Molecular markers of therapeutic resistance in breast cancer

Human Pathology (2013) 44, 1421–1428

www.elsevier.com/locate/humpath

Original contribution

Molecular markers of therapeutic resistance in breast cancer☆,☆☆ Ling Zhou MD, PhD a,1 , Yanli Luo MD, PhD b,1 , Ke Li MD, PhD c,1 , Ling Tian PhD b , Min Wang MD a , Chuanyuan Li ScD d,⁎, Qian Huang MD, PhD c,⁎ a

Department of Surgery, Shanghai First People's Branch Hospital, Shanghai Jiaotong University, Shanghai 200081, China Experimental Research Center, First People's Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200080, China c Department of Surgery, First People's Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200080, China d Department of Dermatology, Medical Center, Duke University, Durham NC 27710, USA b

Received 4 September 2012; revised 5 October 2012; accepted 10 October 2012

Keywords: ALDH1; Breast cancer; Caspase-3; Cox-2; Triple-negative breast cancer

Summary Resistance to chemotherapy and endocrine therapy is a serious obstacle in the treatment of breast cancer. Highly specific biomarkers for predicting therapeutic resistance have not yet been identified. In this study, the amounts of aldehyde dehydrogenase 1, cleaved caspase 3, cyclooxygenase 2, phosphorylated Akt, Ki-67, and H2AX proteins were measured by immunohistochemical staining in 113 breast cancer tissues, and their predictive ability for therapeutic resistance was investigated. The patients were receiving chemotherapy (n = 30), endocrine therapy (n = 22), or combined chemotherapy and endocrine therapy (n = 61). Expression of aldehyde dehydrogenase 1, cleaved caspase 3, and cyclooxygenase 2 correlated significantly with a higher relapse rate (P b .05 or P b .01) and shorter survival (P b .01 or P b .001) in triple-negative patients receiving chemotherapy. In addition, cyclooxygenase 2 expression was an independent predictor of a poor prognosis (P b .05). On the other hand, aldehyde dehydrogenase 1 expression correlated significantly with shorter survival in patients receiving combined therapy (P b .01) but showed no association with relapse. No correlation was observed between Ki-67, phosphorylated Akt, and H2AX expression and survival or relapse in any group of patients. These data suggest that aldehyde dehydrogenase 1, cleaved caspase 3, and cyclooxygenase 2 are useful markers for therapeutic resistance in breast cancer. © 2013 Elsevier Inc. All rights reserved.

☆

Disclosure: All authors declare no conflicts of interest. Funding: This project was supported by grants from the National Natural Science Foundation (81120108017, 81172030), the National Basic Research Program of China (2010CB529902), and Shanghai Hongkou District Public Health Bureau (1001-01). ⁎ Corresponding authors. C. Li, is to be contacted at the Department of Dermatology, Medical Center, Duke University, Durham NC 27710, USA. Tel.: +1 919 613 8754; Q. Huang, Experimental Research Center, First People's Hospital, School of Medicine, Shanghai Jiaotong University, 85 Wujin Road, Shanghai 200080, China. Tel.: +86 21 63240090. E-mail addresses: [email protected] (C. Li), [email protected] (Q. Huang). 1 These authors equally contributed to this work. ☆☆

0046-8177/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humpath.2012.10.027

