Breast cancer, neoadjuvant chemotherapy and residual disease

Clin Transl Oncol (2010) 12:461-467 DOI 10.1007/s12094-010-0538-0 E D U C AT I O N A L S E R I E S Blue Series MOLECULAR AND CELLULAR BIOLOGY OF CANCER Breast cancer, neoadjuvant chemotherapy and residual disease Mariana Chávez-MacGregor · Ana María González-Angulo Received: 14 April 2010 / Accepted: 24 May 2010 Abstract Neoadjuvant systemic therapy (NST) has become part of the standard treatment of patients with locally advanced breast cancer. Patients who achieve a pathologically complete response (pCR) after NST have improved outcomes compared with patients with residual disease at the primary tumor site or the lymph nodes. Achieving a pCR after NST correlates with improved disease-free and overall survival; therefore the amount of residual disease is a prognostic predictor, and it is an area of ongoing research. In this article, we review the literature on NST to highlight the importance of pCR as a prognostic indicator. We also review the definition of pCR and describe the association between different patient and tumor characteristics, including the breast cancer subtype classification, and its response to chemotherapy. We expand on the clinical impact of residual disease and comment on the importance of quantifying it and the current treatment recommendations for patients with residual disease after NST. Keywords Neoadjuvant chemotherapy · Residual disease · Residual cancer burden · Breast cancer M. Chávez-MacGregor Division of Cancer Medicine The University of Texas MD Anderson Cancer Center Houston, USA A.M. González-Angulo (쾷) Departments of Breast Medical Oncology and Systems Biology The University of Texas MD Anderson Cancer Center 1515 Holcombe Boulevard Unit 1354 TX 77030-4009, Houston, USA e-mail: agonzalez@mdanderson.org Introduction Breast cancer is the second most common cause of cancer death in the United States; in 2008, more than184,000 new cases were diagnosed [1]. According to data from the Surveillance Epidemiology and End Results (SEER) program, approximately 30% of invasive breast cancer cases have nodal involvement at the time of diagnosis [2, 3]. Women with node involvement and particularly those with locally advanced breast cancer (LABC) require multidisciplinary therapy to optimize patient care. Preoperative or neoadjuvant systemic chemotherapy (NST) is part of the standard of care in patients with LABC. It has several advantages, including increased rates of conservation surgery and administration of chemotherapy through an intact vascular system. Having a primary tumor during administration of systemic chemotherapy has the advantage of response evaluation. Also, several clinical trials have evaluated the impact of using additional chemotherapy in patients with residual disease and outcome. Patients who achieve a pathologically complete response (pCR) during NST have improved outcomes compared with patients with residual disease at the primary tumor site or the lymph nodes. With that in mind, several clinical trials have tried additional chemotherapy in patients in the latter group in order to improve outcomes. However, no consistent improvement has been observed with a midcourse chemotherapy switch based on the presence or lack of response to NST. Therefore, treatment plan should be devised at the outset and should not be altered unless there is clear evidence of disease progression. In this article, we review the literature on NST to highlight the importance of pCR as a prognostic indicator. We describe and analyze the association between different tumor subtypes and their response to chemotherapy and expand on the direction of ongoing research focusing on residual disease. 462 Clin Transl Oncol (2010) 12:461-467 Fig. 1 Evidence of prognostic importance of axillary lymph node status after neoadjuvant chemotherapy. Reprinted with permission. ©2008 American Society of Clinical Oncology. All rights reserved. Henessy BT, et al: J Clin Oncol 23 (36):9304-9311 Neoadjuvant chemotherapy: pathologically complete response (pCR) and residual disease NST is part of the standard treatment for patients with LABC [4] and is increasingly being used in patients with operable breast cancer. NST improves the rates of breast conservation and allows clinical monitoring of in vivo tumor responses. It has the theoretical advantages of early initiation of systemic therapy, delivery of drugs through an intact vasculature, and the opportunity to study the biological effects of chemotherapy in vivo. However, the use of NST does result in the alteration of standard and well-validated pathological prognostic markers, such as tumor size and number of lymph nodes involved. Most importantly, the survival rates of women treated