Cooverexpression of ERBB1 and ERBB4 receptors predicts poor clinical outcome in pN+ oral squamous...

RESEARCH PAPER

Cooverexpression of ERBB1 and ERBB4 receptors predicts poorclinical outcome in pN+ oral squamous cell carcinomawith extranodal spread

Sabrina Daniela Silva • Moulay A. Alaoui-Jamali •

Michael Hier • Fernando Augusto Soares •

Edgard Graner • Luiz Paulo Kowalski

Received: 25 February 2013 / Accepted: 7 December 2013 / Published online: 15 December 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Overexpression of members of the ErbB

receptor family is common in oral squamous cell carcino-

mas (OSCC); however, their prognostic value for aggres-

sive OSCC has been debated. Extranodal spread to cervical

lymph nodes is the most significant prognostic indicator in

OSCC. In the present study, we investigated the clinical

significance of single versus paired overexpression of

members of the ErbB receptor family in 82 OSCC patients

with lymph nodes metastasis, with or without capsular

rupture (CR) followed by at least 10 years. Immunohisto-

chemistry analysis revealed a common overexpression of

ErbB1 (P = 0.021), ErbB2 (P = 0.001), ErbB4 (P =

0.048), as well as MMP-2 (P = 0.043) in OSCC cases with

CR?. Increased expression of ErbB1 was associated with

MMP-2 in tumors with advanced clinical stages, including

poorly differentiated (grade III) tumors (P \ 0.050). Vas-

cular embolization was associated with MMP-2 (P = 0.021)

and MMP-13 (P = 0.010) overexpression. Survival analysis

revealed a lower survival probability in tumors over-

expressing ErbB1 (P = 0.038), ErbB4 (P = 0.043), and

MMP-12 (P = 0.050). As well a strong association was

observed in cases with high risk of recurrence and strong

immunostaining for ErbB1 (P = 0.017), ErbB4

(P = 0.008), MMP-1 (P = 0.003), MMP-2 (P = 0.016),

MMP-10 (P = 0.041), and MMP-13 (P = 0.005). Stratified

multivariate survival analysis revealed a strong prognostic

interdependence of ErbB1 and ErbB4 cooverexpression in

predicting the worst overall and disease-free survivals

(P = 0.0013 and P = 0.0004, respectively). Taken toge-

ther, these results support a cooperation of ErbB1, ErbB4,

and members of the MMP family in predicting OSCC

invasion and poor clinical outcomes.

Keywords Oral squamous cell carcinoma � ErbB �MMP � Capsular rupture � Extranodal spread �Prognostic factor

Introduction

Oral squamous cell carcinomas (OSCC) represent the sixth

most frequent cancer worldwide [1–3]. Despite of

improvements in surgery, radiotherapy, and chemotherapy

over the last decade, the survival rates have improved

marginally, and the survival probabilities for patients with

OSCC is among the lowest of the major cancers [3–5].

S. D. Silva (&) � M. Hier

Department of Otolaryngology-Head and Neck Surgery - Sir

Mortimer B. Davis-Jewish General Hospital, McGill University,

Montreal, QC, Canada

e-mail: [email protected]

S. D. Silva � M. A. Alaoui-Jamali

Departmens of Medicine and Oncology, and Lady Davis

Institute for Medical Research and Segal Cancer Centre of the

Jewish General Hospital, McGill University, Montreal, QC,

Canada

S. D. Silva � L. P. Kowalski

Department of Head and Neck Surgery and

Otorhinolaryngology, A.C. Camargo Cancer Center, Sao Paulo,

SP, Brazil

F. A. Soares

Department of Anatomic Pathology A.C. Camargo Cancer

Center and Faculty of Dentistry, University of Sao Paulo,

Sao Paulo, SP, Brazil

E. Graner

Department of Oral Diagnosis, School of Dentistry of Piracicaba,

University of Campinas (UNICAMP), Piracicaba, SP, Brazil

123

Clin Exp Metastasis (2014) 31:307–316

DOI 10.1007/s10585-013-9629-y

Currently, the only specific prognostic factors routinely

considered for treatment decision are the clinical stage, site

of the primary tumor, and presence of nodal or distant

metastasis [6–8]. Even when a combination of surgical and

non-surgical approaches is used, more than 50 % of OSCC

patients experience cancer recurrence either locally, at

regional lymph nodes, or at distant sites [3, 5, 9]. The

presence of regional metastasis decreases survival by

almost 50 %, and invasion beyond the lymph nodes with

capsular rupture (CR) significantly predicts high mortality,

representing the most significant adverse prognostic indi-

cator for OSCC patients [5, 9–12].

Several signaling mechanisms, in particular those asso-

ciated with ErbB tyrosine kinase receptors, have been

implicated in OSCC oncogenesis [13–19]. Overexpression

or amplification of specific members of the ErbB family

such as ErbB1/EGFR/HER1 (epidermal growth factor

receptor), ErbB2/neu/HER2, ErbB3/HER3, and ErbB4/

HER4, has been associated with regulation of cell prolif-

eration, survival and differentiation [5, 13, 18, 20–26].

These receptors share a common structure with a large

glycosylated binding extracellular domain, a single

hydrophobic transmembrane region, and a cytoplasmic

domain with tyrosine kinase activity [27]. The signal

transduction is mediated through binding of the ligand with

receptor, except for ErbB2 that has no known ligand but

can be activated via heterodimerization and transactivation

through interactions with the other members [23, 28]. A

biological characteristic of ErbB receptors is their affinity

to a multitude of ligands (e.g. EGF, TGF, amphiregulin,

heregulins, betacellulin, epiregulin, and heparin-binding

EGF) [29], as well as their propensity to form various

homo- and heterodimers, which activate distinct signaling

pathways generating a variety of cellular responses [30,

31]. These receptors have a broad range of function in

autocrine and paracrine signaling [31] correlated with

tumor progression and risk of metastasis [13, 18, 21, 25,

26].

