Breast Cancer DOI 10.1007/s12282-013-0515-x
ORIGINAL ARTICLE
Impact of GATA-3 and FOXA1 expression in patients with hormone receptor-positive/HER2-negative breast cancer Yuichi Hisamatsu • Eriko Tokunaga • Nami Yamashita • Sayuri Akiyoshi Satoko Okada • Yuichiro Nakashima • Kenji Taketani • Shinichi Aishima Yoshinao Oda • Masaru Morita • Yoshihiko Maehara
• •
Received: 21 August 2013 / Accepted: 18 December 2013 Ó The Japanese Breast Cancer Society 2014
Abstract Background Determining the indications for adjuvant chemotherapy (CT) in patients with hormone receptor (HR)-positive/HER2-negative breast cancer are difficult. The transcription factors GATA-binding protein 3 (GATA3) and Forkhead-box protein A1 (FOXA1) are crucial for the hormone responsive phenotype of breast cancer. This study evaluated whether the expression of GATA-3 and FOXA1 is a prognostic and predictive marker of outcomes in patients with HR-positive/HER2-negative breast cancer. Methods The expression of GATA-3 and FOXA1 was analyzed immunohistochemically in 214 patients with invasive breast cancer to evaluate the association with the clinicopathological features and the prognosis. Results GATA-3 expression was positively correlated with FOXA1 expression (P \ 0.0001). Both GATA-3 and FOXA1 were positively correlated with ER (P \ 0.0001 each) and PR expression (P = 0.0001 and P = 0.0009, respectively), and inversely correlated with nuclear grade (P = 0.0002 and P = 0.0018, respectively) and Ki67 index (P = 0.0052 and P = 0.0049, respectively). Expression of GATA-3 and FOXA1 was associated with better prognosis. FOXA1 was an independent favorable
Y. Hisamatsu E. Tokunaga (&) N. Yamashita S. Akiyoshi S. Okada Y. Nakashima K. Taketani M. Morita Y. Maehara Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, 812-8582 Fukuoka, Japan e-mail: [email protected] S. Aishima Y. Oda Department of Anatomic Pathology, Pathological Sciences, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
prognostic marker in HR-positive/HER2-negative breast cancer. Disease-free survival rates were similar in patients with HR-positive/HER2-negative breast cancer and high FOXA1 expression given adjuvant hormone therapy (HT) alone and those given CT plus HT. Conclusion GATA-3 and FOXA1 are associated with a less aggressive phenotype and a better prognosis in patients with HR-positive/HER2-negative breast cancer. FOXA1 may be useful in identifying those patients who may not require adjuvant CT. Keywords GATA-3 FOXA1 HR-positive/HER2-negative Adjuvant therapy Breast cancer
Introduction Breast cancer does not comprise a single disease, but rather a heterogeneous mix of diseases. It can be subdivided by gene expression profile into at least four subtypes: luminal A, luminal B, basal-like, and HER2-enriched [1, 2]. These subtypes can be distinguished by immunohistochemistry (IHC) for several markers, including estrogen receptor (ER), progesterone receptor (PR), human epidermal receptor 2 (HER2), and Ki67 [3, 4]. These subtypes must be considered to determine indicators for the use of adjuvant systemic therapy [5]. Endocrine therapy is the most important treatment for patients with hormone receptor (HR, ER and/or PR)-positive breast cancer. A major issue is determining the indicators for adjuvant chemotherapy (CT) in the treatment of HR-positive/HER2-negative patients [5]. Multigene assays such as Oncotype DxÒ and MammaprintÒ are efficient in determining tumor phenotype [6], with the results of
123
Breast Cancer
clinical trials using these techniques anticipated in the near future. However, these assays are very expensive, have not been approved by medical insurance providers in many countries, and have some limitations in their clinical use. Novel strategies are, therefore, needed to provide prognostic and predictive information in a more cost-effective and efficient manner [7]. One such strategy is to develop IHC assays that can identify patients with HR-positive/ HER2-negative subtypes who can safely avoid cytotoxic CT [7]. Among possible markers are those associated with the ER/GATA-binding protein 3 (GATA-3)/Forkhead-box protein A1 (FOXA1) network, which functions normally in regulating hormone sensitivity of breast tissue [7]. The GATA family of transcription factors is critical for the development and differentiation of various cell types in vertebrates [8]. GATA-3 regulates the lineage determination and differentiation of many cell types [9, 10], as well as playing a crucial role during multiple stages of mammary gland development, including the formation of terminal end buds at puberty and luminal cell differentiation [11, 12]. GATA-3 has been implicated in breast oncogenesis, with the highest expression levels of this gene observed in the luminal subtypes of breast cancer [1, 2]. Moreover, low GATA-3 expression is strongly associated with a higher histological grade, positive lymph nodes, ER and PR negative status, and HER2 overexpression, all of which are of poor prognosis [13]. Forkhead-box protein A1, a member of the forkhead family of transcription factors, is expressed in many organs, including the mammary gland. FOXA1 can bind to the promoters of more than 100 genes associated with metabolic processes, the regulation of signaling pathways, and the cell cycle [14, 15]. FOXA1 interacts with the cis-regulatory regions of heterochromatin and enhances the interaction between ERa and chromatin [16, 17]. FOXA1 is required for almost all ER binding events in breast cancer cells [18]. FOXA1 expression has been associated with luminal subtypes and a good prognosis in patients with ER-positive breast cancers [19, 20, 21], and it is an independent factor for prognosis in HR-positive breast cancer [21, 22]. Positivity for both FOXA1 and FOXP1 has been associated with a favorable prognosis in breast cancer patients treated with tamoxifen [20]. Forkhead-box protein A1 expression is strongly and significantly correlated with GATA-3 expression in breast tumors [11], suggesting that these transcription factors interact. The functional cooperation of GATA-3, FOXA1 and ER is considered necessary for normal mammary gland development, particularly the differentiation of mammary stem/ progenitor cells to ER-positive lineage. This network is also necessary for the ERa functional signature specific to luminal type breast cancers. Therefore, the presence of an intact GATA-3/FOXA1/ER network appears to be critical for the hormone responsive phenotype observed in breast cancer [7]. In a study analyzing both FOXA1 and GATA-3, FOXA1 was
123
found to be an independent prognostic marker in patients with ER-positive breast cancer [21]. It is important to validate these important findings in other patient cohorts. Indeed, we also reported that FOXA1 is an independent prognostic factor in HR-positive breast cancer, but we did not assay GATA-3 expression in these patients. The present study, therefore, investigated the expression of GATA-3 and FOXA1 in Japanese patients with HRpositive/HER2-negative breast cancer, as well as assessing the association between their levels of expression and clinical features in these patients. GATA-3 and FOXA1 expression was associated with a less aggressive phenotype, and FOXA1 expression was shown to be an independently favorable prognostic factor. FOXA1 may, therefore, be useful in identifying HR-positive/HER2negative breast cancer patients with an excellent prognosis who may not require adjuvant CT.
Materials and methods Patient information Invasive breast cancer tissue specimens were obtained from 214 consecutive Japanese patients who underwent surgery at the Department of Surgery and Science, Kyushu University Hospital, between 2000 and 2007. Informed consent was obtained from all patients before collecting tissue samples, and the study was approved by the institutional review board of the university. Of these 214 patients, 71 (33.2 %) had pathological Stage I, 116 (54.2 %) had Stage II, and 27 (12.6 %) had Stage III tumors. HT consisted of an aromatase inhibitor, either tamoxifen or toremifene, and an LH-RH agonist for premenopausal women. CT regimens included epirubicin and cyclophosphamide, 5-fluorouracil (5-FU), epirubicin and cyclophosphamide), cyclophosphamide, methotrexate and 5-FU and taxanes, based on clinicopathological findings. Median follow-up was 1742 days. Immunohistochemistry for GATA-3 and FOXA1 The expression of GATA-3 and FOXA1 in formalin-fixed, paraffin-embedded tissue specimens was analyzed by IHC. The sections were deparaffinized with xylene and rehydrated. GATA-3 expression was analyzed as described [23, 24]. Briefly, the sections were incubated in 10 mM citrate buffer (pH 6.0) in an autoclave at 121 °C for 5 min for antigen retrieval. The slides were cooled for 20 min at room temperature and treated for 30 min with 3 % H2O2 in methanol to block endogenous peroxidases. Nonspecific antibody binding was blocked by incubating the sections for 10 min with normal goat serum (Dako, Glostrup, Denmark). The slides were
Breast Cancer Fig. 1 Immunohistochemical analysis of GATA-3 and FOXA1 expression in breast cancer. Representative images showing GATA-3 and FOXA1 expression in ER-positive (a, c) and ER-negative (b, d) breast cancer. ER-positive breast tumors exhibited high GATA-3 a and FOXA1 c expression, whereas ER-negative breast cancers exhibited low GATA-3 b and FOXA1 d expression. Original magnification, 4009. e Positive correlation between GATA-3 and FOXA1 expression (P = 0.7285, P \ 0.0001). f Positive correlations between GATA-3 and FOXA1 expression and HR (ER and/or PR) expression
incubated overnight at 4 °C with mouse monoclonal antiGATA-3 antibody (1:50, Santa Cruz Biochemical, Santa Cruz, CA, USA) and labeled with the Envision Detection System (Dako) for 1 h at room temperature. The sections were developed with 3,30 -diaminobenzidine tetrahydrochloride (DAB plus, Dako) and counterstained with 10 % Mayer’s hematoxylin, dehydrated, and mounted. IHC for FOXA1 was performed as previously described [22]. GATA-3 and FOXA1 expression were scored as the percentage of positively stained nuclei in a maximum of 1000 cells per sample [22]. GATA-3 expression was dichotomized by its median value (77.4 %) as either low (\77.4 %) or high (C77.4 %). Similarly, FOXA1 expression was dichotomized by its median value (71.7 %) as either low (\71.7 %) or high (C71.7 %).
