. Also, HPV infections described here are believedain localized to the cervix. Studies have shownersistent HPV infection may lead to local immunence (11), which suggests that local inflammation atrvix would be minimal as well as the markers ofheral inflammation. In addition, several studiesdescribed HPV-induced anti-inflammatory me-sms such as HPV16 E6 protein inhibiting the ex-ion and signaling of IFNs (12) and IL-18 (13).d, HPV does not have a lytic life cycle within

entiating keratinocytes that would induce an in-atory response. Interestingly, the measured in-

thermcytok

present 75th percentile.

acrjournals.org

atory response in the periphery is not restrictedividuals infected with “high-risk” HPV types suchV16, HPV18, HPV31, and HPV45. We observedomen infected with “low-risk” HPV types alsomarked increase in IL-6, IL-8, TNF-α, and MIP-en compared with controls. This finding suggestshe peripheral inflammatory response is not influ-by the oncogenicity of the HPV type involved.ugh increased levels of inflammatory cytokinesbeen previously noted in a study of cervical canceromen in our current study were healthy and had

al cervical cytology at each visit.r study focused on comparing cytokine measuresage-matched cases and controls, suggesting thatresence of inflammatory cytokines in the periph-associated with HPV persistence and weak lym-

roliferative responses, but not necessarily age.se women ages 20 to 30 years old have the high-evalence of HPV infection (15-18) and also are mostto progress to CIN3 (19), it would be important toct a longitudinal study in younger women (20-30old) to define the associations between lymphopro-tive responses, inflammatory cytokine levels, andpersistence to track these events early on.e of the limitations of our study is that the samplesall collected at the final visit. Future longitudinales will help define whether the increases in in-atory cytokine levels observed are constitutiveuced following a persistent HPV infection. Fur-

ore, we do not know whether the inflammatoryines in the present study place women at risk

1. Comparisons of (A) IL-6,, (C) TNF-α, and (D) MIP-1αplasma from women with

k (HR) or low-risk (LR) HPVecific persistent infection.LR were each comparedHPV− control group.as not a significantce between the HR and LRfor the cytokines shown.parisons between HPV−or LR were highly

ant (P < 0.0001) using then rank-sum test. Vertical

Cancer Epidemiol Biomarkers Prev; 19(8) August 2010 1957

for ptheseThe

havekineswithfutureas hepmator

periphas bflammneedetionsseen iinfectAn

nificabe exthe ppothesignifkinesPBMCthe cyal effeto celcellsanalywill rthoroIn c

matorfectedimmumineprogrflammwometion a

Discl

No p

Grant

Thisfrom tHHSN2ily refleServicenization

Thepaymenadvertisthis fac

Refe1. Mu

huba19

2. Scshointr

Figureunstimuin AIM Vlevels inPooledfor 48 hwith Linthe sup**, P <cytokin

Kemp et al.

Cance1958

rogression to CIN3 or if they are beneficial towomen.re are several possible hypotheses about why casesincreased levels of systemic inflammatory cyto-, such as the presence of simultaneous infectionsother microbial agents, which deserve attention instudies. For example, chronic viral infections such

0.01, between HPV− (controls) and HPV+ (cases) groups for eache (Wilcoxon rank-sum test).

atitis B and C have been shown to induce inflam-y cytokines such as IL-8, TNF-α, and IL-6 in the

Recepublish

hiffman MH, Bauer HM, Hoover RN, et al. Epidemiologic evidencewing that human papillomavirus infection causes most cervicalaepithelial neoplasia. J Natl Cancer Inst 1993;85:958–64.

3. deanwo20

4. Gapamu

r Epidemiol Biomarkers Prev; 19(8) August 2010

hery (20, 21). Furthermore, high body mass indexeen associated with an increase in peripheral in-ation in some studies (22, 23). Future studies ared to define the potential role of concomitant infec-and obesity in the inflammation and immune deficitn this subset of older women with persistent HPVion.other hypothesis about why proliferation was sig-ntly lower in the cases is that their PBMCs mayhausted or more susceptible to cell death due toeripheral cytokine milieu. Support for this hy-sis comes from our cytokine results that showedicantly lower concentrations for all of the cyto-tested in the supernatants from PHA-stimulateds. To find a dramatic decrease in nearly all oftokine values, we would have to suspect a glob-ct on the cells such as an increased susceptibilityl death or anergy. Due to the limited number ofisolated from our study participants, follow-upsis of cell exhaustion or cell death mechanismsequire future studies to examine this hypothesisughly.onclusion, we have described an increase in inflam-y cytokine levels among women persistently in-with HPV, with evidence of marked decrease inne function. Future studies are needed to deter-the role of this inflammatory cytokine profile inession to precancer and cancer and whether this in-atory profile is present in persistent infections inn from this cohort with preserved immune func-nd in younger women.

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

Support

project has been funded in whole or in part with federal fundshe National Cancer Institute, NIH, under contract no.61200800001E. The content of this publication does not necessar-ct the views or policies of the Department of Health and Humans, nor does mention of trade names, commercial products, or orga-s imply endorsement by the U.S. Government.costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedement in accordance with 18 U.S.C. Section 1734 solely to indicatet.

ived 02/19/2010; revised 04/30/2010; accepted 05/24/2010;ed OnlineFirst 07/20/2010.

rencesnoz N, Bosch FX, de Sanjose S, et al. The causal link betweenman papillomavirus and invasive cervical cancer: a population-sed case-control study in Colombia and Spain. Int J Cancer92;52:743–9.

