Int J Clin Oncol DOI 10.1007/s10147-013-0639-1

ORIGINAL ARTICLE

Subcellular localization of Mdm2 expression and prognosis of breast cancer Hyung Seok Park • Ji Min Park • Seho Park • Junghoon Cho • Seung Il Kim • Byeong-Woo Park

Received: 17 July 2013 / Accepted: 1 November 2013 Ó Japan Society of Clinical Oncology 2013

Abstract Background Mouse double minute 2 (Mdm2) is a negative regulator of the tumor suppressor p53. The p53–Mdm2 pathway may play a role in cancer development and prognosis, although the role of p53–Mdm2 in breast cancer remains unclear. Methods p53 and Mdm2 expressions were determined by immunohistochemistry of tissue microarrays of 865 breast cancer patients who underwent surgery. Clinicopathological characteristics and survival data were analyzed. Mdm2 expression was categorized into four groups: negative, cytoplasm positive, nucleus positive, and concurrent nuclear and cytoplasm positive (N?&C?). Results Negative, cytoplasm-positive, nucleus-positive, and N?&C? expressions of Mdm2 were observed in 59.2, 10.9, 27.8, and 2.1 % of patients, respectively. The N?&C? group was associated with larger tumor size, higher grade, negativity for estrogen and progesterone receptors, HER2 positivity, high Ki-67 index, p53 positivity, and triple negative breast cancer. p53-positive tumors showed poorer overall survival than p53-negative tumors. The nucleus-positive and N?&C? groups showed poorer disease-free survival than the negative and

H. S. Park and J. M. Park contributed equally to this work. H. S. Park  J. M. Park  S. Park  J. Cho  S. I. Kim Department of Surgery, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea B.-W. Park (&) Department of Surgery and Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea e-mail: [email protected]

cytoplasm-positive groups. In multivariate analysis, nuclear Mdm2 expression including the N?&C? group was significantly related to poor prognosis. Conclusions Concurrent nuclear and cytoplasmic Mdm2 expression was an independent prognostic factor in patients with breast cancer. Subcellular localization of Mdm2 expression should be considered in the evaluation of Mdm2 in breast cancer. Keywords Breast  Neoplasms  Proto-oncogene proteins c-mdm2  p53  Prognosis

Introduction Breast cancer results from complex interactions of molecular pathways including the down-regulation of tumor suppressor p53 that stabilizes genetic activity and induces apoptosis in tumor cells [1, 2]. Inactivation of p53 is related to oncogenesis, cancer progression, and chemotherapy resistance in breast cancer [3, 4]. Therefore, restoring or stabilizing tumor suppressor activity, which can be achieved by inhibiting the down-regulation of p53, could play a role in treatment strategies for cancer. Mouse double minute 2 (Mdm2) is an E3 ubiquitin ligase that binds p53 [5]. Mdm2 inhibits p53 transcriptional activity [6] and promotes the degradation of p53 protein via ubiquitination [7–9]. Mdm2 can induce carcinogenesis independent of p53 [10]. Moreover, in-vitro studies have suggested that Mdm2 plays a role in the proliferation of estrogen receptor (ER)-positive breast cancer cells [11–13]. Mdm2 protein expression was observed by immunohistochemistry (IHC) in approximately 25–70 % of breast cancers [14–16]. Mdm2 is primarily found in the nucleus

123

Int J Clin Oncol

where it is bound to p53 [14, 17]; cytoplasmic Mdm2 expression has also been observed [18, 19]. There are variants of Mdm2 that are consistently found with wildtype Mdm2 [20–22], and such variants may modulate wildtype Mdm2 function in cell growth and carcinogenesis independently of p53 [10, 22]. The prognostic value of Mdm2 expression is inconsistent. Mdm2 expression in tumors was found to be associated with aggressive clinicopathological features in previous studies [15, 17, 19, 21]. However, other studies have suggested that Mdm2 expression is related to favorable prognostic factors [14, 18, 23]. One study has suggested that patients with Mdm2 expression have better survival rates than those without Mdm2 expression [19], whereas others have reported poor survival with Mdm2 expression [15, 21, 24]. It is difficult to draw conclusions regarding the relationship between Mdm2 expression and prognosis due to the small sample sizes and heterogeneous designs of most studies. The definition of Mdm2 positivity differed betwen studies and there is no consensus for defining a threshold of positivity of Mdm2 by IHC. Some studies considered any subcellular expression, either nuclear or cytoplasmic, as positive for Mdm2 expression [20], whereas others defined only nuclear staining of Mdm2 as positive [14, 16, 17]. These various definitions of Mdm2 positivity preclude drawing any conclusions based on the presence and subcellular localization of Mdm2 expression. Nevertheless, differences in the subcellular localization of Mdm2 expression among tumors may reflect different molecular behaviors in each tumor and should not be overlooked. The purpose of the present study was to investigate clinicopathological features and prognosis according to p53 or the subcellular localization of Mdm2 expression and to evaluate the significance of p53 and Mdm2 expression in patients with breast cancer.

