MAPK overexpression is associated with anthracycline resistance and increased risk for recurrence in patients with triple-negative breast cancer

Transcription

1 Annals of Oncology 19: , 2008 doi: /annonc/mdm522 Published online 15 November 2007 MAPK overexpression is associated with anthracycline resistance and increased risk for recurrence in patients with triple-negative breast cancer Y. Eralp 1 *, D. Derin 1, Y. Ozluk 2, E. Yavuz 2, N. Guney 1, P. Saip 1, M. Muslumanoglu 3, A. Igci 3, S. Kücücük 4, M. Dincer 4, A. Aydiner 1 & E. Topuz 1 1 Department of Medical Oncology, Institute of Oncology; 2 Department of Pathology, Istanbul Medical Faculty; 3 Department of General Surgery, Istanbul Medical Faculty; 4 Department of Radiation Oncology, Institute of Oncology, Istanbul University, Istanbul, Turkey Received 23 July 2007; revised 7 October 2007; accepted 9 October 2007 Background: Triple-negative breast cancer is estimated to account for 15% 20% of all patients with breast cancer and is considered as a prognostically unfavorable subset. The aim of this study is to evaluate the prognostic impact of various molecular factors in patients with triple-negative breast cancer. Patients and methods: Tumor specimens from 109 patients with receptor-negative (estrogen receptor and progesterone receptor) breast cancer were analyzed for mitogen-activated protein kinase (MAPK), epidermal growth factor receptor (EGFR) and phosphoinositol-3-kinase (PI3K) expression by immunohistochemistry. The prognostic significance of these molecular factors, in addition to various prognostic variables, was investigated. Results: Fifteen (13.8%), 38 (34.9%) and 33 patients (30.3%) had positive staining for EGFR, MAPK and PI3K, respectively. MAPK was associated with anthracycline resistance (P = 0.008) and lower MAPK score was significantly associated with shorter disease-free survival (P = 0.029). Survival following relapse was significantly worse for those with a higher MAPK score (P = 0.03). Conclusion: MAPK is a significant prognostic and predictive factor in patients with triple-negative breast cancer. Furthermore, the level of staining among those with a positive MAPK expression may play a prognostic role at different stages of relapse. Further translational research is required to elucidate molecular mechanisms of tumor proliferation in this subset of patients. Key words: chemoresistance, EGFR, MAPK, prognosis, triple-negative breast cancer original article introduction Triple-negative breast cancer, which accounts for 15% of all patients with breast carcinoma, is defined as the subset that does not express hormone receptors and Her-2/neu. The bulk of data indicate that this subgroup of patients may have a poorer prognosis than others with hormone receptor-positive or Her-2/neu-positive disease [1]. Despite the fact that Her-2/ neu is an established poor prognostic factor in breast cancer, lack of Her-2/neu overexpression in conjunction with hormone receptors limits the use of modern agents that would improve the efficacy of standard chemotherapy. As a result, treating triple-negative breast cancer is a challenge for the physician because it is resistant to many effective therapeutic approaches that have resulted in improved survival rates over the last couple of decades. The lack of useful treatment options have led investigators to identify molecular pathways that are involved in the proliferation of these tumors. Elucidating these *Correspondence to: Dr Y. Eralp, Department of Medical Oncology, Institute of Oncology, Istanbul University, Topkapi, Istanbul, Turkey. Tel: ; Fax: ; yeralp@yahoo.com molecular mechanisms will eventually lead to the development of specific targeted agents, which will hopefully improve the outcome. Potential pathways that have been indicated to play a role in the intracellular signaling process of triple-negative tumors are the activated epidermal growth factor receptor (EGFR) mitogen-activated protein kinase (MAPK) and the Akt/PI3K (phosphoinositol-3-kinase) pathways, as well as a possibly defective DNA repair mechanism driven by p53 and BRCA1 mutations [2 4]. MAPK is a major signal-transducing family, consisting of three key cascades, namely Raf-1, extracellular regulated kinase (ERK)-1 and -2 and p38 MAPK molecules, which