SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Aromatase inhibitor shows efficacy for hormone receptor positive postmenopausal breast cancer. We evaluated the activity of 24 weeks of aromatase inhibition with exemestane for primary breast cancer in a neoadjuvant setting. Patients with stage II/IIIA invasive breast cancer with estrogen receptor (ER) and/or progesterone receptor (PgR)-positive status were eligible. Primary endpoints were objective response rate (ORR) and safety. A steroidal aromatase inhibitor exemestane of 25 mg/day was administered for 16 weeks with an 8-week extension. Secondary endpoints were rates of breast-conserving surgery (BCS), and change of Ki67 index and ER/PgR expression in central laboratory analyses. Between March 2006 and December 2007, 116 patients were enrolled. Among those, 102 patients completed 24 weeks of administration. The ORR was 47% (55/116) at Week 16 and 51% (59/116) at Week 24, respectively. No serious toxicity was seen. ORR was associated with ER Allred scores but not with PgR scores. The significant reduction in Ki67 index was confirmed. No progression was experienced in tumors with less than 15% Ki67 index. Pathological response was observed in 28 (30%) of 94 evaluated cases. No statistical correlation between pre-treatment Ki67 index and pathological response was detected; however, a trend of correlation was found between the post-treatment preoperative endocrine prognostic index (PEPI), a prognostic score and the pathological response. At diagnosis, 59 patients (51%) would have required mastectomy but 40 patients were converted to BCS, showing an increase in the rate of BCS (77%). The 24-week aromatase inhibition provided preferable clinical benefits with significant reduction in Ki67 index. More precise mechanisms of the response need to be investigated. (Cancer Sci 2011; 102: 858–865)

Many studies of neoadjuvant chemotherapy for breast cancer have been conducted. These studies have revealed that neoadjuvant chemotherapy allows more women to undergo breast-conserving surgery (BCS) rather than total mastectomy, and prolongs the survival of patients who achieved pathological complete response (pCR).(1–3) However, it has been described that neoadjuvant chemotherapy has a limited effect in hormone receptor-positive patients in terms of pCR rates, and raises safety concerns for elderly patients.(4–7) Therefore, as a treatment strategy, the efficacy and safety of neoadjuvant hormone therapy using aromatase inhibitors (AI) is being assessed in several trials in postmenopausal breast cancer patients.(8–11)

In a phase II randomized study in which neoadjuvant hormone therapy and neoadjuvant chemotherapy were compared in hormone receptor-positive patients, no significant difference in the clinical response rate was observed between these two groups. Notably, the rate of BCS tended to be higher, and the incidence of adverse events was generally lower in the neoadjuvant hormone therapy group than in the neoadjuvant chemotherapy group.(12) These results suggest the benefit of neoadjuvant hormone therapy in hormone-sensitive postmenopausal breast cancer patients.(13) Therefore, it seems that neoadjuvant hormone therapy offers an alternative to neoadjuvant chemotherapy.

However, there are some concerns surrounding the use of neoadjuvant hormone therapy that need to be addressed. First, tumor regression is slower with neoadjuvant hormone therapy than with chemotherapy. In fact, a study investigating the response rate to 6-month neoadjuvant hormone therapy using exemestane reported that the objective response rate (ORR: complete response [CR] + partial response [PR]) continued to increase even after 4 months of treatment.(14) Another concern is that there is no established index for evaluating the efficacy of neoadjuvant hormone therapy. In neoadjuvant chemotherapy, the pCR rate can be used as a surrogate marker for the prognosis of patients.(2) However, it has been reported that, in estrogen receptor (ER)-positive patients, the proportion of patients who achieved a pCR was not significantly correlated with overall survival (OS) or disease-free survival (DFS).(15) In addition, several Phase II studies of neoadjuvant hormone therapy reported that pCR rates were from 0 to about 3%, which were remarkably lower than those expected from the benefit observed in adjuvant hormone therapy.(8,11,12) Therefore, in hormone receptor-positive breast cancer patients, pCR is unlikely to be a useful marker for assessing efficacy or prognosis. A possible alternative marker for neoadjuvant hormone therapy is the percentage of MIB1/Ki67-positive cells (MIB-1/Ki67 labeling index), a cell proliferative index. The Ki67 index after neoadjuvant hormone therapy was shown to correlate with the recurrence rate.(16,17) However, the usefulness of the Ki67 index has not been fully evaluated.

