Increasing negative lymph node count predicts favorable OS and DSS in breast cancer with different lymph node-positive subgroups

Adequate lymph node evaluation is recommended for optimal staging in patients with malignant neoplasms including breast cancer. However, the role of negative lymph nodes (LNs) remains unclear in breast cancer according to N substage (N1, N2, and N3). In this study, for the first time, we analyzed the prognostic significance of negative LNs in breast cancer patients. A critical relationship was observed between negative LN count and survival, independent of patient characteristics and other related molecular variables including estrogen receptor (PR) status, progesterone receptor (ER) status, human epidermal growth factor receptor 2 (HER2) status, depth of tumor invasion and degree of differentiation. This research is of great importance in providing more information about the prognosis of breast cancer by statistical analysis of negative lymph nodes and can serve as a useful supplement to the current pathological system.


Introduction
Breast cancer is one of the most common carcinomas worldwide, especially among females [1]. The prognosis of patients with early stage breast cancer is favorable, but patients with lymph node (LN) metastasis usually have a poor prognosis; the more metastatic lymph nodes identified, the worse the long-term survival [2]. In the American Joint Committee on Cancer (AJCC)/International Union Against Cancer (UICC) tumor-node-metastasis (TNM) classification [3], positive LNs are classified into 3 subgroups: N1 (1-3 LNs), N2 (4-9 LNs) and N3 (10 or more LNs). The staging system can discriminate prognostic groups very well. However, it is difficult to correctly assess a patient's TNM classification. For example, one of the classic questions is the minimum number of LNs that should be dissected. Unfortunately, the number reported in the literature is inconsistent, ranging from 5 to 15 [4][5][6][7][8].
In addition to the total number of LNs excised, many other parameters have been developed and validated for various malignancies, such as number of negative lymph nodes [9,10], ratio of involved to removed nodes (lymph node ratio, LNR) [11,12], and log odds of positive lymph nodes (LODDS) [13,14], to aid the TNM system in predicting the outcome more accurately. However, few studies have focused on the relationship between the number of negative LNs and prognosis of breast cancer. Kuru [9] compared the roles of total number of nodes removed, negative nodes removed, and ratio of positive nodes to removed nodes in breast cancer using small samples from a single institute. Vinh-Hung et al [15] investigated the effect of number of negative nodes on survival in early breast cancer using the Surveillance Epidemiology and End Results (SEER) database. To the best of our knowledge, this study is the first to evaluate the role of negative LN count in breast cancer according to N substage (N1, N2, and N3) using a population-based database.

Patients
Patients were collected from the SEER database (2004-2013), which is a population-based cancer registry in the United States. The National Cancer Institute's SEER Ã Stat software (Version 8.2.0) was applied to identify patients with breast cancer. Included in the study were patients more than 18 years old, who had received surgical treatment and a pathologically confirmed diagnosis of breast cancer. The exclusion criteria were as follows: (1) patients for whom the number of examined or positive regional lymph nodes was 0 or unknown; (2) patients with distant metastasis (M1); (3) patients with a history of prior malignancy; and (4) patients with chemotherapy or radiotherapy before surgery. We retrieved baseline and clinical characteristics, including sex, age, race, tumor grade, T stage according to the 6th and 7th edition of AJCC criteria, estrogen receptor (ER) status, progesterone receptor (PR) status and human epidermal growth factor receptor 2 (HER2) status. All methods were carried out in accordance with relevant guidelines and regulations. All experimental protocols were approved by the Bioethics Committee of the Affiliated XuZhou Hospital of Medical College of Southeast University, China. Informed consent was obtained from all subjects in the SEER database.

