The Breast 23 (2014) 88e93

Contents lists available at ScienceDirect

The Breast
journal homepage: www.elsevier.com/brst

Original article

Serum levels of CEA and CA15-3 in different molecular subtypes and

prognostic value in Chinese breast cancer
San-gang Wu a,1, Zhen-yu He b,1, Juan Zhou c,1, Jia-yuan Sun b, Feng-yan Li b, Qin Lin a,
Ling Guo d, **, Huan-xin Lin b, *
a
Xiamen Cancer Center, Department of Radiation Oncology, The First Affiliated Hospital of Xiamen University, Xiamen, People’s Republic of China
b
State Key Laboratory of Oncology in Southern China, Department of Radiation Oncology, Sun Yat-Sen University Cancer Center, Guangzhou,
People’s Republic of China
c
Xiamen Cancer Center, Department of Obstetrics and Gynecology, The First Affiliated Hospital of Xiamen University, Xiamen, People’s Republic of China
d
State Key Laboratory of Oncology in Southern China, Department of Nasopharyngeal Carcinoma, Sun Yat-Sen University Cancer Center, Guangzhou,
People’s Republic of China

a r t i c l e i n f o a b s t r a c t

Article history: The prognostic significance of preoperative carcinoembryonic antigen (CEA) and cancer antigen 15-3
Received 14 May 2013 (CA15-3) levels in breast cancer is controversial. This study evaluated the prognostic value of preoper-
Received in revised form ative serum CEA and CA15-3 levels in Chinese breast cancer patients. A total of 470 patients with breast
13 October 2013
cancer had preoperative CEA and CA15-3 concentrations measured. The relationships between preop-
Accepted 15 November 2013
erative concentration and clinicopathological factors and outcomes were determined. CEA and CA15-3
levels were increased in 34 (7.2%) and 58 (12.3%) patients, respectively. Elevations of serum CEA and
Keywords:
CA-15-3 levels correlated with the primary tumor size and axillary lymph node status. CEA levels were
Breast cancer
CEA
lower in patients with triple-negative breast cancer than in those with other subtypes (P ¼ 0.002). The 5-
CA15-3 year distant metastasis-free survival (DMFS), disease-free survival (DFS), and overall survival (OS) of CEA-
Prognosis negative vs. CEA-positive patients were 84.1% vs. 54.5% (P < 0.001), 82.7% vs. 54.8% (P < 0.001), and 89.7%
vs. 78.5% (P ¼ 0.007), respectively. The 5-year DMFS, DFS, and OS of CA15-3-negative vs. CA15-3-positive
patients were 84.0% vs. 69.6% (P ¼ 0.002), 83.0% vs. 66.2% (P < 0.001), 90.9% vs. 74.2% (P ¼ 0.005),
respectively. Multivariate analysis of prognosis indicated that CEA and CA15-3 levels were independent
prognostic factors for DMFS (P ¼ 0.021) and DFS (P ¼ 0.032), and DFS (P ¼ 0.014) and OS (P ¼ 0.032),
respectively. Serum levels of CEA and CA15-3 may differ in breast cancer molecular subtypes and pre-
operative levels of CEA and CA15-3 have a significant effect on prognosis in Chinese women with breast
cancer.
Ó 2013 Elsevier Ltd. All rights reserved.

Introduction
Breast cancer is one of the most common malignant tumors in
women [1], and its incidence has been steadily increasing in China
[2]. With advancements in early diagnostic methods and combined
* Corresponding author. State Key Laboratory of Oncology in Southern China,
Department of Radiation Oncology, Sun Yat-Sen University Cancer Center, 651
Dongfeng Road East, Guangzhou 510060, People’s Republic of China. Tel.: þ86 20
87343543; fax: þ86 20 87343392.
** Corresponding author. State Key Laboratory of Oncology in Southern China,
Department of Nasopharyngeal Carcinoma, Sun Yat-Sen University Cancer Center,
651 Dongfeng Road East, Guangzhou 510060, People’s Republic of China. Tel.: þ86
20 87343543; fax: þ86 20 87343392.
E-mail addresses: hecunzhang@163.com (L. Guo), hezhy@sysucc.org.cn
(H.-x. Lin).
1
San-gang Wu, Zhen-yu He and Juan Zhou contributed equally to this work.
modality therapies, the mortality rate of breast cancer has declined
in recent years [3]. However, once treatment failure occurs the
quality of life and the survival rate of patients is significantly
affected. Therefore, it is essential to identify reliable prognostic
factors to guide decision making during the treatment of breast
cancer in order to improve survival rates. Traditional prognostic
factors including tumor size, axillary lymph node status, hormone
receptor status, and human epidermal growth factor receptor 2
(HER2) expression are widely used in clinical practice [4], but these
characteristics do not fully reflect the prognosis of breast cancer.
Circulating tumor markers such as carcinoembryonic antigen
(CEA) and cancer antigen 15-3 (CA15-3) have been studied as
prognostic factors in breast cancer for more than 30 years. Plasma
CEA and CA15-3 are the most common tumor markers used in
breast cancer [5e10], and CEA was the first tumor antigen to be
studied [11]. CEA is a type of cell adhesion molecule, and CEA levels

