Silencing of Irf7 pathways in breast cancer cells promotes bone metastasis through immune escape

Breast cancer metastasis is a key determinant of long-term patient survival. By comparing the transcriptomes of primary and metastatic tumor cells in a mouse model of spontaneous bone metastasis, we found that a substantial number of genes suppressed in bone metastases are targets of the interferon regulatory factor Irf7. Restoration of Irf7 in tumor cells or administration of interferon led to reduced bone metastases and prolonged survival time. In mice deficient in the interferon (IFN) receptor or in natural killer (NK) and CD8(+) T cell responses, metastasis was accelerated, indicating that Irf7-driven suppression of metastasis was reliant on IFN signaling to host immune cells. We confirmed the clinical relevance of these findings in over 800 patients in which high expression of Irf7-regulated genes in primary tumors was associated with prolonged bone metastasis-free survival. This gene signature may identify patients that could benefit from IFN-based therapies. Thus, we have identified an innate immune pathway intrinsic to breast cancer cells, the suppression of which restricts immunosurveillance to enable metastasis.

Tumor PIK3CA mutations, lymphocyte infiltration, and recurrence-free survival (RFS) in early breast cancer (BC): Results from the FinHER trial

507

Background: Tumor PIK3CA mutations (mts) and lymphocyte infiltration (LI) are prognostic in BC, but their importance is unknown in HER2-positive (HER2+) BC treated with adjuvant trastuzumab (T). Methods: The FinHER trial randomized 1010 patients (pts) with pN+ or high-risk pN- BC to 3 cycles of docetaxel (D) or vinorelbine, followed by 3 cycles of FEC. Pts with HER2+ BC were further randomized to 9 weeks of T or no T (n=232). Pts treated with D and T had superior outcome in prior analyses. BC samples were subjected to somatic hotspot mt profiling (Sequenom) and quantification of percentage tumor LI using full-face H&E sections. RFS and interactions with T were explored with Kaplan-Meier and Cox regression analyses. Results: The median FU time was 62 months. 935 (92.6%) and 687 (68.1%) tumors underwent LI and mt analysis, respectively. Correlation of LI assessment between 2 pathologists was 0.78 (p<0.001). 54 mts detected in 13 genes were evaluated. PIK3CA mts (exons 1,2,9,13,18,20) was present in 25.3% (n=174) and ERBB2 mt in 4.5% (n=31) BCs. Only 1 AKT mt was found and none in KRAS, BRAF, NRAS or PTEN. In ER+/HER2-, HER2+ and ER-/HER2- subtypes the prevalence of PIK3CA mts was 30.2%, 19.5% and 9.2%; ERBB2 mt 6.3%, 1.9% and 2.6%, respectively (both p<0.05). Neither PIK3CA nor ERBB2 mts were associated with RFS overall or within BC subgroups. No interaction was found for the benefit with T and presence of PIK3CA mt in HER2+ BC (interaction p=0.17 T vs no T). Increasing LI was associated with favorable RFS in ER-/HER2- BC (continuous score adjusted p=0.032); 3-y RFS was 90% with extensive LI (>30% infiltrated) vs 66% with non-extensive LI (p=0.007). In HER2+ BC, whilst there was no association with prognosis overall, there was a significant interaction with the benefit from T vs. no T and increasing LI (continuous score interaction p=0.042) with 3-yr RFS 96% vs 78% for extensive and non-extensive LI treated with T respectively (p=0.014). Conclusions: We report 3 clinically relevant findings from this large clinical trial dataset: 1) PIK3CA mts were not prognostic; 2) LI is a strong prognostic factor in ER-/HER2- BC and 3) for the first time that extensive LI predicts benefit from adjuvant T.

Progression of genomic signatures in local and metastatic estrogen receptor-positive (ER+) breast cancer: Relevance to palliative treatment

515

Background: Biological progression of ER+ breast cancer accelerates clinical progression and resistance to treatments. Methods: One laboratory used Affymetrix U133A gene expression microarrays to profile 588 biopsy samples from patients with ER+ breast cancer: 74 AJCC Stage I, 155 Stage IIA, 105 Stage IIB, 127 Stage III, 127 Stage IV (27 at presentation, 100 relapsed). We evaluated stage dependence of ER [ESR1, PGR, sensitivity to endocrine therapy (SET) index], proliferation [MKI67, AURKA, genomic grade index (GGI)], invasion [PLAU (uPA)], PI3-kinase (PIK3CA-GS), VEGF, genomic subtype [PAM50, 3-gene classifier (ESR1, ERBB2, AURKA)], and housekeeper control genes. Significance was evaluated through ordinal median regression (P < 0.002, for multiple testing) after adjusting for staging method (clinical or pathologic). Exploratory Cox regression analyses of progression-free survival (PFS) and overall survival (OS) were performed when treatment was hormonal therapy (HT, N=58) or chemotherapy (CT, N=27) after biopsy of metastatic ER+ breast cancer (MBC). Results: Stage progression was associated with reduced SET index and increased proliferation (GGI, MKI67, AURKA) and metabolism (GAPDH). These changes occurred between Stages IIB and III, and Stages III and IV. Luminal B and proliferation subtypes were more prevalent in Stage IV and less in Stage I. Interestingly, invasion (PLAU) genes were lower in MBC. Only SET index demonstrated a significant interaction with treatment (HT or CT) for MBC (PFS: p=0.018). SET was predictive of PFS and OS following HT, as a continuous score (PFS: HR=0.69, 95%CI 0.49 to 0.97, p=0.035; OS: HR=0.61, 95%CI 0.40 to 0.94, p=0.025) or dichotomized at median value (PFS: HR=0.43, 95%CI 0.24 to 0.76, p=0.003; OS: HR=0.37, 95%CI 0.18 to 0.77, p=0.006). Genomic subtype was prognostic for PFS irrespective of treatment type. High PIK3CA-GS expression predicted OS in the HT subset. Conclusions: Stage progression was associated with decreased ER-related transcription (SET) and increased proliferation, grade, higher risk subtype, and metabolism. In MBC samples, only SET index was predictive of PFS and OS with palliative hormonal therapy.

Gene modules and response to neoadjuvant chemotherapy in breast cancer subtypes: a pooled analysis

PURPOSE

To investigate the association between chemotherapy response and gene expression modules describing important biologic processes and druggable oncogenic pathways in breast cancer (BC) subtypes.