couple signals from growth factor or steroid receptors to intracellular transcription factors, leading to expression of genes regulating apoptosis and cellular proliferation. ERK is the most relevant of all, for intracellular mechanisms involved in breast cancer. This pathway is activated mostly due to overexpression of upstream molecules such as EGFR and mutations involving various genes within the cellular cascade, including Ras, Raf, PI3K or Akt. In addition, cross talk between PI3K/Akt pathway and posttranslational activation of ª The Author Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 proapoptotic molecules lead to regulation of apoptosis, depending on the type of stimulus and cell. Furthermore, the Raf/MAP kinase kinase/erk pathway has been shown to induce resistance to doxorubicin and paclitaxel through ectopic activation of Raf in breast cancer cells [5, 6]. Further data generated by modern gene expression profiles have demonstrated a specific subtype of triple-negative tumors, characterized by EGFR or cytokeratin 5/6 overexpression. The so-called basal-like subtype, which is observed in 67% 85% of specimens analyzed indicates that not all triple-negatives are alike, at least in a molecular sense [2, 7, 8]. In summary, triple-negative breast cancer is a heterogenous and poorly characterized subtype with conflicting data on outcome. Furthermore, gene profiling by microarray analysis is not routinely available in many centers. Therefore, in this retrospective study our primary aim was to evaluate and compare the outcome of our patients with triple-negative and basal-like tumors, as assessed by EGFR overexpression by immunohistochemistry. Secondly, we tried to elucidate the prognostic role of two major intracellular pathways (EGFR MAPK and PI3K) involved in tumor proliferation in this cohort with nonmetastatic triple-negative breast cancer. patients and methods patient selection This study included 109 patients who were admitted and treated in our clinic from January 1997 to December None of the patients were metastatic at initial presentation. The data on hormone receptors assessed by routine immunohistochemical staining were obtained through pathology reports. Medical reports were also reviewed to retrieve clinical information on demographics, treatment details and outcome. Next, a total of 106 available archival tissue blocks obtained from these patients during surgery were retrieved and analyzed for expression of EGFR, PI3K and MAPK by immunohistochemistry. immunohistochemistry analysis Immunohistochemical staining was carried out using standard streptavidin biotin-peroxidase method on 3 5 lm thick tissue sections. Then, the sections were incubated with primary antibodies against Her-2/neu (Clone 10A7, Novocastra, Newcastle, UK), PI3K [Clone p85a (B-9), SantaCruz Biotechnology, Europe], EGFR (PharmDx EGFR manual kit, Clone 2-18C9, DAKO, Denmark) and MAPK [Clone p-erk (E-4), Cat No sc-7383, 1/100, 60 min, SantaCruz Biotechnology, Wembley, UK]. Following antibody staining, the sections were incubated with biotinylated goat anti-mouse immunoglobulin (Lab Vision Corp., Biotinylated Goat Anti-Polyvalent, Fremont, CA) followed by labeling with streptavidin-horseradish peroxidase solution (Lab Vision Corp., Large Volume Streptavidin Peroxidase). assessment of expression by immunohistochemistry. Immunostaining for EGFR, PI3K and MAPK was considered positive when at least 10% of tumor cells showed strong membranous staining. A negative Her-2/neu expression was defined as cases that displayed no membraneous staining (negative) or those that either had some staining in <10% of tumor cells or had weak to moderate staining (1+). Those who had moderate staining in >10% of cells (2+) were further evaluated by FISH to determine actual expression level. Intracellular staining for MAPK was evaluated by both a positive/negative assessment on the basis of the percentage of stained cells and a semiquantitatively scoring system according to the percentage of cells stained positive and the degree of staining within each cell. The scoring system was on the basis of grading with respect to the ratio of stained cells to tumor cells counted (1: 1% 10%, 2: 11% 33%, 3: 34% 66% and 4: 67% 100%) and the intensity of staining (0: none, 1: weak, 