VOLUME

33



NUMBER

15



MAY

20

2015

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Predicting Anthracycline Benefit: TOP2A and CEP17—Not Only but Also John M.S. Bartlett, Ontario Institute for Cancer Research; Frances P. O’Malley, St Michael’s Hospital; Kathleen I. Pritchard, Sunnybrook Odette Cancer Centre and University of Toronto, Toronto; John M.S. Bartlett, Frances P. O’Malley, and Lois E. Shepherd, National Cancer Institute of Canada Clinical Trials Group; Lois E. Shepherd, Queen’s University, Kingston, Canada; John M.S. Bartlett, Alison F. Munro, and David A. Cameron, University of Edinburgh, Edinburgh; Christopher C. McConkey, Janet A. Dunn, and Christopher J. Poole, University of Warwick, Coventry; Helena M. Earl and Carlos Caldas, University of Cambridge, Cambridge; Christopher J. Twelves, St James’s University Hospital, Leeds; Daniel W. Rea, University of Birmingham, Birmingham, United Kingdom; Christine Desmedt and Denis P. Larsimont, Université Libre de Bruxelles, Brussels, Belgium; Fatima Cardoso, Champalimaud Cancer Centre, Lisbon, Portugal; Maj-Britt Jensen and Bent Ejlertsen, Rigshospitalet, Copenhagen, Denmark; and Angelo Di Leo, Hospital of Prato, Prato, Italy. Published online ahead of print at www.jco.org on April 20, 2015. Written on behalf of the HER2/TOP2A Meta-Analysis Study Group. Supported by the Ontario Institute for Cancer Research through funding provided by the Government of Ontario; data gathering supported in part by the Brussels Region BruBreast Project, Cancer Research UK, and Scottish Breast Cancer Clinical Trials Group.

John M.S. Bartlett, Christopher C. McConkey, Alison F. Munro, Christine Desmedt, Janet A. Dunn, Denis P. Larsimont, Frances P. O’Malley, David A. Cameron, Helena M. Earl, Christopher J. Poole, Lois E. Shepherd, Fatima Cardoso, Maj-Britt Jensen, Carlos Caldas, Christopher J. Twelves, Daniel W. Rea, Bent Ejlertsen, Angelo Di Leo, and Kathleen I. Pritchard A

Corresponding author: John M.S. Bartlett, PhD, Transformative Pathology, Ontario Institute for Cancer Research, MaRS Centre, South Tower, 101 College St, Suite 800, Toronto, Ontario, Canada M5G 0A3; e-mail: [email protected]. © 2015 by American Society of Clinical Oncology 0732-183X/15/3315w-1680w/$20.00 DOI: 10.1200/JCO.2013.54.7869

1680

S

T

R

A

C

T

Purpose Evidence supporting the clinical utility of predictive biomarkers of anthracycline activity is weak, with a recent meta-analysis failing to provide strong evidence for either HER2 or TOP2A. Having previously shown that duplication of chromosome 17 pericentromeric alpha satellite as measured with a centromere enumeration probe (CEP17) predicted sensitivity to anthracyclines, we report here an individual patient–level pooled analysis of data from five trials comparing anthracyclinebased chemotherapy with CMF (cyclophosphamide, methotrexate, and fluorouracil) as adjuvant chemotherapy for early breast cancer. Patients and Methods Fluorescent in situ hybridization for CEP17, HER2, and TOP2A was performed in three laboratories on samples from 3,846 of 4,864 eligible patients from five trials evaluating anthracyclinecontaining chemotherapy versus CMF. Methodologic differences did not affect HER2-to-CEP17 ratios but necessitated different definitions for CEP17 duplication: ⬎ 1.86 observed copies per cell for BR9601, NEAT, Belgian, and DBCG89D trials and ⬎ 2.25 for the MA.5 trial. Results Fluorescent in situ hybridization data were available in 89.3% (HER2), 83.9% (CEP17), and 80.6% (TOP2A) of 3,846 patient cases with available tissue. Both CEP17and TOP2A treatment-by-marker interactions remained significant in adjusted analyses for recurrence-free and overall survival, whereas HER2 did not. A combined CEP17 and TOP2A–adjusted model predicted anthracycline benefit across all five trials for both recurrence-free (hazard ratio, 0.64; 95% CI, 0.51 to 0.82; P ⫽ .001) and overall survival (hazard ratio, 0.66; 95% CI, 0.51 to 0.85; P ⫽ .005). Conclusion This prospectively planned individual-patient pooled analysis of patient cases from five adjuvant trials confirms that patients whose tumors harbor either CEP17 duplication or TOP2A aberrations, but not HER2 amplification, benefit from adjuvant anthracycline chemotherapy. J Clin Oncol 33:1680-1687. © 2015 by American Society of Clinical Oncology

Terms in blue are defined in the glossary, found at the end of this article and online at www.jco.org. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.

B

INTRODUCTION

The human epidermal growth factor receptor 2 (HER2) oncogene is amplified or overexpressed in approximately 25% of early breast cancers when adjuvant anthracycline use is considered.1 Observational reports have linked anthracycline sensitivity to HER2 positivity.2,3 HER2 amplification is linked to multiple mechanistic pathways (eg, proliferation, dedifferentiation, apoptosis, DNA repair)4 and molecular events, including topoisomerase 2-alpha (TOP2A) alterations and chromosome 17 (Ch17) centromeric duplication.5-7 However, evidence for HER2 and TOP2A as predictive biomarkers of anthracycline benefit is inconsistent across different

studies and assays.2,3,8-10 We have highlighted the difference between the potential of HER2 to identify tumors requiring high-dose or dose-dense chemotherapy and its potential to act as a predictive biomarker of anthracycline benefit.8,11 Recent evidence from a patient-based meta-analysis and our own work challenges previous assumptions that either HER2 or TOP2A represent independent predictive biomarkers of anthracycline sensitivity8,9 and identifies a novel biomarker, duplication of the Ch17 pericentromeric alpha satellite as measured with a centromere enumeration probe (CEP17), as a potential means of selecting patients with breast cancer who will derive benefit from adjuvant chemotherapy containing anthracyclines.8,12 In this report,