PATIENTS AND METHODS

We searched for publicly available gene expression studies evaluating anthracycline with or without taxane-based neoadjuvant chemotherapy and identified eight studies with 996 patients. We computed 17 gene modules and calculated odds ratios (ORs) for pathologic complete response (pCR) for one-unit increases in scaled modules with and without adjustment for clinicopathologic characteristics. Added predictive accuracy was evaluated using the area under the receiver operating characteristic curve (AUC) and integrated discrimination index (IDI). We used the false discovery rate (FDR) to adjust for multiple testing.

RESULTS

High immune module scores were associated with increased pCR probability in all BC subtypes. High module scores of chromosomal instability, phosphatase and tensin homolog (PTEN) loss, and E2F3 transcription factor were associated with increased pCR probability in estrogen receptor (ER) -negative/human epidermal growth factor receptor 2 (HER2) -negative and ER-positive/HER2-negative but not in HER2-positive tumors (interactions between HER2 and each of these modules for their association with pCR: P < .05; FDR, 0.17; trend for interaction between HER2 and PTEN). High values of insulin-like growth factor 1 activation module were associated with increased pCR probability only in ER-positive/HER2-negative tumors (interaction between insulin-like growth factor 1 and ER: P = .002; FDR, 0.03). When adding the immune module to clinicopathologic characteristics, we observed substantial increases in predictive accuracy for pCR in the HER2-positive subtype (IDI, 0.093; P = .004; increase in AUC from 0.760 to 0.836).

CONCLUSION

Different processes and pathways are associated with pCR in different BC subtypes.

Abstract 3541: CD4+ T cells infiltrating human breast cancer are critical for an effective anti-tumor immune response

CD4+ helper T (Th) cells are critical regulators of immune responses but their role in breast cancer is currently unknown. The goal of this project was to characterize CD4+ T lymphocytes infiltrating primary human breast tumors (TIL), analyze T cell signaling pathways influenced by the tumor microenvironment, correlate differential gene expression with the presence of a specific CD4+ T cell subset(s) and identify a profile predictive of positive patient outcome. Affymetrix U133 Plus 2.0 arrays were used to analyze CD4+ T cells freshly isolated from the primary tumor, axillary lymph node (LN) and blood (PB) of patients with ER+ and ER- invasive breast carcinoma in the discovery set (10 patients). Microarrays, flow cytometry and/or qRT-PCR were used to extend and confirm the data in CD4+ TIL isolated from &gt;60 primary breast tumors as well as purified CD4+ T cells from healthy donors PB treated with supernatant (SN) derived from fresh breast tumor tissue. Assessment of the relative distribution of infiltrating mononuclear cells revealed that the CD4+ T cell subset was consistently the principal component. We detected gene expression changes in the CD4+ TIL compared to patient's LN and PB counterparts that reflected altered cellular signaling pathways, including significant suppression of the TCR/CD3 complex and downstream signaling molecules along with a highly restricted pattern of Th cytokine and chemokine expression. A comparison with publically available Th profiling microarray data revealed that the TIL have characteristics of an activated memory effector CD4+ T cell phenotype, with all the major subsets (Th1, Th2, Th17, Treg, Tfh) represented. Although very few differences were detected in a direct comparison between CD4+ TIL from ER+ and ER- tumors, important changes in TIL (including a Th subset skew and differential cytokine expression) from extensive versus minimally infiltrated tumors were observed, with those containing an extensive CD4+ T cell infiltrate associated with a better clinical outcome. Gene expression profiles of freshly isolated TIL were also compared with TIL that had been rested overnight. These experiments revealed that the rested TIL reverse expression in a number of critical cytokine and signaling genes. PB CD4+ T cells from healthy donors treated with primary tumor SN also displayed significant changes in gene expression, gaining &gt;75% similarity to freshly isolated CD4+ TIL. These experiments further found that activation of CD4+ T cells in conjunction with tumor SN treatment rapidly suppressed stimulation-induced changes in gene expression. Together our data suggest that although activated CD4+ T cells infiltrating breast tumors are quickly suppressed, tumors that recruit higher levels of CD4+ T cells, even in the face of rapid immunosuppression, produce a better memory T cell pool that is ultimately capable of maintaining effective anti-tumor immunity in patients.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3541. doi:1538-7445.AM2012-3541

Abstract 5408: CD73 as a target in cancer immunotherapy

It is increasingly being recognised that the tumor microenvironment modulates the effector functions of tumor-infiltrating lymphocytes and consequently suppresses anti-tumor immunity. One immunosuppressive component of the tumor microenvironment is elevated levels of adenosine. The conversion of ATP into adenosine occurs in a stepwise manner essentially through the enzymatic activity of CD39 (NTPDase I) (ATP&gt;AMP) and CD73 (ecto-5′-nucleotidase- AMP&gt;adenosine). CD73 is a glycosylphosphatidyl-inositol (GPI)-linked cell surface enzyme constitutively expressed on endothelial cells, foxp3+ Tregs and subsets of leukocytes, and is considered as the rate-limiting enzyme in the production of extracellular adenosine. We have recently demonstrated that one of the mechanisms contributing to the immunosuppressive accumulation of extracellular adenosine in tumors is the expression of CD73 by tumor cells, but CD73 expression on foxp3+ Tregs is also important for their suppression of anti-tumor immunity. We now report on the role of host-derived CD73 in de novo tumor development in mice. Combinations of anti-CD73 with agonist and T cell checkpoint blockade antibodies demonstrate significant beneficial effects in experimental and de novo models of tumorigenesis. In humans using large cohorts we have shown that CD73 is highest in triple negative breast cancer (TNBC) and expression correlates with an invasion marker and the lack of ER signalling. CD73 expression is also associated with a worse prognosis in TNBC irrespective of treatment and predicts response to anthracycline therapy. The development of CD73 as a target for cancer immunotherapy will be further discussed.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5408. doi:1538-7445.AM2012-5408

Dissecting the heterogeneity of triple-negative breast cancer

Triple-negative breast cancer (TNBC) accounts for 15% to 20% of breast cancers. It is a heterogeneous disease, not only on the molecular level, but also on the pathologic and clinical levels. TNBC is associated with a significantly higher probability of relapse and poorer overall survival in the first few years after diagnosis when compared with other breast cancer subtypes. This is observed despite its usual high sensitivity to chemotherapy. In the advanced setting, responses observed with chemotherapy lack durability. Early-stage clinical studies suggested impressive potential when a poly (ADP-ribose) polymerase (PARP) inhibitor is given for the treatment of advanced TNBC with BRCA gene dysfunction. The molecular complexity of TNBC has led to proposed subclassifications, which will be of great value for the development of targeted therapies. In this review, we discuss the biology of TNBC at the pathologic and the molecular levels. We also elaborate on the role of systemic therapies and the results of the first phase III clinical trial evaluating the addition of iniparib, a novel investigational anticancer agent that does not possess characteristics typical of the PARP inhibitor class, in combination with chemotherapy in advanced TNBC.

