In breast carcinoma, axillary lymph node metastasis is considered the most important prognostic factor (Carter et al., 1989). Methods used to measure the metastatic burden have varied with time. Initially, Berg (1955) classified axillary nodes into three levels, namely, level I below and lateral to pectoralis minor, level II behind, and level III above and medial to the muscle. He also reported that metastases progress in a sequential manner from level I to III, with worsening of prognosis. Carter et al. (1989) considered the number of lymph nodes as a better indicator of metastatic burden and this system was adopted by the WHO pN staging classification. Recently, Vinh-Hung et al. (2009) have reported that the ratio of positive nodes to the total nodes excised [lymph node ratio (LNR)] is a more accurate predictor of survival.
There is a considerable disagreement among surgeons on the extent of lymphadenectomy in breast cancer. In the past, all three levels of axillary lymph nodes were removed in the standard radical mastectomy operation (Halsted, 1895). However, in recent years, more conservative lymphadenectomies have been performed in view of early detection of tumors and widespread use of adjuvant therapy. The current practice in developed countries is to remove nodal levels I and II only (Silberman et al., 2000). The standard recommended minimal number of nodes to be removed in axillary dissection is 10 nodes (Salama et al., 2005), but varies in the literature between 16 andsix (Tavassoéli and Devilee, 2003; Somner et al., 2004). Finally, no lymphadenectomy is performed if the sentinel lymph node is negative (Giuliano et al., 1994). Although this limited lymphadenectomy is acceptable in developed countries in patients presenting with tumors early (Veronesi et al., 1987), it may not be applicable in developing countries because of the late presentation of patients with more extensive lymph node involvement (Nouh et al., 2004). In Egyptian patients, few studies are available on the extent of metastases as well as the distribution of lymph node metastases by levels (Zaghloul et al., 2001; Khafagy et al., 2011).
The aim of the present investigation is to quantify the anatomic distribution of axillary lymph nodes by levels as well as to compare the metastatic burden of breast cancer at different nodal levels as determined by the number of positive nodes as well as the positive LNR. A hypothetical model of incomplete lymphadenectomy will be tested to determine the risk of such operations if performed in Egyptian patients.
Patients and methods
All patients were private patients treated by a single surgeon (M.M.K.) to minimize operator-dependent variations and to ensure maximal retrieval of lymph nodes. Informed consent was obtained from all patients. The study included 110 patients treated during the years 2008–2013. Their clinical and pathological characteristics are presented in Table 1.
In all operations, axillary dissection was complete including the three levels of nodes (Berg, 1955), which were labeled by silk sutures. The interpectoral (Rotter) lymph nodes were separately submitted in a labeled bottle and were evaluated apart from the levels.
The total series (110 patients) was used to study the anatomic distribution of axillary nodes at different levels as well as to study the rate of metastases at different levels. However, cases with positive nodes (56 patients) were used to determine the metastatic burden by two methods, namely, the number of positive nodes in each level (pN) as well as the LNR. The LNR was determined for each level by dividing the number of positive nodes by the total number of excised nodes at that level (Vinh-Hung et al., 2009). In addition, in lymph node-positive cases, the extent of metastases was typed histologically following the criteria of WHO and Veronesi et al. (1987). Micrometastases (<1 mm) were measured using a stage micrometer.
For the statistical evaluation, results were expressed as means and SDs in normally distributed data or as medians and interquartile ranges in abnormally distributed data. χ2, Kruskal–Wallis, and Mann–Whitney tests were used to test for significance. Statistical package for social sciences version 18 was used for analysis (SPSS Inc., IBM, Chicago).
The characteristics of the patients in the present series are outlined in Table 1. The mean age of the patients was 55±12 years. Most of the patients (63.6%) were postmenopausal (>50 years). The average diameter of the tumors was 2.8±1.7 and 63.3% were more than 2 cm in size (>T1). Invasive duct carcinoma (NOS) predominated (73.6%). Of the total of 110 patients, 54 (49.1%) were negative for metastases (N0), 27 (25.5%) were N1 (1–3), 11 (10%) were N2 (4–9), and 18 (16.4%) were N3 (>9).
The anatomic distribution of axillary lymph nodes at different levels is presented in (Table 2). The total number of lymph nodes removed in the 110 cases was 2463 nodes. There were 1196 (48.6%) nodes at the first level, 919 (37.3%) at the second level, and 348 (14.1%) at the third level. The difference in nodal distribution at various levels was statically significant (P<0.001). The majority of axillary lymph nodes (2115, 85.9%) were located at levels I and II. The median number of axillary lymph nodes removed per patient was 21 nodes. The median number of lymph nodes was 12 at the first level, 9 at the second level, and 4 at the third level. The median number of nodes at different axillary levels is presented in Table 2 and the difference among levels was statistically significant (P<0.001).