© 2015 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

Predicting Anthracycline Benefit

CEP17 duplication represents increased CEP17 copies and must be interpreted in this light. Duplication of CEP175,6 reflects centromeric duplication but does not necessarily reflect 17q21 or chromosomal duplication. Recent data suggest that polysomy, defined by CEP17 duplication, represents duplication of subchromosomal regions, including the CEP, rather than whole-chromosome duplication.13 Given previous so-called false dawns,11 we performed an individual-patient pooled analysis of five similarly designed trials, which between them enrolled 4,864 patients, to test the a priori hypothesis that CEP17 duplication predicts for anthracycline benefit; for completeness, we also tested the predictive value of previous candidates, including TOP2A and HER2.

no clear departure from proportional hazards. The statistical analysis plan was prospectively defined as follows: Interactions between HER2, TOP2A, CEP17, and treatment were considered independently in unadjusted and adjusted analyses (stratified by trial and adjusted for treatment, age, nodal status, grade, tumor size, HER2, TOP2A, and CEP17). All interaction terms that retained significance (P ⬍ .05) in adjusted Cox models were entered independently into a single Cox model, adjusting for conventional prognostic factors. Finally, to illustrate the combined effect of these markers, an exploratory model with a single combined biomarker–treatment interaction term was assessed in a Cox model adjusted for known prognostic factors. A second sensitivity analysis, using additional CEP17 data from the DBCG derived from the TOP2A FISH assay, was performed (Data Supplement). The study conformed to REMARK (Recommendations for Tumor Marker Prognostic Studies) reporting guidelines,21 and a full REMARK table is included (Data Supplement).

PATIENTS AND METHODS RESULTS Clinical Trial Eligibility for Pooled Analysis The trials selected for inclusion in this pooled analysis were adjuvant phase III trials comparing CMF (cyclophosphamide, methotrexate, and fluorouracil) with anthracycline-based therapy, in which the responsible trialists were prepared to share individual patient data for analysis. Five trials met these criteria. Patient characteristics, clinical results, and biomarker data have been published separately for each trial (ie, MA.5, DBCG89D [Danish Breast Cancer Cooperative Group 89D], BR9601, NEAT [National Epirubicin Adjuvant Trial], and Belgian trials2,3,8-10,14-16). The Belgian trial compared two different dose-intensities of CMF with cumulative doses of anthracycline; for our pooled analysis, the anthracycline-based arms were pooled because of the small number of patients in this trial. Data Management Individual patient data from each trial were centralized at Edinburgh University (Data Supplement). Manual checks were performed to verify data consistency. Statistical analyses were performed at the Warwick University Clinical Trials Unit (Coventry, United Kingdom). Fluorescent in situ hybridization (FISH) was performed locally for the different trials for HER2 and TOP2A as previously described.2,3,8-10,16 CEP17 status was derived from the DBCG,3 NEAT/BR9601,17 and MA.512 databases and was repeated on the Belgian study using a single-color probe (Abbott Vysis, Maidenhead, United Kingdom). HER2 amplification was defined as HER2-to-CEP17 ratio ⱖ 2.0, TOP2A amplification as TOP2A-to-CEP17 ratio ⱖ 2.0, and TOP2A deletion as TOP2A-to-CEP17 ratio ⬍ 0.8. Local laboratory estrogen and progesterone receptor status and grade were recorded and immunohistochemistry performed centrally for estrogen and progesterone receptor for the NEAT/ BR9601 studies. FISH for HER2, TOP2A, and CEP17 was performed in separate laboratories (BR9601/NEAT and Belgian trials [CEP17] in J.M.S.B. laboratory; DBCG89D by DBCG; MA.5 in F.P.O. laboratory). BR9601/NEAT, DBCG89D, and Belgian trial tumors were scored, counting all cells with a minimum of one CEP17 signal per cell; in MA.5, a minimum of two CEP17 signals were required for cells to be scored. These methodologic differences did not affect HER2-to-CEP17 ratios but necessitated different definitions for CEP17 duplication, defined as ⬎ 1.866 observed copies per cell for DBCG, BR9601, NEAT, and Belgian trials and ⬎ 2.25 for MA.5 trial.18 Statistical Analyses SAS software (version 9.2; SAS Institute, Cary, NC) was used for all statistical analysis. Pearson’s ␹2 test was used to determine associations between markers. Kaplan-Meier and log-rank analyses were used to compare recurrence-free (RFS) and overall survival (OS), defined as originally outlined for the individual clinical trials, between treatment arms and marker-defined subgroups.8,14-16,19,20 Hazard ratios (HRs) and CIs were calculated from logrank statistics and Cox proportional hazards models as appropriate. Cox proportional hazards models were used for evaluation of treatment-bymarker interactions. All P values are two sided. Graphical examination showed www.jco.org

Patient data were retrieved from 4,864 eligible patients; tissue was available for 3,846 patient cases (79.1%). This study population was representative of that of the main trials.2,8-10,14-16, 20 HER2 status was available for 89.3% of patients with tissue available (3,436 tumors), TOP2A status for 80.6% (3,098 tumors), and CEP17 status for 83.9% (3,225 tumors). A total of 1,581 (41.1%) of 3,846 patients relapsed and 1,329 (34.6%) of 3,846 died during follow-up. The advantage of anthracycline-containing polychemotherapy over CMF for both RFS and OS (HR, 0.82; 95% CI, 0.74 to 0.91; P ⬍ .001 and HR, 0.84; 95% CI, 0.76 to 0.94; P ⫽ .002, respectively; stratified by trial) was similar to that in the overall population. All subsequent analyses were restricted to patient cases with available biomarker data or subsets thereof. Prognostic Impact of Biomarkers The impact of each biologic marker evaluated in this study was first tested with respect to RFS and OS in all patients, irrespective of allocated treatment. In unadjusted analyses, HER2 gene amplification, TOP2A deletion, and CEP17 duplication were each significantly associated with worse outcome for both RFS and OS. There was no such relationship with TOP2A amplification, but when both amplification and deletion of TOP2A were combined as a single variable, TOP2Aaltered patient cases had significantly poorer RFS and OS (Table 1; Data Supplement). Despite heterogeneity for the prognostic impact of some markers across studies, there were no strong reasons to omit any studies on the grounds of such heterogeneity. Predictive Significance of Biomarkers Subsequent analyses focused on predictive effects of markers for RFS and OS between patients receiving anthracyclines versus CMF alone. Table 2 and the Data Supplement summarize HR estimates for RFS and OS by treatment and molecular biomarker subgroup, with biomarkers dichotomized as positive or negative. Unadjusted Analyses No statistically significant interaction between treatment with epirubicin and TOP2A (amplification or deletion) was detected for RFS or OS in unadjusted treatment-by-marker analyses (Table 2; Data Supplement). However, significant interactions were observed between HER2 and treatment (RFS: HR, 0.75; 95% CI, 0.60 to 0.95; P ⫽ .02; OS: HR, 0.76; 95% CI, 0.60 to 0.97; P ⫽ .03), combined TOP2A © 2015 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