Evaluation of ER, PR, HER2, and Ki67 Estrogen receptor and PR were considered positive if C1 % of the nuclei in the tumor were positively stained by IHC [3, 4]. Tumors were considered HER2-positive if they were scored as either 3? by IHC or as or 2? with HER2 amplification (ratio [2.0) by fluorescence in situ hybridization. Ki67 was evaluated as described [22], with the Ki67 index dichotomized as high (C14 %, Ki67high) or low (\14 %, Ki67low) [3]. The study population was classified into four breast cancer subtypes based on ER, PR, HER2 and Ki67 status, as described [3, 4]: ‘‘luminal A’’ (HR-positive, HER2negative and Ki67low), ‘‘luminal B’’ (HR-positive and either HER2-positive and/or Ki67high), HER2-enriched (ER-
123
Breast Cancer
negative, PR-negative and HER2-positive), and triple negative (ER-negative, PR-negative and HER2-negative).
P \ 0.05. Disease-free survival (DFS) was defined as the period from diagnosis to death or tumor recurrence.
Statistical analyses
Results
All molecular and IHC analyses were performed by investigators blinded to the clinical data. Statistical analyses were performed using the JMP software package, version 9.0.2 (SAS Institute Inc., Cary, NC). Associations between GATA-3 and FOXA1 expression and clinicopathological characteristics were assessed using v2 tests. Survival curves were plotted using the Kaplan–Meier method and compared using the log-rank test. Survival data were evaluated using a multivariate Cox proportional hazards model. Differences were considered significant at
Association between GATA-3 and FOXA1 expression and clinicopathological characteristics The expression of GATA-3 and FOXA1 was analyzed by IHC (Fig. 1). High GATA-3 and FOXA1 expression was observed in ER-positive breast tumors (Fig. 1a, c), whereas both were low in ER-negative breast cancers (Fig. 1b, d). The levels of expression of GATA-3 and FOXA1 showed a significantly positive correlation (P \ 0.0001, Fig. 1e). GATA-3 and FOXA1 expression varied among patients in
Table 1 Association between GATA-3 and FOXA1 expression and clinicopathological characteristics of breast cancer patients Factors
GATA-3
P-value
Low (n = 137)
High (n = 137)
Age, mean ± SD, years
55.8 ± 1.2
55.0 ± 1.2
0.6334
Tumor size (cm) B2
0.0334
FOXA1
P-value
Low (n = 137)
High (n = 137)
55.7 ± 1.2
55.0 ± 1.2
0.6810 0.0306
42 (39.3)
61 (57.0)
42 (39.3)
61 (57.0)
2\ , B5
52 (48.6)
37 (34.6)
53 (49.5)
36 (33.6)
5\
13 (12.1)
9 (8.4)
12 (11.2)
10 (9.4)
Negative
57 (53.3)
64 (59.8)
Positive
50 (46.7)
43 (40.2)
1
36 (33.6)
52 (48.6)
36 (33.6)
52 (48.6)
2
23 (21.5)
35 (32.7)
25 (23.4)
33 (30.8)
3
48 (44.9)
20 (18.7)
46 (43.0)
22 (20.6)
Negative
49 (45.8)
7 (6.5)
45 (42.1)
11 (10.3)
Positive
58 (54.2)
100 (93.5)
62 (57.9)
96 (89.7)
61 (57.0)
33 (30.8)
46 (43.0)
74 (69.2)
Negative
79 (73.8)
89 (83.2)
Positive
28 (26.2)
18 (16.8)
LN metastasis 0.3342
56 (52.3)
65 (60.7)
51 (47.7)
42 (39.3)
0.2143
Nuclear grade 0.0002
0.0018
ER \0.0001
\0.0001
PR Negative Positive HER2
0.0001
0.0950
59 (55.1)
35 (32.7)
48 (44.9)
72 (67.3)
78 (72.9)
90 (84.1)
29 (27.1)
17 (15.9)
0.0009
0.0448
Subtype \0.0001
‘‘Luminal A’’
28 (26.2)
40 (37.4)
32 (29.9)
36 (33.7)
‘‘Luminal B’’
34 (31.8)
62 (57.9)
33 (30.8)
63 (58.9)
HER2-enriched
20 (18.7)
3 (2.8)
19 (17.8)
4 (3.7)
Triple negative
25 (23.3)
2 (1.9)
23 (21.5)
4 (3.7)
Ki67 index
0.20 ± 0.02
0.14 ± 0.01
0.20 ± 0.01
0.14 ± 0.01
0.0052
FOXA1 Low
80 (74.8)
27 (25.2)
High
27 (25.2)
80 (74.8)
\0.0001
Values are n (%), LN lymph node, ER estrogen receptor, PR progesterone receptor
123
\0.0001
0.0049
Breast Cancer Fig. 2 Survival curves of the entire cohort of breast cancer patients relative to levels of GATA-3 and FOXA1 expression. a OS (P = 0.0143) and b DFS (P = 0.0011) in patients with high versus low GATA-3 expression. c OS (P = 0.0211) and d DFS (P = 0.0001) in patients with high versus low FOXA1 expression
Fig. 3 Survival curves of patients with HR-positive/ HER2-negative breast cancer relative to levels of GATA-3 and FOXA1 expression. a OS (P = NS) and b DFS (P = 0.0094) in patients with high versus low GATA-3 expression. c OS (P = NS) and d DFS (P = 0.0002) in patients with high versus low FOXA1 expression
both the HR-negative and HR-positive groups, but both were significantly higher in HR-positive patients (P \ 0.0001 each, Fig. 1f). The associations between GATA-3 and FOXA1 expression and clinicopathological characteristics were
also evaluated (Table 1). GATA-3 and FOXA1 expression was positively correlated with ER (P \ 0.0001 each) and PR (P = 0.0001 and P = 0.0009, respectively) expression, and negatively correlated with nuclear grade (P = 0.0002 and P = 0.0018, respectively) and Ki67 index
123
Breast Cancer Fig. 4 Associations between combinations of GATA-3 and FOXA1 expression and DFS in all patients and in patients with HR-positive/HER2-negative breast cancer. a Analysis of prognosis relative to both FOXA1 (F) and GATA-3 (G) expression in patients with HR-positive/HER2-negative breast cancer. DFS was similar in patients with F-high (H) and G-high (H) and those with F-high (H) and G-low (L) expression. b, c High FOXA1 expression was not correlated with prognosis in HR-positive/HER2-negative patients with nuclear grade 1 (b), but was significantly correlated with prognosis in HR-positive/HER2-negative patients with grades 2, 3 (P = 0.0002) (c). d, e High FOXA1 expression was associated with significantly better prognosis in HR-positive/ HER2-negative patients negative (P = 0.0004) d and positive (P = 0.0120) e for lymph node metastasis. f DFS was similar in patients with HRpositive/HER2-negative breast cancer with high FOXA1 expression given HT alone or CT plus HT
(P = 0.0052 and P = 0.0049, respectively). Few HER2enriched and triple-negative tumors showed high GATA-3 and FOXA1 expression (both P \ 0.0001). These features suggest that tumors with high GATA-3 and FOXA1 expression are less aggressive than tumors with low expression of both. GATA-3 and FOXA1 expression was not associated with lymph node metastasis. Association between GATA-3 and FOXA1 expression and prognosis The association between GATA-3 and FOXA1 expression and prognosis was evaluated. High GATA-3 expression was associated with a significantly better prognosis than low GATA-3 expression, as shown by both overall survival (OS) [hazard ratio (HR) 0.33, 95 % confidence interval (CI)
123
0.12–0.80; P = 0.0143] (Fig. 2a) and DFS (HR 0.37, 95 % CI 0.19–0.67, P = 0.0011; Fig. 2b). Similarly, high FOXA1 expression was associated with significantly longer OS (HR 0.35, 95 % CI 0.12–0.85, P = 0.0211; Fig. 2c) and DFS (HR 0.30, 95 % CI 0.15–0.55, P = 0.0001; Fig. 2d). The association between GATA-3 and FOXA1 expression and prognosis was also evaluated in patients with HRpositive/HER2-negative breast cancers. The levels of expressions of these two proteins were not associated with significantly longer OS (Fig. 3a, c). In contrast, high GATA-3 expression was associated with a significantly longer DFS than low GATA-3 expression (HR 0.38, 95 % CI 0.17–0.80, P = 0.0094; (Fig. 3b). Similarly, high FOXA1 expression was associated with significantly longer DFS (HR 0.23, 95 % CI 0.10–0.51, P = 0.0002; Fig. 3d).
