SEARCH

SEARCH BY CITATION

Abstract

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

Caveolin-1 (Cav-1) has been extensively characterized in cancer biological research. However, the role of Cav-1 in the interaction between tumor and stromal cells remains unclear. In the present study, we examined Cav-1 expression in tumor cells and stromal cells in breast cancer tissue by immunohistochemical analysis and evaluated its prognostic value in a training cohort. Immunohistochemical analysis of Cav-1 expression was scored as (++), (+) or (−) according to the proportion of positively stained tumor cells (T) and stromal cells (S). Correlation analysis between tumor/stromal Cav-1 expression and clinicopathological parameters revealed that only T(++) Cav-1 status was positively associated with tumor size and histological nodal status (= 0.019 and 0.021, respectively). Univariate analysis revealed that combined T(++)/S(−) status was significantly correlated with unfavorable prognostic outcomes (< 0.001). Multivariate analysis demonstrated that this combined status is an independent prognostic factor for primary breast cancer (= 0.002). Clinical outcomes in different subgroups of breast cancer patients were also strictly dependent on this combined status (< 0.05). The prognostic value of T(++)/S(−) Cav-1 status was also validated in the testing cohort. Collectively, our data indicate that high Cav-1 expression in tumor cells and lack of this expression in stromal cells could help identify a particular subgroup of breast cancer patients with potentially poor survival. Further studies are required to understand the regulatory mechanism of Cav-1 in the tumor microenvironment. (Cancer Sci 2011; 102: 1590–1596)

Breast cancer is the most common female cancer. Late-onset diagnosis, axillary lymph node metastases, tumor size, pathological type and resistance to antitumor therapy indicate a poor prognosis for breast cancer patients. Although treatment strategies for breast cancer have recently made great progress, recurrence and death rates remain unacceptably high.(1) Therefore, molecular biomarkers for recurrence and progression of breast cancer must be explored to help clinicians identify new diagnostic and therapeutic techniques to detect and treat breast cancer.(2)

Caveolins (Cav) are a family of scaffolding proteins that coat 50–100 nm plasma membrane invaginations. The Cav family is composed of three isoforms: Cav-1, Cav-2 and Cav-3. The Cav-1 gene is located on chromosome 7 (locus 7q31.1) and includes three exons (30, 165 and 342 bp) and two introns (1.5 and 32 kb).(3) Cav-1 expression depends on the type of tumor and its expression is downregulated in several human cancers such as sarcoma and lung cancer and might function as a tumor suppressor.(4,5) However, upregulation of Cav-1 expression has been reported in esophageal and pancreatic cancers and is also correlated with histopathological grade and poor prognosis.(6,7)

Cav-1 is mainly involved in vesicular transport, cholesterol homeostasis and signal transduction.(8) Furthermore, it might facilitate DNA repair and stabilize the insulin receptor against degradation. Cav-1 also plays a negative role in cell movement,(9) cellular senescence(10) and cell growth.(11) Endothelial cells from Cav-1−/− mice exhibit a diminished response to angiogenic growth factors.(12) Furthermore, Cav-1 overexpression is sufficient to induce premature cellular senescence in fibroblasts.(13,14) Cancer-associated fibroblasts (CAFs), which are derived from malignant or normal epithelial cells, promote tumor growth.(15)In vitro studies have shown that both stromal and epithelial Cav-1 play a protective role against mammary hyperplasia and tumorigenesis in breast cancer.(11,16,17) In addition, clinical studies have indicated that stromal loss of Cav-1 is a single independent predictor of early breast cancer recurrence and progression.(18,19) However, the value of combined tumor/stromal Cav-1 expression on the outcome of breast cancer patients is largely unknown.

In the present study, we investigated the clinical significance of Cav-1 expression (including tumor and stromal expression) in a training cohort and the correlation between tumor/stromal Cav-1 expression and clinicopathological characteristics. In addition, effects of combined tumor/stromal Cav-1 expression on outcomes in breast cancer patients were investigated. In addition, the prognostic value of combined tumor/stromal Cav-1 expression was also clarified in a testing cohort. Intriguingly, our results indicated that a counter balance of Cav-1 levels in the tumor microenvironment and epithelial compartment were the most strongly influenced clinical outcomes.

Materials and Methods

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

Collection of tissue samples.  Tissue specimens of the training cohort were collected from the Department of Breast Surgery, Kyoto University Hospital (Kyoto, Japan) between July 2000 and February 2006. Informed consent was obtained from all patients prior to specimen collection and all study protocols were approved by the Ethics Committee for Clinical Research, Kyoto University Hospital. The clinical stage was assessed by The Japanese Breast Cancer Society classification.(20) For analysis of survival and follow up, the date of surgery was used to represent the beginning of the follow-up period. All patients who died from diseases other than breast cancer or from unexpected events were excluded from the case collection. Follow ups were terminated in June 2010. The median follow up was 74 months (range, 3–119 months). Clinicopathological parameters of the training cohort are listed in Table 1. In addition, we validated the results using an independent testing cohort of 193 consecutive patients (Table 1) who underwent surgical resection of breast cancer at Osaka Red Cross Hospital (Osaka, Japan). The protocols used in the testing group were approved by the Ethics Committee of the Osaka Red Cross Hospital. Follow ups in the testing group were terminated in January 2011. The median follow up in the testing cohort was 42 months (range, 1–80 months).

