Abstract
Endocrine therapy is the most important treatment of choice for estrogen receptor (ER)‐positive breast cancer. Potential mechanisms for resistance to endocrine therapy involve ER‐coregulatory proteins and cross‐talk between ER and other growth factor–signaling networks. However, the factors and pathways responsible for endocrine therapy resistance, particularly resistance to aromatase inhibitors, have not been clearly established. Sixteen postmenopausal patients with ERα‐positive primary breast cancer were treated daily with 25 mg of exemestane (an aromatase inhibitor) for 6 months. Expressions of ERα, ERβ, progesterone receptor (PgR), androgen receptor (AR), amplified in breast cancer 1 (AIB1), aromatase, epidermal growth factor receptor, human epidermal growth factor receptor type 2, Ki67, cyclin D1, p53, Bcl2, signal transducer and activator of transcription 5 (Stat5), and insulin‐like growth factor binding protein 5 (IGFBP5), and phosphorylations of ERα serine (Ser) 118, ERα Ser167, Akt Ser473, and p44/42 MAPK threonine (Thr) 202/tyrosine (Tyr) 204, were examined by immunohistochemistry on pretreatment tumor biopsies and post‐treatment surgical specimens. Analyses were made to test for correlations with response to exemestane. Of the 16 patients, seven responded and nine retained stable disease. High‐level expression of AIB1 and phosphorylation of Akt Ser473 were significantly associated with a better response to exemestane, suggesting that these factors could be considered as predictors of exemestane response. Expressions of ERα, ERβ, PgR, aromatase, Ki67, cyclin D1, and p53, and phosphorylations of ERα Ser118, ERα Ser167, and p44/42 MAPK Thr202/Tyr204, were decreased, whereas expressions of Stat5 and IGFBP5 were increased in post‐treatment specimens compared to the values in pretreatment biopsies. Thus, the analysis of factors involved in the estrogen‐dependent growth‐signaling pathways may be useful in identifying patients responsive to exemestane. (Cancer Sci 2009)
Endocrine therapy has become the most important treatment option for women with ER‐positive breast cancer. Local aromatization of androgens to estrogens is the primary source of estradiol in the breasts of postmenopausal women. Large‐scale adjuvant clinical trials of nonsteroidal aromatase inhibitors, such as anastrozole and letrozole, and the steroidal aromatase inhibitor exemestane, have shown improved disease‐free survival for patients with early stage breast cancer randomized to 5 years of aromatase inhibitor treatment as compared to 5 years of tamoxifen treatment.( 1 , 2 ) When exemestane was administered following 2–3 years of tamoxifen, this improvement in disease‐free survival was shown to translate into an overall survival benefit above that obtained with 5 years of tamoxifen treatment.( 3 ) Therefore, aromatase inhibitors are now considered to be the gold standard endocrine therapy for hormone‐receptor positive breast cancer in postmenopausal women. However, many breast cancer patients with tumors expressing high levels of ER are unresponsive to aromatase inhibitors, and all patients with advanced disease eventually develop resistance to the therapy. Identification of response predictors for aromatase inhibitors is critical to help reveal such patients prior to treatment.
Recent neoadjuvant aromatase inhibitor studies demonstrated that the expression of the proliferation marker Ki67 after short‐term presurgical endocrine therapy showed a significant correlation with both response to the therapy and recurrence‐free survival.( 4 , 5 )
We previously investigated prognostic and predictive factors for endocrine therapy in ER‐positive breast cancer, and found various molecular predictors, such as phosphorylation of ERα Ser118 and ERα Ser167,( 6 , 7 ) and expression of aromatase,( 8 ) AIB1,( 7 ) p53,( 9 , 10 ) Stat5,( 11 ) and IGFBP5,( 12 ) as well as expression of HER2 and Ki67.( 7 , 9 , 10 ) However, the endocrine therapies in those studies were mostly tamoxifen. In this study, we examined expressions and phosphorylations of molecular markers, including those factors examined by immunohistochemistry on pretreatment biopsies and post‐treatment surgical specimens in postmenopausal patients with ER‐positive primary breast cancer who were treated daily with 25 mg of exemestane for 6 months. Correlations between response to exemestane and expressions and phosphorylations of molecular markers were analyzed to identify predictors for the treatment.
Materials and Methods
Patients and breast cancer tissues. Sixteen postmenopausal patients over age 70 with primary ERα‐positive breast cancer were treated daily with 25 mg of exemestane for 6 months between 2003 and 2006 (Table 1). Clinical measurement of tumor size and nodal status was performed monthly, and the final clinical sonographic measurements were performed 6 months after the start of treatment prior to the planned surgical excision of the tumor. Clinical response was defined as complete response (CR), partial response (PR), or stable disease (SD) according to the Response Evaluation Criteria in Solid Tumors (RECIST, 2000). Pretreatment specimens were taken by core needle biopsies. Post‐treatment specimens were obtained at surgical treatment. The pathological response was assessed as grades 1 to 3 according to the following criteria: 0 (no response), 1 (mild to moderate response), 2 (marked response), 3 (complete response). The study protocol was approved by the institutional review boards and conformed with the guidelines of the 1996 Declaration of Helsinki.
Table 1.
