半定量检测雌激素受体表达情况对于乳腺导管癌和小叶癌辅助内分泌治疗疗效的影响:一项TEAM研究分析

Influence of semi-quantitative oestrogen receptor expression on adjuvant endocrine therapy efficacy in ductal and lobular breast cancer – A TEAM study analysis
作者:Willemien van de Watera, h, Duveken B.Y. Fonteina,
期刊: EUR J CANCER2013年6月期卷

 

Influence of semi-quantitative oestrogen receptor expression on adjuvant endocrine therapy efficacy in ductal and lobular breast cancer – A TEAM study analysis 

  • Willemien van de Waterah
  • Duveken B.Y. Fonteinah
  • Johanna G.H. van Nesa
  • John M.S. Bartlettbe,
  • Elysée T.M. Hillea
  • Hein Puttera
  • Tammy Robsonb
  • Gerrit-Jan Liefersa
  • Rudi M.H. Roumenc
  • Caroline Seynaeved
  • Luc Y. Dirixf
  • Robert Paridaensg
  • Elma Meershoek-Klein Kranenbarga
  • Johan W.R. Nortiera,
  • Cornelis J.H. van de VeldeaCorresponding author contact information
  • a Leiden University Medical Center, Leiden, The Netherlands
  • b Edinburgh University, Edinburgh, United Kingdom
  • c Máxima Medical Centre, Veldhoven, The Netherlands
  • d Erasmus MC Daniel Den Hoed, Rotterdam, The Netherlands
  • e Ontario Institute for Cancer Research, Toronto, Canada
  • f St. Augustinus Algemeen Ziekenhuis, Wilrijk, Belgium
  • g UZ Gasthuisberg, Leuven, Belgium
  • http://dx.doi.org/10.1016/j.ejca.2012.07.026, How to Cite or Link Using DOI

Abstract

Background

Multiple studies suggest better efficacy of chemotherapy in invasive ductal breast carcinomas (IDC) than invasive lobular breast carcinomas (ILC). However, data on efficacy of adjuvant endocrine therapy regimens and histological subtypes are sparse. This study assessed endocrine therapy efficacy in IDC and ILC. The influence of semi-quantitative oestrogen receptor (ER) expression by Allred score was also investigated.

Methods

Dutch and Belgian patients enrolled in the Tamoxifen Exemestane Adjuvant Multinational (TEAM) trial were randomized to exemestane (25 mg daily) alone or following tamoxifen (20 mg daily) for 5 years. Inclusion was restricted to IDC and ILC patients. Histological subtype was assessed locally; ER expression was centrally reviewed according to Allred score (ER-poor (<7; n = 235); ER-rich (⩾7; n = 1789)). Primary end-point was relapse-free survival (RFS), which was the time from randomization to disease relapse.

Findings

Overall, 2140 (82%) IDC and 463 (18%) ILC patients were included. RFS was similar for both endocrine treatment regimens in IDC (hazard ratio (HR) for exemestane was 0.83 (95% confidence interval (CI) 0.67–1.03)), and ILC (HR 0.69 (95% CI 0.45–1.06)). Irrespective of histological subtype, patients with ER-rich Allred scores allocated to exemestane alone had an improved RFS (multivariable HR 0.71 (95% CI 0.56–0.89)). In contrast, patients with ER-poor Allred scores allocated to exemestane had a worse RFS (multivariable HR 2.33 (95% CI 1.32–4.11)). Significant effect modification by ER-Allred score was confirmed (multivariable p = 0.003).

Interpretation

Efficacy of endocrine therapy regimens was similar for IDC and ILC. However, ER-rich patients showed superior efficacy to upfront exemestane, while ER-poor patients had better outcomes with sequential therapy, irrespective of histological subtype, emphasising the relevance of quantification of ER expression.

Funding

The Dutch/Belgian TEAM trial is supported by an unrestricted research grant from Pfizer. The sponsors have in no way influenced the design and conduct of the study, collection, management, analysis and interpretation of the data and preparation, review or approval of the manuscript.

