Abstract
OBJECTIVE To evaluate the impact of the number of negative lymph nodes on disease free survival (DFS) in patients with locally advanced gastric cancer.
METHODS A total of 485 patients who underwent surgery for locally advanced gastric cancer (pT3N0-2M0) and had a DFS at least 6 months were enrolled in this retrospective study. The medical records of the patients were reviewed in detail, and the characteristics of the patients and the findings of pathologic examination were analyzed in order to find the potential association with DFS. Subgroup analysis according to pathologic stage was performed. Multivariate analysis using the COX regression method was also conducted in order to identify the independent prognostic factors. The Kaplan-Meier method was used to plot DFS curves. The DFS rate was compared in each subgroup.
RESULTS COX regression analysis showed that the DFS rate of gastric cancer patients with pathologic stage T3N0-2M0 was significantly associated with age, degree of tumor differentiation, tumor location as well as the number of negative lymph nodes. Among patients with stage T3N0M0 disease, who had 1-4 and 5 or more negative lymph nodes, the 2-year DFS rate was 8.3% and 55.6%, respectively. Meanwhile, the 3-year DFS rates of the same group of patients was 0% and 24.9%, respectively (P = 0.025). In the T3N1M0 subgroup, the 2-year DFS rate of patients with 3 or fewer, 4-9, and 10 or more negative lymph nodes was 17.3%, 39.1%, 52.6%, respectively. The 3-year DFS rate in this group was 4.2%, 6.0%, 17.1%, respectively (P < 0.001). In the T3N2M0 subgroup, the 2-year DFS rate of patients with 7 or fewer and 8 or more negative lymph nodes was 11.5% and 35%, respectively. The 3-year DFS rate of the same group of patients with 8 or more negative lymph nodes was also significantly improved (0.8% vs. 5%, respectively; P = 0.015).
CONCLUSION For gastric cancer patients with pathologic stage T3N0-2M0, the number of negative lymph nodes is an independent prognostic factor for DFS. The number of negative lymph nodes may reflect the level of regional lymph node dissection or the accuracy of the pathologic staging.
keywords
Introduction
Gastric cancer is one of the most common cancers in China. Despite significant advances in surgical techniques and both preoperative and postoperative adjuvant therapies, the prognosis of patients remains unsatisfactory due to the increased incidence of recurrence after radical surgery. Numerous studies have reported that recurrence rates of patients after surgery are as high as 30%, and 60%-80% of them have regional lymph node metastases. The status of regional lymph nodes is strongly associated with the risk of postoperative recurrence. The purpose of this study is to examine the association between the number of negative lymph nodes and the prognosis, particularly DFS, in postoperative patients with locally advanced gastric cancer.
Patients and Methods
Patients
From January 2002 to December 2007, 2010 patients with primary gastric cancer received treatment in our institution. Each patient had a definitive diagnosis after pathologic examination following surgery (tumor anatomic location, tumor grade, the status of regional lymph nodes and tumor grade), and the pathologic stage was then determined according to the 6th AJCC Cancer Staging Manual (pT1-4N0-3M0). In the 2,010 patients, 512 patients who were classified as pathologic stage pT3N0-2M0 and had a period of DFS at least 6 months were eligible for the study. Twenty-seven patients with pathologic stage pT3N2M0 were excluded because of the small size of the sample. All of the 485 enrolled patients underwent the standard D1/D2 radical operation, among which 462 were given postoperative adjuvant systemic chemotherapy consisting of fluoropyrimidine and/or cisplatin. None of the enrolled patients received adjuvant radiotherapy.
Methods
Disease recurrence was defined as evidence of biopsy-proven relapse and/or evidence of relapse confirmed by examination during follow-up (CT, X-ray, PET-CT, and MRI) after radical surgery. DFS was defined as the time interval from the date of surgery to the date when the relapse was confirmed. The DFS was calculated on the basis of the age of the patients and the average number of negative lymph nodes. According to the DFS, patients were divided into 4 groups based on their age (≤ 49 years, 50-59 years, 60-69 years, ≥ 70 years). Moreover, the patients were stratified by the number of negative lymph nodes identified, and after stratified, patients in the pT3N0 subgroup were separated into 2 levels (1-4, ≥ 5) and those in pT3N1 subgroup were separated into 3 levels (≤ 3, 4-9, ≥ 10). The pT3N2 subgroup was separated into 2 levels (≤ 7, ≥ 8). The relationship between the number of negative lymph nodes and DFS was analyzed.
Statistical analysis
Multivariate analysis was performed using the COX regression method in order to search for independent prognostic factors in each group. A TA value of P ≤ 0.01 was considered statistically significant. The Kaplan-Meier method was used to plot DFS curves which showed the DFS rates in each subgroup. The DFS rates were then compared, and the results of P ≤ 0.05 were considered different. Data were analyzed using SAS version 17.0 (SAS Institute, Cary, NC).
Results
The influence of the number of negative lymph nodes and other factors on DFS of the patients in the T3N0M0 subgroup
Multivariate analysis revealed that average age, tumor site, tumor grade as well as the number of negative lymph nodes were significant factors affecting DFS. The 2-year DFS rate of patients with 1-4 and 5 or more negative lymph nodes was 8.3% and 55.6%, respectively. Meanwhile, the 3-year DFS rate of the same group of patients was 0% and 24.9%, respectively (P = 0.025).