Materials and methods Patients Formalin-fixed, paraffin-embedded tumor samples were collected between November 1999 and August 2005. Archival hematoxylin-and-eosin-stained slides were reviewed by breast pathologists for each case. IHC staining was interpreted in a blind fashion, without any information regarding clinical parameters or outcomes. Among the 1,200 cases with available tissue samples, we excluded cases with distant metastasis at the time of diagnosis (n = 12); duplicated cases (n = 16); cases with unknown Mdm2 status (n = 145); and cases with ductal carcinoma in situ (n = 39), microinvasive cancer (n = 30), or extensive intraductal components that did not have

123

invasive foci in the sampled specimens (n = 102). As a result, 856 patients with primary invasive breast cancer who underwent definitive surgery in Yonsei University Severance Hospital were enrolled for our final analysis. Patients’ information was retrospectively obtained from medical records. After surgery, local radiotherapy or adjuvant treatments were administered if indicated. Clinical follow-up included medical history, physical examinations, laboratory tests, and radiological imaging every 6–12 months for detection of relapse. Tumor stage was based on criteria of the Sixth American Joint Committee on Cancer. Histological grade was assessed by the modified Bloom–Richardson Classification [25]. Subtypes were classified into four groups based on the 2011 St. Gallen Conference consensus [26]. Disease-free survival (DFS) was measured from the date of the first curative surgery to the date of the first locoregional or systemic recurrence, or death without any type of relapse. Overall survival (OS) was measured from the date of the operation to the date of the last follow-up or death from any cause. The Institutional Review Board of Yonsei University Severance Hospital approved this study. Tissue microarray Tissue microarray (TMA) blocks were constructed using formalin-fixed, paraffin-embedded tumor samples as detailed in a previous study [27]. IHC staining was carried out using TMA. In brief, 5-lm-thick sections were deparaffinized and rehydrated. After treatment with 3 % hydrogen peroxide solution for 10 min to block endogenous peroxidase, sections were pretreated in 10 mM citrate buffer in a microwave oven for 20 min for antigen retrieval. After incubation with primary antibodies against p53 (D0-7, 1:100; DAKO), Mdm2 (polyclonal, 1:100; Abcam), ER (SP1, 1:100; Thermo Scientific), PR (PgR 636, 1:50; DAKO, Glostrup, Denmark), human epidermal growth factor receptor 2 (HER2) (polyclonal, 1:1,500; DAKO), and Ki-67 (MIB-1, 1:100; DAKO), immunodetection was performed with biotinylated anti-mouse/rabbit immunoglobulin followed by peroxidase-labeled streptavidin using a labeled streptavidin biotin kit with 3,30 -diaminobenzidine chromogen as the substrate. The slides were counterstained with Harris hematoxylin. Mdm2 expression was categorized into four groups according to subcellular localization in tumor cells: negative, no Mdm2 expression in cells; cytoplasm-positive, isolated cytoplasmic expression of Mdm2; nucleus-positive, isolated nuclear expression of Mdm2; and N?&C?, concurrent nuclear and cytoplasmic expression of Mdm2 (Fig. 1). Tumors in which C10 % of cells exhibited nuclear and/ or cytoplasmic staining were defined as positive for Mdm2

Int J Clin Oncol Fig. 1 Examples of immunohistochemical staining of Mdm2 in invasive ductal carcinoma of the breast. a negative, b cytoplasm positive, c nucleus positive, d nucleus and cytoplasm positive

nuclear and/or cytoplasmic expression. Tumors with C10 % of cells presenting nuclear staining were defined as positive for p53. Tumors with C1 % of cells showing nuclear staining were considered positive for ER and PR [28]. An arbitrary cut-off score of 10 % was used for Ki-67 expression. HER2 status was evaluated using the HercepTestTM (DAKO) and was interpreted as 0, 1?, 2?, or 3? [28]. HER2 was considered positive in cases with an IHC score of 3? or gene amplification by fluorescence in-situ hybridization (FISH) regardless of the HER2 IHC result. FISH analysis for HER2 was analyzed using the method described in a previous study [27]. Statistical analyses Differences between the groups were evaluated by chisquared or Fisher’s exact test. Continuous variables were compared using one-way analysis of variance with a Bonferroni multiple comparison test. Survival curves were plotted using the Kaplan–Meier method and group differences in survival time were investigated by a log-rank test. Cox proportional hazards models were used to identify variables that were independently associated with survival. The assumption of the models was tested by log-minus-log survival function, which suggested that the proportional hazards assumption was acceptable. All models were adjusted for factors that were significantly different in univariate analysis and two different models were tested according to Mdm2 categorization. All statistical tests were two-sided and a P value \0.05 was considered statistically

significant. PASW for Windows version 20.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analysis.