Table 1.   Clinicopathological parameters of the training and testing cohort
ParameterVariablen (100%)
  1. †Patients with ductual carcinoma in situ (DCIS) were excluded. ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PgR, progesterone receptor.

Training cohort
 Age>5075 (72)
≤5029 (28)
 Tumor size≤2 cm24 (23)
2–5 cm51 (49)
>5 cm10 (10)
Unknown17 (18)
 Histological nodal statusPositive46 (44)
Negative56 (54)
Unknown2 (2)
 ERPositive80 (77)
Negative24 (23)
 PgRPositive67 (64)
Negative37 (36)
 HER2†Positive18 (18)
Negative81 (82)
 ER/PgR/HER2†Triple negative12 (12)
Others92 (88)
 Grade†Grade 112 (12)
Grade 241 (39)
Grade 344 (42)
Unknown2 (7)
 RecurrenceYes22 (21)
No72 (69)
Unknown10 (10)
 DeathYes7 (7)
No97 (93)
Testing cohort
 Age>50147 (76)
≤5045 (23)
Unknown1 (1)
 Tumor size≤2 cm85 (44)
2–5 cm97 (50)
>5 cm11 (6)
 Histological nodal statusPositive79 (41)
Negative109 (56)
Unknown5 (3)
 ERPositive139 (72)
Negative50 (26)
Unknown4 (2)
 PgRPositive95 (49)
Negative95 (49)
Unknown3 (2)
 HER2Positive22 (11)
Negative170 (88)
Unknown1 (1)
 ER/PgR/HER2Triple negative35 (18)
Others154 (80)
Unknown4 (2)
 GradeGrade 158 (30)
Grade 250 (26)
Grade 385 (44)
 RecurrenceYes28 (14)
No165 (86)
 DeathYes15 (8)
No178 (92)

Immunohistochemical analysis.  Immunohistochemical analysis was performed as described previously.(18,19) In brief, slides were incubated with an anti-Cav-1 monoclonal antibody (1:800; Cell Signaling Technology, Danvers, MA, USA). The signals were detected by Envision kit (Dako, Glostrup, Denmark) and the sections were counterstained with haematoxylin. Negative control sections were incubated with phosphate-buffered saline plus 1% bovine serum albumin instead of primary antibody. Endothelial cells were used as internal positive controls because these cells commonly express Cav-1 in cancerous regions. Results of the analyses were evaluated by two pathologists, who were independent and blinded to the clinical features of the study. We determined three hot-spots at ×400 magnification, calculated the number of Cav-1-stained tumor cells (T) and stromal cells (S) and graded the cells as follows: negative expression (−), ≤5%; low expression (+), 5–50%; and high expression (++), >50%. For analyzing combined tumor/stromal Cav-1 expression, we determined tumors with a high tumor hot-spot grade that were stromal negative as T(++)/S(−).

Statistical analysis.  The correlation between different types of Cav-1 expression was evaluated using Spearman’s test. The correlation between Cav-1 and clinicopathological parameters was evaluated using the Kruskal–Wallis test. Disease-free survival was estimated using the Kaplan–Meier estimate and a comparison of stratified survival curves was performed using log-rank tests. Cox analysis was used to evaluate the correlation between Cav-1 and disease-free survival in the presence of various potential prognostic factors for disease-free survival. Differences were considered statistically significant at < 0.05.(18,19)

Results

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

Immunohistochemical analysis of Cav-1 expression in the training cohort.  All patients were Japanese women and their clinicopathological characteristics are listed in Table 1. Twenty-two patients developed recurrence and seven of them died as a result. Distributions of recurrence and survival parameters are also indicated in Table 1. Table S1 summarizes the number of patients in each subgroup stratified by Cav-1 grade. T(−) Cav-1 expression was observed in 16%, T(+) Cav-1 expression was observed in 67% and T(++) Cav-1 expression was observed in 17% of breast cancer patients. Strong positive staining showed a prevalent membrane pattern associated with cytoplasm positive. For total stromal Cav-1 expression, 57% were S(++/+) and 43% were S(−). In total, 72 patients were S(++/+) and 32 showed S(−) expression stratified by stromal hot-spot Cav-1 expression. For combined Cav-1 expression score grading, 5% of breast cancer patients were T(++)/S(−) and 16% were T(+)/S(−). Representative examples are illustrated in Figure 1.

image

Figure 1.  Immunohistochemical analysis of caveolin-1 (Cav-1) expression (×400). (a) Tumor hot-spot expression. (b) Diffuse Cav-1 expression in stroma and (c) negative stromal Cav-1 expression. Endothelial cells indicate positive immunostaining for Cav-1 used as internal positive controls (arrows).