Clinicopathological characteristics of patients and primary breast tumors
| Number of patients | ||
|---|---|---|
| Total number of patients | 16 | |
| Age at diagnosis (years) | Mean ± SD | 79.1 ± 6.3 |
| Range | 70 to 88 | |
| Tumor size (cm) | <3.0 | 12 |
| ≥3.0 | 4 | |
| Lymph node status (N) | Negative | 16 |
| Positive | 0 | |
| Nuclear grade | 1 | 12 |
| 2 | 3 | |
| 3 | 1 | |
| HER2 | 0 | 5 |
| 1+ | 10 | |
| 2+ | 1 | |
| 3+ | 0 | |
| EGFR | 0 | 16 |
| Type of surgery | Breast‐conserving surgery | 15 |
| Mastectomy | 1 | |
| Adjuvant therapy | Exemestane | 16 |
| Follow up (months) | Mean ± SD | 46.0 ± 17.0 |
| Range | 18 to 67 |
EGFR, epidermal growth factor receptor; HER2, human epidermal growth factor receptor type 2.
Immunohistochemical (IHC) analysis. One 4‐μm section of each submitted paraffin block was stained first with hematoxylin–eosin to verify that an adequate number of invasive carcinoma cells were present and that the fixation quality was adequate for IHC analysis. Serial sections (4 μm) were prepared from selected blocks and float‐mounted on adhesive‐coated glass slides for staining of expression of ERα,( 13 ) ERβ1,( 14 ) ERβcx/β2,( 14 ) PgR,( 13 ) AR, AIB1,( 7 ) aromatase,( 15 ) EGFR, HER2,( 9 ) Ki67,( 9 ) cyclin D1,( 16 ) p53,( 9 ) Bcl2,( 16 ) Stat5,( 11 ) and IGFBP5,( 12 ) and phosphorylation of ERα Ser118,( 7 ) ERα Ser167,( 7 ) Akt Ser473,( 7 ) and p44/42 MAPK Thr202/Tyr204,( 7 ) as described previously. Primary antibodies used in this study are listed in Table 2.
Table 2.
List of antibodies used for the immunohistochemical markers
| Markers | Antibodies | References |
|---|---|---|
| ERα | 1D5; Dako, Carpinteria, CA, USA | 13 |
| pERα Ser118 | No. 2515; Cell Signaling, Baverly, MA, USA | 7 |
| pERα Ser167 | No. 2514; Cell Signaling | 7 |
| ERβ1 | Generated rabbit polyclonal antibodies | 14 |
| ERβcx/β2 | Generated rabbit polyclonal antibodies | 14 |
| PgR | 636; Dako | 13 |
| AR | 441, sc‐7305; Santa Cruz Biotechnology, Santa Cruz, CA, USA | ND |
| AIB1 | Clone 34; BD Biosciences, San Jose, CA, USA | 7 |
| Aromatase | Kindly provided by Nobuhiro Harada | 15 |
| HER2 | c‐erbB‐2; Dako | 9 |
| EGFR | EGFR; Dako | ND |
| pAkt | No. 9277; Cell Signaling | 7 |
| pMAPK | No. 9101; Cell Signaling | 7 |
| Ki67 | MIB‐1; Dako | 9 |
| Cyclin D1 | bcl‐1 Ab‐4; Neo Markers, Fremont, CA, USA | 16 |
| p53 | PAb1801; Novocastra, Newcastle, UK | 9 |
| Bcl2 | Clone 124; Dako | 16 |
| Stat5 | Stat5b (G‐2), sc‐1656; Santa Cruz Biotechnology | 11 |
| IGFBP5 | H‐100, sc‐13093; Santa Cruz Biotechnology | 12 |
AIB1, amplified in breast cancer 1; AR, androgen receptor; EGFR, epidermal growth factor receptor; ER, estrogen receptor; HER2, human epidermal growth factor receptor type 2; IGFBP5, insulin‐like growth factor binding protein 5; pAkt, phosphorylated Akt; pER, phosphorylated ER; PgR, progesterone receptor; pMAPK, phosphorylated MAPK, Ser, serine; Stat5, signal transducer and activator of transcription 5.
Immunostained slides were scored after the entire slide was evaluated by light microscopy. The expression of ERα, ERβ1, ERβcx/β2, PgR, and AR, and the phosphorylation of ERα Ser118 and ERα Ser167, were scored by assigning proportion and intensity scores, according to Allred’s procedure.( 17 ) In brief, a proportion score represented the estimated proportion of tumor cells staining positive as follows: 0 (none), 1 (<1/100), 2 (1/100 to 1/10), 3 (1/10 to 1/3), 4 (1/3 to 2/3), and 5 (>2/3). Any brown nuclear staining in breast epithelium counted towards the proportion score. An intensity score represented the average intensity of the positive cells as follows: 0 (none), 1 (weak), 2 (intermediate), and 3 (strong). The proportion and intensity scores were then added to obtain a total score which could range from 0 to 8. Tumors with scores ≥3 for ERα were included in this study. Expression of AIB1, aromatase, Ki67, cyclin D1, p53, Bcl2, Stat5, and IGFBP5, and phosphorylation of Akt and p44/42 MAPK, were scored by assigning proportion scores as follows: 0 (none), 1 (<1/100), 2 (1/100 to 1/10), 3 (1/10 to 1/3), 4 (1/3 to 2/3), and 5 (>2/3). Expression of Stat5 and IGFBP5 was assessed by cytoplasmic and nuclear staining, respectively. Expression of AIB1, Ki67, cyclin D1, p53, and Bcl2, and phosphorylation of Akt and p44/42 MAPK, were assessed by nuclear staining only. Expression of aromatase was assessed by cytoplasmic staining in cancer cells and stromal cells, respectively. Immunostaining of EGFR and HER2 was evaluated using the same method as is employed by the HercepTest (Dako, Glostrup, Denmark). To determine the score of EGFR and HER2 expression, the membrane staining pattern was estimated and scored on a scale of 0 to 3.