2: moderate and 3: strong). By adding these to grades, an expression score ranging from 0 to 7 was obtained. Scores from 0 to 3 were considered low expression, whereas, 4 7 were accepted as high expression level. statistical analysis Chi-square and Fisher s exact tests were used to evaluate possible associations between covariates (age younger or older than 50 years, grade 2 versus 3, stages I II versus III, lymph node involvement positive versus negative, staining for Her-2/neu, PI3K and MAPK positive versus negative) and prognostic outcomes such as recurrence, distant metastases and death. Univariate correlations between prognostic variables and survival outcomes were carried out using the Kaplan Meier method. Variables were also evaluated for independent correlations on survival by Cox regression analysis. Overall survival (OS) was calculated as the time from the date of diagnosis to the date of death or last contact. Disease-free survival (DFS) was calculated as the time from the date of diagnosis to the date of initial relapse or the date of last follow-up evaluation. SPSS software (SPSS version , SPSS Inc., Chicago, IL) was used for all statistical evaluations. results patient characteristics Median age of the cohort was 47, ranging between 28 and 81. Fifty-nine patients were premenopausal at presentation. The majority of patients had invasive ductal carcinoma (84.4%). Forty-two patients (38.5%) had family history of cancer. The majority (n = 21, 50%) had a first- or second-degree relative with lung or laryngeal carcinoma and eight (19%) had a firstdegree relative with breast or ovarian cancer. There was a significantly higher ratio of patients with a family history among those who were younger than 50 years (P = 0.029). Patient characteristics are listed in Table 1. Systemic chemotherapy was given to 102 patients (93.6%). All but 13 (11.9%) received anthracycline-based regimens as Table 1. Patient characteristics Annals of Oncology Patient characteristics n % Grade II III Stage I II III Nodal involvement Negative Positive Positive lymph nodes, n Percentages may not add up to 100 because of missing data. n, number of patients. 670 Eralp et al. Volume 19 No. 4 April 2008

3 Annals of Oncology adjuvant chemotherapy. Taxanes were not available for adjuvant treatment at that time and used as second-line treatment after relapse. After a median follow-up period of 53 months (7.5 96), 31 patients had relapse (28.4%). Major organ involvement including liver, lungs and brain were detected in 61% (n = 19) of patients as initial sites of recurrence. Thirteen patients (11.9%) developed recurrence within 12 months after anthracycline-based adjuvant treatment, thus defined as having anthracycline-resistant disease. At the final follow-up examination, 68 patients (62.4%) are alive with no evidence of disease, while 10 had (9.2%) relapse and 20 (18.3%) had died due to disease progression. Eleven patients were lost to follow-up. molecular factors Immunohistochemistry evaluations for EGFR, PI3K and MAPK are summarized in Table 2. There was no correlation observed between molecular factors and disease characteristics such as age, grade, stage or lymph node involvement. MAPK overexpression, however, was found to be correlated with anthracycline resistance (P = 0.008). Moreover, among those with MAPK overexpression, a lower score was associated with a higher recurrence rate (P = 0.036). Only in patients with node-negative disease, EGFR expression was correlated with MAPK expression (P = 0.027). survival DFS and OS at 5 years were 69.3% 6 4.8% and 82.2% 6 4.2%, respectively. Median survival is not reached for either end point Table 2. Summary of results for expression of molecular factors by immunohistochemistry Molecular factors n % EGFR Negative Positive PI3K Negative Positive MAPK Negative Positive MAPK% a MAPK score Analyses were carried out on 106 available tumor blocks. Percentages may not add up to 100 due to technical errors resulting in lack of immunoreactivity. a Ratio of tumor cells with immunoreactivity for MAPK staining. EGFR, epidermal growth factor receptor; PI3K, phosphoinositol-3-kinase; MAPK, mitogen-activated protein kinase. original article at this time. Univariate analyses revealed nodal involvement, anthracycline resistance