Results The mean age of the total study population (n = 856) at diagnosis was 48.9 years (SD = 10.4) The median followup time was 86.0 months. Analysis of the localization patterns of Mdm2 showed that 59.2 % of patients were Mdm2 negative, 10.9 % were cytoplasm positive, 27.8 % were nucleus positive, and 2.1 % were C?&N?. The mean age of the negative group was 49.8 years, whereas that of the cytoplasm-positive, nucleus-positive, and N?&C? groups were 47.2, 47.7, and 48.6 years, respectively. A half of the N?&C? group had stage III disease, although less than 30 % of the other groups had stage III disease. Compared with the negative group, the cytoplasm-positive, nucleus-positive, and N?&C? groups had higher grades and more hormone receptor-negative tumors. The proportion of high-grade tumors in the cytoplasm-positive group was higher than that in the nucleuspositive group, but lower than in the N?&C? group. HER2 overexpression was observed in 38.9 % of N?&C?, 35.3 % of nucleus-positive, 29.0 % of cytoplasm-positive, and 18.7 % of Mdm2-negative cases, and this difference was significant. The cytoplasm-positive and N?&C? groups contained a high percentage (79.6 and 77.8 %) of highly proliferative Ki-67-positive cancers, whereas 46.4 % of the nucleus-positive group and 36.4 %

123

Int J Clin Oncol Table 1 Patient clinicopathological features according to Mdm2 expression patterns in tumor cells Negative (n = 507)

Cytoplasm positive (n = 93)

Nucleus positive (n = 238)

Cytoplasm and nucleus positive (n = 18)

49.87 ± 10.6

47.28 ± 9.27

47.79 ± 10.49

48.06 ± 9.34

0.02

T1

231 (45.6)

33 (35.5)

114 (47.9)

5 (27.8)

0.09

CT2

276 (54.4)

60 (64.5)

124 (52.1)

13 (72.2)

255 (50.3)

47 (50.5)

124 (52.1)

7 (38.9)

252 (49.7)

46 (49.5)

114 (47.9)

11 (61.1)

I

156 (30.8)

22 (23.7)

76 (31.9)

2 (11.1)

II

248 (48.9)

54 (58.1)

109 (45.8)

7 (38.9)

III

103 (20.3)

17 (18.3)

53 (22.3)

9 (50.0)

Age, years (mean ± SD)

P

T stage

Nodal status Negative Positive Stage

0.74

0.03

Histological grade I/II

404 (79.7)

42 (45.2)

162 (68.1)

7 (38.9)

III

103 (20.3)

51 (54.8)

76 (31.9)

11 (61.1)

\0.001

ER Negative (\1 %)

109 (21.5)

50 (53.8)

75 (31.5)

11 (61.1)

Positive (C1 %)

398 (78.5)

43 (46.2)

163 (68.5)

7 (38.9)

\0.001

PR \0.001

Negative (\1 %)

159 (31.4)

55 (59.1)

103 (43.3)

11 (61.1)

Positive (C1 %)

348 (68.6)

38 (40.9)

135 (56.7)

7 (38.9)

412 (81.3) 95 (18.7)

66 (71.0) 27 (29.0)

154 (64.7) 84 (35.3)

11 (61.1) 7 (38.9)

\0.001

\0.001

HER2 Negative Positive Ki-67 (n = 854) Low (\10 %)

322 (63.6)

19 (20.4)

127 (53.6)

4 (22.2)

High (C10 %)

184 (36.4)

74 (79.6)

110 (46.4)

14 (77.8)

p53 (n = 832) Negative

380 (76.9)

51 (56.0)

136 (59.4)

6 (33.3)

Positive

114 (23.1)

40 (44.0)

93 (40.6)

12 (66.7)

\0.001

Subtypes (n = 855) Luminal A

321 (63.4)

19 (20.4)

102 (42.9)

3 (16.7)

Luminal B

86 (17.0)

28 (30.1)

66 (27.7)

5 (27.8)

HER2

36 (7.1)

11 (11.8)

27 (11.3)

2 (11.1)

TNBC

63 (12.5)

35 (37.6)

43 (18.1)

8 (44.4)

BCS

141 (27.8)

31 (33.3)

65 (27.3)

4 (22.2)

Mastectomy

366 (72.2)

62 (66.7)

173 (72.7)

14 (77.8)

Operation method

\0.001

0.64

Chemotherapy

0.06

No

77 (15.2)

7 (7.5)

29 (12.2)

0 (0)

Yes

430 (84.8)

86 (92.5)

209 (87.8)

18 (100)

No

144 (28.4)

52 (55.9)

93 (39.1)

11 (61.1)

Yes

363 (71.6)

41 (44.1)

145 (60.9)

7 (38.9)

No

269 (53.1)

44 (47.3)

131 (55.0)

7 (38.9)

Yes

238 (46.9)

49 (52.7)

107 (45.0)

11 (61.1)

