Utility of the Modified and High‐Sensitivity Modified Glasgow Prognostic Scores for Hypopharyngeal Squamous Cell Carcinoma

Abstract

Objective To determine whether the modified Glasgow prognostic score (mGPS) and high-sensitivity mGPS (HS-mGPS) could predict outcomes among patients with hypopharyngeal squamous cell carcinoma (HSCC). Study Design Retrospective cohort study. Setting Affiliated university hospital. Methods We reviewed the records of 115 patients with histologically confirmed HSCC between March 2007 and December 2019. Univariate and multivariable Cox proportional hazard analyses were performed for overall survival (OS) and disease-free survival (DFS). Results The 5-year OS rates were 84.0% for the mGPS0 group, 47.8% for the mGPS1 group, and 17.9% for the mGPS2 group ( P < .0001), while the 5-year OS rates were 86.7% for the HS-mGPS0 group, 69.0% for the HS-mGPS1 group, and 22.2% for the HS-mGPS2 group ( P < .001). The mGPS and HS-mGPS were both associated with OS in the univariate analyses, although only the HS-mGPS was independently associated with OS (hazard ratio, 2.68 [95% CI, 1.19-6.05]; P < .05). The 5-year DFS rates were 75.8% for the mGPS0 group, 53.0% for the mGPS1 group, and 13.8% for the mGPS2 group ( P < .001), while the 5-year DFS rates were 79.8% for the HS-mGPS0 group, 56.8% for the HS-mGPS1 group, and 11.6% for the HS-mGPS2 group ( P < .001). The mGPS and HS-mGPS were both associated with DFS in the univariate analyses, although only the HS-mGPS was independently associated with DFS (hazard ratio, 2.35 [95% CI, 1.03-5.37]; P < .05). Conclusion Our study suggests that the HS-mGPS is useful as prognostic factor in HSCC.

Utility of High-Sensitivity Modified Glasgow Prognostic Score in Cancer Prognosis: A Systemic Review and Meta-Analysis

“…The sample sizes ranged from 70 to 1625 (median: 163). Among them, 13 studies [ 7 , 14 , 15 , 16 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ] (4651 cases) used multivariable analysis and 4 studies [ 30 , 31 , 32 , 33 ] (1177 cases) used univariable analysis in survival outcome measurement. This meta-analysis analyzed multiple cancer types, including gastrointestinal cancer (3 gastric cancers [ 16 , 30 , 31 ], 1 hepatocellular carcinoma [ 28 ], 1 esophageal cancer [ 14 ], 1 gallbladder cancer [ 32 ], and 1 colorectal cancer [ 29 ]), head and neck cancer (1 head and neck cancer [ 22 ], 1 oropharyngeal cancer [ 26 ], 1 hypopharyngeal cancer [ 27 ], and 1 oral cancer [ 7 ]), soft tissue sarcoma (3 soft tissue sarcoma [ 21 , 23 , 33 ] and 1 neuroblastoma [ 24 ]), non-small-cell lung cancer [ 15 ] (n = 1), and prostate cancer [ 25 ] (n = 1).…”

“…Among them, 13 studies [ 7 , 14 , 15 , 16 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ] (4651 cases) used multivariable analysis and 4 studies [ 30 , 31 , 32 , 33 ] (1177 cases) used univariable analysis in survival outcome measurement. This meta-analysis analyzed multiple cancer types, including gastrointestinal cancer (3 gastric cancers [ 16 , 30 , 31 ], 1 hepatocellular carcinoma [ 28 ], 1 esophageal cancer [ 14 ], 1 gallbladder cancer [ 32 ], and 1 colorectal cancer [ 29 ]), head and neck cancer (1 head and neck cancer [ 22 ], 1 oropharyngeal cancer [ 26 ], 1 hypopharyngeal cancer [ 27 ], and 1 oral cancer [ 7 ]), soft tissue sarcoma (3 soft tissue sarcoma [ 21 , 23 , 33 ] and 1 neuroblastoma [ 24 ]), non-small-cell lung cancer [ 15 ] (n = 1), and prostate cancer [ 25 ] (n = 1). A total of 12 studies reported cancer stage according to the TNM staging system, while 6 studies reported stage according to other staging systems, including the French Federation of Cancer Centers Sarcoma Group grading system, the Barcelona Clinic Liver Cancer staging system, the Japanese Gastric Cancer Treatment Guidelines, and the International Neuroblastoma Staging System.…”

“…Fifteen studies [ 7 , 14 , 15 , 16 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ] (20 datasets and 5545 patients) reported an association between the HS-mGPS and OS. The pooled analysis detected a significant association between a high HS-mGPS and poor OS in cancer patients (HR = 2.17; 95% CI: 1.80–2.60; p < 0.001) with moderate heterogeneity ( I 2 = 65.5%; p H < 0.001; Figure 2 ).…”

“…Three studies [ 7 , 26 , 27 ] with 524 patients reported a prognostic role of the HS-mGPS for DFS. The pooled results indicated the HS-mGPS to be an independent factor that can predict adverse DFS (HR = 1.96; 95% CI: 1.48–2.58; p < 0.001) with no heterogeneity ( I 2 = 0%; p H = 0.459; Figure 3 a).…”

“… Forest plot indicating the pooled HRs to assess the influence of the HS-mGPS on OS. Abbreviations: HR, hazard ratio; CI, confidence interval; HS-mGPS, high-sensitivity modified Glasgow Prognostic Score [ 7 , 14 , 15 , 16 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. …”