Keywords

  • Breast cancer
  • Endocrine therapy
  • Lobular carcinoma
  • Oestrogen receptors

1. Introduction

Treatment decisions in breast cancer therapy are increasingly dependent on a growing number of molecular tumour characteristics, consequently paving the way towards individual targeted treatment regimens. Invasive ductal breast carcinoma (IDC), is by far the most common type of breast cancer.1 The second largest group, found in 10–15% of all female breast cancers is invasive lobular breast carcinoma (ILC); ILC patients tend to be older, and tumours are frequently larger in size, oestrogen and progesterone receptor (ER and PR) positive and human epidermal growth factor receptor 2 (HER2), p53 and epidermal growth factor receptor (EGFR) negative when compared to IDC.2 ILC is also frequently associated with higher ER expression than IDC.3 and 4

With regard to endocrine therapy, several reports have endorsed the association between higher levels of ER expression and the prediction of efficacy of endocrine therapy.5, 6 and 7 However, the association between histological subtype and efficacy of adjuvant endocrine therapy is seldom addressed, as reports have frequently combined IDC and ILC in these efficacy analyses. In 1987, Smith et al. demonstrated similar survival after endocrine treatment with tamoxifen or ovarian ablation in IDC and ILC.8 Another study investigated pathologic parameters and adjuvant tamoxifen response in IDC and ILC. But interaction analyses did not show significant differences in response to tamoxifen in IDC compared with ILC, and inclusion was restricted to premenopausal patients of which less than 2% were treated with adjuvant chemotherapy.4 A study by Rakha and colleagues, on the other hand, demonstrated a better response to various forms of endocrine therapy (ET) (tamoxifen, aromatase inhibitors, ovarian ablation or high-dose progestins) in ILC than IDC.9 Thus, data comparing treatment response between the two major subtypes are scarce and inconsistent, and lack data on quantitative hormone receptor expression.

Upfront treatment with aromatase inhibitors (AI) or sequential therapy with tamoxifen followed by an AI for a total of 5 years is currently the most widely applied endocrine treatment regimen for postmenopausal hormone receptor positive (HR+) early breast cancer patients.10 However, recommendations on prescribing adjuvant endocrine therapy make no distinction between IDC and ILC or quantitative hormone receptor expression.11 In a recent report of the Tamoxifen Exemestane Adjuvant Multinational (TEAM) trial by Bartlett and colleagues, a non-significant benefit for exemestane compared with tamoxifen was established in patients with high ER expression based on Allred scores, although follow-up was limited to 2.75 years.12 The TransATAC (Arimidex, Tamoxifen Alone or in Combination) trial also demonstrated a significant improvement in time to relapse with respect to increasing ER expression in patients treated with anastrozole, while this could not be established for tamoxifen-treated patients.13 In ER poor patients, recurrence rates at 2 years of follow up were lower for those allocated tamoxifen versus anastrozole, but by about 5 years, rates were similar. No analyses by histological subtype were performed.

 

Current evidence is too inconsistent to establish a preferential treatment for histological subtypes in clinical practice. The reported variability in evidence regarding outcome after ET in IDC and ILC may be due to relatively small numbers of included ILC cases, as well as the retrospective nature of many studies previously reported. As ILC is more often ER-positive and frequently associated with stronger ER expression than IDC, we investigated whether the histological subtype would translate to differential efficacies of two endocrine treatment regimens in Dutch and Belgian patients included in the TEAM trial on the basis of quantitative ER expression using the Allred score.12 and 14

2. Methods

2.1. International TEAM trial design

The TEAM trial is a randomized, adjuvant, phase 3, multinational, open label study conducted in postmenopausal women with oestrogen and/or progesterone receptor positive tumours. Patients were randomized to receive either exemestane 25 mg once-daily for 5 years or tamoxifen 20 mg once-daily for 2.5–3 years, followed by exemestane 25 mg once-daily for 2.5–2 years, for a total of 5 years. Participants were enrolled in Belgium, the Netherlands, United Kingdom, Ireland, United States of America, Japan, Greece, Germany and France. Eligibility criteria have been published in earlier reports.10 and 15 In short, postmenopausal patients with histologically confirmed breast adenocarcinoma, who completed local therapy with curative intent, i.e. without evidence of metastatic disease, were eligible. Appropriate approvals from the ethical committees were obtained. All patients provided written informed consent, and additional consent was obtained from patients in the tamoxifen group for the switch to exemestane.

