Endothelin B receptor expression correlates with tumour angiogenesis

and prognosis in oesophageal squamous cell carcinoma

Running title: Endothelin B Receptor Expression in OSCC

T Tanaka 1， M Sho* ぺT Takayama 1， K Wakatsuki 1， S Matsumoto 1， K Migita 1， M

Ito 1， K Hamada 2， Y Nakajima 1

1 Deparlment 0; SUI 習ery ，Nara Medical University ， Nara ， Japan and 2Division of

Clinical and 加ves tJ 匂ative Medicine ， Nara Medical Universi ty， 840 Shijo-cho ，

Kashihara ， Nara ， 634-8522 ， Japan

*Corresponden 偲 :

Masayuki ShO ， M.D 圃， Ph.D. ，

Depa はment of Surgery ， Nara Medical Unive 陪 ity ，840 Shijo-cho ， Kashihara ， Nara ，

634-8522 ， Japan.

Phone:81 圃 744-29 ・8863; Fax:81 聞 744 ・24 ・6866;

E-mail: m 輔副o@naramed ・u.ac.jp

1

Abstract

Background: The endothelin 同 axis has been shown to play a pivotal role in

several human malignancies. The aim of this study was to clarify the clinical

impor 祖nce of endoth 剖in B re 偲 ptor (ETBR) in human oesophageal squamous

cell 叩 rcinoma (OSCC).

Methods: We evaluated ETBR expression in 107 patients with OSCC by

immunohistochemistry. Microvessel density (MVO) and Lymphatic vessel

density (LVO) were assessed by C031 and 02 ・40 immunostaining ， respectively.

Furthermore ， C04 ， C08 ， and C045RO+ tumour-infiltrating Iymphocytes (TILs)

were immunohistochemically analysed.

Results: Sixty-one (57%) 伺 ses showed high expression of E丁目R. ETBR

expression was ∞rrelated with several clini ∞pathological 幅ctors including

tumour di 宵'erentiation ，tumour depth ， and Iymph node me 旬stasis. The overall

and disease 時 specific survival rates we 陪 significantly lower in patients with high

ETBR expression than patients with low expression. Furthermore ， multivariate

analysis revealed that ETBR status was an independent prognostic factor for

patient surviva l. Mechanistic analysis indicated that MVO was significantly higher

in tumour tissues with high E丁目R expression compared to those with low

expression ， suggesting that angiogenesis may be a key mechanism in tumour

2

progression and metastasis of OSCC mediated by ETBR expression 圃 By

contrast ， the 陪 were no significant correlations between TILs and ETBR

expresslon.

Conclusion: ETBR plays a pivotal role in oesophagealωncer and may be

therapeutic target for this intractable malignancy.

Keywords: Endothelin B receptor ， oesophageal 伺 ncer ，angiogenesis

3

Oesophageal cancer is 1目前ly aggressive and is 由es lX血 leading cause of cancer deaths

in the world (Parkin et a! ， 2005). Al though several advancements in 仕le trea 凶 ent

in c1 uding chemotherapy and radiotherapy have been achieved ，仕le prognosis of patients

remains poor. Even in curatively resected patients ，也e5 幽year survival rate is be1 0w 50%

after s町 gery (Courrech Staal et a! ， 2009). In addition ， most cases are diagnosed at

advanced s匂.g e with lymphatic and hematogenous dissemination (Okines et a! ， 2010).

百lerefore ，to elucidate the underlying mechanisms of tumour progressio n， and to

ident 埼rnew biomarkers and 由erapeutic targets for oesophageal cancer are critically

問lportant to improve patients' prognosis.

Th e endothelial cell-derived peptide endothelin (ET) was discovered as a potent

vasocons 住ictor in 1988 (Yanagisawa et a! ， 1988). ET family includes 由ree 21 ・・ammo

acid peptides ， ET-l ， ET ・2 and E下3，which bind to two G-protein 圃 coupled receptors ，

endothelin receptor type A (ETAR) and endothelin receptor type B (ETBR) (Levin ，

1995). The system of 曲目e白ree endothelin peptides and two receptors is 向島町.ed to as

the endo 仕lelin axis (ET 幽 axis). Extensive studies have revealed the various roles of

endothelin system in cardiovasc u1 ar and renal disorders (F elds 飴in & Romero ， 2007;

Iglarz & Clozel ， 2007; Lehrke et a! ， 2001 ょTomobe et a! ， 1988). Furthermore ，血e

ET-axis has also been shown to play sigrrificant roles in a variety of human

malignancies in c1 uding ovarian ， prostate ， cervical ， breast ， colorectal ， lung cancers ， and

4

mel 阻 oma (N elson et al ， 2003). ET-axis has been reve a1 ed to re gu1 ate 加mo 町 grow 血

and metas 匂sis via various mechanisms in c1 uding cell prol 出 :ratio n， angiogenesis ，

antiapoptotic activity ， and immune modulation (Bagnato et al ， 2008; Buckanovich et al ，

2008; He rrm. ann et al ， 2006; Nelson et al ， 2003; Spinella et al ， 2002; W' 叫fmg et al ，

2004). Whil st upregt 白tion ofE 'f.・ 1回:p ression has been consistently r，句0巾 d in various

m a1 ignancies ，血e different expression and function of its r，邸eptors ETAR and ETBR

have been shown 血 distinct 加mo 町 s. Th us ， each rece 戸or has unique roles and its

ftmction may be d叩endent on cancer cell type. Furthermore ， selective antagonists for

each receptor as well as dual ETARAτ 'B R ant 昭onist have been widely investigate d，

and some of them were evalua 総d in clinic a1出 als (Bagnato et al ， 2008).

