Table 2.
Between‐studies correlations corresponding to the study‐level surrogate relationship within each treatment contrast for each model, ρ t is across‐treatment correlations describing the treatment‐level surrogacy obtained from the models allowing for exchangeability, and deviance information criteria (DIC) values corresponding to each model fitted to aCRC data. Where only one value is given for the between‐studies correlation within a treatment contrast (models BRMA, 1d and 2d), the parameters are common across the treatment contrasts
| study‐level surrogacy | ||||||
|---|---|---|---|---|---|---|
| model | AB | AC | BC | BD | ρ t | DIC |
| between‐studies correlations | ||||||
| BRMA | ‐0.67 (‐0.85, ‐0.41) | NA | 43.9 | |||
| bvNMA 1a | ‐0.43 (‐0.84, 0.16) | ‐0.79 (‐0.95, ‐0.46) | ‐0.02 (‐0.88, 0.88) | ‐0.28 (‐0.94, 0.65) | NA | 40.5 |
| bvNMA 1b | ‐0.61 (‐0.9, ‐0.07) | ‐0.79 (‐0.95, ‐0.5) | ‐0.3 (‐0.91, 0.65) | ‐0.25 (‐0.89, 0.61) | NA | 41.8 |
| bvNMA 1c | ‐0.6 (‐0.88, ‐0.11) | ‐0.75 (‐0.93, ‐0.44) | ‐0.27 (‐0.89, 0.66) | ‐0.25 (‐0.87, 0.6) | NA | 52.3 |
| bvNMA 1d | ‐0.75 (‐0.9, ‐0.51) | NA | 39.8 | |||
| bvNMA 2a | ‐0.44 (‐0.85, 0.13) | ‐0.78 (‐0.95, ‐0.47) | ‐0.03 (‐0.89, 0.88) | ‐0.28 (‐0.94, 0.67) | ‐0.33(‐0.92, 0.56) | 39.5 |
| bvNMA 2b | ‐0.62 (‐0.9, ‐0.09) | ‐0.78 (‐0.94, ‐0.48) | ‐0.3 (‐0.91, 0.66) | ‐0.28 (‐0.89, 0.59) | ‐0.33 (‐0.92, 0.57) | 40.0 |
| bvNMA 2c | ‐0.6 (‐0.89, ‐0.13) | ‐0.73 (‐0.93, ‐0.42) | ‐0.25 (‐0.89, 0.68) | ‐0.25 (‐0.87, 0.59) | ‐0.33 (‐0.91, 0.56) | 56.6 |
| bvNMA 2d | ‐0.75 (‐0.9, ‐0.5) | ‐0.36 (‐0.92, 0.52) | 38.4 | |||
| intercepts | ||||||
| BRMA | ‐0.05 (‐0.16, 0.06) | |||||
| bvNMA 1a | ‐0.28 (‐0.42, ‐0.13) | ‐0.03 (‐0.18, 0.12) | 0.05 (‐1.26, 0.96) | 0.12 (‐0.32, 0.53) | ||
| bvNMA 1b | ‐0.25 (‐0.39, ‐0.11) | ‐0.03 (‐0.17, 0.11) | 0.13 (‐0.15, 0.42) | 0.11 (‐0.25, 0.48) | ||
| bvNMA 1c | ‐0.25 (‐0.38, ‐0.12) | ‐0.04 (‐0.18, 0.1) | 0.12 (‐0.15, 0.42) | 0.11 (‐0.13, 0.37) | ||
| bvNMA 1d | ‐0.22 (‐0.32, ‐0.11) | ‐0.04 (‐0.16, 0.08) | 0.18 (0.07, 0.29) | 0.16 (‐0.03, 0.35) | ||
| bvNMA 2a | ‐0.26 (‐0.4, ‐0.11) | ‐0.03 (‐0.18, 0.11) | 0.07 (‐0.8, 0.95) | 0.13 (‐0.16, 0.44) | ||
| bvNMA 2b | ‐0.23 (‐0.37, ‐0.09) | ‐0.03 (‐0.17, 0.11) | 0.12 (‐0.16, 0.42) | 0.12 (‐0.14, 0.39) | ||
| bvNMA 2c | ‐0.23 (‐0.36, ‐0.1) | ‐0.04 (‐0.18, 0.09) | 0.11 (‐0.17, 0.41) | 0.12 (‐0.09, 0.35) | ||
| bvNMA 2d | ‐0.2 (‐0.31, ‐0.1) | ‐0.04 (‐0.16, 0.08) | 0.16 (0.06, 0.27) | 0.16 (‐0.01, 0.33) | ||
Note: A, chemotherapy alone; B, antiangiogenic treatments targeting vascular endothelial growth factor (anti‐VEGF) therapies + chemotherapy; C, epidermal growth factor receptor inhibitor (EGFRi) + chemotherapy; D, EGFRi + anti‐VEGF therapies + chemotherapy
Abbreviations: BRMA, bivariate random effects meta‐analysis; bvNMA, bivariate network meta‐analysis.