Download figure to PowerPoint

Next, correlations between the different types of Cav-1 expression were analyzed. Stromal hot-spot Cav-1 expression was significantly correlated with total stromal Cav-1 expression (R2 = 0.517, < 0.001). However, stromal hot-spot Cav-1 expression was weakly correlated (R2 = 0.081, = 0.003) with tumor hot-spot expression. No significant difference was observed between total stromal and tumor hot-spot Cav-1 expression (R2 = 0.029, = 0.083). Because of the strong correlation between stromal hot-spot and total Cav-1 expression, we used the former as a representative for the following analysis.

Correlations between Cav-1 expression and clinicopathological parameters in the training cohort. Table 2 summarizes the correlation between Cav-1 expression and the clinicopathological parameters of breast cancer patients in the training cohort. T(++) Cav-1 expression was positively associated with tumor size and histological nodal status (= 0.019 and 0.021, respectively). S(−) Cav-1 expression was independent of histological nodal and human epidermal growth factor receptor 2 (HER2) status (= 0.385 and 0.055, respectively). No significant correlations were found between stromal hot-spot and tumor hot-spot Cav-1 expression and age, tumor stage, grade, estrogen receptor (ER) status or progesterone receptor (PgR) status (> 0.05). T(++)/S(−) Cav-1 expression was positively associated with tumor size and histological nodal status (< 0.001 and 0.012, respectively). No significant correlations were found between T(++)/S(−) Cav-1 expression and age, tumor stage, grade, ER, PgR or HER2 status (> 0.05).

Table 2.   Associations between caveolin-1 (Cav-1) expression and clinicopathological parameters in the training and testing cohorts
ParameterVariableTumor hot-spotStromal hot-spotTumor/stromalTumor/stromal
Cav-1 expression (training cohort)Cav-1 expression (training cohort)Cav-1 expression (training cohort)Cav-1 expression (testing cohort)
T(+/−)T(++)PS(−)S(++/+)PT(++)/S(−)OthersPT(++)/S(−)OthersP
  1. †Patients with ductual carcinoma in situ (DCIS) were excluded. ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PgR, progesterone receptor; S, stromal hot-spot; T, tumor hot-spot.

Tumor size 27.0 ± 12.836.3 ± 17.40.01930.5 ± 13.128.2 ± 14.60.49836.6 ± 6.920.9 ± 3.3<0.000140.9 ± 45.423.8 ± 12.4<0.001
Age (%)≤5024 (28)5 (28)0.99110 (31)19 (26)0.6121 (20)28 (28)0.68813 (59)134 (79)0.129
>5062 (72)13 (72)22 (69)53 (74)4 (80)71 (72)9 (41)36 (21)
T status (%)T0/T124 (34)2 (12)0.0955 (21)21 (33)0.2580 (0)26 (31)0.2535 (23)80 (47)0.033
T2/T3/T447 (66)14 (88)19 (79)42 (67)3 (100)58 (69)17 (77)91 (53)
Histological nodal status (%)Positive34 (40)12 (71)0.02116 (52)30 (42)0.3855 (100)41 (42)0.0125 (25)104 (62)0.002
Negative51 (60)5 (29)15 (48)41 (58)0 (0)56 (58)15 (75)64 (38)
Grade† (%)Grade 112 (15)0 (0)0.1402 (7)10 (15)0.2640 (0)12 (13)0.2084 (18)54 (32)0.113
Grade 231 (39)10 (59)11 (37)30 (45)4 (80)37 (40)5 (23)45 (26)
Grade 337 (46)7 (41)17 (56)27 (40)1 (20)43 (47)13 (59)72 (42)
ER (%)Positive69 (80)11 (61)0.08121 (66)59 (82)0.3593 (60)77 (78)0.36013 (62)126 (75)0.201
Negative17 (20)7 (39)11 (34)13 (18)2 (40)22 (22)8 (38)42 (25)
PgR (%)Positive58 (67)9 (50)0.16219 (59)48 (67)0.4763 (60)64 (65)0.47110 (48)85 (50)0.817
Negative28 (33)9 (50)13 (41)24 (33)2 (40)35 (35)11 (52)84 (50)
HER2† (%)Positive15 (19)3 (17)0.8559 (30)9 (13)0.0552 (40)16 (17)0.1973 (29)19 (33)0.667
Negative66 (81)15 (83)21 (70)60 (87)3 (60)78 (83)18 (71)152 (67)

Prognostic value of Cav-1 expression in the training cohort. Figure 2a–d illustrates the Kaplan–Meier curves of disease-free survival for the training cohort constructed on the basis of the Cav-1 expression level. T(++), S(−) and T(++)/S(−) status correlated closely with poor disease-free survival (= 0.009, 0.019 and <0.001, respectively). No major differences were observed between S(+) and S(++) or between T(+) and T(−) Cav-1 expression on the predictive value of disease-free survival (> 0.05; data not shown). T(+)/S(−) status also did not influence disease-free survival.

image

Figure 2.  Disease-free survival curves of the training cohort stratified by (a) stromal hot-spot caveolin-1 (Cav-1) expression status: S(−) vs S(++/+; = 0.009, log-rank test); (b) tumor hot-spot Cav-1 expression status: T(++) vs T(+/−; = 0.019, log-rank test); (c) combined Cav-1 expression: T(++)/S(−) vs Others 1 (< 0.001, log-rank test); and (d) combined Cav-1 expression: T(+)/S(−) vs Others 2 (= 0.662, log-rank test). Disease-free survival curves of the testing cohort stratified by (e) combined Cav-1 expression: T(++)/S(−) vs Others (< 0.001, log-rank test).

