Results of the network meta-analyses for the primary outcomes (OS and PFS) and grade 3 or grade 4 adverse events
| Primary NMA of RCTs | Sensitivity NMA of RCTs and observational studies | ||
| Overall survival | Probability most effective (%) | HR versus everolimus (95% credible interval) | |
| Lenvatinib+everolimus | 61 | 0.61 (0.36 to 0.96) | 0.61 (0.36 to 0.96) |
| Cabozantinib | 28 | 0.66 (0.53 to 0.82) | 0.66 (0.53 to 0.83) |
| Nivolumab | 10 | 0.74 (0.57 to 0.93) | 0.74 (0.57 to 0.93) |
| Axitinib | – | – | 1.14 (0.95 to 1.37) |
| Sorafenib | – | – | 1.38 (1.12 to 1.68) |
| BSC | 2 | 1.90 (0.61 to 4.53) | 1.90 (0.60 to 4.56) |
| Progression-free survival | Probability most effective (%) | HR versus everolimus (95% credible interval) | |
| Lenvatinib+everolimus | 67 | 0.47 (0.26 to 0.77) | 0.47 (0.26 to 0.77) |
| Cabozantinib | 34 | 0.51 (0.41 to 0.63) | 0.51 (0.41 to 0.63) |
| Axitinib | – | – | 0.84 (0.70 to 1.00) |
| Sorafenib | – | – | 1.17 (0.95 to 1.43) |
| BSC | 0 | 3.06 (2.31 to 3.97) | 3.06 (2.31 to 3.97) |
| Grade 3 or four adverse events | Probability least harmful (%) | OR versus everolimus (95% credible interval) | |
| Lenvatinib+everolimus | 0 | 2.67 (1.05 to 5.68) | – |
| Cabozantinib | 0 | 1.66 (1.18 to 2.27) | – |
| Nivolumab | 100 | 0.40 (0.29 to 0.55) | – |
SC, best supportive care; NMA, network meta-analysis; OS, overall survival; PFS, progression-free survival; RCT, randomised controlled trial.
Numbers in bold are statistically significant.