Breast cacinoma is one of the most general malignant oncology all around the world, and caused a large number of death of women every year. Although the treatment of breast cancer has been developed to a great extent, about 30 % of early breast cancer patients will relapse after adjuvant therapies. Breast cancer is a disease which is heterogeneous and has been identified into distinct molecularly subtypes according to some gene expression studies, the multiple treatment setting to the disease may have implications across prognostic (Oh et al., 2006). Classified by gene profiles, the breast cancer is mainly divided into four subgroups, which included Luminal A, Luminal B, HER2-enriched and basal-like (Sorlie et al., 2001). Certain IHC markers (ER, PR, HER 2 and Ki67 index) can improve the accuracy of predicting prognosis and related treatment information, on the other hand, they are also clinical easily accessible. As a result, these markers have been used in order to determine the subtypes of breast cancer.

Luminal B breast cancer is classified by the characteristic of ER positive and (or) PR positive, HER-2 positive and with any Ki-67 level, or ER positive and (or) PR positive, with HER-2 negative and traditionally Ki-67≥14% by IHC classification. The 14% is regarded as the cutoff point to the identification Luminal A and Luminal B breast cancers in Europe (Goldhirsch et al., 2011). However, recently, the Ki67 index was considered high when 20% or more cells were stained (Tamaki et al., 2014), thus twenty percentage is considered as the cutoff point of Ki67 index in the review, therefore, ER positive, PR positive,HER-2 negative with Ki-67＜20% is the diagnostic code of Luminal A subtype of breast cacinoma, and 1% is the cutoff of positive tumor cells that were nuclear stained (ER positivity).Though PR used to be regarded as a prognostic factor,  its prognostic value is considered to be nonsignificant after long-term follow-up (Bardou et al., 2003). The deficiency of PR expression may be unusual signs of growth factor signaling, and can also be seen as one of the mechanisms of anti-estrogen resistance (Rakha et al., 2007). The breast cancer with R-positive and PR-negative, which was defined in RNA profiles, is distinctive subgroup of breast cancer, the reason is that despite the positive expression of ER in breast cancer of this subgroup, there is a certain degree of invasiveness and poor prognosis (Creighton et al., 2009).

The use of molecular markers, including Ki-67 and p53, to forecast the prognosis of breast cancer patients with ER-positive may contribute to a reasonable explanation for patients with ER-positive but failed endocrine therapy and poor prognosis (Millar et al., 2011). This article focuses on the role of PR, HER-2, Ki 67 and p53 in the breast cancer with ER-positive and PR-negative.

2 Methods

2.1 Patients

Data collection were from the breast cancer patients that received surgical treatment with simple mastectomy、sentinel lymph nodes biopsy or modified radical mastectomy between January 2006 and December 2012.The results of the retrospective study did not affect treatment decision of any patient, and the information from each patient’s medical history were retrieved carefully. All breast cancer specimens after resection were examined pathologically. All the breast cancer is the subtype of infiltrating ductal carcinoma in pathology. The main eliminate criteria were male breast cancer, bilateral breast cancer and other histological types of breast cancer. All of the patients have not accepted any chemical therapy until diagnosed breast cancer.

2.2 Immunohistochemistry and Molecular subtypes

The World Health Organization Classification of Tumors criteria for classification of breast cancer are used. The classification of tumors is based on the United States Joint Committee on Cancer (AJCC) Handbook on Cancer staging (Edge, 2009). Pathological data were used to report histologically, including ER and PR status, HER-2 status and Ki-67 index were determined by immunohistochemical method during operation. We used the monoclonal antibody to determine the status of ER and PR, and HER2 were evaluated immunohistochemically by at least two experienced pathologists. As to HER2 evaluation, following membranous staining was graded: score 0, 1+, 2+and 3+. Scores of 0 and 1+ were defined as no HER-2 amplified or HER-2 negative; A score of 2+ was further subjected to Double color silver stain In Situ Hybridization (DISH) assays to assess gene amplification. When score was 3+, it was defined as HER-2 positive breast cancer and trastuzumab should be used (Wolff et al., 2013). when 20% or more cells were stained, Ki67 was considered to be high. Two pathologists evaluated the stained slides independently.

