This article is mentioned in:

Introduction

Breast cancer is one of the most common malignancies and leading causes of cancer-associated mortalities among females (1). Based on DNA microarray techniques, breast cancer is classified into five subtypes: Luminal A; luminal B; normal breast-like; human epidermal growth factor receptor 2 (HER2/neu) overexpressing; and basal-like (2). The subtype that is immunohistochemically characterized by the lack of expression of the estrogen receptor (ER), progesterone receptor (PR) and HER2 is defined as triple negative breast cancer (TNBC) (3).

TNBC, which accounts for 10–15% of breast cancers, is considered to exhibit an aggressive clinical behavior and poor prognosis, due to the insensitivity of the cancer to endocrine and targeted therapy (4–12). Therefore, chemotherapy is a significant therapy for such cancers. The treatment options for TNBC include anthracyclines, taxanes, platinum and alkylating agents (13–16). However, there is no standard chemotherapy regimen for TNBC.

Based on National Comprehensive Cancer Network guidelines, the adjuvant 5-fluorouracil, epirubicin and cyclophosphamide (FEC) chemotherapy regimen is the suggested regimen for breast cancer. However, there is no conclusion regarding the clinicopathological and demographical factors of TNBC patients that are suitable for adjuvant FEC chemotherapy. Therefore, the aim of the present retrospective study is to analyze the association between the clinicopathological and demographical characteristics and the survival of TNBC patients that receive FEC adjuvant chemotherapy.

Materials and methods

Patients

In total, 956 patients were diagnosed with TNBC, and 25.0% (239/956) of these patients had received adjuvant FEC chemotherapy with surgery, modified radical mastectomy or breast-conserving surgery plus a sentinel lymph node examination at the Harbin Medical University (Harbin, Heilongjiang, China) between April 2001 and December 2008. These 239 patients were included in the present study. Primary cancers were evaluated in accordance with the 7th edition of the American Joint Committee on Cancer (AJCC) tumor-node-metastasis (TNM) staging system. The inclusion criteria included females with histologically confirmed ER, PR and HER2-negative breast cancer, between stages I and IIIA. The criteria for TNBC were 0% expression for ER, 0% expression for PR and HER2 expression of 0 or 1+ only. Patients with an immunohistochemical score for HER-2 neu of 2+ demonstrated no HER2 gene amplification by fluorescence in situ hybridization. These patients had not undergone neoadjuvant chemotherapy and radiotherapy prior to surgery. The exclusion criteria included stage IV disease and a history of other cancers. Pathology reports were obtained from the medical record room of the Harbin Medical University. Data on the demographical and clinical characteristics, extent of disease, and types of surgery, chemotherapy and radiotherapy were collected from the medical records. The data also included the menopausal status based on the hormone level and age, family history of cancer in first and second-degree relatives, tumor size and lymph node status. The number of pregnancies consisted of full-term pregnancies, non-full-term pregnancies and miscarriages. Clinical data of these patients were used for survival analysis.

Treatment

All patients underwent conservative surgery or a modified radical mastectomy. Chemotherapy consisting of adjuvant FEC was administered to patients (500 mg/m2 cyclophosphamide, day 1; 75 mg/m2 epirubicin, day 1; 500 mg/m2 5-fluorouracil, days 1 and 8, every 3 weeks, for at least 4 cycles).

Statistical Analysis

The vital status of the study group was assessed from the medical record room. January 30, 2013 was the follow-up completion date. Univariate and multivariate Cox proportional hazard models were used to assess the effects of variables on patient survival (Table I). The parameters were then tested using the multivariate Cox proportional hazards model, which was performed to identify the independent variables for predicting survival. Hazard ratios (HR) and the 95% confidence intervals (CIs) were recorded for each factor. Disease-free survival (DFS) time was calculated from the date of surgery resection to the date of the final follow-up or relapse. Overall survival (OS) time was defined as the elapsed time between the date of the surgery and the date of last follow-up or mortality. The cut-off value was selected using the receiver operating characteristic (ROC) curve analysis (Fig. 1). Survival was estimated using the Kaplan-Meier method (Fig. 2). Spearman's χ2 test was used to analyze categorical variables (Table II). P<0.05 was considered to indicate a statistically significant difference. SPSS 19.0 software for Windows (IBM SPSS, Armonk, NY, USA) was used for statistical analysis.