In human OSCC cases, most studies have focused on

single ErbB receptor but the contribution of cooverex-

pression of multiple ErbB, which is relevant to ErbB het-

erodimerization and downstream signaling, has not been

thoroughly investigated in relation to metastasis. In this

study, we evaluated the expression patterns of all members

of ErbB receptors, as well as their downstream target

MMPs in a cohort of 82 OSCC patients with progressive

disease and a follow-up over 10 years. Taking in consid-

eration that extranodal spread in cervical lymph nodes

represents a significant prognostic indicator in OSCC, we

selected sets of tumors presenting lymph nodes metastasis

(pN?) with capsular rupture (CR?) versus negative ex-

tranodal spread (CR-) to establish ErbB association with

relevant clinicopathological features of advanced OSCC.

Materials and methods

Study population

A retrospective study was performed by analyzing 82 OSCC

paraffin-embedded tissue specimens from 25 OSCC patients

with lymph nodes metastasis (pN?) with capsular rupture

(CR?) and 57 patients with metastasis without extranodal

spread (CR-). All patients were treated at the Department of

Head and Neck Surgery and Otorhinolaryngology, A.C.

Camargo Cancer Center, Sao Paulo, Brazil. The eligibility

criteria included previously untreated patients, without a

second primary tumor and submitted to treatment in the

single institution. All patients were advised of the procedures

and provided written informed consent. The National Human

Research Ethics Committee approved this study. The medi-

cal records of all patients were examined to obtain detailed

clinicopathological data (clinical stage, histological grade,

vascular embolization, perineural infiltration, and extranodal

spread), including information regarding lifestyle (smoking

habit and alcohol consumption), and demographic data (age,

gender, and race) (Table 1). The tumors were re-staged

according to the 2002 version of the International Union

Against Cancer (TNM) classification [32] and grouped as

early clinical stage (I ? II) or advanced clinical stage

(III ? IV). All cases were followed-up after treatment and

the disease recurrence was histologically confirmed. The

histological grade was determined on the basis of classifi-

cation proposed by the World Health Organization [33].

Vascular embolization was classified according to the pre-

sence or absence of neoplastic cells, located in the wall or

lumen of blood or lymphatic vessels, perineural infiltration

considered present when the tissue adjacent to the peri and/or

intra-tumoral nerves were involved by the neoplastic cells.

Tissue Microarray Platform (TMA)

To construct the tissue microarray (TMA), core biopsies

were taken from previously defined areas of the primary

tumor, with a Tissue Microarrayer (Beecher Instruments�,

Silver Springs, USA). Tissue cores with 1.0 mm were

punched from each specimen and arrayed in duplicate on a

recipient paraffin block. Each core was spaced 0.2 mm

apart. Blocks were then cut (3 lm) and transferred with an

adhesive tape to coated slides for subsequent UV cross-

linkage (Instrumedics Inc�, Hackensack, NJ). Slides were

dipped in a layer of paraffin to prevent oxidation, and kept

in a -20 �C freezer.

Immunohistochemistry

The sections were deparaffinized, rehydrated in graded etha-

nol solutions. Thereafter, sections were treated with

308 Clin Exp Metastasis (2014) 31:307–316

123

endogenous peroxidase quenching (0.3 % H2O2 for 15 min)

and blocked for avidin/biotin (DAKO Biotin Blocking Sys-

tem� Dako Corporation, Carpinteria, CA) and protein

(DAKO Protein Block Serum-Free�, Dako), 20 min each

prior to primary antibody incubation. Pressure cooker antigen

retrieval consisted of one period at 125 �C for 30 min and

90 �C for 10 min in 10 mM citric acid solution (pH 6.0) fol-

lowed by a washing step with phosphate-buffered saline

(PBS). The incubations with the primary antibodies diluted in

PBS were conducted overnight at 4 �C: anti-ErbB1/EGFR

(Dako 1:400), anti-ErbB2 (Dako 1:10,000), anti-ErbB3

(Labvision Corporation, Fremont, CA, USA 1:300), anti-

ErbB4 (Neomarkers Corporation, Fremont, CA, USA 1:300),

anti-MMP-1 (Neomarkers Corporation 1:400), anti-MMP-2

(Neomarkers Corporation 1:200), anti-MMP-9 (Neomarkers

Corporation 1:200), anti-MMP-10 (Neomarkers Corporation

1:50), anti-MMP-12 (Abcam, Cambridge, MA, USA – 1:500),

and anti-MMP-13 (Neomarkers Corporation 1:100). The

sections were washed and incubated with secondary anti-

bodies (Advanced TM HRP Link, Dako Cytomation, K0690,

Denmark) for 30 min followed by the polymer detection

system (Advanced TM HRP Link, Dako Cytomation) for

30 min at room temperature. Reactions were developed with a

solution containing 0.6 mg/ml of 3,30-diaminobenzidine

tetrahydrochloride (DAB, Sigma, St Louis, MO) and 0.01 %

H2O2 and then counter-stained with Mayer’s hematoxylin,

dehydrated and mounted with a glass coverslip. Positive

controls (a tissue known to contain the antigen under study)

were included in all reactions in accordance with manufac-

turer0s protocols. The negative control consisted in omitting

the primary antibody and incubating slides with PBS; and

replacing the primary antibody with normal serum. The

immunohistochemical reactions were performed in duplicate

on different TMA levels, representing fourfold redundancy

for each case. The second slides were 25 sections deeper than

the first, resulting in at least 250 lm of distance between the

two sections with different cell samples for each tumor.

Immunohistochemical scoring

Immunohistochemical scoring was blinded to the outcome

and clinical aspects of each tumor specimen. After scan-

ning each tumor specimen in low power field to choose the

most stained area, at least five fields were evaluated under

high power. The presence of a clearly visible dark brown

precipitation was considered positive. Positivity for ErbB

family members was identified as a sharply demarcated cell

membrane staining or a diffuse intracytoplasmic labeling.

ErbB overexpression was considered positive when the

30 % of the tumor cells showed uniform intense mem-

branous staining according to the recommendations of The

American Society of Clinical Oncology/College of Amer-

ican Pathologists (ASCO/CAP) published in January 2007

[34]. In brief: 0 = no staining; 1? = faint or barely per-

ceptible membranous staining; 2? = weak membranous

staining; 3? = strong complete membranous staining.

