Clinicopathologic Characteristics of Gastric Cancer Patients according to the Timing of the Recurrence after Curative Surgery

The incidence of gastric cancer has recently been gradually decreasing, and particularly the proportion of cases of early gastric cancer among all the cases of gastric cancer is rapidly on the rise. Yet the mortality rate due to gastric cancer is still high.(1, 2) Surgical resection is currently the only radical treatment for gastric cancer, and the 5-year survival rate is continuously on the increase because of performing radical resection, extended lymphadenectomy, postsurgical chemotherapy and adjuvant immunotherapy. Nevertheless, death is due to recurrence in 60~70% of the cases of gastric cancer.(3) In some cases, it can be resected again, which allows long term survival. Yet gastric cancer recurs as diverse patterns in several organs, and it is difficult to distinguish the recurrence patterns. Hence, this is treated by conservative treatments, and prognosis is poor.(4-6) The level of tumor infiltration and the presence or absence of lymph node metastasis are currently the important prognostic factors, but there are insufficient studies on the risk factors pertinent to the time of recurrence or the recurrence patterns. To understand the factors associated with recurrence and to help appropriately treat the at-risk group after surgery, we analyzed the clinicopathological characteristics of patients with the same disease stages according to the time of recurrence.

From June 1992 to March 2009, in the Department of Surgery at Hanyang University, 1,795 patients were diagnosed with gastric cancer and 428 of them died. Among them, we excluded the patients that the cause of death was not known (n=50), the patients who died due to accidents (n=11), the patients who died due to cancer in other organs (n=17) and the patients who died of other benign diseases (n=39). Finally, this study was conducted on 311 patients who died due to recurrent gastric cancer (Table 1). A radical resection was done for the patients without distant metastasis and without infiltration to the adjacent organs, the patients with infiltration and without macroscopic residual cancer and the patients who underwent lymphadenectomy higher than D2.

Table 1
Patterns of recurrence after curative surgery in gastric cancer

• Click for larger image

These 311 patients were divided to 1) 72 patients (23.2%) (the early recurrence group) whose recurrence time was within 1 year after surgery, 2) 92 patients (29.6%) (the late recurrence group) whose recurrence occurred 3 years after surgery and 3) 147 patients (47.2%) (the mid recurrence group) who had recurrent disease between 1 year to 3 years after surgery (Fig. 1), and the clinicopathological characteristics of the early recurrence group and the late recurrence group were compared and analyzed.

Fig. 1
Frequency of recurrence after curative operation.

• Click for larger image

History taking, physical examination, general blood tests, biochemistry tests, tumor marker tests, chest X-rays, gastroscopic examination, abdominal ultrasonography, abdominal computed tomography and PET-CT were regularly performed at the postsurgical ambulatory follow-up.

The recurrence patterns were classified as local recurrence, peritoneal recurrence and hematogenous recurrence. The local recurrence group was defined as cases that according to the clinical findings or various test findings, the gastric cancer recurred in the remaining stomach, the anastomosis areas and the adjacent organs and lymph nodes. According to physical examination, abdominal plain radiography and abdominal computed tomography, peritoneal recurrence was defined as the cases with tumors within the abdominal cavity, the cases with cancers detected by an ascites test, the cases in which diffuse metastasis was detected within the abdominal cavity during resurgery and the cases with tumor palpated on the rectal shelf by physical examination. Hematogenous recurrence was defined as the cases with hematogenous metastasis to distant organs such as the liver, bone, brain, abdominal wall etc., and this was detected by chest X-ray, abdominal sonography, a total bone scan, other radiologic tests and biopsy.

SPSS 13.0 was used for all the statistical analyses. Student's t-test was applied for comparing the various clinicopathological characteristics associated with recurrence. Logistic regression tests were applied for the multivariate analysis. P-values <0.05 were considered to be statistically significant.

The incidence and mortality rate of gastric cancer are on the decrease, and due to the development of diagnostic methods, the rate of discovering early gastric cancer as well as the 5-year survival rate are gradually increasing.(1, 2) Numerous studies on the clinicopathological characteristics that exert effects on the prognosis of gastric cancer have been conducted, and prognosis can be predicted by the level of the infiltration of tumors to the stomach wall and the presence of lymph node metastasis and distant metastasis, which are the basis of the classification of the disease stage of gastric cancer.(7) Nevertheless, despite the improved diagnostic methods and extended therapeutic resection, the percent of cases that recur within 2 years after surgery has been reported to be 60~70%(3-6) In such a manner, most cases of mortality are associated with recurrence. Studies on the factors associated with the recurrence of gastric cancer are ongoing, but the information on this is still not sufficient.