Statistical analysis. The Mann–Whitney U‐test and the unpaired t‐test were used to compare the IHC scores of molecular markers with response to exemestane. To examine the change of expression and phosphorylation status between pretreatment and post‐treatment tumors, the one‐sample Wilcoxon signed rank test was used.
Results
Clinical and pathological responses. The patients’ characteristics are summarized in Table 1. All patients completed the 6‐month treatment. Of the 16 patients, seven responded and nine retained SD according to the clinical evaluation (Table 3). A pathological response (grades I to III) was obtained in eight patients. One patient presented both a clinical CR and a pathological CR. No patient showed progressive disease. A significant correlation was seen between clinical and pathological responses (P = 0.035) (Table 3).
Table 3.
Correlation between clinical and pathological responses
| Clinical response | Pathological response (grade) | ||||
|---|---|---|---|---|---|
| 0 | 1 | 2 | 3 | Total | |
| SD | 6 (2)† | 3 | 0 | 0 | 9 |
| PR | 2 (0)† | 2 | 2 | 0 | 6 |
| CR | 0 | 0 | 0 | 1 | 1 |
| total | 8 | 5 | 2 | 1 | 16 |
†Number of cases that no pathological response was observed. CR, complete response; PR, partial response; SD, stable disease.
High expression of AIB1 and high‐level phosphorylation of Akt Ser473 are significantly associated with better responses to exemestane. We first analyzed the potential correlation between IHC scores of molecular markers in tumors before treatment and response to exemestane. ERα Allred expression scores were 5 or greater in all tumors before treatment. All tumors were negative for EGFR. As shown in Table 4, patients with primary tumors that had higher expressions of AIB1 and high‐level phosphorylations of Akt Ser473 responded significantly to the treatment according to the clinical evaluation (Mann–Whitney U‐test, P = 0.037 and P = 0.0006, respectively). Although IHC scores of HER2 were higher in responders compared to nonresponders (scores 1 or 2 vs 0 or 1), none of them were overexpressed (Fig. 1). Similarly, the expression levels of p53 were below 10% in all tumors, although IHC scores of p53 were higher in responders than in nonresponders (scores 0 to 2 vs 0 or 1) (Fig. 2). The other molecular markers, including phosphorylation of ERα and expression of PgR (Fig. 3a), did not affect the clinical response to exemestane. When the response was evaluated pathologically, phosphorylation of Akt Ser473 was the only factor that affected the response (Mann–Whitney U‐test, P = 0.020) (Table 5). We concluded that higher expression of AIB1 and high‐level phosphorylation of Akt Ser473 are predictors for an exemestane response.
Table 4.
Correlation between immunohistochemical scores of biological markers in tumors before treatment and response to exemestane
| PR and CR (n = 7) Mean ± SD (median; range) | SD (n = 9) Mean ± SD (median; range) | P‐values† | P‐values‡ | |
|---|---|---|---|---|
| ERα | 7.6 ± 1.1 | 7.7 ± 0.7 | 0.82 | 0.84 |
| pERα Ser118 | 7.1 ± 1.1 | 7.4 ± 0.8 | 0.58 | 0.55 |
| pERα Ser167 | 7.7 ± 0.8 | 7.2 ± 0.8 | 0.16 | 0.24 |
| ERβ1 | 6.9 ± 1.5 | 6.6 ± 0.7 | 0.28 | 0.60 |
| ERβcx/β2 | 4.9 ± 2.4 | 3.7 ± 2.8 | 0.41 | 0.38 |
| PgR | 4.9 ± 3.6 | 5.1 ± 2.6 | 0.79 | 0.87 |
| AR | 3.0 ± 1.3 | 2.7 ± 0.7 | 0.77 | 0.52 |
| AIB1 | 3.0 ± 2.1 | 0.6 ± 1.7 | 0.037* | 0.020* |
| Aromatase (cancer) | 4.6 ± 0.8 | 4.8 ± 0.7 | 0.44 | 0.58 |
| Aromatase (stroma) | 3.6 ± 1.0 | 4.0 ± 1.0 | 0.44 | 0.40 |
| HER2 | 1.1 ± 0.4 | 0.4 ± 0.5 | 0.016* | 0.010* |
| pAkt | 4.0 ± 1.5 | 0.3 ± 1.0 | 0.0006* | <0.0001* |
| pMAPK | 3.7 ± 1.8 | 2.3 ± 1.9 | 0.12 | 0.17 |
| Ki67 | 2.4 ± 0.5 | 2.6 ± 0.7 | 0.81 | 0.70 |
| Cyclin D1 | 4.9 ± 0.4 | 4.8 ± 0.4 | 0.70 | 0.71 |
| p53 | 1.0 ± 1.0 | 0.1 ± 0.3 | 0.038* | 0.025* |
| Bcl2 | 3.1 ± 0.8 | 3.0 ± 1.1 | 0.96 | 0.78 |
| Stat5 (cytoplasm) | 2.1 ± 2.7 | 2.8 ± 2.6 | 0.63 | 0.64 |
| Stat5 (nuclei) | 4.9 ± 0.4 | 3.7 ± 2.1 | 0.17 | 0.17 |
| IGFBP5 (cytoplasm) | 3.1 ± 2.3 | 2.0 ± 2.5 | 0.39 | 0.36 |
| IGFBP5 (nuclei) | 4.6 ± 0.8 | 3.2 ± 2.1 | 0.18 | 0.13 |
*P < 0.05 is considered significant. †Mann–Whitney U‐test. ‡Unpaired t‐test. AIB, amplified in breast cancer 1; AR, androgen receptor; ER, estrogen receptor; HER2, human epidermal growth factor receptor type 2; IGFBP5, insulin‐like growth factor binding protein 5; pAkt, phosphorylated Akt; pER, phosphorylated ER; PgR, progesterone receptor; pMAPK, phosphorylated MAPK, Ser, serine; Stat5, signal transducer and activator of transcription 5.