\0.001

Hormone therapy

Radiation therapy

0.39

Mdm2 mouse double minute 2, ER estrogen receptor, PR progesterone receptor, HER2 human epidermal growth factor receptor 2, TNBC triple negative breast cancer, BCS breast-conserving surgery

123

Int J Clin Oncol

Fig. 2 Survival curves according to p53 status

of the negative group showed a relatively low incidence of Ki-67-positive cancers. p53 status in IHC was available in 832 patients. Of them, 593 (66.9 %) patients were negative for p53, and 259 patients (31.1 %) were positive for p53. In the analysis regarding Mdm2 expression, patients with cytoplasmic and/or nuclear Mdm2 expression showed more p53 positivity than those with negative Mdm2 expression (p53 positivity: 23.1 % in the negative, 44.0 % in the cytoplasm-positive, 40.6 % in the nucleus-positive, and 66.7 % in the N?&C? group, respectively) (Table 1). Tumor size and nodal status were not significantly different among the four groups. There was a significant difference in subtypes according to Mdm2 expression (P \ 0.001): Mdm2-negative cancers were characterized by Luminal A; cytoplasm-positive and N?&C? cancers were typical of triple-negative breast cancer (TNBC); and the nucleuspositive group was intermediate. Overall, the cytoplasmpositive and N?&C? groups showed more aggressive features than the nucleus-positive and negative groups. Most patients (86.8 %) were treated with systemic chemotherapy, approximately one third of patients (28.2 %) underwent breast conservation surgery (BCS), and almost half of the patients (47.3 %) received radiation therapy. Two thirds of patients (65.0 %) were treated with endocrine therapy. Endocrine therapy was performed in a significantly higher proportion of Mdm2-negative patients (P \ 0.001). There was a trend towards more chemotherapy for cytoplasmic or nuclear Mdm2-positive patients (P = 0.06), probably because of the underlying tumor characteristics. There was no significant difference in DFS between patients with p53-negative and those with p53-positive tumors, while patients with p53-negative tumors showed significantly better OS than those with p53-positive tumors (P = 0.04) (Fig. 2).

The 5-year DFS of patients in the negative, cytoplasmpositive, nucleus-positive, and N?&C? groups were 85.8, 87.1, 78.6, and 59.5 %, respectively (P = 0.005). The 5-year OS according to Mdm2 expression patterns was 92.1, 90.3, 85.7, and 65.5 %, respectively (P = 0.01). The DFS of the cytoplasm-positive group was comparable to that of the negative group and better than that of the nucleus-positive and N?&C? groups. The OS of the N?&C? group was the poorest. OS was not significantly different between cytoplasm-positive and nucleus-positive groups (Fig. 3). When a simplified categorization was used for the survival analysis, nuclear with/without cytoplasmic expression vs. isolated cytoplasmic expression or negative expression, nuclear expression with/without cytoplasmic expression of Mdm2 was significantly associated with poor survival (Fig. 4). Table 2 shows the results of the multivariate analyses. Age, ER, PR, HER2, Ki-67, grade, p53, and hormone therapy were all significantly different between groups (Table 1); thus we used these variables as covariates in the multivariate analyses. Even though the status of chemotherapy was marginally different between the groups (p = 0.06, Table 1), it did not reach the threshold of statistical significance defined under Methods. Thus, we did not include the status of chemotherapy as a covariate in the multivariate analyses. Concurrent nuclear and cytoplasmic expression of Mdm2 was an independent prognostic factor for DFS and OS (Model-1 in Table 2). The nucleus-positive group showed a trend to poor DFS, but this was not significant for OS (Model-1 in Table 2). Nuclear expression with/without cytoplasmic expression was significantly associated with poor survival (Model-2 in Table 2). However, cytoplasmic expression and p53 status in breast tumor cells was not significantly associated with survival in the multivariate

123

Int J Clin Oncol

Fig. 3 Survival curves according to subcellular localization of Mdm2 expression (four-group analysis)

analyses (Table 2). The multivariate analyses also revealed that PR and HER2 were statistically independent prognostic factors (all p \ 0.05 in Model-1 and 2) (Table 2). However, age, ER, Ki-67, grade, p53 expression, and status of hormone therapy were not statistically independent prognostic factors in the multivariate analyses (Table 2).