2.2. Design of the current study

Inclusion in the present study was restricted to Dutch and Belgian patients, due to the central review of tumour characteristics in this population (n = 3167). Patients with a lobular or ductal tumour subtype (n = 2603) were eligible for inclusion. The remaining patients had invasive mixed (n = 134, 4.2%), unspecified (n = 24, 0.8%) and unknown (n = 406, 12.8%) subtypes. The database was locked on 23rd February 2011. Data collection was performed locally on specific study case report forms, and transferred to the central data center in Leiden, the Netherlands. Histological subtype was determined by the local pathologist conform international criteria as described in the literature.16 The use of E-cadherin was not used routinely. Auditing of pathology reports was performed centrally in Leiden. ER and PR expression was reassessed and quantified centrally in Edinburgh, United Kingdom. Detailed information is provided in a recent report by Bartlett and colleagues.12 Oestrogen and progesterone receptor expression were evaluated as binary categories according to the Allred score, with PR and ER cutoff at the median in line with those reported by Bartlett et al. (PR-poor ⩽4 versus PR-rich ⩾5; ER-poor ⩽6 versus ER-rich ⩾7).12 Patients with unknown ER Allred scores (n = 108) were not included in the survival analyses by Allred score, irrespective of their PR score. To assess the robustness of the findings, additional analyses were performed using the ER histoscore.12 ER histoscore was dichotomized at the median. ER receptor expression and histoscore could be determined for 2461 patients (94.5%).

2.3. End-points

The primary end-point of the current study was relapse free survival (RFS), which was defined as the time from randomization to the earliest documentation of disease relapse. Disease relapse was defined as tumour recurrence (locoregional or distant), ipsilateral or contralateral breast cancer. Ductal carcinoma in situ was not considered a relapse.10

2.4. Statistical analysis

Statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL) and R statistical package. To compare differences in proportions between patients with lobular and ductal breast tumours, Pearson chi-square test and the Independent sample t-test were used. Kaplan–Meier curves were plotted and Cox proportional hazard models were used to assess survival differences with respect to treatment arm. Survival analyses were stratified by histological subtype and ER Allred score. Interaction terms between treatment arm and histological subtype, and between treatment arm and ER Allred score respectively, were included in the Cox proportional hazard model to test for the presence of effect modification. Multivariable survival analyses by histological type were adjusted for age, histological grade (Bloom Richardson grade 1;2;3), T-stage (1;2;3,4), nodal stage (negative; positive), ER Allred score (ER-rich; ER-poor), PR Allred score (PR-rich; PR-poor), HER2 status (positive; negative), most extensive surgery (mastectomy; wide local excision), radiotherapy (yes; no), and chemotherapy (yes; no). Multivariable analyses by ER Allred score were adjusted for age, histological grade (Bloom Richardson grade 1;2;3), T-stage (1;2;3,4), nodal stage (negative; positive), PR Allred score (PR-rich; PR-poor), HER2 status (positive; negative), most extensive surgery (mastectomy; wide local excision), radiotherapy (yes; no), and chemotherapy (yes; no). To evaluate the interaction between randomized treatment and ER expression in more detail, we performed a sliding window subpopulation treatment effect pattern plot (STEPP) analysis. 17 and 18 All statistical tests were two-sided. Ap-value of <0.05 was considered statistically significant.

3. Results

Overall, 2603 patients from the Netherlands and Belgium who were included in the multinational TEAM trial, were included in this analysis, of which 2140 (82%) presented with IDC and 463 (18%) with ILC. Baseline characteristics of the included patients are shown by histological subtype in Table 1. Patients with a lobular subtype were significantly older (p = 0.004), had larger tumours (p < 0.001) and more frequently were node positive (p = 0.034) and had an unknown tumour differentiation grade (p < 0.001). Moreover, patients with ILC more frequently underwent a mastectomy (p < 0.001) and received less chemotherapy (p = 0.024). No differences were observed for other therapies applied.

 

Table 1. Patient characteristics by histological subtype. The bold values are those characteristics that were statistically significantly different between ductal and lobular patients.

  Ductal (n = 2140)
Lobular (n = 463)
p Value
  n % n %  
Age (mean, SD) 65.0 9.2 66.3 9.4 0.004
 
Grade (BR)
 Well (G1) 318 14.9 58 12.5 <0.001
 Moderate (G2) 897 41.9 236 51.0  
 Poor (G3, G4) 848 39.6 88 19.0  
 Unknown 77 3.6 81 17.5  
 
T stage         <0.001
 1 1041 48.6 133 28.7  
 2 1002 46.8 251 54.2  
 3, 4 96 4.5 77 16.6  
 Unknown 0 0.0 2 0.4  
 
N stage         0.034
 Negative 662 30.9 120 25.9  
 Positive 1475 69.0 343 74.1  
 Unknown 3 0.1 0 0.0  
 
ER Allred score         0.724
 Low 235 11.0 56 12.1  
 High 1789 83.6 380 82.1  
 Unknown 116 5.4 27 5.8  
 