Breast Cancer Table 2 Multivariate analysis of recurrence in all patients and in HR-positive/HER2-negative patients Variables
Parameters
HR
95 % CI
pvalue
All cases pT factor
1 vs. 2,3
0.84
0.44–1.62
0.5912
LN metastasis
Negative vs. positive
4.10
2.13–8.42
0.0001
Nuclear grade
1 vs. 2,3
3.16
1.41–7.59
0.0046
Ki67 index
Low vs. high
1.16
0.60–2.32
0.6576
Subtype
0.50
0.23–1.04
0.0648
GATA-3
Luminal A, B vs. others High vs. low
1.74
0.83–3.81
0.1478
FOXA1
High vs. low
2.77
1.26–6.37
0.0103
HR-positive/HER2-negative cases T factor
1 vs. 2,3
0.73
0.28–1.92
0.5137
LN metastasis
Negative vs. positive
4.34
1.71–12.20
0.0016
Nuclear grade
1 vs. 2,3
2.65
1.06–6.89
0.0375
Ki67 index GATA-3
Low vs. high High vs. low
1.26 1.26
0.53–2.99 0.52–3.16
0.5910 0.6117
FOXA1
High vs. low
4.22
1.65–11.76
0.0023
HR hazard ratio, CI confidence interval, pT pathological T
The study next analyzed the prognosis associated with combinations of GATA-3 and FOXA1 expression in patients with HR-positive/HER2-negative breast cancer (Fig. 4a). As expected, low expression of both GATA-3 and FOXA1 was correlated with poor patient prognosis, whereas DFS was longer in patients with high than low expression of FOXA1, regardless of the level of GATA-3 expression. These findings suggested that FOXA1 expression has a greater effect on prognosis than GATA-3 expression in patients with HR-positive/HER2-negative breast cancer. Multivariate Cox hazard analyses with models that included pathological T (pT) stage, lymph node metastasis, nuclear grade, subtype, Ki67 index, GATA-3, and FOXA1, all of which were significantly associated with DFS in univariate analysis, revealed that high FOXA1 expression was independently associated with better patient prognosis in the entire cohort (HR 2.77, 95 % CI 1.26–6.37, P = 0.0103). Importantly, multivariate analysis in models including these factors showed that FOXA1 expression in patients with HR-positive/HER2-negative breast cancer was independently associated with a good prognosis (HR 4.22, 95 % CI 1.65–11.76, P = 0.0023; Table 2). Conversely, GATA-3 expression was not an independent prognostic factor in the total cohort or in patients with HRpositive/HER2-negative breast cancer.
Impact of FOXA1 expression on the prognosis of HRpositive/HER2-negative breast cancer according to nuclear grade or lymph node metastasis In the entire patient cohort, FOXA1 expression was significantly associated with a lower nuclear grade. High FOXA1 expression was also associated with a significantly better prognosis than low FOXA1 expression in patients with grades 2 and 3 HR-positive/HER2-negative breast cancer (HR 0.18, 95 % CI 0.06–0.48, P = 0.0002; Fig. 4c), but not in patients with nuclear grade 1 (Fig. 4b). This result, therefore, indicates that FOXA1 expression has a significant effect on a good prognosis in patients with HR-positive/HER2-negative breast cancers with high nuclear grade. The relationship between FOXA1 expression and prognosis according to nodal status was also examined. High FOXA1 expression was associated with a significantly better prognosis in HR-positive/HER2-negative patients without (HR 0.29, 95 % CI 0.12–0.60, P = 0.0004; Fig. 4d) and with (HR 0.36, 95 % CI 0.14–0.89, P = 0.0120; Fig. 4e) lymph node metastasis. These results, therefore, show that FOXA1 expression significantly enhances the prognosis of patients with HRpositive/HER2-negative breast cancer, regardless of lymph node metastasis. Impact of FOXA1 expression on prognosis associated with adjuvant therapy The impact of FOXA1 expression on prognosis was also assessed in patients with HR-positive/HER2-negative breast cancer receiving adjuvant therapy. The groups of these patients with high and low expression of FOXA1 had similar tumor size, rates of lymph node metastasis, nuclear grade, Ki67 index, and adjuvant therapy regimens (Table 3). Among patients with high FOXA1 expression, DFS was similar in those given HT alone and those given CT plus HT (Fig. 4f).
Discussion Patients with HR-positive/HER2-negative breast cancers with high histological grade, high proliferation profile, and low ER and PR expression are usually treated with cytotoxic CT [5]. However, these factors often provide confusion in selecting those patients who require adjuvant CT. The ER/GATA-3/FOXA1 network includes factors that can be evaluated by IHC to identify patients with HRpositive/HER2-negative subtypes who could safely avoid cytotoxic CT [7]. FOXA1 and GATA-3 expression predict
123
Breast Cancer Table 3 Association between GATA-3 and FOXA1 expression and clinicopathological characteristics in patients with HR-positive/HER2negative breast cancer Factors
GATA-3
P-value
FOXA1
P-value
Low (n = 54)
High (n = 87)
51.7 ± 1.6
54.0 ± 1.3
B2 2\ , B5
26 (48.2) 20 (37.0)
51 (58.6) 28 (32.2)
5\
8 (14.8)
8 (9.2)
Negative
29 (53.7)
51 (58.6)
Positive
25 (46.3)
36 (41.4)
1
27 (50.0)
46 (52.9)
29 (52.7)
44 (51.2)
2
14 (25.9)
29 (33.3)
14 (25.5)
29 (33.7)
3
13 (24.1)
12 (13.8)
12 (24.8)
13 (15.1)
Negative
2 (3.7)
2 (2.3)
2 (3.6)
2 (2.3)
Positive
52 (96.3)
85 (97.7)
53 (96.4)
84 (97.7)
Negative
14 (25.9)
23 (26.4)
0.9465
13 (23.6)
24 (27.9)
0.5723
Positive Ki67 index
40 (74.1) 0.12 ± 0.01
64 (73.6) 0.13 ± 0.01
0.7604
42 (76.4) 0.12 ± 0.01
62 (72.1) 0.13 ± 0.01
0.5392
None
2 (3.7)
4 (4.6)
0.9534
3 (5.5)
3 (3.5)
0.7482
HT only
31 (57.4)
48 (55.2)
30 (54.6)
49 (57.0)
CT only
6 (11.1)
8 (9.2)
7 (12.7)
7 (8.1)
HT plus CT
15 (27.8)
27 (31.0)
15 (27.3)
27 (31.4)
Age, mean ± SD, years
Low (n = 55)
High (n = 86)
0.2755
51.6 ± 1.6
54.1 ± 1.3
0.2289
0.4046
27 (49.1) 21 (38.2)
50 (58.1) 27 (31.4)
0.5749
7 (12.7)
9 (10.5)
Tumor size (cm)
LN metastasis 0.5670
30 (54.5)
50 (58.1)
25 (45.5)
36 (41.9)
0.6745
Nuclear grade 0.2737
0.4456
ER 0.6300
0.6514
PR
Adjuvant therapy
Values are n (%), LN lymph node, ER estrogen receptor, PR progesterone receptor, CT chemotherapy, HT hormone therapy
a favorable outcome in breast cancer patients, and their expression correlates with a differentiated, luminal tumor subtype [13, 19–22]. The present study found that the levels of expression of GATA-3 and FOXA1 were significantly correlated, and that the expression of both was associated with a less aggressive phenotype and a good prognosis, both in the entire patient cohort and in patients with HR-positive/ HER2-negative breast cancer, findings consistent with previous results [13, 19]. FOXA1, but not GATA-3, was found to be an independent prognosis factor in patients with HR-positive/HER2-negative breast cancer. Patients with high FOXA1 expression derived little benefit from adding CT to HT. Therefore, concurrent evaluation of the expression of both GATA-3 and FOXA1 showed that FOXA1 was a superior prognostic marker in patients with HR-positive/HER2-negative breast cancer. FOXA1 and GATA-3 expression showed significant positive correlation [11]. Chromatin immunoprecipitation suggests that GATA-3 may function upstream of FOXA1
123