Discussion This study clearly demonstrates differences in tumor characteristics including grade, hormone receptor, HER2 overexpression, Ki-67, and subtypes according to subcellular localization of Mdm2 expression. Additionally,

123

patients with N?&C? and nucleus-positive expression of Mdm2 experienced poorer survival than those with negative or cytoplasmic expression of Mdm2. Concurrent nuclear and cytoplasmic expression of Mdm2 was a statistically independent prognostic factor of breast cancer in multivariate analysis. Interestingly, cytoplasm-positive breast cancers showed fewer favorable tumor biological features than nucleus-positive cancers (Table 1), but OS of these patients was comparable to that of Mdm2-negative and nucleus-positive cancers and their DFS was better than that of nucleus-positive and N?&C? cancers. This suggests that cytoplasmic translocation of Mdm2 may function to inhibit tumor progression. However, the mechanism underlying the paradoxical situation of aggressive clinicopathological features but favorable outcomes in cytoplasmic Mdm2-positive cancers is unclear. The fact that 92 % of patients in the cytoplasm-positive group received chemotherapy could be a factor influencing the favorable outcome of this group (Table 1). Furthermore, although N?&C? tumors include cytoplasmic Mdm2 localization in the tumor cells, the prognoses of these tumors were the poorest. This phenomenon could be explained by a previous study that reported different levels of Mdm2 showing different regulatory actions for p53 activity: low Mdm2 levels in the nucleus leads to a release of Mdm2 through p53 sumoylation, which may be related to modification of p53 activity, while high Mdm2 levels in the nucleus cause degradation of p53 [29]. The N?&C? group reflects a high level of Mdm2 in tumor cells; the prognosis of the N?&C? group may be related to poor survival caused by p53 degradation with a high level of Mdm2. However, further investigation is necessary to clearly predict these mechanisms. The previous positive association between Mdm2 expression and high-grade tumors in breast cancer partly supports our results [30]. However, some authors have reported that Mdm2 expression correlates with favorable prognostic parameters [18]. The limitations of most studies include: small sample size, not considering the Mdm2expression patterns, using different detection methods, and/ or lacking a survival analysis. These are possible explanations for the contradictory results. We have summarized our results and those of previous studies that have investigated the correlation between Mdm2 expression or gene amplification and prognostic factors (Table 3). Because of differences in design, it is difficult to directly compare our study with other studies. Our Mdm2 positivity rate of 40.8 % is consistent with previous studies showing approximately 40–70 % positivity (Table 3). Most studies showed the absence or low expression of Mdm2 in normal breast cells [17, 23], indicating that Mdm2 expression in tumor cells may be a surrogate for instability of the tumor suppressor pathway. In our study, Mdm2 expression was

Int J Clin Oncol

Fig. 4 Survival curves according to subcellular localization of Mdm2 expression (two-group analysis)

Table 2 Multivariate analysis according to Mdm2 expression patterns DFS P

OS HR

95.0 % CI for HR Lower

P

HR

Upper

95.0 % CI for HR Lower

Upper

MODEL-1 (Mdm2 four groups) Age

0.25

1.00

0.99

1.02

0.007

1.02

1.00

1.04

ER (- vs. ?) PR (- vs. ?)

0.99 0.007

1.00 0.56

0.57 0.37

1.76 0.85

0.93 0.01

0.97 0.55

0.51 0.33

1.84 0.90

HER2 (- vs. ?)

0.03

1.44

1.02

2.03

0.44

1.17

0.78

1.75

Ki-67 (low vs. high)

0.50

0.88

0.61

1.26

0.37

0.82

0.53

1.26

HG (I/II vs. III)

0.30

0.81

0.56

1.19

0.23

0.76

0.49

1.18

p53 (negative vs. positive)

0.49

0.88

0.62

1.26

0.74

1.06

0.71

1.60

HTx (no vs. yes)

0.92

1.02

0.61

1.70

0.48

0.81

0.45

1.45

0.81

0.45

1.46

0.73

0.88

0.45

1.73

Mdm2 expression pattern Negative

Ref

Cytoplasm-positive

0.49

Ref

Nucleus-positive

0.07

1.36

0.97

1.92

0.13

1.36

0.91

2.05

N?&C?

0.008

2.95

1.31

6.61

0.001

4.24

1.84

9.76

MODEL-2 (Mdm2 two groups) Age

0.25

1.00

0.99

1.02

0.008

1.02

1.00

1.04

ER (- vs. ?)

0.95

1.01

0.58

1.78

0.99

0.99

0.52

1.88

PR (- vs. ?) HER2 (- vs. ?)

0.008 0.04

0.57 1.41

0.37 1.00

0.86 1.99

0.01 0.58

0.55 1.11

0.33 0.74

0.90 1.67

Ki-67 (low vs. high)

0.52

0.89

0.62

1.27

0.46

0.85

0.56

1.29

HG (I/II vs. III)

0.32

0.83

0.57

1.20

0.27

0.78

0.51

1.21

p53 (negative vs. positive)

0.55

0.89

0.63

1.27

0.60

1.11

0.74

1.65

HTx (no vs. yes)

0.96

1.01

0.60

1.68

0.45

0.80

0.44

1.42

Mdm2 expression pattern (N- vs. N?)

