вход по аккаунту



код для вставкиСкачать
Expression of Interleukin-6, Interleukin-6 Receptor,
and Glycoprotein 130 Correlates with Good Prognoses
for Patients with Breast Carcinoma
Aldona Karczewska, M.D., Ph.D.1
Sergiusz Nawrocki, M.D.1
Danuta Brȩborowicz, M.D., Ph.D.1
Violetta Filas, Ph.D.2
Andrzej Mackiewicz, M.D., Ph.D.1
Department of Cancer Immunology, Chair of Oncology, University School of Medical Sciences at
Great Poland Cancer Center, Poznań, Poland.
Department of Pathology, Chair of Oncology, University School of Medical Sciences, Poznań, Poland.
BACKGROUND. Interleukin-6 (IL-6) is secreted by normal epithelial breast cells but
not by oncogene-transformed cells. Interleukin-6 is able to inhibit growth of breast
carcinoma cells in culture. Interleukin-6 exerts its activity via two receptor subunits, IL-6R and glycoprotein 130 (gp130). The expression of these receptor subunits in breast tumors has been studied, but there are no previous reports of their
prognostic significance, to the authors’ knowledge.
METHODS. mRNA of IL-6, IL-6R, and gp130 was studied in 75 tumor samples
obtained from breast carcinoma patients. Patients were followed for a maximum of
71 months (median follow-up, 61 months; 60 patients were followed for a minimum of 5 years or died during the observation period). Prognostic factors were
analyzed in univariate and multivariate analysis.
RESULTS. mRNA specific to IL-6, IL-6R, and gp130 was detected in 57%, 53%, and
71% of breast carcinoma tissues, respectively. Expression was strongly correlated
with earlier stages of the disease. In univariate analysis, expression of IL-6 and its
receptor subunits proved to be a positive prognostic factor for overall survival (OS)
and disease free survival (DFS). IL-6R and gp130 expression were good independent prognostic factors for OS. The 5-year OS of all patients was 66%. The 5-year
OS in IL-6, IL-6R, and gp130 positive groups was 95%, 94%, and 90%, respectively,
whereas in negative groups it was 26%, 31%, and 9%, respectively.
CONCLUSIONS. Expression of IL-6, IL-6R, and gp130 in breast carcinoma tissue is
associated with earlier stages of the disease. In advanced stages, expression of IL-6
and its receptor subunits predicts better prognosis. Cancer 2000;88:2061–71.
© 2000 American Cancer Society.
KEYWORDS: breast carcinoma, interleukin-6, interleukin-6 receptor, prognostic
factors, prognosis.
Supported by the State Committee for Scientific
Research (Warsaw) and Foundation for Polish Science (Warsaw).
Address for reprints: Andrzej Mackiewicz, M.D.,
Ph.D., Department of Cancer Immunology, Chair of
Oncology, University School of Medical Sciences at
Great Poland Cancer Center, ul. Garbary 15, 61866 Poznań, Poland.
Received March 9, 1999; revisions received October 4, 1999 and January 14, 2000; accepted
January 14, 2000.
© 2000 American Cancer Society
he role of cytokines and growth factors in tumor development and
progression has been the focus of a lot of attention recently. A
variety of factors secreted by normal cells in the tumor microenvironment or by tumor cells themselves may stimulate or inhibit tumor
growth. Cytokines and growth factors can contribute directly to the
proliferation of cancer cells and tumor progression in a number of
ways. These include constitutive autocrine production of stimulatory
factors by tumors, induction of stimulatory factors production, increased sensitivity to autocrine or paracrine factors, or acquisition of
resistance to cytokines and growth factors.1–3 In addition, cytokines
may stimulate or inhibit specific immune responses.4 These factors
have been reported to play a significant role in breast carcinoma
progression.5 Moreover, some growth factors and cytokines have
CANCER May 1, 2000 / Volume 88 / Number 9
been proven to be useful as prognostic indicators in
the course of the disease.
Interleukin-6 (IL-6) is currently a subject of intensive studies as a promoting or inhibitory factor in
various types of tumors. Interleukin-6 is a multifunctional cytokine produced by nonmalignant cells such
as T lymphocytes, fibroblasts, or monocytes and a
number of malignant cells, e.g., melanoma or renal
cell carcinoma.6,7 It activates the development of
megakariocytes and platelets formation, proliferation
of multipotent stem cells, activates T, B, and natural
killer cells and regulates expression of acute phase
proteins in the liver. Interleukin-6 exerts its activity
through the membrane-bound receptor complex consisting of IL-6 binding protein (IL-6R, ␣-subunit and
glycoprotein 130 (gp130, ␤-subunit that functions as a
signal transducer. Interleukin-6 first binds to IL-6R
and then the complex attracts gp130 molecules that
dimerize leading to a signal transduction.8 Both receptor subunits also exist in soluble forms, lacking transmembrane and cytoplasmic domains. Soluble IL-6R
(sIL-6R) is agonistic to IL-6, enhancing its effect on
cells possessing both ␣- and ␤-subunits. The IL-6/sIL-6R
complex also is able to activate cells lacking membrane
bound IL-6R but possessing gp130, thus significantly
broadening the biologic activity of IL-6.9,10
Interleukin-6 has been shown to inhibit proliferation, increase motility, and prolong the interval between mitosis and readherence of daughter cells to
their neighbors and the substratum of breast carcinoma cells.11–13 Interleukin-6 was found to be secreted
by normal breast epithelial cells, whereas breast carcinoma cells demonstrated impaired IL-6 expression.14 In addition, when normal breast epithelial cells
and some breast carcinoma cell lines secreting IL-6
were transformed with oncogenes, they lost the ability
to produce IL-6.15 Similarly, when some breast carcinoma cell lines responding to IL-6 were highly transformed with oncogenes, as above, they lost the IL-6
receptor, and thus IL-6 responsiveness.16 Recent studies have demonstrated expression of IL-6 receptor ␣and ␤-subunits in breast carcinoma tissues.12,17 However, there are no reports on the expression of IL-6 and
its receptor subunits in breast carcinoma microenvironment at various clinical stages of the disease or the
relation of these factors to the disease progression and
clinical course of the disease.