with adjuvant and neoadjuvant chemotherapy are considered to be equivalent [5]. Clinical tumor responses are observed in 70–90% of patients who receive NST, with rates varying according to the type of chemotherapy and number of cycles [6]. Tumor responses obtained after NST correlate with outcome. The correlation between pCR and long-term disease-free (DFS) and overall (OS) survival has been confirmed in virtually all clinical trials and is present regardless of tumor subtype and initial clinical stage of disease [5, 7]. pCR, therefore, has been adopted as a reliable intermediate marker of longterm prognosis and is used as an endpoint for neoadjuvant trials. It is generally accepted that the definition of pCR should include patients without residual disease in the breast (pT0). However, the presence of nodal metastases, minimal residual invasive cancer, and residual in situ carcinoma in not well defined. Our group at The University of Texas, MD Anderson Cancer Center, defines pCR as no evidence of invasive carcinoma in the breast or lymph nodes. Clin Transl Oncol (2010) 12:461-467 463 Fig. 2 Evidence of the prognostic effect of ductal carcinoma in situ (DCIS) in the residual pathological specimen after neoadjuvant chemotherapy. Reprinted with permission. ©2008 American Society of Clinical Oncology. All rights reserved. Mazouni C, et al: J Clin Oncol 25(19):2650-2655 It has been described that when there is no residual invasive cancer in the breast, the number of involved axillary lymph nodes is inversely related to survival. Conversely, patients who convert to a node-negative status after NST have excellent survival, even if there is residual disease in the breast [7–9] (Fig. 1). Including patients with positive lymph nodes in the definition of pCR weakens its prognostic value. Also, there is no evidence that residual in situ carcinoma increases the risk of relapse or has any effect on clinical outcome [10, 11] (Fig. 2). Therefore, patients with residual carcinoma in situ should not be included in the definition of pCR. As discussed, pCR is associated with improved outcomes. However, when evaluating tumor response to chemotherapy, the dichotomization of responses as pCR or residual disease is simplistic, because the patients categorized as having residual disease include a broad range of actual responses to frank resistance [12]. Symmans et al. [13] developed a new model to quantify residual disease. They estimated the residual cancer burden (RCB) as a continuous variable derived from the primary tumor dimensions, tumor-bed cellularity, and axillary nodal burden. The authors noted that patients had almost a twofold increase in relapse risk for each unit of increase in the RCB index, and when the RCB index was included in a model that included clinical and treatment covariates, the overall predictive power of the model improved. When RBC was categorized (RCB-0, RCB-I, RCB-II, RCB-III) different risk groups were identified, with women having RCB-0 and RCB-1 having excellent prognosis (Fig. 3). RCB is a new independent risk factor that improves the prediction of distant relapse after NST. RCB is a new tool for pathologists and is a novel method for reporting findings from posttreatment specimens; it is important, as it represents a new method to evaluate and understand residual disease. Given the clinical relevance of residual disease, different trials have used response as a decision-making aid. The first prospective randomized trial using this strategy was the Aberdeen trial [14] in which 162 patients were treated preoperatively with four cycles of an anthracycline-based regimen. After 12 weeks of treatment, 68% of patients who had a clinical response were randomly assigned to either four more cycles of the same regimen or four cycles of docetaxel. All patients who did not have a clinical response were treated with docetaxel. The change to docetaxel in the responders increased the clinical response rate after eight cycles from 68% to 94% and the pCR rate from 16% to 34%. However, the pCR rate in nonresponders was only 2%, and these patients had statistically significantly worse progression-free survival (PFS) at 2 years than did responders. Thomas et al. [15] evaluated the use of alternate, non-cross-resistant adjuvant chemotherapy based on pCR after the administration of three cycles of an anthracyclinebased chemotherapy. In this study, patients with <1 cm3 residual tumor at the time of surgery received five additional cycles of the same regimen; those with residual tumors >1 cm3 were randomized to the same or an alternate regimen. Among patients with residual tumors >1 cm3, those receiving the alternate, non-cross-resistant regimen seemed to have higher DFS and OS; however, this difference