The influence of the number of negative lymph nodes and other factors on DFS in the T3N1M0 subgroup
Multivariate analysis showed that average age, tumor site, tumor grade as well as the number of negative lymph nodes were significantly associated with DFS (Table 1). The 2-year DFS rate of patients with 3 or fewer, 4-9, and 10 or more negative lymph nodes was 17.3%, 39.1%, 52.6%, respectively. The 3-year DFS rate in this group was 4.2%, 6.0%, 17.1%, respectively (P < 0.001) (Fig. 1).
COX regression analysis of prognostic factors in gastric cancer patients in the T3N1M0 group.
The influence of the number of negative lymph nodes on DFS in the T3N1M0 subgroup.
The influence of the number of negative lymph nodes and other factors on DFS in the T3N2M0 group
Multivariate analysis showed that the factors significantly affecting DFS included average age, tumor site, tumor grade as well as the number of negative lymph nodes. The 2-year DFS rate for patients with 7 or fewer and 8 or more negative lymph nodes was 11.5% and 35%, respectively. The 3-year DFS rate of patients with 8 or more negative lymph nodes was significantly improved (0.8% vs. 5%, respectively; P = 0.015).
Discussion
Disease stage is regarded as the most important prognostic factor in gastric cancer patients. The 5-year overall survival rate of early stage gastric cancer patients who underwent radical surgery may be as high as 90%. However, locally advanced gastric cancer patients carry an unfavorable overall survival rate ranging from 20% to 60%. Accurate pathologic staging is thus vitally important in order to evaluate prognosis and to choose the appropriate treatment for each patient.
The two major TNM classification systems for gastric cancer routinely used in clinical practice include the AJCC/UICC (American Joint Committee on Cancer and the International Union Against Cancer) and the JGCA (Japanese Gastric Cancer Association). The definition of regional lymph nodes is the principal difference and the main controversy[1,2]. The definition proposed by the AJCC/UICC is based on the absolute number of positive lymph nodes, while that put forward by the JGCA is based on the location of positive lymph nodes. Advantages exist in both systems[1,3]. Recently, a growing body of evidence in the literature[4-6] suggests that the AJCC/UICC staging system may predict prognosis more precisely and that the rate of positive lymph nodes is an independent predictive factor of prognosis[7-9].
Precise pathologic lymph node staging (pN staging) depends on the complete examination of local and regional lymph nodes. The extent of the lymph node dissection is considered an indicator of the extent of the radical surgery performed and of the accuracy of pN staging. Japanese oncologists prefer D2 dissection over D1/D0 dissection because they maintain that D2 may lead to improved overall survival; however their preference differs from that of Western oncologists who prefer D1 dissections[10]. A relatively less extensive lymph node dissection results in a decreased number of lymph nodes examined and consequently limits the accuracy of pN staging. Recently, there have not been any consensus views regarding the minimum number of lymph nodes required for examination in order to achieve adequate pN staging. The AJCC/UICC TNM classification systems recommend that at least 15 lymph nodes need to be examined. In cases of which less than 15 lymph nodes are examined, lymph node staging should be determined according to the number of positive lymph nodes identified. According to the AJCC/UICC TNM classification systems, pN2 is defined as at least 7 positive lymph nodes while pN3 is defined as at least 15 positive lymph nodes. Nevertheless, we do not believe that examination of a total of 15 lymph nodes is sufficient to obtain accurate nodal staging. In fact, it has recently been shown that the more lymph nodes examined, the greater the possibility that positive lymph nodes will be found[11]. Furthermore, it has been demonstrated that the number of lymph nodes examined has significant prognostic impact on prognosis in each pN category of gastric carcinoma[6,12].
The number of lymph nodes stated in pathology reports depends not only on the extent of tumor dissection but also on the extent of the lymph node examination performed by the pathologists. The probability of catching negative lymph nodes is much lower than that of catching positive lymph nodes, which can lead to the neglect on the part of pathologists in detecting positive lymph nodes that are unremarkable in appearance. This suggests that if enough negative lymph nodes are examined, then greater accuracy in nodal staging can be achieved, which has been clearly demonstrated in research on colon cancer[13].
In our study, it has been shown that the DFS of patients with pT3N0-2 is significantly associated with age, degree of tumor differentiation, and tumor location as well as the number of negative lymph nodes. Even though the number of the lymph nodes examined in some cases was less than 15, the number of negative lymph nodes affected the DFS in each subgroup. With the increase in the number of negative lymph nodes examined, the DFS was improved significantly. This suggests that in addition to the total number of examined lymph nodes recommended by the NCCN, the number of negative lymph nodes detected was also an indicator in evaluating the extent of the radical procedure and that fewer negative lymph nodes would result in a deviation in pN staging.
It has been shown that 80% of gastric cancer patients have recurrences which are found at autopsy[14]. In the postoperative pathologic data of our study, the number of negative lymph nodes was found to have a close relationship with postoperative DFS in pT3N0-2M0 gastric cancer patients. We can conclude that the pN staging system based on the number of positive lymph nodes is more reasonable and that the number of negative lymph nodes could be taken as an indicator of the accuracy of pN staging even if the total number of lymph nodes examined is less than 15. Further studies with a larger sample size are needed to confirm the findings in our study, particularly in cases where more lymph nodes were examined.
- Received September 21, 2009.
- Accepted November 30, 2009.
- Copyright © 2009 by Tianjin Medical University Cancer Institute & Hospital and Springer