Therefore, we decided to 1) assess the mRNA expression of IL-6 and its receptor subunits in breast
carcinoma tissue (microenvironment) and in primary
breast carcinoma cells (in primary culture) obtained
from patients at various clinical stages of the disease;
and 2) evaluate the clinical usefulness of the factors
studied as prognostic indicators of breast carcinoma
progression to predict a poor prognosis or better survival.
Seventy-five women aged 36 –74 years (median age, 61
years) with primary breast carcinoma without measurable distant metastases at diagnosis were treated surgically between 1992 and 1994 at the Great Poland
Cancer Center (Poznań, Poland), and followed up for a
maximum of 71 months (median follow-up, 61
months; 60 patients were followed up for a minimum
of 5 years or died during the observation period). The
patients were classified into staging groups according
to the American Joint Committee on Cancer TNM
staging system (AJCC-1989).18 The patients underwent
primary surgical therapy consisting of modified radical mastectomy, breast-conserving surgery with axillary lymph nodes dissection, or toilet mastectomy after primary chemotherapy in the Stage III group. The
patients treated with conservative surgery had adjuvant radiation therapy (50 gray [Gy] with cobalt-60 or
6 MeV photons plus a 10 –15-Gy electron boost in
cases with positive surgical margins on microscopic histopathologic examination). All patients with metastases
in axillary lymph nodes (N⫹) had adjuvant chemotherapy (cyclophosphamide, methotrexate, 5-fluorouracil or
5-fluorouracil, doxorubicin, cyclophosphamide) or adjuvant hormonotherapy. The adjuvant chemotherapy was
given to 25 patients. The adjuvant hormonal treatment
was given to 61 patients (in 13 patients, the hormonal
treatment followed the chemotherapy). The adjuvant
radiotherapy after mastectomy was applied in all patients with T3/T4 tumors and/or patients with metastases in more than three lymph nodes. Twenty-five
patients died during the observation. We observed
recurrences in 6 patients and metastases in 28 patients. In five patients with local recurrences, metastases were noticed later whereas in one patient metastases were noticed at the same time as metastases.
The surgically excised primary breast carcinoma tissue
was divided in two or three parts. One sample was
used for histopathologic examination, another was
frozen in liquid nitrogen for RNA isolation, and the
third one was used for establishing primary cultures of
breast carcinoma. Informed consent was obtained
from subjects.
The histopathologic diagnosis was reexamined in
1999 and assessed according to the Elston scoring
system. All 75 tumors were retyped and regraded; nuclear grade and mitotic index were derived from Elston scores. Reexamination revealed 65 cases of infiltrating ductal or lobular cancer (or mixed types), and
10 other types: 2 medullar, 2 tubular, 2 metaplastic, 2
Prognostic Value of IL-6 and Receptors/Karczewska et al.
gelatinous, 1 papillary, and 1 mucinous carcinoma.
Tumors were graded into three general categories
G1–G3 (G1: well differentiated, overall Elston score
1–5; G2: moderately differentiated, overall Elston score
6 –7 and G3: poorly differentiated, overall Elston score
8 –9). Because estrogen and progesterone receptor status has been assessed in our center since 1995, and
routinely since 1998, estrogen receptor (ER) status was
determined in 70 tumors and progesterone receptor
(PR) status in 63 tumors in stored paraffin embedded
specimens in 1999. Immunohistochemical demonstration of ER and PR in paraffin sections was performed using the immunoperoxidase staining procedure with monoclonal mouse anti-human estrogen
receptor antibodies (Dako A/S, Glostrup, Denmark)
and monoclonal mouse anti-human progesterone receptor antibodies (Novocastra Laboratories Ltd., Newcastle upon Tyne, UK). Carcinomas with no or weak
staining (⬍ 10% positive cells) were considered as
receptor negative (ER⫺ or PR⫺).
Frozen tissue was homogenized in the presence of
liquid nitrogen, and total cellular RNA was isolated as
described by Chomczynski and Sacchi.19
Primary Cultures of Breast Carcinoma Cells
Primary cultures were derived from primary breast
carcinoma tissue specimens. Cultures were derived
from 17 ductal carcinomas, 1 lobular carcinoma, 1
medullar carcinoma, and 1 papillary carcinoma. Eight
of the 20 patients died of the disease within 5 years.
Tissue specimens supplied in sterile conditions were
minced into 2–3-mm pieces and digested with collagenase (0.14%) and DNAase Type I (0.1%). The isolated cells were cultured as monolayers in a DMEM/
F12 (1:1) medium supplemented with 5% fetal calf
serum, antibiotics (gentamycine), transferin (25 ␮g/
mL), insulin (3 ␮g/mL), 17␤-estradiol (10 nM), dexamethasone (10 nM), triiodothyronin (10 nM), prostaglandin F2a (100 ng/mL), fibronectin (100 ng/mL),
progesterone (10 nM), and prolactine (10 ng/mL). The
medium used was specially selected to support the
growth of breast carcinoma cells and suppress the
growth of fibroblasts.20 –22 To verify for possible contamination of the breast carcinoma cells with other
cell types in culture, we immunostained cytokeratins
with monoclonal antibodies (MNF-116; Dako A/S).
The success rate of establishing primary cultures was
approximately 50%. The cells were grown until confluency, and total RNA was extracted as above.19
Polymerase Chain Reaction Primers
All primer sets were synthesized by Eurogentec (Brussels, Belgium). The sequence of IL-6 and GAPDH
primers was adopted from published articles.23,24 Am-
plification of reverse transcripts gave products of 630
(IL-6) and 368 (GAPDH) base pairs (bp), respectively.
Other primers were designed using the Oligo Version
5.0 program. The IL-6R primers were: upstream, 5⬘
TGC TGA CCA GTC TGC CAG GAG A 3⬘; downstream,
5⬘ ACA CTA CTG GCG ACG CAC ATG 3⬘; product 485
bp. The gp130 primers were: upstream, 5⬘ TGT TGA
CGT TGC AGA CTT GG 3⬘; downstream, 5⬘ TCT GGA
GGC AAG CCT GAA AT 3⬘; product 380 bp.