1681

Bartlett et al

Table 1. Biomarker Analysis and Relationship With OS and RFS Adjusted by Trial Patients Marker HER2 Normal Amplified TOP2A Normal Amplified TOP2A Normal Deleted TOP2A Normal AB Ch17CEP Normal Duplicated

RFS

No.

%

2,604 832

75.8 24.2

2,825 273

91.2 8.8

2,784 314

89.9 10.1

2,511 587

81.1 18.9

2,254 971

69.9 30.1

OS HR

95% CI

Pⴱ

⬍ .001

1.4

1.54 to 1.96

⬍ .001

0.83 to 1.22

.96

1.11

0.91 to 1.35

.31

1.58

1.34 to 1.87

⬍ .001

1.72

1.44 to 2.04

⬍ .001

1.32

1.15 to 1.51

⬍ .001

1.46

1.27 to 1.68

⬍ .001

1.22

1.08 to 1.37

.001

1.27

1.12 to 1.44

⬍ .001

HR

95% CI

P

1.51

1.35 to 1.70

1.01



Abbreviations: AB, combined TOP2A deletions and amplifications; CEP, centromere enumeration probe; Ch17, chromosome 17; HR, hazard ratio; OS, overall survival; RFS, relapse-free survival. ⴱ Unadjusted Cox regression analysis.

abnormalities and treatment (RFS: HR, 0.69; 95% CI, 0.52 to 0.90; P ⫽ .007; OS: HR, 0.71; 95% CI, 0.54 to 0.94; P ⫽ .02), and CEP17 and treatment (RFS: HR, 0.68; 95% CI, 0.54 to 0.86; P ⫽ .001; OS: HR, 0.70; 95% CI, 0.55 to 0.90; P ⫽ .005; Table 2; Data Supplement). Although there was some heterogeneity between trials for interaction HRs, four of five trials showed similar HRs for CEP17 (Table 2; Data Supplement). Adjusted Analyses In separate adjusted Cox regression analyses of treatment-bymarker interactions (HER2, TOP2A abnormalities, and CEP17, re-

spectively), the interaction term for HER2 was not significant for either RFS or OS, whereas those for TOP2A and CEP17 retained significance when stratified by trial and adjusted for confounding factors (Table 3). Because the TOP2A and CEP17 interaction models were based on the same patient cases, they were compared using the likelihood ratio ␹2. Differences in likelihood ratio ␹2 between models (RFS, 1.81; OS, 0.81) were too small to provide a basis for preferring either interaction term over the other. Therefore, a single Cox regression model was performed to test effects of TOP2A and CEP17 interactions with treatment simultaneously after adjustment for confounding factors (Table 4; Data Supplement). The interaction for

Table 2. Treatment-by-Marker Interactions for RFS and OS Adjusted by Trial RFS

OS

Anthracycline Versus CMF Marker

HR

HER2 Normal Amplified TOP2A Normal Amplified TOP2A Normal Deleted TOP2A Normal AB Ch17CEP Normal Duplicated

95% CI

Pⴱ

0.88 0.69

0.78 to 1.00 0.57 to 0.84

.03 .95

0.83 0.59

0.74 to 0.94 0.41 to 0.86

.14 .58

0.84 0.63

0.74 to 0.95 0.46 to 0.87

.49 .09

0.87 0.62

0.76 to 0.99 0.50 to 0.79

.28 .84

0.93 0.63

0.81 to 1.06 0.52 to 0.76

.23 .19

Treatment by Marker

Anthracycline Versus CMF

HR

95% CI

P

0.75

0.60 to 0.95

.02

0.73

0.74

0.69

0.68

0.50 to 1.07

0.53 to 1.04

0.52 to 0.90

0.54 to 0.86

HR

95% CI

Pⴱ

0.92 0.72

0.80 to 1.05 0.59 to 0.88

.19 .96

0.86 0.66

0.76 to 0.98 0.45 to 0.97

.35 .64

0.88 0.67

0.77 to 1.00 0.48 to 0.93

.56 .18

0.90 0.65

0.78 to 1.03 0.51 to 0.84

.33 .78

0.95 0.68

0.82 to 1.10 0.55 to 0.83

.64 .49

.11

.08

Treatment by Marker HR

95% CI

P

0.76

0.60 to 0.97

.03

0.78

0.53 to 1.16

.23

0.73

0.52 to 1.04

.08

0.71

0.54 to 0.94

.02

0.70

0.55 to 0.90

.005

.007

.001

Abbreviations: AB, combined TOP2A deletions and amplifications; CEP, centromere enumeration probe; Ch17, chromosome 17; CMF, cyclophosphamide, methotrexate, and fluorouracil; HR, hazard ratio; OS, overall survival; RFS, relapse-free survival. ⴱ Heterogeneity between trials.

