1. Background
Breast cancer is the most common cancer among women worldwide. A significant fraction of patients often die of metastatic disease. About 519000 women died in 2004 due to breast cancer. Although breast cancer is thought to be a disease of the developed world, a high percentage (69%) of all breast cancer deaths occurs in developing countries (1, 2). Breast cancer at early stages does not show any symptoms (1). This feature emphasizes the importance of regular breast exams. As the cancer grows, symptoms may include: breast lump, change in the size, shape or feel of the breast or nipple and fluid secretion from the nipple (3-6).
Breast cancer prognostic factors include axillary nodal status, clinical stage, size and grade of the tumor, hormone receptor status, and presence of lymphovascular involvement (7). Involvement of the regional lymph nodes is a major predictive factor of metastatic disease. Adjuvant therapies such as chemotherapy, radiotherapy, hormone therapy, and monoclonal antibodies reduce the incidence of metastasis greatly (8-10).
A predictive factor could be defined as any measurement related to the response to any given therapy. Prognostic factors play a key role in optimizing treatment for breast cancer patients as it leads to general use of adjuvant therapy (11). Estrogen (ER) and progesterone (PR) receptors, and human epidermal growth factor receptor 2 (HER2) are definitely listed as both prognostic and predictive factors (12-20). Hormone therapy and monoclonal antibodies are performed based on the results of immunohistochemistry (IHC) and / or immunofluorescence, which reveal corresponding receptors (estrogen, progesterone and Her2) in tumors (21-26). Although mentioned adjuvant therapies would not be routinely applied on metastatic lymph nodes, mismatch (qualitative and quantitative) of these receptors between the primary tumor and metastatic lymph nodes could result in secondary resistance to adjuvant treatments and occurrence of metastatic disease.
Even though several studies have noted the discrepancy between the immunohistochemistry findings of primary breast tumor and its metastases (27), there still is lack of enough studies comparing the results of primary breast tumor and lymphadenopathy immunohistochemistry. It seems that the receptor status of metastatic lymph nodes would be better criteria for administering adjuvant therapies. Also due to lack of sufficient studies in this field, we studied the qualitative evaluation of ER, PR and Her2 receptors in metastatic lymph nodes of breast cancer patients in Omid hospital of Mashhad from 2005 to 2010.
2. Methods
2.1. Patients
It was a cross sectional study and all the patients with involved regional lymph nodes (N +) were selected and a special check list was filled for them. We evaluated the health records of all the breast cancer patients who referred to our teaching hospital during five years (2005 - 2010).
Sampling method was simple non-possible, and sample size was calculated as 50 based on Mcnemar test and PASS software. Inclusion criteria were: all breast cancer patients who had axillary lymphadenopathy and Immunohistochemical process were performed on their tumors from 2005 to 2010. Exclusion criteria were: 1) unavailability of lymph node pathology samples, and 2) technical errors in staining procedures of lymph node pathology sample.
2.2. Method
Pathology samples of involved nodes were collected from the related laboratories and were assessed by Immunohistochemical markers (Skan Teb Asia) for ER, PR and Her2. The findings were compared with Immunohistochemical results of primary tumors and each change in percentage and staining was considered as discrepancy. We performed Immunohistochemistry for all the lymph node samples of patients upon whom IHC was done previously in the same center and with the same technique to prevent possible mistakes, comparing to the lymph node and tumor IHCs performed by one pathologist.
The interpretation of IHC findings was semi-qualitative, based on Red-score (from 1 to 8) for ER and PR, and each change from mentioned score in percentage or in staining rate was considered as discrepancy. IHC interpretation of Her2 was based on staining rate of cell membrane and it ranged from 0 to +3. All +3 cases were considered positive, 0 and +1cases were considered negative, and +2 cases were considered controversy and candidate for the Fish test. Finally, the gained findings were compared with primary tumor results.
