Disponibleenlignesur
ScienceDirect
www.sciencedirect.com
ORIGINAL ARTICLE
The impact of postoperative inflammatory biomarkers on oncologic outcomes of
bladder cancer
L’impact des marqueurs systémiques d’inflammation sur les résultats oncologique du cancer de la vessie
S. Albisinni
a,∗,1, I. Moussa
a,1, F. Aoun
b, T. Quackels
a, G. Assenmacher
b, A. Peltier
b, T. Roumeguère
aaUniversityClinicsofBrussels,DepartmentofUrology,HôpitalErasme,UniversitéLibrede Bruxelles,RoutedeLennik808,Brussels,Belgium
bUrologyDepartment,Jules-BordetInstitute,UniversitéLibredeBruxelles,Brussels,Belgium
Received6November2018;accepted14February2019 Availableonline4April2019
KEYWORDS Bladdercancer;
Inflammation;
Recurrence;
Survival
Summary
Introduction.—Theclinicalimpactofinflammatorybiomarkershasbeenevaluatedinurothe- lialbladdercancer.However,dataarelimitedtopreoperativevaluesandthereispaucityof evidenceoftheroleofpostoperativemeasurementofthosebiomarkers.Theaimofthecurrent studywastodeterminetheassociationofinflammatorybiomarkersasneutrophil-to-lymphocyte ratio(NLR),monocyte-to-lymphocyteratio(MLR),hemoglobintoplateletratio(HPR)andC- reactiveprotein(CRP),beforeandafterradicalcystectomy,withrecurrenceandsurvivalof bladdercancer.
Materialandmethods.—Weprospectivelyevaluated134patientsundergoingradicalcystec- tomyforinvasivebladdercancerbetweenJanuary2013andJanuary2018.Theinflammatory biomarkersweremeasured10daysbeforesurgeryandat1,6and12monthspostoperatively.
Kaplan—MeiercurvesandCoxproportionalhazardsandlogisticregressionmodelswereusedto evaluatetheassociationbetweenthedifferentinflammatorybiomarkersandrecurrencefree survival(RFS),cancerspecificsurvival(CSS)andoverallsurvival(OS).
Results.—The median follow-up time was 21.1months (5—37 mo). On multivariate analy- sis,preoperativeNLR>3.88wasassociatedtolocally-advanceddisease(>pT3)andNLR>3.88 andHPR<0.039were significantlyassociatedtonodepositivedisease.PostoperativeNLRat 3months>4.68(HR:2.37,95%CI:1.08—4.47,P=0.03)wasassociatedwithareducedRFS.A postoperative NLRat3months>4.68(P=0.04) anda postoperative HPRat3months<0.029 (P=0.001)wereassociatedwithasignificantreductioninCSSandOS.
∗Correspondingauthor.
E-mailaddress:simone.albisinni@erasme.ulb.ac.be(S.Albisinni).
1S.AlbisinniandI.Moussahaveequallycontributedtothiswork.
https://doi.org/10.1016/j.purol.2019.02.008
1166-7087/©2019ElsevierMassonSAS.Allrightsreserved.
Conclusion.—PostoperativeNLR andHPR at3months appear to beclosely associated with RFS,CSSandOS.Furtherstudiesareneededonthesepostoperativemarkerstoestablishthe potentialimpactoftheseinflammatorybiomarkersonatailoredtherapeuticapproachforeach patient.
Levelofevidence.—3.
©2019ElsevierMassonSAS.Allrightsreserved.
MOTSCLÉS Cancerdelavessie; Inflammation; Récidive; Survie
Résumé
Introduction.—L’impactclinique demarqueurs del’inflammation adéjà étérapporté pour lecancer dela vessie. Cependant,lesdonnées sontlimitéesaux valeurspréopératoireset il existepeud’évaluation durôle dela mesurepostopératoirede cesmarqueurs inflamma- toires.Le butde laprésente étudeestde déterminerl’associationdes marqueurstelsque lerapportneutrophiles/lymphocytes(NLR),lerapportmonocytes/lymphocytes(MLR),lerap- porthémoglobine/plaquettes(HPR)etlaprotéineC-réactive(CRP)avantetaprèscystectomie radicaleaveclarécidiveetlasurvieducancerdelavessie.
Matérieletméthodes.—Nousavonsévaluéprospectivement134patientsopérésd’unecystec- tomieradicalepouruncancerdelavessiemusculo-invasifentrejanvier2013etjanvier2018.
Lesmarqueursinflammatoiresontétémesurés10joursavantlachirurgieetà1,6et12mois aprèsl’intervention.DescourbesdeKaplan—Meieraveclesmodèlesderégressionlogistiqueet derisquesproportionnelsdeCoxontétéutiliséspourévaluerl’associationentrelesdifférents marqueurset la surviesansrécidive(SSR),la surviespécifiqueaucancer (SSC)et lasurvie globale(SG).
Résultats.—Le suivi médian a été de 21,1 mois (5—37 mois). En analyse multivariée, un NLR>3,88en préopératoire était associé à unemaladie localementavancée (>pT3) et un NLR>3,88etHPR<0,039étaientassociésàuneatteinteganglionnaire(pN+).LeNLRpostopéra- toireà3mois>4,68(HR:2,37,IC95%:1,08—4,47,p=0,03)étaitassociéàuneréductiondela SSR.UnNLRpostopératoireà3mois>4,68(p=0,04)etunHPRpostopératoireà3mois<0,029 (p=0,001)étaientassociésàuneréductionsignificativedelaSSCetdelaSG.
