| Japanese Journal of Clinical Oncology | Pages |
Introduction
Patients And Methods
Patients and Follow-up
Immunohistochemistry
Statistical Analysis
Results
Discussion
Acknowledgements
References
Prognostic Indicators for Breast Cancer Patients with One to Three Regional Lymph Node Metastases, with Special Reference to Alterations in Expression Levels of bcl-2, p53 and c-erbB-2 Proteins
Patients with primary breast carcinoma with one to three axillary lymph node metastases but without distant metastases (n1-3) in Japan have been shown to have a 10-year disease-free survival rate of >60%. It would be reasonable to divide n1-3 Japanese breast cancer patients into groups with high- or low-risk for recurrence and to consider post-operative adjuvant therapy. In the present study, we analyzed 228 consecutive Japanese patients with n1-3 breast cancer who underwent radical mastectomy and were followed up for a median time of 11.0 years. The expression of bcl-2, p53 and c-erbB-2 proteins in the primary tumors was examined immunohistochemically and their prognostic roles were also analyzed along with conventional clinicopathologic indicators. bcl-2 expression was correlated with positive estrogen receptor status and inversely correlated with p53, c-erbB-2 and histologic grade. Univariate analysis showed that bcl-2, p53 and c-erbB-2 expression were prognostic indicators of the patient's group as well as node status, histologic grade, tumor size, age at diagnosis, menopausal status and estrogen receptor status. Cox's regression analysis demonstrated that the number of nodes involved, menopausal status, p53 and bcl-2 were independent predictors for overall survival and that histologic grade and the number of nodes involved were independent predictors for disease-free survival. These results suggest that bcl-2 expression in combination with p53 and c-erbB-2 expression, the number of lymph node metastases, histologic grade and menopausal status are useful in selecting subgroups of n1-3 breast cancer patients with good or poor prognoses.
INTRODUCTION
The number of ipsilateral axillary lymph nodes with metastases is accepted as the most powerful prognostic indicator for patients with primary breast cancer who have received surgical therapies (1 ). In the pathological staging system adopted by both the International Union against Cancer (UICC) and the American Joint Committee on Cancer Staging and End Results Reporting (AJC), the status of ipsilateral axillary lymph nodes is classified based on the number of metastatic lymph nodes (0, 1-3, >= 4), the size of the metastatic tumors ( <= 0.2 cm, >0.2 cm and <= 2.0 cm, >2.0 cm) and the fixation to one another or to other structures (2 ,3 ). In the Japanese pathological n-factor system, the status of ipsilateral axillary lymph node metastasis is classified into n1[alpha] (1-3 metastatic nodes) and n1[beta] ( >= 4 metastatic nodes) (4 ). Primary breast cancers with one to three regional axillary lymph node metastases (n1-3 breast cancers) but without evidence of distant metastases are generally accepted as a group with an intermediate degree of spread, showing a poorer prognosis than the group with no lymph node metastases but showing a better prognosis than the group with four or more lymph node metastases.
Several studies on breast cancer reported that the prognoses of Japanese women were better than those of women in the United States or European countries (5 ,6 ). The same tendency was also observed in n1-3 breast cancer. Compared with the 5-year and 10-year disease-free survival (DFS) rates of 50-65% and 36% respectively, reported by the National Surgical Adjuvant Breast and Bowel Project and Yale-New Haven Hospital (7 ,8 ), those reported in the National Cancer Center Hospital, Japan, are 76% and 64% respectively. Another Japanese institute reported a similar result for n1-3 breast cancer (9 ). These data suggest that n1-3 breast cancer in Japanese patients has a similar prognosis to node-negative breast cancer in Europe and America. Although post-operative adjuvant chemotherapy is almost always performed for patients with n1-3 breast cancer in Japan, based on such a relatively high recurrence-free rate it would be reasonable to divide n1-3 breast cancer patients into high- and low-risk groups for recurrence and to consider a protocol design in which intensive adjuvant therapy is performed for the high-risk group but not for the low-risk group.