Scores 0 and 1? were interpreted as negative and 2? and

3? were considered as positive in this study. Only cyto-

plasmatic staining for MMPs members were considered as

positive. Evaluation of the molecular markers included the

proportion of reactive cells within the tumors and the

staining intensity. Slides were scored by a method descri-

bed in a previous study [35] for (i) intensity of staining (0,

negative; 1, weak; 2, moderate and 3, intense), (ii) per-

centage of epithelial cells staining (0, 0–5 %; 1, 6–25 %; 2,

26–50 %; 3, 51–75 % and 4, 76–100 %). The product of

(i) and (ii) was used as the total multiplied score, where

0–2 indicates a negative score (-), C3 a positive score (?)

Table 1 Clinicopathological characteristics of OSCC patients

Characteristics

Variable Categories n (%)a

Gender Male 70 (85.4)

Female 12 (14.6)

Race Caucasians 68 (82.9)

Non-caucasians 14 (17.1)

Smoking habit No 5 (6.9)

Yes 67 (92.1)

Alcohol consumption No 9 (12.5)

Yes 63 (87.5)

Clinical stage I ? II 27 (32.9)

III ? IV 54 (67.1)

Histological grade I 17 (22.7)

II 26 (34.7)

III 32 (42.6)

Vascular embolization No 51 (69.9)

Yes 22 (30.1)

Perineural infiltration No 36 (49.3)

Yes 37 (50.7)

Surgical margins Negative 67 (85.9)

Positive 11 (14.1)

Any recurrence or metastasis No 45 (54.9)

Yes 37 (45.1)

Local recurrence No 54 (65.9)

Yes 26 (34.1)

Regional recurrece No 71 (86.6)

Yes 11 (13.4)

Distant metastasis No 60 (73.2)

Yes 22 (26.8)

Status Alive 52 (63.4)

Dead 30 (36.6)

a Percentages considering cases with complete information

Clin Exp Metastasis (2014) 31:307–316 309

123

[36]. For statistical analysis, the samples were categorized

into negative versus positive.

Statistical analysis

For frequency analysis in contingency tables, statistical

analyses of associations between variables were performed

by the v2 test or Fisher’s exact test (with significance set at

P \ 0.05) and for continuous variables the non-parametric

Mann–Whitney U test. The overall survival was defined as

the interval between the beginning of treatment (surgery)

and the date of death or the last information for censored

observations. The disease free interval was measured from

the date of the treatment to the date when recurrence was

diagnosed. Overall survival and disease-free survival

probabilities were estimated by the Kaplan–Meier method,

and the log-rank test was applied to assess the significance

of differences among actuarial survival curves with a 95 %

confidence interval. All analyses were performed using the

statistical software package STATA (STATA Corporation,

College Station, TX, USA).

Results

Clinicopathological characteristics of OSCC

with lymph nodes metastasis

The studied population consisted of 82 patients, of which 70

(85.4 %) were male and 12 (14.6 %) female, with a mean age

was of 54.8 years (range 38–78 years). History of alcohol

consumption was observed in 63 patients (87.5 %) and

tobacco smoking in 67 (92.1 %). With regard to the ethnic

group, 68 (82.9 %) were Caucasians and 14 (17.1 %) non-

Caucasians (Table 1). The time of complaint was defined as

the time between the date of recognition of the first sign or

symptom of the disease by the patient and the date of first

visit to a professional who was qualified to refer the patient

for definitive diagnosis and treatment. The mean time of the

complaint was 6 months (range 1–68 months). A total of 27

cases (32.9 %) were at early clinical stage (I ? II) and 54

(67.1 %) at advanced clinical stage (III ? IV) (Table 1).

Twenty-three patients (28 %) were treated by surgery alone.

Advanced tumors were treated by combined-therapeutics

such as surgery associated with adjuvant radiotherapy (49

cases, 59.8 %) or surgery, radiotherapy, and chemotherapy (9

cases, 11 %). Of the 82 eligible cases, vascular embolization

was found in 22 cases (30.1 %), perineural infiltration in 37

(50.7 %), and involved margins in 11 (14.1 %). Seventeen

cases (22.7 %) were histologically well differentiated (grade

I), 26 cases (34.7 %) were moderately differentiated or grade

II, and 32 cases (42.6 %) were poorly differentiated or grade

III (Table 1).

Relationship between OSCC with lymph nodes

metastasis and clinicopathological characteristics

As expected, advanced clinical stage revealed significant

difference between patients with pathological lymph nodes

metastasis (pN?) and extranodal spread (CR?)

(P = 0.021) (Table 2). Patients with metastatic lymph

nodes with extranodal spread had higher risk of recurrence

(P \ 0.001) compared to CR-, which were significantly

associated with local (P \ 0.001) and distant metastasis

(P = 0.004) (Table 2).

Table 2 Correlation between capsular rupture status and clinico-

pathological variables in OSCC samples

Characteristics Group n (%)a

Variable Categories CR- CR?b P

value

Gender Male 48 (68.6) 22 (31.4) 0.655

Female 9 (75) 3 (25)

Race Caucasians 49 (72.1) 19 (27.9) 0.270

Non-

caucasians

8 (57.1) 6 (42.9)

Smoking habit No 3 (60) 2 (40) 0.743

Yes 45 (67.2) 22 (32.8)

Alcohol consumption No 5 (55.6) 4 (44.4) 0.450

Yes 43 (68.3) 20 (31.7)

Clinical stage I ? II 23 (85.2) 4 (14.8) 0.021

III ? IV 33 (61.1) 21 (38.9)

Histological grade I 11 (64.7) 6 (35.3) 0.354

II 16 (61.5) 10 (38.5)

III 25 (78.1) 7 (21.9)

Vascular

embolization

No 38 (74.5) 13 (25.5) 0.346

Yes 14 (63.6) 8 (36.4)

Perineural infiltration No 28 (77.8) 8 (22.2) 0.223

Yes 24 (64.9) 13 (35.1)

Surgical margins Negative 47 (70.1) 20 (29.9) 0.719

Positive 9 (81.8) 2 (18.2)

Any recurrence or

metastasis

No 39 (86.7) 6 (13.3) \0.001

Yes 18 (48.6) 19 (51.4)

Local recurrence No 44 (81.5) 10 (18.5) 0.001

Yes 11 (46.4) 15 (53.6)

Regional recurrece No 52 (73.2) 19 (26.8) 0.053

Yes 5 (45.5) 6 (54.5)

Distant metastasis No 47 (78.3) 13 (21.7) 0.004

Yes 10 (45.5) 12 (54.5)

Status Alive 42 (80.8) 10 (19.2) 0.007

Dead 15 (50) 15 (50)

a Percentages considering cases with complete informationb CR-: capsular rupture negative; CR?: capsular rupture positive


123

Relationship between immunohistochemical markers

and OSCC with lymph nodes metastasis

The distribution of immunohistochemical markers in rela-

tion to patients’ groups (CR? versus CR-) is summarized

in Table 3. The expression levels of ErbB1, ErbB2, ErbB3,

and ErbB4 were 49.4, 34.2, 13.2, and 50 %, respectively.