Conclusion.—LeNLRetleHPRpostopératoiresà3moissemblentêtreétroitementassociésaux risquesdeRFS,CSSetOS.Desétudescomplémentairessontnécessairespourétablirl’impact potentiel de ces marqueursde l’inflammation pour uneapproche thérapeutiqueadaptée à chaquepatient.
Niveaudepreuve.— 3.
©2019ElsevierMassonSAS.Tousdroitsr´eserv´es.
Introduction
Muscleinvasivebladdercancer(MIBC)isahighlyaggressive malignancywithapoorprognosisincaseoflocally-advanced disease[1,2].Despiteadvancesinchemotherapyandaggres- sivesurgical treatment,MIBC-relatedsurvivalremainslow and has not increased in recent decades. Though strictly dependent on stage, overall survival (OS) at 5years is between 40% and 60%, and cancer specific survival (CSS) variesbetween30and50%[3].Identifyingprognosticfactors remainsnecessaryinordertoestablishabetterriskstrati- ficationthusallowingabettertailoredtreatmentforeach patient.
Several studies have evaluated the tumor-induced inflammatoryreactionanditsassociationtotumoraggres- siveness[4—6].Itisknowntodatethatinflammationplays aroleintumorproliferationandthatitsbiologicalexpres- sion might lead to a large amount of significant clinical
informationparticularlyin relationtotumor behaviorand aggressivenessinadditiontodiseaseprognosis.Theefficacy ofBCG intravesicaltherapy and the current revolution of immunotherapybycheckpointinhibitors[7]inbladdercan- cerdemonstratethehighimmunedependenceofurothelial carcinoma,openingthedoortoimplementationofinflam- matorybiomarkers.
Neutrophil-to-lymphocyte ratio (NLR), monocyte-to- lymphocyteratio(MLR),hemoglobintoplateletratio(HPR) andC-reactiveprotein(CRP)areeasy-to-use,readilyavail- able,reproducibleandinexpensivebiomarkers[8—10].They actuallysummarize numericallythe responsearising from the interactionbetween cancer and the systemic inflam- matory response of the host: the amplification of the innateimmunityexpressedbyhyperneutrophiliaand/orthe increase in macrophages, depression of cellular acquired immunityrepresented by induced leukopenia(Th2>Th1), thrombocytosis secondary to increased neutrophils and
IL-6/IL-1secretingmacrophages,addingtoanemiaresulting fromthetumor-inducedchronicinflammatorystate.
Several studies have investigated the clinical value of thesebiomarkers in bladder cancer [8—10]. The majority of the results revealed that they could be used as prog- nostic factors in retrospective studies [11,12], although otherinvestigatorshavereportedcontradictoryresults[10].
Nonetheless, most studies to date have focused on pre- operativeconcentrationofbiomarkers,while thevalueon postoperativefollow-upisunknown.
The objectiveofourstudy wastodeterminethe prog- nosticvalueoftheinflammatorybiomarkersNLR,LMR,HPR andCRP inthe preoperative andpostoperativefollow-up, inordertoassesstheriskofrecurrenceandsurvivalafter cystectomyformuscleinvasivebladdercancer.
Material and methods
Afterinstitutionalreviewboardapproval(Ethicscommittee referenceNo.P2018/254),170patientsundergoingradical cystectomyformuscleinvasivebladdercancerorhigh-risk non-muscleinvasivebladdercanceratErasmehospitaland Jules-BordetInstitute(sameteam),betweenJanuary2013 and January 2018 were prospectively evaluated for the study.
Preoperativedatawerecollected10dayspriortoradical cystectomy.Forpatientsreceivingneoadjuvantchemother- apy (NAC) (n=39), we also collected these same values in the days leading up to the first course of treatment.
Postoperative datawascollectedduring patientfollow-up at1month,3months,6monthsand1yearpostoperatively.
Figure1. Studydesign.
Patientsunderwentfullbloodworkupandbiomarkerswere registered.Inparticular,weevaluatedNLR,MLR,HPRand CRP levels. NLR was calculated by dividing the absolute neutrophilcount(103/L)bythenumberofabsolutelym- phocytes (103/L), LMR was calculated by dividing the numberofabsoluteneutrophilsbythenumberofabsolute monocytesandfinallyHPRbydividingthehemoglobinlevel (ing/L)bythenumberofplatelets(in106/L).
The date of treatment initiation for each patient was definedastheactualdateofRC.Ajuniorurologistrecorded patients’ demographics and Charlson comorbidity index.
When analyzing biomarkers on follow-up after surgery, values were not considered if patients presented a late complication,inparticularurinaryinfection.
All patients underwent RC with bilateral lymph node dissection. Thirty-nine patients who received neoadju- vant chemotherapy had between 2 and 4 cycles of gemcitabine—cisplatin basedtreatment depending ontol- eranceandefficacy,whichwasevaluatedafter2cycles.For postoperativefollow-up,patientswerefollowedat4weeks afterRC, then every3monthsduring thefirst year, every 6monthsinthesecondyearandyearlythereafter.Follow-up consistedonacompletephysicalexamination,bloodworkup andathoraco-abdomino-pelvicCTscan.