Figure 1.
Changes of immunohistochemical (IHC) expression scores for human epidermal growth factor receptor type 2 (HER2) in tumors in response to exemestane treatment. (a) All patients. (b) Responders (PR) and nonresponders (SD) were separately analyzed.
Figure 2.
Immunohistochemical (IHC) expression scores for p53 in tumors before and after exemestane treatment. (a) All patients. (b) Responders (PR) and nonresponders (SD) were separately analyzed.
Figure 3.
(a) Progesterone receptor (PgR) Allred scores in tumors before exemestane treatment for responders and nonresponders. (b) Changes of immunohistochemical expression scores for PgR in tumors in response to exemestane treatment. Responders (PR) and nonresponders (SD) were separately analyzed.
Table 5.
Correlation between immunohistochemical scores of biological markers in tumors before treatment and pathological response
| Grade 1–3 (n = 8) Mean ± SD (median; range) | Grade 0 (n = 8) Mean ± SD (median; range) | P‐values† | P‐values‡ | |
|---|---|---|---|---|
| ERα | 7.5 ± 1.1 | 7.8 ± 0.5 | 0.54 | 0.59 |
| pERα Ser118 | 7.3 ± 1.1 | 7.4 ± 0.8 | 0.85 | 0.80 |
| pERα Ser167 | 7.5 ± 0.6 | 7.4 ± 0.9 | 0.86 | 0.77 |
| ERβ1 | 7.0 ± 0.6 | 6.4 ± 1.7 | 0.37 | 0.26 |
| ERβcx/β2 | 3.4 ± 2.9 | 5.0 ± 2.1 | 0.21 | 0.23 |
| PgR | 5.5 ± 3.5 | 4.5 ± 2.5 | 0.29 | 0.52 |
| AR | 3.0 ± 0.9 | 2.6 ± 1.1 | 0.33 | 0.46 |
| AIB1 | 2.1 ± 2.3 | 1.1 ± 2.1 | 0.39 | 0.38 |
| Aromatase (cancer) | 4.5 ± 0.9 | 4.9 ± 0.4 | 0.44 | 0.30 |
| Aromatase (stroma) | 4.3 ± 0.9 | 3.4 ± 0.9 | 0.077 | 0.073 |
| HER2 | 0.9 ± 0.6 | 0.6 ± 0.5 | 0.42 | 0.41 |
| pAkt | 3.3 ± 2.2 | 0.5 ± 1.4 | 0.020* | 0.013* |
| pMAPK | 2.5 ± 2.3 | 3.4 ± 1.6 | 0.55 | 0.39 |
| Ki67 | 2.8 ± 0.7 | 2.3 ± 0.5 | 0.12 | 0.12 |
| Cyclin D1 | 4.8 ± 0.5 | 4.9 ± 0.4 | 0.54 | 0.55 |
| p53 | 0.9 ± 1.0 | 0.1 ± 0.4 | 0.083 | 0.063 |
| Bcl2 | 3.4 ± 0.9 | 2.8 ± 1.0 | 0.27 | 0.22 |
| Stat5 (cytoplasm) | 3.1 ± 2.6 | 1.9 ± 2.6 | 0.33 | 0.35 |
| Stat5 (nuclei) | 3.6 ± 2.3 | 4.8 ± 0.5 | 0.44 | 0.19 |
| IGFBP5 (cytoplasm) | 3.0 ± 2.5 | 2.0 ± 2.3 | 0.37 | 0.42 |
| IGFBP5 (nuclei) | 3.3 ± 2.3 | 4.4 ± 0.9 | 0.38 | 0.21 |
*P < 0.05 is considered significant. †Mann–Whitney U‐test. ‡Unpaired t‐test. AIB1, amplified in breast cancer 1; AR, androgen receptor; ER, estrogen receptor; HER2, human epidermal growth factor receptor type 2; IGFBP5, insulin‐like growth factor binding protein 5; pAkt, phosphorylated Akt; pER, phosphorylated ER; PgR, progesterone receptor; pMAPK, phosphorylated MAPK; Ser, serine; Stat5, signal transducer and activator of transcription 5.
Comparison of expression and phosphorylation levels of molecular markers in tumors before and after exemestane treatment. We next examined potential correlations between IHC scores of molecular markers in tumors after treatment and response to exemestane. Only 15 paired tumors were analyzed because one patient obtained a pathological CR. Patients who responded to the therapy had tumors that showed high expressions of HER2 or IGFBP5 in both cytoplasm and nuclei (Mann–Whitney U‐test, P = 0.031, P = 0.030 and P = 0.020, respectively) (Table 6). When IHC scores of molecular markers were compared in tumors before and after exemestane treatment, expression levels of ERα, ERβ1, ERβcx/β2, PgR (Fig. 3b), aromatase in stromal cells, Ki67 (Fig. 4a), and cyclin D1, and phosphorylation levels of ERα Ser118, ERα Ser167, and p44/42 MAPK Thr202/Tyr204, were decreased, whereas expression levels of Stat5 and IGFBP5 in cytoplasm were increased in post‐treatment tumors compared to the levels in pretreatment specimens regardless of the treatment response (Table 7). Although expression levels of p53 (Fig. 2a) also were decreased in post‐treatment tumors compared to pretreatment specimens, these changes were not significant. On the other hand, expression levels of AR, AIB1, HER2 (Fig.1a), and Bcl2 did not change significantly during the treatment. Expression levels of Ki67 decreased more in post‐treatment specimens of responders (grades 1 and 2) than in tumors of nonresponders (grade 0) according to the pathological evaluation (P = 0.035) (Fig. 4c). Interestingly, the phosphorylation levels of Akt Ser473 were decreased in the post‐treatment specimens of responders (PR) but not of nonresponders (SD) according to the clinical evaluation (P = 0.030) (Fig. 5b). We suggest that these factors, especially those in the Akt signaling pathway, are involved in the estrogen‐dependent growth.