[11]. Intriguingly, a recent study that evaluated the association of FOXA1 and GATA-3 expression with Oncotype DXÒ recurrences scores in patients with ER-positive, node-negative breast carcinomas revealed that high FOXA1 expression was significantly correlated with low Oncotype DX recurrence scores [25]. Although FOXA1 and GATA-3 seem to interact, they have distinct functions. Foxa1 deficiency causes a defect in hormone-induced mammary ductal invasion associated with a loss of terminal end bud formation and ERa expression [26]. FOXA1, but not GATA-3, regulates ERa expression [26] and is a major determinant of estrogen-ER activity [18]. FOXA1 is necessary for endocrine responsiveness in ERapositive breast cancers [18, 26]. The expression of FOXA1 is significantly correlated with a lower nuclear grade. However, low FOXA1 expression was associated with a shorter DFS, even in HRpositive/HER2-negative patients with nuclear grade 1. This finding suggests that the prognosis of patients with grade 1 breast tumors may be poor if FOXA1 expression is low. In contrast, patients with grade 3 tumors expressing high
Breast Cancer
levels of FOXA1 were found to have a good prognosis, suggesting that high FOXA1 expression is associated with better patient prognosis, regardless of nuclear grade. In addition, low FOXA1 expression was associated with a shorter DFS in HR-positive/HER2-negative patients regardless of nodal status. The prognosis of patients positive for lymph node metastasis expressing a high level of FOXA1 appears to be good. There was no difference in prognosis between HRpositive/HER2-negative breast cancer patients with high FOXA1 expression who received HT only and those who received CT plus HT. Thus, CT in these patients provides little additional benefit, suggesting that adjuvant CT could be omitted in treating these patients. This study had several limitations. First, this study was retrospective with a relatively small sample size. Second, the optimal cutoff defining high and low FOXA1 expression is unclear. Additional studies in larger numbers of patients are needed. In conclusion, GATA-3 and FOXA1 expression was associated with a less aggressive breast cancer phenotype and a better patient prognosis. Evaluation of FOXA1 expression may provide a cost-effective strategy in the risk stratification of breast cancer patients. The better prognosis observed in patients with HR-positive/HER2-negative breast cancer and high FOXA1 expression suggests that adjuvant CT may be safely omitted. To our knowledge, this is the first study to analyze the significance of GATA-3 and FOXA1 expression in Japanese patients with HR-positive/HER2-negative breast cancer. Acknowledgments This study was supported by grants from the Ministry of Education, Culture, Sports Science, and Technology of Japan (Grant No. 23591896). We are grateful to Ms. Yuko Kubota for her valuable technical assistance. Conflict of interest The authors have no conflict of interest to declare.
References 1. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747–52. 2. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001;98:10869–74. 3. Cheang MC, Chia SK, Voduc D, et al. Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst. 2009;101:736–50. 4. Hugh J, Hanson J, Cheang MC, et al. Breast cancer subtypes and response to docetaxel in node-positive breast cancer: use of an immunohistochemical definition in the BCIRG 001 trial. J Clin Oncol. 2009;27:1168–76. 5. Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thurlimann B, Senn HJ. 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. 2011;22:1736–47.
6. Kim C, Paik S. Gene-expression-based prognostic assays for breast cancer. Nat Rev Clin Oncol. 2010;7:340–7. 7. Badve S, Nakshatri H. Oestrogen-receptor-positive breast cancer: towards bridging histopathological and molecular classifications. J Clin Pathol. 2009;62:6–12. 8. Lim KC, Lakshmanan G, Crawford SE, Gu Y, Grosveld F, Engel JD. GATA3 loss leads to embryonic lethality due to noradrenaline deficiency of the sympathetic nervous system. Nat Genet. 2000;25:209–12. 9. Kaufman CK, Zhou P, Pasolli HA, et al. GATA-3: an unexpected regulator of cell lineage determination in skin. Genes Dev. 2003;17:2108–22. 10. Ting CN, Olson MC, Barton KP, Leiden JM. Transcription factor GATA-3 is required for development of the T-cell lineage. Nature. 1996;384:474–8. 11. Kouros-Mehr H, Slorach EM, Sternlicht MD, Werb Z. GATA-3 maintains the differentiation of the luminal cell fate in the mammary gland. Cell. 2006;127:1041–55. 12. Asselin-Labat ML, Sutherland KD, Barker H, et al. GATA-3 is an essential regulator of mammary-gland morphogenesis and luminal-cell differentiation. Nat Cell Biol. 2007;9:201–9. 13. Mehra R, Varambally S, Ding L, et al. Identification of GATA3 as a breast cancer prognostic marker by global gene expression meta-analysis. Cancer Res. 2005;65:11259–64. 14. Carlsson P, Mahlapuu M. Forkhead transcription factors: key players in development and metabolism. Dev Biol. 2002;250: 1–23. 15. Kaestner KH. The hepatocyte nuclear factor 3 (HNF3 or FOXA) family in metabolism. Trends Endocrinol Metab. 2000;11:281–5. 16. Carroll JS, Brown M. Estrogen receptor target gene: an evolving concept. Mol Endocrinol. 2006;20:1707–14. 17. Laganiere J, Deblois G, Lefebvre C, Bataille AR, Robert F, Giguere V. From the Cover: location analysis of estrogen receptor alpha target promoters reveals that FOXA1 defines a domain of the estrogen response. Proc Natl Acad Sci USA. 2005;102:11651–6. 18. Hurtado A, Holmes KA, Ross-Innes CS, Schmidt D, Carroll JS. FOXA1 is a key determinant of estrogen receptor function and endocrine response. Nat Genet. 2011;43:27–33. 19. Badve S, Turbin D, Thorat MA, et al. FOXA1 expression in breast cancer–correlation with luminal subtype A and survival. Clin Cancer Res. 2007;13:4415–21. 20. Ijichi N, Shigekawa T, Ikeda K, et al. Association of doublepositive FOXA1 and FOXP1 immunoreactivities with favorable prognosis of tamoxifen-treated breast cancer patients. Hormon Cancer. 2012;3:147–59. 21. Mehta RJ, Jain RK, Leung S, et al. FOXA1 is an independent prognostic marker for ER-positive breast cancer. Breast Cancer Res Treat. 2012;131:881–90. 22. Hisamatsu Y, Tokunaga E, Yamashita N, et al. Impact of FOXA1 expression on the prognosis of patients with hormone receptorpositive breast cancer. Ann Surg Oncol. 2012;19:1145–52. 23. Hoch RV, Thompson DA, Baker RJ, Weigel RJ. GATA-3 is expressed in association with estrogen receptor in breast cancer. Int J Cancer. 1999;84:122–8. 24. Parikh P, Palazzo JP, Rose LJ, Daskalakis C, Weigel RJ. GATA3 expression as a predictor of hormone response in breast cancer. J Am Coll Surg. 2005;200:705–10. 25. Ademuyiwa FO, Thorat MA, Jain RK, Nakshatri H, Badve S. Expression of forkhead-box protein A1, a marker of luminal A type breast cancer, parallels low Oncotype DX 21-gene recurrence scores. Mod Pathol. 2010;23:270–5. 26. Bernardo GM, Lozada KL, Miedler JD, et al. FOXA1 is an essential determinant of ERa expression and mammary ductal morphogenesis. Development. 2010;137:2045–54.