Recently, abnormalities of tumor-associated genes and their proteins in breast cancer cells have been shown to be correlated with the clinical outcomes for patients. The bcl-2 oncogene was initially discovered as a gene involved in 14;18 translocation common in follicular lymphomas (10 ). This gene encodes an inner mitochondrial membrane protein of 26 kDa and acts specifically by blocking apoptotic cell death (11 ). In addition to malignant lymphoma, the bcl-2 protein has been identified by immunohistochemical methods in a number of malignancies (12 -19 ). In primary breast cancer the expression of bcl-2 is shown to be closely related to estrogen receptor (ER) and progesterone receptor (PgR) statuses and inversely related to the expression of epidermal growth factor receptor, c-erbB-2 protein and nuclear accumulation of p53 protein (15 -19 ). Gasparini et al. suggested that bcl-2 protein status is an independent factor predictive of clinical outcome in patients with node-positive breast cancer (19 ). However, Silvestrini et al. indicated that bcl-2 expression did not have an independent prognostic value compared with other biological prognostic factors and was mainly dependent on p53 expression in a node-negative group (18 ).
The p53 tumor-suppressor gene is involved in the transcription of genes that negatively control cell proliferation (20 ,21 ). Various immunohistochemical studies revealed that 20-45% of breast cancers show nuclear accumulation of mutant p53-protein (22 -24 ) and the nuclear accumulation is shown to be an indicator of a poor clinical outcome for breast cancer patients regardless of their lymph-node status (25 -28 ). Furthermore, c-erbB-2 overexpression was reported to be an indicator of poor prognosis independent of lymph node status (29 -31 ). The determination of these gene expressions in primary tumors would be effective in selecting poor or good prognosis groups among breast cancer patients with similar degrees of lymph node metastasis. We examined bcl-2, p53 and c-erbB-2 alterations immunohistochemically in n1-3 breast cancer and analyzed their prognostic value in combination with conventional clinicopathological parameters by univariate and multivariate analyses in order to divide such patients into groups at high-risk and low-risk of cancer recurrence and death.
PATIENTS AND METHODS
Patients and Follow-up
Among the patients with primary breast carcinoma who underwent surgical therapies at the National Cancer Center Hospital, Tokyo, between January 1980 and December 1984, we selected 228 consecutive cases with one to three axillary lymph node metastases, without evidence of distant metastases (M0) at the time of surgery. The surgical procedures were standard radical mastectomy in 107, extended radical mastectomy in 103 and modified radical mastectomy in 18. In all cases, complete axillary node dissection was performed. Of the 228 patients, 209 received adjuvant chemotherapy, seven received both adjuvant chemotherapy and endocrine therapy (tamoxifen) and one received tamoxifen therapy alone, following surgery, while 11 patients had no adjuvant therapy. The median age of the patients was 50.7 years, ranging from 27 to 75 years.
Patients were followed up in the outpatient clinic of the same hospital or other hospitals at 3-6 month intervals during the first 5 years and at 6-12 month intervals thereafter. Relapses and metastases were monitored by physical examination, serum tumor markers including carcinoembryonic antigen and CA15-3, chest roentgenogram and skeletal surveys. When necessary, mammography of the contralateral breast, abdominal ultrasonography and abdominal and brain computed tomography were carried out. DFS and overall survival (OS) were calculated as the periods from surgery until the date of the first recurrence or death caused by breast cancer spread. The mean follow-up period for 181 patients with no record of death was 11.0 years, ranging from 1.1 years to 15.0 years.
The histopathologic types of the primary tumors, re-evaluated according to the World Health Organization (WHO) criteria, comprised 215 (94.3%) invasive ductal carcinomas, 5 (2.1%) invasive lobular carcinomas, 4 (1.8%) invasive ductal carcinomas with predominantly intraductal components and 4 (1.8%) medullary carcinomas. All of the carcinomas were classified as grade 1, 2, or 3 according to a histological grading system (32 ) in which modifications were given to the WHO classification.
ER and PgR assays were performed using the dextran-coated charcoal technique (cutoff 10 fmol/mg protein). ER and PgR statuses were available from 136 patients and 110 patients respectively.