Download figure to PowerPoint

Kaplan–Meier analysis demonstrated a significant impact of certain clinicopathological prognostic factors such as HER2, tumor stage and ER on disease-free survival (= 0.015, 0.010 and 0.010, respectively). No significant correlations were found between disease-free survival and other clinicopathological factors, including PgR, histological nodal status, grade and age (> 0.05; Table 3). Cox analysis was performed to evaluate whether the correlation between Cav-1 expression and disease-free survival was related to the correlation of Cav-1 expression with other prognostic factors. The results revealed that S(−) Cav-1 expression was not an independent prognostic factor for disease-free survival (= 0.063). No significant correlation was found between T(++) Cav-1 expression and disease-free survival (= 0.114). Consistent with the univariate analysis results, the multivariate analysis revealed that combined T(++)/S(−) status was strongly associated with an unfavorable prognosis (= 0.002; Table 4). We conducted further subgroup analysis stratified by histological nodal, ER and HER2 status, which were associated with poor clinical outcomes. As a result, S(−) Cav-1 showed clear trends for predicting disease-free survival in histological node+ patients (= 0.008). However, T(+) Cav-1 expression did not influence disease-free survival in histological node+ patients (= 0.088). Notably, histological node+ patients with T(++)/S(−) Cav-1 expression exhibited lower disease-free survival (= 0.001). Furthermore, T(++)/S(−) Cav-1 expression also served as an important predictor of disease-free survival for ER+ and HER2+ patients (= 0.001 and 0.045, respectively). Moreover, patients with T(++)/S(−) Cav-1 expression who were in the ER−, HER2−, PgR (+ and −), tumor size (>5 and ≤5 cm), age (>50 and ≤50 years) and grade (Grade 2 and 3) subgroups had poorer disease-free survival (Table S2).

Table 3.   Univariate analyses of factors associated with recurrence in the training and testing cohorts
VariableDisease-free survival (P)
  1. Cav-1, caveolin-1; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PgR, progesterone receptor; S, stromal hot spot; T, tumor hot spot.

Training cohort
 Tumor/stromal Cav-1 expression T(++)/S(−) vs others<0.001
 Stromal hot-spot Cav-1 expression S(++/+) vs S(−)0.009
 Tumor stage (T0/T1 vs T2/T3/T4)0.010
 ER (positive vs negative)0.010
 HER2 (positive vs negative)0.015
 Tumor hot-spot Cav-1 expression T(++) vs T(+/−)0.019
 PgR (positive vs negative)0.054
 Histological nodal (positive vs negative)0.069
 Age (>50 vs≤50)0.411
Testing cohort
 Tumor/stromal Cav-1 expression T(++)/S(−) vs others<0.001
 Histological nodal status (positive vs negative)<0.001
 Grade (1 vs 2 vs 3)<0.001
 HER2 (positive vs negative)<0.001
 Tumor stage (T0/T1 vs T2/T3/T4)0.008
 ER (positive vs negative)0.020
 PgR (positive vs negative)0.054
 Age (>50 vs≤50)0.376
Table 4.   Multivariate analysis of factors that might affect disease-free survival in the training and testing cohorts
 Disease-free survival
HR95% CIP
  1. Cav-1, caveolin-1; CI, confidence interval; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; HR, hazard ratio.

Training cohort
 Stromal hot-spot Cav-1 expression0.3220.098–1.0650.063
 ER1.0920.306–3.9010.893
 HER23.3621.035–10.9230.044
 Tumor stage7.7720.985–61.2880.052
 Histological nodal status1.5360.900–2.6190.115
 Grade3.8071.018–7.4720.021
 Tumor hot spot-Cav-1 expression0.3700.108–1.2690.114
 ER0.8440.248–2.8670.785
 HER24.7781.337–17.0810.016
 Tumor stage6.4370.806–51.3920.079
 Histological nodal status1.4060.814–2.4300.222
 Grade0.3221.220–11.8810.046
 Tumor/stromal Cav-1 expression0.0410.006–0.2970.002
 ER0.7740.232–2.5800.677
 HER23.6651.037–12.9540.044
 Tumor stage7.2340.900–58.1290.063
 Histological nodal status1.3270.753–2.3410.328
 Grade5.8681.523–22.6140.010
Testing cohort
 Tumor/stromal Cav-1 expression0.2490.107–0.5820.001
 ER0.8460.362–1.9820.701
 HER22.1410.875–5.2380.096
 Tumor stage1.5510.965–2.4940.070
 Histological nodal status4.1241.517–11.2080.005
 Histological grade3.1501.416–7.0070.005