123
ORIGINAL ARTICLE
Impact of GATA-3 and FOXA1 expression in patients with hormone receptor-positive/HER2-negative breast cancer Yuichi Hisamatsu • Eriko Tokunaga • Nami Yamashita • Sayuri Akiyoshi Satoko Okada • Yuichiro Nakashima • Kenji Taketani • Shinichi Aishima Yoshinao Oda • Masaru Morita • Yoshihiko Maehara
• •
Received: 21 August 2013 / Accepted: 18 December 2013 Ó The Japanese Breast Cancer Society 2014
Abstract Background Determining the indications for adjuvant chemotherapy (CT) in patients with hormone receptor (HR)-positive/HER2-negative breast cancer are difficult. The transcription factors GATA-binding protein 3 (GATA3) and Forkhead-box protein A1 (FOXA1) are crucial for the hormone responsive phenotype of breast cancer. This study evaluated whether the expression of GATA-3 and FOXA1 is a prognostic and predictive marker of outcomes in patients with HR-positive/HER2-negative breast cancer. Methods The expression of GATA-3 and FOXA1 was analyzed immunohistochemically in 214 patients with invasive breast cancer to evaluate the association with the clinicopathological features and the prognosis. Results GATA-3 expression was positively correlated with FOXA1 expression (P \ 0.0001). Both GATA-3 and FOXA1 were positively correlated with ER (P \ 0.0001 each) and PR expression (P = 0.0001 and P = 0.0009, respectively), and inversely correlated with nuclear grade (P = 0.0002 and P = 0.0018, respectively) and Ki67 index (P = 0.0052 and P = 0.0049, respectively). Expression of GATA-3 and FOXA1 was associated with better prognosis. FOXA1 was an independent favorable
Y. Hisamatsu E. Tokunaga (&) N. Yamashita S. Akiyoshi S. Okada Y. Nakashima K. Taketani M. Morita Y. Maehara Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, 812-8582 Fukuoka, Japan e-mail: [email protected] S. Aishima Y. Oda Department of Anatomic Pathology, Pathological Sciences, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
prognostic marker in HR-positive/HER2-negative breast cancer. Disease-free survival rates were similar in patients with HR-positive/HER2-negative breast cancer and high FOXA1 expression given adjuvant hormone therapy (HT) alone and those given CT plus HT. Conclusion GATA-3 and FOXA1 are associated with a less aggressive phenotype and a better prognosis in patients with HR-positive/HER2-negative breast cancer. FOXA1 may be useful in identifying those patients who may not require adjuvant CT. Keywords GATA-3 FOXA1 HR-positive/HER2-negative Adjuvant therapy Breast cancer
Introduction Breast cancer does not comprise a single disease, but rather a heterogeneous mix of diseases. It can be subdivided by gene expression profile into at least four subtypes: luminal A, luminal B, basal-like, and HER2-enriched [1, 2]. These subtypes can be distinguished by immunohistochemistry (IHC) for several markers, including estrogen receptor (ER), progesterone receptor (PR), human epidermal receptor 2 (HER2), and Ki67 [3, 4]. These subtypes must be considered to determine indicators for the use of adjuvant systemic therapy [5]. Endocrine therapy is the most important treatment for patients with hormone receptor (HR, ER and/or PR)-positive breast cancer. A major issue is determining the indicators for adjuvant chemotherapy (CT) in the treatment of HR-positive/HER2-negative patients [5]. Multigene assays such as Oncotype DxÒ and MammaprintÒ are efficient in determining tumor phenotype [6], with the results of
123
Breast Cancer
clinical trials using these techniques anticipated in the near future. However, these assays are very expensive, have not been approved by medical insurance providers in many countries, and have some limitations in their clinical use. Novel strategies are, therefore, needed to provide prognostic and predictive information in a more cost-effective and efficient manner [7]. One such strategy is to develop IHC assays that can identify patients with HR-positive/ HER2-negative subtypes who can safely avoid cytotoxic CT [7]. Among possible markers are those associated with the ER/GATA-binding protein 3 (GATA-3)/Forkhead-box protein A1 (FOXA1) network, which functions normally in regulating hormone sensitivity of breast tissue [7]. The GATA family of transcription factors is critical for the development and differentiation of various cell types in vertebrates [8]. GATA-3 regulates the lineage determination and differentiation of many cell types [9, 10], as well as playing a crucial role during multiple stages of mammary gland development, including the formation of terminal end buds at puberty and luminal cell differentiation [11, 12]. GATA-3 has been implicated in breast oncogenesis, with the highest expression levels of this gene observed in the luminal subtypes of breast cancer [1, 2]. Moreover, low GATA-3 expression is strongly associated with a higher histological grade, positive lymph nodes, ER and PR negative status, and HER2 overexpression, all of which are of poor prognosis [13]. Forkhead-box protein A1, a member of the forkhead family of transcription factors, is expressed in many organs, including the mammary gland. FOXA1 can bind to the promoters of more than 100 genes associated with metabolic processes, the regulation of signaling pathways, and the cell cycle [14, 15]. FOXA1 interacts with the cis-regulatory regions of heterochromatin and enhances the interaction between ERa and chromatin [16, 17]. FOXA1 is required for almost all ER binding events in breast cancer cells [18]. FOXA1 expression has been associated with luminal subtypes and a good prognosis in patients with ER-positive breast cancers [19, 20, 21], and it is an independent factor for prognosis in HR-positive breast cancer [21, 22]. Positivity for both FOXA1 and FOXP1 has been associated with a favorable prognosis in breast cancer patients treated with tamoxifen [20]. Forkhead-box protein A1 expression is strongly and significantly correlated with GATA-3 expression in breast tumors [11], suggesting that these transcription factors interact. The functional cooperation of GATA-3, FOXA1 and ER is considered necessary for normal mammary gland development, particularly the differentiation of mammary stem/ progenitor cells to ER-positive lineage. This network is also necessary for the ERa functional signature specific to luminal type breast cancers. Therefore, the presence of an intact GATA-3/FOXA1/ER network appears to be critical for the hormone responsive phenotype observed in breast cancer [7]. In a study analyzing both FOXA1 and GATA-3, FOXA1 was
123
found to be an independent prognostic marker in patients with ER-positive breast cancer [21]. It is important to validate these important findings in other patient cohorts. Indeed, we also reported that FOXA1 is an independent prognostic factor in HR-positive breast cancer, but we did not assay GATA-3 expression in these patients. The present study, therefore, investigated the expression of GATA-3 and FOXA1 in Japanese patients with HRpositive/HER2-negative breast cancer, as well as assessing the association between their levels of expression and clinical features in these patients. GATA-3 and FOXA1 expression was associated with a less aggressive phenotype, and FOXA1 expression was shown to be an independently favorable prognostic factor. FOXA1 may, therefore, be useful in identifying HR-positive/HER2negative breast cancer patients with an excellent prognosis who may not require adjuvant CT.
Materials and methods Patient information Invasive breast cancer tissue specimens were obtained from 214 consecutive Japanese patients who underwent surgery at the Department of Surgery and Science, Kyushu University Hospital, between 2000 and 2007. Informed consent was obtained from all patients before collecting tissue samples, and the study was approved by the institutional review board of the university. Of these 214 patients, 71 (33.2 %) had pathological Stage I, 116 (54.2 %) had Stage II, and 27 (12.6 %) had Stage III tumors. HT consisted of an aromatase inhibitor, either tamoxifen or toremifene, and an LH-RH agonist for premenopausal women. CT regimens included epirubicin and cyclophosphamide, 5-fluorouracil (5-FU), epirubicin and cyclophosphamide), cyclophosphamide, methotrexate and 5-FU and taxanes, based on clinicopathological findings. Median follow-up was 1742 days. Immunohistochemistry for GATA-3 and FOXA1 The expression of GATA-3 and FOXA1 in formalin-fixed, paraffin-embedded tissue specimens was analyzed by IHC. The sections were deparaffinized with xylene and rehydrated. GATA-3 expression was analyzed as described [23, 24]. Briefly, the sections were incubated in 10 mM citrate buffer (pH 6.0) in an autoclave at 121 °C for 5 min for antigen retrieval. The slides were cooled for 20 min at room temperature and treated for 30 min with 3 % H2O2 in methanol to block endogenous peroxidases. Nonspecific antibody binding was blocked by incubating the sections for 10 min with normal goat serum (Dako, Glostrup, Denmark). The slides were
Breast Cancer Fig. 1 Immunohistochemical analysis of GATA-3 and FOXA1 expression in breast cancer. Representative images showing GATA-3 and FOXA1 expression in ER-positive (a, c) and ER-negative (b, d) breast cancer. ER-positive breast tumors exhibited high GATA-3 a and FOXA1 c expression, whereas ER-negative breast cancers exhibited low GATA-3 b and FOXA1 d expression. Original magnification, 4009. e Positive correlation between GATA-3 and FOXA1 expression (P = 0.7285, P \ 0.0001). f Positive correlations between GATA-3 and FOXA1 expression and HR (ER and/or PR) expression
incubated overnight at 4 °C with mouse monoclonal antiGATA-3 antibody (1:50, Santa Cruz Biochemical, Santa Cruz, CA, USA) and labeled with the Envision Detection System (Dako) for 1 h at room temperature. The sections were developed with 3,30 -diaminobenzidine tetrahydrochloride (DAB plus, Dako) and counterstained with 10 % Mayer’s hematoxylin, dehydrated, and mounted. IHC for FOXA1 was performed as previously described [22]. GATA-3 and FOXA1 expression were scored as the percentage of positively stained nuclei in a maximum of 1000 cells per sample [22]. GATA-3 expression was dichotomized by its median value (77.4 %) as either low (\77.4 %) or high (C77.4 %). Similarly, FOXA1 expression was dichotomized by its median value (71.7 %) as either low (\71.7 %) or high (C71.7 %).