Full-length cDNAs for the human IL-6 and its receptor
subunits were used as positive controls for polymerase
chain reaction (PCR). cDNAs encoding IL-6, IL-6R,
gp130, and GADPH were obtained from Dr. S. RoseJohn (I. Med. Klinik, J. Gutenberg Universität, Mainz,
Reverse Transcription–PCR
Total RNA (5 ␮g) extracted from tissue and cultured
cells was used for reverse transcription (RT) into cDNA
with murine MMLV reverse transcriptase (Promega,
Madison, WI) and oligo(dT) primers (Promega) according to manufacturer instructions. The quality of
total RNA after extraction was checked on the 2%
agarose gel, and the purity was verified using spectroscopy.
The cDNA obtained from RT was amplified by 35
cycles PCR (Perkin-Elmer, Norwalk, CT) with Taq DNA
polymerase (Biometra GmbH, Goettingen, Germany)
by using specific glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers to control the effects of
RT reaction. GADPH positive cDNAs then were amplified using IL-6 or its receptor specific primers (listed
above). The 10-␮L PCR reaction mixture contained 1
␮L of RT samples, 1 ␮L 10-fold PCR buffer (10 mM
Tris/HCl, pH 8.8, 1.5 mM MgCl2, 50 mM KCl, 0.1
Triton X-100), 0.4 ␮L of 200 ␮M dNTP mixture, 0.2 ␮L
of 1 ␮M 3⬘ and 5⬘ primers, and 0.5 ␮L of Taq DNA
polymerase. Each sample was amplified for a specified
number of cycles with the following program: denaturation (20 s at 94 °C), annealing (30 s at 55 °C for
GAPDH and gp130, 30 s at 61 °C for IL-6R, and 30 s at
65 °C for IL-6) and extension (20 – 45 s at 72 °C). Plasmids listed above were used as positive controls. As a
template to check for DNA contamination, PCR using
isolated total mRNA (before RT) was performed, revealing no amplification. Each sample was analyzed
three times.
Polymerase chain reaction products were electrophoresed in the presence of ethidium bromide in 1.5%
agarose gels. After the electrophoresis, the gels were
photographed using ultraviolet transilumination. The
size of corresponding PCR products was estimated
CANCER May 1, 2000 / Volume 88 / Number 9
Expression of IL-6, IL-6R, and gp130 Related to Clinical Stages
Stage (no. of patients)
IL-6 (ⴙ) (%)
IL-6 (ⴚ) (%)
IL-6R (ⴙ) (%)
IL-6R (ⴚ) (%)
gp130 (ⴙ) (%)
gp130 (ⴚ) (%)
I (4)
IIA (23)
IIB (28)
IIIA (10)
IIIB (10)
3 (75)
22 (95)
15 (54)
2 (20)
2 (20)
1 (25)
1 (5)
13 (46)
8 (80)
8 (80)
3 (75)
20 (87)
13 (46)
3 (30)
2 (20)
1 (25)
3 (13)
15 (54)
7 (70)
8 (80)
4 (100)
23 (100)
22 (79)
2 (20)
2 (20)
0 (0)
0 (0)
6 (21)
8 (80)
8 (80)
IL-6: interleukin-6; IL-6R: interleukin-6 receptor; gp130; glycoprotein 130; (⫹): expression of specific mRNA; (⫺): no expression of specific mRNA.
using Lambda DNA EcoR1 and Hind III Marker (Promega). Moreover, the specificity of the amplified
cDNA sequences was confirmed by restriction mapping.
Statistical Analyses
The relation between the presence of IL-6, gp130, and
IL-6R mRNAs in tumor samples and established breast
carcinoma prognostic factors (lymph node status, tumor grade, nuclear grade, mitotic index, histopathologic type, tumor size, ER , PR, and menopausal status)
was tested using the chi-square test.
Survival analysis was performed using the
Kaplan–Meier method, the long rank test, and the Cox
proportional hazards regression model. The patterns
of overall survival and disease free survival (recurrence
and metastases free survival) were estimated by
means of the product-limit method (Kaplan–Meier).
The role of each prognostic variable (IL-6, gp130, IL6R, lymph node status, tumor grade, nuclear grade,
mitotic index, histopathologic type, tumor size, ER,
PR, menopausal status, adjuvant hormonal treatment,
adjuvant chemotherapy) and their effect on overall
survival and disease free survival was tested in the
univariate analysis by means of the log rank test. Their
joint effect on overall survival was tested in multivariate analysis using the Cox regression model. All analyses were performed using the Statistica 5.5 software
package (Statsoft Inc., Tulsa, OK).
Expression of IL-6, IL-6R, and gp130 Transcripts in
Breast Carcinoma Tissue
A specific mRNA for IL-6 was detected in 44 of 75
(57%) breast carcinoma tissues (Table 1). Significant
difference (chi-square ⫽ 20.02; P ⬍ 0.0001) was observed in the expression of IL-6 mRNA between earlier
stages (I and IIA) and more advanced stages (IIB, IIIA,
and IIIB). In Stages I and IIA, IL-6 mRNA expression
was observed in 93% of cases, whereas in Stages IIB
and IIIA and IIIB (all three stages combined) in 39% of
cases. However, with two exceptions (8%), none of the
patients who died of the disease (Stages IIB, IIIA, and
IIIB) showed IL-6 mRNA expression.
A specific gp130 mRNA was detected in 53 of 75
(71%) breast carcinoma tissues (Table 1), whereas
IL-6R mRNA was found in 40 of 75 specimens (53%).
Similarly, as in the case of cytokine mRNA expression,
significant differences in the expression of receptor
subunits between stages of the disease were shown.