1682

© 2015 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

Predicting Anthracycline Benefit

Table 3. Adjusted Treatment-by-Marker Regression Analyses for RFS and OS Corrected for Clinicopathologic Variables (n ⫽ 2,797) HER2 Amplification Factor

HR

95% CI

TOP2A Alteration P

HR

CEP17 Duplication

95% CI

P

HR

95% CI

P

0.74 to 0.97 1.79 to 2.15 1.27 to 1.57 1.09 to 1.33 0.99 to 1.00 1.18 to 1.61 0.99 to 1.51 1.00 to 1.30 0.50 to 0.89

.02 ⬍ .001 ⬍ .001 ⬍ .001 .38 ⬍ .001 .06 .05 .005

0.88 1.96 1.40 1.21 1.00 1.38 1.01 1.36 0.67

0.76 to 1.02 1.79 to 2.15 1.26 to 1.56 1.09 to 1.33 0.99 to 1.00 1.18 to 1.61 0.86 to 1.20 1.15 to 1.61 0.52 to 0.86

.09 ⬍ .001 ⬍ .001 ⬍ .001 .35 ⬍ .001 .89 ⬍ .001 .002

0.77 to 1.03 1.81 to 2.20 1.30 to 1.63 1.14 to 1.41 1.00 to 1.02 1.32 to 1.82 1.01 to 1.56 0.99 to 1.31 0.50 to 0.90

.12 ⬍ .001 ⬍ .001 ⬍ .001 .04 ⬍ .001 .04 .07 .008

0.91 1.99 1.45 1.27 1.01 1.55 1.04 1.32 0.71

0.77 to 1.06 1.81 to 2.20 1.29 to 1.62 1.14 to 1.41 1.00 to 1.02 1.32 to 1.82 0.87 to 1.24 1.11 to 1.58 0.54 to 0.93

.23 ⬍ .001 ⬍ .001 ⬍ .001 .05 ⬍ .001 .66 .002 .01

RFS Treatment Node involvement Size Grade Age HER2 amplification TOP2A AB Ch17CEP duplication Treatment-by-marker interaction

0.82 1.96 1.41 1.21 1.00 1.52 1.01 1.14 0.81

0.71 to 0.95 1.79 to 2.14 1.27 to 1.56 1.09 to 1.33 0.99 to 1.00 1.26 to 1.85 0.86 to 1.20 1.00 to 1.31 0.63 to 1.05

.009 ⬍ .001 ⬍ .001 ⬍ .001 .39 ⬍ .001 .89 .05 .11

Treatment Node involvement Size Grade Age HER2 amplification TOP2A AB Ch17CEP duplication Treatment-by-marker interaction

0.87 1.99 1.45 1.27 1.01 1.23 1.04 1.14 0.80

0.74 to 1.02 1.81 to 2.19 1.30 to 1.63 1.14 to 1.41 1.00 to 1.02 1.41 to 2.11 0.87 to 1.24 0.99 to 1.31 0.61 to 1.04

.09 ⬍ .001 ⬍ .001 ⬍ .001 .04 ⬍ .001 .65 .07 .10

0.85 1.96 1.41 1.21 1.00 1.38 1.22 1.14 0.67 OS 0.89 1.99 1.46 1.27 1.01 1.55 1.26 1.14 0.67

NOTE. Cox regression including one interaction term (marker ⫻ treatment) stratified by trial in 2,797 patients. Abbreviations: AB, combined TOP2A deletions and amplifications; CEP, centromere enumeration probe; Ch17, chromosome 17; HR, hazard ratio; OS, overall survival; RFS, relapse-free survival.

CEP17 duplication and anthracycline therapy remained significant for RFS (P ⫽ .01) and was of borderline significance for OS (P ⫽ .06). The TOP2A treatment-by-marker interaction remained significant for RFS (P ⫽ .03) and OS (P ⫽ .03). The unadjusted treatment effects were large in the 561 patient cases with CEP17 duplication but normal TOP2A and in the 260 patient cases with abnormal TOP2A but normal CEP17 (0.65 and 0.62, respectively, for RFS; 0.68 and 0.63, respectively, for OS). Therefore, we extended the analysis by grouping together CEP17 and TOP2A abnormalities (Fig 1). Both across all five trials and within individual trials, this combined CEP17–TOP2A profile consistently identified patients who experienced benefit from addition of anthracyclines to their polychemotherapy (Fig 1); effects were consistent across trials and molecular subgroups (Data Supplement). DISCUSSION

The data presented here provide evidence for a potentially clinically useful predictive biomarker identifying patients who derive benefit from anthracycline-based polychemotherapy in the adjuvant treatment of early breast cancer. This individual patient–level adjusted pooled analysis of five clinical trials, with centrally validated biomarker evaluation, demonstrated that duplication of CEP17 and/or alteration of TOP2A were independent predictive markers of anthracycline sensitivity, with consistent results across five trials. Other suggested biomarkers, notably HER2, were not strongly predictive of anthracycline benefit, as shown previously by a similar meta-analysis.9 The relative benefit derived from anthracyclines in the different patient populations analyzed here was highly consistent across five studies and molecular subgroups (luminal A and B, HER2, and triple negative). www.jco.org

Patients whose tumors exhibited CEP17 and TOP2A abnormalities exhibited a 38% reduction in risk of relapse when treated with anthracycline-containing chemotherapy relative to those treated with CMF alone, a benefit not found for those women whose tumors were normal for CEP17 and TOP2A (Fig 1). The effect reported here differs from previously reported preliminary results17 based on four trials, in that the fifth trial in our pooled analysis (ie, DBCG89D trial) supported the inclusion of the TOP2A gene in the analysis (Data Supplement). Similarly, in contrast to the analyses previously performed across these five trials,9 the combination of CEP17 and TOP2A almost entirely eliminated the heterogeneity across trials observed when either was used as a single marker. Only the small number of patient cases (40 in total) from the Belgian study showed potential heterogeneity in this combined analysis. These results present two challenges in seeking to understand mechanisms that explain the observed results. First, although in vitro evidence supports a role for TOP2A in predicting anthracycline benefit, it is only when both deletions and amplifications of this gene were combined in clinical studies that the predictive effect was observed. This result contradicts a mechanism of anthracycline benefit linked to TOP2A gene expression, and indeed, studies exploring TOP2A gene expression have not shown robust prediction of anthracycline benefit.11 Second, CEP17 duplication, identified by FISH analysis, did not identify a potential mechanism underlying anthracycline sensitivity. The CEP17 probe binds the Ch17 alpha satellite pericentromeric polymorphic repeat, which both enumerates copy number (hence, CEP) and identifies amplification or duplication of the pericentromeric region without chromosome duplication. Ch17 houses multiple genes central to breast cancer and DNA repair, including BRCA1, © 2015 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