0960-9776/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.breast.2013.11.003
 S. Wu et al. / The Breast 23 (2014) 88e93
in the blood are usually increased once the cancer has metastasized
[12]. CA15-3, a member of the mucin-1 (MUC-1) family of glyco-
proteins, is over-expressed in cancers and altered glycosylation of
CA15-3 makes it a useful tumor marker [13].
In recent years, the prognostic value of preoperative CEA and
CA15-3 levels in breast cancer has gained much attention. Study has
shown that preoperative plasma CEA levels combined with CA15-3
levels may provide useful information for diagnosis and treatment
of breast cancer [14]. However, the prognostic significance of pre-
operative CEA and CA15-3 levels in breast cancer remains contro-
versial. Previous reports have suggested that preoperative serum
levels of CEA and CA15-3 are related to prognostic factors including
tumor size, axillary lymph node status, and hormone receptor
status in breast cancer patients [5e10]. Accordingly, the European
Group on Tumor Markers has recommended the CEA and CA15-3
levels be used for assessing prognosis, the early detection of dis-
ease progression, and treatment monitoring in breast cancer [14].
In contrast, a report by Maric et al. [15] failed to support this
conclusion. In addition, there is evidence showing that these
markers are not sufficiently sensitive or specific for early diagnosis
of breast cancer [16,17]. As a result, the American Society of Clinical
Oncology (ASCO) does not recommend the use of CEA and CA15-3
in screening, diagnosis, staging, and treatment monitoring of breast
cancer [18].
Therefore, we conducted a retrospective analysis of clinico-
pathological data of Chinese women with breast cancer to explore
the relationships between preoperative plasma CEA and CA15-3
levels and traditional clinicopathological prognostic factors, as
well as the efficacy of treatment.
Patients and methods
Study subjects and inclusion criteria
We performed a retrospective analysis of data from breast
cancer patients treated at the Sun Yat-sen University Cancer Center
between January 2004 and December 2007. All subjects who were
enrolled in the study gave written informed consent for the use of
their medical records for research purposes, and the study design
and method were approved by the Ethics Committee of Sun Yat-sen
University Cancer Center. Inclusion criteria were: female; unilateral
invasive breast cancer; underwent mastectomy or breast-
conserving surgery; CEA and CA15-3 levels were determined
before surgery; tumor completely removed by surgery with no
residual tumor as indicated by pathologic evaluation; appropriate
adjuvant chemotherapy, adjuvant radiotherapy and endocrine
therapy administered based on international guidelines; complete
results of estrogen receptor (ER), progesterone receptor (PR), HER2,
Ki-67, and histologic grade; and no sign of metastasis or secondary
cancer in other organs at diagnosis.
Measurement of CEA and CA15-3 levels and immunohistochemical
evaluation
Venous blood (5 mL) was collected from all patients before
treatment. Serum samples were obtained by centrifugation, placed
in freezing tubes, and immediately stored at 80  C. Serum CEA
and CA15-3 levels were determined using an automatic electro-
chemistry luminescence immunoassay system (ROCHE E170;
Roche, Germany). The cut-off values of CEA and CA15-3 were
5.0 ng/mL and 25 U/mL, respectively, and the value was considered
positive or negative for the marker if the level was above or below
the cut-off value, respectively.
ER- and PR-positive were defined as the presence of >10%