did not achieve statistical significance. In the recently published GeparTrio trial [16], patients received two 3-week 464 Clin Transl Oncol (2010) 12:461-467 Fig. 3 Prognostic effect of different residual cancer burden (RCB) categories after neoadjuvant chemotherapy Reprinted with permission. ©2008 American Society of Clinical Oncology. All rights reserved. Symmans WF, et al: J Clin Oncol 25(28):4414-4422 cycles of docetaxel (75 mg/m2), doxorubicin (50 mg/m2), and cyclophosphamide (500 mg/m2) (TAC). Patients whose tumors did not decrease in size by at least 50% were randomly assigned to receive four additional cycles of TAC or four cycles of vinorelbine (25 mg/m2) and capecitabine (2,000 mg/m2) (NX). Of the 2,090 patients enrolled, 622 (29.8%) were classified as nonresponders. Sonographic response rate was 50.5% for patients in the TAC arm and 51.2% for patients in the NX arm. Similar rates of pCR (5.3% vs. 6.0%) and breast-conserving surgery (57.3% vs. 59.8%) were seen. Based on the results of such clinical trials, no consistent improvement in outcome has been observed with a midcourse chemotherapy switch based on the presence or lack of response to NST. Therefore, treatment plans should be devised at the outset and should not be altered unless there is clear evidence of disease progression. In a recently published retrospective analysis aimed at identifying progression predictors, Caudle et al. reported that from 1,928 patients, 59 (3%) had evidence of progression while receiving NST. Factors predictive of disease progression included African American race, advanced tumor stage, high nuclear grade, high Ki-67, and estrogen/progesterone receptor (ER/PR) negative status [17]. Patients with evidence of disease progression during preoperative treatment should be switched to an alternate regimen, offered local therapy, or considered for investigational approaches. To date, no trial has shown that additional chemotherapy given to patients with residual disease improves outcome, and there is no role for postoperative chemotherapy if a full course of standard chemotherapy that includes a taxane and anthracycline regimen for 6 months was completed preoperatively [14, 16, 18] unless it is administered as part of a clinical trial. Tumor subtypes and response to treatment It is known that several patient and tumor characteristics at diagnosis correlate with tumor response to treatment and with the probability of recurrence or death in patients with breast cancer. Factors associated with adverse prognosis include the extent of disease and the involvement of regional lymph nodes, high nuclear grade, and increased proliferative fraction [19–21]. Differences also occur according to histological subtype. Invasive lobular carcinomas are more likely to be hormone-receptor positive and have lower nuclear grade than invasive ductal carcinomas. The pCR rate seems to be significantly lower in patients with invasive lobular carcinoma [22]. However, the overall outcome of these tumors after adjusting for patient and tumor characteristics is better than the outcome for invasive ductal carcinomas [22]. Despite achieving a pCR, differences in outcome according to initial stage have been reported. After evaluating 489 patients with pCR, Dawood et al. [23] reported that clinical stage at diagnosis was associated with improved RFS and OS estimates. Overall, patients who achieved a pCR had a low rate of recurrence; however, those with stage IIIB/IIIC disease had worse outcomes than patients with stage I/II/II- Clin Transl Oncol (2010) 12:461-467 IA despite all having achieved pCR after NST [23]. ER/PR positivity correlates with prolonged DFS and OS, although hormonal receptors are much better predictors of the utility of endocrine therapies than prognosis. Human epidermal growth factor receptor 2 (HER-2)/neu has emerged as an important prognostic factor associated with lymph node involvement and relative resistance to hormonal therapy and certain types of chemotherapy [24–26]. It is well known that breast cancer is a heterogeneous collection of diseases with different biology. The use of gene-expression profiling has led to the discovery of different molecular subtypes that have phenotypic diversity with regard to multiple clinical outcomes [27, 28]. The molecular subtypes include at least two luminal subtypes (A and B) characterized by a high expression of hormonereceptor-related genes; the basal-like subtype characterized by a high expression of genes common to the breast myoepithelium, high expression of proliferation genes, and low expression of hormone receptor and HER-2 signatures; and the HER-2/neu-enriched type [29–31]. The basal-like and the HER-2-enriched subtypes have shown the poorest prognosis, with