2.3. Data Collection
The method of gathering data was library and check list form was the tool used in this study. Obtained data were analyzed by using SPSS 16.0 and Chi-Square MCnemar statistical test was performed in significant level of 0.05.
2.4. Ethical Consideration
We obtained consent from hospital officials and patients who participated in this study. Also patients were ensured about confidentiality of their recorded information and privacy. The code of the ethical consideration according to our office is IR.MUMS.REC.1390,40.
3. Results
The youngest patient was 27 years and the oldest was 77 years, and the mean age of patients was 48.02 years. Average size of tumor was 4.798 cm (Table 1).
Table 1.Age and Tumor Size in Patients with Breast Cancer
| Variables | No. | Minimum | Maximum | Mean | Standard Deviation |
|---|---|---|---|---|---|
| Age | 49 | 27 | 77 | 48.02 | 12.592 |
| Size | 46 | 1.0 | 17.0 | 4.798 | 3.5321 |
Primary metastasis was positive for progesterone (PR) in 26 patients, for estrogen (ER) in 25 individuals, and for Her2 in 49 patients. Wilcoxon signed-rank test showed that there were significant differences between IHCs of primary tumor and metastasis for all three PR (P < 0.001), ER (P < 0.001) and Her2 (P = 0.003) (Table 2).
Table 2.IHCs of PRIMARY TUMOr and METASTasis in PR, ER and Her2 (Wilcoxon Signed Ranks Test)a
| IHC | Positive | Negative | P Value, Wilcoxon Signed Ranks Test |
|---|---|---|---|
| PR | 26 (52) | 24 (48) | 0.000 |
| ER | 25 (50) | 25 (50) | 0.000 |
| Her2 | 49 (100) | - | 0.003 |
aValues are expressed as No. (%).
The comparison of primary tumor and lymphadenopathy immunohistochemistry showed that there was a significant difference in ER (P = 0.003) between them. ER was positive for primary tumor in 25 (50%) of the patients and it was positive for metastasis lymphadenopathy in 39 (78%) of individuals. Also chi-square test showed that there was a significant difference in PR between primary tumor and metastasis (P = 0.011). Positive PR was observed in 26 (52%) patients for primary tumor and it was seen in metastasis lymphadenopathy of 38 (76%) of patients (Table 3).
Table 3.IHCs of Primary Tumor and Metastasis in PR, ER (Chi-Square Test)a
| IHC | Primary Tumor | Metastasis Lymphadenopathy | P Value, Wilcoxon Signed Ranks Test | ||
|---|---|---|---|---|---|
| Positive | Negative | Positive | Negative | ||
| PR | 26 (52) | 24 (48) | 38 (76) | 12 (24) | 0.011 |
| ER | 25 (50) | 25 (50) | 39 (78) | 11 (22) | 0.003 |
aValues are expressed as No. (%).
We compared the patients who had similar Her2 results in primary sample and metastasis with patients who had different findings for grade variables, LVST and, LN1-cod. Kruskal-Wallis test showed that there is no significant difference between these two groups (P > 0.05). Also we performed a comparison between patients with similar PR and ER results of primary tumor and metastatic lymphadenopathy, and patients with dissimilar findings. Kolmogorow-Smirnow-Test showed that there was no significant difference between these mentioned groups in ER and PR (P > 0.05).
T-test indicated that there was not any association between the above mentioned groups (those that had same results in tumor sample and metastatic lymphadenopathy, and those with different results) with variables such as age, tumor size, number of removed nodes and number of involved nodes. Also there was no significant difference in ER, PR and Her2 (P > 0.05). Twenty-four, seventeen and nine patients were pN1, pN2 and pN3, respectively.
Cohen’s kappa coefficient was used to determine the association. Discrepancies of immunohistochemical ER, PR and Her2 results in primary tumor and lymphadenopathy were 24% (P = 0.000), 32% (P = 0.002), and 48%, respectively. Overall, a total of 34 patients (68%) had discrepancy for one of the receptors (Table 4).