and MAPK score as significant prognostic factors for DFS, while recurrence and anthracycline resistance remained the only significant variables for OS (Table 3). Neither EGFR nor PI3K expression were observed to have an impact on either survival end point at this time. Median survival after recurrence was [95% confidence interval (CI) ] months in patients who developed relapse from breast cancer. Patients with a positive staining for MAPK showed an insignificant trend for inferior outcome compared with others without MAPK activation. Among those who had MAPK overexpression, a higher score was significantly associated with lower survival following relapse compared with those with a lower expression score (Table 3). Cox regression analysis revealed anthracycline resistance as the only independent prognostic factor for OS [hazard ratio Table 3. Various prognostic factors with an influence on OS, DFS and survival after initial recurrence: univariate analysis Factor n OS 5 years, % (6SD) P OS Anthracycline resistance Resistant 13 2 years, 64.6 (14.3) < Sensitive 80 5 years, 89.9 (4.1) Recurrent disease Relapse (10.0) < No evidence of disease a DFS Anthracycline resistance Resistant 13 1 years, 30.8 (12.8) < Sensitive 80 5 years, 77.3 (5.2) Stage I (9.1) 0.06 II (6.0) III (9.7) Nodal involvement Negative (6.1) 0.05 Positive (6.8) MAPK score (11.9) (8.4) OS after initial recurrence Median 6 SD (95% CI) MAPK Negative ( ) NS Positive ( ) MAPK score ( ) b Significant variables were stratified with respect to potential confounding factors such as age, grade, stage or nodal involvement. a Survival rate could not be computed because there were no events. b Survival rate could not be computed since both patients were dead in 6 months. OS, overall survival; DFS, disease-free survival; SD, standard deviation; MAPK, mitogen-activated protein kinase; CI, confidence interval; NS, not significant. Volume 19 No. 4 April 2008 doi: /annonc/mdm

4 (HR) 15.8, P = 0.014; 95% CI ]. Multivariate analyses for DFS is not included since the number of events in the anthracycline-sensitive group (n = 1) is not enough to allow for meaningful comparison. Nevertheless, when anthracycline resistance is removed from the model (nodal involvement, disease extent locally advanced and early stage and MAPK expression included), MAPK emerges as the only independent factor with a HR: 4.7 (P = 0.048). discussion Recent advances in genetic profiling have led identification of distinct prognostic groups in breast cancer, namely, luminal A and B, normal breast-like, Her-2/neu-positive and basal-like subtypes. Although triple-negative phenotype as determined by the lack of expression of estrogen receptor (ER) and progesterone receptor (PR) and Her-2/neu by immunohistochemistry is considered as a surrogate for basal-like breast tumors, the term basal like by definition requires a positive immunostaining for basal cytokeratins or EGFR [1, 3]. Due to growing interest in this subgroup, we investigated the outcome of our patients with triple-negative and basal-like tumors, as assessed by those with a positive expression for EGFR by immunohistochemistry within the triple-negative subtype. Among 109 patients with triplenegative tumors, EGFR immunoreactivity was observed in 13.8% of cases, which is quite lower than previous studies reporting expression rates in the range of 37% 48% [9, 10]. Although, racial difference may account for this indiscrepancy, technical errors may also be a confounding factor as storage conditions may affect quality of staining and unautomated immunohistochemistry is amenable to interobserver variation. Thus, it is possible that the use of archival material in our study may have led to errors in evaluating the protein expression. Nevertheless, the majority of our patients with triple-negative breast carcinoma had high-grade, ductal tumors and had a predilection for visceral recurrence as well, which are in accordance with the predefined clinical characteristics of basallike tumors [1, 11 15], indicating that the basal-like and null subtypes (ER, PR, Her-2/neu and EGFR negative) share common clinical features. Furthermore, there was a striking incidence of family history seen in 50% of patients. It has been previously reported that familial BRCA-1-mutant tumors