From these circumstances, we conducted the present study in Japanese patients with hormone receptor-positive postmenopausal breast cancer who received neoadjuvant hormone therapy using exemestane for 24 weeks to assess tumor response and safety of the treatment. We also evaluated the Ki67 index and expression of hormone receptors to determine its potential use as a marker to predict clinical and histopathological response in a central laboratory. Preoperative endocrine prognostic index (PEPI),(16) a prognostic index, was determined in each individual and the relationship with clinical and pathological responses was investigated.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Patients.  Postmenopausal women aged 55–75 years with operable, Stage II or IIIA, histologically confirmed invasive breast cancers were enrolled. Patients were confirmed positive for ER or progesterone receptor (PgR) by immunohistochemical staining (≥10% nuclear staining was defined as positive). Expression of human epidermal growth factor receptor 2 (HER2) was determined immunohistologically with the HercepTest (Dako, Glostrup, Denmark). Positive in HER2 status was defined as either 3+ or 2+ with confirmed c-erbB2 gene amplification by the FISH test. All patients were judged by their primary physicians as having a good performance status (PS; 0–1) and an indication for neoadjuvant hormone therapy after consideration of other treatment options such as surgical therapy and neoadjuvant chemotherapy.

This study was performed in accordance with the Declaration of Helsinki and the Ethical Guidelines for Clinical Research of the Ministry of Health, Labour and Welfare of Japan. Approval was obtained from the institutional review board at each study center. Written informed consent was obtained from all patients before enrolment.

Treatment scheme.  The patients’ lesions were measured by palpation, ultrasound and computed tomography or magnetic resonance imaging. Surgical procedures were determined based on the initial examination; axillary lymph node metastasis was also assessed.

Patients were initially treated with 25 mg of exemestane (Aromasin®; Pfizer Inc. Tokyo, Japan) once daily, orally, for 16 weeks. Clinical response was assessed by comparing the longest diameter of the target lesions with the baseline measurement based on Response Evaluation Criteria in Solid Tumors (RECIST) criteria. Patients with progressive disease (PD) were withdrawn from the study and the remainder continued to receive exemestane for a further 8 weeks, for a total treatment period of 24 weeks. At Week 24, the clinical response was re-evaluated using the same criteria as at Week 16. Patients classified as showing CR, PR or stable disease (SD) at Week 24 underwent surgery as appropriate; patients classified as PD either underwent surgery or commenced another treatment. After surgery, patients classified as CR, PR or SD continued to receive exemestane for postoperative adjuvant hormone therapy for ≥5 years, including the neoadjuvant treatment period. Radiotherapy and drug therapy other than hormone therapy could be given concomitantly at the investigator’s discretion. Postoperative treatment was not pre-specified for patients with PD.

Study end points.  The primary end points were objective response rates (ORR) and safety after 16 and 24 weeks of treatment in intent to treatment analysis. Secondary end points were rates of breast-conserving surgery and mastectomy, nodal status, biomarker changes, pathological response and Allred score. Correlations between the pre-treatment Ki67 labeling index and its changes by treatment and therapeutic effects were also investigated.

Safety assessments.  Adverse events (defined as the development of a new medical condition or the deterioration of a pre-existing medical condition) were recorded every 4 weeks, and were graded according to the National Cancer Institute, Common Toxicity Criteria version 3.0. Pre-specified adverse events were hot flushes, sweating, headache, dizziness, fatigue, nausea/vomiting, appetite loss, weight gain, hypertension, vaginal bleeding, joint pain and bone fracture.

Central biomarker analysis.  In order to determine the suitability for further immunohistochemical (IHC) analyses and then for the evaluation of pathological response, initially, one 4-μm section of each submitted paraffin blocks of pre- and post-treatment specimens of 107 patients who underwent surgery were stained with H&E to verify an adequate number of invasive breast carcinoma cells and the quality of fixation for this study. Serial tissue sections were then prepared from selected blocks and immunohistochemistry was performed to immunolocalize ER, PgR, HER2 and Ki67 as described previously.(18–20) In brief, IHC staining was performed by streptavidin–biotin amplification method using a Histofine Kit (Nichirei, Tokyo, Japan). The Ki67 was stained after overnight preparation using the following antibody dilution: 1:100 (Dako). The ER, PgR and HER2 were stained automatically (Ventana, Tucson, Arizona, USA). The immunostained slides were independently evaluated by three of the authors (NC, TS, HS) who were blinded to clinical outcome of individual patients. The immunoreactivity of ER and PgR was scored by assigning proportion and intensity scores according to Allred’s procedure.(18) The membrane staining pattern was estimated in HER2 immunostaining and scored on a scale of 0–3.(19) Evaluation of Ki67 was performed by counting 1000 carcinoma cells or more from each patient and the percentage of immunoreactivity was subsequently determined by a labeling index.(20)

Pre-operative endocrine prognostic index (PEPI).  According to an algorithm proposed by Ellis’s group, we calculated the total PEPI score for each patient. Briefly, the PEPI score is the sum of the risk points derived from the pathological T stage, pathological nodal stage, Ki67 level and ER Allred score status of the surgical specimen.(16) High PEPI scores correlate with high risk of relapse.