Statisticalanalysis:uni-andmultivariatelogisticregres- sions were performed to assess an association between preoperativebiomarkers(NLR,MLR,HPR,CRP)andlocally- advanced disease (pT3/4), node positive disease (pN+).
Multivariateanalyses wereadjustedtoage,gender,smok- ing status (none, active or ex-smoker), kidney function (measured asCKD-EPI) andneoadjuvant therapy.Uni-and multivariate Cox regressions were calculated to estimate therelationshipbetweenthesevariablesandtheRFS,CSS, and OSpost-RC. Significant factorson univariateanalyses wereincorporated in amultivariate modeltotest inflam- matorybiomarkers.Kaplan-Meiercurveswereconstructed toevaluatetheeffectofpreviouslysignificantvariableson RFS,CSSandOS.Log-ranktestswereusedtoassessdiffer- enceacrossgroups. Biomarkerswerecategorizedin<75th percentilevs.>75percentileforNLRandCRPandin<25th percentilevs.>25thpercentile for MLR andHPR. We also investigatedthepotentialeffectofthesedifferentinflam- matorybiomarkers(CRP,NLR,MLR,andHPR)asapredictor of a NAC response. This response has been defined as a
‘‘down-staging’’,thatis,thetransitionfromapre-NACpT2 tumor stage toa post NACpT0 or pTis tumor stage. This relationship wasanalyzed throughthe implementationof uni- andmultivariatelogisticregressions.Allanalyses and graphicswereperformed using theStata softwareversion 12.Atwo-sidedP<0.05definedstatisticalsignificance.
Results
Afterexclusionof36patients(Fig.1)duetonon-eligibility, 134 patients were available for analysis. General charac- teristicsofthecohortareillustratedinTable1.Themedian follow-uptimewas21.1months(5—36.8).Forty-sixpatients (32%)hadarecurrencewithinamediandurationof6months (2—14.5).Thirty-sixpatientsdiedofbladdercancerand46 patientsdiedofanycauseduringtheavailablefollow-up.
Table1 Pathologicalfeaturesofthecohort(n=134).
Age(year)Median(IQR) 72(64—77)
Mean±SD 70±10
≥65 98(73%)
<65 36(27%)
Gender
M 110(82%)
F 24(18%)
BMI(kg/m2)
Median(IQR) 25.6(22.3—29)
Mean±SD 25.9±5.14
≥25 72(54%)
<25 62(46%)
Tobacco
No 36(27%)
Active 54(40%)
Former 45(33%)
Charlsoncomorbidityindex
Median(IQR) 6(5—7)
Mean±SD 6.49±2.39
Neoadjuvantchemotherapy
Yes 39(29%)
No 95(71%)
pTa
pT0 9(7%)
pTa 9(7%)
pTis 4(3%)
pT1 16(12%)
pT2 37(27%)
pT3 39(29%)
pT4 20(15%)
pNa
pNx 3(2%)
pN0 99(74%)
pN1 13(10%)
pN2—pN3 19(14%)
LymphadenectomyMedian(IQR) 17(12—23)
Mean±SDb 18.14±9.12
<10 23(17%)
≥10 111(83%)
Histology
Urothelialpure 124(94%)
Epidermoïdvariant 2(1%)
Sarcomatoïdvariant 2(1%)
Mixed 3(2%)
Adenocarcinoma 2(1%)
Plasmocytoïdvariant 1(1%)
Tumorgrade
Low(I+II) 2(2%)
High(III) 122(91%)
Undetermined 10(7%)
CISassociate
Yes 59(44%)
No 75(56%)
Surgicalmargins
Positive 3(2%)
Negative 131(98%)
Prostateadenocarcinoma
Yes 34(25%)
No 100(75%)
a TNMstagingsystem.
b Standarddeviation.
Table2 Cohortpreoperativeinflammatorybiomarkers.
Cohort(n=134) CohortwithoutNAC(n=95) P-value Creatinine(mg/dL)
Median(IQR) 1.00(0.90—1.20) 1.07(0.9—1.28) 0.70
Mean±SD 1.25±1.83 1.33±2.15
CKD-EPI(mL/min/1.73m2)
Median(IQR) 64(53.75—77) 64(52—70) 0.36
Mean±SD 65.63±19.68 63.30±18.71
CRP(mg/L)
Median(IQR) 4.15(1.80—9.15) 4.70(2.15—10.45) 0.19
Mean±SD 10.74±18.55 11.90±18.58
Neutrophilcount(×103/L)
Median(IQR) 4.80(3.60—5.90) 5.12(4.05—6.16) 0.11
Mean±SD 5.10±2.53 5.53±2.46
Lymphocytecount(×103/L)
Median(IQR) 1.65(1.19—2.00) 1.62(1.17—1.98) 0.77
Mean±SD 1.73±0.78 1.70±0.79
Monocytecount(×103/L)
Median(IQR) 0.70(0.52—0.83) 0.70(0.50—0.80) 0.45
Mean±SD 0.74±0.28 0.70±0.23
Haemoglobin(g/dL)
Median(IQR) 13(11.03—14.10) 13.30(11.50—14.50) 0.21
Mean±SD 12.66±1.97 12.96±2.06
Platelets(×103/L)
Median(IQR) 244(194—299) 251(196—304) 0.76
Mean±SD 264.11±102.66 268.38±108.30
NLR
Median(IQR) 3(1.86—3.88) 3.24(2.49—4.11) 0.12
Mean±SD 3.89±4.57 4.26±4.52
HPR
Median(IQR) 0.05(0.04—0.07) 0.06(0.04—0.07) 0.60
Mean±SD 0.06±0.03 0.06±0.03
LMR
Median(IQR) 2.34(1.69—3.02) 2.33(1.71—3.10) 0.79
Mean±SD 2.53±1.29 2.60±1.36
BiomarkerconcentrationsareshowninTables2(preoper- ative)and3(postoperative).Nosignificantdifferenceacross patient having received neoadjuvant chemotherapy and thosewhounderwentimmediatecystectomywasobserved (allP>0.05).