immunostained malignant cells were. HER2-positivity was
89
indicated by a 3þ or 2þ score from the immunohistochemical
evaluation, and was confirmed using a fluorescence in situ hy-
bridization (FISH) test for HER2. The Ki-67 staining was considered
negative, weakly-positive, positive, and strongly-positive if the
percentage of cells stained among the cells in the field selected in
the tissue section were <10%, between 10% and 25%, between 26%
and 50%, and >50%, respectively. A cut-off point of 25% was used to
distinguish between the categories of low and high proliferative
tumors. We could not exactly define breast cancer intrinsic sub-
types with immunohistochemistry in all tumors [19] since Ki-67
assessment was not available as an exact value. Thus, we classi-
fied breast cancer intrinsic subtypes as follows: luminal A (ERþ
and/or PRþ, HER2); luminal B (ERþ and/or PRþ, HER2þ); HER2
positive (ER and PR, HER2þ); and triple-negative (ER and PR,
Her-2) [20].
Follow-up and study endpoints
Patients were followed up at an interval of 3e6 months, with
the day pathological diagnosis was performed considered as the
first day of follow-up. The primary study endpoints were locore-
gional recurrence-free survival (LRFS; locoregional recurrence with
or without distant metastasis), distant metastasis-free survival
(DMFS), disease-free survival (DFS), and overall survival (OS).
Locoregional recurrence was defined as pathologically confirmed
relapse on the chest wall, supra- and infraclavicular fossa, axillary
area, or internal mammary region. Distant metastasis was
confirmed using 2 medical imaging methods, and pathology
assessment if needed. Spread of the primary cancer to sites distant
from the locoregional recurrence sites were considered indicative
of distant metastasis. The mortality endpoint was defined as breast
cancer-related death. The survival condition of the patients was
obtained by reviewing their medical records, via a phone consul-
tation, or through a mailed letter.
Statistical analysis
All data were analyzed using the SPSS 16.0 statistical software
package. The c2 and Fisher’s exact probability tests were used to
analyze the differences between qualitative data. The Kaplane
Meier method was employed to generate survival plots and to
compare survival rates, and statistical significance was determined
using the log-rank test. Stepwise Cox regression analysis was used
for multivariate analysis. Factors that were significant in the uni-
variate analysis were included in the stepwise Cox regression
analysis. P values below 0.05 were considered significant.
Results
Patient and treatment characteristics
There were 1021 patients treated for early breast cancer be-
tween 2004 and 2007, of who 470 met criteria for inclusion in
the study. The clinicopathological characteristics of the patients
and the treatments are shown in Table 1. The median age of
disease onset was 48 years (range, 25e73 years). The patholog-
ical classification of 449 cases (95.5%) was invasive ductal carci-
noma. Among all the cases, 262 (55.7%) were classified as luminal
A, 72 (15.3%) as luminal B, 73 (15.5%) as HER2 positive, and 63
(13.5%) as triple-negative. Ki-67 was >25% in 239 cases (50.9%),
and 442 (94.0%) were classified as histologic grade II and above. A
total of 441 (93.8%) patients underwent total mastectomy, and 29
(6.2%) received breast-conserving surgery. The median number of
axillary lymph nodes removed was 16. A total of 452 patients
(96.2%) received adjuvant chemotherapy, and among them 12
90 S. Wu et al. / The Breast 23 (2014) 88e93

