Table 1.
Pooled hazard ratios for subgroup analysis of overall survival.
| Study characteristic | All included studies | Limited stage disease patients only | Extensive stage disease patients only |
|---|---|---|---|
| Study design | |||
| RCT | 0.88 (95% CI 0.65–1.18), n = 5 studies | 0.87 (95% CI 0.65–1.12), n = 2 studies | 0.93 (95% CI 0.02–48.93), n = 2 studies |
| RCS | 0.58 (95% CI 0.54–0.62), n = 106 studies | 0.56 (95% CI 0.5–0.64), n = 70 studies | 0.55 (95% CI 0.48–0.64), n = 24 studies |
| PCS | 0.39 (95% CI 0.28–0.54), n = 7 studies | 0.44 (95% CI 0.24–0.81), n = 5 studies | 0.46 (95% CI 0.22–0.94), n = 1 study |
| nRCT | 1.25 (95% CI 0.75–2.09), n = 2 studies | 1.03 (95% CI 0.54–1.96), n = 2 studies | 1.71 (95% CI 0.79–3.72), n = 1 study |
| Treatment response to first-line therapy | |||
| CR | 0.70 (95% CI 0.48–0.98), n = 13 studies | 0.57 (95% CI 0.45–0.71), n = 10 studies | 1.14 (95% CI 0.00–263.19), n = 2 studies |
| CR/PR | 0.65 (95% CI 0.56–0.75), n = 26 studies | 0.55 (95% CI 0.48–0.61), n = 19 studies | 0.71 (95% CI 0.47–1.08), n = 6 studies |
| CR/PR/SD | 0.55 (95% CI 0.42–0.72), n = 12 studies | 0.59 (95% CI 0.42–0.82), n = 7 studies | 0.49 (95% CI 0.25–0.97), n = 5 studies |
| CR/PR/SD/PD | 0.51 (95% CI 0.34–0.78), n = 7 studies | 0.50 (95% CI 0.29–0.87), n = 5 studies | 0.48 (95% CI 0.46–4.172), n = 2 study |
| NR | 0.57 (95% CI 0.52–0.62), n = 62 studies | 0.56 (95% CI 0.53–0.60), n = 37 studies | 0.56 (95% CI 0.46–0.68), n = 13 studies |
| Use of brain baseline brain CT/MRI | |||
| Yes | 0.60 (95% CI 0.53–0.68), n = 49 studies | 0.56 (95% CI 0.52–0.60), n = 34 studies | 0.60 (95% CI 0.42–0.86), n = 13 studies |
| No, NR, only in a subset of patients | 0.58 (95% CI 0.51–0.64), n = 171 studies | 0.59 (95% CI 0.52–0.66), n = 44 studies | 0.61 (95% CI 0.50–0.73), n = 14 studies |
| MRI confirmation of no brain metastases at restaging | |||
| Yes | 0.75 (95% CI, 0.50–1.12), n = 9 studies | 0.68 (95% CI 0.45–1.02), n = 7 studies | 1.27 (95%CI 0.70–2.29), n = 1 study |
| No | 0.62 (95% CI 0.54–0.71), n = 19 studies | 0.59 (95% CI 0.50–0.71), n = 14 studies | 0.69 (95% CI 0.52–0.91), n = 4 studies |
| NR | 0.57 (95% CI 0.52–0.62), n = 84 studies | 0.58 (95% CI 0.52–0.65), n = 51 studies | 0.57 (95% CI 0.48–0.69), n = 20 studies |
| Use of platinum-based therapy | |||
| Yes | 0.56 (95% CI 0.51–0.62); n = 74 | 0.58 (95% CI 0.50–0.63), n = 51 studies | 0.65 (95% CI 0.51–0.83), n = 14 studies |
| No | 0.59 (95% CI 0.27–1.30); n = 4 studies | 0.55 (95% CI 0.13–2.32), n = 3 studies | NA |
| Not administered to all patients | 0.58 (95% CI 0.45–0.74); n = 15 studies | 0.57 (95% CI 0.43–0.76), n = 10 studies | 0.61 (95% CI 0.29–1.29), n = 5 studies |
| NR | 0.65 (95% CI 0.56–0.75); n = 27 studies | 0.72 (95% CI 0.64–0.85), n = 14 studies | 0.56 (95% CI 0.45–0.69), n = 6 studies |
| AHRQa | |||
| Good | 0.67 (95% CI 0.47–0.96), n = 7 studies | 0.68 (95% CI 0.45–1.02), n = 7 studies | 0.66 (95% CI 0.24–1.75), n = 3 studies |
| Fair | 0.55 (95% CI 0.38–0.78), n = 9 studies | 0.62 (95% CI 0.49–0.78), n = 6 studies | 0.67 (95% CI 0.20–2.24), n = 3 studies |
| Poor | 0.57 (95% CI 0.53–0.62), n = 99 studies | 0.56 (95% CI 0.54–0.59), n = 63 studies | 0.55 (95% CI 0.47–0.66), n = 22 studies |
| RoBb | |||
| Low | 0.93 (95% CI 0.02–49.90), n = 2 studies | NA | 0.93 (95% CI 0.02–48.99), n = 2 studies |
| Some concerns | 0.82 (95% CI 0.31–2.13), n = 2 studies | 0.86 (95% CI 0.57–1.58), n = 1 study | NA |
| High | 0.90 (95% CI 0.40–2.05), n = 1 study | 0.90 (95% CI 0.41–1.98), n = 1 study | NA |
AHRQ = Agency for Health Research and Quality; CR = complete response; CT = computed tomography; MRI = magnetic resonance imaging; NA = not applicable; NR = not reported; nRCT = non-randomised controlled trial; PCS = prospective cohort study; PR = partial response; SD = stable disease; RCS = retrospective cohort study; RCT = randomised controlled trial; RoB = risk of bias.
a
AHRQ only reported for non-randomised trials.
b
RoB only reported for randomised controlled trials.