Prognostic significance of a systemic inflammatory response in patients receiving first-line palliative chemotherapy for recurred or metastatic gastric cancer

Recurred or metastatic gastric cancer has a very poor prognosis, but chemotherapy can improve survival and possibly provide significant palliation of symptoms. Despite the recently reported benefits of chemotherapy, the 5-year survival rate for recurred or metastatic gastric cancer remains at ~ 5-20% [1‐3]. Despite an often short and poor overall survival, there is marked heterogeneity in the duration of survival among patients. Therefore, there have been continuing efforts to investigate the prognostic factors related to survival [4, 5].

C-reactive protein (CRP) is an acute phase protein which is synthesized by hepatocytes and;its levels in the serum increase during inflammatory diseases [6]. Cancer growth and resultant invasion induce local tissue damage, disturb local homeostasis, and cause systemic acute phase responses. Although the exact mechanism by which systemic inflammation arises in cancer patients remains to be clarified, it is generally accepted that cancer associated inflammation is modulated by cancer cells, host stromal cells, and their interactions [7].

There is increasing evidence that the presence of a systemic inflammatory response, as evidenced by elevated concentrations of CRP, is a prognostic factor in patients with advanced cancer [8‐10]. High CRP levels are common in patients with advanced disease, because advanced cancer is often associated with an inflammatory response [7].

The Glasgow prognostic score (GPS) was introduced as a useful predictor for survival in patients with cancer by Forrest in 2003, and consists of the combination of 2 values, CRP and albumin [4, 11]. It has been reported that GPS was associated with prognosis in various types of cancer including non small cell lung cancer, gastric cancer, colorectal cancer, pancreatic cancer, and breast cancer. Nearly all studies evaluated the prognostic value of preoperative CRP in resettable tumors [12]. Clinical factors such as liver metastasis, carcinomatosis peritonei, and bone metastasis are easily identifiable and imply that a patient has an advanced cancer. In this study, we evaluated the relationships between carcinomatosis peritonei, liver metastasis, bone metastasis, ECOG PS, albumin, CRP, GPS, and progression free survival (PFS), and overall survival (OS) in patients with recurrent or metastatic gastric cancer receiving first-line palliative chemotherapy.

This study included a total of 402 patients who received first-line chemotherapy in the Department of Oncology at Chonnam National University Hwasun Hospital (Gwangju, Republic of Korea). We collected follow-up patient data from the cancer registry. All data were prospectively recorded and only the survival data was updated at the time of analyses. The median follow-up time was 11.4 months and ranged from 1.1 months to 58.5 months. The baseline clinical characteristics of the patients at the start of first-line chemotherapy are shown in Table 1 and Table 2. The median age of the patients was 59 years, with a range of 19-80 years. A total of 203 patients (72.9%) were male, and 73 patients (48.2%) had an ECOG PS of 2-3. A total of 125 patients (31.1%) had liver metastases, and 159 patients (39.6%), and 28 patients (7.0%) had peritoneal metastases and bone metastases, respectively. A total of 140 patients (34.9%) had an elevated CRP concentration (> 1 mg/dL), and 77 patients (19.2%) had hypoalbuminemia (< 3.5 mg/dL).

The most commonly used first-line chemotherapy regimens were taxanes and cisplatin (n = 191, 47.5%). The specific chemotherapy regimens are shown in Table 3. A complete response to first-line chemotherapy was achieved in 25 patients (6.2%), and a partial response was achieved in 120 patients (29.9%), giving an overall response rate of 36.1% and a disease control rate of 70.9%. The results of chemotherapy are shown in Table 4. The median progression free survival was 4.4 months (95% CI 4.018-4.782), and the median overall survival was 11.8 months (95% CI 10.727-12.873).

The univariate analysis demonstrated that 5 clinical factors were significantly associated with a shorter PFS; these factors included bone metastases, albumin < 3.5 mg/dL, CRP > 1 mg/dL, PS ≥ 2, GPS ≥ 1 (Table 1). The univariate analysis also demonstrated that 5 clinical factors were significantly associated with a shorter OS; these factors included peritoneal metastasis, albumin < 3.5 mg/dL, CRP >1 mg/dL, PS ≥ 2, and GPS ≥ 1 (Table 2).

Cox multivariate analysis for PFS and OS was performed. Among the clinical factors, those with P < 0.2 were selected and included in the multivariate regression analysis using the Cox proportional hazards regression model. The result of the analysis identified the independent prognostic factors for PFS and OS. The independent poor prognostic factors for PFS were PS ≥ 2 (HR 1.37, 95% CI 1.02-1.84; P = 0.035), CPR >1 mg/dL (HR 1.64 95% CI 1.28-2.09; P = 0.001), and bone metastasis (HR 1.74, 95% CI 1.14-2.65; P = 0.009). The independent prognostic factors for OS were PS ≥ 2 (HR 1.33, 95% CI 1.01-1.76; P = 0.037), GPS 1 (HR 1.75, 95% CI 1.37-2.26; P = 0.001), GPS 2 (HR 1.79, 95% CI 1.29-2.47; P = 0.001), and bone metastasis (HR 1.61 95% CI 1.08-2.39; P = 0.017) (Table 5). Figure 1 shows the survival curve for patients according to GPS.