Table 1
General characteristics of study population and correlation between serum CA15-3 and CEA levels and clinicopathological factors.

Characteristic n CEA CA15-3

CEA high (%) CEA normal (%) P CA15-3 high (%) CA15-3 normal (%) P

Age (y)
&35 52 4(7.7) 48(92.3) 0.892 9(17.3) 43(82.7) 0.248
>35 418 30(7.2) 388(92.8) 49(11.7) 369(88.3)
Tumor size
T1 187 6(3.2) 181(96.8) <0.001* 14(7.5) 173(92.5) <0.001*
T2 233 16(6.9) 217(93.1) 28(12.0) 205(88.0)
T3 20 5(25.0) 15(75.0) 7(35.0) 13(65.0)
T4 30 7(23.3) 23(76.7) 9(30.0) 21(70.0)
Nodal status
N0 296 11(3.7) 285(96.3) <0.001* 21(7.1) 275(92.9) <0.001*
N1 67 6(9.0) 61(91.0) 15(22.4) 52(77.6)
N2 58 8(13.8) 50(86.2) 13(22.4) 45(77.6)
N3 49 9(18.4) 40(81.6) 9(18.4) 40(81.6)
Molecular subtype
LuminalA 262 13(5.0) 249(95.0) 0.002* 27(10.3) 235(89.7) 0.513
LuminalB 72 9(12.5) 63(87.5) 11(15.3) 61(84.7)
HER2 positive 73 11(15.1) 62(84.9) 11(15.1) 62(84.9)
Triple-negative 63 1(1.6) 62(98.4) 9(14.3) 54(85.7)
Ki-67 (%)
&25 231 13(5.6) 218(94.4) 0.186 22(9.5) 209(90.5) 0.069
>25 239 21(8.8) 218(91.2) 36(15.1) 203(84.9)
Histologic grade
I 28 1(3.6) 27(96.4) 0.402 3(10.7) 25(89.3) 0.662
II 241 21(8.7) 220(91.3) 27(11.2) 214(88.8)
III 201 12(6.0) 189(94.0) 28(13.9) 173(86.1)
Adjuvant chemotherapy
None 18 1(5.6) 17(94.4) 0.620 2(11.1) 16(88.9) 0.613
Yes 452 33(7.3) 419(92.7) 56(12.4) 396(87.6)
Adjuvant radiotherapy
None 393 26(6.6) 367(93.4) 0.242 42(10.7) 351(89.3) 0.014*
Yes 77 8(10.4) 69(89.6) 16(20.8) 61(79.2)
Endocrine therapy
None 136 12(8.8) 124(91.2) 0.396 20(14.7) 116(85.3) 0.320
Yes 334 22(6.6) 312(93.4) 38(11.4) 296(88.6)

CEA, carcinoembryonic antigen; CA15-3, cancer antigen 15-3; HER2, human epidermal growth factor receptor 2.
*P < 0.05 indicates a significant difference.

received a regimen consisting of cyclophosphamide, metho-
trexate, and 5-fluorouracil (CMF), and 440 received regimens
with anthracycline and/or taxane. Forty patients (8.5%) received
neoadjuvant chemotherapy with anthracycline and/or taxane.
The median number of chemotherapy cycles was 6 (range, 6e8
cycles). There were 334 patients (71.6%) with positive hormone
receptor status who received endocrine therapy. None of the
patients with HER2 positive cancers received trastuzumab (Her-
ceptin) treatment. A total of 77 patients (16.4%) received adju-
vant radiotherapy within 6 months after mastectomy or breast-
conserving surgery.
Relationship between CEA and CA15-3 levels and clinicopathological
data
Table 1 shows the correlations between CEA and CA15-3 levels
and clinicopathological factors. CEA and CA15-3 levels were
elevated in 34 (7.2%) and 58 (12.30%) cases, respectively. The
analysis indicated that elevation of CEA was positively correlated
with the size of the primary tumor (P < 0.001) and axillary lymph
node status (P < 0.001). The elevation of CEA levels was signifi-
cantly greater in patients with HER2 positive tumors (12.5% in
luminal B and 15.1% in HER2 positive tumors) than in patients with
other subtypes (5.0% in luminal A and 1.6% in triple-negative tu-
mors), whereas elevation in patients with the triple-negative sub-
type (1.6%) was significantly lower than that in patients with other
subtypes (5.0%e15.1%) (P ¼ 0.002). Increased CEA levels were not
associated with age, Ki-67, or pathological grade. CA15-3 levels
were positively correlated with the size of the primary tumor
(P < 0.001) and axillary lymph node status (P < 0.001), but had no
significant correlation with the other factors. Since adjuvant
treatment was not determined by tumor marker levels, but by the
international guidelines, more patients with elevated CA15-3 levels
received adjuvant radiotherapy (P ¼ 0.014); there was no statisti-
cally difference in other adjuvant treatments according to CA 15-3
and CEA levels.
Survival and prognostic value of CEA and CA15-3 levels
The median follow-up duration of the 470 patients was 49.9
months (range, 13e80 months). Locoregional recurrence occurred
in 29 patients, with a 5-year LRFS of 92.7%. Distant metastases
developed in 70 cases, with 5-year DMFS and DFS rates of 82.2% and
80.8%, respectively. Forty-five patients died, with a 5-year OS rate
89.0%.
Univariate analysis showed that the 5-year DMFS, DFS, and OS of
CEA-negative vs. CEA-positive patients were 84.1% vs. 54.5%
(P < 0.001), 82.7% vs. 54.8% (P < 0.001), and 89.7% vs. 78.5%
(P ¼ 0.007), respectively (Table 2 and Fig. 1). The 5-year DMFS, DFS,
and OS of CA15-3-negative vs. CA15-3-positive patients were 84.0%
vs. 69.6% (P ¼ 0.002), 83.0% vs. 66.2% (P < 0.001), 90.9% vs. 74.2%
(P ¼ 0.005), respectively (Table 2 and Fig. 1). Multivariate analysis
using the Cox model indicated CEA to be an independent prognostic
factor for DMFS (P ¼ 0.021) and DFS (P ¼ 0.032), and CA15-3 to be
an independent prognostic factor for DFS (P ¼ 0.014) and OS
(P ¼ 0.032). Traditional clinicopathological factors such as lymph
node status and histologic grade were also shown to have inde-
pendent prognostic value (Table 3).
 S. Wu et al. / The Breast 23 (2014) 88e93 91