decreased DFS and OS. Patients with luminal A breast cancer have better prognosis compared with all other groups, and the luminal B subtype seems to have an intermediate outcome. Luminal B tumors, although expressing hormone-receptor–related gene signature, do so at a lower level than luminal A tumors and have a variable expression of the HER-2 signature; therefore, they are more proliferative than the luminal A subtype [28–32]. It is important to mention that there is no standardized and uniformly accepted molecular assay to assign molecular class to breast cancer. The original intrinsic subtype predictor has undergone important methodological changes in each subsequent publication. The genes used for classification, the prediction algorithm, and the reference or training population vary from reference to reference. Because of the limited availability of microarray expression analysis, immunohystochemical (IHC) markers have been used as a surrogate. This simplification has the limitation that the prognostic power of the different breast cancer subtypes is based on a complex gene expression signature. The marker combinations that best matched the molecular profiles segregate tumors into four types: ER+ and/or PR+, HER-2/neu normal for luminal A; ER+ and/or PR+, HER-2/neu+ or high Ki-67 for luminal B; ER–, PR–, HER-2/neu normal, cytokeratin 5/6+, and/or epidermal growth factor recetptor (EGFR)+ for basal-like type and ER–, PR– and HER-2/neu+ for the HER-2/neuenriched subtype [33–35]. These subtypes overlap largely with luminal A, B, basaloid and HER-2-like molecular subtypes, but the overlap is only about 70–80%. Different groups have explored the effect of molecular subgroups or breast cancer subtypes and treatment response. In a retrospective study evaluating the prognostic value of pCR in relation to hormone receptor status, Guarneri et al. [36], observed that from 1,731 patients treated with NST, a pCR was observed in 24% of patietns with hornone-receptor-negative status and only in 8% of those 465 with hormone-receptor-positive tumors (p<0.001). Patients who achieve a pCR after NST have improved OS regardless the tumor subtype [36–38]. However, it has been observed that the rate of pCR differs considerably: the luminal A subtype has low rates of pCR; in contrast, basal-like and HER-2-enriched tumors have much higher rates of pCR [33, 36, 39, 40]. The luminal B subtype was associated with an intermediate response rate. The poor prognosis and inferior survival rates of patients with basal-like and HER-2/ neu-enriched tumors appears to contrast with the high pCR rates observed in these patients [36, 37]. The unfavorable outcomes seen in patients with basal-like and HER-2/neuenriched tumors are caused by a higher frequency of recurrence and death in those who did not achieve a pCR [37, 38]. It has been demonstrated that tumors with a high expression of proliferation-related genes are associated with a high pCR rate after NST; however, this may also increase the capacity of the tumors to metastasize [41]. There is variability between breast cancer subtypes and the degree of chemotherapy sensitivity. Different groups have developed various methods to estimate the probability of treatment response. Predictive models have been developed using clinical variables, cell lines with known drug sensitivities, and supervised analysis of gene expression data of human cancers. Recently, Lee et al. [42] compared the performance of different established models used to predict pCR to NST. Using a cohort of 100 breast cancer patients treated with anthracycline and taxane-based NST, the authors observed that the optimized cell-line-derived predictor [in vitro coexpression extrapolation (COXEN)] [43] was not predictive and that a clinical normogram [44] and a human-cancer-derived genomic predictor (DLDA30) [45] had similar performances [area under the curve (AUC) 0.73]. The in vivo COXEN that used informative genes from cell lines but was trained on a human data set also had a predictive value (AUC 0.67). All predictors had lesser performance than seen in the original reports. Intense research on gene expression, complementary DNA (cDNA) microarrays, and proteomics might provide predictive tools of greater value in the not too distant future. Results from ongoing research are promising; however, this technology is still investigational and its results are not used routinely to determine clinical practice. Molecular characterization of residual disease There is a growing interest within the scientific community in understanding the biology and characteristics of residual disease. Guarneri et al. [46] evaluated the effect of NST on tumor biomarker expression and the prognostic role of treatment-induced variation