A three-gene model to robustly identify breast cancer molecular subtypes

BACKGROUND

Single sample predictors (SSPs) and Subtype classification models (SCMs) are gene expression-based classifiers used to identify the four primary molecular subtypes of breast cancer (basal-like, HER2-enriched, luminal A, and luminal B). SSPs use hierarchical clustering, followed by nearest centroid classification, based on large sets of tumor-intrinsic genes. SCMs use a mixture of Gaussian distributions based on sets of genes with expression specifically correlated with three key breast cancer genes (estrogen receptor [ER], HER2, and aurora kinase A [AURKA]). The aim of this study was to compare the robustness, classification concordance, and prognostic value of these classifiers with those of a simplified three-gene SCM in a large compendium of microarray datasets.

METHODS

Thirty-six publicly available breast cancer datasets (n = 5715) were subjected to molecular subtyping using five published classifiers (three SSPs and two SCMs) and SCMGENE, the new three-gene (ER, HER2, and AURKA) SCM. We used the prediction strength statistic to estimate robustness of the classification models, defined as the capacity of a classifier to assign the same tumors to the same subtypes independently of the dataset used to fit it. We used Cohen κ and Cramer V coefficients to assess concordance between the subtype classifiers and association with clinical variables, respectively. We used Kaplan-Meier survival curves and cross-validated partial likelihood to compare prognostic value of the resulting classifications. All statistical tests were two-sided.

RESULTS

SCMs were statistically significantly more robust than SSPs, with SCMGENE being the most robust because of its simplicity. SCMGENE was statistically significantly concordant with published SCMs (κ = 0.65-0.70) and SSPs (κ = 0.34-0.59), statistically significantly associated with ER (V = 0.64), HER2 (V = 0.52) status, and histological grade (V = 0.55), and yielded similar strong prognostic value.

CONCLUSION

Our results suggest that adequate classification of the major and clinically relevant molecular subtypes of breast cancer can be robustly achieved with quantitative measurements of three key genes.

Elucidating prognosis and biology of breast cancer arising in young women using gene expression profiling

PURPOSE

Breast cancer in young women is associated with poor prognosis. We aimed to define the role of gene expression signatures in predicting prognosis in young women and to understand biological differences according to age.

EXPERIMENTAL DESIGN

Patients were assigned to molecular subtypes [estrogen receptor (ER)(+)/HER2(-); HER2(+), ER(-)/HER2(-))] using a three-gene classifier. We evaluated whether previously published proliferation, stroma, and immune-related gene signatures added prognostic information to Adjuvant! online and tested their interaction with age in a Cox model for relapse-free survival (RFS). Furthermore, we evaluated the association between candidate age-related genes or gene sets with age in an adjusted linear regression model.

RESULTS

A total of 3,522 patients (20 data sets) were eligible. Patients aged 40 years or less had a higher proportion of ER(-)/HER2(-) tumors (P < 0.0001) and were associated with poorer RFS after adjustment for breast cancer subtype, tumor size, nodal status, and histologic grade and stratification for data set and treatment modality (HR = 1.34, 95% CI = 1.10-1.63, P = 0.004). The proliferation gene signatures showed no significant interaction with age in ER(+)/HER2(-) tumors after adjustment for Adjuvant! online. Further analyses suggested that breast cancer in the young is enriched with processes related to immature mammary epithelial cells (luminal progenitors, mammary stem, c-kit, RANKL) and growth factor signaling in two independent cohorts (n = 1,188 and 2,334).

CONCLUSIONS

Proliferation-related prognostic gene signatures can aid treatment decision-making for young women. However, breast cancer arising at a young age seems to be biologically distinct beyond subtype distribution. Separate therapeutic approaches such as targeting RANKL or mammary stem cells could therefore be needed.

Characterization and clinical evaluation of CD10+ stroma cells in the breast cancer microenvironment

PURPOSE

There is growing evidence that interaction between stromal and tumor cells is pivotal in breast cancer progression and response to therapy. Based on earlier research suggesting that during breast cancer progression, striking changes occur in CD10(+) stromal cells, we aimed to better characterize this cell population and its clinical relevance.

EXPERIMENTAL DESIGN

We developed a CD10(+) stroma gene expression signature (using HG U133 Plus 2.0) on the basis of the comparison of CD10 cells isolated from tumoral (n = 28) and normal (n = 3) breast tissue. We further characterized the CD10(+) cells by coculture experiments of representative breast cancer cell lines with the different CD10(+) stromal cell types (fibroblasts, myoepithelial, and mesenchymal stem cells). We then evaluated its clinical relevance in terms of in situ to invasive progression, invasive breast cancer prognosis, and prediction of efficacy of chemotherapy using publicly available data sets.

RESULTS

This 12-gene CD10(+) stroma signature includes, among others, genes involved in matrix remodeling (MMP11, MMP13, and COL10A1) and genes related to osteoblast differentiation (periostin). The coculture experiments showed that all 3 CD10(+) cell types contribute to the CD10(+) stroma signature, although mesenchymal stem cells have the highest CD10(+) stroma signature score. Of interest, this signature showed an important role in differentiating in situ from invasive breast cancer, in prognosis of the HER2(+) subpopulation of breast cancer only, and potentially in nonresponse to chemotherapy for those patients.

CONCLUSIONS

Our results highlight the importance of CD10(+) cells in breast cancer prognosis and efficacy of chemotherapy, particularly within the HER2(+) breast cancer disease.