Validation of the prediction power of T(++)/S(−) Cav-1 status in the testing cohort.  In the testing cohort, all patients were also Japanese women and their clinicopathological characteristics are listed in Table 1. T(++)/S(−) Cav-1 expression was observed in 11% and other expressions were observed in 89% breast cancer patients. T(++)/S(−) Cav-1 expression was significantly related to tumor size (< 0.001), tumor stage (= 0.033) and histological nodal status (= 0.002). We could not find a statistically significant association between T(++)/S(−) Cav-1 expression and age, grade, ER, PgR or HER2 (> 0.05; Table 2).

We then examined the association of T(++)/S(−) Cav-1 expression with the clinical outcome. Kaplan–Meier survival analysis showed patients with T(++)/S(−) Cav-1 expression had shorter disease-free survival than those with other expressions (< 0.001, Fig. 2e). The significant impact of clinicopathological prognostic factors such as tumor stage, ER and HER2 on disease-free survival (= 0.008, 0.020 and <0.001, respectively) was also validated. These results were consistent with the above findings. In addition, a significant correlation between disease-free survival and grade or histological nodal status was observed in the testing cohort (< 0.001 and <0.001, respectively; Table 3).

Table 4 provides the multivariate analyses of factors related to patient disease-free survival. Cox analysis indicated that T(++)/S(−) Cav-1 status was an independent predictor of disease-free survival (= 0.001), as were histological nodal status (= 0.005) and grade (= 0.005). Moreover, the role of T(++)/S(−) Cav-1 expression in disease-free survival in the ER (+ and −), HER2 (+ and −), PgR (+ and −), tumor size (>5 and ≤5 cm), age (>50 and ≤50 years) and grade (Grade 1 + 2 and 3) subgroups is shown in Table S2.

Discussion

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

The Cav-1 gene is colocalized at the D7S522 locus on human chromosome 7q31.1 and is commonly deleted in breast, colon, kidney, prostate, ovary, head and neck cancers. Thus, it seems feasible to propose that the Cav-1 gene might serve as a candidate tumor suppressor gene.(21) In the present study, we focused on breast cancer patients to determine the correlation of tumor/stromal Cav-1 expression with clinicopathological parameters and survival. Cav-1 expression was evaluated semi-quantitatively based on the proportion of positively stained tumor and stromal cells. We found that tumor Cav-1 demonstrated a prevalent membrane pattern associated with cytoplasm positive and that T(−) Cav-1 expression was noted in 16% of cases, which is consistent with previous reports.(22) The stromal Cav-1 expression rate and pattern in the present study are similar to a previous report.(23)

According to previous reports, total tumor Cav-1 expression has no prognostic value in primary breast cancer patients.(18,24) In the present study, we examined stromal hot-spot, total stromal and tumor hot-spot Cav-1 expression. First, we analyzed correlations between stromal hot-spot and total stromal and tumor Cav-1 expression and found a weak but significant correlation between stromal and tumor expression, indicating that Cav-1 expression could be regulated differently between tumor and stromal cells and that Cav-1 might influence different functions in those cells.(25,26) T(++) Cav-1 expression was positively associated with tumor size and histological nodal status. Previous reports revealed that tumor Cav-1expression was negatively associated with HER2 status.(27–29) Several studies have indicated that Cav-1 might function as a negative signal transduction regulator to HER2/neu and that it might play a negative regulatory role in mammary tumor development. In addition, activation of HER2/neu might downregulate Cav-1 expression in vitro.(30,31) However, this finding was not supported by our results. Stromal Cav-1 expression showed no significant correlation with any of the clinicopathological parameters, which was inconsistent with a previous report,(18) and therefore we focused on combined tumor/stromal Cav-1 expression. T(++)/S(−) was observed frequently in large-size tumors and histological node+ cases, indicating that tumor/stromal Cav-1 expression is involved in breast cancer progression. Interestingly, survival analyses revealed that patients with T(++)/S(−) Cav-1 expression had the shortest disease-free survival among various Cav-1 expression subgroups. Multivariate analysis confirmed an independent prognostic value of the combined status. Consistent with these results, T(++)/S(−) Cav-1 expression was significantly related to tumor size, histological nodal status and disease-free survival in the testing cohort. Besides, the positive correlation between T(++)/S(−) Cav-1 expression and tumor stage was also indicated in the testing cohort. A possible explanation for the discrepancy could be due to the difference in sample size. Furthermore, T(++)/S(−) Cav-1 expression also impacted the clinical outcomes stratified by ER status, PgR status, HER2 status, tumor size, age, histological nodal status and grade in both the training and testing cohorts. Therefore, these results indicate that combined Cav-1 status has a more potent prognostic value than either stromal or tumor hot-spot alone. We believe that these results are important when considering breast cancer biology. Given the limited number of cases, prospective studies with long-term follow-up data are warranted.(32)