Concerning the impact of inflammatory biomarkers on the prediction of locally-advanced disease and pos- itive lymph nodes, on univariate logistic regressions (Table 4), preoperative CRP≥9.1, CKD-EPI≤60, preop- erative NLR≥3.88 and preoperative HPR≤0.039 were significantly associated to locally-advanced bladder can- cer (≥pT3) and nodal invasion (pN+). On multivariate analysis, only preoperative NLR>3.88 was associated to locally-advanceddiseaseandNLRandHPRweresignificant predictorsofnodepositivedisease.
Regarding recurrence free survival (RFS) (Table 5), an elevated CRP>9.1 (HR2.34, 95% CI: 1.21—4.53, P=0.01) as well as an increased NLR>4.68 at 3months (HR:
2.37, 95% CI: 1.08—4.47, P=0.03) were both significantly associated to recurrence on multivariate Cox regres- sions, corrected for pT. Nodal status was not included in the multivariate model due to elevated collinear- ity with pT stage≥pT3 (Spearman’s =0.44, P<0.001).
Kaplan—Meier curves illustrate the impact of preopera- tive CPR and NLR values 3 months after surgery on RFS (Fig.2).
Analyzing risk factors for CSS (Table 6) CRP>9.1, pre- operative NLR>3.88, preoperative HPR<0.039, NLR at 3months>4.68,HPRat3months<0.029andfinallyMLRat 3months<1.74appeartobeassociatedtoCSSonunivariate analysis(Fig.3).Onmultivariate analysisNLRat 3months (HR:2.94;95%CI:1.15—9.16;P=0.05)aswellasHPRat3 months(HR:3.22;95%CI:1.41—7.33;P=0.005)weresigni- ficantlyassociatedtoCSS.
Finally, 46 deaths due toall causes were recorded on available follow-up. Analysesexploring risk factorsfor OS (Table 7) confirmed that preoperative HPR<0.039, NLR at 3months>4.68, HPR at 3months<0.029 and MRL at 3months<1.74wereallsignificantlyassociatedtosurvival on univariate analysis. However, as for CSS, only NLR at 3months>4.68(HR:2.91;95%CI:1.06—7.94,P=0.04)and HPR at 3months<0.029 (HR: 3.66; 95% CI: 1.77—7.56, P=0.001)remainedsignificantlyassociatedonmultivariate analysis.Kaplan—Meiercurves(Fig.4)illustratetheassoci- ationbetweenNLRandHPRvaluesat3monthsaftersurgery andOS.
Table3 Postoperativeinflammatorybiomarkers.
Cohort(n=134) Cohortwithout
NAC(n=95)
P-value Neutrophilcount
(×103/L)to 1month
Median(IQR) 5.40(3.70—7.90) 5.99(3.89—8.22) 0.12
Mean±SD 6.23±3.62 6.41±2.57
Lymphocyte count(×103/L) to1month
Median(IQR) 1.50(1.10—1.90) 1.51(1.08—1.93) 0.85
Mean±SD 1.64±0.76 1.63±0.76
Monocytecount (×103/L)to 1month
Median(IQR) 0.70(0.53—0.90) 0.70(0.57—0.90) 0.54
Mean±SD 0.75±0.32 0.76±0.30
Haemoglobin (g/dL)to1month
Median(IQR) 10.35
(9.40—11.50)
10.35 (9.30—11.50)
0.25
Mean±SD 10.46±1.38 10.35±1.43
Platelets (×103/L)to 1month
Median(IQR) 369.50
(277.00—479.30)
402
(318.8—495.50)
0.76
Mean±SD 388.00±162.67 419.14±158.60
NLRMedian(IQR) Mean±SD
1month 3.43(2.46—5.58)
4.70±3.99
3.45(2.70—5.66) 5.06±4.46
0.17
3month 2.85(2.05—4.68)
4.95±10.05
3.09(1.99—5.19) 5.71±11.75
1year 2.83(2.23—4.62)
4.02±3.72
3.09(2.32—4.86) 4.50±4.27 HPRMedian(IQR)
Mean±SD
1month 0.03(0.02—0.04)
0.03±0.02
0.03(0.02—0.03) 0.03±0.02
0.87
3month 0.04(0.03—0.06)
0.05±0.03
0.04(0.03—0.05) 0.05±0.03
1year 0.06(0.04—0.07)
0.06±0.02
0.06(0.04—0.07) 0.05±0.02 LMRMedian(IQR)
Mean±SD
1month 2.22(1.57—2.91)
2.63±2.49
2.20(1.48—2.86) 2.62±2.79
0.76
3month 2.51(1.74—3.50)
2.88±1.87
2.50(1.56—3.82) 2.88±2.03
1year 2.51(1.66—3.40)
2.86±2.34
2.47(1.59—3.08) 2.80±2.68
Discussion
Systemicinflammationplaysamajorroleintumordevelop- mentandprogression.Infact,cancercanbecomparedto achronicinflammatorydiseasewherethecancercellsare
abletomodulatetheinnateandacquiredimmuneresponses toinduceamicroenvironment favorabletoitsgrowthand invasion[4].
