Abstract
Background/Aim: The Glasgow prognostic score (GPS) is a predictor of outcome for several cancer types. The present study examined the significance of modified GPS (mGPS) in the prognosis of patients undergoing palliative surgery for stage IV gastric cancer. Patients and Methods: A total of 42 patients with stage IV gastric cancer treated with palliative gastrectomy and gastrojejunostomy were included in the study. Univariate and multivariate analyses were performed to evaluate the relationship between clinicopathological factors and cancer-specific survival (CS). Results: Among patients who underwent palliative surgery including gastrectomy and gastrojejunostomy, univariate analysis of CS identified the following significant risk factors: surgical treatment, chemotherapy and mGPS, and multivariate analysis revealed that mGPS was independently-associated with CS. In particular, among patients who underwent palliative gastrectomy, mGPS was shown to be the strongest independent predictive factor for CS. Conclusion: The mGPS was an independent predictive factor for survival in patients who underwent palliative surgery for stage IV incurable gastric cancer, especially for those who underwent palliative gastrectomy.
Gastric cancer is the most common malignant disease in Japan, and the second most common cause of cancer-specific death worldwide. Stage IV gastric cancer has a poor prognosis, and patients with metastatic gastric cancer are currently not considered candidates for surgical treatment and are most often offered systemic chemotherapy. Although palliative surgery, including simple gastrectomy and gastrojejunostomy, is performed for incurable, advanced, and metastatic gastric cancer to relieve distressing symptoms, including hemorrhage from a tumor, pain, and inability to eat, the outcome is poor (1). Easily obtainable and simple biomarkers may help select patients for appropriate treatment strategies for stage IV gastric cancer. Histological markers such as tumor size, differentiation, lymphoid metastasis, and vessel invasion are well-established tumor-related prognostic factors. Weight loss and performance status are also patient-related factors associated with survival in advanced gastric cancer (2, 3). However, the use of these parameters as prognostic indicators remains problematic because they are often ill-defined and subject to bias (3, 4).
The association between hypoalbuminemia and poor prognosis in patients with carcinoma is well-recognized (5, 6). In addition, there is increasing evidence that the host inflammatory response to cancer cells plays a critical role in the development and progression of numerous types of cancers (7, 8). To date, the inflammation-based prognostic score–the Glasgow prognostic score (GPS), which is defined by serum C-reactive protein (CRP) and albumin levels–has been demonstrated to be a prognostic marker for patients with gastrointestinal carcinomas. Although the GPS appears to be a simple prognostic score of malignancy because it can be easily determined, only few studies of the postoperative prognosis of gastric cancer using GPS have been reported (9-11), and a detailed evaluation of patients who have undergone palliative surgery for stage IV gastric cancer has not been performed. This study aimed to test the hypothesis that the GPS is useful for predicting postoperative prognosis in patients who have undergone palliative surgery for stage IV incurable gastric cancer.
Patients and Methods
A total of 193 patients with gastric cancer underwent surgical resection or gastrojejunostomy at the Department of Surgery, Social Insurance Yokohama Central Hospital from March 2006 to March 2013. Of these, 42 with stage IV gastric cancer were included in this study. Patients who died within 30 days of surgery and who had other malignancies that might have increased their CRP levels were excluded from this study. We excluded six patients for the following reasons: two had concurrent malignancies and four had incomplete clinical data. The remaining 36 cases included in this study had adequate clinical information and follow-up data. Palliative gastrectomy was defined as resection of the primary stomach lesion along with regional lymph node (LN) dissection but no resection of lesions considered incurable, such as peritoneal dissemination and residual metastases of LNs, liver, or distant organs, to reduce the tumor volume or relieve discomfort associated with gastric cancer. Distal gastrectomy was performed for tumors located in the lower or middle third of the stomach, whereas total gastrectomy was performed for tumors located in the upper third or whole stomach. Gastrojejunostomy was performed to enable oral intake in unresectable locally advanced gastric antral cancer. The pathological diagnosis and classifications were made according to the seventh edition of the TNM Classification of Malignant Tumors (12).
Comparison of clinicopathological factors between the low-modified Glasgow Prognostic Score (mGPS) group and the high-mGPS group.
The potential prognostic factors for advanced gastric cancer were as follows: age (<70 vs. ≥70 years); sex (male vs. female); body mass index (<20 vs. ≥20 kg/m2); Eastern Cooperative Oncology Group performance status (PS) (0-1 vs. ≥2); histological subtype (well- and moderately-differentiated adenocarcinoma vs. poorly- and undifferentiated adenocarcinoma); number of metastatic sites (≤1 vs. ≥2), metastatic sites such as LNs, liver, lung, and peritoneum; peritoneal cytology (positive vs. negative); carcinoma embryonic antigen level (<5 vs. ≥5 ng/ml); modified GPS (mGPS: 0-1 vs. 2); palliative surgical treatment (gastrectomy vs. gastrojejunostomy); and chemotherapy (none vs. given). Medical records were retrospectively reviewed to examine these factors.