Proposals for uniform collection of biospecimens from neoadjuvant breast cancer clinical trials: timing and specimen types

In this Personal View, we outline proposals for uniform collection of biospecimens obtained in neoadjuvant breast cancer trials undertaken by the Breast International Group (BIG) and the National Cancer Institute-sponsored North American Breast Cancer Group (NABCG). These proposals aim to standardise collection of high-quality specimens, with respect to both type and timing, to enhance and allow integration of results obtained from neoadjuvant trials done by several groups. They should be considered in parallel with recommendations for tissue-specimen collection and handling previously developed by BIG and NABCG. We propose that tumour tissue (formalin-fixed, paraffin-embedded and samples dedicated for molecular studies) should be taken at baseline, 1-3 weeks after the start of treatment, and at definitive surgery, with clear prioritisation in the study protocol of number, order, and preservation of samples to be gathered. This step should be accompanied by blood collection (plasma, serum, and whole blood) whenever possible. We advocate strongly a move towards one diagnostic and research biopsy procedure in all women with breast cancers potentially suitable for neoadjuvant treatment. If possible, patients should be referred at the outset to specialised centres to give them the opportunity to participate in neoadjuvant clinical trials, thereby avoiding several biopsy procedures.

PD03-10: Gene Modules and Response to Neoadjuvant Chemotherapy in Breast Cancer: A Meta-Analysis

Background: We and others have shown that high expression of proliferation and immune related genes are associated with pathological complete response (pCR) after neoadjuvant chemotherapy in breast cancer (BC). Here, we performed a meta-analysis to validate these findings and to interrogate the association between pCR and several other gene expression modules beyond standard clinico-pathological characteristics in BC subtypes.

Methods: We searched for publicly available gene expression studies evaluating anthracycline ± taxane-based neoadjuvant chemotherapy. We identified 7 studies with complete genomic and clinico-pathological data including pCR totaling 788 patients. Relapse-free survival (RFS) data were available for 427 patients. We used gene expression data generated from pretreatment biopsies and computed 17 gene modules corresponding to proliferation-driven signatures, immune response, stroma activation, phosphatase and tensin homolog (PTEN) loss, chromosomal instability, and several other oncogenic pathways. We calculated odds ratios (OR) for pCR for one-unit increases in scaled modules, all adjusting for pretreatment clinico-pathological characteristics. Moreover, we tested for interactions between gene modules and ER or HER2 status for their association with pCR after adjusting for clinicopathological characteristics. We used the false discovery rate (FDR) to adjust for multiple testing.

Results: We observed pCR in 178(22.6%) of 788 patients: 112(30.1%) of 372 with ER−/HER2− BC, 41(36%) of 113 with HER2+ BC, and 25(8.2%) of 303 with ER+/HER2− BC. High values of the proliferation-driven modules were associated with increased pCR rates in ER−/HER2− (e.g. Gene70 OR=2.34, 95%CI 1.22−4.59, p=0.01, FDR=0.10) and ER+/HER2− (e.g. Gene70 OR=3.26, CI 1.13−9.60, p=0.03, FDR=0.16), but not in the HER2+ subtype (e.g. Gene70 OR=0.82, CI 0.23−2.67, p=0.74, FDR=0.86). There was a trend for interaction between proliferation-driven modules (e.g. Gene70) and HER2 status (p=0.08, FDR=0.23). Interestingly, we demonstrated a strong association between high values of immune module and increased pCR rates in the HER2+ (OR=6.58, CI 2.20−23.40, p&lt;0.01, FDR=0.03) and ER−/HER2− (OR=1.75, CI 1.12−2.76, p=0.02, FDR=0.10) but not in the ER+/HER2− subtype (OR=1.50, CI 0.63−3.42, p=0.35, FDR=0.49). Moreover, high values of AKT activation module were associated with decreased pCR rates in HER2+ (OR=0.38, CI 0.14−0.98, p=0.05, FDR=0.29) but not in the ER−/HER2− (OR=0.87, CI 0.56−1.37, p=0.56, FDR=0.68) or ER+/HER2− (OR=1.81, CI 0.65−5.30, p=0.27, FDR=0.47) subtype. Interactions between immune module and HER2 status and between AKT module and HER2 status were nominally significant (p=0.04 and p=0.05, respectively), but came with an FDR of 0.23. Of interest after a median follow-up of 3.47 years, (95% CI 3.18−3.70 years) patients with pCR had significantly longer RFS irrespective of BC subtypes (HR=0.20, 95% CI 0.08 to 0.50, p&lt;0.001).

Conclusion: Different biological processes namely proliferation, immune response and AKT activation are associated with pCR in different BC subtypes. Our results suggest that new drugs that modulate immune responses in ER−/HER2− and HER2+ BC and target AKT activation in HER2+ BC might be evaluated in these subtypes.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD03-10.

OT1-02-08: The PERSEPHONE Trial: Duration of Trastuzumab with Chemotherapy in Women with HER2 Positive Early Breast Cancer. Changing the Randomisation Point To Address Potential Barriers to Recruitment

Background

Persephone is a phase III randomised controlled trial comparing six months of trastuzumab (9 doses) to the standard 12 month duration (18 doses) in patients with HER2 positive early breast cancer in respect of disease free survival, safety and cost-effectiveness. The trial opened to recruitment in October 2007 but soon showed that meeting recruitment targets was challenging, reaching only 20–30 patients per month. A key issue seemed to be the mandatory requirement to randomise before patients started trastuzumab treatment. Successful accrual of patients in the PHARE trial, a Persephone sister study, run by the National Institute for Cancer, Paris had not incorporated this criteria.

Method: In September 2009, following accrual of 316 patients, this potential barrier to recruitment was addressed by a major protocol amendment which relaxed the eligibility criteria to allow randomisation of patients who had received up to 9 doses of trastuzumab.

Results: To date, 1334 patients have been randomised into PERSEPHONE. After the amendment, monthly recruitment increased to 40–50 patients and, more recently, to 60–70 patients. Of the 1018 patients recruited since September 2009, 450 (44%) patients had received at least 1 dose of trastuzumab pre-randomisation (see Table).

P2-18-01: The Magnitude of Trastuzumab Benefit in HER2−Positive (HER2+) Lobular Breast Carcinoma (BC): Results of a HERA Trial Sub-Group Analysis

Background: Invasive lobular carcinoma (ILC) represents the second most common BC subtype and is often characterized as hormone receptor positive and HER2−negative. However a small subset of ILC is found to be HER2+. Isolated case reports have demonstrated high responsiveness to trastuzumab in patients with advanced HER2+ ILC. The HERA trial compares 1 or 2 years of trastuzumab treatment with observation and 1 versus 2 years of trastuzumab treatment after standard chemotherapy in women with HER2+ breast cancer. We sought to evaluate the incidence of HER2+ ILC and the magnitude of trastuzumab benefit in HER2+ ILC in the context of the HERA trial. In addition, we sought to describe the pattern of hormone receptor positivity in the subsets of ILC and invasive ductal carcinoma (IDC) Methods: The database used in the analysis had a clinical cut-off date of 9th June 2008 and 4-year median follow-up (Gianni et al., 2011). Patients randomized to the 1-year trastuzumab and observation arms were included in the present analysis. Central assessment of hormone receptor status was considered. Histological BC subtype was assessed locally.