Table 6.
Correlation between immunohistochemical scores of biological markers in tumors after treatment and response to exemestane
| PR (n = 6) Mean ± SD (median; range) | SD (n = 9) Mean ± SD (median; range) | P‐values† | P‐values‡ | |
|---|---|---|---|---|
| ERα | 6.8 ± 1.9 | 6.9 ± 1.1 | 0.58 | 0.94 |
| pERα Ser118 | 4.3 ± 1.8 | 4.8 ± 1.7 | 0.55 | 0.63 |
| pERα Ser167 | 5.5 ± 1.8 | 5.6 ± 1.4 | 0.95 | 0.95 |
| ERβ1 | 3.0 ± 3.0 | 3.9 ± 2.7 | 0.59 | 0.56 |
| ERβcx/β2 | 1.7 ± 2.0 | 1.4 ± 2.2 | 0.79 | 0.84 |
| PgR | 2.5 ± 2.3 | 2.3 ± 2.0 | 0.90 | 0.88 |
| AR | 2.5 ± 0.5 | 2.9 ± 1.1 | 0.56 | 0.42 |
| AIB1 | 2.3 ± 2.6 | 0.4 ± 1.3 | 0.08 | 0.08 |
| Aromatase (cancer) | 4.8 ± 0.4 | 4.4 ± 1.1 | 0.67 | 0.44 |
| Aromatase (stroma) | 2.3 ± 0.8 | 3.1 ± 1.5 | 0.27 | 0.26 |
| HER2 | 1.0 ± 0.0 | 0.4 ± 0.5 | 0.031* | 0.024* |
| pAkt | 1.7 ± 2.6 | 1.0 ± 1.3 | 0.89 | 0.52 |
| pMAPK | 1.3 ± 1.4 | 1.4 ± 1.5 | 0.80 | 0.89 |
| Ki67 | 1.5 ± 0.5 | 1.7 ± 0.7 | 0.69 | 0.63 |
| Cyclin D1 | 3.8 ± 1.8 | 4.0 ± 1.4 | 0.95 | 0.85 |
| p53 | 0.2 ± 0.4 | 0.0 ± 0.0 | 0.22 | 0.23 |
| Bcl2 | 3.3 ± 1.6 | 1.9 ± 1.3 | 0.10 | 0.08 |
| Stat5 (cytoplasm) | 5.0 ± 0.0 | 3.2 ± 2.4 | 0.07 | 0.10 |
| Stat5 (nuclei) | 5.0 ± 0.0 | 4.6 ± 0.7 | 0.13 | 0.16 |
| IGFBP5 (cytoplasm) | 5.0 ± 0.0 | 3.3 ± 2.1 | 0.030* | 0.7 |
| IGFBP5 (nuclei) | 5.0 ± 0.0 | 3.7 ± 1.6 | 0.020* | 0.06 |
*P < 0.05 is considered significant. †Mann–Whitney U‐test. ‡Unpaired t‐test. AIB1, amplified in breast cancer 1; AR, androgen receptor; ER, estrogen receptor; HER2, human epidermal growth factor receptor type 2; IGFBP5, insulin‐like growth factor binding protein 5; pAkt, phosphorylated Akt; pER, phosphorylated ER; PgR, progesterone receptor; pMAPK, phosphorylated MAPK; Ser, serine; Stat5, signal transducer and activator of transcription 5.
Figure 4.
Immunohistochemical (IHC) expression scores for Ki67 in tumors before and after exemestane treatment. (a) All patients. (b) Responders (PR) and nonresponders (SD) with clinical evaluation were separately analyzed. (c) Responders (grades 1 and 2) and nonresponders (grade 0) with pathological evaluation were separately analyzed.
Table 7.