Statistical analysis.  The target sample size of this study (110 patients) was calculated based on clinical data obtained in previous studies of aromatase inhibitors and assumptions regarding the expected number of dropouts. Tumor response was evaluated by summary statistics and calculated together with 95% confidence intervals. The distribution of adverse events was summarized and their incidence rates calculated for each grade of severity (grades 1–4). Univariate and multivariate analyses were performed with a logistic regression model, Pearson’s chi-squared test and multiple logistic regression models, respectively.

Results

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Patients.  Between March 2006 and December 2007, 116 patients were enrolled; their baseline characteristics are inlineed in Table 1. All patients were defined as ER-positive; 80 (68.9%) were PgR-positive and 3(2.5%) were HER2-positive by investigator evaluation. During the first 16 weeks, ten patients discontinued neoadjuvant exemestane treatment because of PD (four patients), investigator decision (one patient), adverse events (two patients, one of whom was not evaluable at Week 16), or the patient’s decision (three patients, two of whom were not evaluable at Week 16) (Fig. 1). A total of 106 patients were included in the 8-week extension and 102 patients completed 24 weeks of exemestane neoadjuvant treatment. Of 102 patients who completed the extension study, 99 underwent surgery.

Table 1.   Patient characteristics
Factorn (%)
  1. ER, estrogen receptor; PgR, progesterone receptor; HER2, human epidermal growth factor receptor 2.

Age, years (median, range) 64 (55–79)
Prior treatmentNone
Tumor stage
 T2110 (95)
 T3  6 (5)
Nodal status
 N0 91 (78)
 N1 23 (20)
 Unknown  3 (2)
Clinical stage
 IIA 89 (77)
 IIB 23 (20)
 IIIA  4 (3)
Tumor diameter, mm (median, range)
 Caliper 32 (12–74)
 Ultrasound 27.3 (15–102)
ER status
 ER+116 (100)
 ER−  0
PgR status
 PgR+ 80 (69)
 PgR− 36 (31)
HER2 status
 HER2+  3 (3)
 HER2−101 (87)
 Not evaluated 12 (10)
image

Figure 1.  Patient registration and the treatment flow of 24 weeks. PD, progressive disease; pts, patients; PR, partial response; SD, stable disease.

Download figure to PowerPoint

Clinical response.  The clinical response was determined by the investigators evaluation based on the combination of caliper measurement and other image modalities such as ultrasound (US), computed tomography (CT) and MRI as defined by protocol. In intent to treat (ITT) analysis with 116 patients, at Week 16, 55 patients (47.4%) achieved PR and 54 patients (46.6%) showed SD. Four patients (3.4%) were considered to have PD (Table 2). The ORR at Week 24 analysis was 50.9%. In detail, no patient achieved CR, 59 (50.9%) achieved PR, 41 (35.3%) had SD and PD was noted in eight patients (6.9%), including four PD cases at Week 16. There was no significant difference in ORR between Weeks 16 and 24 (P = 0.54, McNemar’s test). As a reference, ORR in patients who could complete the 24-week exemestane course was 57.8% (59/102). Although ORR at 24 weeks treatment was about 50%, most patients experienced shrinkage of the tumor with no regard to the evaluation with caliper or ultrasound as shown in the Waterfall plot analysis (Fig. 2).

Table 2.   Clinical response after 16 weeks and 24 weeks of treatment
After 16 weeks evaluation (number of patients)After 24 weeks evaluation (number of patients)
  1. PR, partial response; SD, stable disease; PD, progressive disease.

PR45PR59
SD14  
PR7SD41
SD34  
PR1PD4
SD3  
PR2Not evaluated12
SD3 
PD4 
Not evaluated3  
Total116
image

Figure 2.  Waterfall plot analysis of clinical response at 24 weeks evaluated by caliper and ultra sound (US). Horizontal axis indicates data from each patient and vertical axis, the reduction rate of tumor size evaluated by indicated modality. Negative values on vertical axis indicates tumor progression.

Download figure to PowerPoint

Rate of conversion to breast conserving surgery (BCS).  Based on assessments before neoadjuvant hormone therapy, 59 (50.9%) and 57 (49.1%) of 116 patients were indicated for total mastectomy and BCS, respectively (Table 3). At Week 24, 19 (16.4%) and 89 (76.7%) patients underwent total mastectomy and BCS, showing an increase in the rate of BCS. Of the 59 patients originally indicated for total mastectomy, 14 underwent total mastectomy, 40 were converted to BCS and five received no surgical treatment because of multiple reasons as already described. Of the 57 patients originally indicated for BCS, 49 underwent BCS, five underwent total mastectomy and three received no surgical treatment, respectively. Among five patients whose surgery were converted from BCS to mastectomy after neoadjuvant treatment, four were due to the patient’s preference for mastectomy rather than BCS, and one patient showed progression of the primary tumor.