Bladder cancer is highly immune-modulated, as demonstrated by the impressive results of modern
Table4 Uni-andmultivariatelogisticregressionpredictinglocally-advanceddisease(≥pT3)andnodepositivedisease (pN+).
Prediciting≥pT3disease Univariate Multivariate
OR(95%CI) P-value OR(C95%CI) P-value
Age≥77 0.93(0.45—1.95) 0.85
Gender 0.92(0.37—2.22) 0.84
Activesmoking 0.37(0.15—0.9) 0.03 0.56(0.23—1.37) 0.88
Formersmoking 0.43(0.17—1.09) 0.07
Neoadjuvantchemotherapy 0.8(0.62—1.04) 0.1
CRPpreoperative≥9.1 2.23(1.01—4.94) 0.05 1.81(0.75—4.36) 0.18
CKD-EPI≥60 0.43(0.20—0.89) 0.02 0.49(0.22—1.09) 0.08
NLRpreoperative≥3.88 4.40(1.88—10.29) 0.001 3.23(1.3—8.01) 0.01
HPRpreoperative≤0.039 2.46(1.05—5.72) 0.04 1.60(0.61—4.20) 0.34
LMRpreoperative≤1.69 2.09(0.94—4.64) 0.07
PredictingpN+disease Univariate Multivariate
OR(95%CI) P-value OR(95%CI) P-value
Age≥77 0.99(0.42—2.34) 0.99
Gender 1.24(0.42—3.63) 0.70
Activesmoking 0.64(0.24—1.71) 0.38
Formersmoking 0.84(0.31—2.26) 0.73
Neoadjuvantchemotherapy 0.89(0.66—1.21) 0.47
CRP≥9.1 2.18(0.93—5.14) 0.07
CKD-EPI≥60 0.38(0.17—0.86) 0.02 0.47(0.19—1.15) 0.10
NLRpreoperative≥3.88 3.55(1.50—8.45) 0.004 2.81(1.11—7.12) 0.03
HPRpreoperative≤0.039 4.51(1.86—10.96) 0.004 3.68(1.40—9.68) 0.008 LMRpreoperative≤1.69 1.31(0.53—3.23) 0.55
NLR3M4.68 2.05(0.75—5.62) 0.16
HPR3M0.029 2.82(1.04—7.69) 0.04 2.82(1.04—7.69) 0.04
LMR3M1.74 1.57(0.57—4.35) 0.39
Inboldarestatisticallysignificantassociations.
immunotherapies in metastatic patients [13]. Given ourcurrentinabilitytoadequatelystratifypatientsandtai- loradjuvanttherapiesafterradicalcystectomy,biomarkers areurgentlydemandedinthissetting.Indeed,attimesthe
answer is easier than what we believe:assuch, a simple bloodtestanalyzingneutrophils,lymphocytes,hemoglobin and platelet levels could be a major help in identifying patientsatriskorrecurrenceandlethaldisease,requiring
Table5 Coxregressionforrecurrencefreesurvival.
Univariate Multivariate
HR(95%CI) P-value HR(95%CI) P-value
Neoadjuvantchemotherapy 0.55(0.26—1.15) 0.11
≥pT3 6.15(1.6—9.78) 0.0001 5.23(2.49—11.09) <0.0001
pN+ 5.05(2.65—9.63) 0.0001
Lymphadenectomy≥10 1.08(0.45—2.58) 0.86
Surgicalmargin 4.22(1.01—17.92) 0.05
CRP≥9.1 2.77(1.51—5.08) 0.001 2.34(1.21—4.53) 0.01
Creatinine 1.83(0.89—3.76) 0.10
CKD-EPI≥60 0.78(0.41—1.49) 0.45
NLRpreoperative
≥3.88
2.12(1.12—4) 0.02 1.51(0.78—2.92) 0.22
HPRpreoperative
≤0.039
1.95(0.85—3.23) 0.05 1.05(0.50—2.22) 0.90 LMRpreoperative
≤1.69
1.19(0.61—2.33) 0.26
Table5(Continued)
Univariate Multivariate
HR(95%CI) P-value HR(95%CI) P-value
NLRpostoperative1month
≥5.58
1.82(0.96—3.46) 0.07 NLRpostoperative3months
≥4.68
2.46(1.19—508) 0.02 2.37(1.08—4.74) 0.03
HPRpostoperative1month
≤0.021
1.45(0.75—2.77) 0.27 HPRpostoperative3months
≤0.029
1.81(0.89—3.69) 0.10 LMRpostoperative1month
≤1.57
1.47(0.75—2.87) 0.26 LMRpostoperative3months
≤1.74
2.13(1.06—4.28) 0.03 1.85(0.89—6.22) 0.09
Bold:statisticallysignificant.
Figure2. Recurrencefreesurvivalaccordingtoinflammatorymarkers.