1683

Bartlett et al

Table 4. Adjusted Treatment-by-Marker Regression Analyses for RFS and OS for Both CEP17 and TOP2A Interaction Terms Corrected for Clinicopathologic Variables (n ⫽ 2,797) Factor

HR

95% CI

P

0.93 1.97 1.40 1.21 1.00 1.38 1.17 1.32 0.73 0.71

0.80 to 1.08 1.80 to 2.15 1.26 to 1.56 1.09 to 1.33 0.99 to 1.00 1.18 to 1.60 0.95 to 1.45 1.11 to 1.56 0.54 to 0.98 0.55 to 0.93

.34 ⬍ .001 ⬍ .001 ⬍ .001 .39 ⬍ .001 .14 .002 .03 .01

0.96 1.99 1.45 1.27 1.01 1.55 1.21 1.28 0.72 0.76

0.81 to 1.14 1.81 to 2.20 1.29 to 1.62 1.14 to 1.41 1.00 to 1.02 1.31 to 1.82 0.97 to 1.52 1.06 to 1.53 0.53 to 0.98 0.58 to 1.01

.64 ⬍ .001 ⬍ .001 ⬍ .001 .04 ⬍ .001 .09 .009 .03 .06

RFS Treatment Node involvement Size Grade Age HER2 amplification TOP2A AB Ch17CEP duplication TOP2A ⫻ anthracycline interaction Ch17 ⫻ anthracycline interaction OS Treatment Node involvement Size Grade Age HER2 amplification TOP2A AB Ch17CEP duplication TOP2A ⫻ anthracycline interaction Ch17 ⫻ anthracycline interaction

Abbreviations: AB, combined TOP2A deletions and amplifications; CEP, centromere enumeration probe; Ch17, chromosome 17; HR, hazard ratio; OS, overall survival; RFS, relapse-free survival.

TP53, HER2, and TOP2A.2,22-26 The 17q12-q21 chromosomal region exhibits extensive genetic instability. Although HER2 is the most widely studied gene in the region, amplifications and deletions of neighboring genes, including GRB7, CDC6, and TOP2A, are frequent.27 However, CEP17 duplication reflects chromosomal instability and spindle assembly checkpoint disregulation, which are linked in vitro to anthracycline sensitivity and could readily explain the observed clinical effect11 while linking TOP2A abnormalities (deletions and amplifications) and CEP17. We found no evidence for HER2 as a predictive biomarker for anthracycline benefit, consistent with current literature.9 Initial reports that HER2 expression might predict anthracycline benefit were linked to anthracycline dose escalation or acceleration in retrospective analyses of two trials, in which all patients received anthracyclines.28,29 Only subsequently were HER2 and TOP2A tested as biomarkers to identify response to anthracycline-based compared with CMF-based chemotherapies in a number of smaller trials3,8,9,16,30-34 (as reviewed by Munro et al11). Critically, only one of these earlier studies was prospectively powered to test the primary hypothesis that HER2 predicted for anthracycline benefit,8 and this study (based on one trial in our pooled analysis), consistent with the recent meta-analysis, failed to support the use of HER2 to predict anthracycline benefit.9 The extent to which different methodologies and modest study sizes contributed to the uncertainty is unclear. In contrast, the combination of CEP17 duplication and TOP2A amplification or deletion provided a unifying biomarker for anthracycline benefit across multiple trials in which previous biomarker data were heterogeneous3,8,10,11,34-37 and potentially provides a criterion to 1684

© 2015 by American Society of Clinical Oncology

select patients likely to benefit from anthracycline-containing therapy. Although the biologic mechanism underlying this effect remains to be confirmed, chromosome instability is linked to both CEP17 duplication and amplification and deletion of TOP2A and may provide a mechanistic rationale for the combination of these markers to predict anthracycline benefit. Furthermore, given the numerous possible effects of anthracyclines, it suggests that there may not be a single gene or pathway that identifies anthracycline-sensitive cancers. It has recently been suggested that the HER2-positive population of patients, who are clear candidates for trastuzumab, do not require any anthracycline treatment, based on data from the BCIRG006 (Breast Cancer International Research Group) trial.38 This study, which recruited only patients with HER2 amplification, randomly assigned patients to docetaxel, carboplatin, and Herceptin (trastuzumab; Genentech, South San Francisco, CA) or anthracycline and cyclophosphamide followed by docetaxel and herceptin and compared outcome with a control arm of anthracycline plus cyclophosphamide followed by docetaxel.38 In patients whose tumors were amplified for both HER2 and TOP2A, no superiority of any regimen (⫾ trastuzumab or ⫾ anthracyclines) was demonstrated. This result is open to interpretation in a number of ways. It may suggest that in HER2- or TOP2A-amplified breast cancers, TOP2A amplification is required for benefit from anthracyclines; equally, it may suggest that TOP2A amplification is a marker of trastuzumab resistance, because patients whose tumors had amplification of both HER2 and TOP2A did not benefit significantly from the addition of tratuzumab. Finally, in HER2-positive cancers, TOP2A amplification may be a marker of generalized chemosensitivity, because chemotherapy with docetaxel, carboplatin, and trastuzumab was no less effective than either comparison arm. The conclusion—that a small group (4% to 6%) of HER2- and TOP2A-amplified patient cases may be spared anthracyclines— does not inform treatment choice in HER2-negative breast cancers and does not address the benefit from anthracycline-based therapy observed in CEP17-abnormal HER2-negative patients in this study. However, our study cannot fully address whether the patients who benefit from anthracyclines also benefit from the further addition of taxanes. It remains possible that there are subpopulations of patients whose tumors are sensitive to either anthracyclines or taxanes. Of interest, the N9831 adjuvant trastuzumab trial suggested prolonged RFS in patients with CEP17 duplication treated with doxorubicin plus cyclophosphamide and sequential paclitaxel compared with patients without duplication. In trastuzumab-treated patients, this effect was lost, a result strikingly similar to that observed for TOP2A in the BCIRG006 trial. Although this study again was restricted to patients with HER2-positive tumors, the CEP17 treatment interaction observed in the control arm is provocative and suggests the results observed in our study should be tested in taxane-based clinical trials, particularly for HER2-negative patient cases. However, for patients with normal CEP17 copy number, there is no therapeutic advantage in using anthracyclines rather than CMF, which may be particularly important if taxanes are not prescribed or are contraindicated. Adequate quality control programs for the analysis of CEP17 and TOP2A within the context of current HER2 FISH-based diagnostic procedures would be required to support the implementation of this finding in routine clinical practice, but existing evidence suggests that this could be achieved. A critical appraisal of the effects of this JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