overlap significantly with triple-negative tumors and share a common cytogenetic features, such as p53 mutations and c-myc amplification [1]. Despite the prior belief that EGFR expression is linked to poor prognosis [3, 16 18], there is an emerging data indicating that triple-negative patients show similar relapse rates and outcome [8]. A similar clinicopathological overview by Kim et al. [9] shows that patients with basal-like tumors show a similar survival rate compared with other subtypes including hormone receptorpositive tumors and a better outcome than those with receptornegative and Her-2/neu-positive tumors. Parallel to these data, the OS of our patients with triple-negative or basal-like tumors are not significantly different compared with hormone receptor-negative tumors with Her-2/neu overexpression, despite improved DFS, when adjusted for stage, grade, age and nodal involvement (data not shown). Although Annals of Oncology EGFR expression does not seem to confer a poorer outcome in our patient group, the number of patients with a positive EGFR staining is too small to make comparisons between subgroups; therefore, the data should be interpreted with caution. Considering the aggressive nature of this distinct patient group with a high proliferative capacity and limited therapeutic options, our patients have shown a favorable outcome with 82% surviving through 5 years. Nevertheless, survival following initial relapse is poor (median 18 months), reflecting the inherent aggressive course, as 60% of patients developed visceral metastatic disease at initial relapse, and the lack of effective treatment options, especially in anthracycline-resistant patients. Anthracycline resistance was shown to play a significant role in tumor recurrence and was determined as an independent poor prognostic factor for OS in our patients with triple-negative tumors. Although not reaching statistical significance limits due to small sample size, patients with resistance to anthracycline-based adjuvant chemotherapy had shorter survival following initial relapse (13 versus 21 months). Since anthracyclines are not given as secondline treatment for relapse, a possible explanation is that anthracycline resistance is a potential predictive factor for chemoresistance in general or that the molecular-deriving mechanisms leading to chemoresistance are similar. In an attempt to elucidate the molecular pathways involved in anthracycline resistance and outcome, we have found that MAPK expression was in fact correlated with anthracycline resistance (P = 0.008). Furthermore, the level of expression seemed to outline different stages of disease progression, as lower expression score was correlated with a higher risk of initial recurrence (P = 0.036), and lower DFS (51% versus 87% at 5 years, P = 0.029). On the other hand, higher scores were predicted for decreased survival after initial recurrence (P = 0.03). In the latter group, the numbers, however, are too small to be conclusive about any existing relationship. Intracellular signaling through the Ras MAPK pathway has been observed in a wide range of breast tumors and has been linked to nongenomic estrogen-mediated tumor growth, and induction of ER-negative phenotype, in addition to resistance to hormonal agents, such as tamoxifen [5, 19 21]. MAPK overexpression has also been associated with growth factorrelated and anchorage-independent tumor proliferation by increased heat shock protein expression in triple-negative tumors [22]. In our group, we have observed MAPK expression in 28% of primary tumors, which is slightly higher than previously reported rates ranging between 17% and 24% [23, 24], which include both hormone receptor-negative and - positive cases. The higher ratio of MAPK expression in our patients with hormone receptor-negative tumors is in concordance with in vitro data, indicating that active MAPK signaling is correlated with ER negativity and MAPK activation results in induction of receptor-negative phenotype [25, 26]. In addition, parallel to our findings, increased MAPK signaling has been reported to be associated with lower DFS, although not an independent factor per se [27 29]. Furthermore, Esteva et al. [24] have proposed a scoring system for MAPK expression and indicated that higher level of expression predicts shorter relapse-free survival in nodepositive patients with Her-2/neu-positive tumors. In our study, 672 Eralp et al. Volume 19 No. 4 April 2008