Gp130 mRNA was found in all Stage I and IIA cases
(100%), in 22 of 28 Stage IIB patients (79%), in 2 of 10
Stage IIIA patients (20%), and in 2 of 10 Stage IIIB
patients (20%) (P ⬍ 000.1). Samples of only 5 of the 25
patients (20%) who died of the disease demonstrated
the gp130 mRNA expression. Interleukin-6R mRNA
was expressed in 23 of 27 (85%) specimens from patients with Stages I and IIA and in 18 of 48 (37%)
samples obtained from patients with Stages IIB, IIIA,
and IIIB (chi-square ⫽ 15.9; P ⫽ 0.0001). Only 2 of the
25 patients (8%) who died of the disease showed IL-6R
mRNA expression in breast carcinoma tissues.
Expression of IL-6, gp130, and IL-6R Transcripts in
Primary Cultures of Breast Carcinoma
The mRNA expression of IL-6 and its receptor subunits in cultured cells and corresponding tissues is
shown in Table 2. Interleukin-6 mRNA was observed
in the cells of four cultures derived from nine IL-6
positive breast carcinoma tissues. None of the cultures
derived from tissues obtained from the patients who
died within 5 years showed detectable levels of IL-6
IL-6R mRNA was found in 5 cultures derived from
9 IL-6R positive breast carcinoma tissues, whereas
gp130 was found in 8 of 16. Neither IL-6R nor gp130
was found in cultures derived from tissues obtained
from the patients who died within 5 years.
Interleukin-6, IL-6R, and gp130 as Prognostic Factors in
Survival Analysis
There was a strong correlation between the expression
of the individual transcripts (IL-6 vs. IL-6R: chisquare ⫽ 24.51, P ⬍ 0.0001; IL-6 vs. gp130: chi-
Prognostic Value of IL-6 and Receptors/Karczewska et al.
Expression of Cytokines and Their Receptor Subunits in Cultures Derived from Tumor Tissue and Clinical Outcome
Culture no.
IL-6: interleukin-6; IL-6R: interleukin-6 receptor; gp130: glycoprotein 130; ⫹: expression of cytokines.
square ⫽ 31.55, P ⬍ 0.001; IL-6R vs. gp130: chisquare ⫽ 29.77, P ⬍ 0.001). The expression of IL-6,
gp130, and IL-6R was negatively correlated with
lymph node status, e.g., N(⫹) with IL-6 negative tumors and N(⫺) with IL-6 positive tumors (chisquare ⫽ 20.24, P ⬍ 0.0001). The expression of IL-6,
gp130, and IL-6R correlated with smaller tumor size
(e.g., IL-6 and T2 vs. T3⫹T4: chi-square ⫽ 10.73, P ⫽
0.0011). Expression of IL-6, gp130, and IL-6R did not
correlate with overall grade, nuclear grade, or mitotic
index (e.g., IL-6 and nuclear grade: chi-square ⫽ 0.16,
P ⫽ 0.68; IL-6 and mitotic index: chi-square ⫽ 1.26,
P ⫽ 0.26). Expression of IL-6, gp130, and IL-6R did not
correlate with menopausal status (e.g., IL-6 and
menopausal status: chi-square ⫽ 1.34, P ⫽ 0.25), estrogen receptor status (e.g., IL-6 and estrogen receptor
status: P ⫽ 0.99), and progesterone status (e.g., IL-6
and progesterone receptor status: chi-square ⫽ 0.84,
P ⫽ 0.36).
In the survival analysis, overall survival (OS) was
defined as observation time in days from the surgery
(or the first course of chemotherapy in Stage III patients) to death of the disease or to the last follow-up.
Disease free survival (DFS) was defined as the time of
observation from the beginning of the treatment as
defined above to the diagnosis of local recurrence or
metastasis. There were 15 censored observations in
terms of analysis of the 5-year OS and 13 censored
observations in the DFS analysis. The following variables (prognostic factors) were included in the analysis: histopathologic type (infiltrating ductal/lobular vs.
other types), lymph node status (N⫹, N⫺), tumor
stage (T1, T2, T3, T4), overall histopathologic grading
(G1, G2, G3), nuclear grade (nG1, nG2, nG3), mitotic
index (mG1, mG2, mG3), menopausal status (pre-,
post-), estrogen receptor status (ER⫹, ER⫺), progesterone receptor status (PR⫹, PR⫺), IL-6 status (⫹, ⫺),
gp130 status (⫹, ⫺), and IL-6R status (⫹, ⫺). In addition, the prognostic impact of adjuvant hormonal
treatment and adjuvant chemotherapy was tested.
The patients characteristics in terms of T, N, G, nG,
mG, ER, and PR category and menopausal status related to IL-6, gp130, and IL-6R expression are summarized in the Table 3.