Immunohistochemistry
We examined routinely processed formalin-fixed and paraffin-embedded tissue blocks of 228 carcinomas, of which 53 were obtained by excisional biopsy and 175 were obtained by mastectomy. Cut sections 3µm thick were deparaffinized in xylene and rehydrated in graded alcohol. After blocking endogenous peroxidase activity with a methanol solution of 0.3% hydrogen peroxidase for 10 minutes, the slides were briefly rinsed in distilled water and treated in a microwave oven (H 2500 Microwave Processor, Bio-Rad, Hercules, CA) for 10 minutes at 90°C in 0.01 M citrate buffer (citric acid monohydrate, pH 6.0) or in 5% zinc sulfate heptahydrate to retrieve antigens of bcl-2 and p53. The slides, cooled down to room temperature in buffer for 30 minutes, were washed in phosphate-buffered saline (PBS). The primary antibodies used in the present study were: an anti-bcl-2 mouse monoclonal antibody (Dakopatts, Denmark), an anti-p53 rabbit polyclonal antibody (Rsp53, Nichirei Co., Tokyo, Japan) and an anti-c-erbB-2 rabbit polyclonal antibody (Nichirei). After preincubation in 2% normal swine serum in PBS, the sections were incubated overnight with the primary antibody. Biotinylated anti-IgG immunoglobulin (Vector Laboratories Inc., Burlingame, CA) at 1:200 dilution and avidin-biotin-peroxidase complex (Vector) at 1:100 dilution were added in sequence to the sections for 30 minutes each. The slides were then incubated in 3.3'-diaminobenzidine tetrahydrochloride and hydrogen peroxide as chromogens for 10 minutes and counterstained with hematoxylin by Mayer's method. The sections were washed three times with PBS between each step for 10 minutes. Negative controls were obtained by omitting the primary antibody. Normal ductal epithelial cells or lymphocytes were used as internal positive controls for bcl-2 expression. Breast cancer cases with a p53 mutation or c-erbB-2 amplification were also used as positive controls (23 ,31 ).
Compared with the intensity of immunoreaction in normal ductal epithelial cells and lymphocytes used as internal controls, the intensity of bcl-2 protein expression was divided into three categories: strong, an intensity similar to or greater than the internal controls; weak, an intensity inferior to that of internal controls; and negative, not stained. According to the staining intensity and the proportion of cancer cells showing positive immunoreactivity, we judged bcl-2 expression as strong positive (++) when 50% or more of cancer cells were strongly stained; weak positive (+) when >10% but <50% of cancer cells were strongly stained, or >10% of cancer cells were weakly stained; negative (-) when not stained or <10% of cancer cells were stained. p53 nuclear immunoreaction and c-erbB-2 protein expression were judged as positive (+) or overexpressed when >10% of cancer cells revealed distinct nuclear and membrane staining respectively. Otherwise, they were judged as negative (-).
Statistical Analysis
The association between bcl-2 expression and clinicopathologic variables was assessed by a [chi]2 test. DFS and OS curves starting from the date of surgery were computed by the Kaplan-Meier method (33 ) and the differences between the curves were analyzed by using the generalized Wilcoxon test (34 ). P values <= 0.05 were regarded as significant. Cox's proportional hazards model was used for multivariate analysis to test the independent significance of each parameter on prognosis (35 ). All statistical analyses were performed using the Statview 4.11 J package (Abacus Concepts Inc., Berkeley, CA) on a Macintosh personal computer.
RESULTS
Of the 228 breast carcinomas analyzed, 87 cases (38%) and 72 cases (32%) revealed strong and weak bcl-2 protein expression respectively. Forty-one tumors (18%) in our series showed nuclear accumulation of p53 protein and 32 tumors (14%) revealed overexpression of c-erbB-2 oncoprotein.