Sanjose S, Diaz M, Castellsague X, et al. Worldwide prevalenced genotype distribution of cervical human papillomavirus DNA inmen with normal cytology: a meta-analysis. Lancet Infect Dis07;7:453–9.

2. A, cytokine levels in supernatants collected from 48-hlated PBMCs. Pooled supernatants from 2 × 105 PBMCs culturedmedium for 48 h were collected and frozen at −80°C. B, cytokinesupernatants collected from 48-h PHA-stimulated PBMCs.

supernatants from 2 × 105 PBMCs stimulated with PHA (2 μg/mL)were collected and frozen at −80°C. Samples were testedco-plex bead assay to examine the cytokine concentrations inernatants. Bars represent interquartile range. *, P < 0.0001;

rcia-Pineres AJ, Hildesheim A, Herrero R, et al. Persistent humanpillomavirus infection is associatedwith a generalizeddecrease in im-ne responsiveness in older women. Cancer Res 2006;66:11070–6.

Cancer Epidemiology, Biomarkers & Prevention

5. Guinn65–

6. Blogro13

7. Pe10929

8. BrayeacerRe

9. HepoproPu

10. Cade

11. Dooncytmo199

12. Kohum12:

13. Chsio18.

14. ChkinInv

15. Fraspwid

16. Hetypin

17. LainfCa

18. Monaog

19. Wafacthemi

20. Pointan

21. FarelAn

22. Fema

HPV and Inflammatory Cytokines

www.a

idotti LG, Chisari FV. Noncytolytic control of viral infections by theate and adaptive immune response. Annu Rev Immunol 2001;19:91.be GC, Schiemann WP, Lodish HF. Role of transformingwth factor β in human disease. N Engl J Med 2000;342:50–8.stkaS,KrauseCD,SarkarD,WalterMR, Shi Y, Fisher PB. Interleukin-and related cytokines and receptors. Annu Rev Immunol 2004;22:–79.tti MC, Rodriguez AC, Schiffman M, et al. Description of a seven-r prospective study of human papillomavirus infection andvical neoplasia among 10000 women in Guanacaste, Costa Rica.v Panam Salud Publica 2004;15:75–89.rrero R, Schiffman MH, Bratti C, et al. Design and methods of apulation-based natural history study of cervical neoplasia in a ruralvince of Costa Rica: the Guanacaste Project. Rev Panam Saludblica 1997;1:362–75.stle PE, Schiffman M, Gravitt PE, et al. Comparisons of HPV DNAtection by MY09/11 PCR methods. J Med Virol 2002;68:417–23.an T, Herd K, Street M, et al. Human papillomavirus type 16 E7coprotein expressed in peripheral epithelium tolerizes E7-directedotoxic T-lymphocyte precursors restricted through human (anduse) major histocompatibility complex class I alleles. J Virol9;73:6166–70.romilas AE, Li S, Matlashewski G. Control of interferon signaling inan papillomavirus infection. Cytokine Growth Factor Rev 2001;

157–70.

o YS, Kang JW, Cho M, et al. Down-modulation of IL-18 expres-n by human papillomavirus type 16 E6 oncogene via binding to IL-FEBS Lett 2001;501:139–45.

Re23. Tra

rol

acrjournals.org

opra V, Dinh TV, Hannigan EV. Circulating serum levels of cyto-es and angiogenic factors in patients with cervical cancer. Cancerest 1998;16:152–9.nceschi S, Herrero R, Clifford GM, et al. Variations in the age-ecific curves of human papillomavirus prevalence in women world-e. Int J Cancer 2006;119:2677–84.rrero R, Castle PE, Schiffman M, et al. Epidemiologic profile ofe-specific human papillomavirus infection and cervical neoplasiaGuanacaste, Costa Rica. J Infect Dis 2005;191:1796–807.zcano-Ponce E, Herrero R, Munoz N, et al. Epidemiology of HPVection among Mexican women with normal cervical cytology. Int Jncer 2001;91:412–20.lano M, Posso H, Weiderpass E, et al. Prevalence and determi-nts of HPV infection among Colombian women with normal cytol-y. Br J Cancer 2002;87:324–33.ng SS, Zuna RE, Wentzensen N, et al. Human papillomavirus co-tors by disease progression and human papillomavirus types instudy to understand cervical cancer early endpoints and deter-

nants. Cancer Epidemiol Biomarkers Prev 2009;18:113–20.lyak SJ, Khabar KS, Rezeiq M, Gretch DR. Elevated levels oferleukin-8 in serum are associated with hepatitis C virus infectiond resistance to interferon therapy. J Virol 2001;75:6209–11.lasca K, Ucciferri C, Dalessandro M, et al. Cytokine patterns cor-ate with liver damage in patients with chronic hepatitis B and C.n Clin Lab Sci 2006;36:144–50.sta A, D'Agostino R, Jr., Williams K, et al. The relation of body fatss and distribution to markers of chronic inflammation. Int J Obeslat Metab Disord 2001;25:1407–15.

yhurn P, Wood IS. Adipokines: inflammation and the pleiotropice of white adipose tissue. Br J Nutr 2004;92:347–55.

Cancer Epidemiol Biomarkers Prev; 19(8) August 2010 1959