A few previous studies have addressed the role of ET-axis in oesophage a1 c組問r

(1曲 iba 由iet al ， 2003; Jiao et al ， 2008; Jiω et al ， 2007). 百 ese s旬dies have 晶own 也at

tissue or circula 也l8 sぽum.偲:p ression of endothelin has a significant prognostic va1 ue in

oesophageal sq 回 mous cell carcinoma (OSCC). A 組ldy has also shown 血at a d瑚 1

recept ぽ釦旬gonist inhibi 匂dmigr 甜on of human oesophageal c組 Cぽ ce l1血 vitro (Jiao

et al ， 2007). Fur 也釘more ，a recent study h部 iden 出ied aberrant promoter me 血c}' lation of

EDNRB gene ， which encodes ETBR in hu 血鑑18 ，in OSCC (Zhao et al ， 2009). Al though

these studies have sugges 旬d a potent ia1 role of ETBR 血 OSCC ，li 枇le is known about its

c1 inical impor 加 lce. In白 lS S' 加dy ，we ev a1 uated 加mo 町 ETBR expression to investigate

5

iぉclinical si 伊i:fi cance in OSCC.

MATERIALS AND METHODS

Patients

We examined 107 patients with oesophageal squamous cell carcinoma. 百le patients

underwent s明 ery in c1 uding subtotal oesophage 伽 my wl 也 dissection of regional

lymph nodes at Nara Medical University Hospital between Novemb 町 1995 and May

2007. None of them have received pre-operative 出 a回 en t， such 出 radiation or

chemotherapy. Th e p低ient's median age w 部 61 years (range ， 42 ・75). Postoperative

follow-up data were obtained 企om all patien 旬.百 le pathologic features of 也e

specimens were class i:fi ed based on 血e 7也 edition of 也e pa 也ologic a1

加血O町 '-node-met 部蜘is (百品。c1 ass i:fi cation of the In temational Union Against Cancer

(U ICC). Docume 脚 d informed consent w: 部 obtained from individ ua1 patients for use of

也.eir tiss 田 samples and c1 inic a1 record.

Immunohistochemical staining

Formalin-fixe d， paraffm-embedded tissues were cut inω 与問 sections ，deparaf 百nised

and rehydrat 吋血 a graded series of e血ano 1. Imm unohis 加chemic a1 staining for ETBR

6

W 出 Pぽ formed wi 也 a Dako Envision™ kit (D AK. O Cy tomatio n， Tokyo ， Japan).

An tigen retriev a1 was performed by heating tissue sections using a t釘get retriev a1

sol 叫io n， pH 9・o (O AKO). Then， the samples w 釘'e incubated for 5 min in peroxidase

blocking solution (D AK. O) to inhibit endogenous peroxidase and washed 由rice in fresh

PBS ， each of 5 min duration. 百le an ti-・ hu 血 an ETBR antibody (LS-A54 ， MBL ， USA)

diluted 1 :50 wi 血An tibody Diluent (D AK. O) was added and incubated at 37 0 C for 30

min. Aft er 也，e sections w 町 ewashed 由rice in PBS ， each of 5 min duratio n， a subsequent

reaction was carried out using second antibodies (O AKO) at 37 0 C for 30 min. The

sections were then washed thrice in PBS and 血e ∞lour w 部 displayed subsequently

wi 出 diaminobenzidine (DAB) (D AK. O) for approximately 5 min and rinsed in dis 世led

wat 既 Sections were counterstained with haematoxyl in， dehydrated in ethanol ， cle 釘 'ed

in xylene and coverslipped. Fur 也ermore ，for the ev a1 uation of tumour angiogenesis and

加mo 町白血回白18 lymphocy 旬s (百 Ls) ，immunohistρchemic a1 瑚血ings s町 CD31 ，

D2 ・40 ，CD4 ， CD8 ， and CD45RO were carried out by the same method as above.

Monoclon a1 antibodies were purchased 企om Abcam (Tokyo ， Japan); anti-CD31

(JC /7 0A) ， anti ・D240 (0 2・40) ，arr 佐CD4 (mA b51312) ， anti-CD8 (144B) ， and

組 lti ・CD45RO (U CH ・L1).

E 玄traction oftotal RNAs and real-time reverse transcriptase PCR analysis

7

To 泊lRNAw 邸 isola 句d 田 ing RNAspin Mini (GE He a1 thcare ， Tokyo ， Japan) and the

frrst-strand cDNA was synthesised 金om 1 同 RNA using a High Capaci 勿 cDNA

Rev ぽ se Transcription Kit (Applied Biosystems ， Fos 伽 City ，CA ， USA) ， according to

出e instructions of 血e manufacturer. Re a1 -time quantitative PCR analysis was ca:n ed

out using an ABI Pr ism 7700 sequence detec 加[" system (Applied Biosystems). All

primer/probe 田ts were purchased from Applied Biosystems. PCR was carried out using

血e TaqMan Univ' 町'Sa1 PCR Mast ぽ Mix (Applied Biosystems) using 1 J.U of cDNA in a

20μ1 fma1 reaction volume. 百le PCR thermal cycle conditions were as follows: initial

S旬:p at 95 0 C for 10 m in， followed by 40 cycles of 95 0 C for 15 s and 60 0C for 1 min. The

expression level of 血.e houseke 叩 ing gene (3 2-mi ぽ oglob 叫m w 節 measured as an

inte rna1 refer 聞 .ce with a s刷伽d curve to determine the 血tegrity of 旬mplate RNA for

a1 1 specimens. 官le ratio of the mRN A level of each g阻 e was ca1 culated as follows:

(absolute ∞py number of each gene) / (absolute copy number of (3 2叩 icroglobulin).