In this study, GPS 1 was most commonly found to be a result of an elevated CRP (87 of 111, 78.4%). Among the patients with GPS 1, the patients with increased CRP and normal albumin level had a better median progression survival and median overall survival than the patients with normal CRP and decreased albumin level; however, the difference was not statistically significant (Table 6).

CRP is a nonspecific but sensitive marker of systemic inflammatory response, and might be expressed in selected tumor cells [13]. The biological basis for the correlation between elevations in this general marker of systemic inflammation and disease risk and outcome are not completely understood. The liberation of multiple proinflammatory cytokines, including interleukin-1 (IL-1), IL-6, and tumor necrosis factor-α from the tumor environment eventually results in the induction of CRP synthesis from the liver and other tissues [14]. Based on many recent studies, it is now widely accepted that an elevated CRP value is a reliable indicator of poor prognosis for certain malignant tumors including melanoma, non-Hodgkin's lymphoma, colorectal, lung, prostate and ovary [12, 15, 16].

Several studies demonstrated that low albumin concentrations are significantly associated with poorer survival in patients with gastric cancer [17, 18]. In patients with gastric cancer, it is recognized that there may be reduced caloric intake due to stenosis of the cardia or pylorus. However, several recent reports showed that the systemic inflammatory response plays a major role in the progressive nutritional and functional decline of patients with cancer [19]. Indeed, measurement of the systemic inflammatory response, particularly CRP, has been included in the definition of cancer cachexia, together with weight loss and reduced calorie intake [20]. It thus appears that the development of hypoalbuminemia is secondary to an ongoing systemic inflammatory response and poor cancer specific survival is secondary to the systemic inflammatory response in patients with gastric cancer [21]. GPS can thus reflect both the presence of a systemic inflammatory response and the progressive nutritional decline of the patient with advanced cancer. In this study, GPS 1 was most commonly associated with increased CRP (87 of 111, 78.4%). Our results also demonstrated that increased CRP was associated with shorter PFS, and a GPS ≥ 1 was associated with shorter OS in advanced gastric cancer in palliative chemotherapy settings.

In gastric cancers, almost all studies evaluated the prognostic value of the preoperative CRP in resettable tumors [12]. In a recent study of 204 patients who underwent curative resection of gastric carcinoma, preoperative CRP elevations were found to be independently predictive of shorter overall survival [22]. To our knowledge, only 2 studies demonstrated the relationship between GPS and cancer specific survival in unresectable gastric cancer [4, 5]. These studies showed that GPS was a significant independent predictor of cancer survival in unresectable gastroesophageal cancer. However, these studies also included esophageal cancer and diverse treatment modalities such as chemotherapy, radiotherapy, and endoscopic laser were used for palliative treatment. In our study, we only enrolled patients receiving first-line palliative chemotherapy for gastric adenocarcinoma. We excluded gastroesophageal junction cancer, because the underlying pathophysiology of cancers of the gastroesophageal junction and the corresponding therapeutic approach differ substantially from those for gastric cancer proper [23].

Although gastric cancer is thought to spread more commonly to the peritoneum or liver than to bone, the reported incidence of skeletal lesions varies between 1 and 45% [24]. The presence of bone metastasis has consistently been associated with poor prognosis in studies of Korean gastric cancer patients. In their report analyzing 1,455 metastatic gastric cancer patients who received first-line chemotherapy, Lee et al. reported that the presence of bone metastasis was an independent poor prognostic factor [18]. In our study, bone metastasis was also independently associated with poor prognosis (median OS 8.2 vs. 12.1 months, bone metastasis (+) vs. (-), P = 0.017).

The mechanism by which a systemic inflammatory response might influence cancer survival is not clear. It may be that an elevated systemic inflammatory response is associated with a poor local immune response to the tumor and therefore increased lymph node spread and metastasis [21]. Also, the presence of a systemic inflammatory response and the associated nutritional decline may influences tolerance and compliance with active treatment [25, 26]. Andreyev et al. reported that the poorer outcome of chemotherapy in advanced gastrointestinal cancer patients with weight loss appears to be as a result of receiving less chemotherapy due to toxicity rather than poorer tumor response [27]. In this study, among the patients with GPS 1, the patients with normal CRP and decreased albumin level (n = 24) had a tendency to have a poorer median progression free survival and median overall survival than the patients with increased CRP and normal albumin level (n = 87). In this group of the patients, decreased albumin level might be associated with the chronic nutritional decline and poor ECOG PS.

The presence of a systemic inflammatory response as evidenced by the CRP and GPS, appears to be a useful indicator of prognosis in patients receiving first-line chemotherapy for metastatic gastric cancer. Poor performance status and bone metastasis were also associated with poor prognosis. CRP, GPS, and bone metastasis are easily identifiable and objective clinical parameters. These parameters could facilitate the individual patient risk assessment and thus, a more appropriate prediction of survival for each patient.