Table 2
Prognostic factors for survival in univariate analyses.

Characteristic LRFS DMFS DFS OS

5-year (%) P 5-year (%) P 5-year (%) P 5-year (%) P

Age (y)
&35 93.8 0.923 81.7 0.569 81.7 0.775 85.3 0.292
>35 92.5 82.3 80.7 89.5
Tumor size
T1eT2 93.6 0.007* 84.8 <0.001* 83.4 <0.001* 91.2 <0.001*
T3eT4 85.0 60.7 59.0 69.5
Nodal status
N0eN1 95.1 <0.001* 88.7 <0.001* 87.7 <0.001* 94.5 <0.001*
N2eN3 84.0 59.4 56.7 70.2
Molecular subtype
LuminalA 94.0 0.542 84.1 0.143 82.8 0.161 91.1 0.242
LuminalB 90.7 78.1 76.7 88.5
HER2-enriched 92.2 76.5 75.2 83.5
Triple-negative 89.8 86.2 84.6 87.7
Ki-67 (%)
&25 93.1 0.553 85.6 0.089 83.2 0.234 92.1 0.082
>25 92.3 78.8 78.5 86.1
Histologic grade
I 100 0.043* 100 0.005* 100.0 <0.001* 100 <0.001*
II 94.2 84.3 83.9 93.4
III 89.6 77.0 74.2 82.2
CEA (ng/ml)
&5 93.2 0.333 84.1 <0.001* 82.7 <0.001* 89.7 0.007*
>5 84.2 54.5 54.8 78.5
CA15-3 (U/ml)
&25 93.3 0.115 84.0 0.002* 83.0 <0.001* 90.9 0.005*
>25 88.4 69.6 66.2 74.2
Adjuvant chemotherapy
None 100 0.269 94.4 0.263 94.4 0.220 94.4 0.594
Yes 92.4 81.7 80.3 88.8
Adjuvant radiotherapy
None 93.3 0.131 82.4 0.530 81.6 0.200 90.2 0.193
Yes 89.3 81.1 77.2 83.4
Endocrine therapy
None 91.1 0.268 81.0 0.262 79.5 0.250 85.2 0.089
Yes 93.4 82.9 81.6 91.5

CEA, carcinoembryonicantigen; CA15-3, cancer antigen 15-3; LRFS, locoregional recurrence-free survival; DMFS, distant metastasis-free survival; DFS, disease-free survival;
OS, overall survival.
*P < 0.05 indicates a significant difference.