Results: Of the 1703 women randomized to one-year of trastuzumab and 1698 to observation, 5.5% (n=187) and 94.5% (n=3213) were diagnosed as HER2+ ILC and IDC, respectively. Central hormone receptor status was available in 88.3% (n = 2838) of IDC and 86.1% (n=161) of ILC. ER and/or PR positivity was more common in ILC than IDC (63.4% [102/161] vs. 46.3% [1314/2838]; p&lt;0.001). Allred scores for ER are shown in the table below. The DFS hazard ratios comparing one year of trastuzumab versus observation were 0.63 (95% CI 0.34−1.14) for ILC and 0.77 (95% CI 0.67−0.89) for IDC. There was no evidence of an interaction between histological subtype and trastuzumab benefit (interaction [subtype lobular and subtype ductal and not lobular] p=0.49).

Low residual proliferation after short-term letrozole therapy is an early predictive marker of response in high proliferative ER-positive breast cancer

The gene expression grade index (GGI) is a 97-gene algorithm that measures proliferation and divides intermediate histological grade tumors into two distinct groups. We investigated the association between early changes in GGI and clinical response to neoadjuvant letrozole and compared this to Ki67 values. The paired gene expression data at the beginning and after 10-14 days of neoadjuvant letrozole treatment were available for 52 post-menopausal patients with estrogen receptor (ER)-positive breast cancer. Baseline values and changes in GGI, Ki67, and RNA expression modules representing oncogenic signaling pathways were compared to sonographic tumor volume changes after 3 months of treatment in the subsets of patients defined by high and low baseline GGI. The clinical response was observed in 80% genomic low-grade (24/30) and 59% genomic high-grade (13/22) tumors (P=0.10). Low residual proliferation after 10-14 days of neoadjuvant letrozole therapy, measured by either GGI or Ki67, was associated with sonographic response in genomic high-grade (GGI, P=0.003; Ki67, P=0.017) but not genomic low-grade (GGI, P=0.25; Ki67, P=1.0) tumors. The analysis of expression modules suggested that sonographic response to letrozole in genomic high-grade tumors was associated with an early reduction in IGF1 signaling (unadjusted P=0.018). The major conclusion of this study is that the early assessment of proliferation after short-term endocrine therapy may be useful to evaluate endocrine responsiveness, particularly in genomic high-grade ER-positive breast cancer.

Tamoxifen in early-stage estrogen receptor-positive breast cancer: overview of clinical use and molecular biomarkers for patient selection

Tamoxifen was the first targeted anticancer agent for breast cancer patients and its effects on reduction of breast cancer events and improvement in overall survival are undisputed. Hence, it has long been considered an essential part of patient care. Recent results of several large adjuvant hormonal trials evaluating the use of aromatase inhibitors in comparison with the previous standard of five years of tamoxifen has led to a paradigm shift, ensuring the inclusion of an aromatase inhibitor as part of standard endocrine therapy for most postmenopausal women diagnosed today with estrogen receptor-positive breast cancer. However, one could argue that despite statistically significant improvements in breast cancer events, an overall survival advantage has not been clear. In this review, we discuss recent genomic and molecular data pertaining to estrogen receptor-positive breast cancer and how this knowledge may aid clinicians to prescribe adjuvant hormonal treatment in the future. A combination of gene expression and genetic aberration markers may be most useful in discerning a population that is still appropriate for adjuvant tamoxifen treatment.

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context

Background

Different microarray studies have compiled gene lists for predicting outcomes of a range of treatments and diseases. These have produced gene lists that have little overlap, indicating that the results from any one study are unstable. It has been suggested that the underlying pathways are essentially identical, and that the expression of gene sets, rather than that of individual genes, may be more informative with respect to prognosis and understanding of the underlying biological process.

Results

We sought to examine the stability of prognostic signatures based on gene sets rather than individual genes. We classified breast cancer cases from five microarray studies according to the risk of metastasis, using features derived from predefined gene sets. The expression levels of genes in the sets are aggregated, using what we call a set statistic. The resulting prognostic gene sets were as predictive as the lists of individual genes, but displayed more consistent rankings via bootstrap replications within datasets, produced more stable classifiers across different datasets, and are potentially more interpretable in the biological context since they examine gene expression in the context of their neighbouring genes in the pathway. In addition, we performed this analysis in each breast cancer molecular subtype, based on ER/HER2 status. The prognostic gene sets found in each subtype were consistent with the biology based on previous analysis of individual genes.

Conclusions

To date, most analyses of gene expression data have focused at the level of the individual genes. We show that a complementary approach of examining the data using predefined gene sets can reduce the noise and could provide increased insight into the underlying biological pathways.

Early Assessment of Proliferation by the Genomic Grade Index (GGI) Predicts Response to Neo-Adjuvant Letrozole

Background: Expression of the proliferation antigen Ki-67 after short-term pre-surgical endocrine treatment is predictive of long-term relapse free survival in post-menopausal women with estrogen-receptor (ER) positive breast cancer (Dowsett et al., JNCI 2007). There are no established predictive markers of clinical response to neo-adjuvant endocrine therapy.Methods: Whole genome expression profiles (Affymetrix HG-U133A) from paired tumor biopsies before and after 10 to 14 days of neo-adjuvant letrozole were compared to tumor volume changes over 3 months of treatment for 52 post-menopausal women with hormone receptor positive breast cancer to evaluate early predictive markers of clinical response.Results: Clinical response was observed in 80% Luminal A (24/30) and 59% Luminal B (13/22) tumors (p=0.1). Baseline tumor proliferation assessed by GGI or Ki-67 mRNA expression were not predictive of clinical response (p=0.29 and p=0.73 respectively). The absolute value of GGI at Day 10 to 14 was predictive of clinical response (p=0.001), particularly for luminal B (p=0.001) but not luminal A (p=0.35) tumors. The change in GGI from baseline to Day 10 to 14 was marginally predictive of response (p=0.063). Absolute day 10 to 14 Ki67 mRNA expression and the change in Ki67 mRNA were not associated with clinical response (p=0.54 and p=0.58 respectively). Baseline, absolute Day 10 to 14, and change in the expression of the ER gene (ESR1) or an integrative gene module reflecting ER signaling were not predictive of clinical response.Conclusion: Expression of GGI after 10 to 14 days of neo-adjuvant letrozole and the change in GGI with short-term treatment are predictive of clinical response observed after 3 months of therapy. Early assessment of GGI may be useful to determine which post-menopausal patients with highly-proliferative ER-positive disease benefit from endocrine therapy.