Table 1
| Parameter | No. of cases | bcl-2 expression (%) | P value | |
| ++ or + | - | |||
| Age (years) | ||||
| <50 | 118 | 90 (76) | 28 (24) | 0.04 |
| >= 50 | 110 | 69 (63) | 41 (37) | |
| Menopausal status | ||||
| Pre- | 120 | 92 (77) | 28 (23) | 0.03 |
| Post- | 108 | 67 (62) | 41 (38) | |
| Histologic grade | ||||
| 1 | 5 | 4 (80) | 1 (20) | <0.0001 |
| 2 | 87 | 80 (92) | 7 (8) | |
| 3 | 136 | 75 (55) | 61 (45) | |
| TNM stage | ||||
| I | 55 | 46 (84) | 9 (16) | 0.05 |
| II | 136 | 91 (67) | 45 (33) | |
| III | 37 | 22 (59) | 15 (41) | |
| No. of nodes | ||||
| 1 | 122 | 87 (71) | 35 (29) | 0.635 |
| 2 | 63 | 43 (68) | 20 (32) | |
| 3 | 43 | 29 (67) | 14 (33) | |
| Tumor size | ||||
| <= 2 cm | 96 | 76 (79) | 20 (21) | 0.02 |
| >2 cm | 132 | 83 (63) | 49 (37) | |
| ER status | ||||
| Positive | 70 | 60 (86) | 10 (14) | <0.0001 |
| Negative | 66 | 33 (50) | 33 (50) | |
| Unknown | 92 | 66 (72) | 26 (28) | |
| PgR status | ||||
| Positive | 25 | 22 (88) | 3 (12) | 0.07 |
| Negative | 95 | 61 (64) | 34 (36) | |
| Unknown | 108 | 76 (70) | 32 (30) | |
| Nuclear p53 accumulation | ||||
| Positive | 41 | 20 (49) | 21 (51) | 0.0007 |
| Negative | 187 | 139 (74) | 48 (26) | |
| c-erbB-2 overexpression | ||||
| Positive | 32 | 10 (31) | 22 (69) | <0.0001 |
| Negative | 196 | 149 (76) | 47 (24) | |
Table 2
| Variable | No. of cases | No. of recurrences | No. of deaths | ||
| n (%) | P value | n (%) | P value | ||
| bcl-2 expression | |||||
| ++ | 87 | 14 (16) | 9 (10) | ||
| + | 72 | 15 (21) | <0.05* | 11 (15) | <0.001* |
| - | 69 | 22 (32) | 21 (30) | ||
| p53 nuclear accumulation | |||||
| + | 41 | 16 (39) | <0.01 | 14 (34) | <0.001 |
| - | 187 | 35 (19) | 27 (14) | ||
| c-erbB-2 overexpression | |||||
| + | 32 | 12 (38) | <0.05 | 11 (34) | <0.01 |
| - | 196 | 39 (20) | 30 (15) | ||
| No. of metastatic lymph nodes | |||||
| 1 | 122 | 24 (20) | 15 (12) | ||
| 2 | 63 | 9 (14) | <0.001[dagger] | 11 (18) | <0.01[dagger] |
| 3 | 43 | 18 (42) | 15 (35) | ||
| Tumor size | |||||
| >= 2 cm | 13 (22) | <0.01 | 9 (9) | <0.01 | |
| >2 cm | 38 (29) | 32 (24) | |||
| Histologic grade | |||||
| 1 | 5 | 0 (0) | 0 (0) | ||
| 2 | 87 | 10 (12) | <0.01[dagger] | 8 (9) | <0.01[dagger] |
| 3 | 136 | 41 (30) | 33 (24) | ||
| TNM stage | |||||
| I | 55 | 7 (13) | <0.05[Dagger] | 5 (9) | <0.05[Dagger] |
| II | 136 | 31 (23) | 27 (20) | ||
| III | 37 | 13 (35) | 9 (24) | ||
| Age (years) | |||||
| <50 | 118 | 18 (15) | <0.01 | 12 (10) | <0.001 |
| >= 50 | 110 | 33 (30) | 29 (26) | ||
| Menopausal status | |||||
| Pre- | 120 | 22 (18) | 0.07 | 13 (11) | <0.01 |
| Post- | 108 | 29 (27) | 28 (26) | ||
| ER status | |||||
| + | 70 | 11 (16) | <0.01 | 9 (13) | <0.01 |
| - | 66 | 22 (33) | 20 (30) | ||
| PgR status | |||||
| + | 25 | 5 (20) | NS | 5 (20) | NS |
| - | 95 | 23 (24) | 21 (22) | ||
| Adjuvant therapy | |||||
| yes | 217 | 48 (22) | NS | 38 (18) | NS |
| no | 11 | 3 (27) | 3 (27) | ||
The immunohistochemical results of bcl-2 expression were associated with the histologic grade and the statuses of p53, c-erbB-2 and ER (Table 1 ). With regard to histologic grade, bcl-2 expression was positive in 80% (4 of 5) and 92% (80 of 87) of grade 1 and 2 groups respectively, whereas the positivity was only 55% (75 of 136) in grade 3 group (P < 0.0001). bcl-2 was positive in 60 of 70 (86%) tumors which were ER positive but in only 50% (33 of 66) of ER negative tumors (P < 0.0001). Furthermore, bcl-2 expression showed inverse associations with nuclear p53 accumulation (P = 0.0007) and c-erbB-2 overexpression (P < 0.0001).