Table 3.   Rates of breast conserving surgery in pre-treatment estimation and surgery undergone after treatment
  Post treatment (underwent)Total
MastectomyBCSWithout surgery
  1. BCS, breast conserving surgery.

Estimation pre-treatmentMastectomy1440559 (50.9%)
BCS549357 (49.1%)
Total19 (16.4%)89 (76.7%)8 (6.9%)116

Safety.  The most frequently seen adverse events were an abnormal increase in liver enzyme levels (SGOT, SGPT, ALP), hot flushes, joint pain, hypoalbuminuria and elevated creatinine and bilirubin levels. None of these adverse events was deemed to be severe in intensity. The only Grade 3 adverse events were elevated liver enzymes in four cases. No other adverse events of Grade 3 or 4 were noted in this study. Overall, two patients discontinued the study during the initial 16-week phase because of adverse events. One patient had Grade 3 AST and ALT elevations, and the other patient had Grade 2 AST and Grade 3 ALT elevations. No patient discontinued the study during the 8-week extension because of adverse events. Details are described in a Data S1.

Centrally evaluated pathological response.  Tissue sections from 94 patients among 107 surgical specimens, from pretreatment core needle biopsies and final surgical specimens, were available to be assessed for changes in cellularity and degree of fibrosis in H&E stained slides. Pathological response was categorized using the modified criteria previously described by Miller et al.,(21) and assessed as follows: complete when there was no evidence of malignant cell at the original tumor site, partial response when histological decrement in cellularity and/or increment in fibrosis was detected, or no change/non-response. All of the pathological responders were partial response 28 cases (29.8%) while non-responders comprised 66 cases (70.2%).

Centrally evaluated ER/PgR Allred scores and Ki67 labeling index.  Paraffin embedded slides for biomarker studies were submitted to the Department of Pathology, Tohoku University School of Medicine, which served as the central laboratory as described in Patients and Methods. Allred scores of ER and PgR staining before treatment (102 samples and 83 samples, respectively, were available from 116 enrolled patients) were analyzed for evaluating the correlation to clinical response. Clinical objective response was observed in patients with score 5 or greater in ER expression, and had a tendency to increase in higher score group (Fig. 3a). However, it was shown that in any PgR score patients could have a favorable clinical response.

image

Figure 3.  (a) Clinical response rates and centrally evaluated ER and PgR Allred scores. Numbers above the graph indicate total patient counts in each Allred score group. (b) Pathological response rates and centrally evaluated ER and PgR Allred scores. Numbers above the graph indicate total patient counts in each Allred score group.

Download figure to PowerPoint

Allred scores of ER and PgR staining of the same population were also assessed for correlation to pathological response (Fig. 3b). For ER, there was the same tendency that pathological responses were observed in higher Allred score group such as 6 or more. In PgR evaluation, there was no obvious correlation of pathological response to PgR score. Ninety-three pairs of core needle biopsies before treatment and tumor tissues after surgery were applied to Ki67 index evaluation.

Figure 4a shows the scatter plot of pre-treatment and post-treatment Ki67 indices with information of clinical response. Plots located under the curve of y = x indicate the tumors that Ki67 decreased by neo-adjuvant exemestane treatment. At first, there was no correlation between the pre-treatment Ki67 index and clinical responses (PR versus others, P = 0.52). Overall, significant reduction in the Ki67 index was observed at Week 24 compared to the baseline (Median [range]: pre, 11 [0–68]; post, 3 [0–51], P < 0.0001, paired Student’s t-test). Analysis of the Ki67 index according to clinical response revealed that the Ki67 index was significantly decreased in patients with both PR and SD (P < 0.0001 for both). In patients who achieved PR, the median Ki67 index decreased from 9 (range 0–47) to 2 (range 0–37) after neoadjuvant treatment with exemestane, while that in patients with SD decreased from 8 (range 1–68) to 3 (range 0–51). No association was observed between changes in Ki67 index and clinical responses. A noteworthy observation in Figure 4a was that there were no PD patients during the 24-week treatment period, if pretreatment tumor expressed a Ki67 index of 15% or less.

image

Figure 4.  (a) Correlation between pre- and post-treatment Ki67 index and clinical response. PD, progressive disease; PR, partial response; SD, stable disease (n = 93). (b) Correlation between pre-and post-treatment Ki67 index and pathological response. NR, non-response; PR, Partial response (n = 90).