Predicting Anthracycline Benefit

A

Anthracycline Events/Patients %

CMF Events/Patients %

TOP2A/CEP17 normal 23/72 BR9601 36/84 31.9 171/559 NEAT 168/537 30.6 55/106 MA5 51/112 51.9 9/13 Belgian 1/2 69.2 61/152 DBCG 68/157 40.1 319/902 Stratified 324/892 35.4 Heterogeneity between 5 groups χ24 = 3.3; P = .51 TOP2A/CEP17 abnormal BR9601 22/76 34/68 28.9 NEAT 51/194 82/213 26.3 MA5 46/96 70/111 47.9 Belgian 6/17 4/8 35.3 DBCG 73/153 114/189 47.7 Stratified 198/536 304/589 36.9 Heterogeneity between 5 groups χ42 = 1.7; P = .80 TOP2A/CEP17 interaction TOP2A/CEP17 normal all 319/902 324/892 35.4 TOP2A/CEP17 abnormal all 198/536 304/589 36.9 Stratified 517/1,438 628/1,481 36.0 Interaction between 2 groups χ21 = 11.4; P < . 001

42.9 31.3 45.5 50.0 43.3 36.3

0.41 to 1.14 0.76 to 1.17 0.83 to 1.78

0.92 0.95

0.65 to 1.30 0.81 to 1.11

P

0.28 to 0.80 0.44 to 0.88 0.45 to 0.94 0.13 to 2.22 0.52 to 0.92 0.53 to 0.76 .001

0.95 0.63 0.79

36.3 51.6 42.4

0.81 to 1.11 0.53 to 0.76 0.71 to 0.89 .001

0.5

1.0

1.5

2.0

CMF better

CMF Events/Patients %

TOP2A/CEP17 normal 21/72 BR9601 29/84 29.2 142/559 NEAT 137/537 25.4 43/106 MA5 34/112 40.6 8/13 Belgian 1/2 61.5 63/152 DBCG 69/157 41.4 277/902 Stratified 270/892 30.7 Heterogeneity between 5 groups χ42 = 4.4; P = .36 TOP2A/CEP17 abnormal BR9601 21/76 29/68 27.6 NEAT 44/194 71/213 22.7 MA5 38/96 56/111 39.6 Belgian 5/17 4/8 29.4 DBCG 77/153 114/189 50.3 Stratified 185/536 274/589 34.5 Heterogeneity between 5 groups χ42 = 1.7; P = .79 TOP2A/CEP17 interaction TOP2A/CEP17 normal all 277/902 270/892 30.7 TOP2A/CEP17 abnormal all 185/536 274/589 34.5 Stratified 462/1,438 544/1,481 32.1 Interaction between 2 groups χ21 = 9.1; P = . 003

0.69 0.94 1.21

0.47 0.62 0.65 0.54 0.69 0.63

50.0 38.5 63.1 50.0 60.3 51.6

0.0

Anthracycline Events/Patients %

CI

.51

Anthracycline better

B

HR

34.5 25.5 30.4 50.0 43.9 30.3

HR

CI

0.78 0.97 1.48

0.45 to 1.36 0.76 to 1.22 0.95 to 2.32

0.91 0.99

0.65 to 1.28 0.84 to 1.17

P

.90 0.54 0.63 0.69 0.42 0.75 0.67

42.6 33.3 50.5 50.0 60.3 46.5

0.31 to 0.94 0.44 to 0.90 0.46 to 1.04 0.09 to 1.87 0.56 to 0.99 0.56 to 0.81 < .001

30.3 46.5 36.7

0.99 0.67 0.83

0.84 to 1.17 0.56 to 0.81 0.73 to 0.94 .003

0.0

0.5

1.0

Anthracycline better

1.5

2.0

CMF better

Fig 1. Forest plot illustrating treatment-by-marker effects overall and by trial for TOP2A/chromosome 17 pericentromeric alpha satellite as measured with a centromere enumeration probe (CEP17) normal combined, TOP2A/CEP17 abnormal combined, and all patients on (A) recurrence-free and (B) overall survival. CMF, cyclophosphamide, methotrexate, and fluorouracil; DBCG, Danish Breast Cancer Cooperative Group; HR, hazard ratio; NEAT, National Epirubicin Adjuvant Trial.

biomarker in trials randomly assigning patients to receive anthracyclines or not in the context of taxane-based chemotherapy would be valuable. In conclusion, we provide evidence, in a retrospective individualpatient intent-to-treat pooled analysis of five clinical trials, that abnorwww.jco.org

mal CEP17 and/or TOP2A in the primary tumor may identify patients who will benefit from polychemotherapy regimens, including an anthracycline. TOP2A and CEP17 can be measured accurately and reproducibly and seem to separate two patient populations. Existing in situ hybridization diagnostic assays provide a robust and © 2015 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

1685

Bartlett et al

readily applicable method for patient selection. Although evidence from other trials may suggest that the substitution of trastuzumab for anthracyclines in HER2-positive patients may negate this effect, in HER2-negative tumors, CEP17 and TOP2A seem to select a significant subgroup of patients for whom anthracyclines are of benefit while simultaneously identifying patients for whom anthracyclines may be safely omitted.