Figure 3. Cancer specific survival according to inflammatory markers.
aggressiveadjuvant therapy. In metastatic kidney cancer, absolutelaboratoryvaluesofhemoglobinandplatelets[14]
arecurrentlybeingconsideredinriskcalculatorsandhelp indeterminingthebestclinicalpathwayforpatients.
Of course, ideally it would be best tostratify patient before major abdominal surgery. Nonetheless, it is clear thatpTNMstaging,obtainedafterspecimenremoval,isand willmostprobablyremainthemostsignificantpredictorof oncologicoutcomes.As such,itwouldbelogictowaitthe postoperativephasetostratifyourpatients,oncetheTNM stagingisavailable.Inthiscontext,althoughmultiplestud- ieshavedemonstratedastrongassociationofinflammatory biomarkersasNLRtobladdercanceroutcomes,thereisa truepaucity of dataonthe postoperativeconcentrations.
Moreover,beforemajorabdominalsurgeryascystectomyfor cancer,multipleconfoundingfactors(stress,insomnia,uri- narybleeding) could potentially altertheseresults, while at1—3monthsaftersurgery,onecouldspeculatethatsuch confoundersareprobablylesssignificant.Therefore,taking intoaccountthe burdenofsurgery addingtoitspotential complications, aswell asthe relatively vast literature on preoperativeinflammatorybiomarkersinbladdercancer,we focusedontheevaluationofpotentialprognosticbiomarkers inbothpreoperativeandmostofallpostoperativesettings forradicalcystectomy.
Figure4. Overallsurvivalaccordingtoinflammatorymarkers.
First,weobservedthatthepreoperativeNLRwasasso- ciated with a pT3—4 stage and a pN+status, and that preoperative HPR was associated to node positive dis- ease.Second,weexploredriskfactorsforstrongoncologic outcomesasRFS,CSSandOS.PreoperativeCRPandpostop- erativeNLRat3monthswereassociatedwithasignificant reductioninRFS.Furthermore,postoperativeNLRandHPR at 3monthswereassociatedwithareductioninboth CSS andOS.
Theseresultswereindeedinlinewiththosefoundinthe literature.Infact,twolargemeta-analysesreportedthata highpreoperativeNLRwasassociatedwithapT3—4stage, a pN+, as well as reductions in CSS and OS after radical cystectomy [12,15]. For preoperative HPR, there is less data.LaCroceetal.havedemonstratedthatthisratiocan beastrong prognosticfactor forOSandCSSaftersurgery [16].Tangetal.alsoevaluatedHPRinthesettingofhigh- gradeNMIBC(pT1G3)andconcludedthatit representeda powerfulpredictorwithastrongassociationwithbothRFS and OS [8].However, these results should be interpreted with caution as several studies have also highlighted the absence of correlation between preoperative NLR and the ‘‘outcomes’’ related to bladder cancer after surgery, pointinginparticulartoalackofprecision inthemethod- ologies used and the lack of specificity of these markers
Table6 Coxregressionforcancerspecificsurvival(CSS).
Univariate Multivariate
HR(95%CI) P-value HR(95%CI) P-value
Age 0.78(0.37—1.67) 0.52
Gender 1.00(0.44—2.30) 0.99
BMI≥30 0.38(0.12—1.26) 0.11
Charlsoncomorbidityindex≥8 1.15(0.56—2.35) 0.70
Clavien≥2 1.50(0.76—2.87) 0.25
Neoadjuvantchemotherapy 0.35(0.06—0.91) 0.03 0.53(0.19—1.45) 0.12
Adjuvantchemotherapy 2.34(1.2—4.55) 0.01
≥pT3 8.75(3.61—21.19) <0.0001 8.01(3.12—20.59) <0.0001
pN+ 5.18(2.65—10.12) <0.0001
Lymphadenectomy≥10 0.57(0.26—1.26) 0.16
CRP≥9.1 2.08(1.05—4.15) 0.04 1.41(0.64—3.11) 0.40
Creatinine 1.12(1.02—1.24) 0.07
CKD-EPI≥60 0.74(0.36—1.5) 0.41
NLRpreoperative
≥3.88
2.09(1.02—4.26) 0.04 1.16(0.55—2.43) 0.70
HPRpreoperative
≤0.039
2.16(1.08—4.36) 0.03 1.57(0.69—3.60) 0.29
LMRpreoperative
≤1.69
1.20(0.43—2.12) 0.91 NLR1month
≥5.58
1.9(0.94—3.82) 0.07 NLR3months
≥4.68
3.79(1.81—7.94) <0.0001 2.94(1.15—9.16) 0.05 HPR1month
≤0.021
1.88(0.94—3.79) 0.07 HPR3months
≤0.029
3.05(1.45—6.40) 0.003 3.22(1.41—7.33) 0.005 LMR1month
≤1.57
1.68(0.82—3.42) 0.16 LMR3months
≤1.74
2.17(1.40—4.15) 0.03 1.75(0.62—4.94) 0.29
Bold:statisticallysignificant.
[10,17]. With respect to preoperative CRP, our findings wereconsistentwiththestudydonebyLeparaetal.,which foundasignificantrelationshipwithCSS,RFS,andOS[18].