2.3 Chemotherapy regimens and surgical treatment

There was available treatment information from 320 patients. Taken together, 265 patients were given chemotherapy while another 55 patients received endocrinotherapy alone. The regimens of chemotherapy were determined according to the preference of clinician, TC (Docetaxel 75 mg/m² and Cyclophosphamide 600 mg/m² iv on the first day of every 21 days for 4 cycles); TEC (Epirubicin 60 mg/m² and  Cyclophosphamide 600 mg/m² iv on the first day of every cycle, Docetaxel 75mg/m² IV on the second of every cycle, cycle is 21 days and cycles for 6 times); EC-TH (Epirubicin 60 mg/m² and Cyclophosphamide 600 mg/m² IV on the first day of every  21 days for 4 cycles followed by trastuzumab 8mg/kg IV for the first time on the first day and Docetaxel 75 mg/m² IV on the second day of every 21 days' cycle for 4 cycles, trastuzumab 6mg/kg IV since the second cycle of TH and the following one year of trastuzumab alone) was used for patients with Her-2 amplificated. Schemes of endocrine therapy were determined according to menstrual status of breast cancer patients, tamoxifen, aromatase inhibitors and goserelin were applied to patients as endocrinotherapy up to physical conditions.

2.4 Follow-up

The recurrence of breast cancer was defined as the first time for confirming invasive region recurrence, or distant metastasis of any site. The basic end point of the study were the incidence of local recurrence, distant metastasis, and breast cancer-related death. Local recurrence includes recurrence of the chest wall and the skin of the ipsilateral breast. Distant metastasis is defined as the recurrence of all sites except the location included in the local recurrence. For all first-found recurrences, biopsy is necessary but not mandatory if technical conditions permit. In addition, all the recurrence included in this study were local recurrence or liver metastasis. The diagnosis of local recurrence and distant metastasis was based on clinical manifestation, physical examination and imaging examination. It should be noted in particular that in this study, only the recurrence of the invasive component was considered to be a recurrence, and only the recurrence of the tumor in situ was not considered a recurrence.

2.5 Statistical analysis

Statistical analyses were performed to deal with collected clinical materials. In order to assess the association between clinical categorical and ordinal variables respectively, the Fisher's exact test and the Chi square test were applied. As to determine the associations between Luminal subtypes or PR status and OS, DDFS and LRFS, survival curves were plotted by Kaplan-Meier method and Log-rank test was used. Cox proportional hazards regression modeling was applied to analyze every single factor associated with LRFS (local relapse free survival), DDFS (distant disease-free survival) and OS (overall survival). The difference was statistically significant when P-value＜0.05.

The study group of our study comprised 320 patients with invasive breast carcinoma. 126 of 224 Luminal B breast cancer patients was identified as having a negative PR status in tumors (56.25%). We considered twenty percentage as the cutoff point for PR. Therefore, 96 patients with positive PR status tumor were classified as Luminal A subtype, another 98 patients were classified to the Luminal B subtype. According to the subtype of breast cancers, the characteristics of patients and tumor in this cohort are summarized in Table 1.

Table 1 Luminal Subtype and Clinical Pathologic

Luminal B subtype breast cancers were associated with youth, high expression of P53 protein (≥10%), high histological grade, large tumor size and involvement of axillary lymph nodes. It is shown that Luminal B subtype had a higher percentage (79.1%) of high expression of P53 protein than did Luminal A (20.9%), and the diversity was significant (P＜0.05).Survival analysis for Luminal A and Luminal B subtype was shown in Figure 1, luminal A like breast cancers had a better prognosis including DDFS, LRFS,OS than that of Luminal B like breast cancers.

Figure 1 Survival analysis for different Luminal subtype

Table 2 indicated the characteristics of Luminal B cancers: Ki 67 index ≥20% was defined as positive in this study, it was highly expressed in Luminal B like breast cancer than that in Luminal A like, for there were 178 out of 224 Luminal B patients were positive (Ki67 index ≥20%). Ki-67 index was distinctly associated with PR status (P＜0.05), and it was more properly to have a higher Ki 67 level than in Luminal B like breast cancers with PR positivity (53.4%VS46.6%). 17.3% of Luminal B PR positive breast cancers were with Her-2 overexpressed, while 29.4% were Her-2 enriched in that of Luminal B PR negative breast cancers in spite of insignificance P-value (P=0.05377). There was no significantly disserve between PR status and other clinical factors listed in table 2 except Ki 67 index.