Two distinct patterns of ErbB receptors positivity were

identified as a sharply demarcated cell membrane staining

or a diffuse intracytoplasmic labeling (Fig. 1a, b, c and d).

The intracytoplasmic expression of ErbB receptors did not

correlate with clinicopathological features. However,

positive correlation with paired ErbB receptor cooverex-

pression was found in these samples (ErbB1–ErbB4

P \ 0.0001, ErbB2–ErbB4 P = 0.016) (data not shown).

Interesting, the expression of ErbB1 (P = 0.021), ErbB2

(P = 0.001), and ErbB4 (P = 0.048) occurred more fre-

quently in patients whose tumors present extranodal spread

(CR?) in comparison to CR- (Table 3).

In addition to ErbB receptors, intracytoplasmic MMPs

were highly expressed in the studied OSCC cases (Fig. 1e, f,

g, h, i, and j; Table 3). Levels of expression of MMP-1,

MMP-2, MMP-9, MMP-10, MMP-12, and MMP-13 were

67.9, 52, 56, 73.2, 80.5, and 70.7 %, respectively.

Noticeable, MMP-2 overexpression was the most significant

(P = 0.043) in patients with extranodal spread (CR?) than

CR- (Table 3). Additionally, a significant association was

observed among MMP family members (MMP-1-MMP-

12, P = 0.001; MMP-1-MMP-13, P \ 0.0001; MMP-2-

MMP-9, P = 0.003; MMP-10-MMP-13, P = 0.012;

MMP-12-MMP-13, P = 0.008), and between ErbB4 with

MMP-9 (P = 0.002) (data not shown).

Significant positive associations were found between

ErbB1 and MMP-2 cooverexpression and advanced clinical

stages of OSCC (P \ 0.050). MMP-2 expression was also

most common in poorly differentiated grade III tumors

(P = 0.050) in comparison with lower grade tumors. OS-

CCs presenting vascular embolization were significantly

associated with MMP-2 (P = 0.021) and MMP-13 immu-

nostaining (P = 0.010) (Table 4).

Relationship between ErbB and OSCC with lymph

nodes metastasis and prognosis

At the end of the follow-up period, 52 (63.4 %) patients

were alive and 30 (36.6 %) died during the follow-up,

among which 28 patients (93.3 %) died due to OSCC and 2

(6.7 %) for other causes (Table 1). The overall survival

time varied from 1 to 144.4 months (mean of 67 months).

The 5-year rates for overall survival and disease free sur-

vival were 56 and 48 %, respectively. A significantly lower

survival probability was observed in patients whose tumors

have overexpression of ErbB1 (log-rank test, P = 0.038),

ErbB4 (log-rank test, P = 0.043), and MMP-12 (log-rank

test, P = 0.050) (Fig. 2a, b, and c, respectively). Further-

more, the stratified multivariate survival analysis showed

that cooverexpression of ErbB1 and ErbB4 receptors

resulted in a worst overall survival probability (log-rank

test, P = 0.0013) (Fig. 2d) considering all the possible

combination among the markers investigated (data not

shown).

In addition, 37 patients (45.1 %) had tumor recurrence

during the study course within a mean time of 24.8 months

(range 1–88.3 months). Local recurrences were detected in

28 patients (34.1 %), while 11 cases (13.4 %) had regional

lymph nodes involved, and 22 patients (26.8 %) had local

and/or regional recurrence and developed distant metasta-

sis to the lung, bone, or brain (Table 1). Association

between a higher risk of recurrence was observed in tumors

presenting overexpression of ErbB1 (log-rank test,

P = 0.017) (Fig. 2e), ErbB4 (log-rank test, P = 0.008)

(Fig. 2f), MMP-1 (log-rank test, P = 0.003) (Fig. 2h),

MMP-2 (log-rank test, P = 0.016) (Fig. 2i), MMP-10 (log-

rank test, P = 0.041) (Fig. 2j), and MMP-13 (log-rank test,

P = 0.005) (Fig. 2k). Equally important, our results show

that ErbB1 and ErbB4 was able to strongly predict the

disease free survival, since patients whose tumors showed

Table 3 Correlation between ErbB and MMP immunopositivity and

capsular rupture status in OSCC samples

Characteristics Group n (%)a

Variable Categories CR- CR?b P value

ErbB1 Negative 32 (82.1) 7 (17.9) 0.021

Positive 22 (57.9) 16 (42.1)

ErbB2 Negative 41 (82) 9 (18)

Positive 11 (43.3) 15 (57.7) 0.001

ErbB3 Negative 44 (66.7) 22 (33.3) 0.398

Positive 8 (80) 2 (20)

ErbB4 Negative 30 (78.9) 8 (21.1) 0.048

Positive 22 (57.9) 16 (42.1)

MMP-1 Negative 16 (61.5) 10 (38.5) 0.309

Positive 40 (72.7) 15 (27.3)

MMP-2 Negative 29 (80.6) 7 (19.4) 0.043

Positive 23 (59) 6 (41)

MMP-9 Negative 32 (78) 9 (22)

Positive 20 (58) 14 (41.2) 0.061

MMP-10 Negative 15 (68.2) 7 (31.8) 0.874

Positive 42 (70) 18 (30)

MMP-12 Negative 14 (87.5) 2 (12.5) 0.063

Positive 43 (65.2) 23 (34.8)

MMP-13 Negative 17 (70.8) 7 (29.2) 0.867

Positive 40 (69) 18 (31)

a Percentages considering cases with complete informationb CR-: capsular rupture negative; CR?: capsular rupture positive


123

the positive expression of both receptors also had a higher

risk of recurrence compared to patients with negative

immunostaining (log-rank test, P = 0.0004) (Fig. 2g).