PR Allred score⁎⁎         0.062
 Low 503 25.1 88 20.8  
 High 1498 74.9 336 79.2  
 Unknown 139 6.5 39 8.4  
 
HER2         <0.001
 Positive 276 12.9 21 4.5  
 Negative 1719 80.3 394 85.1  
 Unknown 145 6.8 48 10.4  
 
Most extensive surgery         <0.001
 Mastectomy 1104 51.6 332 71.7  
 WLE 1036 48.4 131 28.3  
 
Adjuvant radiotherapy         0.397
 Yes 1348 63.0 282 60.9  
 No 789 36.9 181 39.1  
Unknown 3 0.1 0 0.0  
 
Adjuvant chemotherapy         0.024
 Yes 698 32.6 126 27.2  
 No 1441 67.4 337 72.8  
 
Randomization         0.152
 Exemestane 1096 51.2 220 47.5  
 Sequential 1044 48.8 243 52.5  

BR, Bloom Richardson; ER, estrogen receptor; PR, progesterone receptor; WLE, wide local excision.

Low: <7; High: ⩾7.

⁎⁎

Low: <5; High: ⩾5.

At database lock, there were 338 (16%) relapses in patients with IDC and 83 (18%) in patients with ILC. Five-year RFS was 86% for IDC and 83% for ILC (IDC functioned as a reference, ILC univariate hazard ratio (HR) 1.15 (95% confidence interval (CI) 0.90–1.46), p = 0.257; ILC multivariable HR 1.18 (95% CI 0.89–1.57),p = 0.257). Table 2 shows efficacy analyses stratified by histological subtypes. In patients with IDC, RFS was not significantly different for either treatment regimen (sequential therapy functioned as a reference, univariate HR for exemestane: 0.83 (95% CI 0.67–1.03), p = 0.089; multivariable HR 0.90 (95% CI 0.72–1.13), p = 0.373). Comparable results were observed in patients with ILC (univariate HR 0.69 (95% CI 0.45–1.06), p = 0.089; multivariable HR 0.61 (95% CI 0.37–1.01), p = 0.053). In order to assess whether efficacy analyses were modified by histological subtype, we included an interaction term between randomization and histological subtype. No significant interaction was observed (multivariable p = 0.148).

Table 2. Relapse free survival by treatment regimen, stratified by histological subtype.

  5 years survival Univariate HR (95% CI) p Value Multivariable⁎⁎ HR (95% CI) p Value
IDC     0.089   0.373
 Tam → Exe 84% 1 (reference)   1 (reference)  
 Exemestane 87% 0.83 (0.67–1.03)   0.90 (0.72–1.13)  
 
ILC     0.089   0.053
 Tam → Exe 80% 1 (reference)   1 (reference)  
 Exemestane 86% 0.69 (0.45–1.06)   0.61 (0.37–1.01)  

IDC, invasive ductal breast carcinoma; ILC, invasive lobular breast carcinoma; HR, hazard ratio; CI, confidence interval.

Tamoxifen followed by exemestane.

⁎⁎

HR adjusted for age, histological grade, T stage, nodal stage, estrogen receptor (ER) Allred score, progesterone receptor (PR) Allred score, HER2 status, most extensive surgery, radiotherapy, and chemotherapy.

The number of patients having an ER-rich tumour was 1789 (89.5%) in IDC patients, and 380 (87.6%) in ILC patients (p = 0.234). Median ER Allred scores were similar in IDC and ILC patients (7 versus 7), mean ER Allred score was slightly lower in IDC patients (6.89 versus 6.98, p = 0.042). Due to the fact that a previous study showed better RFS in patients with ER-rich Allred scores, and because patients with ILC more often present with ER-positive disease, survival by ER Allred score was assessed in patients with ductal and lobular subtypes separately.

For both histological subtypes, ER-rich Allred scores showed superior outcomes for upfront exemestane in both univariate and multivariable analyses (Web Table 1). In contrast, in both IDC and ILC patients with ER-poor Allred scores, those who were allocated exemestane had a worse relapse free survival. No significant effect modification by histological subtype was observed for ER-rich Allred scores (p = 0.201) and ER-poor Allred scores (p = 0.799).