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 2016.

Gene expression profiling identifies activated growth factor signaling in poor prognosis (Luminal-B) estrogen receptor positive breast cancer

BACKGROUND

Within estrogen receptor-positive breast cancer (ER+ BC), the expression levels of proliferation-related genes can define two clinically distinct molecular subtypes. When treated with adjuvant tamoxifen, those ER+ BCs that are lowly proliferative have a good prognosis (luminal-A subtype), however the clinical outcome of those that are highly proliferative is poor (luminal-B subtype).

METHODS

To investigate the biological basis for these observations, gene set enrichment analysis (GSEA) was performed using microarray data from 246 ER+ BC samples from women treated with adjuvant tamoxifen monotherapy. To create an in vitro model of growth factor (GF) signaling activation, MCF-7 cells were treated with heregulin (HRG), an HER3 ligand.

RESULTS

We found that a gene set linked to GF signaling was significantly enriched in the luminal-B tumors, despite only 10% of samples over-expressing HER2 by immunohistochemistry. To determine the biological significance of this observation, MCF-7 cells were treated with HRG. These cells displayed phosphorylation of HER2/3 and downstream ERK and S6. Treatment with HRG overcame tamoxifen-induced cell cycle arrest with higher S-phase fraction and increased anchorage independent colony formation. Gene expression profiles of MCF-7 cells treated with HRG confirmed enrichment of the GF signaling gene set and a similar proliferative signature observed in human ER+ BCs resistant to tamoxifen.

CONCLUSION

These data demonstrate that activation of GF signaling pathways, independent of HER2 over-expression, could be contributing to the poor prognosis of the luminal-B ER+ BC subtype.

Correlation of PIK3CA mutation-associated gene expression signature (PIK3CA-GS) with deactivation of the PI3K pathway and with prognosis within the luminal-B ER+ breast cancers

533

Background: The phosphathidylinositol-3-kinase (PI3K) signaling pathway is frequently deregulated in tumor biology and is an attractive target for cancer therapy. Our aim was to characterize the molecular and clinical outcome effects of PIK3CA mutations in breast cancer (BC). Methods: We analyzed 173 BC samples for PIK3CA mutations. Corresponding gene expression profiles were used to understand its effects on the PI3K pathway. We validated a PIK3CA-GS in 2 independent BC cohorts (n = 183) with known PIK3CA mutation status and evaluated its correlation with clinical outcome in 1748 BC samples stratified by treatment and subtype. Results: 26% of BCs had a PIK3CA mutation. Tumors with PIK3CA mutation demonstrated a distinct gene expression signature (p = 0.03 after 1000 perm). In 2 datasets it could discriminate PIK3CA mutation carriers from wild-type (ROC 0.68, 0.71, p = 0.001for both). However, the PIK3CA-GS was correlated with deactivation of the PI3K pathway probably through a negative feedback loop. This observation was supported by: 1) the PIK3CA-GS was significantly correlated with gene expression changes induced by PI3K inhibitors (Connectivity Map, Gene set enrichment analyses) and 2) the PIK3CA-GS was anti-correlated with a GS of PTEN loss (R = -0.3; Saal et al, 2007). Higher levels of the PIK3CA signature were observed in HER-2+ and estrogen receptor positive (ER+), luminal BC subtypes. Whilst there was no association with mutation status alone and prognosis, increasing expression of the PIK3CA-GS (suggesting deactivation) was significantly associated with better clinical outcome in both untreated (p = 0.04) and particularly ER+, luminal-B, tamoxifen only-treated (p = 0.004) BC. Multivariate analysis (HR: 0.4; 95%CI: 0.3–0.7; p = 0.002) confirmed that the PI3KCA-GS provided independent prognostic information. Conclusions: Paradoxically, the PIK3CA-GS correlates with inhibition of the PI3K pathway in ER+ BC and identifies a subgroup of luminal B BCs with a favorable outcome. The PIK3CA-GS may be a better indicator of PI3K pathway dysfunction than mutation status, potentially indicating patients who may benefit from combined endocrine therapy and PI3K inhibition.

No significant financial relationships to disclose.

Identification of functional networks of estrogen- and c-Myc-responsive genes and their relationship to response to tamoxifen therapy in breast cancer

Background

Estrogen is a pivotal regulator of cell proliferation in the normal breast and breast cancer. Endocrine therapies targeting the estrogen receptor are effective in breast cancer, but their success is limited by intrinsic and acquired resistance.

Methodology/Principal Findings

With the goal of gaining mechanistic insights into estrogen action and endocrine resistance, we classified estrogen-regulated genes by function, and determined the relationship between functionally-related genesets and the response to tamoxifen in breast cancer patients. Estrogen-responsive genes were identified by transcript profiling of MCF-7 breast cancer cells. Pathway analysis based on functional annotation of these estrogen-regulated genes identified gene signatures with known or predicted roles in cell cycle control, cell growth (i.e. ribosome biogenesis and protein synthesis), cell death/survival signaling and transcriptional regulation. Since inducible expression of c-Myc in antiestrogen-arrested cells can recapitulate many of the effects of estrogen on molecular endpoints related to cell cycle progression, the estrogen-regulated genes that were also targets of c-Myc were identified using cells inducibly expressing c-Myc. Selected genes classified as estrogen and c-Myc targets displayed similar levels of regulation by estrogen and c-Myc and were not estrogen-regulated in the presence of siMyc. Genes regulated by c-Myc accounted for 50% of all acutely estrogen-regulated genes but comprised 85% (110/129 genes) in the cell growth signature. siRNA-mediated inhibition of c-Myc induction impaired estrogen regulation of ribosome biogenesis and protein synthesis, consistent with the prediction that estrogen regulates cell growth principally via c-Myc. The ‘cell cycle’, ‘cell growth’ and ‘cell death’ gene signatures each identified patients with an attenuated response in a cohort of 246 tamoxifen-treated patients. In multivariate analysis the cell death signature was predictive independent of the cell cycle and cell growth signatures.