Discussion

OSCC progression to locoregional and distant metastasis is

multifactorial and involves both intrinsic factors innate to

cancer cells and factors contributed by tumor microenvi-

ronment and the host. This progression is also affected by

OSCC heterogeneity, which greatly defines the cancer cell

biological behavior. This heterogeneity likely contributes

to a wider tumor response rates seen for tumors of the same

stage and with the same treatment. At present, therapeutic

decisions are based on clinicopathological parameters,

including clinical stage, site of the primary tumor, and

presence of nodal or distant metastasis. Although useful,

these factors often fail to differentiate between more and

less aggressive lesions. Treatment failures occur in many

patients due to the high local recurrences and neck

metastases rates, despite of the aggressive multimodality

therapeutic options available.

Nowadays, the literature has shown that lymph nodes

metastases with extranodal spread represent the most

significant adverse prognostic indicator in OSCC patients

[5, 9–12]. Therefore, the present study evaluated the

expression of ErbB and MMP family members in OSCC

patients with (CR?) or without (CR-) extranodal spread

in their metastatic lymph nodes (pN?) in order to assess

the prognostic impact in patients followed up at least by

10 years. We demonstrated that ErbB and MMP family

members are highly cooverexpressed in OSCC samples and

specific combination of ErbB1 and ErbB4 receptors predict

poor outcomes in OSCC with extranodal spread.

ErbB receptor-associated signaling regulates tumor cell

proliferation, survival, migration, invasion, and metastasis

[36]. Although ErbB1 overexpression has been reported in

OSCC [18, 21, 22], the combination of ErbB1 and ErbB4

cooverexpression reported in this study revealed to repre-

sent a strong prognostic predictor for OSCC with extran-

odal spread. Here, the cooverexpression of ErbB1 and

ErbB4 highlights the contribution of ErbB cooperation in

OSCC in contrast to previous studies focusing only in a

single receptor.

As noted in the introduction, ErbB receptors have a

propensity for heterodimerization, which account for their

activity in carcinogenesis. Indeed we have reported earlier

a great impact of ErbB receptor heterodimers on gene

transcriptional profiling compared to homodimers [37]. In

Fig. 1 Representative immunohistochemical reactions for ErbB1 (a),

ErbB2 (b), ErbB3 (c), ErbB4 (d), MMP-1 (e), MMP-2 (f), MMP-9

(g), MMP-10 (h), MMP-12 (i), and MMP-13 (j) in OSCC samples.

Two distinct patterns of ErbB positivity were identified: a sharply

demarcated membrane staining (a and b) and an intracytoplasmic

labeling (c and d). A clear intracytoplasmic labeling for all studied

MMP family members was identified. Original magnification: 9200


123

this context, the correlation of ErbB1 and ErbB4 coover-

expression in progressive OSCC could be explained by

distinct signaling attributed to this paired combination. For

instance, ErbB1 receptor couple efficiently to multiple

activated downstream signaling, including phospholipase

C, mitogen-activated protein kinases (MAPK), protein

kinase C, phosphatidylinositol-3 kinase (PI3K) and Janus

kinase 2/signal transducer and activator of transcription

pathways. In contrast, ErbB4 couples primarily to MAPK

and PI3K pathways [31]. However, ErbB4 phosphorylation

has been reported to induce a more sustained activation of

the Ras-MAPK signaling compared to the others ErbB

members [38, 39]. ErbB4 is the only receptor tyrosine

kinase cleaved by a two-step proteolytic process releasing

an 80-kDa intracellular domain with an active tyrosine

kinase and nuclear localization capabilities [41]. Studies

performed in preclinical models have shown that the sol-

uble 80-kDa domain of ErbB4 can directly associate with

STATs in the nucleus and acts as a transcriptional co-

activator [40]. Furthermore, formation of ErbB1and ErbB4

heterodimer following activation by neuregulin can lead to

a stronger activation of serine/threonine phosphorylation

compared to activation of ErbB1; this lead to receptor

tyrosine phosphorylation and recruitment of Shc/Grb2,

which is crucial for neoplastic transformation [41, 42].

This cooperative signaling also applies to the broad

alteration seen for MMPs, which are downstream of ErbB

signaling. Multiple MMPs regulate extracellular matrix

(ECM) remodeling as well as other aspects of tumor

microenvironment such as angiogenesis [43–47]. Several

Table 4 Relationship between molecular markers and clinical stage, histological grade, and vascular embolization of OSCC samples

Variable Clinical stage n (%)a Histological grade n (%)a Vascular embolizationa

I ? II III ? IV P value I II III P value Negative Positive P value

ErbB1

Negative 17 (44.7) 21 (55.3) 0.050 6 (16.7) 13 (36.1) 17 (47.2) 0.307 24 (70.6) 10 (29.4) 0.999

Positive 9 (23.7) 29 (76.3) 11 (31.4) 12 (34.3) 12 (34.3) 24 (70.6) 10 (29.4)

ErbB2

Negative 19 (38.8) 30 (61.2) 0.305 8 (17) 20 (42.6) 19 (40.4) 0.125 29 (64.4) 16 (35.6) 0.288

Positive 7 (26.9) 19 (73.1) 9 (37.5) 6 (25) 9 (37.5) 17 (77.3) 5 (22.7)

ErbB3

Negative 22 (33.8) 43 (66.2) 0.703 15 (24.6) 22 (36.1) 24 (39.3) 0.945 40 (69) 18 (31) 0.890

Positive 4 (40) 6 (60) 2 (20) 4 (40) 4 (40) 6 (66.7) 3 (33.3)

ErbB4

Negative 16 (43.2) 21 (56.8) 0.124 6 (17.1) 14 (40) 15 (42.9) 0.416 24 (70.6) 10 (29.4) 0.729

Positive 10 (26.3) 28 (73.7) 11 (30.6) 12 (33.3) 13 (36.1) 22 (66.7) 11 (33.3)

MMP-1

Negative 9 (36) 16 (64) 0.774 8 (33.3) 7 (29.2) 9 (37.5) 0.336 18 (81.8) 4 (18.2) 0.131

Positive 18 (32.7) 37 (67.3) 9 (18) 19 (38) 22 (44) 32 (64) 18 (36)