Evaluation of ETBR expr ，回sion

All c邸 esw ぽ e 01 部 si 五ed into two groups 部 cording 旬也ep 町 'cent of positively stained

cells of ETB R. The ev a1 uation w 邸 performed by authorised pathologists who had no

knowledge of the patients' clinic a1 sぬ加s and outcome. Low ETBR immunoreactivity

W 回 defmed as sぬ血血g of く 50% of 加盟O町 cells ， whereas high ETBR

8

im.m unoreactivity was defmed 部 staining of ~ 50% of tumour cells.

Evaluation ofmicrovessel density 明fVD) and Iymphatic vessel density (L VD)

Vessel count w 田部sessed by light microscopy 血 areas of the 加血O町 containing 也e

highest numb 町 s of cap i1 1aries. 百le hi gh1 y v田α.u ar areas were identified at low power

(x40 m 碍;nifi cation). After six are 邸 of most abundant positively stained vessels were

iden 出ie d， a vessel count w師陣formed on a x200 fiel d，組 d the average count of six

fields w 部品加mined 出血.e MVD or LVD respectively 部 previously described (Ik: eda

etal ， 1999; Matsumoto etal ， 2007; Weidneretal ， 1991).

Evaluation oftumou r- infiltrating IymphocytesσILs)

We selec 七ed five are 部 Wl 血血e most abundant positively stained cells in each tissue

und ぽ x40ma 郡出cation. We 也m ∞unted these cells and ca1 culated the mean number

of each samples stained by ant トCD4 ，CD8 ， and CD45RO antibody ， respectively.

Statistical analysis

Comparisons amo 噌也e c1凶 ca1 and pathologic a1 fea 加res were ev a1 uated 出血giand

Fisher's exact te 蹴 .S 凶 stical significance between two groups of param 甜 ic da 飴 W 出

evaluated using an unpaired Student's t-tes t. S町 vival c町 ves were estima 旬d using the

9

Kaplan-Meier metho d， and the si ，伊ificances of diff ぽ ences between survi va1 c町 ves

were dete rm.in ed using 10 含rank tes t. M u1 tivariate ∞mparisons of surviv a1 dis 出butions

were made using Cox proportio na1 h田沼町d models. A P va1 ue < 0.05 w 田 considered to

indicate statistic a1 1y signific 阻 .ce.

RESULTS

ETBR e玄P時 ssion in human 0倒 ophag 伺 1 squamous cell carcinoma. We frrst

examined 也.e expression of ETBR on 107 oesophage a1 squa 血 ous cell carcinoma tissues

by immunohis 旬chemistry. Six 全y-one (57.0%) cases showed high expression of ETB R.

ETBR expression w 邸 identified 血血e cytopl 部 m of carcinoma cells (Figure 1). We

then compared 也e relative expression of ETBR between oesophage a1 cancer and

non-c 組 .cer tissues using av ai1 able 企ozen tissues by q回出回tive re a1 -time PCR ana1 ysis.

百le ETBR expression in oesophage a1 carcinoma t凶suesw 剖 significantly high 紅白組m

non-carcinoma tissues (P = 0.036; Figure 2A). F町也.ermore ，也e ETBR expression of

cancer tissue was consistently hi g1姐 y 血組曲at ofnon ・cancer tissue in each individu a1

patientwI 血 OSCC (Figure 2B). 官lese da 土a suggested 也at ETBR might have some role

in OSCC and may be a potential 也.erapeutic 旬rge t.

10

Correlation between ETBR exprl 倒 sion and clinicopathological findings. To clari 命

the c1 inic a1 significance of ETBR in OSCC ， we ∞mpared ETBR e: 却 ression wi 也

c1 inicopathologic a1 features. As a res u1t， th ぽ 'e was no significant cαTelation between

ETBR expression and sev ぽa1 clinicopathologic a1 variables including gender ， age ， and

distant metastatic status (Table 1). In con 国民 ETBR e} 中Iressl 佃 was significan t1 y

correlated with 加血O町 differentiatio n，加 mour dep th， lymph node metas 飽sis ，lymphatic

血v出 io n， venous invasio n，阻 d 加mour stage (P = 0.013 ，0.041 ，0.037 ，0.002 ，0.019 ， and

0.036 ， respectively). Moreover ，larger 加mo 町 and metastatic lymph nodes w 釘 'e found

more often in p仰ents wi 血 highETBR 回 pression (P = 0.012 and 0.016 ， respectively).

Th us ， ETBR 郎 pression in OSCC may play a critic a1 role in 旬血our growth and

metas 阻SIS.

Prognostic value of ETBR 偲 pr l倒 sion in OSCC. Next we examined the prognostic

imp 町tance of ETBR expression in patients with OSCC. 百le over a11 and

disease 圃 specific surviv a1 rates were significan t1 y lower in patients wi 也 high ETBR

expression 血an in those wi 血 low expression (P = 0.003 and 0.002; Figure 3). 百le

5・year ove ra11 surviv a1 rate for ETBR high expression patients was 56.2 % and for low

expression patients w 部 20 .1%. In血 is s加dy ，gender ，加mσ 1U" status ， lymph node

metastasis and VenOUS invasion were a1 so found to have a prognostic va1 ue for patient

11

survival of OSCC. To det ぽmine ind 叩 endent variables among these prognostic f1郎 tors ，

we performed a m u1 tivariate analysis using Cox propo 出onal hazard models. Th e

analysis revealed 血at， in addition to gender and lymph node metastasis ， ETBR

e却 ression was one of 血e independent prognostic factors for the patients with OSCC

(Table 2). Taken toge 血er ，ETBRis 白nctional and plays a significant role in OSCC.