Conclusions/Significance

These functionally-based gene signatures can stratify patients treated with tamoxifen into groups with differing outcome, and potentially identify distinct mechanisms of tamoxifen resistance.

Predicting prognosis using molecular profiling in estrogen receptor-positive breast cancer treated with tamoxifen

Background

Estrogen receptor positive (ER+) breast cancers (BC) are heterogeneous with regard to their clinical behavior and response to therapies. The ER is currently the best predictor of response to the anti-estrogen agent tamoxifen, yet up to 30–40% of ER+BC will relapse despite tamoxifen treatment. New prognostic biomarkers and further biological understanding of tamoxifen resistance are required. We used gene expression profiling to develop an outcome-based predictor using a training set of 255 ER+ BC samples from women treated with adjuvant tamoxifen monotherapy. We used clusters of highly correlated genes to develop our predictor to facilitate both signature stability and biological interpretation. Independent validation was performed using 362 tamoxifen-treated ER+ BC samples obtained from multiple institutions and treated with tamoxifen only in the adjuvant and metastatic settings.

Results

We developed a gene classifier consisting of 181 genes belonging to 13 biological clusters. In the independent set of adjuvantly-treated samples, it was able to define two distinct prognostic groups (HR 2.01 95%CI: 1.29–3.13; p = 0.002). Six of the 13 gene clusters represented pathways involved in cell cycle and proliferation. In 112 metastatic breast cancer patients treated with tamoxifen, one of the classifier components suggesting a cellular inflammatory mechanism was significantly predictive of response.

Conclusion

We have developed a gene classifier that can predict clinical outcome in tamoxifen-treated ER+ BC patients. Whilst our study emphasizes the important role of proliferation genes in prognosis, our approach proposes other genes and pathways that may elucidate further mechanisms that influence clinical outcome and prediction of response to tamoxifen.

Clinical application of the 70-gene profile: the MINDACT trial

The 70-gene profile is a new prognostic tool that has the potential to greatly improve risk assessment and treatment decision making for early breast cancer. Its prospective validation is currently ongoing through the MINDACT (Microarray in Node-Negative Disease May Avoid Chemotherapy) trial, a 6,000-patient randomized, multicentric trial. This article reviews the several steps in the development of the profile from its discovery to its clinical validation.

Incidence of Malignancies in Heart and/or Lung Transplant Recipients: A Single-Institution Experience

BACKGROUND

The purpose of this study was to determine the incidence and type of malignancies in heart and/or lung transplant recipients at a single institution in Victoria, Australia, and to compare these findings with the non-transplant general Victorian population.

METHODS

Recipients of heart and/or lung transplants at the Alfred Hospital between February 1989 and January 2004 were cross-referenced with the Victorian Cancer Registry. The medical records of all patients with a cancer diagnosis by January 1, 2005 were reviewed. Data were collected on baseline demographics, including cancer type, stage, treatment and survival. Cancer incidence was then compared with rates found in the Victorian population.

RESULTS

There were 907 transplants (Tx) conducted between February 1989 and January 1, 2004 on 905 patients, which included 424 heart (HTx), 56 heart-lung (HLTx), 200 single-lung (SLTx), and 227 double-lung (DLTx) procedures. Of these patients, 606 (67%) were male and 299 (33%) were female. Mean age at transplantation was 46.4 years (range 12.6 to 70.4 years). Four hundred twenty-four (47%) deaths have occurred. Median survival for all patients after transplantation was 8.6 years. One hundred two cancers were confirmed, translating to a 7.1-fold increased incidence compared with the non-transplant population. The most common cancer diagnoses were lymphoproliferative disorders (692 per 100,000 person-years), head and neck cancer (336 per 100,000 person-years) and lung cancer (251 per 100,000 person-years). Compared with the non-transplant population this translates into a 26.2-, 21.0- and 9.3-fold increased risk for developing these cancers, respectively, after cardio-pulmonary transplantation.

CONCLUSIONS

Certain malignancies are more common after heart and/or lung transplantation. The most predominant in our cohort were lymphoproliferative disorders, head and neck cancer and lung cancer.

Investigating tamoxifen resistance in the luminal B estrogen receptor positive breast cancer subtype: Tailoring treatment in hormone responsive breast cancer

10603

Background: We recently reported that the two estrogen receptor (ER) positive breast cancer (BC) molecular subtypes can be defined by their expression of proliferation genes using a gene expression index (GGI): the luminal A and B subtypes have low and high levels respectively (J Clin Onc, in press). When treated with adjuvant tamoxifen, luminal A tumors have a good prognosis, however the clinical outcome of the luminal B subtype was poor. This study aimed to explain the biological basis for these observations using global gene expression profiling and an in vitro model of ER+ BC. Methods: 246 ER+ BC samples from women treated with adjuvant tamoxifen monotherapy were analyzed with affymetrix gene expression arrays and evaluated using gene set enrichment analysis (GSEA). ER+ MCF-7 BC cells (control) treated with tamoxifen (TAM) and heregulin (HRG) were used to investigate molecular pathways identified using GSEA. Results: We found that a gene set suggesting ERBB2 pathway activation was significantly enriched in the luminal B subtype (p=0.02). Only 10% of samples overexpressed HER2 by immunohistochemistry, suggesting that activation of HER2 signaling pathways is independent of HER2 overexpression and may contribute to TAM resistance in this subtype. To validate this hypothesis, MCF-7 cell-lines were treated with HRG (HRG-MCF7) to create a model of ERBB2 pathway activation. HRG-MCF7 cells displayed phosphorylation of HER2/3 without HER2 overexpression. Treatment with HRG overcame TAM induced cell cycle arrest with higher S-phase fraction (p<0.01) and increased anchorage- independent colony formation (p<0.01). Gene expression profiling confirmed significant enrichment of the ERBB2 gene set (p<0.01) and higher GGI levels (p=0.02) in HRG-MCF7 cells compared with control. Conclusions: HRG-MCF7 cells may be useful as an in vitro model of the TAM resistant luminal B subtype. In this group, targeting activated HER2 signaling may be a helpful treatment strategy despite the lack of HER2 overexpression. Our data suggests that agents like lapatinib may be effective only in the luminal B and not the luminal A tumors, demonstrating the importance of stratifying by subtype in future clinical trials of ER+ disease.

No significant financial relationships to disclose.

Strong time dependence of the 76-gene prognostic signature for node-negative breast cancer patients in the TRANSBIG multicenter independent validation series

PURPOSE

Recently, a 76-gene prognostic signature able to predict distant metastases in lymph node-negative (N(-)) breast cancer patients was reported. The aims of this study conducted by TRANSBIG were to independently validate these results and to compare the outcome with clinical risk assessment.