Comparison of immunohistochemical scores of biological markers in tumors before and after exemestane treatment
| Before Mean ± SD | After Mean ± SD | P‐values† | P‐values‡ | |
|---|---|---|---|---|
| ERα | 7.6 ± 0.9 | 6.9 ± 1.4 | 0.016* | 0.010* |
| pERα Ser118 | 7.3 ± 0.9 | 4.6 ± 1.7 | 0.0014* | <0.0001* |
| pERα Ser167 | 7.4 ± 0.8 | 5.5 ± 1.5 | 0.0039* | 0.0005* |
| ERβ1 | 6.7 ± 1.1 | 3.5 ± 2.7 | 0.0021* | 0.0002* |
| ERβcx/β2 | 4.2 ± 2.6 | 1.5 ± 2.0 | 0.0071* | 0.0012* |
| PgR | 5.0 ± 3.0 | 2.4 ± 2.1 | 0.0089* | 0.0025* |
| AR | 2.8 ± 1.0 | 2.7 ± 0.9 | >0.99 | |
| AIB1 | 1.6 ± 2.2 | 1.2 ± 2.0 | 0.79 | 0.71 |
| Aromatase (cancer) | 4.7 ± 0.7 | 4.6 ± 0.9 | 0.83 | 0.84 |
| Aromatase (stroma) | 3.8 ± 1.0 | 2.8 ± 1.3 | 0.0065* | 0.0021* |
| HER2 | 0.8 ± 0.6 | 0.7 ± 0.5 | 0.059 | 0.058 |
| pAkt | 1.9 ± 2.2 | 1.3 ± 1.9 | 0.29 | 0.43 |
| pMAPK | 2.9 ± 1.9 | 1.4 ± 1.4 | 0.020* | 0.018* |
| Ki67 | 2.5 ± 0.6 | 1.6 ± 0.6 | 0.0013* | 0.0005* |
| Cyclin D1 | 4.8 ± 0.4 | 3.9 ± 1.5 | 0.041* | 0.037* |
| p53 | 0.5 ± 0.8 | 0.07 ± 0.3 | 0.059 | 0.055 |
| Bcl2 | 3.1 ± 0.9 | 2.5 ± 1.6 | 0.30 | 0.26 |
| Stat5 (cytoplasm) | 2.5 ± 2.6 | 3.9 ± 2.0 | 0.035* | 0.027* |
| Stat5 (nuclei) | 4.1 ± 1.7 | 4.7 ± 0.6 | 0.38 | 0.23 |
| IGFBP5 (cytoplasm) | 2.5 ± 2.4 | 4.0 ± 1.8 | 0.033* | 0.029* |
| IGFBP5 (nuclei) | 3.8 ± 1.8 | 4.2 ± 1.4 | 0.36 | 0.44 |
*P < 0.05 is considered significant. †One‐sample Wilcoxon signed rank test. ‡Paired t‐test. AIB1, amplified in breast cancer 1; AR, androgen receptor; ER, estrogen receptor; HER2, human epidermal growth factor receptor type 2; IGFBP5, insulin‐like growth factor binding protein 5; pAkt, phosphorylated Akt; pER, phosphorylated ER; PgR, progesterone receptor; pMAPK, phosphorylated MAPK; Ser, serine; Stat5, signal transducer and activator of transcription 5.
Figure 5.
Immunohistochemical (IHC) phosphorylation scores for Akt in tumors before and after exemestane treatment. (a) All patients. (b) Responders (PR) and nonresponders (SD) with clinical evaluation were separately analyzed. (c) Responders (grades 1 and 2) and nonresponders (grade 0) with pathological evaluation were separately analyzed.
Patients’ outcomes. One patient developed bone metastasis during adjuvant exemestane treatment 36 months after surgery. The treatment was switched from exemestane to tamoxifen when the recurrence was diagnosed, and tamoxifen has been effective against her bone metastasis for more than 8 months. Another patient developed local recurrence in her breast during adjuvant exemestane treatment 45 months after surgery. Both patients showed stable disease with the primary exemestane treatment.
Discussion
We investigated predictive factors for exemestane response using pretreatment tumor biopsies and post‐treatment surgical specimens in ER‐positive primary breast cancer. Although expression levels of ERα were 5 or greater by Allred score in all 16 pretreatment tumors, clinical and pathological responses to exemestane varied. Several clinical studies on exemestane as primary endocrine therapy in operable breast cancer have been reported.( 18 , 19 , 20 ) The duration of neoadjuvant exemestane therapy was 4 months in two studies and 6 months in one study. The clinical response rates in these studies were 34–66%, which are comparable to the results in our present study.
Surprisingly, high expression of AIB1 and high‐level phosphorylation of Akt Ser473 were significantly associated with better responses to the therapy. Moreover, phosphorylation of Akt Ser473 is a convincing predictive factor from the aspects of clinical and pathological responsiveness. Several studies have reported that phosphorylation of Akt predicts worse outcome and tamoxifen resistance in ER‐positive breast cancer.( 21 , 22 , 23 , 24 ) Our previous study did not indicate correlation between Akt phosphorylation and prognosis in ER‐positive breast cancer, although phosphorylation of Akt was strongly and positively associated with phosphorylation of ERα Ser118, ERα Ser167, and MAPK.( 7 ) It was reported that estradiol rapidly activates Akt via the HER2 signaling pathway.( 25 ) Akt might be activated via growth factor signaling pathways, both estrogen‐dependently and estrogen‐independently in breast cancer. We previously reported that high expressions of AIB1 and HER2 were associated with significantly worsened disease‐free survival in ER‐positive breast cancer.( 7 ) Moreover, p53 protein accumulation in primary breast tumors predicted resistance to endocrine therapy and decreased post‐relapse survival in metastatic breast cancer.( 10 ) The cut‐off levels in our previous studies were 3 (overexpressed) for HER2 and more than 10% of cells positive for p53. All 16 tumors in this study were negative for HER2 and p53, although expression levels of HER2 and p53 were slightly higher in responders compared to nonresponders.