Tumor Cav-1 expression with respect to tumorigenesis seems more complex than originally believed. Cav-1 loss-of-function induces ligand-independent hyperactivation of Ras-p42/44 MAPK and Smad signaling pathways as well as enhanced matrix metalloproteinase-2/9 secretion. Each of these pathways is likely to contribute to cell cycle progression, growth factor independence, cell invasiveness and epithelial–mesenchymal transition.(33) Despite extensive evidence supporting the role of Cav-1 as a tumor suppressor, several studies have suggested an alternative view of Cav-1 expression in tumors. In breast cancer, Cav-1 protects tumor cells from anoikis, promotes tumor cell survival and abrogates detachment-induced p53 activation.(34,35) Furthermore, Cav-1 expression is upregulated in multidrug-resistant MCF-7 cells.(36,37) A hypothesis has been proposed to explain the divergent roles of Cav-1; even if an initial loss of Cav-1 is observed in breast cancer, re-expression of Cav-1 at later stages might correlate with more malignant characteristics.(35,38)

Stromal Cav-1 plays a vital role in tumorigenesis. Loss of stromal Cav-1 is an independent predictor for therapeutic resistance and poor prognosis in primary breast cancers.(12,18,19,39) Woodman et al.(12) reported that endothelial cells from Cav-1−/− mice exhibit a disrupted response to angiogenic growth factors. Senescent human diploid fibroblasts exhibit increased levels of the Cav-1 protein.(10) In addition, loss of Cav-1 in stromal cells of various organs directly leads to disorganised stromal compartments and dysfunctional organ systems.(40)

Furthermore, recent studies have revealed a role played by Cav-1 in the interaction between tumor and stromal cells in breast cancer. During tumor formation, cancer cells and adjacent fibroblasts are metabolically coupled. A new model has been proposed in which glycolytic CAF promote tumor growth by secreting energy-rich metabolites that can be taken up by adjacent tumor cells.(41) Loss of Cav-1 in vitro induces metabolic coupling between CAF and tumor cells and leads to the formation of a host–parasite relationship. Martinez-Outschoorn et al.(42) showed that Cav-1 expression is downregulated in fibroblasts co-cultured with MCF-7 cells and that it mediates autophagic/lysosomal degradation. Furthermore, autophagy induced by loss of Cav-1 in fibroblasts provides cancer cells with essential chemical building blocks.(42,43) Loss of stromal Cav-1 fibroblasts protects adjacent cancer cells via decreased apoptosis, increased TP53-induced glycolysis and apoptosis regulator expression.(44) Furthermore, loss of Cav-1 induces oxidative stress in CAF, which is the root cause of mitochondrial dysfunction in CAF and promotes DNA damage. In the present study, the predictive value of T (++)/S(−) was demonstrated in luminal-type cancers and HER2+ cancers. Its value was stratified by an intrinsic subtype and warrants an examination with a greater number of cases.

The regulatory mechanism of Cav-1 expression in breast cancer remains to be elucidated. Pro-autophagic stimuli such as hypoxia, oxidative stress and nuclear factor κB activation might cause the loss of Cav-1.(45) Conversely, multiple factors are present during Cav-1 upregulation.

In conclusion, we provide evidence that T (++)/S(−) Cav-1 expression is closely associated with unfavorable prognostic outcomes in primary breast cancer patients. This particular subgroup seems to be engaged in rapid disease progression. Further studies involving analysis of molecular mechanisms of Cav-1 expression are required. The interaction between tumor and stromal cells and Cav-1 in the tumor microenvironment is also a key issue to investigate. Moreover, new therapies targeting Cav-1 expression might be a novel therapeutic approach, particularly for patients with T (++)/S(−) Cav-1 status.

Acknowledgment

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

N.S.Q. was supported by the State Scholarship Foundation of China (No. 2009659015).