Because of the absence of effect modification, survival analyses were repeated for all patients. Cumulative incidence of breast cancer relapse by treatment regimen for both patients with ER-rich and ER-poor tumours is shown in Fig. 1 (Web Table 2). As shown in Table 3, patients with ER-rich Allred scores and allocated 5 years of exemestane showed a more than 30% improved RFS when compared to patients allocated sequential treatment, which was confirmed in multivariable analyses (HR for exemestane: 0.72 (95% CI 0.57–0.90), p = 0.004). In contrast, ER-poor Allred scores in combination with allocation to exemestane monotherapy was associated with a worse RFS, which was even more pronounced after multivariable analyses (HR for exemestane: 1.85 (95% CI 1.11–3.09), p = 0.019). ER Allred score was a significant effect modifier of endocrine therapy efficacy in all patients (multivariable p for interaction = 0.003). Similar analyses were performed for PR Allred score, but no significant effect modification was observed (p for interaction = 0.137).

Full-size image (22 K)

Fig. 1. Cumulative breast cancer relapse by estrogen receptor Allred score.

 

Table 3. Relapse free survival by treatment regimen, stratified by estrogen receptor (ER) Allred score.

  5 years survival Univariate HR (95% CI) p Value Multivariable⁎⁎ HR (95% CI) p Value
ER-rich     0.001   0.004
 Tam → Exe 80% 1 (reference)   1 (reference)  
 Exemestane 73% 0.69 (0.56-0.86)   0.72 (0.57-0.90)  
 
ER-poor     0.031   0.019
 Tam → Exe 82% 1 (reference)   1 (reference)  
 Exemestane 74% 1.66 (1.05–2.63)   1.85 (1.11–3.09)  

HR, hazard ratio; CI, confidence interval.

Tamoxifen followed by exemestane.

⁎⁎

HR adjusted for age, histological subtype, histological grade, T stage, nodal stage, progesterone receptor (PR) Allred score, HER2 status, most extensive surgery, radiotherapy and chemotherapy. ER-rich was defined as an Allred score of ⩽6; ER-poor was defined as an Allred score of ⩾7.

To assess the robustness of the findings, additional analyses were performed using the ER histoscore instead of the ER Allred score. ER histoscore was dichotomized at the median, which resulted in 1231 patients with ER-rich tumours and 1230 patients with ER-poor tumours. Results are shown in Web Table 3. In both univariate and multivariable analyses, patients with a high histoscore derived a significant benefit from exemestane monotherapy (multivariable HR for exemestane was 0.56 (95% CI 0.39–0.80), p = 0.001). No difference was observed for patients with a low histoscore (multivariable HR for exemestane was 1.04 (95% CI 0.80–1.34), p = 0.779). Nevertheless, significant effect modification by histoscore (p = 0.005) was observed. Additionally, a sliding window STEPP analysis was performed (Fig. 2), which showed a dose-response relation. Moreover, only those patients with the lowest ER histoscores derived benefit from a sequential therapy regimen, while with increasing histoscore, the benefit of exemestane monotherapy increased. Lastly, as ILC patients received less chemotherapy, sensitivity analyses were also performed restricted to patients without prior chemotherapy. These analyses revealed similar results (data not shown).

Full-size image (16 K)

Fig. 2. Sliding window STEPP analysis.

4. Discussion

This study showed no preferential benefit from endocrine therapy based on histological subtype in patients with either IDC or ILC. However, a major finding of the present study was that efficacy of endocrine therapy regimens was modified by ER Allred scores: an improved RFS was demonstrated in ER-rich patients when treated with 5 years of exemestane, whereas ER-poor patients benefited more from the sequential treatment strategy.

Several analyses were performed to assess sensitivity of the used variables and robustness of our findings. First, survival analyses were repeated with ER histoscore instead of ER Allred score, which revealed comparable results. Next, a sliding window STEPP analysis was performed, which showed a dose-response relation between increasing ER histoscore and benefit of exemestane monotherapy. Lastly, analyses were restricted to patients without prior chemotherapy, which did not alter the results.

Despite the numerous differences between IDC and ILC, histological subtype did not modify the efficacy of treatment regimen. As literature suggests different distributions of ER Allred scores for IDC and ILC patients,3 and 4 survival analyses by Allred score for both IDC and ILC separately were performed. Again, similar effects in both IDC and ILC patients were observed, in which higher ER Allred scores were predictive of a better response to exemestane monotherapy, whereas a sequential treatment regimen was preferred in patients with lower ER Allred scores. These findings support the supposition of at least a similar efficacy of endocrine treatment in IDC and ILC. Of note, a prior publication by Pestalozzi et al. revealed changes in survival advantage for IDC and ILC over a 25-year period; up to 6 years, there is a survival benefit for ILC, after which the advantage changes to IDC.19 It is therefore important to interpret our results in the context of longer-term findings.