of such biomarkers. After evaluating tumor samples before and after treatment, the authors observed that NST induced a significant reduction in the expression of ER, PR, Ki-67, and apoptosis. Ki-67 15% and nodal positivity after treatment were predictors 466 of DFS. Patients with those two parameters had a 9.3- and 6.5-fold increase in risk of recurrence and death compared with patients who had no lymph node positivity and had Ki-67 <15. The previously described observation has been confirmed by a different group. Jones et al. [47] also found that after adjusting for patient and tumor characteristics, Ki-67 expression after NST was significantly associated with RFS and OS. These studies suggest that Ki-67 is a strong predictor of outcome for patients with residual disease after NST. Our group at MD Anderson evaluated whether patients with HER-2 overexpressed tumors have a change in HER-2/neu status in the residual disease after receiving trastuzumab-based NST [48]. Despite the small sample size, the authors observed that loss of HER-2 status was present in one third of the patients, and that this change was associated with poor RFS [48]. Creighton et al. [49] explored whether tumor cells surviving after conventional NST were enriched for cells that have high-initiating potential or are capable of self-renewal. Using gene expression signature, the authors confirmed increased expression of mesenchymal markers. This suggests that residual tumor cells that survive after conventional treatment may be enriched for subpopulations of cells with both tumorinitiating and mesenchymal features. It is possible that as we change our treatment modalities and include novel targeted agents in the standard NST regimens we will also modify the biology and characteristics of the residual disease. We consider that it is fundamental to understand the determinants of residual disease. Efforts at performing molecular determination of residual disease will allow us to understand patters of resistance and hopefully develop new targets and improve outcomes in breast cancer patients. Important research is ongoing in an attempt to determine the survival mechanisms of breast cancer after NST. Recommendations and future directions According to recent guidelines [4], there is no role for postoperative chemotherapy if a full course of standard chemotherapy that includes a taxane and anthracycline regimen for 6 months was completed preoperatively unless it is administered as part of a clinical trial. There is no data to confirm that a midcourse chemotherapy switch based on the presence or lack of response to NST leads to improved outcomes. Therefore, treatment plans should be devised at the outset and should not be altered unless there is clear evidence of disease progression. Several clinical trials are randomizing patients with residual disease to receive new therapeutic agents; recruiting is ongoing, and results are eagerly awaited, but this approach remains investigational. Our group at MD Anderson is enrolling patients with HER-2-negative tumors on a randomized phase II study of ixabepilone (40 mg/m2 every 21 days for 6 cycles) versus observation (NCT00877500). Eli- Clin Transl Oncol (2010) 12:461-467 gible patients must have received complete anthracyclineand taxane-based NST and have significant residual disease in the surgical specimen. The German Breast Group is evaluating the use of zoledronic acid in the adjuvant setting in a phase III clinical trial (NCT00512993). A total of 654 patients with stage II and III breast cancer with residual disease in the breast and/or lymph nodes after NST are being randomized to receive zoledronic acid (4 mg every 4 weeks for six doses, followed by every 3 months for eight doses, followed by every 6 months for five doses) versus observation. The ABCDE trial (NCT00925652), a phase III randomized study of adjuvant bevacizumab, metronomic chemotherapy, diet, and exercise after NST in patients with residual disease. The trial will randomize 660 patients into four arms: diet; diet and exercise and bevacizumab (every 3 weeks for 12 months), in combination with metronomic chemotherapy with cyclophosphamide (orally, once daily for 6 months) and methotrexate (twice daily on the first 2 days of each week) with or without diet and exercise intervention. The CIBOMA/2004-01 is another ongoing trial. This study from the Grupo Espanol de Investigacion en Cancer de Mama is a phase III randomized study that will evaluate the use of capecitabine as maintenance versus observation after chemotherapy in 876 patients with triplereceptor-negative tumors. Conclusions Despite significant improvements in breast cancer mortality, there are still large numbers of patients that relapse after standard