**5.1. Glasgow prognostic score and survival**

Glasgow prognostic score has high informativity both in surgically treatable and advanced, unresectable and/or metastatic gastric cancer. Thus, evaluating Glasgow prognostic score, NLR and PLR in 324 patients with resected stage III gastric adenocarcinoma, only Glasgow prognostic score along with TNM stage was independently associated with overall and cancerfree survival [43]. In 207 gastric cancer patients who underwent surgery, GPS along with NLR, PLR, CRP, albumin and TNM stage were significantly associated with overall and cancer-free


**Table 4.** Glasgow prognostic score and its modifications.

survival. However, only GPS and TNM were independent prognostic factors; therefore in this study, GPS was favoured as the most informative SIR parameter [63]. By multivariate analysis, GPS was independent predictor of overall survival in 425 surgically treated gastric cancer patients who had normal serum levels of carcinoembryonic antigen [91]. In a large cohort of 1017 patients subjected to curative resection of gastric cancer, GPS was an independent prognostic factor for overall survival [95].

**5.1. Glasgow prognostic score and survival**

172 Gastric Cancer

Glasgow prognostic score has high informativity both in surgically treatable and advanced, unresectable and/or metastatic gastric cancer. Thus, evaluating Glasgow prognostic score, NLR and PLR in 324 patients with resected stage III gastric adenocarcinoma, only Glasgow prognostic score along with TNM stage was independently associated with overall and cancerfree survival [43]. In 207 gastric cancer patients who underwent surgery, GPS along with NLR, PLR, CRP, albumin and TNM stage were significantly associated with overall and cancer-free

**Score Definition References** Glasgow prognostic score [68]

OR albumin < 35 g/L

mg/L AND albumin < 35 g/L

normal albumin level: CRP > 10 mg/L

hypoalbuminemia: CRP > 10 mg/L

0 CRP < 10 mg/L AND albumin ≥ 35 g/L 1 One high-risk finding: CRP ≥ 10 mg/L

2 Both high-risk findings: CRP ≥ 10

Modified Glasgow prognostic score [31] [94] <sup>0</sup> CRP ≤ 10 mg/L irrespective of

Modified Glasgow prognostic score by Hirashima et al. [92] 0 CRP ≤ 5 mg/L AND albumin ≥ 38 g/L 1 One high-risk finding: CRP > 5 mg/L

High-sensitivity Glasgow prognostic score [93] 0 CRP ≤ 3 mg/L AND albumin ≥ 35 g/L 1 One high-risk finding: CRP > 3 mg/L

2 Both high-risk findings: CRP > 5

2 Both high-risk findings: CRP > 3

Abbreviations: CRP, C-reactive protein.

**Table 4.** Glasgow prognostic score and its modifications.

1 Increased CRP on the background of

2 Increased CRP and

albumin level

AND albumin ≥ 35 g/L

AND albumin < 35 g/L

OR albumin < 38 g/L

OR albumin < 35 g/L

mg/L AND albumin < 38 g/L

mg/L AND albumin < 35 g/L

In a homogeneous group of 88 gastric cancer patients undergoing only surgical treatment, increasing GPS was an independent predictor of worse overall survival and perioperative mortality. The median survival by GPS 0 vs. 1 vs. 2 was 25.2 vs. 15.3 vs. 5.8 months. The perioperative mortality in the same subgroups was 0.0% vs. 20.0% vs. 80.0% [96]. However, the GPS capacity to predict complications is not straightforward. In contrast with the previous report, assessing 1017 patients subjected to curative resection of gastric cancer, GPS was not associated with the incidence of complications [95].

Variations of GPS have been successfully tested. In a large group of 236 gastric cancer patients who underwent gastrectomy, high-sensitivity GPS after surgery was a significant prognostic factor for overall survival while the pre-operative level was less informative [93]. Modified Glasgow prognostic score was an independent prognostic factor for overall and cancer-free survival in 102 consecutive gastric cancer patients treated with resection [97]. Modified GPS was independent predictor of cancer-specific survival in 120 surgically treated gastric cancer patients [98]. The role of modified GPS in stage IV gastric cancer was confirmed by Mimatsu et al., who evaluated cancer-specific survival in 42 patients at stage IV, treated by palliative gastrectomy and chemotherapy. The modified GPS was associated with cancer-specific survival [99]. Pre-operative modified GPS retained prognostic value in elderly patients [92]. Assessing 1710 surgically treated patients with gastric cancer, modified GPS was associated with postoperative mortality [94]. However, high-sensitivity modified GPS was found to be superior prognostic predictor for overall survival compared to modified GPS having especially high prognostic importance in stage I and IV [100].

By some study designs, the informativity of GPS has been estimated lower. In comparison with NLR-PLR score, modified Glasgow prognostic index was not an independent prognostic factor for survival of stage I–I gastric cancer patients [31]. In 224 patients receiving chemotherapy for advanced gastric cancer, NLR and diffuse type histology were independent prognostic factors for overall survival while GPS was not. However, the median survival still was significantly longer in patients having GPS 0 in contrast to those having GPS 1 or 2 [24].

GPS and its variations retain the prognostic value in advanced cases. In 402 patients with advanced gastric adenocarcinoma treated by palliative chemotherapy, GPS was an independent predictor of overall survival [101]. GPS was an independent predictor of cancer-specific and progression-free survival in 83 patients having advanced gastric cancer and receiving chemotherapy. In low GPS group, favourable response to chemotherapy can be obtained [102]. In patients affected by stage IV gastric cancer and treated by chemotherapy, higher modified Glasgow prognostic score was associated with shorter overall survival (along with lower level of albumin, elevated concentration of C-reactive protein, high absolute number of neutrophilic leukocytes and elevated NLR). In multivariate analysis, modified Glasgow prognostic score was identified as an independent prognostic factor along with NLR, presence of lymph node metastasis and histological subtype [69]. In 68 patients affected by advanced gastric cancer and treated by chemotherapy with or without irradiation, high GPS predicted shorter survival [103]. High GPS was an independent prognostic factor in 384 inoperable advanced or metastatic gastric cancer patients treated with chemotherapy. The value of GPS was higher than that of NLR, PLR or CRP [68]. In 125 patients with recurrent or metastatic gastric cancer placed on single agent chemotherapy because of poor performance status, GPS had independent prognostic value [104]. In 91 metastatic or recurrent gastric cancer patients treated by palliative chemotherapy, GPS was significantly associated with survival. The differences were also biologically remarkable: the median survival was 12.3 months if GPS was 0 but only 2.9 if GPS was 2 [105].

Recently, a meta-analysis was carried out including 14 studies and 5579 gastric cancer patients. High GPS was significantly associated with poor overall survival (hazard ratio 1.51; 95% CI 1.37– 1.66), and disease-free survival (HR 1.45; 95% CI = 1.26–1.68) as reported by Zhang et al. [106].

Glasgow prognostic score has been further developed into different complex scores. Thus, complex predictive score regarding survival was elaborated, based on NLR and modified Glasgow prognostic score in patients with metastatic gastric adenocarcinoma treated by chemotherapy, after independent prognostic value of both parameters was justified in a group of 256 patients [26]. The design of studies devoted to GPS in gastric carcinoma is summarised in **Table 5**.