1422

1. Introduction Despite numerous advances in prevention, surgical technology, and various therapies for breast cancer (BC), an estimated 450 000 patients die from the disease annually [1,2]. The survival rate differs greatly depending on cancer type, disease stage, and treatment. Overall, the 5-year survival rate is approximately 84% in the Western world [3] but much lower in developing countries. BC usually is treated with surgery followed by chemotherapy, radiotherapy, or hormone therapy according to its clinical features and risk of recurrence. For example, the presence of estrogen and progesterone receptors on the cancer cells is an important index for hormone treatments because cancer cells that lack such receptors usually are unresponsive to such therapy [4]. Chemotherapy can be given before surgery (neoadjuvant therapy) or after surgery (adjuvant therapy) [5]. Despite the early efficacy of both chemotherapy and endocrine therapy, BCs may recur and form metastases. Therapeutic resistance is one of the major hurdles to successful treatment. It is widely accepted that therapeutic resistance is caused by a series of complex molecular events that include decreased intracellular drug concentrations mediated by drug transporters and metabolic enzymes; impaired cellular responses that affect cell cycle arrest, apoptosis, and DNA repair; and the induction of signaling pathways that promote the progression of cancer cell populations [6]. In addition, the observation that targeting any single molecule is ineffective against chemotherapy-resistant BC suggests that multiple molecular pathways may be responsible for this resistance [7,8]. Identification of biological markers able to predict the sensitivity of BC cells to chemotherapy or endocrine therapy is important for making treatment decisions because patients with poor tumor sensitivity require more aggressive treatment. A recent study demonstrated that aldehyde dehydrogenase 1 (ALDH1) expression was associated with shorter survival and a poor clinical response to neoadjuvant chemotherapy but had no effect on the outcome of hormone receptor–positive BC [8,9]. However, the predictive role of ALDH1 in adjuvant chemotherapy of BC has rarely been reported, especially in patients receiving combined endocrine therapy and chemotherapy. Cyclooxygenase 2 (Cox-2) is involved in drug and endocrine resistance in BC [10-12]. The chemoresistance role of Ki-67 [13] and phosphorylated Akt (p-Akt) [14] has been found only in cultured BC cells, although Akt has been widely demonstrated to be involved in endocrine resistance [15,16]. Our recent studies suggest that cleaved caspase 3 (CC3) is also a proliferation and differentiation factor in cancer cells [17]. Activated caspase 3 exerts its role through activating calcium-independent phospholipase A2 and arachidonic acid production by paracrine signaling. Arachidonic acid is the chemical precursor of prostaglandin E2, a key regulator of tumor growth. However, the role of CC3 in drug and endocrine resistance has not been reported in BC.

L. Zhou et al. We collected 113 breast cancer samples and categorized them into chemotherapy alone, endocrine therapy alone, or combined therapy (chemotherapy and endocrine therapy) groups. The correlation of CC3, ALDH1, Ki-67, p-Akt, Cox2, and H2AX protein expression with overall survival and relapse rate was analyzed in each group.

2. Materials and methods 2.1. Case selection A total of 113 BC samples were collected from January 2003 to December 2009 at the First People's Hospital, Shanghai Jiaotong University. All patients were female with an average age of 56.3 ± 10.6 years and ranging from 33 to 78 years. The histopathologic subtypes were 89 invasive ductal carcinomas, 8 invasive lobular carcinomas, 4 in situ ductal carcinomas, 4 mucinous carcinomas, 3 medullary carcinomas, and 5 other types, such as poorly differentiated carcinoma and metaplastic carcinoma. According to the TNM staging recommended by the American Joint Committee on Cancer, 37 cases were in stage I; 53, in stage II; and 23, in stage III. Lymph node metastasis was evaluated according to standard criteria [18]. Fifty-five patients had regional lymph node metastases. Survival information was collected for all cases. Fifteen patients died during the follow-up, and metastasis and recurrence were seen in 20 cases. The average follow-up period was 4.6 ± 1.95 years with a median of 4.5 years. All patients underwent a modified radical mastectomy with no chemotherapy or radiotherapy before surgery. Chemotherapy, endocrine therapy, or combined therapy was chosen by reference to the National Comprehensive Cancer Network guidelines. Thirty patients (23 estrogen receptor [ER] negative/progesterone receptor [PR] negative/Her-2− and 7 ER−/PR−/Her2+) were given only chemotherapy, 22 ER+/PR+ patients were given endocrine therapy, and 61 patients (44 ER+/PR+, 17 ER+, or PR+ with or without Her-2+) were given combined treatment. In total, 91 patients received regular chemotherapy (11 received cyclophosphamide, methotrexate, and 5-fluorouracil; 52 received cyclophosphamide, epirubicin, and 5-fluorouracil; 28 received paclitaxel and epirubicin), 23 cases received radiotherapy (7 in the chemotherapy group and 16 in the combined therapy group), and 83 cases received endocrine therapy (75 tamoxifen, 7 letrozole, 1 anastrozole).

2.2. Immunohistochemistry staining Immunohistochemistry staining was performed according to a published protocol [19]. Briefly, 4-μm-thick sections were cut from routinely paraffin-embedded tissues. After deparaffinization, hydration, and antigen retrieval, the sections were incubated overnight at 4°C with primary

Predicting therapeutic resistance in breast cancer

1423

Fig. 1 Immunohistochemistry of protein expression. Representative photomicrographs of positive CC3, Cox-2, ALDH1, Ki-67, H2AX, and p-Akt staining. Positive reactions of CC3, Cox-2, ALDH1, and p-Akt were localized mainly in the cytoplasm with some staining at the membrane. Positive reactions for Ki-67 and H2AX were mainly in the nucleus. Original magnification ×400.