A 路 ociation of E 'I冨 Re 玄pr l闘 sion with tumour angiogenesis and Iymphangiogenesis.

To investig 蹴也e unde r1 ying mechanisms ofETBR e却 ression in OSCC ， we examined

加mo 町 angiogenesis and lymphangiogenesis. To 由is en d， we performed

immunohistochemical staining of CD31 and D2 ・40 in 也.e same tissues of OSCC in

which ETBR e: 却，ression w 田 evaluated (Figure 4A and B). We found 也atMVD w 部

significan t1 y high ぽ血 tissues wi 也 high ETBR expression ∞mpared to those with low

expression (8.0 4:土:4 .31 and 4.03 土3.99; P < 0.001; Figure 4C). On the other hand，也 .ere

W 邸 no si 伊 ificant corr e1 ation be 阿 een tumo 町 ETBR expression and LVD (hi gh ETBR

group and low group; 2.96 土:2 .54 and 4.02 :1: 2.70 ， respectively). We 白r血er 鉱Ial yzed the

postoperative recurrence pa 低em. In由 is series ， a to 匂1 of 61 patients had rec 町 rence

d百 ing follow- 叩 period. 官町全y-e 場1t patients had hematogenous metastasis in c1 uding

12 in 也e liv 民 21 血也e lung ， and 8 in the bone. Fur 血ermore ，40 patie 臨 had lymphatic

me 陶旬sis. There were sigt 凶 cant correlations of ETBR 回 pression wi 血 hemaωgenous

12

and lymphatic metastasis (P = 0.042 阻 d 0.025 ， respectively). On the other han d， neither

MVDn 町 LVD correlated wi 血 each metastatic pat ぬm Taken 旬ge 由民 angogglesis

might play some roles in tumour progression ra 出町血組 me 胞stasis in rel 説iontoETB R.

Association between ETBR e豆pression and tumour infiltrating Iymphocyt 倒 .

Finally ， we investigated 也.e immunomodulatory function of ETBR in OSCC. We

performed immunohistochemic a1 analysis of TILs including CD4 ， CD8 ，血dCD45RO.

As ares 叫t血釘'e were no significant c田relations of tumour ETBR expression with TILs

in any T cell subsets (Figure 5).

DISCUSSION

Accumulating evidence demons 回.t es 也at ET-axis plays significant roles 血旬mours by

a number of complex mechanisms in c1 uding cell surviv a1， prolill ぽ atio n， migratio n，

invasio 民 epitheli a1岨 esenchym a1位祖国 itio n， me 血l)' latio n， angiogenesis ， and immune

modulation (Bagnato et al ， 2008; Carducci et al ， 2003; Eltze et al ， 2007; Nelson et al ，

2003; Nelso n， 2003). F班血.ermore ，seve ra1 clinic a1 studies have shown 也atE 下 1 and its

two recep 旬rs ，ETAR and ETB R， are ov 釘 'ex 戸'essed in various actu a1 human 凶 ncer

tissues (Bagna 加 et al ， 2008). Al血 ough a few studies have a1 so r，叩orted a potenti a1 role

ofET ・1 in human oesophage a1 cancer ， the da 同紅'e very li 血自d. Since a basic research

13

has suggested a potenti a1 involvement of EτBR in OSCC ，也is study focused on 血e

c1 inic a1 si 伊 fican 回 of ETBR 叫 Iression 也 OSCC tissues (Zhaο et al ， 2009). As a

resul t， we found sever a1 impo 蜘 1t findings. Firs t， ETBR was highly ex:p ressed in cancer

tissues compared 加 non 凶 C組 問r tissues. Similar results were also reported in several

other 加mo 町 s (Eltze et al ， 2007; Wulfmg et al ， 2004). Secon d，也釘 'e were si 伊.ifi cant

associations of ETBR ex:p ression wi 血 some clinicopa 由ologi ca1 features including

加血O町 difti ぽ 'entiatio n，加血 O町 size ，lymph node metastasis ， and venous invasion. Th us ，

data suggest 血at ETBR ex:p ression in OSCC may play 姐 important role in bo 也知mour

progression and me 旬S旬sis. Thir d， most importantly ， ETBR ex:p ression has a significant

prognostic va1 ue 血 OSCC. 百1e patients wi 血 high ETBR ex:p ression had a worse

prognosis in OVI 町'a1 1 and diseas e- specific surviv a1 compared 加 patients with low

expression. To our knowledge ，也is is 也e 五rstrep 町 tto demons 位 脱 出at ETBR is an

independent prognostic marker for hun1at1 oesophage a1 can ，侃r. Pr evious studies have

also demons 回 ted 也at 加mo 町 ETBR ex:p ression was a negative prognostic marker in

o也.erc 組 .cer (Shen et al ， 2011). On the 0血er han d， ETBR gene e却 ression w 部 reported

as a positive prognostic marker in renal cell carcinoma (W uttig et al ， 2012). Th erefore ，

血.e function of ETBR may be dependent on 如mo 町勿pe.

N ext， we investiga 総d 血.e und 町iying mechanisms in ETBR 阻 pression of OSCC.

We 自rst focused on angiogenesis 白紙 plays a key role in 加mourgrow 也 andme 胎S旬.sis.

