A common first route of spread for breast carcinoma is through the axillary lymph nodes, and the incidence of ALNM increases with larger tumors. Nodal status is the most powerful independent prognostic factor in breast cancer and remains the most important feature for defining risk category. There is evidence that overall survival decreases as the number of positive node increases [4, 29]. According to St. Gallen experts, involvement of four or more nodes in the axilla by itself indicated high-risk, but patients with one to three nodes involved required HER2/neu overexpression or amplification to be included in the high-risk group, with other patients with one to three nodes included in the intermediate-risk category . Attempts have been made to identify factors that may predict an increase risk of nodal involvement in this group of patients. Although nodal micrometastases were prognostically relevant in several studies [31, 32], neither they nor isolated tumor cells in lymph nodes are considered in risk allocation. Increasing size of the tumor has been found to be predictive of ALNM. Even patients with T1a and T1b disease have significant nodal involvement (5-15%) [33–36]. This study found a 29% overall incidence of ALNM in patients with tumors ≤20 mm. There was no difference in the distribution of ALNM frequency among the different histologic types.
Multicentricity has been evaluated as a potential predictive factor of ALNM. When a combined diameter assessment was used, the frequency of lymph node positivity was not significantly different in multifocal versus unifocal cases [37, 38]. However, multicentric and multifocal breast cancer is associated with increased nodal involvement compared to similar unifocal disease and the tendency of breast tumors to metastasize is a reflection of the total tumor load . Classification of patients with breast cancer based on the size of the dominant lesion, without taking into account multicentricity, may not accurately reflect the risk of ALNM in patients with small, screen-detected cancers. This may, in turn lead to over- or under-treatment of some of these patients. Univariate analysis of this study identified multicentricity as a significant factor for ALNM. The number of foci should be considered as an independent prognostic parameter, which is currently not reflected in the TNM classification. Therefore, cases with multicentric tumours should be analyzed separately and should be included in the risk assessment by re-evaluating the current TNM classification.
In univariate analysis, there was a positive relation between nodal involvement and EIC. It is possible that EIC sometimes was associated with unrecognized multifocality of infiltrating cancer predisposing to nodal spread. Further research is required to explore this and other possible explanations.
Lymphovascular invasion has been proved to be the strongest independent predictor of nodal involvement in two series, and the grading system for lymph-vessel tumor embolus is a very useful histological grading system for accurately predicting lymph node metastasis by IDCs [39, 40]. In this study, both univariate and multivariate analysis identified lymphovascular invasion as significant predictors for ALN involvement.
Despite controversial data available on the value of age as a prognostic factor, the prognosis of breast cancer in very young women is generally considered to be unfavorable. Age has been found to be an independent prognostic factor in the multivariate analysis with women aged 35 years or younger having a shorter loco-regional recurrence-free distant relapse and shorter overall survival. When the data was matched for stage and lymph node status, patients ≤35 years of age continued to show a poorer 10-year distant relapse free survival . Ten-year disease free survival (DFS) and overall survival were worse in younger than in older (≥35 years) patients. Of interest, younger patients with ER positive tumors had a poorer DFS than patients with ER negative tumors. In contrast, among older patients the DFS was similar irrespective of ER status. Saghir et al. showed that young age had a negative impact on the survival of patients with positive axillary lymph nodes and positive hormonal receptors . According to Colleoni and colleagues, compared with less young, very young patients with endocrine responsive and node-negative breast cancer have a worse prognosis . Young patients tend to have larger tumor sizes, more positive lymph nodes, more negative hormone receptors, higher tumor grades than their older counterparts [4, 11, 14, 18, 26, 29]. The issue remains controversial and not all studies reported age as an adverse prognostic factor [18–21, 26, 27, 29]. In line with these findings, although the significance is marginal, this study identified young age as a significant predictor of ALNM. This negative predictive effect of age on nodal involvement is consistent with earlier evidence and indicates that breast tumours can be more aggressive in younger women [44–46]. Alternatively, this may be due to the longer period of exposure to the screening service of older women, when compared to the younger women; thus reflecting a cumulative protective effect of repeated screening.