EXPERIMENTAL DESIGN

Gene expression profiling of frozen samples from 198 N(-) systemically untreated patients was done at the Bordet Institute, blinded to clinical data and independent of Veridex. Genomic risk was defined by Veridex, blinded to clinical data. Survival analyses, done by an independent statistician, were done with the genomic risk and adjusted for the clinical risk, defined by Adjuvant! Online.

RESULTS

The actual 5- and 10-year time to distant metastasis were 98% (88-100%) and 94% (83-98%), respectively, for the good profile group and 76% (68-82%) and 73% (65-79%), respectively, for the poor profile group. The actual 5- and 10-year overall survival were 98% (88-100%) and 87% (73-94%), respectively, for the good profile group and 84% (77-89%) and 72% (63-78%), respectively, for the poor profile group. We observed a strong time dependence of this signature, leading to an adjusted hazard ratio of 13.58 (1.85-99.63) and 8.20 (1.10-60.90) at 5 years and 5.11 (1.57-16.67) and 2.55 (1.07-6.10) at 10 years for time to distant metastasis and overall survival, respectively.

CONCLUSION

This independent validation confirmed the performance of the 76-gene signature and adds to the growing evidence that gene expression signatures are of clinical relevance, especially for identifying patients at high risk of early distant metastases.

Definition of clinically distinct molecular subtypes in estrogen receptor-positive breast carcinomas through genomic grade

PURPOSE

A number of microarray studies have reported distinct molecular profiles of breast cancers (BC), such as basal-like, ErbB2-like, and two to three luminal-like subtypes. These were associated with different clinical outcomes. However, although the basal and the ErbB2 subtypes are repeatedly recognized, identification of estrogen receptor (ER) -positive subtypes has been inconsistent. Therefore, refinement of their molecular definition is needed.

MATERIALS AND METHODS

We have previously reported a gene expression grade index (GGI), which defines histologic grade based on gene expression profiles. Using this algorithm, we assigned ER-positive BC to either high-or low-genomic grade subgroups and compared these with previously reported ER-positive molecular classifications. As further validation, we classified 666 ER-positive samples into subtypes and assessed their clinical outcome.

RESULTS

Two ER-positive molecular subgroups (high and low genomic grade) could be defined using the GGI. Despite tracking a single biologic pathway, these were highly comparable to the previously described luminal A and B classification and significantly correlated to the risk groups produced using the 21-gene recurrence score. The two subtypes were associated with statistically distinct clinical outcome in both systemically untreated and tamoxifen-treated populations.

CONCLUSION

The use of genomic grade can identify two clinically distinct ER-positive molecular subtypes in a simple and highly reproducible manner across multiple data sets. This study emphasizes the important role of proliferation-related genes in predicting prognosis in ER-positive BC.

The use of gene-expression profiling to better understand the clinical heterogeneity of estrogen receptor positive breast cancers and tamoxifen response

In a short period of time DNA microarray technology has revolutionized our understanding of human cancer biology. This has been particularly impressive in the field of breast cancer research, where the clinical heterogeneity long observed by physicians seems to be mirrored by different molecular phenotypes exposed by microarray analysis. Gene-expression signatures have been developed to predict prognosis and treatment response and pending adequate validation, are on the verge of entry into the clinical setting. In this review article we explore how gene-expression profiling has influenced our understanding of the ER-positive breast cancers: that proliferation and cell-cycle genes seem to be the strongest predictor for metastasis and relapse in this group, and discuss the various gene predictors and molecular subtype classifications that exist that may help us individualize therapy for these women.

Gene signature evaluation as a prognostic tool: challenges in the design of the MINDACT trial

This Review describes the work conducted by the TRANSBIG consortium in the development of the MINDACT (Microarray In Node negative Disease may Avoid ChemoTherapy) trial. The goal of the trial is to provide definitive evidence regarding the clinical relevance of the 70-gene prognosis signature, and to assess the performance of this signature compared with that of traditional prognostic indicators for assigning adjuvant chemotherapy to patients with node-negative breast cancer. We outline the background work and the key questions in node-negative early-stage breast cancer, and then focus on the MINDACT trial design and statistical considerations. The challenges inherent in this trial in terms of logistics, implementation and interpretation of the results are also discussed. We hope that this article will trigger further discussion about the difficulties of setting up and analyzing trials aimed at establishing the worth of new methods for better selection of patients for cancer treatment.

Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer

BACKGROUND

A 70-gene signature was previously shown to have prognostic value in patients with node-negative breast cancer. Our goal was to validate the signature in an independent group of patients.

METHODS

Patients (n = 307, with 137 events after a median follow-up of 13.6 years) from five European centers were divided into high- and low-risk groups based on the gene signature classification and on clinical risk classifications. Patients were assigned to the gene signature low-risk group if their 5-year distant metastasis-free survival probability as estimated by the gene signature was greater than 90%. Patients were assigned to the clinicopathologic low-risk group if their 10-year survival probability, as estimated by Adjuvant! software, was greater than 88% (for estrogen receptor [ER]-positive patients) or 92% (for ER-negative patients). Hazard ratios (HRs) were estimated to compare time to distant metastases, disease-free survival, and overall survival in high- versus low-risk groups.

RESULTS

The 70-gene signature outperformed the clinicopathologic risk assessment in predicting all endpoints. For time to distant metastases, the gene signature yielded HR = 2.32 (95% confidence interval [CI] = 1.35 to 4.00) without adjustment for clinical risk and hazard ratios ranging from 2.13 to 2.15 after adjustment for various estimates of clinical risk; clinicopathologic risk using Adjuvant! software yielded an unadjusted HR = 1.68 (95% CI = 0.92 to 3.07). For overall survival, the gene signature yielded an unadjusted HR = 2.79 (95% CI = 1.60 to 4.87) and adjusted hazard ratios ranging from 2.63 to 2.89; clinicopathologic risk yielded an unadjusted HR = 1.67 (95% CI = 0.93 to 2.98). For patients in the gene signature high-risk group, 10-year overall survival was 0.69 for patients in both the low- and high-clinical risk groups; for patients in the gene signature low-risk group, the 10-year survival rates were 0.88 and 0.89, respectively.

CONCLUSIONS

The 70-gene signature adds independent prognostic information to clinicopathologic risk assessment for patients with early breast cancer.