14

ET ・1 has been shown to promo ぬ創:t.g iogenesis bo 血 directly and indirectly by inducing

endothelial cell surviv a1， prol 正釘atio n， invasio n， and upre gu1組 ng VEGF production in

the vasc u1 ature 血rough ETBR (Kan 由la 庇et al ， 2009; Sa1ru:世 et al ， 2000). In白 is study ，

we found a si 伊 ificant positive correlation of ETBR expression wi 也ru::t.g iogenesis

ev a1 uated by counting intra 加盟O町 microvessel density. This is consistent wi 血 previous

伽1a indi 宜erent 加血O町仰叫自喝 et al ， 2004). Although several studies have

demonstrated 也at ru::t.g iogenesis has a significant role and prognostic value in OSCC ， the

mechanisms to con 仕01 angiogenesis are not 白lly elucidated (Igara 由i et al ， 1998;

Ki tadai et al ， 1998; Tanigawa et al ， 1997). F班白釘more ，our 白畑a1 so indicated 白紙

也ere was a si 伊 ificant association between ETBR expression and venous invasion.

Taken toge 血er ，加mo 町 ETBR expression may promote neovas cu1 arisation and enhance

veno 田血:v asion. In addition ， ETBR has been shown to play a critical role 血加mour

lymphangiogenesis (Spinella et α1， 2009). However ， we had no significant correlation

between tu 血 O町 ETBR expression and lymphangiogenesis determined by D2-40

immunos 句血血8・百lerefore ，lymphangiogenesis may n叫 be a key mechanism in 旬血O町

progression mediated by ETBR in OSCC.

Th en， we ev a1 uated 也.e association of ETBR wi 由加血O町lIDID田 nty. Pr evious

studies have shown 也atET ・1 axis a1 so has a unique role 加 re gu1 ate immune response in

加mo ぽ environment (Buckanovich et al ， 2008; Gri血 shaw et al ， 2004; G百叫i et 叫

15

2004). E下1 axis mod u1 ate the activati on， di 宜erentiati ∞ and trafficking of

tumour.inftl 位低ing immune 関 Us. Fur 血ermore ，ETBR plays a cruci a1 role in

lymphocy もehoming and OVI ぽ expression of endothel ia1 ETBR in 加mo 町 sprevents T cell

hom 血g. We have recently reported 白紙 TILs ，especi a1 1y CD45RO 噌中Iressing memory

T cells ， have significant prognostic va1 ue in OSCC (Enomoto et al ， 2012). 百lerefore ，

wehypo 也.esised 白紙 ETBR may re gu1 ate 百Ls in OSCC. To c1 ari 命血is possib 出帆 we

ev a1 uated association of tumour expressing ETBR wi 血 TILs in c1 uding CD4 ， CD8 ，姐d

CD45RO by immunohistochemis 位y. As a res u1t， we found no significant ∞rrelations

between 也em. Thus ， 0町 da 句 suggest 血at ETBR may have 1加le role in mod u1 ating

immune response in the progression and me 旬S旬sis in OSCC. However ，白r也er studies

are required to completely ru1 e out the immunologic a1 role ofETBR in OSCC.

Bωed on diverse functions and involvement of ET .axis in a variety of 加血ours ，

sever a1 agents 回 geting E下 axis have been extensively investigated (B agnato et al ，

2008; Kan 白laft et al ， 2009). Fur 仕lermore ，some endothelin an 同 onists have been

c1 inic a11 y 肝a1 uated (Bagnato et al ， 2011). A1 though most of 血em are selective ETAR

m ぬgo 国sts ，ETBR an 旬gonist is a1 so av ai1 able for clinic a1 use and expected of clinic a1

benefi t. Compared to 0血釘伺邸ぽs，血ere were on1 y a few therapeutic agents clinic a11 y

av ai1 able in OSCC. In additi on， OSCC is generally diffic u1 t to treat and c町 eby

nons 町 gic a1 anti 加mourtrea 加 .ents. 百lere It町 e，combination therapy of ETBR an 飽gonist

16

wi 也 other antitumour 喝ents m 町 be desirable for 也IS m回 ctable 加mour.O 町 data

indicates a significant relationship between ETBR and angiogenesis. 百1ぽぱore ，

combination of ETBR with anti-angiogeneic 回 a回 ent may eX!ぽ t a synergistic effect

Since anti-angiogenic 出 a回 ent including anti- VEGF monoclonal antibody is widely

used in clinical canc ぽ也erapy ，也IS 白町apeutic s佐ategy can be relatively easily tested.

Howevl 町，血ere are limitations in 也is study and fur 血.er studies are essential before

clinical application. Th e num. ber of samples evaluated for 也iss 加.dy is rel 組vely smalL

In fa ct， we have 仕'eated more patients with oesophageal canc 町血血e study period. Due

to some re 田 ons including sample limi 旬tions ，we rando m1 y chose s佃 lples of 107

patients and investigated.τ 'h erefore ， a large-scale and care fu1 evaluation for Eτ 'B R is

critically important to confrrm reproduc ゐ丑i勿 ofourda 肱

In con c1 usio n， we have shown 白紙 ETBR is expressed in human oesophageal canc 町

and h田 a si gni:fi cant prognostic value. Thi s study may provide a rationale for

developing a novel cancer therapy targeting ETBR for 血is fa:凶 malignant disease.

However ， more 阻 penmen 凶 and c1ini cal evidence is needed 加 prove 血e si 酔 ificance

ofETBR 田 anewth ぽ apeutic tぽge t.

17

ACKNOWLEDGEMENTS

百us work was supported by the following grants: Gr ant s- in-Aid for Scientific Research

from 血.e Ministry of Educatio n， C叫旬re ，Sports ， Science and Technology of Japan (M.

Sho).