5 Annals of Oncology we used a slightly more detailed scoring system and found that the level of staining among those with a positive MAPK expression may play a prognostic role at different stages of relapse. The discrepancy between these observations may be related to patient groups included. Despite the fact that 56% of the patient group included by Esteva et al. [24] was ER positive, our group comprised a more homogenous group of receptornegative tumors. In fact, it has been shown previously that intracellular activity of the Ras/MAPK pathway involves complex interactions between other signaling pathways such as the PI3K Akt cascade and results in the generation of both proliferative and apoptotic activity through competing mechanisms [6]. Furthermore, hyperactivation of MAPK activity has also been shown to result in cell cycle arrest and apoptosis [6]. These data support our findings that a lower level of activity is associated with chemoresistance and deteriorated survival since hyperactivation reflected by a score of immunoreactivity would be expected to be associated with increased cell death and delayed recurrence. Nevertheless, these suggestions are purely speculative as early recurrence may not only imply anthracycline resistance but also reflect an aggressive disease course with inherent poor prognosis. Therefore, it remains to be determined whether augmented MAPK activity is a prognostic factor for outcome or is the consequence of the cellular response to increased tumor proliferation in uncontrolled disease setting. Excluding a trend for association with survival after initial recurrence, active MAPK expression, however, did not seem to have an impact on any survival end point in our patient group. Nevertheless, our finding are in parallel to previous data and a recent in vitro study by Small et al. [30], who have shown that increased MAPK phosphatase 1 expression which is induced in response to activated MAPK signaling results in chemoresistance to alkylating agents and doxorubicin in breast cancer tumor models. Furthermore, in a microarray-based evaluation by Sotiriou et al. [16], the basal-like subtype had two distinct gene profiles: one showed high expression of topoisomerase II alpha in addition to multiple molecules, which is a target for anthracyclines and the other subgroup overexpressed caveolin in conjunction with various intracellular molecules and growth factors, which is associated with MAPK activation. These findings lend support to a specific subgroup within basal-like tumors that show inherent anthracycline resistance through topoisomerase II alpha downregulation and MAPK activation. Our findings indicate that anthracycline resistance is the most important prognostic factor for survival in patients with triple-negative breast cancer. In the light of these data discussed above, we propose that MAPK expression may play an important role in the generation of chemoresistance, possibly outlining a distinct molecular subgroup among triple-negative tumors and that different molecular mechanisms may be responsible for tumor proliferation before and after initial recurrence in this subgroup, reflecting selection of more aggressive clones as tumors grow and metastatic dissemination occurs. Although the small sample size precludes any definite conclusions to be drawn, this data may be considered as hypothesis generating and should be subject to prospective validation. references original article 1. Cleator S, Heller W, Coombes RC. Triple-negative breast cancer: therapeutic options. Lancet 2007; 8: Korsching E, Packeisen J, Agelopoulos K et al. Cytogenetic alterations and cytokeratin expression patterns in breast cancer: integrating a new model of breast differentiation into cytogenetic pathways of breast carcinogenesis. Lab Invest 2002; 82: Nielsen TO, Hsu FD, Jensen K et al. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 2004; 10: Foulkes WD, Steffansson IM, Chappuis PO et al. Germline BRCA-1 mutations and a basal epithelial phenotype in breast cancer. J Natl Cancer Inst 2003; 95: Santen RJ, Song RX, McPherson R et al. The role of mitogen-activated protein (MAP) kinase in breast cancer. J Steroid Biochem Mol Biol 2002; 80: McCubrey JA, Steelman LS, Abrams SL et al. Roles of the RAF/MEK/ERK and PI3K/PTEN/AKT pathways in malignant transformation and drug resistance. Adv Enzyme Regul 2006; 46: Siziopikou KP, Cobleigh M. The basal subtype of breast carcinomas may represent the group of breast tumors that could benefit from EGFR-targeted therapies. Breast 2007; 16: Hafty BG, Yang Q, Reiss M et al. Locoregional relapse and distant metastasis in conservatively managed triple negative early-stage breast cancer. J Clin Oncol 2006; 24: Kim MJ, Ro JYR, Ahn SH et al. Clinicopathologic significance of the basal-like subtype of breast cancer: a comparison with hormone-receptor and Her-2/neuoverexpressing phenotypes. Hum Pathol 2006; 37: Rakha EA, El-Sayed ME, Green AR et al. Prognostic markers in triple-negative breast cancer. Cancer 2007; 109: Van de Rijn M, Perou CM, Tibshirani R et al. Expression of cytokeratins 17 and 5 identifies a group of breast carcinomas with poor clinical outcome. Am J Pathol 2002; 161: Banarjee S, Reiss-Filho JS, Ashley S et al. Basal-like breast carcinomas: clinical outcome and response to chemotherapy. J Clin Pathol 2006; 59: Livasy CA, Karaca G, Nanda R et al. Phenotypic evaluation of the basal-like subtype of invasive breast carcinoma. Mod Pathol 2006; 19: Osbourne CR, Kannan L, Asfaq R et al. Clinical and pathological classification of basal-like breast cancer. Breast Cancer Res Treat 2005; 94 (Suppl 1): S Rodriguez-Pinilla SM, Sarrio D, Honrado E et al. Prognostic significance of basallike phenotype and fascin expression in node-negative invasive breast carcinomas. Clin Cancer Res 2006; 12: Sotiriou C, Neo SY, McShane LM et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci USA 2003; 100: Sorlie T, Tibshirani R, Parker J et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA 2003; 100: Sorlie T, Perou CM, Tibshirani R et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001; 98: Jeng MH, Shupnik MA, Bender TP et al. Estrogen receptor expression and function in long-term estrogen deprived human breast cancer cells. Endocrinology 1998; 139: Shim WS, Conaway M, Masamura S et al. Estradiol hypersensitivity and mitogenactivated protein kinase expression in long-term estrogen-deprived human breast cancer cells in vivo. Endocrinology 2000; 141: Salh B, Marotta C, Mattewson C et al. Investigation of the MEK-MAP kinase-rsk pathway in human breast cancer. Anticancer Res 1999; 19: Moyano JV, Evans JR, Chen F et al. Alpha-B-crystallin is a novel onco-protein that predicts poor clinical outcome in breast cancer. J Clin Invest 2006; 116: Esteva FJ, Hortobagyi GN, Sahin AA et al. Expression of erbb/her receptors, heregulin and P38 in primary breast cancer using immunohistochemistry. Pathol Oncol Res 2001; 7: Volume 19 No. 4 April 2008 doi: /annonc/mdm

6 Annals of Oncology 24. Esteva FJ, Sahin AA, Smith TL et al. Prognostic significance of phosphorylated P38 mitogen-activated protein kinase and Her-2 expression in lymph nodepositive breast carcinoma. Cancer 2004; 100: Oh AS, Lorant LA, Holloway JN et al. Hyperactivation of MAPK induces loss of ERalpha expression in breast cancer cells. Mol Endocrinol 2001; 15: Creighton CJ, Hilger AM, Murthy S et al. Activation of mitogen-activated protein (MAP) kinase in estrogen receptor alpha-positive breast cancer cells in vitro induces an in vivo molecular phenotype of estrogen receptor negative human breast tumors. Cancer Res 2006; 66: Gee JMW, Robertson JFR, Ellis IO, Nicholson RI. Phosphorylation of ERK1/2 mitogen-activated protein (MAP) kinase is associated with poor response to antihormonal therapy and decreased patient survival in clinical breast cancer. Int J Cancer 2001; 95: Mueller H, Flury N, Eppenberger-Castori S et al. Potential prognostic value of mitogen-activated protein (MAP) kinase activity for disease-free survival of primary breast cancer patients. Int J Cancer 2000; 89: von Lintig FC, Dreilinger AD, Varki NM et al. Ras activation in human breast cancer. Breast Cancer Res Treat 2000; 62: Small GW, Shi YY, Higgins LS, Orlowski RZ. Mitogen-activated protein (MAP) kinase phosphatase-1 is a mediator of breast cancer chemoresistance. Cancer Res 2007; 67: Eralp et al. Volume 19 No. 4 April 2008