1424

L. Zhou et al.

Table 1 Molecular expression and clinicopathologic characteristics of BC patients Characteristic

ET, n (%)

CT, n (%)

ET + CT, n (%)

Total cases Age ≤50 N50 TNM stage I II/III Tumor size (cm) ≤2 N2 Lymph nodes Negative Positive CC3 Negative Positive ALDH1 Negative Positive Ki-67 Negative Positive Cox-2 Negative Positive p-Akt Negative Positive H2AX Negative Positive

22

30

61

3 (13.6) 19 (86.4)

8 (26.7) 22 (73.3)

30 (49.2) 31 (50.8)

17 (77.3) 5 (22.7)

7 (23.3) 23 (76.7)

13 (21.3) 48 (78.7)

17 (77.3) 5 (22.7)

10 (33.3) 20 (66.7)

32 (52.5) 29 (47.5)

22 (100) 0

14 (46.7) 16 (53.3)

22 (36.1) 39 (63.9)

17 (77.3) 5 (22.7)

19 (63.3) 11 (36.7)

52 (85.2) 9 (14.8)

21 (95.5) 1 (4.5)

22 (73.3) 8 (26.7)

50 (82.0) 11 (18.0)

14 (63.6) 8 (36.4)

7 (23.3) 23 (76.7)

33 (54.1) 28 (45.9)

12 (54.5) 10 (45.5)

21 (70.0) 9 (30.0)

38 (62.3) 23 (37.7)

15 (68.2) 7 (31.8)

11 (36.7) 19 (63.3)

37 (60.7) 24 (39.3)

16 (72.7) 6 (27.3)

14 (46.7) 16 (53.3)

49 (80.3) 12 (19.7)

Abbreviations: ET, endocrine therapy; CT, chemotherapy.

antibodies against ALDH1, p-Akt, Her-2, ER, or PR. The anti-ALDH1 antibody (1:50) was purchased from BD Bioscience (San Jose, CA); antibodies for p-Akt (Ser473) (1:50), CC3 (1:100), Ki-67 (1:100), H2AX (1:100), and Her2 (1:100) were purchased from Cell Signaling Technology (Beverley, MA); and anti-ER (ERα) and anti-PR antibodies (1:100 dilution) were from Dako (Carpinteria, CA). After washing with phosphate-buffered saline, sections were incubated with horseradish peroxidase–conjugated goat antirabbit or antimouse secondary antibody (Cell Signaling Technology; 1:200) for 30 minutes. The immune complex was visualized by 3,3′-diaminobenzidine staining. The results were evaluated by 2 blinded investigators. The strength of CC3, p-Akt, Ki-67, H2AX, ER, and PR staining was scored from 1 to 3: 1 for no staining or uncertain weak staining, 2 for weak to moderate staining, and 3 for moderate to strong staining. The percentage of positive cells was calculated from 10 random fields. Cases with both positive cells of 10% or more and scores of 2 or more were considered positive [19]. In this study, Her-2 staining was scored as 0, +, ++, or +++, and a score of +++ was defined as positive. For

ALDH1 staining, any sample with greater than 1% positive cells was considered positive.

2.3. Statistical analysis All analyses were performed using SPSS v 13.0 (SPSS, Inc, Chicago, IL). Categorical variables were assessed using the χ2 and Fischer exact tests. Univariate analyses of overall survival and survival curves were performed following the Kaplan-Meier method, and the Cox proportional hazards model was used for multivariate analysis. P b .05 was considered statistically significant.