18

REFERENCES

Ba gnato A， Lo izidou M ， Pflug BR， Curwen J， Grow ∞抗 J 包01 1) Ro le of 由.e endothelin 部 is

and its antagonists in the trea 凶.e nt of 叩 ncer. 島・J Pharm8co1168(2): 220 ・233

Bagnato A， Spinella F， Ro sano L (2 008) 百.e endothelin 回 isin 回 n田 r: the promise and the

challenges of molecularly targeted therap y. Can J Physiol Pharm8co186(8): 473 ・484

Buckanovich RJ ， Facciabene A， Kim S， Benencia F， Sasaroli D， Bal担 t K， Ka tsaros D，

O'Brien"Jenkins A， Gimotty PA ， Coukos G (2008) Endo 由.elin B re 田p加，r mediates the

endothelial barrierωTcellhom 泊g to tumors and disables immune therapy. Nat Med 14(1):

28 ・36

Ca 吋ucci MA， Padley RJ ， Breul J， Vogelzang NJ ， Zonnenberg BA， Daliani DD ， Schulman CC ，

Nabulsi AA， Humeric 油 ouse RA， Weinberg MA， Schmi 抗札， Nelson JB 白003) Effect of

endothelin"A 四回pωr blockade with atrasentan on 加 mor progression in men with

hormone"refractory pros 阻te cancer: a randomized ， phase IT ， plaω ，bo ・controlled trial. J Clin

Onco121( 4): 679 ・689

Courrech St 制 1 E F， van Coevorden F， Cats A， Al eman BM ， van Velthuysen 阻 " Boot H ，

Pee 加rsMJ ，vanS 姐 dick JW (2009) Outωme oflow"volume surgery for esophageal 叩 ncerin

a high"volume re 島町'al 田 nter. An n Surg Onco116(12): 3219 ・3226

Eltze E， Bertolin M ， Korsch 血g E， W，叫宣 ng P， Ma ggino T， Le lle R (2007) E却 ression and

prognostic relevan 四 of endothelin"B recepωrinvulv 町 cancer. Oncol R ，申 18(2): 305 ・311

Enomoto K， Sho M ， Wakatsuki K， Ta kayama T， Matsumoto S， Nakamura S， Akah ori T，

Tanaka T， Migi 阻K， I加 M ，Nak 司加a Y (2012) Prognostic importan 田 of tumou r- in 五ltrating

memory T cells 血 oesophageal squamous cell carcinoma. Clin E:. 甲 Immuno1168(2): 186 ・191

Felds 旬加 C，Ro mero C (2007) Ro le of endothelins 担 hypertension. Am J Th er 14 (2): 147 ・153

Goseki N ， Ko ike 阻 Yoshida M (1 992) 回目加Ipathologic charact 息，ristics of early stage

esophageal 開削noma. Acomparative study with gastric carcinoma. Cancer69(5): 1088 ・1093

Gr 皿 shaw MJ ， Hagemann T， Ayhan A， Gillett CE ， Binder C， Balkwill FR (2004) A role for

endotheli 且・2 組.d iお四回p加rs in breast tumor 田Uinvasio 且 CancerRes64(7): 2461 ・2468

19

Guruli G， Ptlug BR， Pecher S， Makarenkova V， Shur 也 M R， Nelson JB (2 004) Funct ion and

survival of dendritic 偲 lls depend on endothelin.1 and endothelin 間四ptor au 初田恒.e loops.

Blood104(7): 2107 ・2115

Herrmann E， Bo gemann M ， Bierer S， Eltze E， Hertle L， wl 叫fing C 白006) The endothelin

田 i自 in urologic tumors: mechanisms of tumor biology and therapeutic implications. E:. 叱pert

Rev An ticancer Ther6(1): 73 ・81

Ig 世田hi M ， Dhar D K， Kubota H， Yamamo 加 A，EI. As salO ， Nagasue N (1 998) Th e prognostic

時国fican 田 of microvessel density and thymidine phosphorylase expression in squamous

all carcinoma of the esophagus. 白 ncer 82(7): 1225 ・1232

Iglarz M ， Clozel M 包007) Mechanisms of ET- 1.indu 田 d endothelial dysfunction. J

ぬrdiovasc Pharmaco150(6): 621 ・628

Ikeda N ， Adachi M ， Taki T， Huang C， Hashida H ， Takabayashi A， Sho M ， Nakaj 加 a Y，

Kan ehiro H， His 佃 aga M， Nak 阻.0 H ， 間 yake M (1999) Pr o伊.o stic sig 国自国na of

angiogenesis 血 human pancreatic 伺 n田 r.BrJ ぬncer79(9 ・10): 1553 ・1563

Ishibashi Y， Hanyu N， Nakada K， Suzuki Y， Y;田 namoωT ，Takahashi T， Ka wasaki N，

Ka wakami M ， Matsushima M ， Urashima M (2 003) Endothelin protein expression as a

si 伊出cant prognostic factor in oesophageal squamous 田 II carcinoma. Eur J 白 ncer 39(10):

1409 ・1415

Jiao W， Xu J， Zheng J， Shen Y， Lin L， Li J 位008) Elevation of cirαuat 担g big endothelin.1:

an independent prognostic factor for tumor recurrena and survival in patients with

esophageal squamous all 明白血oma. BMC Cancer8: 334

Jiao W J， Xu J， Pan H， Wang TY， Shen Y (2007) E臨 ct of endothelin.1 in esophageal

squamous 田 H 伺 rc 岨oma invasion and its correlation wi 出回thepsin B. 防'o rld J

Ga8troenterolI3(29): 4002.4005

Kan曲 laft LE ， Facciabene A， Buckanovich RJ ， Coukos G (2 009) Endothelin B reapω r， a

newt 町 get in canar immune therap y. Clin Cancer Re8 15(1 4): 4521 ・4528

Ki ta 血iY， H 町 uma 民'Ib kuto 凶T， τ'ana ka S， Sumii K， C町 V叫ho M ， Kuwabara M ， Yoshida K，