Expression of IL-6, gp130, and IL-6R in the tumor
microenvironment was found to correlate with a favorable prognosis. The survival curves showing cumulative OS and DFS, related to the expression of IL-6,
gp130, and IL-6R are shown in Figure 1. The 5-year
survival rate (Kaplan–Meier analysis) for all patients
was 66%, 95% for the IL-6 positive group, and 26% for
the IL-6 negative group. The 5-year disease free survival rate (Kaplan–Meier analysis) was 62%, 93%, and
16%, respectively. In the IL-6R positive group, 5-year
survival rate was 94%, whereas in the IL-6R negative
group it was 31%, and 5-year DFS rate was 94% and
CANCER May 1, 2000 / Volume 88 / Number 9
Patient Characteristics Related to IL-6, gp130, and IL-6R Expression in the Primary Tumor
IL-6 status
Prognostic variable (75)
Tumor size (%)
T1 (4)
T2 (53)
T3 (8)
T4 (10)
Lymph node status (%)
N (⫹) (45)
N (⫺) (30)
Grade (%)
G1 (12)
G2 (25)
G3 (38)
Nuclear grade (%)
nG1 (6)
nG2 (35)
nG3 (34)
Mitotic index (%)
mG1 (20)
mG2 (24)
mG3 (31)
Menopausal status (%)
Premenopausal (19)
Postmenopausal (56)
ER status (%)
ER (⫹) (29)
ER (⫺) (41)
PR status (%)
PR (⫹) (31)
PR (⫺) (32)
gp130 status
IL-6R status
(ⴙ) (n ⴝ 44)
(ⴚ) (n ⴝ 31)
(ⴙ) (n ⴝ 53)
(ⴚ) (n ⴝ 22)
(ⴙ) (n ⴝ 40)
(ⴚ) (n ⴝ 35)
3 (75)
37 (70)
2 (25)
2 (20)
1 (25)
16 (30)
6 (75)
8 (80)
4 (100)
45 (85)
2 (25)
2 (20)
0 (0)
8 (15)
6 (75)
8 (80)
3 (75)
33 (62)
2 (25)
2 (20)
1 (25)
20 (38)
6 (75)
8 (80)
17 (38)
27 (90)
28 (62)
3 (10)
23 (51)
30 (100)
22 (49)
0 (0)
16 (36)
24 (80)
29 (64)
6 (20)
9 (75)
15 (60)
20 (53)
3 (25)
10 (40)
18 (47)
9 (75)
19 (76)
25 (66)
3 (25)
6 (24)
13 (34)
8 (67)
15 (60)
17 (45)
4 (33)
10 (40)
21 (55)
3 (50)
22 (63)
19 (56)
3 (50)
13 (37)
15 (44)
3 (50)
29 (83)
21 (62)
3 (50)
6 (17)
13 (38)
2 (33)
21 (60)
16 (47)
4 (67)
14 (40)
18 (53)
14 (70)
14 (58)
17 (55)
6 (30)
10 (42)
14 (45)
17 (85)
15 (63)
21 (68)
3 (15)
9 (37)
10 (32)
13 (65)
14 (58)
11 (35)
7 (35)
10 (42)
20 (65)
9 (47)
35 (63)
10 (53)
21 (37)
12 (63)
41 (73)
7 (37)
15 (27)
10 (53)
30 (54)
9 (47)
26 (46)
17 (59)
24 (59)
12 (41)
17 (41)
22 (76)
27 (66)
7 (24)
14 (34)
15 (52)
21 (51)
14 (48)
20 (49)
20 (65)
17 (53)
11 (35)
15 (47)
24 (77)
18 (56)
7 (23)
14 (44)
16 (52)
14 (44)
15 (48)
18 (56)
IL-6: interleukin-6; gp130: glycoprotein 130; IL-6R: interleukin-6 receptor; n: number of patients; (⫹): expression of specific mRNA; (⫺): no expression of specific mRNA; N(⫹): lymph node positive; N(⫺): lymph node
negative; ER: estrogen receptor; PR: progesterone receptor.
22%, respectively. In the gp130 positive group, 5-year
survival rate was 90%, whereas in the gp130 negative
group it was 9%, and 5-year DFS rate was 88% and 0%,
In the univariate analysis significant differences (log
rank test, P ⬍ 0.05) in OS and DFS were related to the
lymph node status (N[⫹] vs. N[⫺], P ⬍ 0.0001), histopathologic type of the tumor (infiltrating ductal/lobular carcinoma vs. other types, P ⫽ 0.002), tumor stage (T2
vs. T3⫹T4, P ⫽ 0.01; T1 patients were not included, and
T3 were grouped with T4 because there were few cases),
hormonal adjuvant treatment (P ⫽ 0.03), the IL-6 status
(IL-6[⫹] vs. IL-6[⫺], P ⬍ 0.0001), the IL-6R status (IL6R[⫹] vs. IL-6R[⫺], P ⬍ 0.0001), and the gp130 status
(gp130[⫹] vs. gp130[⫺], P ⬍ 0.0001). No significant differences were related to the grade of the tumor (G1 vs.
G2⫹G3, P ⫽ 0.08; G1⫹G2 vs. G3 P ⫽ 0.34), nuclear grade
(P ⫽ 0.77), mitotic index (P ⫽ 0.61), ER status (P ⫽ 0.33),
PR status (P ⫽ 0.37), or menopausal status (P ⫽ 0.4). No
significant differences were related to adjuvant chemotherapy (P ⫽ 0.15)
The mRNA expression of IL-6, IL-6R, and gp130,
respectively, was found to be a significant positive prognostic factor in the univariate stratified analysis for subgroups of patients, e.g., IL-6 in the N(⫹) subgroup (P
⫽ 0.00015), IL-6 in the N(⫹)T2, subgroup (P ⫽ 0.008)
(Fig. 2). Patients who had specific IL-6 mRNA in the
tumor microenvironment had better OS and DFS rates.
In addition, if “better” and rare histopathologic types
were excluded and survival analysis was performed on
the group of infiltrating ductal/lobular carcinomas (n
⫽ 65), the IL-6, IL-6R, and gp130 mRNA expression also
were found to be significant positive prognostic factors
(e.g., IL-6 in this subgroup, P ⫽ 0.00012)
In the multivariate analysis, the histopathologic
type, lymph node status, tumor stage, adjuvant hormonal treatment, and the IL-6, IL-6R, and gp130 status were tested as independent prognostic factors for
Prognostic Value of IL-6 and Receptors/Karczewska et al.
FIGURE 1. Overall survival and disease free survival of breast carcinoma patients (n ⫽ 75) in relation to detection of IL-6, IL-6R, and gp130 mRNAs in tumor
tissues are shown. Log rank analysis was done. P ⬍ 0.001 in all graphs. (⫺): cases without expression; (⫹) cases with expression; n: number of cases in each
group; OS: overall survival; IL-6: interleukin-6; gp130: glycoprotein 130; DFS: disease free survival; IL-6R: interleukin-6 receptor.
OS in Cox proportional hazards regression model. In
this model, each regression coefficient beta is the loge
of a relative hazard ratio. Tumor size and adjuvant
hormonal treatment did not prove to be independent
prognostic factors in this group of patients. Lymph
node involvement increased the relative hazard by a
factor of approximately 1.4; histopathology of infilrating ductal/lobular carcinoma increased the hazard by
1.6, whereas the presence of mRNA of IL-6 or IL-6R or
gp130 decreased the relative hazard by a factor of 0.84,
0.26, and 0.14, respectively. Only IL-6R and gp130
have had statistically significant effect on OS considering the 95% confidence interval (Table 4).