Regardless of histological subtype, a significant effect modification by ER Allred score was ascertained. Patients with ER-rich tumours responded better to 5 years of exemestane, and patients with ER-poor tumours were better off with the sequential treatment regimen. It is tempting to speculate on the mechanisms which may explain the results observed. Although not the case in all patients, ER expression can decrease during tamoxifen treatment, also one of the mechanisms that contribute to reduced tolerability to endocrine treatment over time.20 and 21 We may thus contemplate that more rapid exhaustion of ER expression leads to an improved response to a new mechanism that counters the effect of oestrogens on tumour growth, namely through the inhibition of aromatase. Therefore, in ER-poor tumours one can argue that these more readily develop endocrine resistance than ER-rich tumours, and thus the mechanism of switching therapy is more relevant than the therapy itself.

A recent report by Bartlett and colleagues in the international TEAM data suggested a beneficial effect of exemestane for ER-rich patients.12 However, in contrast to the present investigation, no significant treatment effect modification by ER Allred score was observed. This may be explained by the restriction to the two major histological subtypes in the current study. Second, the present findings are based on the Dutch and Belgian TEAM participants only, who may display slightly different characteristics than the combined international data.15 However, as nearly one third of the patients included in the international study are of Dutch/Belgian origin, this is not likely to have a major influence on our findings. Third, the analyses by Bartlett and colleagues were limited to 2.75 years of follow-up, whereas the median follow-up of the current study was 5.2 years.

Data are conflicting on the topic of prognostic value of quantitative hormone receptor expression. Previously, Mauriac and colleagues described prognostic risk factors that are associated with disease relapse, based on findings in the Breast International Group (BIG) 1-98 trial.22 In these analyses, a beneficial effect of letrozole versus tamoxifen was found in patients with poorer prognoses (in terms of tumour burden and aggressiveness), regardless of the inclusion of quantitative ER expression in the analysis.22 In the same cohort of patients, Viale et al. found that patients with lower levels of ER (<80%) had more benefit of letrozole monotherapy than patients with higher levels of ER (⩾80%),23 while this could not be established in another report by Regan et al.24 In contrast, combined analyses of the ABCSG 8 and ARNO 95 trials concluded that treatment efficacy with anastrozole improved with increasing ER expression, while this was not the case in tamoxifen-treated patients.25

While our results remain to be confirmed in other studies, there is reason to believe that the differential response based on ER expression is of clinical relevance. The importance of quantitative associations between hormone receptor content and response to endocrine therapy has been documented in the past, and several semi-quantitative scoring systems have been proposed in order to reduce the number of borderline cases of ER-positive tumours. However, it is still infrequently used to determine treatment regimen clinically.

A major strength of the present investigation is the relatively large cohort of patients with ILC as well as the randomized design of this study, which assures quality controlled data and an extended record of patient and treatment characteristics. More importantly, the centrally reviewed ER expression excluded variation between pathology centres. Since histological subtype was assessed locally, we can not exclude a potential bias in the inclusion of our patients. Inclusion in the TEAM study was restricted to ER-and/or PR-positive breast tumours, therefore it was not possible to elaborate on the proportion of ILC patients with ER-positive tumours compared to IDC patients. It is known that ILC patients more frequently present with ER-positive tumours, which predicts a superior response to endocrine therapy.26, 27, 28 and 29 Although a significant effect modification by ER expression was observed, the sliding window STEPP analysis indicated that only those patients with very low ER expression may be better off with a sequential treatment, whereas patients with very high ER expression derive benefit from exemestane monotherapy. This is in line with the results of our Cox regression analyses, in which the effect in ER-rich patients is more outspoken than in ER-poor patients. The retrospective, unplanned nature of this analysis renders our investigation hypothesis generating, and these results will need to be validated through further investigation prospectively in order to define a possible ER expression cutoff for treatment preference.

There was no difference in efficacy of endocrine treatment between IDC and ILC, and therefore endocrine therapy remains warranted in ILC patients. Instead, the extent of ER positivity may be superior in predicting treatment efficacy than histological subtype. In other words, our hypothesis was not confirmed on the basis of histological subtypes. Alternatively, an effect modification for treatment modality by ER expression in both subtypes was established. The semi-quantitative analysis of ER expression in postmenopausal early breast cancer patients may therefore be an important item in clinical decision-making to determining the optimal therapeutic regimen.

Conflict of interest statement

None declared.