Discussion DFS and OS of breast cancer patients [6,9,19,24e26], favoring the

The results of this study including 470 Chinese women with
breast cancer showed that preoperative serum CEA and CA15-3
levels were independent factors affecting prognosis, in particular
distant recurrence.
Studies of the correlation between preoperative levels of CEA
and CA15-3 and prognostic factors have concentrated on traditional
factors, including T and N stage [5e10,21]. The present study also
supports a significant relationship between these markers and T
and N stage. Reports on the correlation between CEA and CA15-3
and molecular subtypes, however, are rare. Lee et al. [21] re-
ported that CEA expression was positively correlated with HER2
expression, but not related to ER or PR status in breast cancer. Our
study showed that at the time of diagnosis, HER2-positivity pa-
tients (luminal B and HER2 positive tumors) had greater serum CEA
levels as compared to those with other tumor subtypes, whereas
the triple-negative subtype was associated with lower serum CEA
levels than other subtypes This may be explained in part by the
undifferentiated characteristics of triple-negative breast cancers,
and the different biological behaviors of different molecular
subtypes.
Studies by Duffy et al. [22] and Kumpulainen et al. [23] both
found that the CA15-3 expression level is an independent prog-
nostic factor for OS. However, the majority of reports have shown
that both CEA and CA15-3 expression levels significantly affect the
prognostic value of CA15-3 expression level to that of CEA
[5,6,22,27e29]. Our study found that CEA and CA15-3 levels had no
effect on LRFS, but the CEA level affected DMFS and DFS and the
CA15-3 level affected DFS and OS, suggesting that the prognostic
value of the 2 markers varies with the endpoint considered. How-
ever, it should be pointed out that in multivariable analysis,
although CEA and CA15-3 had some independent predictive ability,
the strongest predictors of outcome were nodal status and tumor
grade.
Despite the use of systemic chemotherapy in most patients, the
levels of CEA and CA15-3 still significantly affect the prognosis of
patients. It is possibly that the expression of CEA and CA15-3 are
associated with the potential for micrometastasis of cancer cells.
This further emphasizes the need for the application of individu-
alized treatment to improve survival in patients with positive
expression of both CEA and CA15-3.
The utility of measuring CEA and CA15-3 levels in patients with
breast cancer remains controversial. The European Group on Tumor
Markers has recommended the use of CEA and CA15-3 for the
prediction of prognosis and for the monitoring of recurrence and
therapeutic effects [14]. However, ASCO guidelines suggest that
these markers should only be used for monitoring clinical param-
eters in the treatment of metastatic breast cancer [18]. The National
Comprehensive Cancer Network (NCCN) guidelines do not recom-
mend the use of CEA or CA15-3 as markers for clinical evaluation
92 S. Wu et al. / The Breast 23 (2014) 88e93

Fig. 1. KaplaneMeier survival curves of patients with normal or high CEA and CA15-3 levels. Distant metastasis-free survival (DMFS) according to preoperative carcinoembryonic
antigen (CEA) (A) and cancer antigen 15-3 (CA15-3) (D) levels. Disease-free survival (DFS) according to preoperative CEA (B) and CA15-3 (E) levels. Overall survival (OS) according to
preoperative CEA (C) and CA15-3 (F) levels.

Table 3
Prognostic factors for survival in multivariable analyses.

Characteristic LRFS DMFS DFS OS

HR 95%CI P HR 95%CI P HR 95%CI P HR 95%CI P

Tumor size
T3eT4 vs. T1eT2 1.967 0.814e4.755 0.133 1.412 0.760e2.623 0.274 1.463 0.814e2.630 0.203 1.742 0.844e3.595 0.133
Nodal status
N2eN3 vs. N0eN1 4.199 2.027e8.711 <0.001* 4.697 2.894e7.623 <0.001* 4.82 3.036e7.653 <0.001* 5.617 3.083e10.235 <0.001*
Histologic grade
III vs. IeII 2.304 1.088e4.882 0.029* 2.081 1.281e3.380 0.003* 2.339 1.461e3.744 <0.001* 3.281 1.742e6.179 <0.001*
CEA (ng/mL)
Positive vs. Negative e 2.196 1.126e4.285 0.021* 2.048 1.062e3.948 0.032* 1.385 0.564e3.403 0.477
CA15-3 (U/mL)
Positive vs. Negative e 1.735 0.970e3.101 0.063 1.982 1.149e3.421 0.014* 2.119 1.066e4.214 0.032*

CEA, carcinoembryonic antigen; CA15-3, cancer antigen 15-3; LRFS, locoregional recurrence-free survival; DMFS, distant metastasis-free survival; DFS, disease-free survival;
OS, overall survival; HR, Hazard ratio; CI, Confidence interval.
*P < 0.05 indicates a significant difference.

