Fifty-two RCTs involving 9,387 patients were included.
Combined surgery and chemotherapy versus surgery (14 trials, 4,357 patients).
There was significant heterogeneity between trials (p=0.02) and between categories of chemotherapy (p=0.004), so results are reported for the predefined chemotherapy categories rather than overall. For cisplatin-based regimens, the HR was 0.87 (95% confidence interval, CI: 0.74, 1.02, p=0.08). For other regimens, the HR was 0.89 (95% CI: 0.72, 1.11, p=0.30). For the older long-term alkylating agents, the HR for surgery plus chemotherapy versus surgery was 1.15 (95% CI: 1.04,1.27, p=0.005).
Combined surgery, radiotherapy and chemotherapy versus combined surgery and radiotherapy (7 trials, 807 patients).
The overall HR was 0.98 (p=0.76). The test for heterogeneity was not significant (p=0.73). For the 6 cisplatin-based trials, the HR was 0.94 (95% CI: 0.79,1.11, p=0.46).
Radical radiotherapy plus chemotherapy versus radical radiotherapy (22 trials, 3,033 patients).
The overall HR was 0.90 (p=0.006). The heterogeneity between trials and chemotherapy categories was not significant (p=0.56 and p=0.59 respectively). Eleven of the trials used cisplatin-based drugs, and the HR for these trials was 0.87 (95% CI: 0.79, 0.96, p=0.005). The HRs for all other chemotherapy categories were not significantly different from 1.
Supportive care combined with chemotherapy versus supportive care (11 trials, 1,190 patients).
There was significant heterogeneity both between trials (p<0.0001) and between chemotherapy categories (p=0.003). For the 8 cisplatin-based trials the HR was 0.73 (95% CI: 0.63, 0.85, p<0.0001). The results for the trials using long-term alkylating agents suggested a detrimental effect of chemotherapy, but since there were only 2 trials the result is not significant. HR was 1.26 (95% CI: 0.96, 1.66, p=0.095).
The subgroup analyses were undertaken for cisplatin-based regimens only. No evidence of any differences was found.