Asdiscussed,fewstudiestackledpostoperativeNLR.Two retrospective studies reported that NLR at three or four
monthspost-RCwassignificantlyassociatedwithCSSandOS [19,20].However,wecouldnotidentifyanystudyevaluating postoperativeHPR.
Severalimplicationssprout fromthe presentedresults.
First,the absenceof data oninflammatory markers after
Table7 Coxregressionanalyzingpredictorsofoverallsurvival(OS).
Univariate Multivariate
HR(95%CI) P-value HR(C95%CI) P-value
Age 1.21(0.66—2.21) 0.55
Gender 1.02(0.49—2.12) 0.96
Neoadjuvantchemotherapy 0.30(0.10—0.72) 0.006 0.39(0.02—0.95)a 0.04
Adjuvantchemotherapy 1.66(0.91—3.03) 0.098
Charlsoncomorbidityindex≥8 1.91(1.06—3.42) 0.03
Clavien≥2 1.58(0.89—2.82) 0.12
BMI≥30 0.62(0.26—1.46) 0.27
≥pT3 6.89(3.39—14) <0.0001 6.63(3.04—14.44)b <0.0001
pN+ 3.72(2.07—6.67) <0.0001
Table7(Continued)
Univariate Multivariate
HR(95%CI) P-value HR(C95%CI) P-value
Lymphadenectomy
≥10
0.79(0.37—1.7) 0.54
CRP≥9.1 2.15(1.18—3.93) 0.01 1.38(0.69—2.76) 0.36
NLRpreoperative
≥3.88
2.46(1.33—4.55) 0.004 1.42(0.75—2.68) 0.29 HPRpreoperative
≤0.039
2.06(1.11—3.83) 0.02 1.57(0.76—3.25) 0.22
LMRpreoperative
≤1.69
1.28(0.67—2.45) 0.45 NLR1month
≥5.58
1.58(0.83—3.04) 0.17 NLR3months
≥4.68
3.81(2.00—7.26) <0.0001 2.91(1.06—7.94) 0.04 HPR1month
≤0.021
1.77(0.93—3.36) 0.08 HPR3months
≤0.029
3.41(1.77—6.57) <0.0001 3.66(1.77—7.56) 0.001 LMR1month
≤1.57
1.59(0.83—3.06) 0.16 LMR3months
≤1.74
2.15(1.08—4.26) 0.02 1.84(0.74—4.59) 0.19
Bold:statisticallysignificant.
aPreoperativevalues.
b Postoperativevalues.
surgery,whichwererarely takenintoconsiderationinthe currentlypublishedliterature.Inthecurrentstudypostop- erativeNLRandHPRat 3monthsproved tobestatistically moretightlyassociatedtopatientsurvivalthanpreoperative levels.By extrapolation, these results could suggest that patientswithan increased postoperativeNLR at 3months and/ordecreasedpostoperativeHPRat3months,couldbe at increasedrisk of tumor progression, thus should bene- fitfromclosermonitoring.Moreover,thisfindingopensthe doortopossibleimplicationsforpatientselectionforadju- vanttherapy.Ofnote,subanalysisoftheS-TRACstudy[21]
foradjuvantimmunotherapyinhigh-riskrenalcancerfound thatNLRwaspredictiveofresponsetoadjuvanttreatment.
Limitations of this study include the limited number of study participants and a relatively short follow-up.
Moreover, despite adopting a well-defined methodology, inflammatorybiomarkers present a lack ofspecificity due totheinfiniteconfoundingfactorswhichcouldpotentially biastheirconcentrations,andwhichcannotbealltakeninto considerationwhenanalyzingthesebiomarkers.
Conclusion
Inthis singlecenter prospective study,postoperative NLR and HPR at 3months after cystectomy were significantly associatedto RFS, CSS and OSfor bladder cancer. Larger studies arerequired toconfirm ourfindings and evaluate
theimpactof inflammatorybiomarkersonbladder cancer outcomesandadjuvanttherapy.
Acknowledgments
None.
Disclosure of interest
Theauthorsdeclarethattheyhavenocompetinginterest.
References
[1]AntoniS,FerlayJ,SoerjomataramI,ZnaorA,JemalA,BrayF.
Bladdercancerincidenceandmortality:aglobaloverviewand recenttrends. EurUrol 2017;71:96—108, http://dx.doi.org/
10.1016/j.eururo.2016.06.010.
[2]CumberbatchMGK, Jubber I,Black PC, Esperto F, Figueroa JD,KamatAM,etal.Epidemiologyofbladdercancer:asys- tematicreview and contemporary update of risk factors in 2018. Eur Urol 2018;74:784—95, http://dx.doi.org/10.1016/
j.eururo.2018.09.001.
[3]AlfredWitjesJ,LebretT,CompératEM,CowanNC,DeSantisM, BruinsHM,etal.Updated2016EAUguidelinesonmuscleinva- siveandmetastaticbladdercancer.EurUrol2017;71:462—75, http://dx.doi.org/10.1016/j.eururo.2016.06.020.
[4]Gakis G. The role of inflammation in bladder cancer.
Adv Exp Med Biol 2014;816:183—96, http://dx.doi.org/10.
1007/978-3-0348-0837-88.
[5]Grivennikov SI, Greten FR, Karin M. Immunity, inflamma- tion, and cancer. Cell2010;140:883—99, http://dx.doi.org/
10.1016/j.cell.2010.01.025.
[6]Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860—7,http://dx.doi.org/10.1038/nature01322.