Figure 1. Receiver operating characteristic curves of the effect of lymph node status on TNBC recurrence and mortality. (A) The AUC of lymph node status for TNBC recurrence was 0.676 and P=0.002. (B) The AUC of lymph node status for TNBC-associated mortality was was 0.685 and P=0.001. AUC, area under the curve; TNCB, triple-negative breast cancer.

Figure 2. The effect of clinicopathological factors on the OS and DFS rates of triple negative breast cancer patients that underwent adjuvant chemotherapy with fluorouracil, epirubicin and cyclophosphamide. Prognostic significance was assessed using Kaplan-Meier survival estimates and log-rank tests. Graphs compare the OS rates according to (A) all patients, (B) the stage of disease (P=0.001) and (C) lymph node status (P<0.001). Graphs compare the DFS rates according to (D) all patients, (E) the stage of disease (P=0.005) and (F) lymph node status (P<0.001). OS, overall survival, DFS, disease-free survival.

Table I. The clinicopathological and demographical factors of triple-negative breast cancer patients with adjuvant fluorouracil, epirubicin and cyclophosphamide chemotherapy.

Table I.

The clinicopathological and demographical factors of triple-negative breast cancer patients with adjuvant fluorouracil, epirubicin and cyclophosphamide chemotherapy.

	Overall survival				Disease-free survival
	Univariate		Multivariate		Univariate		Multivariate
Variables	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
Age	0.861 (0.430–1.725)	0.673		0.473	1.308 (0.613–2.793)	0.488		0.757
Menopausal status	1.225 (0.603–2.489)	0.574		0.369	0.836 (0.399–1.754)	0.636		0.974
Number of pregnancies	1.246 (0.613–2.533)	0.544		0.841	2.454 (1.149–5.239)	0.020		0.077
Family history	1.094 (0.438–2.731)	0.848		0.825	0.921 (0.350–2.423)	0.868		0.667
Tumor histology	0.767 (0.183–3.220)	0.717		0.863	0.412 (0.056–3.034)	0.384		0.454
Adjuvant radiotherapy	1.394 (0.422–4.601)	0.585		0.704	2.066 (0.716–5.959)	0.179		0.574
Primary surgery	1.411 (0.380–5.244)	0.607		0.246	0.708 (0.106–4.717)	0.721		0.924
pT stage	1.254 (0.657–2.393)	0.493		0.848	1.238 (0.626–2.450)	0.539		0.957
pN stage	1.996 (1.465–2.720)	<0.001	1.996 (1.465–2.720)	<0.001	1.824 (1.315–2.531)	<0.001	1.824 (1.315–2.531)	<0.001
TNM stage	2.901 (1.640–5.132)	<0.001		0.540	2.442 (1.357–4.395)	0.003		0.375

[i] The variables were used due to the prognostic significance. For univariate analyses, a forward, stepwise selection (forward likelihood ratio) was used and P<0.050 indicated a statistically signifcant difference. For multivariate analyses, a Cox proportional hazards regression model was used. HR, hazard ratio; CI, confidence interval; pT, pathological assessment of primary tumor size pN, pathological assessment of regional lymph nodes; TNM, pathological assessment of anatomic stage.

Table II. Correlation between the pT, pN and TNM stages, and the clinicopathological and demographic characteristics of patients with triple-negative breast cancer patients.

Table II.

Correlation between the pT, pN and TNM stages, and the clinicopathological and demographic characteristics of patients with triple-negative breast cancer patients.

		pT classification, n				pN stage, n					TNM stage, n
Variables	Patients, n	T1	T2	T3	Correlation (P-value)	N0	N1	N2	N3	Correlation (P-value)	I	II	III	Correlation (P-value)
All cases	239
Age					−0.100 (0.122)					0.078 (0.232)				0.031 (0.632)
<48 years	109	34	69	6		67	30	6	6		27	67	15
≥48 years	130	52	74	4		74	28	16	12		37	64	29
Menopausal status					0.051 (0.431)					−0.085 (0.191)				−0.048 (0.463)
Premenopausal	130	44	80	6		61	24	3	11		34	77	19
Postmenopausal	109	42	63	4		80	34	9	7		30	54	25
Number of pregnancies					0.146 (0.024)a					0.177 (0.006)b				0.241 (<0.001)b
≥2	95	26	64	5		47	25	12	11		16	52	27
≤1	144	60	79	5		94	33	10	7		48	79	17
Family history					0.023 (0.725)					0.042 (0.519)				0.046 (0.479)
Yes	40	15	24	1		25	9	6	0		13	20	7
No	199	71	119	9		116	49	16	18		51	111	37
Tumor histology					−0.011 (0.871)					−0.016 (0.807)				−0.006 (0.925)
Ductal adenocarcinoma	219	86	133	0		129	53	19	18		59	119	41
Non-ductal adenocarcinoma	20	0	10	10		12	5	3	0		5	12	3
Adjuvant radiotherapy					−0.021(0.750)					0.199 (0.002)b				0.122 (0.060)
Yes	19	7	12	0		6	4	6	3		5	5	9
No	220	79	131	10		135	54	16	15		59	126	35
Primary surgery					−0.186 (0.004)b					−0.041 (0.529)				−0.147 (0.023)a
Modified radical mastectomy	227	77	140	10		133	55	22	17		57	127	43
Conservative surgery	12	9	3	0		8	3	0	1		7	4	1