treatment. NST has become part of the standard treatment of patients with locally advanced breast cancer and provides an ideal clinical setting for predictive marker research. It is now well established that selection of chemotherapy regimens should take into account tumor subtype, and clinical trials are attempting to incorporate new drugs to the standard regimens, which could provide improved response rates. Achieving a pCR after NST correlates with good outcome; therefore, the amount of residual disease is a prognostic predictor and should also continue to be a source of research. Identifying markers able to better discriminate different prognostic categories among patients with residual disease will allow selecting candidates for additional therapy. Understanding the biology and the clinical implications of residual disease may help us better tailor our management and improve outcomes. Also, understanding the mechanisms involved in invasion, metastasis, survival, and resistance will help us identify and develop specific targets for evaluation in clinical trials. Conflict of interest The authors declare that they have no conflict of interest relating to the publication of this manuscript Acknowledgments This work was supported in part by NCI 1K23CA121994-01 and The Susan G. Komen Foundation KG090341 (to AMG) Clin Transl Oncol (2010) 12:461-467 References 1. Jemal A, Siegel R, Ward E et al (2008) Cancer statistics. CA Cancer J Clin 58:71–96 2. American Cancer Society (2009) Breast cancer facts and figures 2007–2008 Atlanta: American Cancer Society. http://www.cancer.org/downloads/ stt/bcff-final.pdf 3. SEER (2009) Cancer statistics review 1975–2006: National Cancer Institute, DC-CPS, Surveillance Research Program, Cancer Statistics Branch. 4. NCCN (2009) Clinical practice guidelines in oncology. Breast cancer V.I: National Comprehensive Cancer Network. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp 5. Fisher B, Bryant J, Wolmark N et al (1998) Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 16:2672–2685 6. Rastogi P, Anderson SJ, Bear HD et al (2008) Preoperative chemotherapy: updates of National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J Clin Oncol 26:778–785 7. Kuerer HM, Newman LA, Smith TL et al (1999) Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol 17:460–469 8. Hennessy BT, Hortobagyi GN, Rouzier R et al (2005) Outcome after pathologic complete eradication of cytologically proven breast cancer axillary node metastases following primary chemotherapy. J Clin Oncol 23:9304–9311 9. Rouzier R, Extra JM, Klijanienko J et al (2002) Incidence and prognostic significance of complete axillary downstaging after primary chemotherapy in breast cancer patients with T1 to T3 tumors and cytologically proven axillary metastatic lymph nodes. J Clin Oncol 20:1304–1310 10. Jones RL, Lakhani SR, Ring AE et al (2006) Pathological complete response and residual DCIS following neoadjuvant chemotherapy for breast carcinoma. Br J Cancer 94:358–362 11. Mazouni C, Peintinger F, Wan-Kau S et al (2007) Residual ductal carcinoma in situ in patients with complete eradication of invasive breast cancer after neoadjuvant chemotherapy does not adversely affect patient outcome. J Clin Oncol 25:2650–2655 12. Pusztai L (2008) Preoperative systemic chemotherapy and pathologic assessment of response. Pathol Oncol Res 14:169–171 13. Symmans WF, Peintinger F, Hatzis C et al (2007) Measurement of residual breast cancer burden to predict survival after neoadjuvant chemotherapy. J Clin Oncol 25:4414–4422 14. Smith IC, Heys SD, Hutcheon AW et al (2002) Neoadjuvant chemotherapy in breast cancer: significantly enhanced response with docetaxel. J Clin Oncol 20:1456–1466 15. Thomas E, Holmes FA, Smith TL et al (2004) The use of alternate, non-cross-resistant adjuvant chemotherapy on the basis of pathologic response to a neoadjuvant doxorubicin-based regimen in women with operable breast cancer: long-term results from a prospective randomized trial. J Clin Oncol 22:2294–2302 16. von Minckwitz G, Kummel S, Vogel P et al (2008) Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase III randomized GeparTrio trial. J Natl Cancer Inst 100:542–551 467 17. Caudle AS, Gonzalez-Angulo AM, Hunt KK et al (2010) Predictors of tumor progression during neoadjuvant chemotherapy in breast cancer. J Clin Oncol 28:1821–1828 18. von Minckwitz G, Kummel S, Vogel P et al (2008) Intensified neoadjuvant chemotherapy in earlyresponding breast cancer: phase III randomized GeparTrio study. J Natl Cancer Inst 100:552–562 19. Early Breast Cancer Trialists’ Collaborative Group (1992) Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet 339(8785):71–85 20. Allred DC, Harvey JM, Berardo M, Clark GM (1998) Prognostic and predictive