Leo, AstraZeneca (C), Novartis (C), Roche (C) Stock Ownership: None Honoraria: Christopher J. Poole, Roche, Eisai, Genomic Health; Angelo Di Leo, AstraZeneca, Novartis, Roche Research Funding: David A. Cameron, Pfizer (paid to another institution but D.A.C. was trial chief investigator; Christopher J. Poole (NEAT trial was funded in part by Pfizer through another institution/previous employer); Bent Ejlertsen, Amgen, Novartis, Roche; Angelo Di Leo, Pfizer Expert Testimony: None Patents, Royalties, and Licenses: None Other Remuneration: None

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST AUTHOR CONTRIBUTIONS

Although all authors completed the disclosure declaration, the following author(s) and/or an author’s immediate family member(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: None Consultant or Advisory Role: Christopher J. Poole, Roche (C), AstraZeneca (C), Eli Lilly (C), Novartis (C), Genomic Health (C), Gelgene (C), Eisai (C); Angelo Di REFERENCES 1. Bartlett JM, Ellis IO, Dowsett M, et al: Human epidermal growth factor receptor 2 status correlates with lymph node involvement in patients with estrogen receptor (ER) negative, but with grade in those with ER-positive early-stage breast cancer suitable for cytotoxic chemotherapy. J Clin Oncol 25:44234430, 2007 2. Pritchard KI, Shepherd LE, O’Malley FP, et al: HER2 and responsiveness of breast cancer to adjuvant chemotherapy. N Engl J Med 354:2103-2111, 2006 3. Knoop AS, Knudsen H, Balslev E, et al: Retrospective analysis of topoismerase IIa amplifications and deletions as predictive markers in primary breast cancer patients randomly assigned to cyclophosphamide, methotrexate, and fluorouracil or cyclophosphamide, epirubicin and fluorouracil: Danish Breast Cancer Cooperative Group. J Clin Oncol 23:7483-7490, 2005 4. Ross JS, Fletcher JA, Bloom KJ, et al: Targeted therapy in breast cancer: The HER-2/neu gene and protein. Mol Cell Proteomics 3:379-398, 2004 5. Järvinen TA, Kononen J, Peltohuikko M, et al: Expression of topoisomerase II alpha is associated with rapid cell proliferation, aneuploidy, and c-erbB2 overexpression in breast cancer. Am J Pathol 148: 2073-2082, 1996 6. Watters AD, Going JJ, Cooke TG, et al: Chromosome 17 aneusomy is associated with poor prognostic factors in invasive breast carcinoma. Breast Cancer Res Treat 77:109-114, 2003 7. Bartlett JMS, Campbell FM, Mallon EA: Determination of HER2 amplification by in situ hybridization: When should chromosome 17 also be determined? Am J Clin Pathol 130:920-926, 2008 8. Bartlett JM, Munro AF, Dunn JA, et al: Predictive markers of anthracycline benefit: A prospec1686

© 2015 by American Society of Clinical Oncology

Conception and design: John M.S. Bartlett, Janet A. Dunn, David A. Cameron, Helena M. Earl, Christopher J. Poole, Fatima Cardoso, Bent Ejlertsen, Angelo Di Leo, Kathleen I. Pritchard Provision of study materials or patients: John M.S. Bartlett, Fatima Cardoso, Bent Ejlertsen Collection and assembly of data: John M.S. Bartlett, Christopher C. McConkey, Alison F. Munro, Christine Desmedt, Denis P. Larsimont, Frances P. O’Malley, Lois E. Shepherd, Maj-Britt Jensen, Carlos Caldas, Chris J. Twelves, Daniel W. Rea, Bent Ejlertsen, Kathleen I. Pritchard Data analysis and interpretation: John M.S. Bartlett, Christopher C. McConkey, Janet A. Dunn, David A. Cameron, Lois E. Shepherd, Fatima Cardoso, Maj-Britt Jensen, Bent Ejlertsen, Kathleen I. Pritchard Manuscript writing: All authors Final approval of manuscript: All authors

tively planned analysis of the UK National Epirubicin Adjuvant Trial (NEAT/BR9601). Lancet Oncol 11: 266-274, 2010 9. Di Leo A, Desmedt C, Bartlett JM, et al: HER2 and TOP2A as predictive markers for anthracyclinecontaining chemotherapy regimens as adjuvant treatment of breast cancer: A meta-analysis of individual patient data. Lancet Oncol 12:1134-1142, 2011 10. O’Malley FP, Chia S, Tu D, et al: Topoisomerase II alpha and responsiveness of breast cancer to adjuvant chemotherapy. J Natl Cancer Inst 101:644650, 2009 11. Munro AF, Cameron DA, Bartlett JM: Targeting anthracyclines in early breast cancer: New candidate predictive biomarkers emerge. Oncogene 29:5231-5240, 2010 12. Pritchard K, Munro A, O’Malley FP, et al: Chromosome 17 centromere (CEP17) duplication as a predictor of anthracycline response: Evidence from the NCIC Clinical Trials Group (NCIC CTG) MA.5 trial. Breast Cancer Res Treat 131:541-551, 2012 13. Yeh IT, Martin MA, Robetorye RS, et al: Clinical validation of an array CGH test for HER2 status in breast cancer reveals that polysomy 17 is a rare event. Mod Pathol 22:1169-1175, 2009 14. Poole CJ, Earl HM, Hiller L, et al: Epirubicin and cyclophosphamide, methotrexate, and fluorouracil as adjuvant therapy for early breast cancer. N Engl J Med 355:1851-1862, 2006 15. Levine MN, Pritchard KI, Bramwell VH, et al: Randomized trial comparing cyclophosphamide, epirubicin, and fluorouracil with cyclophosphamide, methotrexate, and fluorouracil in premenopausal women with node-positive breast cancer: Update of National Cancer Institute of Canada Clinical Trials Group trial MA5. J Clin Oncol 23:5166-5170, 2005 16. Di Leo A, Gancberg D, Larsimont D, et al: HER-2 amplification and topoisomerase II alpha gene aberra-