a P<0.05

b P<0.01. pT, pathological assessment of primary tumor size; pN, pathological assessment of pathological regional lymph nodes; TNM, pathological assessment of anatomic stage.

Results

The demographic and clinical characteristics and the treatment options of the patients are presented in Table I. The average age of patients was 48.3±9.4 years (median, 49.0 years; range, 26.0–76.0 years). The average follow-up time was 80.1±30.4 months (range, 9.5–144.1 months). The number of pre-menopausal patients was 130 patients, 54.4% of the total. Of the patients, 6 (2.5%) suffered bone metastases and 25 patients (10.5%) suffered visceral metastases. The lymph node metastatic rate of patients with tumor sizes of <2 cm was 27.9% (24/86); however, for tumor sizes of >2 cm, the rate was as high as 48.4% (74/153).

The results of univariate analysis are shown in Table I. The univariate analysis showed that the lymph node status (P<0.001) and stage of disease (P<0.001) were significantly associated with the OS time. The OS time showed no significant difference for age (P=0.673), menopausal status (P=0.574), number of pregnancies (P=0.544), family history (P=0.848), tumor histology (P=0.717), adjuvant radiotherapy (P=0.585), primary surgery (P=0.607) or tumor size (P=0.493). The DFS was significantly associated with the lymph node status (P<0.001) and stage of disease (P=0.003), but not with the other factors. In the multivariate statistical analysis, the significant independent prognostic variable affecting survival, including OS and DFS time, was lymph node status rather than the stage of disease, despite the stage of disease being a well-characterized independent prognostic factor (Table I)

The prognostic value of lymph node status on TNBC recurrence and mortality was assessed using ROC analysis. The association of lymph node status and tumor recurrence or mortality was identified. The cut-off points for OS and DFS in the study population were each 0.5. The area under the curves (AUCs) of the lymph node status for TNBC recurrence and mortality were 0.676 (P=0.002; 95% CI, 0.580–0.791) and 0.685 (P=0.001; 95% CI, 0.565–0.788), respectively (Fig 1.).

The OS rates of patients following diagnosis were 97.5, 92.1 and 71.1% at 12, 36 and 60 months, respectively; the DFS rates were 95.0, 88.3 and 69.0%, respectively (Fig. 2). The Kaplan-Meier survival curves stratified for lymph node status and stage of disease are exhibited in Fig. 2. Patients with N3 or stage III disease tended to demonstrate a shorter OS and DFS compared with patients with N0-2 (OS, P<0.001; DFS, P<0.001) or stage I–II disease (OS, P=0.001; DFS, P=0.005) (Fig. 2.).

The tumor size, lymph node status and stage of disease were all associated with the number of pregnancies. Additionally, the lymph node status and stage of disease were associated with primary surgery, and the lymph node status was associated with adjuvant radiotherapy (Table II).

Discussion

The results of the present study demonstrated that age (45.6 vs. 54.4%) and menopausal status (54.4 vs. 45.6%) did not significantly affect the incidence of TNBC in China. This is not in accordance with the findings of previous studies, which report the prevalence of TNBC among non-African female breast cancer patients as between 10 and 17% and report that TNBC is increased in menopausal females compared with pre-menopausal African females (17–19). Numerous studies have demonstrated that premenopausal African-American females were more prone to develop TNBC (5,9,17,18). Carey et al reported that the morbidity rate of the TNBC subtype among African-American breast cancer patients <50 years old is as high as 39%, whereas among Caucasian females and post-menopausal African-American females, the morbidity rate is 16 and 14% respectively (5,9,18).