MMP-2

Negative 17 (43.6) 22 (56.4) 0.050 12 (31.6) 16 (42.1) 10 (26.3) 0.050 26 (81.2) 6 (18.8) 0.027

Positive 8 (22.9) 27 (77.1) 5 (15.6) 10 (31.3) 17 (53.1) 19 (55.6) 15 (44.4)

MMP-9

Negative 10 (25) 30 (75) 0.083 8 (21.1) 13 (34.2) 17 (44.7) 0.504 23 (62.2) 14 (37.8) 0.236

Positive 15 (44.1) 19 (55.9) 9 (28.1) 13 (40.6) 10 (31.3) 22 (75.9) 7 (24.1)

MMP-10

Negative 7 (31.8) 15 (68.2) 0.860 5 (25) 5 (25) 10 (50) 0.564 15 (78.9) 4 (21.1) 0.319

Positive 20 (33.3) 39 (66.7) 12 (21.8) 21 (38.2) 22 (40) 36 (66.7) 18 (33.3)

MMP-12

Negative 3 (18.8) 13 (81.2) 0.167 3 (23.1) 1 (7.7) 9 (69.2) 0.062 11 (73.3) 4 (26.7) 0.742

Positive 24 (36.9) 41 (63.1) 14 (22.6) 25 (40.3) 23 (37.1) 40 (69) 18 (31)

MMP-13

Negative 5 (20.8) 19 (79.2) 0.121 7 (31.8) 7 (31.8) 8 (36.4) 0.469 20 (90.9) 2 (9.1) 0.010

Positive 22 (38.6) 35 (61.4) 10 (18.9) 19 (35.8) 24 (45.3) 31 (60.8) 20 (39.2)

a Percentages considering cases with complete information


123

studies have demonstrated that metastatic cancer cells often

lose their epithelial properties and present a fibroblast-like

phenotype (epithelial-mesenchymal transition–EMT). The

polarized epithelial cells are converted into motile mes-

enchymal cells followed by degradation of ECM compo-

nents involving MMPs; these steps are necessary for the

metastatic cascade [48–50]. A study by Xu et al. [43]

reported that up-regulation of MMP-2 is accompanied by

advanced clinical stage. Furthermore, this study also

showed increased expression of MMP-2 in poorly differ-

entiated tumors (grade III) in comparison with negative

tumors. In the same line, Patel et al. [51] have shown that

MMP-2 and -9 expression is higher in malignant tissue

from OSCC patients presenting lymph node metastasis. We

have previously demonstrated that MMP-2 and -9 activities

are also associated with lymph nodes recurrence and dis-

tant metastasis [52]. Yorioka et al. [52] confirmed that

gelatinolytic activity of MMP-2 and MMP-9 are associated

with a shortening of the disease free survival. This is in line

with the view that MMP activity is an important compo-

nent of disease progression and metastasis [5, 11, 53].

Finally, the presence of lymph nodes metastases was an

independent predictor factor for shorter overall and dis-

ease-free survivals. In conclusion, our findings support that

ErbB and MMP family members are highly expressed in

OSCC samples and they have significant importance in the

prognostic features being able to predict OSCC invasion

and poor clinical outcomes.

Acknowledgments This work was supported by Fundacao de

Amparo a Pesquisa do Estado de Sao Paulo (FAPESP 06/61039-8 and

CEPID/FAPESP 98/14335). Silva SD was supported by a FAPESP

fellowship (06/61040-6). The authors would like to acknowledge Jose

Ivanildo Neves, Carlos Ferreira Nascimento, and Severino Ferreira

for their technical assistance.

P =0.004

Negative MMP1

Positive MMP1

P= 0.003

P =0.013

Negative ErbB1

Positive ErbB1

P= 0.017

Negative MMP2

Positive MMP2

P= 0.016

Negative MMP10

Positive MMP10

P= 0.041

Negative MMP13

Positive MMP13

P= 0.005

Negative ErbB1

Positive ErbB1

P= 0.038

0

Months

Negative ErbB4

Positive ErbB4

P= 0.05P =0.043

Negative ErbB4

Positive ErbB4

P= 0.008

25 50 75 100 125 0

Months

25 50 75 100 125 0

Months

25 50 75 100 125 0

Months

25 50 75 100 125

0

Months

25 50 75 100 125

0

Months

25 50 75 100 125 0

Months

25 50 75 100 125 0

Months

25 50 75 100 125

0

Months

25 50 75 100 125 0

Months

25 50 75 100 125 0

Months

25 50 75 100 125

a b c d

e

i j k

f g h

Dis

ease

Fre

e su

rviv

alD

isea

se F

ree

surv

ival

Dis

ease

Fre

e su

rviv

alD

isea

se F

ree

surv

ival

Dis

ease

Fre

e su

rviv

al

Dis

ease

Fre

e su

rviv

al

Dis

ease

Fre

e su

rviv

al

Fig. 2 Overall survival analysis and disease free survival. Patients

with lymph nodes metastasis presenting strong ErbB1 (a), ErbB4 (b),

MMP-12 (c), and ErbB1-ErbB4 (d) immunolabeling had shorter

survival rate in comparison with the negative and weakly stained

cases. Patients with strong ErbB1 (e), ErbB4 (f), ErbB1-ErbB4

combination (g), MMP-1 (h), MMP-2 (i), MMP-10 (j), and MMP-13

(k) immunolabeling had higher risk of recurrence in comparison with

the negative and weakly stained cases. (line): negative or weak

immunohistochemical expression for ErbB1 (a and e), ErbB4 (b and

f), ErbB1-ErbB4 combination (d and g), MMP-12 (c), MMP-1 (h),

MMP-2 (i), MMP-10 (j), and MMP-13 (k); (spaced dashes): strong

positive immunostaining for ErbB1 (a and e), ErbB4 (b and f),ErbB1-ErbB4 combination (d and g), MMP-12 (c), MMP-1 (h),

MMP-2 (i), MMP-10 (j), and MMP-13 (k). Kaplan–Meier test


123

Conflict of interest The authors declare that they have no conflict

of interest.