H問問Ka jiyama G， Tahara E (1998) Si 伊出問na of vessel count and va 蜘加 endothelial

growth factor in human es 叩 hageal carcinoma 自. Glin Gancer &84(9): 2195 ・220020

Le hrke 1， Waldherr R， Ri tz E， Wagner J 但001) Re nal endothelin 圃 1 and endo 也elin 四国p加r

type B expression in glomerul 世 diseases with proteinur 抱. J Am Soc Nephro1 12(11):

2321 ・2329

Le vin ER (1995) Endothelins. N Eng1 J Med 333(6): 356 ・363

Matsumoto S， Yamada Y， N町出yoM ，UenoM ，羽maki H， Miki K， Wakatsuki K， Enomoto K，

Yokotani T， Nakajima Y 但007) Pr ognostic significance of platelet'derived gro 帆 hfactor'BB

e却 ression 加 human esophageal squamous 曲 11 carcinomas. An ticancer .R es 27(4B):

2409 ・2414

Nelson J， Bagnato A， Battistini B， Nisen P 白003) The endothelin axis: eme 唱曲g role in

cancer. -̂加 .R ev Cancer 3(2): 110 ・116

Nelson JB (2 003) Endo 也elin inhibition: novel therapy for prostate canω:r. J Uro1170(6 pt

2): S65 ・67; discussion S67 ・68

Okines A， Sharma B， Cunningham D 包010) Peri 叩 erative management of esophageal

can 田 r. Nat .R ev Cli n Onco17(4): 231 ・238

Park 担 DM ，Bray F， Ferlay J， Pi sa 国 P 但005) Global cancer s祖tisti 四， 2002. CA Cancer J

C1in 55 (2): 74 ・108

Salani D， Taraboletti G， Ro阻 .no L， Di Castro V， Bor 閥抗iP， Giavazzi R， Bagna 加 A 包000)

Endothelin'l induce 自組問gio 酔国cphenotype 匝 cultured endo 也.elial 田l1 s and stim 叫ates

neovascul 町 ization in vivo. Am J Pa 品。11157(5): 1703 ・1711

Shen C， Yang L， Yuan X 白011) Endo 出.e lin B re 田pωre 却 ression in human astrocytoma:

association wI 也 cli 国∞pathologi 個 1variables and survival outωmes. In t J Neurosci121(11):

626 ・631

Spinella F， G町 rafa E， Di Castro V， Ro sano L， Ni ，∞tra MR ， Caruso A， Natali PG ， Bagnato A

(2009) Endothelin'l stimulates lymphatic endothelial 田 lls and lymphatic vessels 10 grow

andinvade. 白 ncer Res 69(6): 2669 ・2676

Spinella F， Ro sano L， Di Castro V， Natali PG ， Bagnato A (2 002) Endo 也elin'l induces

vascular endothelial growth 血ctor by inω'easing hypox 鼠胴阻ducible facto r- lalpha in ovarian

21

carcinoma cells. J Bi oJ Chem 277(31): 27850 ・27855

四国gawa N， Matsumura M ， Am aya H， Ki taoka A， Shimomatsuya T， Lu C， Muraoka R，

Tanaka T (1 997) 百四or vascularity ∞rrelates with the prognosis of patients with

esophageal squa 皿 ou 島田H 倒民泊oma. CanCel ・79 包): 220 ・225

'lb mobe Y， Mi yauchi T， Sai 加 A，Yanagisawa M ， Kim ura S， Goto K， Ma saki T (1988) E質問ts of

endothelin on the renal a此ery from spontan 凹 usly hypertensive and Wistar Kyoto rats. Eu r

J Ph l1 rmaco] 152(3): 373 ・374

Weidner N， Semple JP， Welch WR， Folkman J (1991) 引lID or angiogenesis and

metastasis ..∞ r即 lation in 阻va 自ive bre 剖 t田町匝oma. N EngJ JMed 324(1): 1・8

Wulfing P， Ke rst 泊g C， Ti o J， Fischer RJ ， wi 叫fing C， Poremba C， Diallo R， Bocker W， Ki esel L

(2004) Endothelin'I' ， eIl!d othelin 企， and eIl!d othelin 骨 rec ，eptor express 旬nis ∞rrelated with

vascular endothelial growth factor e却 ression and angiogenesi 自国 breast cancer.α in

白 'nc ，θ'r Res 10(7): 2393 ・2400

Wuttig D， Zastrow S， Fus sel S， 'lb ma 阻， Me 阻h町 dtM ，Kalma且K， Junker K， Sa 珂myatavJ ，

Boll K， Hackerm 叫ler J， Ro lle A， Grimm MO ， W 廿血 MP (2012) CD31 ， EDNRB and TSPAN7

are promising progno 自ticmarker 白血clear 園田llrenal 田 II carcinoma revealed by genome'wide

expression analyses ofprimary 印 morsandme 祖 stases. Int J Cancer 131(5): E693 刊 4

Y阻 agisawa M ， Ku 封h田 aH ，K回目a S， 'lb mobe Y， Kobayashi M ， Mitsui Y， Yazaki Y， GoωE

Masaki T (1 988) A novel po 胞nt vaso ∞nstrictor peptide produ 田 d by vascular endothelial

cells. Nature 332(6163): 411 ・415

Zhao BJ ， Sun DG ， Zhang M ，τ加 SN ，Ma X (2 009) Iden 組問.tion of aberrant promoωr

methylation of EDNRB gene in esophageal squamous 偲 II carcinoma. Dis Es ophagus 22(1):

55 ・61

22

FIGURE LEGEND

Figure 1. Expression of ETBR in oesophageal squamous 関 11 carcinoma.