Patients who had both elevated CRP level (>0.5 mg/dl) and hypoalbuminemia (<3.5 g/dl) were assigned an mGPS of 2, and patients with only one or neither of these blood chemistry abnormalities were assigned a score of 1 or 0, respectively (13). We defined patients with a score of 2 as the high mGPS group and those with a score of 1 or 0 as the low mGPS group. The association between mGPS and clinicopathological parameters and cancer-specific survival (CS) was assessed.
Differences between categorical variables were determined using the Chi-squared and Fisher's exact tests. The influence of each prognostic factor identified by univariate analysis was assessed by multivariate analysis using Cox's proportional hazards regression. CS was defined as the time from the first day of surgical treatment until death due to cancer-related causes. Kaplan–Meier analysis and the log-rank test were used to compare mortality for each mGPS group. A value of p<0.05 was considered statistically significant in all analyses. All statistical analyses were performed using StatView (SAS Institute, Inc, Cary, NC, USA).
Results
The mGPS was 0, 1, and 2 for 7 (19.4%), 13 (36.1%), and 16 (44.4%) patients, respectively. Therefore, the low-mGPS group included 20 (55.6%) patients and the high-mGPS group included 16 (44.4%). A comparison of patient characteristics according to mGPS group is shown in Table I. Chemotherapy after surgery was performed more often in the low-mGPS group than in the high-mGPS group. However, there was no significant difference in the other clinicopathological factors.
Table II shows the relationship between the clinicopathological factors and CS among patients who underwent palliative gastrectomy and gastrojejunostomy. On univariate analyses, there were significant differences in surgical treatment (p=0.0358), chemotherapy after surgery (p=0.0103), and mGPS (p=0.0001). On multivariate analyses using the clinicopathological factors that had significant differences in univariate analyses, mGPS [hazard ratio (HR)=0.156, 95% confidence interval (CI)=0.049-0.519, p=0.0023] was the only factor significantly associated with CS.
Univariate and multivariate analyses of cancer-specific survival in patients who underwent palliative gastrectomy and gastrojejunostomy.
Univariate and multivariate analyses of cancer-specific survival in patients who underwent gastrectomy.
Prognostic factors for each palliative surgical treatment are shown in Tables III and IV. Among patients who underwent gastrectomy, mGPS (HR=0.129, 95% confidence interval 0.025-0.663, p=0.0142) was the only independent factor on multivariate analysis (Table III). However, among patients who underwent palliative gastrojejunostomy, there were no significantly clinicopathological factors on univariate analysis (Table IV). mGPS was not a prognostic factor for CS in patients who underwent palliative gastrojejunostomy.
The median follow-up duration was 189 days. Thirty-one (86.1%) patients died and five (13.9%) were censored at the last follow-up date. The median CS was shorter in the high-mGPS group (156 days) than in the low-mGPS group (377 days) (p<0.0001) (Figure 1). Moreover, comparing CS between the low mGPS group and the high mGPS group for each palliative surgical treatment, among patients who underwent palliative gastrectomy, there was a significant difference between the two groups (p<0.0001) (Figure 2). However, among patients who underwent palliative gastrojejunostomy, there was no significant difference (p=0.0897; Figure 3).
Discussion
The rationale for offering palliative surgery to patients with stage IV gastric cancer is that if left, the primary tumor will result in gastric obstruction, perforation, bleeding, or debilitating ascites. Gastric resection should therefore be used for alleviation of gross disease with the goal of improving function and quality of life by removing a bulky symptomatic tumor. Gastric bypass will also enable oral food intake by improving obstructive symptoms. Sadi et al. reported that even in Stage IV gastric cancer, palliative gastrectomy might be associated with a survival advantage in selected groups of patients, especially if combined with adjuvant therapy as part of a multi-modality treatment (1). In general clinical practice, the indications for palliative surgery were determined by attending surgeons based on patients' general health PS, symptoms, extent of disease, and feasibility of resection. However, it is unclear whether palliative surgery or chemotherapy is recommended for all patients with stage IV incurable gastric cancer.
Univariate analysis of cancer-specific survival in patients who underwent gastrojejunostomy.