Generally, recurrence refers to the growth of cancer lesions pertinent to the primary caner after radical resection is performed according to the objective decision of surgeons.(8) The aim of radical resection of gastric cancer is first to completely remove the primary lesions with appropriate margins and second to prevent metastasis of cancer by the complete resection of the lymph nodes in the vicinity of the stomach.(9) In regard to distant metastasis that develops after such radical resection, the possibility is high that cancer has already formed beyond the range of surgery and at the time of surgery. Regarding local metastasis, the possibility of the presence of microscopical residual cancer cells is high. When reviewing the characteristics of recurrence according to the advanced level of cancer, in patients with progressive stomach cancer, the progression routes are diverse such as lymphoid, hematogenous and direct infiltration, peritoneal dissemination, etc. In most cases, the cancer progresses by several routes and not by a single route. On the other hand, in patients with early gastric cancer, progression is caused by overlooking multiple lesions, incomplete resection without securing sufficient margins after radical resection and insufficient lymphadenectomy in most cases.(9, 10)

In most studies, recurrence has been classified as the early recurrence group and the late recurrence group based on 2 years and 3 years, respectively, and the risk factors associated with recurrence or the survival period after recurrence were analyzed.(6, 11-16) This is based on the observation that for approximately 70% of recurred patients, the cancer recurs within 2 years after surgery, and cases that recur after 5 years are rare.(15)

Shiraishi et al.(11) classified 138 patients who died of recurrence after radical stomach resection as 104 cases (75.36%) of the early recurrence group (recurred within 2 years after surgery) and 34 cases (24.64%) of the late recurrence group (recurred after 2 years), and they compared the clinicopathological characteristics. It was observed that in the early recurrence group, the average size of tumors was significantly larger (P<0.01) and the cases with lymph node metastasis (P=0.01), lymph duct infiltration (P<0.01) and advanced disease stages (P=0.01) were significantly more abundant.

In addition, Yokota et al.(14) classified 251 patients who died of recurrence after radical stomach resection as the group that recurred within 2 years and the group that recurred after 2 years. In regard to the incidence of recurrence and the timing, there were 195 patients (77.69%) and 56 patients (22.31%) in each group, respectively, and the early recurrence group was larger. In the early recurrence group, the size of tumor was significantly larger (P=0.029). In addition, serous infiltration (P=0.038), lymph node metastasis (P=0.001), and the incidence of vascular infiltration (P<0.001) were significantly higher in the early recurrence group.

Efforts have recently been made to apply immunohistochemical staining to find predictive factors for recurrence after radical resection of gastric cancer, in addition to the previously determined clinicopathological factors. Kim et al.(17) compared the expressions of c-erbB2, EGFR, MLH1, MSH2 and aquaporins of the recurred group with early gastric cancer after therapeutic radical resection with those of the non-recurrence group with early gastric cancer. They reported that the expression of c-erb B2 was significantly higher in the recurrence group (P=0.024).

In this current study, based on 72 patients (23.2%) whose recurrence time was within 1 year and 92 patients (29.6%) whose recurrence time was longer than 3 years, the patients who died of recurred gastric cancer after radical resection were divided to the early recurrence group, the mid recurrence group and the late recurrence group. Different from most studies(6, 9, 11-13) that classified patients based on 2 years after surgery.

In this study we defined the early recurrence group based on the approximately early 25% cases of all recurrence cases and the late recurrence group based on the approximately late 25% cases of all recurrence cases.

Thus, if recurrence could be detected early based on our study' s data, this could be an important for more aggressive therapies for recurrent lesions rather than just performing radical resection. In the early recurrence group, as compared with the late recurrence group, an advanced disease stage was more prevalent, the mean age was higher, the incidence of vascular, lymph duct and perineural infiltration was higher and the histological types with poor differentiation were more abundant. Particularly, for the patients with a highly advanced gastric cancer (stage III, stage IV), perineural infiltration was abundant in the early recurrence group, and there were significantly more cases with a relatively small number of dissected lymph nodes. In such a manner, although the recurrence time was different according to the disease stages, within a same disease stage, some clinicopathological factors showed significant differences according to the recurrence time.

Lymph node metastasis is a well known prognostic factor that exerts effects on the recurrence rate as well as the overall survival rate. Tumor cells released from primary lesions penetrate the basement membrane, they transverse the interstitial tissues and penetrate the vascular basement membrane, they enter the circulation system and then they form tumor cell embolism. They, in turn, are released again through the basement membrane and form distant metastatic lesions. In addition, it has been reported that perineural infiltration, which has already been recognized as a major factor that mediates effects on tumor recurrence after surgery in patients with pancreas, biliary tract, esophagus and coloreactal cancer, is associated with the recurrence of gastric cancer.(18-20) Bilici et al.(18) analyzed the perineural infiltration of 238 patients who underwent radial gastrectomy, and it was observed that 180 patients (75.6%) showed perineural infiltration. When their clinicopathological characteristics were analyzed and compared with those of the group without perineural infiltration, the average size of tumor was significantly larger (P<0.001), the differentiation grade was worse (P=0.009) and an advanced disease stage (P<0.001), lymph node metastasis (P<0.001) and vascular infiltration (P<0.001) were significantly more abundant. In addition, the average survival period of the group with perineural infiltration was 28.1 months, and it was significantly shorter than 64.9 months of the group without perineural infiltration (P=0.001). Perineural infiltration is a process of tumor infiltration due to the neurotropic properties of tumor cells and it refers to the process of the denaturation of nerve fibers due to tumor cells infiltrating nerve bundles or nerve sheaths in the vicinity of tumors. Examined under an electronic microscope, nerve sheaths, lymph ducts and blood vessels are connected to each other anatomically, and through this, tumor cells infiltrate to adjacent tissues through the nerve sheaths and then they disseminated into the lymphoid system and toward the peritoneal membrane. Therefore, for cases with lymph node metastasis or perineural infiltration, even if radical resection is performed, the possibility of recurrence is high. It has been suggested that the follow-ups should be regularly performed and this may be of help to detect recurrence early.