Represen 祖tive 但 se of positive exp 陪 ssion of ETBR. (A) Original magnification ，

x2 0. (8) Original magn 師団tion ，x100. (C) Representative 伺 se of negative

expression of ETB R. Original magnifi 個目on ，x100.

Figure 2. Comparison of ETBR expression between cancer and non-cancer

tissue of the oesophagus. (A) The cumulative ETBR e却 re 錨 ion in 伺 nωrtissue

was significantly higher compa 陪 d with that in non 剛 cancer tissue as determined

by real-time PCR (n = 10 of each ， P = 0.036). (8) 官1e expression in cancer

tissue is ∞nsistently higher than that in non- 伺 n偲 r tissue of individual

oesophagealcanωr patients 開

Figure 3. Prognosis of oesophageal can 偲 rpatients ac ∞rding to tumour ETBR

expression s泊tus. (A) Overall survival 悶te (P = 0.003). (8) Disease-specific

survival rate (P = 0.002).

Figu 陪 4. Association of tumour angiogenesis with ETBR expression. A: CD31

expression was detected in endothelialωIIs of tumour vasculature (Original

23

magnification ， x( 00). B: Immunohistochemistry for 02-40 showing Iymphatic

channels with staining of endothelium (Original magnification ， x( 00). C:

Microvessel density (MVO) was signi 百回ntly higher in tissues with high ETBR

expression compared to those with lowexpression.

Figure 5. Association of tumour infiltrating Iymphocytes with E丁目R expression.

There were no significant correlations of tumour ETBR expression with the

number of tumour infiltrating Iymphocytes in C04 ， C08 ， and C045RO ，

respectively.

24

A

A B 5. む

1.6 置. Cancer

1.4 E CコNor トCancerc o 1.2 ω ω

3.0 主》@《 10

BGX 2 •

羽ω〉23 0.8

.4。f〉2o 5 2.0 偲

0.6 庇ω E。K

0.4 ~ 1.0

0.2

。 。σJJF r，d' h

2 3 4 5 6 7 8 9 10

Patient

A 100 • •

..L ETBR Low expression

喝授。圃轟 、要 60 i -'''，

5 4 0 . H

~ I .，_ゐ“島情 .1.1 ・-。ω 〉 ‘・輔園

20 1 - - - ，帽輔輔副帽帽 . . . . . . . 欄欄醐個個

Log rank P = 0.003 。。 20 40 60 80 100

Months after operation

B 100

，圃崎、

議史 iわL 山 L ETBR Low expression 、町圃，

4@ 偲』圃圃， 80 i ¥ .

EγBR High expression -・・・・ー

何〉

.~ 60

3 的

。

申'0也立40

a. ω @

g g 20 I Log rank P = 0.002 .-。。。 20 40 60 80 100

Months after operation

C P< 0.001

9 注、-ω 申 8

S: 7 。@ 6

一。5

ω ω ω 4

〉g3 .: 2 ~

。High Low

ETBR expression

詰 4000 5 命コ4

E 300 一一1.11 ω u

ト

ち 200、ーjg 邑コ 100z

G

CD4

割聞 ETBR High expr! 信ssion

cコETBR しow expression

CD8 CD45RO

T cell in 制trating into OSCC

Table 1. Relationship between ETBR expn 闘 sion and clini ∞pathologi 回 Icharacteristics in

oesophageal sequamous cell carcinoma

ETBRexpre 踊 ion

Total High Low

Charact 町 istics (n = 107) (n 園 61) (n =46) p. 司lalue

Gender Male 88 48 40 0.315 Female 19 13 6

Age (years) (mean 土s.d.) 61.1 土7.01 61.0 土:7.66 0.948

Oi 仔ren 世昌司on

Well 39 16 23 0.013 Mode 悶 lte 52 37 15 Poor 16 8 8

Tumour depth pT1 23 9 14 0.041 pT2 21 9 12 pT3 58 39 19 pT4 6 4 1

Lymph node metas 臨sispNO 37 16 21 0.037 pN1 70 45 25

Oistant metastasis MO 106 61 45 0.4 31 M1 。 1

Lymphatic invasion Negative 20 5 15 0.002 Positive 87 56 31

Venous invasion

Nega 伽B 60 28 32 0.019

Positive 47 33 14

Tumours 祖ge

16 6 10 0.036

11 32 14 18

111 42 29 13

IV 17 12 5

Tumour size(mm) 43.7 士18.9 33.5 士17.6 0.012

(mean 士50)

Number of Iymph node metastasis 4.94 士7.15 1.90 士3.40 0.016

(mean 士50)

Table 2. Univanate and mu lUv ariate a陥 Iysis for over 百11 survivalln 107 pa 世間箇 with 個釦凶渦ge 剖 sequamous 国 11 悶悶inoma

Unlvarlate analys l$ Mu 附varlatea 回Iy sl$

Characterlstl 曲 P噌叫ue HR 9S %CI P-v alue

Gender (Fema 聞 vs. M aIe) 0.017 2.967 1.274 -6.944 0.012

Tumour depth (T1晦. T2 -4) 0.003 1.693 1.112- 2. 5'π 0.014

Lym 凶 nodemet 醐踊恒例ovs. N1) <0.001 2.165 1.119-4.1 騎 0.022

Venous invaslon (N 健同ive vs. Positive) 0.015 0.725 0.400 -1.312 0.287

Tumoursize 0.014 0.99 自 0.981- 1.018 0.948

ETBR expression (Low vs. High) 0.0 日3 2.104 1.108- 3.996 0.023

Abb 珂Iviations: HR=hazar 世間tio; 95% CI=95% ∞nfiden 閣 Inter 噛 |