TNBC is prone to local recurrence and distant metastasis. In the present study, 6 patients (2.5%) suffered bone metastases and 25 patients (10.5%) suffered visceral metastases. TNBC has an increased risk of local recurrence or distant metastasis following the final diagnosis (5,17,18,20–22). In the present study, the general disease progression rate is 12.1% (29/239, the local recurrence rate is 6.7% (16/239) and the distant metastasis rate is 8.8% (21/239). These findings suggest that distant metastasis may exhibit a certain organ tendency in TNBC (23–25) and that the specific target organ metastasis may be associated with specific gene expression (26–28). Dent et al observed that the frequency of distant metastasis was significantly increased among patients with TNBC compared with non-TNBC patients (33.9 and 22.4%, respectively), and the risk of distant metastasis was found to be increased in the TNBC group (relative risk = 2.6) (18). There was a gradual increase in the risk of distant metastasis in the TNBC group, with a peak in the 2nd and 3rd years (18), followed by a rapid decline, with a lower risk in the 5th year and no distant metastasis in the 8th year of follow up (29). Previous studies also reported that TNBC was associated with an increased risk of visceral metastasis and a decreased risk of bone metastasis (30,31). These results are similar to those indicated in the present study.

The variation in the tumor size was not of notable importance in patients with TNBC that possessed no distant metastasis, and had received adjuvant FEC chemotherapy. The benefit of cyclophosphamide, methotrexate and 5-fluorouracil (CMF) chemotherapy to patients with triple-negative, node-negative breast cancer is notable (32). One previous study indicated that there was no additional benefit associated with the cyclophosphamide, epirubicin and 5-fluorouracil (CEF) regimen over CMF, suggesting that non-anthracycline regimens may be sufficient in intrinsic subtypes; however, additional studies are required (33).

Therefore, the lymph node status, not stage of disease, was regarded as an independent prognostic indicator for OS and DFS. For this reason, the present study considered that the patients with large numbers of lymph node metastasis may not be suitable for FEC adjuvant chemotherapy, but may receive taxane-containing chemotherapy regimens. A previous study reported that patients with node-positive breast cancer responded better to docetaxel compared with fluorouracil (34).

In total, 40 of the 239 TNBC patients (16.7%) in the present study had a family history of breast cancer, which was not significantly increased compared with the non-TNBC subgroup total of 13% reported by Bhatti et al (35). Haffty et al concluded that TNBC exhibits an increased proportion of positive family history of breast cancer (36). Zhang et al reported that, in China, there was no statistically significant difference in the family history of TNBC and non-TNBC groups (37).

For tumor sizes of <2 cm, the rate of lymph node metastasis was 27.9% (24/86), compared with 48.4% for tumor sizes of >2 cm (74/153). Therefore, tumor size was not indicated to be associated with lymph node metastasis, which is in agreement with several studies (17,36,38). The present study, which assessed the association between tumor size and lymph node metastasis, produced varying results. Kandel et al demonstrated that 50% of patients with TNBC developed lymph node metastases when the average tumor size of TNBC was 2 cm (39). However, Haffty et al suggested that tumor size was not associated with lymph node metastasis (36).

Additionally, in the first year after diagnosis in the present study, the DFS rate for patients with TNBC was 95.0% (227/239), which was increased compared with that in the non-TNBC group of another study (18). This finding may be attributed to the indication that TNBC appears to be more sensitive to chemotherapy compared to non-TNBC (40). Therefore, patients with TNBC may achieve increased short-term DFS rates. In brief, TNBC patients have a poorer OS time and tend to relapse sooner compared with patients with other breast cancer subtypes (41). The present study reports that lymph node status is an effective prognostic parameter for TNBC patients, particularly for those that exhibit the N3 stage of disease; however, the effect of stages of disease is decreased. Therefore, a larger sample size is required in order to verify the results of the present study. In summary, the present study presents evidence that lymph node status may predict the prognosis of TNBC patients receiving FEC adjuvant chemotherapy in China. Patients with N0-2 may obtain the most benefit from FEC.

Acknowledgements

The present study was supported by Harbin Medical University Cancer Hospital (grant no., JJZ2011-02), Heilongjiang Provincial Department of Education Project (grant no., 11541140) and Heilongjiang Provincial Bureau of Health (grant no., 2009-012).

Glossary

Abbreviations

Abbreviations:

References