Download figure to PowerPoint

Correlation of Ki67 index to pathological response was also evaluated in the same manner (Fig. 4b). In patients who achieved pathological partial response (PR), the median Ki67 index decreased from 10 (range 0–55) to 2 (range 0–34) after neoadjuvant treatment, while that in patients with NR decreased from 12 (range 1–68) to 4 (range 0–51). Statistically, the Ki67 index dropped significantly in both pathological responders and non-responders (P < 0.0001). There was no statistical correlation between pre-treatment Ki67 index and pathological response. Nevertheless, all cases that showed increases of Ki67 index after the treatment were evaluated as pathological non-responders.

The results of univariate and multivariate analysis with respect to clinical and pathological response are summarized in Table 4. Young age, small tumor size and high ER score were associated with clinical response: PR + SD versus PD.

Table 4.   Univariate and multivariate analysis with respect to clinical and pathological response
(a) Outcome = clinical response (PR versus SD + PD)
VariablesUnivariate analysisMultivariate analysis (full model)Multivariate analysis (stepwise)
P-valueOdds95%CIP-valueOdds95%CIP-valueOdds95%CI
Age0.2330.9630.9041.0240.2950.9590.8841.0370.2330.9630.9041.024
T0.5160.5500.0883.4280.7260.6830.0716.507    
N (N2–3 vs N0)0.8920.9440.4132.1590.3081.7260.6095.166    
ER (score)0.3561.2530.7772.0710.2691.3600.7902.422    
PR (score)0.7691.0300.8421.2610.9010.9860.7831.232    
HER2 (positive versus negative)*0.189   0.2005553.5970.0000.000    
Ki67 index0.2540.9840.9561.0110.3580.9830.0000.000    
Model P-value    0.617   0.233   
Model R2    0.050   0.009   
(b) Outcome = clinical response (PR + SD versus PD)
VariablesUnivariate analysisMultivariate analysis (full model)Multivariate analysis (stepwise)
P-valueOdds95%CIP-valueOdds95%CIP-valueOdds95%CI
Age0.0670.9000.7941.0070.2270.9030.7411.0630.0720.8670.7171.012
T0.0080.1080.0150.7580.0230.0320.0010.6080.0060.0190.0010.306
N (N2–3 vs N0)0.0100.1760.0410.7540.3890.4360.0562.935    
ER (score)0.0242.2271.1174.5580.0093.3181.33911.1360.0024.0461.67412.841
PR (score)0.1611.2250.9151.6030.6491.0950.7081.579    
HER2 (positive versus negative)*0.668   0.6992242.6340.0000.000    
Ki67 index0.0500.9590.9201.0000.0850.9520.0000.0000.1020.9580.9051.009
Model P-value    0.008   0.001   
Model R2    0.367   0.353   
(c) Outcome = pathological response
VariablesUnivariate analysisMultivariate analysis (full model)Multivariate analysis (stepwise)
P-valueOdds95%CIP-valueOdds95%CIP-valueOdds95%CI
  1. PR, partial response; SD, stable disease; PD, progressive disease. *All HER2 positive cases (n = 2) were PR, so that odds calculation could be unstable.

Age0.3370.9650.8931.0380.4980.9690.8791.062    
T0.2520.000  0.2060.0000.0003.280    
N (N2–3 vs N0)0.0580.2970.0801.1010.2740.4660.0000.0000.0460.2970.0650.979
ER (score)0.3291.3250.7672.5450.2821.4530.0000.000    
PR (score)0.5101.0810.8641.4090.7711.0400.0000.000    
HER2 (positive versus negative)0.1565.0000.43457.5470.5812.4470.0000.000    
Ki67 index0.5550.9910.9571.0210.4280.9820.0000.000    
Model P-value    0.379   0.046   
Model R2    0.087   0.035   

The relationship between the PEPI score and responses is described in Table 5. There was no correlation between PEPI score and clinical response (P = 0.99, chi-squared test). Nevertheless, a trend was found that that patients with PEPI score of 4 or more unlikely to have pathological response (P = 0.053, chi-squared test).

Table 5.   Association between tumor response and preoperative endocrine prognostic index
ResponsePEPIP value (chi-squared test)
01–34−
  1. PEPI, preoperative endocrine prognostic index; PR, partial response; SD, stable disease.

