TABLE 4.
The subgroup analysis according to whether or not surgery and different stages.
| Group | Analysis | N | References | Random-effects model | Heterogeneity | ||
|---|---|---|---|---|---|---|---|
| HR (95%CI) | p | I 2 (%) | p | ||||
| GPS = 0 vs. GPS = 1 or 2 | Subgroup 1 | ||||||
| Surgery | 7 | (17, 21, 23, 24, 27, 29, 32) | 1.79 (1.08–2.97) | .024 | 79.00 | ≤.001 | |
| Non-surgery | 7 | (7, 14, 15, 18, 20, 28, 33) | 1.59 (1.21–2.10) | ≤.001 | 73.20 | ≤.001 | |
| Subgroup 2 | |||||||
| Stage I–II | 2 | (17, 24) | 1.72 (.92–3.22) | .087 | 44.50 | .180 | |
| Stage I–III | 3 | (21, 27, 32) | 1.56 (.45–5.35) | .482 | 91.80 | ≤.001 | |
| Stage III–IV | 4 | (7, 14, 15, 33) | 1.73 (1.43–2.09) | ≤.001 | 18.40 | .299 | |
| Stage I–IV | 5 | (18, 20, 23, 28, 29) | 1.41 (.78–2.52) | .251 | 79.60 | ≤.001 | |
| GPS = 0 vs GPS = 1 | Subgroup 1 | ||||||
| Surgery | 2 | (26, 30) | 6.80 (.38–122.45) | .194 | 95.90 | ≤.001 | |
| Non-surgery | 4 | (19, 22, 25, 31) | 1.47 (.95–2.26) | .082 | 75.60 | .006 | |
| Subgroup 2 | |||||||
| Stage I–III | 2 | (26, 30) | 6.80 (.38–122.45) | .194 | 95.90 | .009 | |
| Stage III–IV | 5 | (16, 19, 22, 25, 31) | 1.56 (1.05–2.31) | .026 | 70.70 | ≤.001 | |
| GPS = 0 vs GPS = 2 | Subgroup 1 | ||||||
| Surgery | 2 | (26, 30) | 5.31 (.37–75.45) | .218 | 95.10 | ≤.001 | |
| Non-surgery | 4 | (19, 22, 25, 31) | 1.83 (.73–4.57) | .195 | 94.70 | ≤.001 | |
| Subgroup 2 | |||||||
| Stage I–III | 2 | (26, 30) | 5.30 (.37–75.32) | .218 | 93.60 | ≤.001 | |
| Stage III–IV | 5 | (16, 19, 22, 25, 31) | 2.23 (1.17–4.26) | .015 | 84.70 | ≤.001 | |
GPS, Glasgow Prognostic Score; N, number of studies; HR, hazard ratio; 95% CI, 95% confidence interval, p, p-values of Q test; OS, overall survival; VS, versus.