The clinical course of breast carcinoma is very variable. The most commonly used prognostic factors are
conventional clinical and histopathologic tumor char-
CANCER May 1, 2000 / Volume 88 / Number 9
FIGURE 2. Overall survival of breast carcinoma patients in relation to detection of IL-6 mRNA in tumor tissues is shown. Subgroup analysis: subgroup analysis
in N[⫹] and N[⫺] patients and in N[⫹]T2 patients; OS: overall survival; N(⫹): patients with metastases in lymph nodes; N(⫺): patients without metastases in lymph
nodes; T2: patients with T2 tumors; (⫺): cases without expression; (⫹): cases with expression; n: number of cases in each subgroup; IL-6: interleukin-6. Log rank
analysis was done. P ⬍ 0.001.
Multivariate Analysis of Prognostic Factors in Cox Regression Model
Prognostic factor
Standard error
Hazard ratio (exponent
beta) (95% CI)
N ⫹/⫺
IL-6 ⫹/⫺
IL-6R ⫹/⫺
gp130 ⫹/⫺
1.36 (0.4–2.32)
0.84 (0.14–1.54)
0.26 (0.05–0.47)
0.14 (0.00–0.30)
1.56 (0.00–3.15)
CI: confidence interval; N ⫹/⫺: lymph node status; IL-6: interleukin-6; IL-6R: interleukin-6 receptor;
gp130: glycoprotein 130; HISPAT: histopathology.
Infiltrating ductal and/or lobular ca vs. other types.
acteristics such as tumor size, lymph node involvement, grade of the tumor, and estrogen and progesteron status of the tumor. These characteristics are
used widely for both prognosis and the choice of treatment option. However, in groups of patients homogenous in terms of the above prognostic factors, we
have found patients with dramatically different outcomes. Advances in molecular biology allow us to
study different biologic and biochemical characteristics of the tumor microenvironment that might be
responsible for the clinical course of the disease. We
have found that expression of IL-6, IL-6R, and gp130
in the tumor microenvironment is a strong indicator
of a favorable clinical course of the disease. Transcripts of IL-6 were found in 57% of breast carcinoma
tissues (microenvironment) studied; however, significant differences were shown between various clinical
stages of the disease (Table 1). Interleukin-6 mRNA
was detected in a majority of breast carcinoma tissues
obtained from patients in the early stages of the disease, whereas it was detected only in a fraction of
tissues excised from patients in stages more advanced
in terms of tumor size and lymph node involvment.
Contradictory results of IL-6 mRNA expression in
breast carcinoma tissue have been reported in literature. Marrogi et al. were not able to show IL-6 mRNA
in any of the 19 cases studied, whereas some of the
patients were at early stages of the disease.25 In contrast, Crichton et al. found IL-6 transcripts in all the six
breast carcinoma specimens that they studied.17
Green et al. reported IL-6 in 58% of 77 cases studied,
but its expression did not correlate with some clinical
parameters such as tumor size, state of differentiation,
lymph node metastases, or menopausal status.26
However, these authors did not correlate IL-6 expression with the clinical course of the disease. Similarly,
Venetsanakos et al. observed IL-6 transcripts in approximately 50% of the breast carcinoma cases studied, but they did not correlate its expression with the
clinical stage.27
RNA transcripts of IL-6R were detected in the
microenvironment of approximately 50% of the breast
carcinoma specimens studied. Similarly as in the case
of IL-6 in the early stages of the disease, IL-6R was
expressed significantly more frequently than in later
stages. Glycoprotein 130 mRNA was found in more
than two-thirds of the cases studied. In the early
stages of the disease, gp130 was expressed in all the
cases studied, whereas in the later stages it progressively declined down to 20% in Stage IIIA or IIIB. There
has been only one report as far as we know so far
demonstrating expression of IL-6R and gp130 mRNAs
in breast carcinoma tissue.12 In this study, IL-6R was
detected in 80% and gp130 in 96% of the tissues. The
authors did not present the clinical status of the pa-
Prognostic Value of IL-6 and Receptors/Karczewska et al.
tients studied, but the pattern of IL-6R and gp130
expression resembled that which we observed in the
early stages of the disease.
Coexpression of both IL-6 receptor subunits was
observed in 50% of the tissues studied, mostly in specimens resected in the early stages of the disease. This
finding indicates that cells present in the tumor microenvironment only in the half of breast carcinoma
tissues might be potentially sensitive to IL-6.
Breast carcinoma tissue may be infiltrated by immune cells. Also other cells such as fibroblasts might
be present in the tumor microenvironment. These
cells, when activated, are potent producers of IL-6.28
Moreover, they may possess IL-6R and gp130 and thus
be the source of the mRNA of these factors in the
tissues studied. Conversely, Basolo et al. reported (using hybridization in situ) that breast carcinoma cells
but not infiltrating immune cells expressed mRNA of
several studied cytokines.14 Because we did not perform in situ hybridization and immunohistochemical
techniques are not sensitive enough to detect IL-6
receptors (especially IL-6R), to elucidate if the breast
carcinoma cells are expressing mRNA of IL-6 and its
rececptor subunits, we decided to derive primary cell
cultures from tumor tissues. In fact, only approximately 50% of the cultures derived from the tissues
that were positive for IL-6, IL-6R, or gp130 mRNAs
demonstrated the expression of corresponding factors. However, coexpression of IL-6R and gp130 mRNAs (mostly at the early stages of the disease) was
observed in only 20% of the cultures, indicating that
these cells might express the functional IL-6 receptor
complex. The last suggestion is drawn on the basis of
the report that demonstrated correlation between the
expression of mRNA of IL-6 receptor subunits and the
responsiveness of breast carcinoma cells to IL-6.11 Primary cultures of solid tumor cells has its limitations,
because certain types of clones are lost during the
culture and some cell clones may overgrow the others.