Evaluation of ER, PR, HER2, and Ki67 Estrogen receptor and PR were considered positive if C1 % of the nuclei in the tumor were positively stained by IHC [3, 4]. Tumors were considered HER2-positive if they were scored as either 3? by IHC or as or 2? with HER2 amplification (ratio [2.0) by fluorescence in situ hybridization. Ki67 was evaluated as described [22], with the Ki67 index dichotomized as high (C14 %, Ki67high) or low (\14 %, Ki67low) [3]. The study population was classified into four breast cancer subtypes based on ER, PR, HER2 and Ki67 status, as described [3, 4]: ‘‘luminal A’’ (HR-positive, HER2negative and Ki67low), ‘‘luminal B’’ (HR-positive and either HER2-positive and/or Ki67high), HER2-enriched (ER-
123
Breast Cancer
negative, PR-negative and HER2-positive), and triple negative (ER-negative, PR-negative and HER2-negative).
P \ 0.05. Disease-free survival (DFS) was defined as the period from diagnosis to death or tumor recurrence.
Statistical analyses
Results
All molecular and IHC analyses were performed by investigators blinded to the clinical data. Statistical analyses were performed using the JMP software package, version 9.0.2 (SAS Institute Inc., Cary, NC). Associations between GATA-3 and FOXA1 expression and clinicopathological characteristics were assessed using v2 tests. Survival curves were plotted using the Kaplan–Meier method and compared using the log-rank test. Survival data were evaluated using a multivariate Cox proportional hazards model. Differences were considered significant at
Association between GATA-3 and FOXA1 expression and clinicopathological characteristics The expression of GATA-3 and FOXA1 was analyzed by IHC (Fig. 1). High GATA-3 and FOXA1 expression was observed in ER-positive breast tumors (Fig. 1a, c), whereas both were low in ER-negative breast cancers (Fig. 1b, d). The levels of expression of GATA-3 and FOXA1 showed a significantly positive correlation (P \ 0.0001, Fig. 1e). GATA-3 and FOXA1 expression varied among patients in
Table 1 Association between GATA-3 and FOXA1 expression and clinicopathological characteristics of breast cancer patients Factors
GATA-3
P-value
Low (n = 137)
High (n = 137)
Age, mean ± SD, years
55.8 ± 1.2
55.0 ± 1.2
0.6334
Tumor size (cm) B2
0.0334
FOXA1
P-value
Low (n = 137)
High (n = 137)
55.7 ± 1.2
55.0 ± 1.2
0.6810 0.0306
42 (39.3)
61 (57.0)
42 (39.3)
61 (57.0)
2\ , B5
52 (48.6)
37 (34.6)
53 (49.5)
36 (33.6)
5\
13 (12.1)
9 (8.4)
12 (11.2)
10 (9.4)
Negative
57 (53.3)
64 (59.8)
Positive
50 (46.7)
43 (40.2)
1
36 (33.6)
52 (48.6)
36 (33.6)
52 (48.6)
2
23 (21.5)
35 (32.7)
25 (23.4)
33 (30.8)
3
48 (44.9)
20 (18.7)
46 (43.0)
22 (20.6)
Negative
49 (45.8)
7 (6.5)
45 (42.1)
11 (10.3)
Positive
58 (54.2)
100 (93.5)
62 (57.9)
96 (89.7)
61 (57.0)
33 (30.8)
46 (43.0)
74 (69.2)
Negative
79 (73.8)
89 (83.2)
Positive
28 (26.2)
18 (16.8)
LN metastasis 0.3342
56 (52.3)
65 (60.7)
51 (47.7)
42 (39.3)
0.2143
Nuclear grade 0.0002
0.0018
ER \0.0001
\0.0001
PR Negative Positive HER2
0.0001
0.0950
59 (55.1)
35 (32.7)
48 (44.9)
72 (67.3)
78 (72.9)
90 (84.1)
29 (27.1)
17 (15.9)
0.0009
0.0448
Subtype \0.0001
‘‘Luminal A’’
28 (26.2)
40 (37.4)
32 (29.9)
36 (33.7)
‘‘Luminal B’’
34 (31.8)
62 (57.9)
33 (30.8)
63 (58.9)
HER2-enriched
20 (18.7)
3 (2.8)
19 (17.8)
4 (3.7)
Triple negative
25 (23.3)
2 (1.9)
23 (21.5)
4 (3.7)
Ki67 index
0.20 ± 0.02
0.14 ± 0.01
0.20 ± 0.01
0.14 ± 0.01
0.0052
FOXA1 Low
80 (74.8)
27 (25.2)
High
27 (25.2)
80 (74.8)
\0.0001
Values are n (%), LN lymph node, ER estrogen receptor, PR progesterone receptor
123
\0.0001
0.0049
Breast Cancer Fig. 2 Survival curves of the entire cohort of breast cancer patients relative to levels of GATA-3 and FOXA1 expression. a OS (P = 0.0143) and b DFS (P = 0.0011) in patients with high versus low GATA-3 expression. c OS (P = 0.0211) and d DFS (P = 0.0001) in patients with high versus low FOXA1 expression
Fig. 3 Survival curves of patients with HR-positive/ HER2-negative breast cancer relative to levels of GATA-3 and FOXA1 expression. a OS (P = NS) and b DFS (P = 0.0094) in patients with high versus low GATA-3 expression. c OS (P = NS) and d DFS (P = 0.0002) in patients with high versus low FOXA1 expression
both the HR-negative and HR-positive groups, but both were significantly higher in HR-positive patients (P \ 0.0001 each, Fig. 1f). The associations between GATA-3 and FOXA1 expression and clinicopathological characteristics were
also evaluated (Table 1). GATA-3 and FOXA1 expression was positively correlated with ER (P \ 0.0001 each) and PR (P = 0.0001 and P = 0.0009, respectively) expression, and negatively correlated with nuclear grade (P = 0.0002 and P = 0.0018, respectively) and Ki67 index
123
Breast Cancer Fig. 4 Associations between combinations of GATA-3 and FOXA1 expression and DFS in all patients and in patients with HR-positive/HER2-negative breast cancer. a Analysis of prognosis relative to both FOXA1 (F) and GATA-3 (G) expression in patients with HR-positive/HER2-negative breast cancer. DFS was similar in patients with F-high (H) and G-high (H) and those with F-high (H) and G-low (L) expression. b, c High FOXA1 expression was not correlated with prognosis in HR-positive/HER2-negative patients with nuclear grade 1 (b), but was significantly correlated with prognosis in HR-positive/HER2-negative patients with grades 2, 3 (P = 0.0002) (c). d, e High FOXA1 expression was associated with significantly better prognosis in HR-positive/ HER2-negative patients negative (P = 0.0004) d and positive (P = 0.0120) e for lymph node metastasis. f DFS was similar in patients with HRpositive/HER2-negative breast cancer with high FOXA1 expression given HT alone or CT plus HT
(P = 0.0052 and P = 0.0049, respectively). Few HER2enriched and triple-negative tumors showed high GATA-3 and FOXA1 expression (both P \ 0.0001). These features suggest that tumors with high GATA-3 and FOXA1 expression are less aggressive than tumors with low expression of both. GATA-3 and FOXA1 expression was not associated with lymph node metastasis. Association between GATA-3 and FOXA1 expression and prognosis The association between GATA-3 and FOXA1 expression and prognosis was evaluated. High GATA-3 expression was associated with a significantly better prognosis than low GATA-3 expression, as shown by both overall survival (OS) [hazard ratio (HR) 0.33, 95 % confidence interval (CI)
123
0.12–0.80; P = 0.0143] (Fig. 2a) and DFS (HR 0.37, 95 % CI 0.19–0.67, P = 0.0011; Fig. 2b). Similarly, high FOXA1 expression was associated with significantly longer OS (HR 0.35, 95 % CI 0.12–0.85, P = 0.0211; Fig. 2c) and DFS (HR 0.30, 95 % CI 0.15–0.55, P = 0.0001; Fig. 2d). The association between GATA-3 and FOXA1 expression and prognosis was also evaluated in patients with HRpositive/HER2-negative breast cancers. The levels of expressions of these two proteins were not associated with significantly longer OS (Fig. 3a, c). In contrast, high GATA-3 expression was associated with a significantly longer DFS than low GATA-3 expression (HR 0.38, 95 % CI 0.17–0.80, P = 0.0094; (Fig. 3b). Similarly, high FOXA1 expression was associated with significantly longer DFS (HR 0.23, 95 % CI 0.10–0.51, P = 0.0002; Fig. 3d).