References

1. Curado MP, Hashibe M (2009) Recent changes in the epidemi-

ology of head and neck cancer. Curr Opin Oncol 21:194–200

2. Parkin DM, Bray F, Ferlay J, Pisani P (2005) Global cancer

statistics, 2002. CA Cancer J Clin 55:74–108

3. Hardisson D (2003) Molecular pathogenesis of head and neck

squamous cell carcinoma. Eur Arch Otorhinolaryngol

260:502–508

4. Cojocariu OM, Huguet F, Lefevre M, Perie S (2009) Prognosis

and predictive factors in head-and-neck cancers. Bull Cancer

96:369–378

5. Choi S, Myers JN (2008) Molecular pathogenesis of oral squa-

mous cell carcinoma: implications for therapy. J Dent Res

87:14–32

6. Takes RP, Rinaldo A, Silver CE, Piccirillo JF, Haigentz M Jr,

Suarez C, Van der Poorten V, Hermans R, Rodrigo JP, Devaney

KO, Ferlito A (2010) Future of the TNM classification and

staging system in head and neck cancer. Head Neck

32:1693–1711

7. Gospodarowicz MK, Miller D, Groome PA, Greene FL, Logan

PA, Sobin LH (2004) The process for continuous improvement of

the TNM classification. Cancer 100:1–5

8. Greene FL (2002) The American Joint Committee on Cancer:

updating the strategies in cancer staging. Bull Am Coll Surg

87:13–15

9. Kowalski LP, Sanabria A (2007) Elective neck dissection in oral

carcinoma: a critical review of the evidence. Acta Otorhinolar-

yngol Ital 27:113–117

10. Shaw RJ, Lowe D, Woolgar JA, Brown JS, Vaughan ED, Evans

C, Lewis-Jones H, Hanlon R, Hall GL, Rogers SN (2010)

Extracapsular spread in oral squamous cell carcinoma. Head

Neck 32:714–722

11. Rosenthal EL, Matrisian LM (2006) Matrix metalloproteases in

head and neck cancer. Head Neck 28:639–648

12. Greenberg JS, Fowler R, Gomez J, Mo V, Roberts D, El Naggar

AK, Myers JN (2003) Extent of extracapsular spread: a critical

prognosticator in oral tongue cancer. Cancer 97:1464–1470

13. Silva SD, Cunha IW, Nishimoto IN, Soares FA, Carraro DM,

Kowalski LP, Graner E (2009) Clinicopathological significance

of ubiquitin-specific protease 2a (USP2a), fatty acid synthase

(FASN), and ErbB2 expression in oral squamous cell carcinomas.

Oral Oncol 45:e134–e139

14. Silva SD, Cunha IW, Rangel AL, Jorge J, Zecchin KG, Agostini

M, Kowalski LP, Coletta RD, Graner E (2008) Differential

expression of fatty acid synthase (FAS) and ErbB2 in nonma-

lignant and malignant oral keratinocytes. Virchows Arch

453:57–67

15. Lafky JM, Wilken JA, Baron AT, Maihle NJ (2008) Clinical

implications of the ErbB/epidermal growth factor (EGF) receptor

family and its ligands in ovarian cancer. Biochim Biophys Acta

1785:232–265

16. Syrigos KN, Zalonis A, Kotteas E, Saif MW (2008) Targeted

therapy for oesophageal cancer: an overview. Cancer Metastasis

Rev 27:273–288

17. Wei Q, Sheng L, Shui Y, Hu Q, Nordgren H, Carlsson J (2008)

EGFR, HER2, and HER3 expression in laryngeal primary tumors

and corresponding metastases. Ann Surg Oncol 5:1193–1201

18. Kassouf W, Black PC, Tuziak T, Bondaruk J, Lee S, Brown GA,

Adam L, Wei C, Baggerly K, Bar-Eli M, McConkey D, Czerniak

B, Dinney CP (2008) Distinctive expression pattern of ErbB

family receptors signifies an aggressive variant of bladder cancer.