Inflammation seems to play an important role in the development and progression of various types of cancer. Recently, increasing evidence has demonstrated the important role of inflammation in advanced-stage cancer (14). Elevation of serum CRP level indicating the presence of systemic inflammation adversely affects the prognosis of patients with upper digestive tract cancer (15). Hypoalbuminemia is often observed in patients with advanced cancer, and is a good predictor of malnutrition and cachexia. In gastric cancer, hypoalbuminemia is considered an independent predictor of poor prognosis (16). However, hypoalbuminemia is likely to develop secondary to an increase in serum CRP level (17). Crumley et al. demonstrated that the relationship of low serum albumin level and reduced survival in patients with gastric cancer was dependent on an elevated CRP level (18). The mGPS combines two variables, CRP level and albumin level. Therefore, the mGPS may enable a better appreciation of inflammation and malnutrition influenced by the tumor. Since Forrest et al. first published an evaluation of GPS in patients with inoperable non-small cell lung cancer, there is increasing evidence for a role of the systemic inflammatory response in predicting survival in various cancer types, independent of tumor stage (19). In gastric cancer, Nozoe et al. reported higher GPS and tumor stage to be independent prognostic indicators for worse prognosis in patients with curatively resected gastric cancer (9). Kubota et al. also reported a high GPS to be a significant predictor of short- and long-term survival after curative surgery for gastric cancer, excluding Stage IV cancer (11). However, no studies demonstrate the significance of mGPS in predicting the outcome after palliative surgery for stage IV gastric cancer. The result of this study indicates that CS among patients with Stage IV gastric cancer who underwent palliative gastrectomy and who had a high mGPS was worse than that among patients with a low mGPS. Therefore, mGPS may be a particularly useful prognostic indicator in patients who undergo gastrectomy. To date, there are limited reports with extensive clinical evidence of palliative gastrectomy (20); however, this report indicates that mGPS may be a good prognostic marker in patients with stage IV incurable gastric cancer.
Jingxu et al. in 2013 reported a meta-analysis analyzing survival outcomes and for establishing a consensus on whether palliative gastrectomy is suitable for patients with incurable advanced gastric cancer and which type of patients should be selected to undergo palliative gastrectomy (20). This meta-analysis showed that palliative gastrectomy for patients with incurable advanced gastric cancer may be associated with longer survival, especially for patients with Stage M1 gastric cancer. Moreover, it described that palliative gastrectomy combined with hepatic resection for patients with liver metastasis (HR=0.3, 95% CI=0.15-0.61) and those receiving adjuvant chemotherapy (HR=0.63, 95% CI=0.47-0.84) may improve survival. However, inflammation and nutritional markers, CRP and albumin, were not included in the analysis in this study. In the present study, liver metastasis, chemotherapy, and mGPS were significantly associated with CS factors in univariate analysis, and the mGPS was identified as a significant independent prognostic factor for CS in patients who underwent palliative gastrectomy in multivariate analysis.
Relationship between modified Glasgow Prognostic Score (mGPS) and cancer-specific survival in patients with stage IV gastric cancer.
However, serum albumin and CRP levels and the mGPS were not significant prognostic factors in patients who underwent palliative gastrojejunostomy in this study. Fujitani et al. described that surgical palliation was beneficial in patients with a PS of 0-1 and those with recurrent disease, in terms of improved quality of life and prolonged survival in patients with advanced gastric cancer and those presenting with gastrointestinal obstruction (21). However, serum CRP levels and hypoalbuminemia were not independent factors of overall survival in this study. Takeno et al. reported a multi-center retrospective analysis of clinical outcomes and indications for palliative gastrojejunostomy in unresectable advanced gastric cancer (22). They described that poorer PS, preoperative chemotherapy, and a high CRP level were significant independent predictors of poor survival. These reports suggest that PS is a good independent predictor of survival in patients with advanced gastric cancer undergoing palliative gastrojejunostomy. However, in the present study, PS was not a predictor for CS, although the number of patients who underwent palliative gastrojejunostomy was small.
Recently, endoscopic stent placement as a non-surgical palliative treatment has been reported in patients with malignant gastric obstruction (23). This palliative treatment offers several advantages over gastrojejunostomy because it does not require general anesthesia and is less invasive because a laparotomy is not required. Moreover, it is less costly and associated with shorter Hospital duration (23). Nishikawa et al (24) described that prognosis of patients undergoing gasrtojejunostomy in advanced gastric cancer was poor, as the GPS increased from 0 to 2. They suggested that endoscopic stenting should be considered as an alternative to gastrojejunostomy for patients with a GPS of 2 based on the poor prognosis of these patients.
Relationship between modified Glasgow Prognostic Score (mGPS) and cancer-specific survival in patients who underwent gastrectomy.
Relationship between modified Glasgow Prognostic Score (mGPS) and cancer-specific survival in patients who underwent gastrojejunostomy.
In conclusion, preoperative mGPS can be easily obtained from a simple, inexpensive, and non-invasive laboratory examination. The mGPS may be used as a prognostic indicator together with traditional risk factors and prognostic markers for patients who undergo palliative gastrectomy for Stage IV incurable gastric cancer.
Footnotes
-
Conflicts of Interest
The Authors have no conflicts of interest to declare.
- Received March 7, 2014.
- Revision received April 10, 2014.
- Accepted April 11, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