before the initiation of treatment for breast cancer [30]. Based on
our study results, preoperative CEA and CA15-3 levels might be
considered independent factors affecting prognosis in patients with
breast cancer, at least in Chinese women.
There are several limitations to this study that should be
considered. First, our conclusions are based on the results of a
retrospective study, with bias in subject selection and a relatively
short duration of follow-up. Furthermore, the effects of CEA and
CA15-3 levels (positive and negative) on prognosis were studied
qualitatively, but not quantitatively. In addition, none of the pa-
tients with HER2 positive breast cancer were treated with trastu-
zumab. Further analyses, including a prospective study design and
quantitative evaluation of CEA and CA15-3 levels, will help clarify
the utility of these serum biomarkers as prognostic indicators of
clinically-relevant endpoints.
In summary, the present study showed that preoperative CEA
levels were higher in patients with HER2 positive breast cancer, and
lower in those with triple-negative breast cancer. Preoperative CEA
and CA15-3 levels have a significant effect on prognosis in Chinese
women with breast cancer. Further prospective studies are required
to evaluate the significance of these findings to provide more in-
formation regarding therapeutic decision-making in the clinical
setting.
Grant support
This study was supported by a grant from the Youth Foundation
of the First Affiliated Hospital of Xiamen University (No.
XYY2012005) and the Sci-Tech Office of Guangdong Province (No.
2008B060 600019).
 S. Wu et al. / The Breast 23 (2014) 88e93
Conflict of interest disclosures
The authors have no financial conflicts of interest to declare.
Acknowledgments
None.
References
[1] DeSantis C, Siegel R, Bandi P, Jemal A. Breast cancer statistics, 2011. CA Cancer
J Clin 2011;61(6):409e18.
[2] Li J, Zhang BN, Fan JH, Pang Y, Zhang P, Wang SL, et al. A nation-wide
multicenter 10-year (1999e2008) retrospective clinical epidemiological study
of female breast cancer in China. BMC Cancer 2011;11:364.
[3] Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin
2010;60(5):277e300.
[4] Soerjomataram I, Louwman MW, Ribot JG, Roukema JA, Coebergh JW. An
overview of prognostic factors for long-term survivors of breast cancer. Breast
Cancer Res Treat 2008;107(3):309e30.
[5] Molina R, Augé JM, Escudero JM, Filella X, Zanon G, Pahisa J, et al. Evalu-
ation of tumor markers (HER-2/neu oncoprotein, CEA, and CA 15.3) in
patients with locoregional breast cancer: prognostic value. Tumour Biol
2010;31(3):171e80.
[6] Park BW, Oh JW, Kim JH, Park SH, Kim KS, Kim JH, et al. Preoperative CA 15-3
and CEA serum levels as predictor for breast cancer outcomes. Ann Oncol
2008;19(4):675e81.
[7] Lumachi F, Basso SM, Brandes AA, Pagano D, Ermani M. Relationship between
tumor markers CEA and CA 15-3, TNM staging, estrogen receptor rate and MIB-1
index in patients with pT1-2 breast cancer. Anticancer Res 2004;24(5B):3221e4.
[8] Molina R, Filella X, Zanon G, Pahisa J, Alicarte J, Munoz M, et al. Prospective
evaluation of tumor markers (c-erbB-2 oncoprotein, CEA and CA 15.3) in pa-
tients with locoregional breast cancer. Anticancer Res 2003;23(2A):1043e50.
[9] Molina R, Auge JM, Farrus B, Zanón G, Pahisa J, Muñoz M, et al. Prospective
evaluation of carcinoembryonic antigen (CEA) and carbohydrate antigen 15.3
(CA 15.3) in patients with primary locoregional breast cancer. Clin Chem
2010;56(7):1148e57.
[10] Samy N, Ragab HM, El Maksoud NA, Shaalan M. Prognostic significance of
serum Her2/neu, BCL2, CA15-3 and CEA in breast cancer patients: a short
follow-up. Cancer Biomark 2010;6(2):63e72.
[11] Gold P, Freedman SO. Demonstration of tumor-specific antigens in human
colonic carcinomata by immunological tolerance and absorption techniques.
J Exp Med 1965;121:439e62.
[12] Benchimol S, Fuks A, Jothy S, Beauchemin N, Shirota K, Stanners CP. Carci-
noembryonic antigen, a human tumor marker, functions as an intercellular