J Urol 179:353–358

19. Silva SD, Agostini M, Nishimoto IN, Coletta RD, Alves FA,

Lopes MA, Kowalski LP, Graner E (2004) Expression of fatty

acid synthase, ErbB2 and Ki-67 in head and neck squamous cell

carcinoma. A clinicopathological study. Oral Oncol 40:688–696

20. Marcu LG, Yeoh E (2009) A review of risk factors and genetic

alterations in head and neck carcinogenesis and implications for

current and future approaches to treatment. J Cancer Res Clin

Oncol 135:1303–1314

21. Zhang H, Berezov A, Wang Q, Zhang G, Drebin J, Murali R,

Greene MI (2007) ErbB receptors: from oncogenes to targeted

cancer therapies. J Clin Invest 117:2051–2058

22. Holbro T, Civenni G, Hynes NE (2003) The ErbB receptors and

their role in cancer progression. Exp Cell Res 284:99–110

23. O-charoenrat P, Rhys-Evans PH, Modjtahedi H, Eccles SA

(2002) The role of c-erbB receptors and ligands in head and neck

squamous cell carcinoma. Oral Oncol 38:627–640

24. Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell

100:57–70

25. Zeren T, Inan S, Seda Vatansever H, Ekerbicer N, Sayhan S

(2008) Significance of tyrosine kinase activity on malign trans-

formation of ovarian tumors: a comparison between EGF-R and

TGF-alpha. Acta Histochem 110:256–263

26. Pu J, McCaig CD, Cao L, Zhao Z, Segall JE, Zhao M (2007) EGF

receptor signalling is essential for electric-field-directed migra-

tion of breast cancer cells. J Cell Sci 120:3395–3403

27. Klapper LN, Kirschbaum MH, Sela M, Yarden Y (2000) Bio-

chemical and clinical implications of the ErbB/HER signaling

network of growth factor receptors. Adv Cancer Res 77:25–79

28. Penuel E, Schaefer G, Akita RW, Sliwkowski MX (2001)

Structural requirements for ErbB2 transactivation. Semin Oncol

28:36–42

29. Riese DJ 2nd, Stern DF (1998) Specificity within the EGF family/

ErbB receptor family signaling network. BioEssays 20:41–48

30. Olayioye MA, Neve RM, Lane HA, Hynes NE (2000) The ErbB

signaling network: receptor heterodimerization in development

and cancer. EMBO J 19:3159–3167

31. Yarden Y, Sliwkowski MX (2001) Untangling the ErbB signal-

ling network. Nat Rev Mol Cell Biol 2:127–137

32. O’Sullivan B, Shah J (2003) New TNM staging criteria for head

and neck tumors. Semin Surg Oncol 21:30–42

33. Wahi PN, Cohen B, Luthra UK, Torloni H (1971) Histological

typing of oral and oropharyngeal tumours. World Health Orga-

nization, Geneva 28

34. Wolff AC, Hammond ME, Schwartz JN, Hagerty KL, Allred DC,

Cote RJ, Dowsett M, Fitzgibbons PL, Hanna WM, Langer A,

McShane LM, Paik S, Pegram MD, Perez EA, Press MF, Rhodes

A, Sturgeon C, Taube SE, Tubbs R, Vance GH, van de Vijver M,

Wheeler TM, Hayes DF (2007) American Society of Clinical

Oncology/College of American Pathologists. merican Society of

Clinical Oncology/College of American Pathologists guideline

recommendations for human epidermal growth factor receptor 2

testing in breast cancer. Arch Pathol Lab Med 131:18–43

35. Kononen J, Bubendorf L, Kallioniemi A, Barlund M, Schraml P,

Leighton S, Torhorst J, Mihatsch MJ, Sauter G, Kallioniemi OP(1998) Tissue microarrays for high-throughput molecular profil-

ing of tumor specimens. Nat Med 4:844–847

36. Geiger TR, Peeper DS (2009) Metastasis mechanisms. Biochim

Biophys Acta 1796:293–308

37. Alaoui-Jamali MA, Song DJ, Benlimame N, Yen L, Deng X,

Hernandez-Perez M, Wang T (2003) Regulation of multiple tumor

microenvironment markers by overexpression of single or paired

combinations of ErbB receptors. Cancer Res 63:3764–3774

38. Muraoka-Cook RS, Feng SM, Strunk KE, Earp HS 3rd (2008)

ErbB4/HER4: role in mammary gland development,


123

differentiation and growth inhibition. J Mammary Gland Biol

Neoplasia 13:235–246

39. Ortega MC, Bribian A, Peregrın S, Gil MT, Marın O, de Castro F

(2012) Neuregulin-1/ErbB4 signaling controls the migration of

oligodendrocyte precursor cells during development. Exp Neurol

235:610–620

40. Olayioye MA, Graus-Porta D, Beerli RR, Rohrer J, Gay B, Hynes

NE (1998) ErbB-1 and ErbB-2 acquire distinct signaling prop-

erties dependent upon their dimerization partner. Mol Cell Biol

18:5042–5051

41. Clark DE, Williams CC, Duplessis TT, Moring KL, Notwick AR,

Long W, Lane WS, Beuvink I, Hynes NE, Jones FE (2005)

ErbB4/HER4 potentiates STAT5A transcriptional activity by

regulating novel STAT5A serine phosphorylation events. J Biol

Chem 280:24175–24180

42. Salcini AE, McGlade J, Pelicci G, Nicoletti I, Pawson T, Pelicci

PG (1994) Formation of Shc-Grb2 complexes is necessary to

induce neoplastic transformation by overexpression of Shc pro-

teins. Oncogene 9:2827–2836

43. Xu J, Rodriguez D, Petitclerc E, Kim JJ, Hangai M, Moon YS,

Davis GE, Brooks PC (2001) Proteolytic exposure of a cryptic

site within collagen type IV is required for angiogenesis and

tumor growth in vivo. J Cell Biol 154:1069–1079

44. Giannelli G, Falk-Marzillier J, Schiraldi O, Stetler-Stevenson

WG, Quaranta V (1997) Induction of cell migration by matrix

metalloprotease-2 cleavage of laminin-5. Science 277:225–228

45. Ke Z, Lin H, Fan Z, Cai TQ, Kaplan RA, Ma C, Bower KA, Shi

X, Luo J (2006) MMP-2 mediates ethanol-induced invasion of

mammary epithelial cells over-expressing ErbB2. Int J Cancer

119:8–16

46. Kim IY, Yong HY, Kang KW, Moon A (2009) Overexpression of

ErbB2 induces invasion of MCF10A human breast epithelial cells

via MMP-9. Cancer Lett 275:227–233

47. Bao W, Fu HJ, Jia LT, Zhang Y, Li W, Jin BQ, Yao LB, Chen

SY, Yang AG (2010) HER2-mediated upregulation of MMP-1 is

involved in gastric cancer cell invasion. Arch Biochem Biophys

499:49–55

48. Aglund K, Rauvala M, Puistola U, Angstrom T, Turpeenniemi-

Hujanen T, Zackrisson B, Stendahl U (2004) Gelatinases A and B

(MMP-2 and MMP-9) in endometrial cancer-MMP-9 correlates

to the grade and the stage. Gynecol Oncol 94:699–704

49. Dutton JM, Graham SM, Hoffman HT (2002) Metastatic cancer

to the floor of mouth: the lingual lymph nodes. Head Neck

24:401–405

50. Davidson B, Goldberg I, Gotlieb WH, Kopolovic J, Ben-Baruch

G, Nesland JM, Berner A, Bryne M, Reich R (1999) High levels

of MMP-2, MMP-9, MT1-MMP and TIMP-2 mRNA correlate

with poor survival in ovarian carcinoma. Clin Exp Metastasis

17:799–808

51. Patel BP, Shah SV, Shukla SN, Shah PM, Patel PS (2007)

Clinical significance of MMP-2 and MMP-9 in patients with oral

cancer. Head Neck 29:564–572

52. Yorioka CW, Coletta RD, Alves F, Nishimoto IN, Kowalski LP,

Graner E (2002) Matrix metalloproteinase-2 and -9 activities

correlate with the disease-free survival of oral squamous cell

carcinoma patients. Int J Oncol 20:189–194

53. Kohrmann A, Kammerer U, Kapp M, Dietl J, Anacker J (2009)

Expression of matrix metalloproteinases (MMPs) in primary

human breast cancer and breast cancer cell lines: new findings

and review of the literature. BMC Cancer 9:188


123

Cooverexpression of ERBB1 and ERBB4 receptors predicts poor clinical outcome in pN+ oral squamous...

Documents

Transcript of Cooverexpression of ERBB1 and ERBB4 receptors predicts poor clinical outcome in pN+ oral squamous...