Table 2 PR status and clinical Pathologic in Luminal B patients

It is shown in Table 3 that there is association between PR negativity and DDFS or OS in Luminal B breast cancer patients, while there was obviously correlation between PR status and LRFS, DDFS and OS when we divided all 320 specimens into PR (+) and PR (-) groups as shown in Figure 2. In Luminal B breast cancer patients, the young (≤35) candidates were more likely to result in high risk of local recurrence, distant metastasis and died of disease, the same as other clinical factors exactly as Her-2 overexpression, high Ki 67 index and histological grade, large tumor size and involvement of lymph nodes. Besides, we could find that P53 protein accumulation was also correlated with DDFS, LRFS and OS of Luminal B breast cancers (Table 3).

Table 3 Clinical-Pathologic Characteristics and Outcome in Luminal B patients

Figure 2 Survival analysis for different PR status

As previously mentioned, Luminal B breast cancer was classified by having positive ER and (or) positive PR, positive HER-2 and with any Ki-67 level, or positive ER and (or) positive PR, negative HER-2 and Ki-67 level ≥20% by IHC classification. Among this retrospective study, PR negativity was totally considered as Luminal B breast cancer, therefore, we analyzed subgroup of Luminal B PR negative breast cancers ulterior. Clinical pathologic characteristics and outcome of PR negative patients were demonstrated in Table 4. High expression of P53 protein (≥10%) was related to DDFS and OS, the rest of the clinical factors were entirely associated with DDFS, LRFS and OS of Luminal B PR negative breast cancers.

Table 4 Clinical-Pathologic Characteristics and Outcome in PR negative patients

Table 5 shows that when the Ki-67 was high, it correlated with 118 patients (66.3%) who were with lymph nodes involved; however, when Ki-67 was found at low level, it correlated with only 39 patients (27.5%) whose lymph nodes had also been involved. Related to this correlation between high Ki-67 and lymph node involvement, we can draw a that Luminal like breast cancer patients had a higher involvement of axillary lymph nodes when Ki-67 was at high level.

Table 5 Ki67 status and lymph node

Previous studies have shown that the combination of gene expression profiles with pathological features and clinical manifestations can provide important information for the prediction of different subtypes of breast cancer. In recent years, through the comprehensive analysis of breast cancer, we have deepened the understanding of Luminal B breast cancer and found that it is a unique type of breast cancer different from luminal A or ER negative type. In this study, immunohistochemical markers (ER, PR, HER-2 and Ki 67) were used to classify Lumina A and Lumina B breast cancer. As we all know, Luminal A like breast cancer usually has a favorable outcome, However, some Luminal B like breast cancer represents a heterogeneous disease with dreadful outcome. These Luminal B like breast cancer are a clinically important subgroup associated with poor outcome whether with systemic adjuvant therapy or not, and our study essentially identified good prognosis luminal A cases from poorer prognosis luminal B cases as demonstrated in Figure 1 that any prognostic outcome of Luminal B breast cancer including LRFS, DDFS and OS is much more unfavorable than that of Luminal A.

PR is an estrogen-regulated gene, Estrogen is involved in the synthesis of estrogen in both normal breast cells and breast cancer cells. The expression of ER and PR represented a functional PR pathway (Jacobsen et al., 2005). The loss of PR may be a signal of unusual growth factor signal. It may be one of the mechanisms of anti-estrogen resistance. Several previous clinical studies have shown that molecular crosstalk between ER and growth factor signaling pathway may lead to down-regulation of PR, and membrane activation may cause steroid signal transduction and tamoxifen resistance (Cui et al., 2005; Osborne et al., 2005; Zhang et al., 2009). However, the mechanism of drug resistance remains controversial.

Data of whether PR have prognostic value are somewhat conflicting. The pre-study results of Bardou indicated that (Bardou et al., 2003), PR-negative was the independent prognostic factor for tamoxifen resistance and an independent predictor of the recurrence of breast cancer. Moreover, a retrospective analysis of Cancelello also found tha