0.01

1.49

1.08

2.06

0.02

1.54

1.06

2.24

Mdm2 mouse double minute 2, DFS disease-free survival, OS overall survival, HR hazard ratio, CI confidential interval, ER estrogen receptor, PR progesterone receptor, HER2 human epidermal growth factor receptor 2, HG histologic grade, HTx hormone therapy, Ref reference value, N?&C? concurrent nuclear and cytoplasmic positive, N- nuclear negative, N? nuclear positive

123

Int J Clin Oncol Table 3 Summary of correlations between Mdm2 expression or gene amplification and prognosis of breast cancer in the literature Mdm2 expression or gene amplification in tumor

Featuresa

Survival

28

IHC N? 68 % (15/28)

:ER positivity

N/A

1995

66

IHC? 40.9 % (27/66)

:ER positivity

N/A

Bueso-Ramos et al.

1996

33

RT-PCR 73 % (24/33)

:Nuclear grade

N/A

Western-blot 71 % (15/21)

:Metastasis no relation to HR

Lukas et al.

2001

Authors

Year

Marchetti et al.

1995

Baunoch et al.

N

IHC N? 48 % (10/21) 38

RT-PCR wild-type 97 % (37/38)

:p53 positivity

Poor OS with variantsb

:PR negativity

Variants 34 % (13/38) Hori et al.

2002

104

IHC N?/C? 49 % (51/104)

;Grade

RT-PCR variants (-) 51.9 % (27/66)

;LN metastasis

N/A

:ER positivity

RT-PCR variants (?) 48.1 % (40/104) Al-Kuraya et al.

2004

1847

Vestey et al.

2004

103

FISH amplification 5.7 % of all cases

Not associated with survival

IHC C? 54 % (49/91)

:Ki-67

N? 47 % (43/92)

:HER2 (only DCIS)

Better OS and DFS with cytoplasmic positivec

990

IHC N? 26 % (253/990)

180

IHC N? 37 % (66/180)

Different features according to coexpression of p53, Mdm4, bcl2, and p21

Different according to co-expression of p53, Mdm4, bcl2, and p21

Abdel-Fatah et al.

2010

Araki et al.

2010

45

IHC? 65 % of the samples

75 % of positive cases presented LN or distant metastasis

N/A

Choschzick et al.

2010

572

FISH amplification 7 % (40/ 572)

:Tumor size

Poor OS with Mdm2 amplification in ER? breast cancerc

KarrayChouaykh et al.

2011

42

IHC? 38 % (16/42)

:Tumor size

Poor OS with Mdm2-positived

Park et al.

Present study

856

IHC N?/C? 40.8 % (349/856)

;Age

Poor OS and DFS with nuclear positivec

C? 10.9 % (93/856)

:HR negativity

N? 29.9 % (256/856)

:HER2

:Grade

:Ki-67 IHC immunohistochemistry, N? nucleus positive, ER estrogen receptor, N/A not available, RT-PCR reverse transcription polymerase chain reaction of Mdm2 mRNA, HR hormone receptor, PR progesterone receptor, N?/C? nuclear and/or cytoplasmic expression of Mdm2, LN lymph node, C? cytoplasm positive, HER2 human epidermal growth factor receptor 2, DCIS ductal carcinoma in situ, DFS disease-free survival, OS overall survival, FISH fluorescence in situ hybridization a

Clinicopathological features associated with Mdm2 expression or gene amplification

b

Statistically significant in univariate analysis

c

Statistically significant in both univariate and multivariate analyses

d

Statistically significant in multivariate analysis

associated with more aggressive features which might reflect inhibitory activity of the p53 by Mdm2 in tumor cells. The 31 % incidence of p53 nuclear staining in breast cancer cells was similar to that of the previous study [19]. However, it is not consistent with the finding that p53 positivity was more prevalent in tumors with Mdm2 expression because Mdm2 down-regulates p53. This might be due to non-functional p53 variants, Mdm2 function

123

independent of p53, or functional modification by sumoylation [10, 31, 32]. This study included a survival analysis using two different categorizations of Mdm2 expression, dichotomization and quadrichotomization, and all categorizations had significant statistical power. Interestingly, although p53 status was only significant for OS in univariate analysis and the significance disappeared in the multivariate analysis,