3. Results 3.1. Treatments and their relations to protein expression and clinical characteristics Immunohistochemistry of ER, PR, and Her-2 expression is performed routinely in the clinical laboratory to guide cancer management. Immunohistochemistry staining revealed that positive reactions of CC3, ALDH1, Cox-2, and p-Akt were mainly in the cytoplasm, with some staining at the membrane, whereas positive reactions for Ki-67 and H2AX were mainly in the nucleus (Fig. 1). In BC patients who received endocrine therapy, more than 86% were older than 50 years, 77.3% had stage I tumors, no lymph node metastases were found, and the tumor size in 77.3% was less than 2 cm. In BC patients who received chemotherapy, 73.3% of the patients were more than 50 years old, 76.6% had stage II/III tumors, 53.3% had lymph node metastases, and the tumor exceeded 2 cm in 66.7% of cases. In BC patients who received combined therapy, 50.8% were more than 50 years old, 78.7% had stage II/III tumors, 63.9% had lymph node metastases, and the tumor was greater than 2 cm in 47.5% (Table 1).

Table 2 Univariate and multivariate analyses of overall survival of 30 BC patients who underwent chemotherapy Variable

CC3 ALDH1 Ki-67 Cox-2 p-Akt H2AX

Univariate

Multivariate

P

P

.001 ⁎⁎ .001 ⁎⁎ .516 .000 ⁎⁎⁎ .603 .889

.055 .295 .017 ⁎

RR

95% CI

8.335

0.954, 72.808

13.892

1.590, 121.347

NOTE. Statistical analyses for differences in survival time for positive versus negative expression of each marker. Abbreviation: RR, relative risk. ⁎ P b .05. ⁎⁎ P b .01. ⁎⁎⁎ P b .001.

Predicting therapeutic resistance in breast cancer

1425

3.2. Correlation of protein expression with survival Kaplan-Meier survival analysis revealed that CC3, ALDH1, and Cox-2 correlated significantly with overall survival time (P = .001, .001, and .000, respectively; Table 2, Fig. 2). Specifically, patients with CC3, ALDH1, and Cox-2 expression survived for a shorter time than did patients without expression of these proteins. Cox multivariate analysis showed that Cox-2 was an independent prognostic factor in patients receiving chemotherapy (P = .017; Table 2), and CC3 showed a tendency to be an independent prognostic factor (P = .055). These results suggest that CC3, ALDH1, and Cox-2 expression is associated with drug resistance. Among them, Cox-2 is an independent predictive factor for drug resistance. In contrast, p-Akt, Ki-67, and H2AX exhibited no significant correlation with overall survival time in patients given chemotherapy (Table 2). Among the 22 patients receiving endocrine treatment, no patient had died by the end of the observation period. This finding is not suitable for Kaplan-Meier survival or Cox multivariate analysis. Kaplan-Meier survival analysis revealed that only ALDH1 exhibited significant correlation with overall survival time in patients receiving combined chemotherapy and endocrine therapy (P = .003; Table 3, Fig. 3). Patients with ALDH1 expression survived a shorter time than patients without ALDH1 expression. Cox multivariate analysis showed that ALDH1 was an independent prognostic factor in patients receiving combined treatment (P = .009; Table 2). These results suggest that ALDH1 expression is associated with drug and endocrine resistance.

3.3. Correlation of protein expression with relapse Recurrence is another event reflecting therapeutic resistance in tumors. We analyzed the relapse rate in patients receiving chemotherapy, endocrine therapy, and combined therapy and its associations with molecular markers. Fisher exact test showed no differences in relapse rate between

Table 3 Univariate and multivariate analyses of overall survival of 61 BC patients who underwent chemotherapy and endocrine therapy Variable

CC3 ALDH1 Ki-67 Cox-2 p-Akt H2AX Fig. 2 Survival analysis in BC patients receiving chemotherapy. Kaplan-Meier plots of overall survival in patients receiving chemotherapy with positive or negative CC3, ALDH1, and Cox-2 expression. Significant correlations were observed.

Univariate

Multivariate

P

P

RR

95% CI

.009 ⁎⁎

6.759

1.607, 28.433

.254 .003 ⁎⁎ .343 .979 .996 .279

NOTE. Statistical analyses for differences in survival time for positive versus negative expression of each marker. ⁎⁎ P b .01, difference in survival time between positive and negative ALDH1 expression.

1426

L. Zhou et al. patients with and without expression of CC3, ALDH1, Ki67, p-Akt, Cox-2, or H2AX in both the endocrine therapy and combined therapy groups (data not shown). In patients receiving chemotherapy, expression of CC3, ALDH1, and Cox-2 protein correlated significantly with a high relapse rate (P = .042, P = .003, and P = .008, respectively). In contrast, no relation with relapse was observed in patients with Ki-67, p-Akt, or H2AX expression (Table 4). Cox multivariate analysis showed that CC3 (P = .038) and ALDH1 (P = .005) were independent predictors of relapse in patients receiving chemotherapy (Table 5).