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure Statement
  8. References
  9. Supporting Information
  • 1
    Gluz O, Liedtke C, Gottschalk N, Pusztai L, Nitz U, Harbeck N. Triple-negative breast cancer – current status and future directions. Ann Oncol 2010; 21: 712.
  • 2
    Yamashiro H, Toi M. Update of evidence in chemotherapy for breast cancer. Int J Clin Oncol 2008; 13: 37.
  • 3
    Shatz M, Liscovitch M. Caveolin-1: a tumor-promoting role in human cancer. Int J Radiat Biol 2008; 84: 17789.
  • 4
    Cantiani L, Manara MC, Zucchini C et al. Caveolin-1 reduces osteosarcoma metastases by inhibiting c-Src activity and met signaling. Cancer Res 2007; 67: 767585.
  • 5
    Wikman H, Kettunen E, Seppänen JK et al. Identification of differentially expressed genes in pulmonary adenocarcinoma by using cDNA array. Oncogene 2002; 21: 580413.
  • 6
    Kato K, Hida Y, Miyamoto M et al. Overexpression of caveolin-1 in esophageal squamous cell carcinoma correlates with lymph node metastasis and pathologic stage. Cancer 2002; 94: 92933.
  • 7
    Suzuoki M, Miyamoto M, Kato K et al. Impact of caveolin-1 expression on prognosis of pancreatic ductal adenocarcinoma. Br J Cancer 2002; 87: 11404.
  • 8
    Cohen AW, Hnasko R, Schubert W, Lisanti MP. Role of caveolae and caveolins in health and disease. Physiol Rev 2004; 84: 134179.
  • 9
    Zhang W, Razani B, Altschuler Y et al. Caveolin-1 inhibits epidermal growth factor-stimulated lamellipod extension and cell migration in metastatic mammary adenocarcinoma cells (MTLn3). Transformation suppressor effects of adenovirus-mediated gene delivery of caveolin-1. J Biol Chem 2000; 275: 2071725.
  • 10
    Park WY, Park JS, Cho KA et al. Up-regulation of caveolin attenuates epidermal growth factor signaling in senescent cells. J Biol Chem 2000; 275: 2084752.
  • 11
    Sotgia F, Schubert W, Pestell RG, Lisanti MP. Genetic ablation of caveolin-1 in mammary epithelial cells increases milk production and hyper-activates STAT5a signaling. Cancer Biol Ther 2006; 5: 2927.
  • 12
    Woodman SE, Ashton AW, Schubert W et al. Caveolin-1 knockout mice show an impaired angiogenic response to exogenous stimuli. Am J Pathol 2003; 162: 205968.
  • 13
    Volonte D, Zhang K, Lisanti MP, Galbiati F. Expression of caveolin-1 induces premature cellular senescence in primary cultures of murine fibroblasts. Mol Biol Cell 2002; 13: 250217.
  • 14
    Chretien A, Piront N, Delaive E, Demazy C, Ninane N, Toussaint O. Increased abundance of cytoplasmic and nuclear caveolin 1 in human diploid fibroblasts in H(2)O(2)-induced premature senescence and interplay with p38alpha(MAPK). FEBS Lett 2008; 582: 168592.
  • 15
    Ostman A, Augsten M. Cancer-associated fibroblasts and tumor growth – bystanders turning into key players. Curr Opin Genet Dev 2009; 19: 6773.
  • 16
    Qian N, Ueno T. Is dysfunction of caveolin-1 a link between systemic sclerosis and breast cancer, opening a window on both etiologies? Arch Med Res 2010; 41: 297301.
  • 17
    Williams TM, Sotgia F, Lee H et al. Stromal and epithelial caveolin-1 both confer a protective effect against mammary hyperplasia and tumorigenesis: caveolin-1 antagonizes cyclin D1 function in mammary epithelial cells. Am J Pathol 2006; 169: 1784801.
  • 18
    Witkiewicz AK, Dasgupta A, Sotgia F et al. An absence of stromal caveolin-1 expression predicts early tumor recurrence and poor clinical outcome in human breast cancers. Am J Pathol 2009; 174: 202334.
  • 19
    Witkiewicz AK, Dasgupta A, Nguyen KH et al. Stromal caveolin-1 levels predict early DCIS progression to invasive breast cancer. Cancer Biol Ther 2009; 8: 10719.
  • 20
    Sakamoto G, Inaji H, Akiyama F et al ; Japanese Breast Cancer Society. General rules for clinical and pathological recording of breast cancer 2005. Breast Cancer 2005; 12(Suppl): S127.
  • 21
    Liu P, Rudick M, Anderson RG. Multiple functions of caveolin-1. J Biol Chem 2002; 277: 412958.
  • 22
    Perrone G, Altomare V, Zagami M et al. Caveolin-1 expression in human breast lobular cancer progression. Mod Pathol 2009; 22: 718.
  • 23
    Sloan EK, Ciocca DR, Pouliot N et al. Stromal cell expression of caveolin-1 predicts outcome in breast cancer. Am J Pathol 2009; 174: 203543.
  • 24
    Liedtke C, Kersting C, Bürger H, Kiesel L, Wülfing P. Caveolin-1 expression in benign and malignant lesions of the breast. World J Surg Oncol 2007; 5: 110.
  • 25
    Rønnov-Jessen L, Bissell MJ. Breast cancer by proxy: can the microenvironment be both the cause and consequence? Trends Mol Med 2009; 15: 513.
  • 26
    Mercier I, Casimiro MC, Wang C et al. Human breast cancer-associated fibroblasts (CAFs) show caveolin-1 downregulation and RB tumor suppressor functional inactivation: implications for the response to hormonal therapy. Cancer Biol Ther 2008; 7: 121225.