[7]RassyEE,BakounyZ,AounF,HaddadFG,SleilatyG,AssiT,etal.
Anetworkmeta-analysisofthePD(L)-1inhibitorsinthesal- vagetreatmentofurothelialbladdercancer.Immunotherapy 2018;10:657—63,http://dx.doi.org/10.2217/imt-2017-0190.
[8]Tang G, Zhen Y, Xie W, Wang Y, Chen F, Qin C, et al. Preoperative hemoglobin-platelet ratio can signi- ficantly predict progression and mortality outcomes in patientswithT1G3bladder cancerundergoingtransurethral resection of bladder tumor. Oncotarget 2018;9:18627—36, http://dx.doi.org/10.18632/oncotarget.23896.
[9]YoshidaT, KinoshitaH, YoshidaK, MishimaT,YanishiM,Inui H, et al. Prognostic impact of perioperative lymphocyte- monocyte ratio in patients with bladder cancer under- going radical cystectomy. Tumour Biol 2016;37:10067—74, http://dx.doi.org/10.1007/s13277-016-4874-8.
[10] Ojerholm E, Smith A, Hwang W-T, Baumann BC, Tucker KN, Lerner SP, et al. Neutrophil-to-lymphocyte ratio as a bladder cancer biomarker: assessing prognostic and pre- dictive value in SWOG 8710. Cancer 2017;123:794—801, http://dx.doi.org/10.1002/cncr.30422.
[11] TempletonAJ,McNamaraMG, ˇSerugaB,Vera-BadilloFE,Aneja P,Oca˜naA,etal.Prognosticroleofneutrophil-to-lymphocyte ratioinsolidtumors:asystematicreviewandmeta-analysis.
J Natl Cancer Inst 2014;106:dju124, http://dx.doi.org/
10.1093/jnci/dju124.
[12] Lucca I, Jichlinski P, Shariat SF, Rouprêt M, Rieken M, Kluth LA, et al. The neutrophil-to-lymphocyte ratio as a prognosticfactor for patients with urothelial carcinoma of the bladder following radical cystectomy: validation and meta-analysis. Eur Urol Focus 2016;2:79—85, http://dx.
doi.org/10.1016/j.euf.2015.03.001.
[13] Bellmunt J, de Wit R, Vaughn DJ, Fradet Y, Lee J-L,Fong L,etal.Pembrolizumabassecond-linetherapyforadvanced urothelialcarcinoma.NEnglJMed2017;376:1015—26,http://
dx.doi.org/10.1056/NEJMoa1613683.
[14]HengDYC,XieW,ReganMM,WarrenMA,GolshayanAR,Sahi C, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascu- lar endothelial growth factor-targeted agents: results from a large, multicenter study. J Clin Oncol 2009;27:5794—9, http://dx.doi.org/10.1200/JCO.2008.21.4809.
[15]Li X, Ma X, Tang L, Wang B, Chen L, Zhang F,et al. Prog- nostic value of neutrophil-to-lymphocyte ratio in urothelial carcinomaoftheupperurinarytractandbladder:asystem- aticreviewandmeta-analysis.Oncotarget2017;8:62681—92, http://dx.doi.org/10.18632/oncotarget.17467.
[16]La Croce G, Moschini M, Nini A, Paolo D, Ventimiglia E, Di Trapani E, et al. MP58-10 preoperative hemoglobin to platelet ratio as a predictor of survival after radical cys- tectomy.JUrol2017;197:e776,http://dx.doi.org/10.1016/j.
juro.2017.02.1805.
[17]Demirtas¸ A, Sabur V, Akınsal EC, Demirci D, Ekmekcioglu O, Gulmez I, et al. Can neutrophil-lymphocyte ratio and lymphnodedensitybeusedasprognosticfactorsinpatients undergoing radical cystectomy? Sci World J 2013;2013:1—5, http://dx.doi.org/10.1155/2013/703579.
[18]Lepara Z, Junuzovic D, Sadovi´c S, Lepara O, Spahovic H, Omerbegovic D. Elevated CRP level in serum as potential biomarker and predictor of progression of bladder can- cer. Eur Urol Suppl 2016;15:e1324, http://dx.doi.org/10.
1016/S1569-9056(16)30188-9.
[19]KangM,JeongCW,KwakC,KimHH,KuJH.The prognostic significance of the early postoperative neutrophil-to- lymphocyte ratio in patients with urothelial carcinoma of the bladder undergoing radicalcystectomy. Ann Surg Oncol 2016;23:335—42, http://dx.doi.org/10.1245/s10434-015- 4708-8.
[20]JiménezMP,PerezSM,JorgePN,GonzálezJM,KimLD,Marrero UN,etal.1044Neutrophil-to-lymphocyteratioasaprognostic factor for survivalinpatientswithbladdercancerundergo- ingradicalcystectomy.EurUrolSuppl2016;15,http://dx.doi.
org/10.1016/S1569-9056(16)61045-X[e1044—e1044a].
[21]Motzer RJ, Ravaud A, Patard J-J, Pandha HS, George DJ, Patel A, et al. Adjuvant sunitinib for high-risk renal cell carcinoma after nephrectomy: subgroup analyses and updated overall survival results. Eur Urol 2018;73:62—8, http://dx.doi.org/10.1016/j.eururo.2017.09.008.