4. Discussion Currently, biological markers have not been identified to predict therapeutic resistance in BC. In this study, a Table 4 Relationship between relapse and clinicopathologic parameters of 30 BC patients who underwent chemotherapy Prognostic factor

Fig. 3 Survival analysis in BC patients receiving combined chemotherapy and endocrine therapy. Kaplan-Meier plots of overall survival in patients receiving combined chemotherapy and endocrine therapy with positive or negative CC3, ALDH1, and Cox-2 expression. Significant correlations were observed in patients with ALDH1 expression.

Age (y) ≤50 N50 TNM stage I II/III Tumor size (cm) ≤2 N2 Lymph nodes Negative Positive CC3 Negative Positive ALDH1 Negative Positive Ki-67 Negative Positive Cox-2 Negative Positive p-Akt Negative Positive H2AX Negative Positive

No. of cases

Relapse

P

Yes

No

8 22

3 6

5 16

.666

7 23

1 8

6 15

.393

10 20

3 6

7 14

1.000

14 16

3 6

11 10

.440

19 11

3 6

16 5

.042 ⁎

22 8

3 6

19 2

.003 ⁎⁎

7 23

1 8

6 15

.393

21 9

3 6

18 3

.008 ⁎⁎

11 19

2 7

9 12

.419

14 16

5 4

9 12

.694

NOTE. Statistical analyses using Fisher exact test for differences in survival time for positive versus negative expression of each protein. ⁎ P b .05. ⁎⁎ P b .01.

Predicting therapeutic resistance in breast cancer Table 5 Univariate and multivariate analysis of disease-free survival of 30 BC patients who underwent chemotherapy Variable

CC3 ALDH1 Ki-67 Cox-2 p-Akt H2AX

Univariate

Multivariate

P

P

RR

95% CI

.008 ⁎⁎ b.001 ⁎⁎⁎ .457 .003 ⁎⁎ .337 .503

.039 ⁎ .005 ⁎⁎

5.481 7.493

1.089, 27.588 1.828, 3.744

.917

NOTE. Statistical analyses for differences in survival time for positive versus negative expression of each marker. ⁎ P b .05. ⁎⁎ P b .01. ⁎⁎⁎ P b .001.

significant association of ALDH1, CC3, and Cox-2 expression with poor prognosis as well as a high recurrence rate was observed in patients with triple-negative BC who were receiving chemotherapy. In addition, ALDH1 was predictive of resistance to combined chemotherapy and endocrine therapy. In this study, patients with 1% or higher ALDH1-positive BC cells who received chemotherapy alone survived a shorter time and had a higher recurrence rate than patients with less than 1% ALDH1-positive cells. This finding suggests that ALDH1 aids in granting drug resistance to BC cells. In contrast, no patient died in the group receiving endocrine therapy alone, nor did cancer recur during the observation period. Therefore, the survival and recurrence rates cannot be used for evaluating the predictive role of these molecular markers in endocrine therapy. Among 22 BC patients receiving endocrine therapy, only 1 patient showed ALDH1 expression. Therefore, our study provided no evidence regarding whether ALDH1 is involved in resistance to endocrine treatment. However, a previous study demonstrated that high ALDH1 expression is not associated with the outcome of endocrine therapy in hormone receptor– positive BC [9]. We noted that ALDH1 expression correlated with a shorter survival time in patients receiving combined chemotherapy and endocrine therapy (n = 61). Therefore, resistance to combined therapy may reflect only resistance to chemotherapy. It is unclear why ALDH1 grants resistance to chemotherapy but not to endocrine therapy if it is not reflective of bias caused by the small sample size. Recently, ALDH1 expression and activity have been used successfully as a marker of cancer stem cells in a variety of cancers, including BC [20,21]. If the chemoresistant cells are indeed cancer stem cells, targeting treatment at these cells would be the most logical way to overcome drug resistance and could improve the outcome of BC treatment. A novel finding in this study was that high concentrations of CC3 correlated significantly with shorter survival time and a high recurrence rate in patients receiving chemotherapy. To our knowledge, this is the first study to report the