Significant but weaker associations were also observed between bcl-2 expression and other variables, including younger age (<50 years, P = 0.04), premenopausal status (P = 0.03), smaller tumor size ( <= 2 cm, P = 0.02) and an earlier TNM stage (P = 0.051). However, there was no association of bcl-2 expression with PgR status or with the number of regional lymph nodes involved.
During the period of observation, 51 patients had recurrences and 41 patients died from breast cancer spread. Six patients died from other causes: two from cardiovascular diseases, two from cerebrovascular disease, one from gastric carcinoma with liver metastases and one from rectal carcinoma. These six cases were treated as censored. DFS and OS rates at 10 years after surgery in the total patient population were 77% and 82% respectively. There was no difference in recurrence or death rate between the groups with and without adjuvant therapies (Table 2 ). To avoid the variation which might be caused due to the effetiveness of adjuvant therapy, 11 patients who did not receive any adjuvant therapy were excluded from the survival analyses.
Univariate analysis revealed differences in both the DFS and OS curves between the bcl-2 (++) or (+) group and the bcl-2 (-) group (P < 0.05 and P < 0.001), between the groups with and without nuclear p53 immunoreactivity (P < 0.01 and P < 0.001) (Figs 1 , 2 ), and between the groups with and without c-erbB-2 overexpression (P < 0.05 and P < 0.01). The 15-year DFS and OS rates were respectively 82% and 87% in the bcl-2 (++) or (+) group compared with 67% and 69% in the bcl-2 (-) group; 59% and 66% in the p53 positive group, compared with 81% and 85% in the p53 negative group; 62% and 66% in the group with c-erbB-2 overexpression, compared with 81% and 85% in the group without c-erbB-2 overexpression. Combining bcl-2 with p53 expression, we found that bcl-2 enhanced the prognostic value of p53 (Fig. 3 ). The 139 patients who were bcl-2 positive, (++) or (+), and p53 negative showed a 15-year DFS rate of 85%, whereas the 21 patients with negative bcl-2 and positive p53 showed a 15-year DFS rate of only 55% (P = 0.0059).
DISCUSSION
The inhibition of apoptosis by the activation of the bcl-2 oncogene and the inactivation of the p53 tumor-suppressor gene in cells have suggested roles in tumor development. At the experimental level, the wild-type p53 protein has been shown to regulate apoptosis by transactivating the bax gene, which induces a decrease in bcl-2 expression and induction of apoptosis (36 ,37 ). In breast cancers, decrease in the expression of bcl-2 oncoprotein was shown to be correlated with the accumulation of p53 protein in cancer cell nuclei and both of these protein alterations are detected in 20-40% of cases. These alterations were also associated with a high histologic grade of carcinoma cells and negative ER status, i.e., highly aggressive or poorly differentiated tumors regardless of lymph node status. The nuclear accumulation of p53 protein is usually shown to be derived from a mutation of the p53 gene. If the association between the bcl-2 and p53 holds true for in vivo cases, nuclear accumulation of mutant p53 could play a role in the inhibition of apoptosis on the bcl-2-mediated pathway. However, in half of the breast cancer cases with p53 nuclear immunoreaction, bcl-2 protein expression was reduced. Histological observations have revealed that the number of apoptotic cells is larger in breast cancers with higher mitotic rate (38 ). Therefore, nuclear accumulation of mutant p53 does not appear to be effective in inhibiting apoptosis in breast cancers of high histologic grade. Conversely, it might be that the enhanced apoptosis mediated by the loss of bcl-2 expression is important in maintaining the rapid growth of the cancer cells with a high turnover rate. On the other hand, in low grade or well differentiated breast cancers, the bcl-2 protein was constantly expressed and appeared to be involved in the inhibition of apoptosis.