factors in breast cancer by immunohistochemical analysis. Mod Pathol 11:155–168 21. Nemoto T, Vana J, Bedwani RN et al (1980) Management and survival of female breast cancer: results of a national survey by the American College of Surgeons. Cancer 45:2917–2924 22. Cristofanilli M, Gonzalez-Angulo A, Sneige N et al (2005) Invasive lobular carcinoma classic type: response to primary chemotherapy and survival outcomes. J Clin Oncol 23:41–48 23. Dawood S, Broglio K, Kau SW et al (2008) Prognostic value of initial clinical disease stage after achieving pathological complete response. Oncologist 13:6–15 24. Buzdar AU, Ibrahim NK, Francis D et al (2005) Significantly higher pathologic complete remission rate after neoadjuvant therapy with trastuzumab, paclitaxel, and epirubicin chemotherapy: results of a randomized trial in human epidermal growth factor receptor 2-positive operable breast cancer. J Clin Oncol 23:3676–3685 25. Ravdin PM, Chamness GC (1995) The c-erbB-2 proto-oncogene as a prognostic and predictive marker in breast cancer: a paradigm for the development of other macromolecular markers--a review. Gene 159:19-27 26. Slamon DJ, Clark GM, Wong SG et al (1987) Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235(4785):177–182 27. Huber KE, Carey LA, Wazer DE (2009) Breast cancer molecular subtypes in patients with locally advanced disease: impact on prognosis, patterns of recurrence, and response to therapy. Semin Radiat Oncol 19:204–210 28. Perou CM, Sorlie T, Eisen MB et al (2000) Molecular portraits of human breast tumours. Nature 406(6797):747–752 29. Sorlie T, Perou CM, Tibshirani R et al (2010) Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 98:10869–10874 30. Sorlie T, Tibshirani R, Parker J et al (2003) Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A 100:8418–8423 31. Sotiriou C, Neo SY, McShane LM et al (2003) Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci U S A 100:10393–10398 32. Fan C, Oh DS, Wessels L et al (2006) Concordance among gene-expression-based predictors for breast cancer. N Engl J Med 355:560–569 33. Carey LA, Perou CM, Livasy CA et al (2006) Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295:2492– 2502 34. Nielsen TO, Hsu FD, Jensen K et al (2004) Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 10:5367–5374 35. Cheang MC, Chia SK, Voduc D et al (2009) Ki-67 index, HER-2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst 101:736-750 36. Guarneri V, Broglio K, Kau SW et al (2006) Prognostic value of pathologic complete response after primary chemotherapy in relation to hormone receptor status and other factors. J Clin Oncol 24:1037–1044 37. Carey LA, Dees EC, Sawyer L et al (2007) The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res 13:2329–2334 38. Liedtke C, Mazouni C, Hess KR et al (2008) Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol 26:1275–1281 39. Rouzier R, Perou CM, Symmans WF et al (2005) Breast cancer molecular subtypes respond differently to preoperative chemotherapy. Clin Cancer Res 11:5678–5685 40. Sawaki M, Ito Y, Akiyama F et al (2006) High prevalence of HER-2/neu and p53 overexpression in inflammatory breast cancer. Breast Cancer 13:172–178 41. Gianni L, Zambetti M, Clark K et al (2005) Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer. J Clin Oncol 23:7265–7277 42. Lee JK, Coutant C, Kim YC et al (2010) Prospective comparison of clinical and genomic multivariate predictors of response to neoadjuvant chemotherapy in breast cancer. Clin Cancer Res 1:711–718 43. Lee JK, Havaleshko DM, Cho H et al (2007) A strategy for predicting the chemosensitivity of human cancers and its application to drug discovery. Proc Natl Acad Sci U S A 104:13086–13091 44. Rouzier R, Pusztai L, Delaloge S et al (2005) Nomograms to predict pathologic complete response and metastasis-free survival after preoperative chemotherapy for breast cancer. J Clin Oncol 23:8331–8339 45. Hess KR, Anderson K, Symmans WF et al (2006) Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer. J Clin Oncol 24:4236–4244 46. Guarneri V, Piacentini F, Ficarra G et al (2009) A prognostic model based on nodal status and Ki-67 predicts the risk of recurrence and death in breast cancer patients with residual disease after preoperative chemotherapy. Ann Oncol 20:1193–1198 47. Jones RL, Salter J, A’Hern R et al (2009) The prognostic significance of Ki67 before and after neoadjuvant chemotherapy in breast cancer. Breast Cancer Res Treat 116:53–68 48. Mittendorf EA, Wu Y, Scaltriti M et al (2009) Loss of HER2 amplification following trastuzumabbased neoadjuvant systemic therapy and survival outcomes. Clin Cancer Res 15:7381–7388 49. Creighton CJ, Li X, Landis M et al (2009) Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci U S A 106:13820–13825