The rates of axillary node metastases in patients as well as at different levels were studied in the entire series of 110 patients including patients with negative nodes. In this series, 56 patients had lymph node metastases, a rate of 50.9%. The rates of node metastases at axillary levels I, II, and III were 50.9, 34.5, and 20%, respectively. About half of the patients had positive nodes at level I at presentation. We also used the entire series to determine the risk of leaving residual positive nodes if hypothetical incomplete lymphadenectomies were performed (Table 3). As shown, hypothetical lymphadenectomy of levels I and II would only leave residual metastases at level III in 22 patients (20%).
The metastatic burden, which describes the intensity of metastases, was determined in each axillary level of 56 patients with positive nodes. It was calculated using two methods, namely, the absolute number of nodes with metastases (pN) and the metastatic LNR.
A total of 2463 nodes were studied. Metastases were detected in 447 nodes. Skip metastases were observed in five patients (4.5%), in whom there were level II metastases without level I affection. The median number of metastatic lymph nodes in node-positive cases was 4, whereas the median numbers per level were 3 for level I, 4 for level II, and 3 for level III. The median LNR for positive patients was 0.18, whereas the median LNRs per level were 0.3 for level I, 0.5 for level II, and 1 for level III (Table 4). These differences in LNR between levels were statistically significant (P<0.001). Most of the node-positive patients (55.4%), according to LNR, were considered to be at low risk (≤0.2), whereas 28.6% were at an intermediate risk (0.2–0.65) and 16% were at a high risk (>0.65%). As shown in Table 4, the highest metastatic burden was observed at level I, followed by levels II and III in that order of frequency and the difference in metastatic burden was statistically significant (P<0.001).
The histologic types of metastases observed in the present study are presented in Table 5 and Figs 1–3. The majority of patients presented with macrometastases and extracapsular spread (23.2 and 75%, respectively). However, micrometastases were rarely encountered; it was observed only in one patient (1.8%).
In the present series, Rotter (interpectoral) lymph nodes were present in 21 (19.1%) of the total of 110 patients studied. The mean number of the excised nodes was 1.6 (range of 1–2). Metastases were present only in four patients (3.6%).
The frequency of multiple level affection in this series is presented in Table 6. The risk of apical node involvement related to positivity of lower nodal levels is shown in Table 7.
In Egyptian patients, few studies have been carried out on axillary metastases in breast cancer. In two previous reports (Nouh et al., 2004; Alieldin et al., 2014), the main aim was to determine the frequency of node positivity without any evaluation of their level distribution. In three other reports (Zaghloul et al., 2001; Khafagy et al., 2007, 2011), the metastatic burden was determined at different levels as measured by the number of positive nodes (pN). To our knowledge, the present report is the first to quantify the anatomic distribution of axillary nodes as well as to determine the metastatic burden by LNR at different levels. Such a ratio is considered to be superior for the measurement of metastatic burden (Vinh-Hung et al., 2009; Elkhodary et al., 2014). In our study, the number of nodes removed was optimal (>10 nodes in all patients, with a mean value of 22.4 nodes).
All previous studies on axillary node metastases by levels were restricted to positive patients (Veronesi et al., 1987; Khafagy et al., 2011; Elkhodary et al., 2014). In the present report, we included patients with negative nodes. This provided us with a large material to study the anatomic distribution of lymph nodes; moreover, we could measure the rate of node positivity in the entire series, which reflects the actual situation of patients at presentation. Finally, the inclusion of node-negative patients enabled us to develop a hypothetical lymphadenectomy model to determine the risk of residual disease in case of incomplete lymphadenectomy in Egyptian patients.
The age of the patients and tumor stages in the present study were different from those in other reports from National Cancer institute (NCI), Cairo (Nouh et al., 2004), as well as western series (Jatoi, 1999). Thus, in our series, the mean age of the patients was 55 years, with a predominance of the postmenopausal age group (>50 years), 63.6%. In the largest Egyptian series (3747 patients) reported by Nouh et al. (2004), the premenopausal age group (<50 years) predominated (58.9%). Also, tumors tended to be operated at earlier stages in the present series, as T1 represented 38.3%. Conversely, in the series of Zaghloul et al. (2001), T1 represented only 2%. The rate of metastases was 50.9% in this series, which is lower than previous reports on Egyptian patients from the NCI, in which the rate of metastases varied between 70.6% (Nouh et al., 2004) and 80% (Zaghloul et al., 2001). This difference is probably because of differences in the socioeconomic status of the patients between series. The present seriesincluded entirely private patients, whereas NCI is mainly a free hospital draining a lower socioeconomic group of patients.
Compared with western reports, the present series showed a comparable predominance of postmenopausal patients, but the rate of lymph node metastases (50.9%) was much higher than that reported in the West (25–40%) (Thomssen et al., 1998; Jatoi, 1999).
The average number of axillary lymph nodes retrieved in the present study was 22.4 node per patient, which is comparable with the result of Veronesi et al. (1987) (20.5 nodes) and Khafagy et al. (2011) (24.1 nodes). The anatomic distribution of the total nodes of all patients in the axillary levels in our study was 48.6% in level I, 37.3% in level II, and 14.1% in level III, which is comparable with Berg (1955), who reported a nodal distribution of 45% in level I, 35% in level II, and 20% in level III.