This may result in a loss of cells producing certain
factors. Thus, the number of breast carcinoma cells
producing IL-6 or expressing IL-6R or gp130 in vivo
could be even higher.
Statistical analysis demonstrated strong correlation between the mRNA expression of all the factors
studied in the breast carcinoma microenvironment.
This fact and earlier reports demonstrating loss of IL-6
and IL-6R expression in oncogene-transformed epithelial and breast carcinoma cells suggest that IL-6,
IL-6R, and gp130 might be indicators of molecular
progression of breast carcinoma or even play a role in
these processes in either the autocrine or the paracrine manner. Thus, they may reflect or lead to acquisition of invasiveness and metastatic potential by
breast carcinoma cells in vivo. Conversely, the IL-6
and/or IL-6/sIL-6R complex present in the tumor microenviroment (even if secreted by nontransformed
cells) may stimulate immune response to cancer cells
or have a direct antiproliferative effect.11 The antiproliferative mechanism of IL-6 or the IL-6/sIL-6R complex may be associated with regulation of cell cycle by
promoting the expression of p21 and p27 as it was
demonstrated in melanoma.29 Interleukin may inhibit
breast carcinoma cells growth synergistically with
other factors such as IL-1. Danforth and Sgagias have
shown that IL-1 and IL-6 can act additively to inhibit
growth in the absence or presence of estradiol.30 The
role of IL-6 and its receptors in modulating immune
response to breast carcinoma cells is not fully understood. Interleukin-6 was reported to induce responses
against tumor cells and decrease tumorogenicity and
metastatic potential of various tumors. Delivery of
IL-6 into tumor microenironment by using IL-6 gene
transfer resulted in inhibition of tumor growth and
elucidation of specific immune responses.31–33
We correlated the expression of the factors studied with parameters such as tumor size, lymph node
status, histopathologic type and grade, nuclear grade,
mitotic index, ER status, PR status, or menopausal
status. To evaluate the clinical usefulness of IL-6, IL6R, and gp130 as prognostic factors in breast carcinoma, we followed up our patients for 5 years and
analyzed their OS and DFS. We found that the expression of IL-6 and its receptors in breast carcinoma
microenvironment strongly correlates with longer DFS
and OS and is a good prognostic factor. In our study,
patients demonstrating IL-6, IL-6R, and gp130 mRNAs
lived longer regardless of the clinical stage of the disease, histopathologic type or grade of the tumor, tumor size, and lymph node status of the disease. When
the material was divided into subgroups, the number
of patients was lower, but in several subgroups homogenous in terms of lymph node involvement, tumor size or histopathology, the differences in OS were
strikingly statistically significant. Interleukin-6 positivity was correlated with far better survival in the N(⫹)
subgroup, in the T2N(⫹) subgroup, and in the infiltrating ductal and/or lobular subgroup. In univariate
analysis, significant differences in DFS and OS were
correlated with the following prognostic factors:
lymph node status, histopathologic type, tumor size,
IL-6 expression, IL-6R, gp130 expression, and adjuvant
hormonal treatment. In the analyzed group of patients, there were no significant differences in OS and
DFS related to ER or PR status, which is not surprising
because hormonal receptor status is a strong prognostic predictor of response to treatment with hormonal
agents but a rather weak prognostic factor in terms of
CANCER May 1, 2000 / Volume 88 / Number 9
survival analyses.34 Subgroup analyses suggested that
the “new” prognostic factors are independent from
histopathologic type, lymph node status, and tumor
size. In addition, to confirm the above findings, we
analyzed the data in the Cox model. In multivariate
analysis, IL-6R and gp130 remained independent
prognostic factors, and the prognostic impact of the
gp130 and IL-6R status on OS was much higher (as
pronounced by hazard ratios) than the lymph node
status and histopathologic type. The strongest positive
prognostic factor was the gp130 expression then the
To conclude, we found that expression of IL-6,
IL-6R, and gp130 in breast carcinoma tissue is associated with less advanced stages of the disease.
Expressions of IL-6 and its receptor subunits, IL-6R
and gp130, in cancer tissue are correlated and the
presence of their specific transcripts (mRNAs) is a
remarkable positive prognostic factor. Interleukin-6R and gp130 are independent from other well
established factors such as lymph node involvement, tumor size, histopathologic type, grade, and
ER and PR status.
Sporn MB, Roberts A. Autocrine growth factors and cancer.
Nature 1985;313:745–7.
2. Browder TM, Dunbar CE, Nienhuis A. Private and public
autocrine loops in neoplastic cells. Cancer Cells 1989;1:9 –17.
3. Kerbel RS. Expression of multi-cytokine resistance and
multi-growth factor independence in advanced stage metastatic cancer. Am J Pathol 1992;141:519 –24.
4. Tepper RI, Mule JJ. Experimental and clinical studies of
cytokine gene-modified tumor cells. Hum Gene Ther 1994;
5:153– 64.
5. De Jong JS, Van Diest PJ, Van Der Valk P, Baak JPA. Expression of growth factors, growth inhibiting factors, and their
receptors in invasive breast cancer. I. An inventory in search
of autocrine and paracrine loops. J Pathol 1998;184:44 –52 .
6. Kishimoto T, Hirano T. Molecular regulation of B lymphocytes response. Annu Rev Immunol 1988;6:485–512.
7. Takenawa J, Kaneko Y, Fukumoto M, Fukatsu A, Hirano T,
Fukuyama H, et al. Enhanced expression of interleukin 6 in
primary human renal cell carcinomas. J Natl Cancer Inst
1991;83:1668 –72.
8. Heinrich PC, Behrmann I, Mueller-Newen G, Schaper F,
Graeve L. Interleukin 6-type cytokine signaling through the
gp130/JAK/STAT pathway. Biochem J 1998;334:297–314.