Acknowledgements

The authors would like to thank Pfizer, and The Dutch Cancer Society (KWF 2007-3968).

 

 

Appendix A. Supplementary data

 

Supplementary data 1.  Webtable 1. Relapse free survival by treatment regimen, stratified by ER Allred score and histological subtype; Webtable 2. Relapse free survival by treatment regimen, stratified by ER histoscore (dichotomized at the median).

 

 

 

References

    • 1
    • A.R. Dixon, I.O. Ellis, C.W. Elston, R.W. Blamey
    • A comparison of the clinical metastatic patterns of invasive lobular and ductal carcinomas of the breast

    • Br J Cancer, 63 (4) (1991), pp. 634–635

    •  | 
    • 2
    • G. Arpino, V.J. Bardou, G.M. Clark, R.M. Elledge
    • Infiltrating lobular carcinoma of the breast: tumor characteristics and clinical outcome

    • Breast Cancer Res, 6 (3) (2004), pp. R149–R156

    • 3
    • M. Colleoni, G. Viale, A. Goldhirsch
    • Lessons on responsiveness to adjuvant systemic therapies learned from the neoadjuvant setting

    • Breast, 18 (Suppl 3) (2009), pp. S137–S140

    •  |   | 
    • 4
    • K. Jirstrom, L. Ryden, L. Anagnostaki et al.
    • Pathology parameters and adjuvant tamoxifen response in a randomised premenopausal breast cancer trial

    • J Clin Pathol, 58 (11) (2005), pp. 1135–1142

    • 5
    • A. Goldhirsch, J.N. Ingle, R.D. Gelber et al.
    • Thresholds for therapies: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2009

    • Ann Oncol, 20 (8) (2009), pp. 1319–1329

    •  | 
    • 6
    • J.M. Harvey, G.M. Clark, C.K. Osborne, D.C. Allred
    • Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer

    • J Clin Oncol, 17 (5) (1999), pp. 1474–1481

    • 7
    • International Breast Cancer Study Group (IBCSG)
    • Endocrine responsiveness and tailoring adjuvant therapy for postmenopausal lymph node-negative breast cancer: a randomized trial

    • J Natl Cancer Inst, 94 (14) (2002), pp. 1054–1065

    • 8
    • D.B. Smith, A. Howell, J. Wagstaff
    • Infiltrating lobular carcinoma of the breast: response to endocrine therapy and survival

    • Eur J Cancer Clin Oncol, 23 (7) (1987), pp. 979–982

    •  |   | 
    • 9
    • E.A. Rakha, M.E. El-Sayed, D.G. Powe et al.
    • Invasive lobular carcinoma of the breast: response to hormonal therapy and outcomes

    • Eur J Cancer, 44 (1) (2008), pp. 73–83

    •  |   | 
    • 10
    • C.J. van de Velde, D. Rea, C. Seynaeve et al.
    • Adjuvant tamoxifen and exemestane in early breast cancer (TEAM): a randomised phase 3 trial

    • Lancet, 377 (9762) (2011), pp. 321–331

    •  |   | 
    • 11
    • A. Goldhirsch, W.C. Wood, A.S. Coates et al.
    • Strategies for subtypes–dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011

    • Ann Oncol, 22 (8) (2011), pp. 1736–1747

    •  | 
    • 12
    • J.M. Bartlett, C.L. Brookes, T. Robson et al.
    • Estrogen receptor and progesterone receptor as predictive biomarkers of response to endocrine therapy: a prospectively powered pathology study in the Tamoxifen and Exemestane Adjuvant Multinational trial

    • J Clin Oncol, 29 (12) (2011), pp. 1531–1538

    •  | 
    • 13
    • M. Dowsett, C. Allred, J. Knox et al.
    • Relationship between quantitative estrogen and progesterone receptor expression and human epidermal growth factor receptor 2 (HER-2) status with recurrence in the Arimidex, Tamoxifen, Alone or in Combination trial

    • J Clin Oncol, 26 (7) (2008), pp. 1059–1065

    •  | 
    • 14
    • A. Qureshi, S. Pervez
    • Allred scoring for ER reporting and it’s impact in clearly distinguishing ER negative from ER positive breast cancers

    • J Pak Med Assoc, 60 (5) (2010), pp. 350–353

    • 15
    • J.G. van Nes, C. Seynaeve, S. Jones et al.
    • Variations in locoregional therapy in postmenopausal patients with early breast cancer treated in different countries