Prognostic factors in ER‐positive breast cancer, which we have demonstrated previously – such as phosphorylation of ERα Ser118 and ERα Ser167,( 7 ) and expression of Stat5( 11 )– were not predictors for the response to exemestane as primary endocrine therapy for 6 months. We suggest that the discordance between predictors in this study and prognostic factors in our previous studies may be due to different types of endocrine therapy. Most patients were treated with tamoxifen in our previous studies of prognosis prediction, whereas exemestane (an aromatase inhibitor) was used in this study. The mechanisms of endocrine therapy are different for aromatase inhibitors, which block estrogen synthesis in peripheral adipose tissues, and selective ER modulators such as tamoxifen, which bind to ER and block the action of estrogen. The data from recent randomized clinical trials have shown that aromatase inhibitors substantially improve disease‐free survival as compared to tamoxifen in postmenopausal women in the adjuvant setting. Viale and colleagues demonstrated that the magnitude of improved disease‐free survival for letrozole versus tamoxifen was greater for patients with high tumor Ki67 expression than for those with low tumor Ki67 expression. Thus, predictors and/or prognostic factors may be different for different types of endocrine therapy. In addition, predictors for response to primary endocrine therapy evaluated in pretreatment biopsies may not be surrogate markers for prognosis in ER‐positive breast cancer. Ellis and colleagues described that the aromatase inhibitor letrozole, but not tamoxifen, was effective in HER2‐positive tumors as neoadjuvant endocrine therapy in ER‐positive breast cancer.( 26 ) On the other hand, a high HER2 expression is associated with a high risk of recurrence for 5 years after treatment with the aromatase inhibitors anastrozole and letrozole, as well as tamoxifen, in adjuvant endocrine therapy.( 27 , 28 ) Therefore, response predictors for aromatase inhibitors assessed with a pretreatment tumor biopsy may not be prognostic markers for adjuvant endocrine therapy.
Our study indicated that expression levels of ERα, ERβ1, ERβcx/β2, PgR, aromatase in stromal cells, Ki67, and cyclin D1, and phosphorylation levels of ERα Ser118, ERα Ser167, and p44/42 MAPK Thr202/Tyr204, were decreased, whereas expression levels of Stat5 and IGFBP5 in cytoplasm were increased in post‐treatment specimens compared to the levels in pretreatment tumors. It is suggested that inhibition of aromatase activity and estrogen production by exemestane affects molecules that are present downstream of ER signaling pathways regardless of the treatment response. However, expression of AIB1, a predictor for exemestane, was not affected during the treatment. In contrast, phosphorylation levels of Akt Ser473 and expression levels of Ki67 were more reduced in post‐treatment specimens of responders than in those from the tumors of nonresponders. Dowsett and colleagues demonstrated that measurements of tumor Ki67 level after short‐term endocrine treatment may improve the prediction of recurrence‐free survival by integrating the prognostic value of Ki67 level at baseline with changes in Ki67 level that are associated with treatment benefits.( 5 ) It is reasonable to assume that a response to neoadjuvant endocrine therapy affects prognosis when it is known that the same treatment has an effect when it is given as adjuvant therapy. On the other hand, it has been reported that Ki67 is a prognostic factor in postmenopausal women with ER‐positive early breast cancer who were treated with letrozole or tamoxifen as adjuvant therapy, because high Ki67 expression levels in primary breast tumors were associated with worse disease‐free survival.( 29 )
One patient developed bone metastasis during adjuvant exemestane treatment, and tamoxifen has been effective for the recurrent disease. It is not clear whether this phenomenon is intrinsic or whether it represents an acquired resistance to exemestane, although the patient showed SD with the neoadjuvant exemestane therapy. Because tumor biology changes during a long‐term endocrine therapy, predictors for a short‐term endocrine treatment response may not have prognostic value for adjuvant endocrine therapy.
In conclusion, the present data indicate that high expressions of AIB1 and high‐level phosphorylation of Akt Ser473 are predictors for exemestane response, and that phosphorylation of Akt Ser473 is a convincing predictive factor from the aspects of clinical and pathological responsiveness. Our findings will be helpful when preoperative endocrine therapy is planned for women with ER‐positive breast cancer. It might be necessary to consider predictors and prognostic factors for endocrine therapy separately in ER‐positive breast cancer.
Disclosure Statement
Hirotaka Iwase received lecture fees from AstraZeneca, Pfizer, and Chugai‐Roche, and received research funding from AstraZeneca, Takeda, Chugai‐Roche, and Taiho. The other authors have no conflict of interest.
Acknowledgment
We thank Mariko Nishio for her excellent technical assistance.