tions as predictive markers in node-positive breast cancer patients randomly treated either with an anthracycline-based therapy or with cyclophosphamide, methotrexate, and 5-fluorouracil. Clin Cancer Res 8:1107-1116, 2002 17. Bartlett JM, Munro A, O’Malley FP, et al: Duplication of chromosome 17 CEP predicts for anthracycline benefit: Evidence from an international meta-analysis of 4 adjuvant breast cancer trials for the HER2/TOP2A meta-analysis study group. EJC Suppl 8:121, 2010 18. Ma Y, Lespagnard L, Durbecq V, et al: Polysomy 17 in HER-2/neu status elaboration in breast cancer: Effect on daily practice. Clin Cancer Res 11:4393-4399, 2005 19. Piccart MJ, Di Leo A, Beauduin M, et al: Phase III trial comparing two dose levels of epirubicin combined with cyclophosphamide with cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer. J Clin Oncol 19:3103-3110, 2001 20. Ejlertsen B, Mouridsen HT, Jensen MB, et al: Improved outcome from substituting methotrexate with epirubicin: Results from a randomised comparison of CMF versus CEF in patients with primary breast cancer. Eur J Cancer 43:877-884, 2007 21. McShane LM, Altman DG, Sauerbrei W, et al: REporting recommendations for tumour MARKer Prognostic Studies (REMARK). Br J Cancer 93:387391, 2005 22. Fraser JA, Reeves JR, Stanton PD, et al: A role for BRCA1 in sporadic breast cancer. Br J Cancer 88:1263-1270, 2003 23. Fraser M, Leung BM, Yan X, et al: p53 is a determinant of X-linked inhibitor of apoptosis protein/Aktmediated chemoresistance in human ovarian cancer cells. Cancer Res 63:7081-7088, 2003 24. McClendon AK, Osheroff N: DNA topoisomerase II, genotoxicity, and cancer. Mutat Res 623: 83-97, 2007 JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

Predicting Anthracycline Benefit

25. Olivier M, Petitjean A, Marcel V, et al: Recent advances in p53 research: An interdisciplinary perspective. Cancer Gene Ther 16:1-12, 2009 26. Greenberg RA: Recognition of DNA double strand breaks by the BRCA1 tumor suppressor network. Chromosoma 117:305-317, 2008 27. Arriola E, Marchio C, Tan DS, et al: Genomic analysis of the HER2//TOP2A amplicon in breast cancer and breast cancer cell lines. Lab Invest 88:491-503, 2008 28. Muss HB, Thor AD, Berry DA, et al: c-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 330:1260-1266, 1994 29. Del Mastro L, Bruzzi P, Nicolo G, et al: HER2 expression and efficacy of dose-dense anthracyclinecontaining adjuvant chemotherapy in breast cancer patients. Br J Cancer 93:7-14, 2005 30. Di Leo A, Biganzoli L, Claudino W, et al: Topoisomerase II alpha as a marker predicting anthracyclines’ activity in early breast cancer patients:

Ready for the primetime? Eur J Cancer 44:27912798, 2008 31. Di Leo A, Chan S, Paesmans M, et al: HER-2/ neu as a predictive marker in a population of advanced breast cancer patients randomly treated either with single-agent doxorubicin or single-agent docetaxel. Breast Cancer Res Treat 86:197-206, 2004 32. Paik S, Bryant J, Park C, et al: erbB-2 and response to doxorubicin in patients with axillary lymph node- positive, hormone receptor-negative breast cancer. J Natl Cancer Inst 90:1361-1370, 1998 33. Paik S, Bryant J, Tan-Chiu E, et al: HER2 and choice of adjuvant chemotherapy for invasive breast cancer: National Surgical Adjuvant Breast and Bowel Project protocol B-15. J Natl Cancer Inst 92:19911998, 2000 34. Pritchard KI, Messersmith H, Elavathil L, et al: HER-2 and topoisomerase II as predictors of response to chemotherapy. J Clin Oncol 26:736-744, 2008

35. Bartlett JMS, Munro A, Dunn JA, et al: Chromosome 17 polysomy (Ch17) as a predictor of anthracycline response: Emerging evidence from the UK NEAT adjuvant breast cancer trial. Cancer Res 69:74S, 2009 36. Bartlett JM, Munro A, Cameron DA, et al: Type 1 receptor tyrosine kinase profiles identify patients with enhanced benefit from anthracyclines in the BR9601 adjuvant breast cancer chemotherapy trial. J Clin Oncol 26:5027-5035, 2008 37. Mueller RE, Parkes RK, Andrulis I, et al: Amplification of the TOP2A gene does not predict high levels of topoisomerase II alpha protein in human breast tumor samples. Genes Chromosomes Cancer 39:288-297, 2004 38. Press MF, Sauter G, Buyse M, et al: Alteration of topoisomerase II-alpha gene in human breast cancer: Association with responsiveness to anthracycline-based chemotherapy. J Clin Oncol 29:859-867, 2011

■ ■ ■

GLOSSARY TERMS

anthracyclines: a class of antineoplastic agents derived from Streptomyces bacterium used to treat a variety of hematologic and solid malignancies. Anthracyclines have a well-established doserelated risk of cardiomyopathy and congestive heart failure. Anthracyclines include agents like daunorubicin, doxorubicin, epirubicin, and idarubicin.

www.jco.org

fluorescent in situ hybridization (FISH): in situ hydridization is a sensitive method generally used to detect specific gene sequences in tissue sections or cell preparations by hybridizing the complementary strand of a nucleotide probe to the sequence of interest. FISH uses a fluorescent probe to increase the sensitivity of in situ hybridization.

© 2015 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at NEW YORK UNIVERSITY MED CTR on May 19, 2015 Copyright © 2015 Americanfrom Society of Clinical Oncology. All rights reserved. 128.122.253.212

1687

Predicting Anthracycline Benefit: TOP2A and CEP17-Not Only but Also.

Evidence supporting the clinical utility of predictive biomarkers of anthracycline activity is weak, with a recent meta-analysis failing to provide st...
153KB Sizes 4 Downloads 6 Views