adhesion molecule. Cell 1989;57(2):327e34.
[13] Duffy MJ, Evoy D, McDermott EW. CA 15-3: uses and limitation as a biomarker
for breast cancer. Clin Chim Acta 2010;411(23e24):1869e74.
[14] Molina R, Barak V, van Dalen A, Duffy MJ, Einarsson R, Gion M, et al. Tumor
markers in Breast Cancer European Group on tumor markers recommenda-
tions. Tumour Biol 2005;26(6):281e93.
93
[15] Maric P, Ozreti c P, Levanat S, Oreskovic S, Antunac K, Beketi c-Oreskovi
c L.
Tumor markers in breast cancer-evaluation of their clinical usefulness. Coll
Antropol 2011;35(1):241e7.
[16] Hou MF, Chen YL, Tseng TF, Lin CM, Chen MS, Huang CJ, et al. Evaluation of
serum CA27.29, CA15-3 and CEA in patients with breast cancer. Kaohsiung J
Med Sci 1999;5(9):520e8.
[17] Clinton SR, Beason KL, Bryant S, Johnson JT, Jackson M, Wilson C, et al.
A comparative study of four serological tumor markers for the detection of
breast cancer. Biomed Sci Instrum 2003;39:408e14.
[18] Harris L, Fritsche H, Mennel R, Norton L, Ravdin P, Taube S, et al. American
Society of Clinical Oncology 2007 update of recommendations for the use of
tumor markers in breast cancer. J Clin Oncol 2007;25(33):5287e312.
[19] Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thürlimann B, Senn HJ, et al.
Strategies for subtypes e dealing with the diversity of breast cancer: high-
lights of the St. Gallen International Expert Consensus on the primary therapy
of early breast cancer 2011. Ann Oncol 2011;22(8):1736e47.
[20] Houssami N, Macaskill P, von Minckwitz G, Marinovich ML, Mamounas E.
Meta-analysis of the association of breast cancer subtype and pathologic
complete response to neoadjuvant chemotherapy. Eur J Cancer 2012;48(18):
3342e54.
[21] Lee JS, Park S, Park JM, Cho JH, Kim SI, Park BW. Elevated levels of preoperative
CA 15-3 and CEA serum levels have independently poor prognostic signifi-
cance in breast cancer. Ann Oncol 2013;24(5):1225e31.
[22] Duffy MJ, Duggan C, Keane R, Hill AD, McDermott E, Crown J, et al. High
preoperative CA 15-3 concentrations predict adverse outcome in node-
negative and node-positive breast cancer: study of 600 patients with histo-
logically confirmed breast cancer. Clin Chem 2004;50(3):559e63.
[23] Kumpulainen EJ, Keskikuru RJ, Johansson RT. Serum tumor marker CA 15.3
and stage are the two most powerful predictors of survival in primary breast
cancer. Breast Cancer Res Treat 2002;76(2):95e102.
[24] Ebeling FC, Schmitt UM, Untch M, Nagel D, Fateh-Moghadam A, Stieber P,
et al. Tumour markers CEA and CA 15-3 as prognostic factors in breast
cancer e univariate and multivariate analysis. Anticancer Res 1999;19(4A):
2545e50.
[25] Cañizares F, Sola J, Pérez M, Tovar I, De Las Heras M, Salinas J, et al. Preop-
erative values of CA 15-3 and CEA as prognostic factors in breast cancer: a
multivariate analysis. Tumour Biol 2001;22(5):273e81.
[26] Brenner B, Siris N, Rakowsky E, Fenig E, Sulkes A, Lurie H. Prediction of
outcome in locally advanced breast cancer by post-chemotherapy nodal sta-
tus and baseline serum tumour markers. Br J Cancer 2002;87(12):1404e10.
[27] Uehara M, Kinoshita T, Hojo T, Akashi-Tanaka S, Iwamoto E, Fukutomi T. Long-
term prognostic study of carcinoembryonic antigen (CEA) and carbohydrate
antigen 15-3 (CA 15-3) in breast cancer. Int J Clin Oncol 2008;13(5):447e51.
[28] Ebeling FG, Stieber P, Untch M, Nagel D, Konecny GE, Schmitt UM, et al. Serum
CEA and CA 15-3 as prognostic factors in primary breast cancer. Br J Cancer
2002;86(8):1217e22.
[29] Gion M, Boracchi P, Dittadi R, Biganzoli E, Peloso L, Mione R, et al. Prognostic
role of serum CA15.3 in 362 node-negative breast cancers. An old player for a
new game. Eur J Cancer 2002;38(9):1181e8.
[30] National Comprehensive Cancer Network [Web page]. National Comprehen-
sive Cancer Network (NCCN) clinical practice guidelines in oncology, breast
cancer. Version 1. Available at: http://www.nccn.org/professionals/physician_
gls/pdf/breast.pdf; 2011 [cited April 10 2013].