In the total series, the relative frequencies of metastases at different axillary levels were 53.2% in level I, 33.6% in level II, and 13.2% in level III. Thus, the present study confirms that level I includes most of the metastatic nodes. This result is similar to the level distribution reported by Berg (1955), namely, 60% in level I, 31% in level II, and 9% in level III.
In our study, we adopted the three risk group stratifications of LNR and the cut-off values reported by Vinh-Hung et al. (2009), namely, high risk (>0.65%), intermediate risk (0.2–0.65),and low risk (≤0.2). According to this risk stratification by LNR, most of the node-positive patients (55.4%) were at a low risk (≤0.2), whereas 28.6% were at an intermediate risk (0.2–0.65) and 16% were at a high risk (>0.65%). Our results are comparable with those of Vinh-Hung et al. (2009), who reported 5% of patients to be at a low risk, 30.2% at an intermediate risk, and 13.8% at a high risk.
It can be concluded from this study that Egyptian patients with operable breast cancer present at a late stage (63.6% of tumors are >T1), and 50.9% have positive nodes. Axillary nodes are commonly located (85.9%) at levels I and II and most metastases (86.8%) affect these two levels. However, level III also involves metastases in 20% of patients at presentation, hence the importance of performing complete axillary lymphadenectomy in Egyptian patients.
Conflicts of interest
There are no conflicts of interest.
Alieldin NH, Abo-Elazm OM, Bilal D, Salem SE, Gouda E, Elmongy M, Ibrahim AS (2014). Age at diagnosis in women with non-metastatic breast cancer: Is it related to prognosis? J Egypt Natl Canc Inst 26:23–30.
Berg JW (1955). The significance of axillary node levels in the study of breast carcinoma. Cancer 8:776–778.
Carter CL, Allen C, Henson DE (1989). Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases.. Cancer 63:181–187.
Elkhodary TR, Ebrahim MA, Hatata EE, Niazy NA (2014). Prognostic value of lymph node ratio in node-positive breast cancer in Egyptian patients. J Egypt Natl Canc Inst 26:31–35.
Giuliano AE, Kirgan DM, Guenther JM, Morton DL (1994). Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 220:391–398discussion 398–401.
Halsted W (1895). The results of operations for the cure of cancer of breast performed at the Johns Hopkins Hospital from June 1889 to January 1894. Johns Hopkins Hosp Bull 4:297.
Jatoi I (1999). Management of the axilla in primary breast cancer. Surg Clin North Am 79:1061–1073.
Khafagy M, Mostafa A, Fakhr I (2011). Distribution of axillary lymph node metastases in different levels and groups in breast cancer, a pathological study. J Egypt Natl Canc Inst 23:25–30.
Khafagy MM, Kairy T, Shalaan M (2007). Forgotten sites of axillary dissection in breast cancer patient. Kasr El Aini J Surg 8:111–122.
Nouh MA, Ismail H, El-Din NH, El-Bolkainy MN (2004). Lymph node metastasis in breast carcinoma: clinicopathological correlations in 3747 patients. J Egypt Natl Canc Inst 16:50–56.
Salama JK, Heimann R, Lin F, Mehta N, Chmura SJ, Singh R, Kao J (2005). Does the number of lymph nodes examined in patients with lymph node-negative breast carcinoma have prognostic significance? Cancer 103:664–671.
Silberman AW, Sarna GP, Palmer D (2000). Adjuvant radiation trials for high-risk breast cancer patients: adequacy of lymphadenectomy. Ann Surg Oncol 7:357–360.
Somner JE, Dixon JM, Thomas JS (2004). Node retrieval in axillary lymph node dissections: recommendations for minimum numbers to be confident about node negative status. J Clin Pathol 57:845–848.
Tavassoéli FA, Devilee P (2003). Pathology and genetics of tumours of the breast and female genital organs (IARC WHO Classification of Tumours) Lyon: IARC press; 2003.
Thomssen C, Oppelt P, Jänicke F, Ulm K, Harbeck N, Höfler H, et al. (1998). Identification of low-risk node-negative breast cancer patients by tumor biological factors PAI-1 and cathepsin L. Anticancer Res 183C2173–2180.
Veronesi U, Rilke F, Luini A, Sacchini V, Galimberti V, Campa T, et al. (1987). Distribution of axillary node metastases by level of invasion. An analysis of 539 cases. Cancer 59:682–687.
Vinh-Hung V, Verkooijen HM, Fioretta G, Neyroud-Caspar I, Rapiti E, Vlastos G, et al. (2009). Lymph node ratio as an alternative to pN staging in node-positive breast cancer. J Clin Oncol 27:1062–1068.
©2014Egyptian Journal of Pathology
Zaghloul AS, Ghomeim WA, Abd El-Moneim T (2001). Patterns of axillary lymph node metastasis from breast cancer in Egyptian patients. J Egypt Natl Cancer Inst 13:1–8.