Breast Cancer Table 2 Multivariate analysis of recurrence in all patients and in HR-positive/HER2-negative patients Variables
Parameters
HR
95 % CI
pvalue
All cases pT factor
1 vs. 2,3
0.84
0.44–1.62
0.5912
LN metastasis
Negative vs. positive
4.10
2.13–8.42
0.0001
Nuclear grade
1 vs. 2,3
3.16
1.41–7.59
0.0046
Ki67 index
Low vs. high
1.16
0.60–2.32
0.6576
Subtype
0.50
0.23–1.04
0.0648
GATA-3
Luminal A, B vs. others High vs. low
1.74
0.83–3.81
0.1478
FOXA1
High vs. low
2.77
1.26–6.37
0.0103
HR-positive/HER2-negative cases T factor
1 vs. 2,3
0.73
0.28–1.92
0.5137
LN metastasis
Negative vs. positive
4.34
1.71–12.20
0.0016
Nuclear grade
1 vs. 2,3
2.65
1.06–6.89
0.0375
Ki67 index GATA-3
Low vs. high High vs. low
1.26 1.26
0.53–2.99 0.52–3.16
0.5910 0.6117
FOXA1
High vs. low
4.22
1.65–11.76
0.0023
HR hazard ratio, CI confidence interval, pT pathological T
The study next analyzed the prognosis associated with combinations of GATA-3 and FOXA1 expression in patients with HR-positive/HER2-negative breast cancer (Fig. 4a). As expected, low expression of both GATA-3 and FOXA1 was correlated with poor patient prognosis, whereas DFS was longer in patients with high than low expression of FOXA1, regardless of the level of GATA-3 expression. These findings suggested that FOXA1 expression has a greater effect on prognosis than GATA-3 expression in patients with HR-positive/HER2-negative breast cancer. Multivariate Cox hazard analyses with models that included pathological T (pT) stage, lymph node metastasis, nuclear grade, subtype, Ki67 index, GATA-3, and FOXA1, all of which were significantly associated with DFS in univariate analysis, revealed that high FOXA1 expression was independently associated with better patient prognosis in the entire cohort (HR 2.77, 95 % CI 1.26–6.37, P = 0.0103). Importantly, multivariate analysis in models including these factors showed that FOXA1 expression in patients with HR-positive/HER2-negative breast cancer was independently associated with a good prognosis (HR 4.22, 95 % CI 1.65–11.76, P = 0.0023; Table 2). Conversely, GATA-3 expression was not an independent prognostic factor in the total cohort or in patients with HRpositive/HER2-negative breast cancer.
Impact of FOXA1 expression on the prognosis of HRpositive/HER2-negative breast cancer according to nuclear grade or lymph node metastasis In the entire patient cohort, FOXA1 expression was significantly associated with a lower nuclear grade. High FOXA1 expression was also associated with a significantly better prognosis than low FOXA1 expression in patients with grades 2 and 3 HR-positive/HER2-negative breast cancer (HR 0.18, 95 % CI 0.06–0.48, P = 0.0002; Fig. 4c), but not in patients with nuclear grade 1 (Fig. 4b). This result, therefore, indicates that FOXA1 expression has a significant effect on a good prognosis in patients with HR-positive/HER2-negative breast cancers with high nuclear grade. The relationship between FOXA1 expression and prognosis according to nodal status was also examined. High FOXA1 expression was associated with a significantly better prognosis in HR-positive/HER2-negative patients without (HR 0.29, 95 % CI 0.12–0.60, P = 0.0004; Fig. 4d) and with (HR 0.36, 95 % CI 0.14–0.89, P = 0.0120; Fig. 4e) lymph node metastasis. These results, therefore, show that FOXA1 expression significantly enhances the prognosis of patients with HRpositive/HER2-negative breast cancer, regardless of lymph node metastasis. Impact of FOXA1 expression on prognosis associated with adjuvant therapy The impact of FOXA1 expression on prognosis was also assessed in patients with HR-positive/HER2-negative breast cancer receiving adjuvant therapy. The groups of these patients with high and low expression of FOXA1 had similar tumor size, rates of lymph node metastasis, nuclear grade, Ki67 index, and adjuvant therapy regimens (Table 3). Among patients with high FOXA1 expression, DFS was similar in those given HT alone and those given CT plus HT (Fig. 4f).
Discussion Patients with HR-positive/HER2-negative breast cancers with high histological grade, high proliferation profile, and low ER and PR expression are usually treated with cytotoxic CT [5]. However, these factors often provide confusion in selecting those patients who require adjuvant CT. The ER/GATA-3/FOXA1 network includes factors that can be evaluated by IHC to identify patients with HRpositive/HER2-negative subtypes who could safely avoid cytotoxic CT [7]. FOXA1 and GATA-3 expression predict
123
Breast Cancer Table 3 Association between GATA-3 and FOXA1 expression and clinicopathological characteristics in patients with HR-positive/HER2negative breast cancer Factors
GATA-3
P-value
FOXA1
P-value
Low (n = 54)
High (n = 87)
51.7 ± 1.6
54.0 ± 1.3
B2 2\ , B5
26 (48.2) 20 (37.0)
51 (58.6) 28 (32.2)
5\
8 (14.8)
8 (9.2)
Negative
29 (53.7)
51 (58.6)
Positive
25 (46.3)
36 (41.4)
1
27 (50.0)
46 (52.9)
29 (52.7)
44 (51.2)
2
14 (25.9)
29 (33.3)
14 (25.5)
29 (33.7)
3
13 (24.1)
12 (13.8)
12 (24.8)
13 (15.1)
Negative
2 (3.7)
2 (2.3)
2 (3.6)
2 (2.3)
Positive
52 (96.3)
85 (97.7)
53 (96.4)
84 (97.7)
Negative
14 (25.9)
23 (26.4)
0.9465
13 (23.6)
24 (27.9)
0.5723
Positive Ki67 index
40 (74.1) 0.12 ± 0.01
64 (73.6) 0.13 ± 0.01
0.7604
42 (76.4) 0.12 ± 0.01
62 (72.1) 0.13 ± 0.01
0.5392
None
2 (3.7)
4 (4.6)
0.9534
3 (5.5)
3 (3.5)
0.7482
HT only
31 (57.4)
48 (55.2)
30 (54.6)
49 (57.0)
CT only
6 (11.1)
8 (9.2)
7 (12.7)
7 (8.1)
HT plus CT
15 (27.8)
27 (31.0)
15 (27.3)
27 (31.4)
Age, mean ± SD, years
Low (n = 55)
High (n = 86)
0.2755
51.6 ± 1.6
54.1 ± 1.3
0.2289
0.4046
27 (49.1) 21 (38.2)
50 (58.1) 27 (31.4)
0.5749
7 (12.7)
9 (10.5)
Tumor size (cm)
LN metastasis 0.5670
30 (54.5)
50 (58.1)
25 (45.5)
36 (41.9)
0.6745
Nuclear grade 0.2737
0.4456
ER 0.6300
0.6514
PR
Adjuvant therapy
Values are n (%), LN lymph node, ER estrogen receptor, PR progesterone receptor, CT chemotherapy, HT hormone therapy
a favorable outcome in breast cancer patients, and their expression correlates with a differentiated, luminal tumor subtype [13, 19–22]. The present study found that the levels of expression of GATA-3 and FOXA1 were significantly correlated, and that the expression of both was associated with a less aggressive phenotype and a good prognosis, both in the entire patient cohort and in patients with HR-positive/ HER2-negative breast cancer, findings consistent with previous results [13, 19]. FOXA1, but not GATA-3, was found to be an independent prognosis factor in patients with HR-positive/HER2-negative breast cancer. Patients with high FOXA1 expression derived little benefit from adding CT to HT. Therefore, concurrent evaluation of the expression of both GATA-3 and FOXA1 showed that FOXA1 was a superior prognostic marker in patients with HR-positive/HER2-negative breast cancer. FOXA1 and GATA-3 expression showed significant positive correlation [11]. Chromatin immunoprecipitation suggests that GATA-3 may function upstream of FOXA1
123