  • 27
    Savage K, Lambros MB, Robertson D et al. Caveolin 1 is overexpressed and amplified in a subset of basal-like and metaplastic breast carcinomas: a morphologic, ultrastructural, immunohistochemical, and in situ hybridization analysis. Clin Cancer Res 2007; 13: 90101.
  • 28
    Sagara Y, Mimori K, Yoshinaga K et al. Clinical significance of Caveolin-1, Caveolin-2 and HER2/neu mRNA expression in human breast cancer. Br J Cancer 2004; 91: 95965.
  • 29
    Park SS, Kim JE, Kim YA et al. Caveolin-1 is down-regulated and inversely correlated with HER2 and EGFR expression status in invasive ductal carcinoma of the breast. Histopathology 2005; 47: 62530.
  • 30
    Couet J, Sargiacomo M, Lisanti MP. Interaction of a receptor tyrosine kinase, EGF-R, with caveolins. Caveolin binding negatively regulates tyrosine and serine/threonine kinase activities. J Biol Chem 1997; 272: 3042938.
  • 31
    Kim YN, Wiepz GJ, Guadarrama AG, Bertics PJ. Epidermal growth factor-stimulated tyrosine phosphorylation of caveolin-1. Enhanced caveolin-1 tyrosine phosphorylation following aberrant epidermal growth factor receptor status. J Biol Chem 2000; 275: 748191.
  • 32
    Decensi A, Guerrieri-Gonzaga A, Gandini S et al. Prognostic significance of Ki-67 labeling index after short-term presurgical tamoxifen in women with ER-positive breast cancer. Ann Oncol 2011; 22: 5827.
  • 33
    Sotgia F, Rui H, Bonuccelli G, Mercier I, Pestell RG, Lisanti MP. Caveolin-1, mammary stem cells, and estrogen-dependent breast cancers. Cancer Res 2006; 66: 1064751.
  • 34
    Fiucci G, Ravid D, Reich R, Liscovitch M. Caveolin-1 inhibits anchorage-independent growth, anoikis and invasiveness in MCF-7 human breast cancer cells. Oncogene 2002; 21: 236575.
  • 35
    Ravid D, Maor S, Werner H, Liscovitch M. Caveolin-1 inhibits anoikis and promotes survival signaling in cancer cells. Adv Enzyme Regul 2006; 46: 16375.
  • 36
    Lavie Y, Fiucci G, Liscovitch M. Up-regulation of caveolae and caveolar constituents in multidrug-resistant cancer cells. J Biol Chem 1998; 273: 323803.
  • 37
    Lavie Y, Liscovitch M. Changes in lipid and protein constituents of rafts and caveolae in multidrug resistant cancer cells and their functional consequences. Glycoconj J 2000; 17: 2539.
  • 38
    Quest AF, Gutierrez-Pajares JL, Torres VA. Caveolin-1: an ambiguous partner in cell signaling and cancer. J Cell Mol Med 2008; 12: 113050.
  • 39
    Sotgia F, Williams TM, Schubert W et al. Caveolin-1 deficiency (−/−) conveys premalignant alterations in mammary epithelia, with abnormal lumen formation, growth factor independence, and cell invasiveness. Am J Pathol 2006; 168: 292309.
  • 40
    Yang G, Timme TL, Naruishi K et al. Mice with cav-1 gene disruption have benign stromal lesions and compromised epithelial differentiation. Exp Mol Pathol 2008; 84: 13140.
  • 41
    Pavlides S, Tsirigos A, Vera I et al. Transcriptional evidence for the “Reverse Warburg Effect” in human breast cancer tumor stroma and metastasis: similarities with oxidative stress, inflammation, Alzheimer’s disease, and “Neuron-Glia Metabolic Coupling”. Aging (Albany NY) 2010; 2: 18599.
  • 42
    Martinez-Outschoorn UE, Pavlides S, Whitaker-Menezes D et al. Tumor cells induce the cancer associated fibroblast phenotype via caveolin-1 degradation: implications for breast cancer and DCIS therapy with autophagy inhibitors. Cell Cycle 2010; 9: 242333.
  • 43
    Pavlides S, Tsirigos A, Vera I et al. Loss of stromal caveolin-1 leads to oxidative stress, mimics hypoxia and drives inflammation in the tumor microenvironment, conferring the “reverse Warburg effect”: a transcriptional informatics analysis with validation. Cell Cycle 2010; 9: 220119.
  • 44
    Martinez-Outschoorn UE, Trimmer C, Lin Z et al. Autophagy in cancer associated fibroblasts promotes tumor cell survival: role of hypoxia, HIF1 induction and NFκB activation in the tumor stromal microenvironment. Cell Cycle 2010; 9: 351533.
  • 45
    Bonuccelli G, Whitaker-Menezes D, Castello-Cros R et al. The reverse Warburg effect: glycolysis inhibitors prevent the tumor promoting effects of caveolin-1 deficient cancer associated fibroblasts. Cell Cycle 2010; 9: 196071.

Supporting Information

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

Table S1. The numbers of patients in each subgroup stratified by Cav-1 grade.

Table S2. Univariate analyses of Cav-1 expression associated with disease-free survival of the training and testing cohort.

FilenameFormatSizeDescription
CAS_1985_sm_tS1.doc39KSupporting info item
CAS_1985_sm_tS2.doc50KSupporting info item

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