Table 4.Discrepancies of Immunohistochemical ER, PR and Her2 Results in Primary Tumor and Lymphadenopathy
| IHC | Tumor | Lymph Node | No. (%) | Kappa Coefficient |
|---|---|---|---|---|
| ER | - | - | 12 (24) | 0.51* |
| - | + | 12 (24) | ||
| + | - | 0 (0) | ||
| + | + | 26 (52) | ||
| PR | - | - | 10 (20) | 0.36* |
| - | + | 15 (30) | ||
| + | - | 1 (2) | ||
| + | + | 24 (48) | ||
| Her2 | + | + | 13 (26.5) | 0.253* |
| + | ++ | 11 (22.4) | ||
| + | +++ | 3 (6.1) | ||
| ++ | + | 4 (8.2) | ||
| ++ | ++ | 6 (12.2) | ||
| ++ | +++ | 5 (10.2) | ||
| +++ | + | 0 (0) | ||
| +++ | ++ | 1 (2) | ||
| +++ | +++ | 6 (12.2) |
4. Discussion
We evaluated the pathology samples of fifty patients who suffered from breast cancer.
All regional lymphadenopathy samples were tested by IHC for ER, PR and Her2 markers and finally we compared them with IHC findings of related primary tumors.
Atiken et al. (2009) studied the difference in expression of ER, PR and Her2 receptors between primary tumors and lymph nodes via immunofluorescence (28). Approximately in fifty percent of patients’ samples of breast tumors and related nodes differed at least for one receptor. They claimed that considering lymph node receptor could be a more precise method for adjuvant treatment. Correspondingly, we showed that the difference between IHC of primary tumor and IMLNs for ER, PR and Her2 was 32% (P = 0.000), 24% (P = 0.002) and 48%, respectively.
Nedergaard assessed the ER receptor expression between primary tumor and metastatic lymph node (29). Discrepancy was observed for 21% of patients. They concluded that it was because of lack of ER receptors in metastatic cells and it could justify the failure of hormone therapy.
Cardoso et al. (2001) performed a similar study for the assessment of predictive biological makers (30). They showed that there is no marker with 100% association with both samples. They studied primary tumor and regional lymph nodes in the axillary area of the patients with breast cancer. For each marker, the percentage of stained cells (fatality) and its severity were assessed. IHC assessment was performed by using monoclonal antibodies against topo Π-alpha, Hsp27 (heat shock protein) HSP 70, HER2, p53 and bcl2.
Dissimilarly, Cho et al. (2008) showed that discrepancy was less than what we gained for expression of Her2 receptor between primary tumor and lymph node. They studied the marker expression status of HER 2, EGFR in primary tumor and lymph node metastasis in CYCLIN-D1 region (axillary) using IHC and CISH in 73 patients (31).
Although many studies indicate that Her2 and other hormone receptors play effective role in patient management therapy as we showed (32-34), there are researchers who believe that receptor status in recurrent tumors does not pose predictable value based on the analysis of hormone receptors in primary stage (27, 35, 36).
We showed significant differences between IHC findings of biomarkers in lymphadenopathy of primary tumor and metastatic node (P < 0.05). This could change the future of adjuvant treatments and the choice of best method for patients with cancer.
4.1. Conclusion
While the receptor status of metastatic lymph nodes would be better for selection of adjuvant therapies and there are not enough studies comparing the results of primary breast tumor and lymphadenopathy IHC, we decided to evaluate discrepancy between IHC results of primary tumor and ipsilateral metastatic lymph nodes (IMLNs). In our study it was significant. Since metastatic clones in MLNs are potential sources of systemic metastasis, routine IHC on MLNs could play an important role in determining prognosis, indication for FISH and finally choosing adjuvant treatment. The youngest patient was 27 years and the oldest was 77 years; mean age of patients was 48.02 years. Average size of tumor was 4.798 cm (Table 1).