    • Br J Surg, 97 (5) (2010), pp. 671–679

    •  | 
    • 16
    • NHS Cancer Screening Programmes, The Royal College of Pathologists. Guidelines for pathology reporting in breast disease. NHSBSP Publication No 58; 2005.
    •  
    • 17
    • A.A. Lazar, B.F. Cole, M. Bonetti, R.D. Gelber
    • Evaluation of treatment-effect heterogeneity using biomarkers measured on a continuous scale: subpopulation treatment effect pattern plot

    • J Clin Oncol, 28 (29) (2010), pp. 4539–4544

    •  | 
    • 18
    • M. Bonetti, D. Zahrieh, B.F. Cole, R.D. Gelber
    • A small sample study of the STEPP approach to assessing treatment-covariate interactions in survival data

    • Stat Med, 28 (8) (2009), pp. 1255–1268

    •  | 
    • 19
    • B.C. Pestalozzi, D. Zahrieh, E. Mallon et al.
    • Distinct clinical and prognostic features of infiltrating lobular carcinoma of the breast: combined results of 15 International Breast Cancer Study Group clinical trials

    • J Clin Oncol, 26 (18) (2008), pp. 3006–3014

    •  | 
    • 20
    • M.J. Ellis, V.J. Suman, J. Hoog et al.
    • Randomized phase II neoadjuvant comparison between letrozole, anastrozole, and exemestane for postmenopausal women with estrogen receptor-rich stage 2 to 3 breast cancer: clinical and biomarker outcomes and predictive value of the baseline PAM50-based intrinsic subtype–ACOSOG Z1031

    • J Clin Oncol, 29 (17) (2011), pp. 2342–2349

    •  | 
    • 21
    • S.R. Johnston, G. Saccani-Jotti, I.E. Smith et al.
    • Changes in estrogen receptor, progesterone receptor, and pS2 expression in tamoxifen-resistant human breast cancer

    • Cancer Res, 55 (15) (1995), pp. 3331–3338

    • 22
    • L. Mauriac, A. Keshaviah, M. Debled et al.
    • Predictors of early relapse in postmenopausal women with hormone receptor-positive breast cancer in the BIG 1–98 trial

    • Ann Oncol, 18 (5) (2007), pp. 859–867

    •  | 
    • 23
    • G. Viale, M.M. Regan, P. Dell’Orto et al.
    • Which patients benefit most from adjuvant aromatase inhibitors? Results using a composite measure of prognostic risk in the BIG 1–98 randomized trial

    • Ann Oncol, 22 (10) (2011), pp. 2201–2207

    •  | 
    • 24
    • M.M. Regan, K.N. Price, A. Giobbie-Hurder, B. Thurlimann, R.D. Gelber
    • Interpreting Breast International Group (BIG) 1–98: a randomized, double-blind, phase III trial comparing letrozole and tamoxifen as adjuvant endocrine therapy for postmenopausal women with hormone receptor-positive, early breast cancer

    • Breast Cancer Res, 13 (3) (2011), p. 209

    •  | 
    • 25
    • R. Jakesz, W. Jonat, M. Gnant et al.
    • Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial

    • Lancet, 366 (9484) (2005), pp. 455–462

    •  |   | 
    • 26
    • A. Katz, E.D. Saad, P. Porter, L. Pusztai
    • Primary systemic chemotherapy of invasive lobular carcinoma of the breast

    • Lancet Oncol, 8 (1) (2007), pp. 55–62

    •  |   | 
    • 27
    • T. Petit, M. Wilt, M. Velten et al.
    • Semi-quantitative evaluation of estrogen receptor expression is a strong predictive factor of pathological complete response after anthracycline-based neo-adjuvant chemotherapy in hormonal-sensitive breast cancer

    • Breast Cancer Res Treat, 124 (2) (2010), pp. 387–391

    •  | 
    • 28
    • A. Purushotham, S. Pinder, M. Cariati, M. Harries, A. Goldhirsch
    • Neoadjuvant chemotherapy: not the best option in estrogen receptor-positive, HER2-negative, invasive classical lobular carcinoma of the breast?

    • J Clin Oncol, 28 (22) (2010), pp. 3552–3554

    •  | 
    • 29
    • S.A. Farese, S. Aebi
    • Infiltrating lobular carcinoma of the breast: Systemic treatment

    • Breast Dis, 30 (2009), pp. 45–52

学科代码:肿瘤学   关键词:EJC全文 EJC
顶一下(0
您可能感兴趣的文章
    发表评论网友评论(0)
      发表评论
      登录后方可发表评论,点击此处登录