Table 3. Pooled and subgroup analysis of main results for the meta-analysis of overall survival (OS).
Categories | Trials (Patients) | HR (95% CI) | I2 (%) | Ph | Z | P |
---|---|---|---|---|---|---|
OS | 15 (4238) | 1.89 (1.56-2.29)R | 62.9% | 0.001 | 6.58 | < 0.001 |
Cancer type | ||||||
CRC | 3 (1050) | 2.25 (1.40-3.61)R | 72.5% | 0.026 | 3.35 | 0.001 |
NSCLC | 2 (477) | 1.77 (1.33-2.35) | 0.0% | 0.891 | 3.91 | < 0.001 |
HCC | 2 (300) | 2.25 (1.47-3.43) | 0.0% | 0.521 | 3.75 | < 0.001 |
BC | 2 (977) | 1.52 (1.23-1.88) | 0.0% | 0.804 | 3.83 | < 0.001 |
CC | 2 (315) | 2.00 (0.85-4.74) | 18.6% | 0.268 | 1.58 | 0.114 |
EC | 2 (188) | 2.56 (1.72-3.79) | 44.4% | 0.180 | 4.67 | < 0.001 |
Others | 2 (931) | 1.22 (1.03-1.44)R | 61.3% | 0.108 | 2.29 | 0.022 |
Clinical stage | ||||||
I–IV | 7 (2302) | 1.65 (1.33-2.04)R | 57.1% | 0.030 | 4.53 | < 0.001 |
I–III | 4 (693) | 2.48 (1.60-3.85)R | 64.1% | 0.039 | 4.04 | < 0.001 |
II–III | 2 (928) | 1.76 (1.33-2.32) | 0.0% | 0.775 | 3.98 | < 0.001 |
I–II | 2 (315) | 2.00 (0.85-4.74) | 18.6% | 0.268 | 1.58 | 0.114 |
Study region | ||||||
Eastern Asia | 10 (2336) | 1.89 (1.45-2.45)R | 62.0% | 0.005 | 4.77 | < 0.001 |
Europe | 5 (1902) | 1.92 (1.43-2.60)R | 65.9% | 0.020 | 4.28 | < 0.001 |
Blinding status * | ||||||
Yes | 13 (3938) | 1.86 (1.51-2.28)R | 65.5% | 0.001 | 5.93 | < 0.001 |
NR | 2 (300) | 2.25 (1.47-3.43) | 0.0% | 0.521 | 3.75 | < 0.001 |
Sample size | ||||||
≥ 100 | 11 (3884) | 1.80 (1.45-2.23)R | 65.9% | 0.001 | 5.34 | < 0.001 |
< 100 | 4 (354) | 2.24 (1.67-2.99) | 0.0% | 0.423 | 5.43 | < 0.001 |
NOS score | ||||||
≤ 6 | 8 (1032) | 2.38 (1.94-2.93) | 18.4% | 0.285 | 8.23 | < 0.001 |
> 6 | 7 (3206) | 1.43 (1.28-1.60) | 46.4% | 0.082 | 6.14 | < 0.001 |
Blinding status represented that the evaluation of the tumor-troma ratio was blinded to the clinical outcomes.
OS overall survival, CRC colorectal cancer; NSCLC non small cell lung cancer; HCC hepatocellular carcinoma; BC breast cancer; CC cervival cancer; EC esophagus cancer; NOS Newcastle-Ottawa Scale.