Gene expression profiles do not consistently predict the clinical treatment response in locally advanced breast cancer Therese Sørlie, 1 Charles M. Perou, 3 Cheng Fan, 3 Stephanie Geisler, 5 Turid Aas, 5 Andrew Nobel, 4 Gun Anker, 5 Lars A. Akslen, 6 David Botstein, 7 Anne-Lise Børresen-Dale, 1,2 and Per Eystein Lønning 5 1 Department of Genetics, Institute for Cancer Research, Rikshospitalet-Radiumhospitalet Medical Center; 2 Medical Faculty, University of Oslo, Oslo, Norway; 3 Lineberger Comprehensive Cancer Center and Departments of Genetics and Pathology and Laboratory Medicine and 4 Department of Statistics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; 5 Department of Medicine, Section of Oncology, Haukeland University Hospital; 6 The Gade Institute, Section for Pathology, University of Bergen, Bergen, Norway; and 7 Department of Molecular Biology, 140 Carl Icahn Laboratory, Princeton University, Princeton, New Jersey Abstract Neoadjuvant treatment offers an opportunity to correlate molecular variables to treatment response and to explore mechanisms of drug resistance in vivo . Here, we present a statistical analysis of large-scale gene expression patterns and their relationship to response following neoadjuvant chemotherapy in locally advanced breast cancers. We analyzed cDNA expression data from 81 tumors from two patient series, one treated with doxorubicin alone (51) and the other treated with 5-fluorouracil and mitomycin (30), and both were previously studied for correlations between TP53 status and response to therapy. We observed a low frequency of progressive disease within the luminal A subtype from both series (2 of 36 versus 13 of 45 patients; P = 0.0089) and a high frequency of progressive disease among patients with luminal B type tumors treated with doxorubicin (5 of 8 patients; P = 0.0078); however, aside from these two observations, no other consistent associ- ations between response to chemotherapy and tumor subtype were observed. These specific associations could possibly be explained by covariance with TP53 mutation status, which also correlated with tumor subtype. Using supervised analysis, we could not uncover a gene profile that could reliably (>70% accuracy and specificity) predict response to either treatment regimen. [Mol Cancer Ther 2006;5(11):2914–8] Introduction Resistance to cytotoxic compounds is a main reason for therapy failure in most malignancies, including breast cancer. In vitro experiments as well as studies in animal models have shown that mutations in the TP53 gene are associated with chemoresistance (1). Molecular studies of tumors from patients treated with neoadjuvant chemother- apy using either doxorubicin monotherapy or 5-fluoroura- cil and mitomycin (FUMI) in concert revealed that TP53 mutations affecting the DNA-binding domain of the protein correlate with drug resistance (2–4). However, neither in these tumors nor in the studies reported by others (5) did mutations in TP53 unequivocally predict drug resistance, suggesting that other interactions and genes must be involved (6). By subjecting the same tumors characterized for TP53 mutations in relation to chemotherapy response to DNA microarray analysis, we were able to classify tumors into five distinct subtypes based on their gene expression patterns (7). This classification showed prognostic effect with respect to relapse-free as well as overall survival in our cohort (8) and also in series of patients examined by other investigators (9). The prognostic significance of gene ex- pression profiles has been well documented with respect to breast cancer (10–13) as well as other malignancies (14–17). Although these findings confirm the biological relevance of such genomic analyses, a prognostic factor provides no specific information about responsiveness to specific treat- ments and should be distinguished from a ‘‘predictive factor’’ (18, 19). Knowledge about the value of genome-wide expression analyses in predicting treatment response in breast cancer has resulted in at least to two studies corre- lating gene expression profiles with sensitivity to taxane monotherapy (20, 21) and three studies (22–24) reporting sensitivity to anthracycline combination regimens con- taining either cyclophosphamide or a taxane. However, the predictive powers achieved in any of these studies do not allow clinical implementation without further evaluations. Received 3/7/06; revised 6/4/06; accepted 9/25/06. Grant support: Norwegian Cancer Society, Norwegian Research Council, ‘‘SalusAnsvar’’ Award (A-L. Børresen-Dale), National Cancer Institute Breast Specialized Program of Research Excellence program grant P50-CA58223-09A1, andBreastCancerResearch Foundation(C.M.Perou). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Note: The analyses of the data presented in this report are original work of the authors and have not been presented previously. Some molecular data on these cases have been used in two previous studies (see refs. 3 and 4), and the raw microarray data have been published in a different context (see ref. 9). Requests for reprints: Per Eystein Lønning, Department of Medicine, Section of Oncology, Haukeland University Hospital, N-5021 Bergen, Norway. Phone: 47-55975000; Fax: 47-55973599. E-mail: per.lonning@helse-bergen.no Copyright C 2006 American Association for Cancer Research. doi:10.1158/1535-7163.MCT-06-0126 2914 Mol Cancer Ther 2006;5(11). November 2006 Research. on January 13, 2022. © 2006 American Association for Cancer mct.aacrjournals.org Downloaded from