[11]. Intriguingly, a recent study that evaluated the association of FOXA1 and GATA-3 expression with Oncotype DXÒ recurrences scores in patients with ER-positive, node-negative breast carcinomas revealed that high FOXA1 expression was significantly correlated with low Oncotype DX recurrence scores [25]. Although FOXA1 and GATA-3 seem to interact, they have distinct functions. Foxa1 deficiency causes a defect in hormone-induced mammary ductal invasion associated with a loss of terminal end bud formation and ERa expression [26]. FOXA1, but not GATA-3, regulates ERa expression [26] and is a major determinant of estrogen-ER activity [18]. FOXA1 is necessary for endocrine responsiveness in ERapositive breast cancers [18, 26]. The expression of FOXA1 is significantly correlated with a lower nuclear grade. However, low FOXA1 expression was associated with a shorter DFS, even in HRpositive/HER2-negative patients with nuclear grade 1. This finding suggests that the prognosis of patients with grade 1 breast tumors may be poor if FOXA1 expression is low. In contrast, patients with grade 3 tumors expressing high
Breast Cancer
levels of FOXA1 were found to have a good prognosis, suggesting that high FOXA1 expression is associated with better patient prognosis, regardless of nuclear grade. In addition, low FOXA1 expression was associated with a shorter DFS in HR-positive/HER2-negative patients regardless of nodal status. The prognosis of patients positive for lymph node metastasis expressing a high level of FOXA1 appears to be good. There was no difference in prognosis between HRpositive/HER2-negative breast cancer patients with high FOXA1 expression who received HT only and those who received CT plus HT. Thus, CT in these patients provides little additional benefit, suggesting that adjuvant CT could be omitted in treating these patients. This study had several limitations. First, this study was retrospective with a relatively small sample size. Second, the optimal cutoff defining high and low FOXA1 expression is unclear. Additional studies in larger numbers of patients are needed. In conclusion, GATA-3 and FOXA1 expression was associated with a less aggressive breast cancer phenotype and a better patient prognosis. Evaluation of FOXA1 expression may provide a cost-effective strategy in the risk stratification of breast cancer patients. The better prognosis observed in patients with HR-positive/HER2-negative breast cancer and high FOXA1 expression suggests that adjuvant CT may be safely omitted. To our knowledge, this is the first study to analyze the significance of GATA-3 and FOXA1 expression in Japanese patients with HR-positive/HER2-negative breast cancer. Acknowledgments This study was supported by grants from the Ministry of Education, Culture, Sports Science, and Technology of Japan (Grant No. 23591896). We are grateful to Ms. Yuko Kubota for her valuable technical assistance. Conflict of interest The authors have no conflict of interest to declare.
References 1. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747–52. 2. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001;98:10869–74. 3. Cheang MC, Chia SK, Voduc D, et al. Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst. 2009;101:736–50. 4. Hugh J, Hanson J, Cheang MC, et al. Breast cancer subtypes and response to docetaxel in node-positive breast cancer: use of an immunohistochemical definition in the BCIRG 001 trial. J Clin Oncol. 2009;27:1168–76. 5. Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thurlimann B, Senn HJ. 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. 2011;22:1736–47.
6. Kim C, Paik S. Gene-expression-based prognostic assays for breast cancer. Nat Rev Clin Oncol. 2010;7:340–7. 7. Badve S, Nakshatri H. Oestrogen-receptor-positive breast cancer: towards bridging histopathological and molecular classifications. J Clin Pathol. 2009;62:6–12. 8. Lim KC, Lakshmanan G, Crawford SE, Gu Y, Grosveld F, Engel JD. GATA3 loss leads to embryonic lethality due to noradrenaline deficiency of the sympathetic nervous system. Nat Genet. 2000;25:209–12. 9. Kaufman CK, Zhou P, Pasolli HA, et al. GATA-3: an unexpected regulator of cell lineage determination in skin. Genes Dev. 2003;17:2108–22. 10. Ting CN, Olson MC, Barton KP, Leiden JM. Transcription factor GATA-3 is required for development of the T-cell lineage. Nature. 1996;384:474–8. 11. Kouros-Mehr H, Slorach EM, Sternlicht MD, Werb Z. GATA-3 maintains the differentiation of the luminal cell fate in the mammary gland. Cell. 2006;127:1041–55. 12. Asselin-Labat ML, Sutherland KD, Barker H, et al. GATA-3 is an essential regulator of mammary-gland morphogenesis and luminal-cell differentiation. Nat Cell Biol. 2007;9:201–9. 13. Mehra R, Varambally S, Ding L, et al. Identification of GATA3 as a breast cancer prognostic marker by global gene expression meta-analysis. Cancer Res. 2005;65:11259–64. 14. Carlsson P, Mahlapuu M. Forkhead transcription factors: key players in development and metabolism. Dev Biol. 2002;250: 1–23. 15. Kaestner KH. The hepatocyte nuclear factor 3 (HNF3 or FOXA) family in metabolism. Trends Endocrinol Metab. 2000;11:281–5. 16. Carroll JS, Brown M. Estrogen receptor target gene: an evolving concept. Mol Endocrinol. 2006;20:1707–14. 17. Laganiere J, Deblois G, Lefebvre C, Bataille AR, Robert F, Giguere V. From the Cover: location analysis of estrogen receptor alpha target promoters reveals that FOXA1 defines a domain of the estrogen response. Proc Natl Acad Sci USA. 2005;102:11651–6. 18. Hurtado A, Holmes KA, Ross-Innes CS, Schmidt D, Carroll JS. FOXA1 is a key determinant of estrogen receptor function and endocrine response. Nat Genet. 2011;43:27–33. 19. Badve S, Turbin D, Thorat MA, et al. FOXA1 expression in breast cancer–correlation with luminal subtype A and survival. Clin Cancer Res. 2007;13:4415–21. 20. Ijichi N, Shigekawa T, Ikeda K, et al. Association of doublepositive FOXA1 and FOXP1 immunoreactivities with favorable prognosis of tamoxifen-treated breast cancer patients. Hormon Cancer. 2012;3:147–59. 21. Mehta RJ, Jain RK, Leung S, et al. FOXA1 is an independent prognostic marker for ER-positive breast cancer. Breast Cancer Res Treat. 2012;131:881–90. 22. Hisamatsu Y, Tokunaga E, Yamashita N, et al. Impact of FOXA1 expression on the prognosis of patients with hormone receptorpositive breast cancer. Ann Surg Oncol. 2012;19:1145–52. 23. Hoch RV, Thompson DA, Baker RJ, Weigel RJ. GATA-3 is expressed in association with estrogen receptor in breast cancer. Int J Cancer. 1999;84:122–8. 24. Parikh P, Palazzo JP, Rose LJ, Daskalakis C, Weigel RJ. GATA3 expression as a predictor of hormone response in breast cancer. J Am Coll Surg. 2005;200:705–10. 25. Ademuyiwa FO, Thorat MA, Jain RK, Nakshatri H, Badve S. Expression of forkhead-box protein A1, a marker of luminal A type breast cancer, parallels low Oncotype DX 21-gene recurrence scores. Mod Pathol. 2010;23:270–5. 26. Bernardo GM, Lozada KL, Miedler JD, et al. FOXA1 is an essential determinant of ERa expression and mammary ductal morphogenesis. Development. 2010;137:2045–54.
123