1427 role of CC3 in drug resistance in BC patients. Classically, CC3 is an important apoptotic molecule. However, recent studies suggest that CC3 is also a proliferation and differentiation factor in cancer cells [17,19]. Our most recent study also demonstrated that CC3 is a proliferation marker and a predictor of poor prognosis in pancreatic cancer [22]. Therefore, it is not surprising that the proliferative feature of CC3 grants drug resistance to BC cells. For the reasons discussed above, the role of CC3 in resistance to endocrine therapy was not conclusive. Unlike ALDH1, CC3 expression did not correlate with resistance to combined chemotherapy and endocrine therapy, suggesting that CC3 plays a relatively weaker role in drug resistance than does ALDH1. Cox-2, Akt, and Ki-67 are involved in tumor cell proliferation. One study reported that Cox-2 is a key player in genotoxic drug-induced resistance in BC cells [23]. Moreover, Cox-2 inhibitors can prevent or reduce the development of a drug-resistant phenotype in BC cells treated with doxorubicin [10,11]. The role of Cox-2 in drug resistance was proposed to be associated with its roles in cancer cell proliferation, resistance to apoptosis, and tumor angiogenesis [24]. Consistent with previous reports, Cox-2 expression exhibited a significant correlation with overall survival time in BC patients receiving chemotherapy. Furthermore, Cox-2 is an independent predictive factor of poor prognosis in patients receiving chemotherapy. Although Cox-2 expression was detected in 50% of patients receiving endocrine therapy alone and 37% of patients receiving combined treatment, no resistance to endocrine therapy or combined treatment was observed. Akt is activated by PI3K through phosphorylation at its Ser473 residue, and activated Akt stimulates tumor cell proliferation or blocks apoptotic signaling in tumor cells. Recent studies demonstrate that PI3K/Akt signaling plays a crucial role in BC cell survival and low therapeutic efficacy of cisplatin [25]. In this study, no correlation of Akt phosphorylation with the survival of patients receiving chemotherapy or endocrine therapy was observed. The Ki-67 protein is commonly a proliferative marker of tumor cells. However, the role of Ki-67 in multidrug resistance in BC has not been validated [13]. Consistent with previous studies, Ki-67 showed no role in predicting resistance to chemotherapy or endocrine therapy in the current study. H2AX is a histone H2A variant that is essential for DNA repair [26]. The role of H2AX in predicting therapeutic resistance in BC has rarely been reported. In this study, no association between H2AX expression and chemotherapeutic or endocrine efficacy was observed. We acknowledge that a lack of correlation of pAkt, Ki-67, and H2AX expression with therapeutic resistance was revealed in the current study; however, bias attributable to the small sample size cannot be excluded. In conclusion, ALDH1, CC3, and Cox-2 are predictive markers for drug resistance in triple-negative BC. Among them, Cox-2 expression is an independent predictive factor for drug resistance, whereas ALDH1 is a predictive marker for therapeutic resistance to combined chemotherapy and

1428 endocrine therapy. Further studies are necessary to explore whether ALDH1, CC3, and Cox-2 collaborate in mediating drug resistance.

References [1] Gluz O, Liedtke C, Gottschalk N, Pusztai L, Nitz U, Harbeck N. Triple-negative breast cancer—current status and future directions. Ann Oncol 2009;20:1913-27. [2] Gucalp A, Traina TA. Triple-negative breast cancer: adjuvant therapeutic options. Chemother Res Pract 2011;2011:696208. [3] Garcia M, Jemal A, Ward EM, et al. Global Cancer Facts & Figures 2007. Atlanta, GA: American Cancer Society; 2007. [4] Pal SK, Childs BH, Pegram M. Triple negative breast cancer: unmet medical needs. Breast Cancer Res Treat 2011;125:627-36. [5] Downs-Holmes C, Silverman P. Breast cancer: overview & updates. Nurse Pract 2011;36:20-6. [6] Kutanzi KR, Yurchenko OV, Beland FA, Checkhun VF, Pogribny IP. MicroRNA-mediated drug resistance in breast cancer. Clin Epigenetics 2011;2:171-85. [7] Senderowicz AM. Novel direct and indirect cyclin-dependent kinase modulators for the prevention and treatment of human neoplasms. Cancer Chemother Pharmacol 2003;52:S61-73. [8] Gong C, Yao H, Liu Q, et al. Markers of tumor-initiating cells predict chemoresistance in breast cancer. PLoS One 2010;5:e15630. [9] Hashimoto K, Shimizu C, Tsuda H, et al. Immunohistochemical detection of breast cancer stem cells in hormone receptor-positive breast cancer and their role in response to endocrine therapy and clinical outcome. Oncology 2012;82:168-74. [10] Zatelli MC, Molè D, Tagliati F, Minoia M, Ambrosio MR, degli Uberti E. Cyclo-oxygenase 2 modulates chemoresistance in breast cancer cells involving NF-kappaB. Cell Oncol 2009;31:457-65. [11] Zatelli MC, Luchin A, Tagliati F, et al. Cyclooxygenase-2 inhibitors prevent the development of chemoresistance phenotype in a breast cancer cell line by inhibiting glycoprotein p-170 expression. Endocr Relat Cancer 2007;14:1029-38. [12] Falandry C, Debled M, Bachelot T, et al. Celecoxib and exemestane versus placebo and exemestane in postmenopausal metastatic breast cancer patients: a double-blind phase III GINECO study. Breast Cancer Res Treat 2009;116:501-8. [13] Linn SC, Pinedo HM, van Ark-Otte J, et al. Expression of drug resistance proteins in breast cancer, in relation to chemotherapy. Int J Cancer 1997;71:787-95.