Int J Clin Oncol

concurrent nuclear and cytoplasmic expression of Mdm2 remained significant in terms of DFS and OS in both the univariate and multivariate analyses. In particular, patients with nuclear expression of Mdm2 showed poorer survival than those without nuclear expression irrespective of cytoplasmic expression of Mdm2 (Model 2 in Table 2). Therefore, the nuclear expression of Mdm2 including the N?&C? group should be considered as a standard criterion of categorization. Nuclear Mdm2 expression with/ without cytoplasmic expression could be a better discriminatory prognostic factor in patients with breast cancer than p53 status. Questions remain regarding the mechanism of the different subcellular localizations and the reasons for their prognostic differences. One possible explanation is that nuclear Mdm2 expression may reflect enhanced degradation of p53 as a result of nuclear-to-cytoplasmic shuttling of the p53–Mdm2 complex and the resultant instability of the tumor suppressor [33]. In contrast, cytoplasmic Mdm2 expression may represent a non-functional or partially functional p53–Mdm2 pathway, which could result in enhanced stability of the tumor suppressor. Interaction of the p53–Mdm2 complex with the small ubiquitin-related modifier could be another explanation of the mechanism of the subcellular localization of Mdm2 protein. Sumoylation modifies protein functions and is known to be a regulator of Mdm2–p53 degradation [32]. Sumoylation mediates nuclear localization of some proteins: HIPK1 and TEL [34]. Rapid shuttling of Mdm2 between the nucleus and cytoplasm may be crucial for efficient degradation of p53, which could be modulated by sumoylation [35]. Mdm2 variants that function independently of p53 might be another explanation for isolated cytoplasmic expression of Mdm2, and have no prognostic value for p53 status. Different mechanical action at different nuclear Mdm2 levels could be a candidate to explain why nuclear Mdm2 expression including the N?&C? group indicates a poor prognosis. Identifying the mechanism of the subcellular localization of Mdm2 expression is a subject for further investigation. This study has some limitations. Firstly, its retrospective nature might cause a selection bias. Secondly, even though a previous study demonstrated that mRNA level correlates with protein expression of Mdm2 [17], we have not yet shown that mRNA levels are associated with subcellular localization of Mdm2 expression in tumor cells. Nevertheless, the relatively large cohorts that we used to conduct the analyses is a strength of our study, through which multivariate analyses adjusted for significant covariates in terms of clinicopathological features suggested that Mdm2 expression may be a statistically independent prognostic factor. In addition to the prognostic value of subcellular localization of Mdm2 expression, inhibition of Mdm2 could have potential as a novel treatment for cancer by

restoring p53 activity and inducing apoptosis of tumor cells [36, 37]. Further studies should be carried out to verify the mechanisms and roles of Mdm2 expression with respect to localization patterns in tumor cells in conjunction with its possible role as a therapeutic molecular target.

Conclusions Our study indicates that Mdm2 expression is a common event in breast cancer. Nuclear or cytoplasmic expression of Mdm2 was associated with aggressive tumor behavior and poor survival compared with negative Mdm2 expression. Concurrent nuclear and cytoplasmic expression or nuclear expression with/without cytoplasmic expression of Mdm2 were a statistically independent prognostic factor in patients with breast cancer regardless of p53 status, whereas cytoplasmic Mdm2 expression showed comparable prognosis with negative Mdm2 expression. Therefore, the subcellular localization of Mdm2 expression should be considered in the clinical evaluation of breast cancer. Acknowledgments A major part of this study was presented at the 35th Annual San Antonio Breast Cancer Symposium poster session, 4–8 December 2012 in San Antonio, TX. This research has been supported by the Korea Breast Cancer Foundation, and was supported by the Brain Korea 21 Project for Medical Science at Yonsei University. Conflict of interest of interest.

The authors declare that they have no conflict

References 1. Toledo F, Wahl GM (2006) Regulating the p53 pathway: in vitro hypotheses, in vivo veritas. Nat Rev Cancer 6:909–923 2. Vousden KH, Prives C (2009) Blinded by the light: The growing complexity of p53. Cell 137:413–431 3. O’Connor PM, Jackman J, Bae I et al (1997) Characterization of the p53 tumor suppressor pathway in cell lines of the National Cancer Institute anticancer drug screen and correlations with the growth-inhibitory potency of 123 anticancer agents. Cancer Res 57:4285–4300 4. Scata KA, El-Deiry WS (2007) p53, BRCA1 and breast cancer chemoresistance. Adv Exp Med Biol 608:70–86 5. Fang S, Jensen JP, Ludwig RL et al (2000) Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53. J Biol Chem 275:8945–8951 6. Oliner JD, Kinzler KW, Meltzer PS et al (1992) Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 358:80–83 7. Haupt Y, Maya R, Kazaz A et al (1997) Mdm2 promotes the rapid degradation of p53. Nature 387:296–299 8. Kubbutat MH, Jones SN, Vousden KH (1997) Regulation of p53 stability by Mdm2. Nature 387:299–303 9. Fakharzadeh SS, Trusko SP, George DL (1991) Tumorigenic potential associated with enhanced expression of a gene that is amplified in a mouse tumor cell line. EMBO J 10:1565–1569