Pathological responder9113P = 0.112 (0–3 vs 4−:0.053)
Pathological non-responder142621
Clinical PR132212P = 0.988 (0–3 vs 4−: 0.88)
Clinical SD101710

Discussion

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

The objectives of neoadjuvant hormone therapy for breast cancer are to increase the likelihood for patients to undergo BCS rather than mastectomy, and to expect benefits from a drug that is used in adjuvant therapy. With the introduction of third-generation AI such as anastrozole, exemestane and letrozole, the response rate in hormone-sensitive breast cancer patients has increased. In addition, several clinical studies have reported that neoadjuvant hormone therapy using AI improves the rate of BCS. For example, in the P024 study,(9) 4 months of neoadjuvant hormone therapy using letrozole was compared with tamoxifen in 337 postmenopausal patients with hormone receptor-positive early breast cancer. The ORR and the rate of BCS were significantly higher in the letrozole group (55% and 45%, respectively) than in the tamoxifen group (36% and 35%, respectively). Similarly, in the large-scale PROACT study, 314 patients received only neoadjuvant hormone therapy with anastrozole or tamoxifen for 3 months, and the ORR and rate of BCS were significantly higher in the anastrozole group (49.7% and 43.0%, respectively) than in the tamoxifen group (39.7% and 30.8%, respectively).(10) In the ABCSG-17 study, which used exemestane,(8) the ORR was 34% in hormone receptor-positive breast cancer patients who received 4 months of neoadjuvant treatment with exemestane, and the rate of BCS was 76%.

In the present study, the ORR for exemestane was 47.4% at Week 16 and 50.9% at Week 24 in ITT analysis, which were comparable with the results of the previous studies. Although the response rate at Week 24 was slightly higher than that at Week 16, this difference was not significant. Notably, of the 55 patients with PR at Week 16, 45 patients maintained PR at Week 24 and, of the 54 patients with SD at Week 16, 14 had PR at Week 24 and 35 had SD. These results suggest that 24 weeks of continuous treatment with exemestane induces sustained tumor regression. The response rate in patients who could complete 24 weeks of exemestane was 57.8%.

The proportion of patients suitable for BCS was 49.1% in the evaluation performed before treatment, but improved to 76.7% after 24 weeks of neoadjuvant hormone therapy. Notably, among 59 patients who are initially candidates for mastectomy, 40 patients (67.8%) could undergo BCS. This observation is almost identical to the recent phase II study in which 30 (65%) of 46 patients who were initially marginal for BCS underwent BCS after 16–24 weeks of treatment with letrozole.(22) These improvements seem to be due to universal tumor shrinkage in the majority of the patients, as shown in the Waterfall plot analysis. Toxicity was acceptable. Therefore, treatment with exemestane for 24 weeks was effective in promoting tumor regression and improving BCS rates, with an acceptable tolerability. It has also been reported that continuing letrozole in responding patients beyond 3–4 months achieves further clinical reduction in tumor size.(23) A treatment period of 24 weeks is considered to achieve the efficacy and safety levels required for neoadjuvant therapy.

The ER status is an established predictive factor for the response to endocrine therapy. In the central laboratory evaluation of ER IHC, there was a tendency for higher ER Allred scores in a tumor to correlate with preferable response both for clinical and pathological outcomes. Limitation of this finding was all patients enrolled to this study were above score 4 in the ER evaluation and 65.7% were at score 8. Progesterone receptor expression before treatment predicted neither clinical response nor pathological response. Although further studies are required with a central laboratory determination, this result indicates that tumors with lower or absent expression of PgR should not be excluded from neoadjuvant AI treatment.

The percentage of MIB1/Ki67-positive cells (Ki67 index) is considered to be a prognostic factor for breast cancer patients. In the P024 study, although no correlation was observed between the Ki67 index before neoadjuvant hormone therapy and the recurrence rate, the Ki67 index after neoadjuvant hormone therapy (at surgery) was correlated with the recurrence rate.(16) The Ki67 index and progression free survival (PFS) has also been studied in patients receiving neoadjuvant hormone therapy with anastrozole, with reported similar findings.(17) Miller et al.(21) evaluated 63 postmenopausal breast cancer patients after 3 months of neoadjuvant hormone therapy with letrozole and reported that the clinical response rate was 85% (41 of 48) in patients with a ≥40% decrease in the Ki67 index, and that a ≥40% decrease in Ki67 index was observed in 11 (70%) of 15 patients with SD.