Conceptually, prognostic factors of n1-3 breast carcinoma could be classified into three categories: 1, the extent of local spread of cancer cells including regional lymph node metastasis, which is represented by the size of the primary tumor and the number of metastatic lymph nodes, visible microscopically; 2, the indicators of biological properties or aggressiveness of the cancer cells, which is represented by the histologic grade, expression levels of bcl-2, p53, and/or c-erbB-2 proteins and the ER status; 3, host factors, e.g., menstrual status and age. All of these factors were shown by univariate analysis to be indicators of patient prognosis. Even in this subgroup, which is restricted to n1-3 breast cancers, the number of metastatic nodes had independent prognostic value. The rates of recurrence and death in patient groups with metastases to one or two axillary lymph nodes were 20% or lower at 15 years after surgery, whereas the rates were 42% and 35% respectively in the group with metastases to three lymph nodes. It is important to predict the prognosis of breast cancer patients as accurately and comprehensively as possible by assessing and relating multiple categories. For example, the combination of the number of metastatic lymph nodes, histologic grade, bcl-2 expression and p53 expression would be helpful for identifying the group of n1-3 breast cancer patients at high risk of recurrence and death. In addition, our univariate analysis results suggested that we can identify a low-risk group. For example, the 15-year DFS rate was 84% or higher and the 15-year OS rate was 90% or higher in the groups of grade 1 and 2, stage I, and <50 years old with a high level expression of bcl-2 protein. In such low-risk groups, it appears to be worth examining the validity of not performing adjuvant therapy as an arm of choice. Although the number of cases was not sufficient, there was no significant difference in the outcomes between 217 patients who received adjuvant therapy and 11 without adjuvant therapy.
Multivariate analysis showed that the number of lymph node metastases, menopausal status, p53 and bcl-2 were all independent prognostic predictors for OS, whereas only histologic grade and the number of nodes were significant predictors for DFS. Because bcl-2, p53, c-erbB-2 and the histological grade are tightly correlated, the prognostic impact of some of these parameters is absorbed in the effect of the others in multivariate analysis. The present multivariate analysis data are therefore not absolute, and we consider that each of these factors indicating cancer cell aggressiveness is nearly equally effective in predicting the prognosis for n1-3 breast cancer patients.
In summary, our study showed that the number of metastatic lymph nodes, tumor size, age, histological grade, bcl-2, p53, c-erbB-2 and ER are significant prognostic factors for n1-3 breast cancer. The combination of these factors, for example bcl-2 and p53, can identify the subgroups at low or high risk of relapse.
Acknowledgements
This research was supported by Grants-in-Aid for Cancer Research and for the 2nd-term Comprehensive 10-year Strategy for Cancer Control from the Ministry of Health and Welfare in Japan. The authors thank Dr Rikiya Abe, Dr Atsuo Tsuchiya and Dr Izo Kimijima, Department of Surgery II, Fukushima Medical College, for their helpful comments and suggestions. We also thank Ms Y. Yamauchi for cutting tissue blocks and Dr N. Katsumata for statistical advice.
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 May 1998
Copyright© Japanese Journal of Clinical Oncology, 1997.
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 May 1998
Copyright© Japanese Journal of Clinical Oncology, 1997.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Khoury, M. A. Naujokas, D. Zuo, V. Sangwan, M. M. Frigault, S. Petkiewicz, D. L. Dankort, W. J. Muller, and M. Park HGF Converts ErbB2/Neu Epithelial Morphogenesis to Cell Invasion Mol. Biol. Cell, February 1, 2005; 16(2): 550 - 561. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-J. Zhang, I. Kimijima, M. Onda, M. Kanno, H. Sato, T. Watanabe, A. Tsuchiya, R. Abe, and S. Takenoshita Tamoxifen-induced Apoptosis in Breast Cancer Cells Relates to Down-Regulation of bcl-2, but not bax and bcl-XL, without Alteration of p53 Protein Levels Clin. Cancer Res., October 1, 1999; 5(10): 2971 - 2977. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||