L. Zhou et al. [14] Knuefermann C, Lu Y, Liu B, et al. HER2/PI-3K/Akt activation leads to a multidrug resistance in human breast adenocarcinoma cells. Oncogene 2003;22:3205-12. [15] Tokunaga E, Kimura Y, Mashino K, et al. Activation of PI3K/Akt signaling and hormone resistance in breast cancer. Breast Cancer 2006; 13:137-44. [16] Generali D, Fox SB, Brizzi MP, et al. Down-regulation of phosphatidylinositol 3′-kinase/AKT/molecular target of rapamycin metabolic pathway by primary letrozole-based therapy in human breast cancer. Clin Cancer Res 2008;14:2673-80. [17] Végran F, Boidot R, Oudin C, Riedinger JM, Bonnetain F, LizardNacol S. Overexpression of caspase-3s splice variant in locally response to neoadjuvant chemotherapy advanced breast carcinoma is associated with poor response to neoadjuvant chemotherapy. Clin Cancer Res 2006;12:5794-800. [18] Weaver DL. Pathology evaluation of sentinel lymph nodes in breast cancer: protocol recommendations and rationale. Mod Pathol 2010;23: S26-32. [19] Huang Q, Li F, Liu X, et al. Caspase 3–mediated stimulation of tumor cell repopulation during cancer radiotherapy. Nat Med 2011; 17:860-6. [20] Douville J, Beaulieu R, Balicki D. ALDH1 as a functional marker of cancer stem and progenitor cells. Stem Cells Dev 2009;18:17-25. [21] Tanei T, Morimoto K, Shimazu K, et al. Association of breast cancer stem cells identified by aldehyde dehydrogenase 1 expression with resistance to sequential Paclitaxel and epirubicin-based chemotherapy for breast cancers. Clin Cancer Res 2009;15:4234-41. [22] Luo Y, Qiu Z, Tian L, et al. Identification of novel predictive markers for the prognosis of pancreatic ductal adenocarcinoma. HUM PATHOL 2013;44:69-76. [23] Das S, Ongusaha PP, Yang YS, Park JM, Aaronson SA, Lee SW. Discoidin domain receptor 1 receptor tyrosine kinase induces cyclooxygenase-2 and promotes chemoresistance through nuclear factor-kappaB pathway activation. Cancer Res 2006;66:8123-30. [24] Glover JA, Hughes CM, Cantwell MM, Murray LJ. A systematic review to establish the frequency of cyclooxygenase-2 expression in normal breast epithelium, ductal carcinoma in situ, microinvasive carcinoma of the breast and invasive breast cancer. Br J Cancer 2011; 105:13-7. [25] Johnson-Holiday C, Singh R, Johnson EL, Grizzle WE, Lillard Jr JW, Singh S. CCR9-CCL25 interactions promote cisplatin resistance in breast cancer cell through Akt activation in a PI3K-dependent and FAK-independent fashion. World J Surg Oncol 2011;9:46. [26] Fillingham J, Keogh MC, Krogan NJ. GammaH2AX and its role in DNA double-strand break repair. Biochem Cell Biol 2006;84:568-77.