123

Int J Clin Oncol 10. Steinman HA, Burstein E, Lengner C et al (2004) An alternative splice form of Mdm2 induces p53-independent cell growth and tumorigenesis. J Biol Chem 279:4877–4886 11. Brekman A, Singh KE, Polotskaia A et al (2011) A p53-independent role of Mdm2 in estrogen-mediated activation of breast cancer cell proliferation. Breast Cancer Res 13:R3 12. Sanchez M, Picard N, Sauve K et al (2013) Coordinate regulation of estrogen receptor beta degradation by Mdm2 and CREBbinding protein in response to growth signals. Oncogene 32:117–126 13. Kim K, Burghardt R, Barhoumi R et al (2011) MDM2 regulates estrogen receptor alpha and estrogen responsiveness in breast cancer cells. J Mol Endocrinol 46:67–79 14. Marchetti A, Buttitta F, Girlando S et al (1995) mdm2 gene alterations and mdm2 protein expression in breast carcinomas. J Pathol 175:31–38 15. Karray-Chouayekh S, Baccouche S, Khabir A et al (2011) Prognostic significance of p16INK4a/p53 in Tunisian patients with breast carcinoma. Acta Histochem 113:508–513 16. Abdel-Fatah TM, Powe DG, Agboola J et al (2010) The biological, clinical and prognostic implications of p53 transcriptional pathways in breast cancers. J Pathol 220:419–434 17. Bueso-Ramos CE, Manshouri T, Haidar MA et al (1996) Abnormal expression of MDM-2 in breast carcinomas. Breast Cancer Res Treat 37:179–188 18. Hori M, Shimazaki J, Inagawa S et al (2002) Overexpression of MDM2 oncoprotein correlates with possession of estrogen receptor alpha and lack of MDM2 mRNA splice variants in human breast cancer. Breast Cancer Res Treat 71:77–83 19. Vestey SB, Sen C, Calder CJ et al (2004) p14ARF expression in invasive breast cancers and ductal carcinoma in situ–relationships to p53 and Hdm2. Breast Cancer Res 6:R571–R585 20. Hori M, Shimazaki J, Inagawa S et al (2000) Alternatively spliced MDM2 transcripts in human breast cancer in relation to tumor necrosis and lymph node involvement. Pathol Int 50:786–792 21. Lukas J, Gao DQ, Keshmeshian M et al (2001) Alternative and aberrant messenger RNA splicing of the mdm2 oncogene in invasive breast cancer. Cancer Res 61:3212–3219 22. Bartel F, Taubert H, Harris LC (2002) Alternative and aberrant splicing of MDM2 mRNA in human cancer. Cancer Cell 2:9–15 23. Baunoch D, Watkins L, Tewari A et al (1996) MDM2 overexpression in benign and malignant lesions of the human breast. Int J Oncol 8:895–899 24. Choschzick M, Heilenkotter U, Lebeau A et al (2010) MDM2 amplification is an independent prognostic feature of node-

123

25.

26.

27.

28.

29.

30.

31.

32.

33.

34. 35. 36.

37.

negative, estrogen receptor-positive early-stage breast cancer. Cancer Biomark 8:53–60 Elston CW, Ellis IO (1991) Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 19:403–410 Goldhirsch A, Wood WC, Coates AS et al (2011) Strategies for subtypes—dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 22:1736–1747 Park S, Park HS, Koo JS et al (2012) Breast cancers presenting luminal B subtype features show higher discordant human epidermal growth factor receptor 2 results between immunohistochemistry and fluorescence in situ hybridization. Cancer 118:914–923 Hammond ME, Hayes DF, Dowsett M et al (2010) American Society of Clinical Oncology/College Of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. J Clin Oncol 28:2784–2795 Carter S, Bischof O, Dejean A et al (2007) C-terminal modifications regulate MDM2 dissociation and nuclear export of p53. Nat Cell Biol 9:428–435 Al-Kuraya K, Schraml P, Torhorst J et al (2004) Prognostic relevance of gene amplifications and coamplifications in breast cancer. Cancer Res 64:8534–8540 Galmarini CM, Clarke ML, Falette N et al (2002) Expression of a non-functional p53 affects the sensitivity of cancer cells to gemcitabine. Int J Cancer 97:439–445 Nishida T, Yamada Y (2011) The nucleolar SUMO-specific protease SMT3IP1/SENP3 attenuates Mdm2-mediated p53 ubiquitination and degradation. Biochem Biophys Res Commun 406:285–291 O’Keefe K, Li H, Zhang Y (2003) Nucleocytoplasmic shuttling of p53 is essential for MDM2-mediated cytoplasmic degradation but not ubiquitination. Mol Cell Biol 23:6396–6405 Muller S, Hoege C, Pyrowolakis G et al (2001) SUMO, ubiquitin’s mysterious cousin. Nat Rev Mol Cell Biol 2:202–210 Melchior F, Hengst L (2000) Mdm2-SUMO1: is bigger better? Nat Cell Biol 2:E161–E163 Wang H, Ma X, Ren S et al (2011) A small-molecule inhibitor of MDMX activates p53 and induces apoptosis. Mol Cancer Ther 10:69–79 Maeda T, Hobbs RM, Merghoub T et al (2005) Role of the protooncogene Pokemon in cellular transformation and ARF repression. Nature 433:278–285

Subcellular localization of Mdm2 expression and prognosis of breast cancer.

Mouse double minute 2 (Mdm2) is a negative regulator of the tumor suppressor p53. The p53-Mdm2 pathway may play a role in cancer development and progn...
520KB Sizes 0 Downloads 0 Views