In the present study, the Ki67 index decreased significantly between baseline and endpoint but the change ratio showed no significant correlation with clinical objective response. As in previous reports, the pretreatment Ki67 index had no predictive value for the probability of clinical objective response and pathological response, indicating that patients who have higher Ki67 tumors still may achieve clinical and/or pathological responses with neoadjuvant exemestane treatment. On the other hand, there were no clinical PD patients during the 24-week treatment period (patients withdraw due to PD and PD, Fig. 4a), if a primary tumor expressed a Ki67 index of 15% or less. In addition, an increase of Ki67 index after treatment, even among tumors with of Ki67 index less than 15%, meant no chance of pathological response. The prognostic impact of clinical non-responders and of pathological non-responders is not fully understood yet, and these are crucial issues to be investigated with a long-term follow-up.(24,25) In this study, a PEPI score(16) was used as a prognostic indicator, and correlation to clinical and pathological responses was investigated. The results implied that pathological non-response might correlate to the poor prognosis, which was shown as a PEPI score of 4 or higher. In future studies, the relationship between pathological response and biomarker changes will be assessed on survival outcomes.

From our investigation of the Ki67 index, monitoring of changes in Ki67 index during treatment, as well as primary treatment stratification with the initial Ki67 index, is essentially important to identify a more appropriate combination of neoadjuvant hormone therapy and chemotherapy, and also to realize the selection of an optimal regimen for each individual patient. For the future, it is warranted to analyze hormone receptor associations and crosstalks with other growth axes such as the HER family in further depth.

In conclusion, this study revealed that 24 weeks of neoadjuvant treatment with exemestane is safe and effective in patients with postmenopausal, hormone receptor-positive breast cancer, achieving ORR values of 50.9%. Furthermore, the number of patients considered candidates for BCS was increased by neoadjuvant treatment. The Ki67 index decreased significantly in patients with PR and SD after treatment. The treatment of hormone responsive breast cancer patients will be personalized with clinical and/or pathological response and biomarker determinations for greater efficacy.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Trial investigators (excluding authors): Iwate Medical University, G. Wakabayashi, Iwate; Tsukuba University, H. Bando, Ibaragi; St. Luke’s International Hospital, S. Nakamura, Tokyo; Kumamoto City Hospital, R. Nishimura, Kumamoto; Nihon University Itabashi Hospital, S. Amano, Tokyo; Sapporo Medical University, T. Ohmura, Hokkaido; Gunma Prefectural Cancer Center, Y. Yanagida, Gunma; Saitama Medical University International Medical Center, T. Saeki, Saitama; Juntendo University Hospital, K. Kojima, Tokyo; Showa University Hospital, T. Sawada, Tokyo; Toho University Omori Medical Center, H. Ogata, Tokyo; International Medical Center of Japan, H. Yasuda, Tokyo; The Cancer Institute Hospital of JFCR, S. Takahashi, Tokyo; Tokyo Metropolitan Fuchu Hospital, M. Takami, Tokyo; Mitsui Memorial Hospital, T. Nishi, Tokyo; Kanagawa Cancer Center, A. Chiba, Kanagawa; Tokai University, Y Tokuda, Kanagawa; Shinshu University Hospital K. Ito, Nagano; Fujita Health University, T. Utsumi, Aichi; Kitakyushu Municipal Medical Center, Keisei Anan, Fukuoka, Japan. This work was supported by The Japanese Foundation for Multidisciplinary Treatment of Cancer. We thank Dr Furuta and Ms Nakajima for their logistical support. Editorial support was provided by Dr Nicholas Smith of Edanz Writing. This study was funded by a research grant from Japan’s Ministry of Health, Labour and Welfare for a study on constructing an algorithm for multimodality therapy with biomarkers for primary breast cancer during formulation of the decision-making process, led by Masakazu Toi (H18-3JIGAN-IPPAN-007, H19-3JIGAN-IPPAN-007).

Disclosure Statement

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Masakazu Toi received lecture fees from Pfizer Japan Inc. Norikazu Masuda received lecture fees from Chugai Pharmaceutical Co. Ltd. Shigehira Saji received lecture fees from Novartis Pharma K.K. and Chugai Pharmaceutical Co. Ltd.

