The Egger’s test result was highly significant (t = 6.60, p < 0.0001), indicating potential asymmetry and suggesting the likelihood of publication bias ( Supplementary File S26 ). - Similarly, in the OS multivariate analysis, the funnel plot showed some studies lying outside the triangular area, consistent with significant heterogeneity.
The test for subgroup differences showed a trend toward statistical significance ( p = 0.07), thus no strong evidence of differential effects based on the method used to define sarcopenia was observed ( Supplementary File S6 ). - Overall survival (multivariate analyses, aHR): In the subgroup analysis for OS based on multivariate regression analyses’ reported outcomes, the pooled aHR for the 29 studies defining sarcopenia using SMI was 1.59 (1.40; 1.80), with substantial heterogeneity (I 2 = 79%, τ 2 = 0.06, p < 0.01).
Although the point estimates between subgroups were different, the test for subgroup differences did not reach statistical significance ( p = 0.10) ( Supplementary File S18 ). - Overall survival (multivariate analyses, aHR): The curative setting subgroup (23 studies) showed a pooled aHR of 1.74 (95% CI, 1.46–2.08), with significant heterogeneity (I 2 = 72%, τ 2 = 0.11, p < 0.01).