Statistical analysis
Data about baseline and clinical characteristics were presented as count and percent values. The selection of the subgroups in the negative number of lymph nodes was determined by Xtile software (Yale University, Version 3.6.1). Overall survival (OS) and disease-specific survival (DSS) were evaluated by the Kaplan-Meier method. OS was calculated from the time of initial diagnosis to last follow-up or death, and DSS was defined as the interval from diagnosis until death due to breast cancer. Multivariate analyses for OS and DSS of breast cancer were performed by Cox regression models with adjusted hazard ratios (HRs). P values < 0.05 were considered statistically significant. All analyses were conducted utilizing PASW Statistics 18 software.
had early T stage (T1 accounting for 37.7%, T2 accounting for 44.7%). The number of negative LNs for the cohort ranged from 0 to 77, and the median number was 8.
During the follow-up period, 21679 (17.2%) patients in the entire cohort died. Overall, 14729 patients died from breast cancer, accounting for 67.9% (14729/21679) of total deaths. Taking into consideration the optimal cutoff number of negative LNs acquired by X-tile software in N1, N2 and N3 substage and the number of negative LNs in each subgroup, four negative LNs were categorized as an interval. The N1, N2 and N3 substage were divided into 4 subgroups: 0-3, 4-7, 8-11 and more than 11. OS and DSS for N1, N2 and N3 substage with 8-11 for OS (P > 0.05). No significant differences were observed among N1 patients with subgroups 4-7, 8-11 and more than 11 for DSS (P > 0.05). All the survival curves indicated that patients with more negative LNs had better outcomes, especially for N2 and N3 substage. The effect of negative LNs on prognosis was evaluated by multivariate analyses (Tables 2  and 3). The number of negative LNs was an independent prognostic factor for OS and CSS in each N substage. A larger number of negative LNs had a more positive effect on the prognosis  For example, sex had no effect on either OS or DSS for N3 substage (P = 0.360, P = 0.495, respectively).

Discussion
The TNM classification system attempts to account for most basic parameters of cancer, to provide guidance for treatment planning and to predict the outcome. The number of positive LNs has been the focus on for a long time. In the present study, for the first time, we analyzed the prognostic significance of negative LNs in breast cancer patients. We observed a critical relationship between negative lymph node count and survival, independent of patient characteristics and other related molecular variables including PR status, ER status, HER2 ststus, depth of tumor invasion and degree of differentiation. Although the number of negative LNs has an apparent effect on survival in each N stage, the mechanism underlying the relationship between negative LNs and survival remains unconfirmed. It has been reported that stage migration probably occurs due to under-evaluation of total lymph nodes, which has an intimate connection with the number of positive and negative LNs, thus influencing on pathological stage. Examining more lymph nodes can more accurately identify metastasis, thus avoiding misclassification of pathological stage [16].
Another possible hypothesis is that the number of negative LNs reflects the host lymphocytic reaction to tumor. The interaction between host and tumor contributes to the size or count of lymph nodes, and the increase of negative LNs indicates that the host takes a predominant position when competing with a tumor, and consequently obtains longer survival. A similar hypothesis has been associated with longer survival in colorectal cancer [17,18].
The number of lymph nodes informs the surgical approach, classification of pathological stage and institutional care. It has been reported that the number of lymph nodes may have a connection with better care overall and may not affect outcome directly [19][20][21][22]. There are also studies showing that the removal of more lymph nodes leads to a better prognosis [11]. Negative lymph nodes, clinically, imply that cancer cells were not observed using microscopy, instead of absolute no-tumor-metastasis, which means that there may be micro-metastasis of cancer cells. In other words, positive lymph nodes examined on routine pathological analysis do not necessarily reflect lymph node metastasis. For approximately 9%-30% of breast cancer patients, there are micro-metastases in their axillary lymph nodes that are undetectable by routine pathological examination [23]. Lymph node micro-metastasis is a strong risk factor for tumor-free survival and metastatic recurrence [24]. Therefore, the increase of the negative lymph node count, to a certain extent, can avoid the adverse effects of lymph node micrometastasis and improve the prognosis of patients [25,26].
The rate of LNR was not used in this study to assess the prognosis of patients. LNR refers to the ratio of the number of metastatic lymph nodes to the total number of lymph nodes cleared. In recent years, many studies have indicated that LNR is superior to pN staging in reflecting axillary lymph node status and predicting prognosis [11,[27][28][29][30][31]. PN staging depends on the number of positive lymph nodes and the total number of lymph nodes cleared. Different from pN staging, LNR balances the potential influencing factors. It can better reflect axillary lymph node stage, as well as prognosis [32,33]. In a meta-analysis study, Woodward also suggested that LNR is better as a prognostic index than the number of metastatic lymph nodes [34].
In addition, we only selected patients from the SEER database to analyze survival and prognostic factors. It is necessary for us to obtain more detailed information to confirm the relationship between negative LNs and survival.
In conclusion, we confirmed, for the first time, the relationship between negative lymph node count and prognosis of breast cancer in substage N2 and N3. The analysis is consistent with the presentation of colorectal cancer: more negative LNs imply longer survival [35][36][37][38]. It is of great importance to provide more information about the prognosis of breast cancer by statistical analysis of negative lymph nodes, which can serve as a useful supplement to the current pathological system.