The potential poor survival or early recurrence associated with CK5/6 and CK17 expression in tumor cells was first reported by Dairkee et al. in 1987 . Although IHC-based assays do not provide so much biological insight into tumor biology as mRNA-based assays that include thousands of genes, this IHC assay allowed classification of tumors into categories based on the associations between intrinsic subtypes and proliferation rates, overall survival, TP53 status, and BRCA1 mutation status [8–10, 15, 48]. In order to facilitate the investigation of the ALNM frequencies in basal-like breast cancer subtype, this study used a refined an IHC-based assay. Basal markers are not routinely used in the standard histological diagnosis of breast cancer. As existing prognostic markers do not identify this group, patients with basal-like and non-basal-like tumours are currently treated similarly. In this study, when all tumors were classified into five groups based on ER/PR/HER2 and basal-cytokeratin expressions, ALNM rate was lower in only basal-like phenotype. In addition, this study identified lack of CK5 expression as an independent predictor of ALNM, with tumors expressing CK5 bearing a significantly smaller risk (adjusted OR = 0.003, 95% CI 0.000-0.23, p = 0.009).
Marked differences in the extent of lymph node involvement, multicentric/multifocal disease, lymphovascular invasion (LVI), and extensive intraductal component were observed among subtypes. Only the basal subtype was found to differ significantly from the luminal A subtype with regard to the risk of nodal metastasis. This subtype was a significant predictor of having involvement of four or more nodes on multivariate analysis. However, some studies have shown in patients without ALNM that expression of basal CK was associated with a poor prognosis [15, 49]. New prognostic markers are quite important for this group of patients since the prognosis for node-negative patients is less clear, and the clinical decision to give or withhold systemic therapy is difficult, depending only on tumor size and grade. CK5 expression appears to be a useful marker to define the group of breast tumors with a poor prognosis even in the absence of ALNM.
Mutations in the tumor suppressor gene, p53, are present in 18-25% of primary breast carcinomas [50, 51]. A previous study found significant association between anti-p53 antibodies and tumor size, histological grade, and the number of axillary lymph nodes involved . TN breast cancers more frequently show p53 nuclear expression; therefore, are likely to harbour TP53 gene mutations more frequently . In this study, patient with P53 expression were present in 49% HER2 subtype, and in 41% basal-like phenotype. However, p53 expression did not emerge as a significant factor associated with ALNM. Expression of p53 could provide information concerning a poor outcome in triple-negative breast cancer. There is no definite answer to optimal management of triple-negative tumors at this moment. In such cases, consideration might well be given to more aggressive or alternative treatment.
Proliferative activity of tumour cells assessed by immunohistochemical Ki-67 expression is one of several prognostic indicators in breast cancer. Intriguingly, all the studies [54–56] have shown a statistical correlation with clinical outcome irrespective of cut-off points. Previous studies have reported significant associations between high Ki-67 index and lymph node status . In contrast, similar to the findings of this study, a recent study have shown that Ki-67 did not appear to be a helpful predictor .
In this study, univariate analysis identified significant associations between ALNM and twelve factors: age, multicentric disease, tumor size, vascular and lymphatic invasion, epithelial hyperplasia, necrosis, in situ carcinoma, perineural invasion, basal-like phenotype, and CK5, CK14 and EGFR expressions. However, only age, CK5 expression and lymphatic/vascular invasion remained to be significant predictors on multivariate analysis. Increasing age and CK5 expression was associated with decreased risk whereas lymphatic and vascular invasion was associated with an increased risk.
Clinical assessment of the axilla, tumor palpability and the method of detection were not uniformly documented in our data set, thus were not included in the analysis. This represents a limitation of this study. Another potential weakness of this report may be that ALN status we based on the results from a mixture of complete axillary node dissection and sentinel lymph node procedure. However, it has previously been shown that predictors for ALN status are independent of how the lymph node resection was performed although the metastatic detection rate in lymph nodes may be higher using the sentinel node procedure since higher number of metastases has been detected thorough histological examination of the sentinel lymph node biopsies . In general, patients can be selected for breast-conserving surgery with a high degree of accuracy by history, physical examination, and diagnostic mammography . This study identified age, CK5 expression and vascular/lymphatic invasion as three variables associated with high-risk for ALNM. The latter two can be identified preoperatively with reasonable accuracy by pathologic evaluation of core needle biopsies . As a result, pathologists can accurately assess the true malignant potential of IDCs by seeking lymph-vessel tumor emboli as part of a histological prognostic classification. In addition, patients with basal-like carcinomas may be less sensitive to standard adjuvant chemotherapy than with other types; necessitating novel therapeutic approaches.