9. Taga T, Hibi M, Hirata Y, Yamasaki K, Yasukawa K, Matsuda
T, et al. Interleukin 6 triggers the association of its receptor
with a possible signal transducer gp130. Cell 1989;11:573–
10. Mackiewicz A, Wiznerowicz M, Roeb E, Karczewska A,
Nowak J, Heinrich PC, et al. Soluble interleukin 6 receptor is
biologicaly active in vivo. Cytokine 1995;7:142–9.
11. Chen L, Schulman LM, Revel M. IL-6-receptors and sensitivity to growth inhibition by IL-6 in clones of human breast
carcinoma cells. J Biol Regul Homeost Agents 1991;4:125–36.
12. Douglas AM, Goss GA, Sutherland RL, Hilton DJ, Berndt MC,
Nicola NA, et al. Expression and function of members of the
cytokine receptor superfamily on breast cancer cells. Oncogene 1997;14:661–9.
Tamm I, Cardinale I, Murphy JS. Decreased adherence of
interleukin 6 treated breast carcinoma cells can lead to
separation from neighbors after mitosis. Proc Natl Acad Sci
USA 1991;88:4414 – 8.
Basolo F, Conaldi PG, Fiore L, Calvo S, Toniolo A. Normal
breast epithelial cells produce interleukins 6 and 8 together
with tumor-necrosis factor: defective IL6 expression in
mammary carcinoma. Int J Cancer 1993;55:926 –30.
Basolo F, Fiore L, Fontanini G, Conaldi G, Calvo S, Falcone
V, et al. Expression and response to interleukin 6 (IL6) in
human mammary tumors. Cancer Res 1996;56:3118 –22.
Silvani A, Ferrari G, Paonessa G, Toniatti C, Parmiani G,
Colombo MP. Down-regulation of interleukin 6 receptor
alpha chain in interleukin 6 transduced melanoma cells
causes selective resistance to interleukin 6 but not to oncostatin M. Cancer Res 1995;55:2200 –5.
Crichton MB, Nichols JE, Zhao Y, Bulun SE, Simpson ER.
Expression of transcripts of interleukin 6 and related cytokines by human breast tumors, breast cancer cells, and
adipose stromal cells. Mol Cell Endocrinol 1996;118:215–
Beahrs OH, Henson DE, Hutter RVP, Kennedy BJ, editors.
American Joint Committee on Cancer. Manual for staging of
cancer. Philadelphia: J.B. Lippincott Company, 1992.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform
extraction. Anal Biochem 1987;162:156 – 60.
Allegra JC, Lippman ME. Growth of a human breast cancer
cell line in serum-free hormone supplement medium. Cancer Res 1978;38:3823–9.
Barnes D, Sato G. Growth of human mammary tumor cells
in a serum-free medium. Nature 1979;281:388 –9.
Caivo F, Brower M, Carney DN. Continous culture and soft
agarose cloning of multiple human breast carcinoma cell
lines in serum-free medium. Cancer Res 1984;44:4553–9.
Van der Velde-Zimmerman D, Roijers FM, Bouwens-Rombouts A, De Weger RA, De Graf PW, Tilanus MGJ, et al.
Molecular test for the detection of tumor cells in blood and
sentinel nodes of melanoma patients. Am J Pathol 1996;149:
759 – 64.
Sorg R, Enczmann J, Sorg U, Heermeier K, Schneider M,
Wernet P. Rapid and sensitive mRNA phenotyping for interleukins (IL-1 to IL-6) and colony-stimulating factors (G-CSF,
M-CSF, GM-CSF) by reverse transcription and subsequent
polymerase chain reaction. Exp Hematol 1991;19:882–7.
Marrogi AJ, Munshi A, Merogi AJ, Ohadike Y, El-Habashi A.
Study of tumor infiltrating lymphocytes and transforming
growth factor-␤ as prognostic factors in breast carcinoma.
Int J Cancer 1997;74:492–501.
Green AR, Green VL, White MC, Speirs V. Expression of
cytokine messenger RNA in normal and neoplastic human
breast tissue: identification of interleukin 8 as a potential
regulatory factor in breast tumors. Int J Cancer 1997;72:937–
Venetsanakos E, Beckman I, Bradley J, Skinner JM. High
incidence of interleukin 10 mRNA but not interleukin 2
mRNA detected in human breast tumors. Br J Cancer 1997;
75:1826 –30.
Van Snik J. Interleukin 6: an overview. Ann Rev Immunol
Prognostic Value of IL-6 and Receptors/Karczewska et al.
29. Kortylewski M, Heinrich PC, Mackiewicz A, Schniertshauer
U, Klingmuller U, Nakajima K, et al. Interleukin-6 and oncostatin M-induced growth inhibition of human A375 melanoma cells is STAT-dependent and involves upregulation
of the cyclin-dependent kinase inhibitor p27/Kip1. Oncogene 1999;24;18:3742–53.
30. Danforth DN, Sgagias MK. Interleukin-1 ␣ and interleukin 6
act additively to inhibit growth of MCF-7 breast cancer cells
in vivo. Cancer Res 1993;53:1538 – 45.
31. Porgador A, Tzehoval E, Katz A, Vadai E, Revel M, Feldman M, et al. Interleukin 6 gene transfection into Lewis
lung carcinoma tumor cells suppresses the malignant
phenotype and confers immunotherapeutic competence
against parental metastatic cells. Cancer Res 1992;52:
3679 – 86.
32. Dougherty GJ, Thacker JD, Lavey RS, Belldegrun A, McBride
WH. Inhibitory effect of locally produced and exogenous
interleukin-6 on tumor growth in vivo. Cancer Immunol
Immunother 1994;38:339 – 45
33. Mackiewicz A, Wiznerowicz M, Roeb E, Karczewska A,
Nowak J, Heinrich PC, et al. Interleukin-6 type cytokines and
their receptors for gene therapy of melanoma. Ann NY Acad
Sci 1995;762:361–74
34. DeVita VT Jr., Hellman S, Rosenberg SA, editors. Cancer:
principles and practice of oncology.5th ed. Philadelphia: JB
Lippincott, 1997;36:1590 –1
Без категории
Размер файла
149 Кб
Пожаловаться на содержимое документа