For making the diagnosis of recurrence, not only imaging tests, but also tumor markers, hepatic function tests, general blood tests and other hematological tests are applied. Currently at our hospital, after radical gastrectomy and depending on the disease stage, different categories of tests and test intervals are applied. In other words, for stage I and stage II patients, general blood tests, hepatic function tests and tumor marker tests are performed at 6 month intervals. Abdominal computed tomography is performed at 6 month intervals up to 18 months after surgery, and at 12 month intervals after 18 months. In addition, gastroscopy is performed for the first time 6 months after surgery and at 12 month intervals afterward. PET-CT is performed 24 months after surgery for the first time. For the highly advanced cancer stage III and stage IV patients, since the risk of recurrence is high, general blood tests, liver function tests and tumor marker tests are performed at 3 month intervals, and abdominal computed tomography is performed 3 months after surgery for the first time and then at 6 month intervals afterward. PET-CT as well as gastroscopy are performed 6 months after surgery for the first time, and at 12 month intervals afterward (Table 6). In such a manner, the follow-up observation is performed differently according to the disease stage in order to detect recurrence after surgery early and to facilitate establishing treatment plans accordingly.

Table 6
Post-operative surveillance schedule according to stage

• Click for larger image

It is clear that an early diagnosis and radical resection during the initial surgery are the most important factors to increase the survival rate and to lower the recurrence rate. It is also true that with the development of chemotherapy, immunotherapy and other diverse therapies, the survival rate and treatment outcomes are improving.(21-23) Nonetheless, gastric cancer recurs after surgery in many patients, and once cancer recurs, the prognosis is very poor because definite treatments have not been established. (24-26) Gastric cancer is different from other cancers, it recurs as asynchronous multicentric cancer rather than locally, and so radical resection is difficult in many cases, and the cancer's responsiveness to systemic treatments such as chemotherapy is poor.(25-27) Gastric cancer shows complex patterns of recurring at diverse times and the studies on factors that can predict the recurrence time are not still sufficient.

However, it has been reported that after radical resection, the survival rate can be increased through efforts to detect recurrence early by regular examination as well as by determining the appropriate range of surgery and performing radical treatments after the definite diagnosis.(28-30) From July 1988 to December 1995, Kim et al.(28) examined the clinicopathological characteristic of patients who developed local metastasis in the remaining stomach after radical gastrectomy, and the gastric cancer was primarily diagnosed by gastroscopy and an upper gastrointestinal series. The CEA level was elevated in 4 cases, and N2 and N3 group lymph node infiltration was abundant. It has been reported that in such local recurrence cases caused by residual cancer, at the time of detection, the progression level of the lesion was severe and so the prognosis was poor. Nonetheless, in regard to clinically significant factors, it is thought that more comprehensive follow-up observation should be performed if the above findings are detected by presurgical tests as well as postsurgical tests. Moreover, it has been suggested that studies on methods that could be of help to improve the survival rate after recurrence are required.

Therefore, not only improving the quality of life according to radical treatments, but also the prevention of recurrence as well as the early detection of recurrence are important as study subjects. It has been shown that the clinicopathological characteristics of primary cancer are associated with the recurrence patterns and the recurrence time, and so if appropriate follow-up observation is performed on the factors that were significant in this study, recurrence could be detected early and then properly treated, and the mortality rate of gastric cancer itself could be decreased. For this, among various clinicopathological factors, efforts should be made to find new predictive factors that are clearly correlated with the recurrence of gastric cancer. In addition, studies on surgery, chemotherapy and other optimal treatment methods according to such correlation are also required.

In gastric cancer patients who recurred after radical resection, according to the disease stages and even in the same disease stage, the characteristics and time of recurrence vary according to diverse clinicopathological factors. Therefore, by considering this, if the interval of postsurgical follow-up observation and the types of tests are individualized, this would be useful to detect recurrence early and it would be of great help to improve the prognosis through appropriate treatments at early times.