References
- 1. Forbes JF, Cuzick J, Buzdar A, Howell A, Tobias JS, Baum M. Effect of anastrozole and tamoxifen as adjuvant treatment for early‐stage breast cancer: 100‐month analysis of the ATAC trial. Lancet Oncol 2008; 9: 45–53. [DOI] [PubMed] [Google Scholar]
- 2. Thurlimann B, Keshaviah A, Coates AS et al. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N Engl J Med 2005; 353: 2747–57. [DOI] [PubMed] [Google Scholar]
- 3. Coombes RC, Hall E, Gibson LJ et al. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004; 350: 1081–92. [DOI] [PubMed] [Google Scholar]
- 4. Dowsett M, Ebbs SR, Dixon JM et al. Biomarker changes during neoadjuvant anastrozole, tamoxifen, or the combination: influence of hormonal status and HER‐2 in breast cancer – a study from the IMPACT trialists. J Clin Oncol 2005; 23: 2477–92. [DOI] [PubMed] [Google Scholar]
- 5. Dowsett M, Smith IE, Ebbs SR et al. Prognostic value of Ki67 expression after short‐term presurgical endocrine therapy for primary breast cancer. J Natl Cancer Inst 2007; 99: 167–70. [DOI] [PubMed] [Google Scholar]
- 6. Yamashita H, Nishio M, Kobayashi S et al. Phosphorylation of estrogen receptor alpha serine 167 is predictive of response to endocrine therapy and increases postrelapse survival in metastatic breast cancer. Breast Cancer Res 2005; 7: R753–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Yamashita H, Nishio M, Toyama T et al. Low phosphorylation of estrogen receptor alpha (ERalpha) serine 118 and high phosphorylation of ERalpha serine 167 improve survival in ER‐positive breast cancer. Endocr Relat Cancer 2008; 15: 755–63. [DOI] [PubMed] [Google Scholar]
- 8. Zhang Z, Yamashita H, Toyama T et al. Quantitative determination, by real‐time reverse transcription polymerase chain reaction, of aromatase mRNA in invasive ductal carcinoma of the breast. Breast Cancer Res 2003; 5: R250–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Yamashita H, Nishio M, Toyama T et al. Coexistence of HER2 over‐expression and p53 protein accumulation is a strong prognostic molecular marker in breast cancer. Breast Cancer Res 2004; 6: R24–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Yamashita H, Toyama T, Nishio M et al. p53 protein accumulation predicts resistance to endocrine therapy and decreased post‐relapse survival in metastatic breast cancer. Breast Cancer Res 2006; 8: R48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Yamashita H, Nishio M, Ando Y et al. Stat5 expression predicts response to endocrine therapy and improves survival in estrogen receptor‐positive breast cancer. Endocr Relat Cancer 2006; 13: 885–93. [DOI] [PubMed] [Google Scholar]
- 12. Mita K, Zhang Z, Ando Y et al. Prognostic significance of insulin‐like growth factor binding protein (IGFBP)‐4 and IGFBP‐5 expression in breast cancer. Jpn J Clin Oncol 2007; 37: 575–82. [DOI] [PubMed] [Google Scholar]
- 13. Yamashita H, Ando Y, Nishio M et al. Immunohistochemical evaluation of hormone receptor status for predicting response to endocrine therapy in metastatic breast cancer. Breast Cancer 2006; 13: 74–83. [DOI] [PubMed] [Google Scholar]
- 14. Sugiura H, Toyama T, Hara Y et al. Expression of estrogen receptor beta wild‐type and its variant ERbetacx/beta2 is correlated with better prognosis in breast cancer. Jpn J Clin Oncol 2007; 37: 820–8. [DOI] [PubMed] [Google Scholar]
- 15. Zhang Z, Yamashita H, Toyama T et al. Semi‐quantitative immunohistochemical analysis of aromatase expression in ductal carcinoma in situ of the breast. Breast Cancer Res Treat 2002; 74: 47–53. [DOI] [PubMed] [Google Scholar]
- 16. Yamashita H, Nishio M, Fujii Y, Iwase H. Dominant‐negative Stat5 inhibits growth and induces apoptosis in T47D‐derived tumors in nude mice. Cancer Sci 2004; 95: 662–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Allred DC, Harvey JM, Berardo M, Clark GM. Prognostic and predictive factors in breast cancer by immunohistochemical analysis. Mod Pathol 1998; 11: 155–68. [PubMed] [Google Scholar]
- 18. 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: 87–94. [DOI] [PubMed] [Google Scholar]
- 19. 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: 203–13. [DOI] [PubMed] [Google Scholar]
- 20. Barnadas A, Gil M, Gonzalez 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: 442–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kirkegaard T, Witton CJ, McGlynn LM et al. AKT activation predicts outcome in breast cancer patients treated with tamoxifen. J Pathol 2005; 207: 139–46. [DOI] [PubMed] [Google Scholar]
- 22. Perez‐Tenorio G, Stal O. Activation of AKT/PKB in breast cancer predicts a worse outcome among endocrine treated patients. Br J Cancer 2002; 86: 540–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Tokunaga E, Kataoka A, Kimura Y et al. The association between Akt activation and resistance to hormone therapy in metastatic breast cancer. Eur J Cancer 2006; 42: 629–35. [DOI] [PubMed] [Google Scholar]
- 24. Tokunaga E, Kimura Y, Oki E et al. Akt is frequently activated in HER2/neu‐positive breast cancers and associated with poor prognosis among hormone‐treated patients. Int J Cancer 2006; 118: 284–9. [DOI] [PubMed] [Google Scholar]
- 25. Stoica GE, Franke TF, Wellstein A et al. Estradiol rapidly activates Akt via the ErbB2 signaling pathway. Mol Endocrinol 2003; 17: 818–30. [DOI] [PubMed] [Google Scholar]
- 26. Ellis MJ, Coop A, Singh B et al. Letrozole is more effective neoadjuvant endocrine therapy than tamoxifen for ErbB‐1‐ and/or ErbB‐2‐positive, estrogen receptor‐positive primary breast cancer: evidence from a phase III randomized trial. J Clin Oncol 2001; 19: 3808–16. [DOI] [PubMed] [Google Scholar]
- 27. Dowsett M, Allred C, Knox J 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 2008; 26: 1059–65. [DOI] [PubMed] [Google Scholar]
- 28. Rasmussen BB, Regan MM, Lykkesfeldt AE et al. Adjuvant letrozole versus tamoxifen according to centrally‐assessed ERBB2 status for postmenopausal women with endocrine‐responsive early breast cancer: supplementary results from the BIG 1‐98 randomised trial. Lancet Oncol 2008; 9: 23–8. [DOI] [PubMed] [Google Scholar]
- 29. Viale G, Giobbie‐Hurder A, Regan MM et al. Prognostic and predictive value of centrally reviewed Ki‐67 labeling index in postmenopausal women with endocrine‐responsive breast cancer: results from Breast International Group Trial 1‐98 comparing adjuvant tamoxifen with letrozole. J Clin Oncol 2008; 26: 5569–75. [DOI] [PMC free article] [PubMed] [Google Scholar]