References

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information
  • 1
    Fisher B, Bryant J, Wolmark N et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16: 267285.
  • 2
    Wolmark N, Wang J, Mamounas E et al. Preoperative chemotherapy in patients with operable breast cancer: nine-year results from National Surgical Adjuvant Breast and Bowel Project B-18. J Natl Cancer Inst Monogr 2001; (30): 96102.
  • 3
    van der Hage JA, van de Velde CJ. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001; 19: 422437.
  • 4
    Bear HD, Anderson S, Brown A et al. The effect on tumor response of adding sequential preoperative docetaxel to preoperative doxorubicin and cyclophosphamide: preliminary results from National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 2003; 21: 416574.
  • 5
    Bear HD, Anderson S, Smith RE et al. Sequential preoperative or postoperative docetaxel added to preoperative doxorubicin plus cyclophosphamide for operable breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 2006; 24: 201927.
  • 6
    Guarneri V, Broglio K, Kau SW et al. Prognostic value of pathologic complete response after primary chemotherapy in relation to hormone receptor status and other factors. J Clin Oncol 2006; 24: 103744.
  • 7
    Aapro M, Monfardini S, Jirillo A, Basso U. Management of primary and advanced breast cancer in older unfit patients (medical treatment). Cancer Treat Rev 2009; 35: 5038.
  • 8
    Mlineritsch B, Tausch C, Singer C et al. Exemestane as primary systemic treatment for hormone receptor positive post-menopausal breast cancer patients: a phase II trial of the Austrian Breast and Colorectal Cancer Study Group (ABCSG-17). Breast Cancer Res Treat 2008; 112: 20313.
  • 9
    Eiermann W, Paepke S, Appfelstaedt J et al. Preoperative treatment of postmenopausal breast cancer patients with letrozole: a randomized double-blind multicenter study. Ann Oncol 2001; 12: 152732.
  • 10
    Cataliotti L, Buzdar AU, Noguchi S et al. Comparison of anastrozole versus tamoxifen as preoperative therapy in postmenopausal women with hormone receptor-positive breast cancer: the Pre-Operative “Arimidex” Compared to Tamoxifen (PROACT) trial. Cancer 2006; 106: 2095103.
  • 11
    Takei H, Suemasu K, Inoue K et al. Multicenter phase II trial of neoadjuvant exemestane for postmenopausal patients with hormone receptor-positive, operable breast cancer: Saitama Breast Cancer Clinical Study Group (SBCCSG-03). Breast Cancer Res Treat 2008; 107: 8794.
  • 12
    Semiglazov VF, Semiglazov VV, Dashyan GA et al. Phase 2 randomized trial of primary endocrine therapy versus chemotherapy in postmenopausal patients with estrogen receptor-positive breast cancer. Cancer 2007; 110: 24454.
  • 13
    NCCN Clinical Practice Guidelines in Oncology. Breast Cancer. [Cited September 2010.] Available from URL: http://www.nccn.org/index.asp.
  • 14
    Barnadas A, Gil M, González S, et al. Exemestane as primary treatment of oestrogen receptor-positive breast cancer in postmenopausal women: a phase II trial. Br J Cancer 2009; 100: 4429.
  • 15
    Ring AE, Smith IE, Ashley S et al. Oestrogen receptor status, pathological complete response and prognosis in patients receiving neoadjuvant chemotherapy for early breast cancer. Br J Cancer 2004; 91: 20127.
  • 16
    Ellis MJ, Tao Y, Luo J et al. Outcome prediction for estrogen receptor-positive breast cancer based on postneoadjuvant endocrine therapy tumor characteristics. J Natl Cancer Inst 2008; 100: 13808.
  • 17
    Dowsett M, Smith IE, Ebbs S et al. Prognostic value of Ki67 expression after short-term presurgical endocrine therapy for primary breast cancer. J Natl Cancer Inst 2007; 99: 16770.
  • 18
    Harvey JM, Clark GM, Osborne CK, Allred DC. 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 1999; 17: 147481.
  • 19
    Wolff AC, Hammond ME, Schwartz JN et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 2007; 25: 11845.
  • 20
    Bouzubar N, Walker KJ, Griffiths K et al. Ki67 immunostaining in primary breast cancer: pathological and clinical associations. Br J Cancer 1989; 59: 9437.
  • 21
    Miller WR, White S, Dixon JM et al. Proliferation, steroid receptors and clinical/pathological response in breast cancer treated with letrozole. Br J Cancer 2006; 94: 10516.
  • 22
    Olson JA Jr, Budd GT, Carey LA et al. Improved surgical outcomes for breast cancer patients receiving neoadjuvant aromatase inhibitor therapy: results from a multicenter phase II trial. J Am Coll Surg 2009; 208: 90614.
  • 23
    Dixon JM, Renshaw L, Macaskill EJ et al. Increase in response rate by prolonged treatment with neoadjuvant letrozole. Breast Cancer Res Treat 2009; 113: 14551.
  • 24
    Burstein HJ, Prestrud AA, Seidenfeld J et al. American Society of Clinical Oncology clinical practice guideline: update on adjuvant endocrine therapy for women with hormone receptor-positive breast cancer. J Clin Oncol 2010; 28: 378496.
  • 25
    Miller WR. Clinical, pathological, proliferative and molecular responses associated with neoadjuvant aromatase inhibitor treatment in breast cancer. J Steroid Biochem Mol Biol 2010; 118: 2736.

Supporting Information

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Data S1. Adverse events (number of patients).

FilenameFormatSizeDescription
CAS_1867_sm_Supplementarydata_e.doc107KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.