Six RCTs (n=1,803) were included in the review. Overall the studies were considered to be of good quality with adequate allocation concealment, ITT analysis and sufficient description of withdrawals and drop-outs. Median follow-up ranged from 33 to 92 months, where reported.
There were no statistically significant differences between tandem and single AHCT in either overall survival (HR 0.94, 95% CI 0.77 to 1.14, Χ2=0.04; six RCTs) or event-free survival (HR 0.86, 95% CI 0.70 to 1.05, Χ2=0.02; six RCTs).
A sensitivity analysis that excluded a study that used concomitant thalidomide in the single AHCT group removed the statistically significant heterogeneity in both analyses and produced a statistically significant benefit of tandem AHCT in event-free survival (HR 0.79, 95% CI 0.70 to 0.89; five RCTs), but there remained no significant difference in overall survival.
There was a statistically significantly higher response rate in the tandem AHCT groups (RR 0.79, 95% CI 0.67 to 0.93; four RCTs). Treatment-related mortality was statistically significantly higher in the tandem AHCT groups (RR 1.71, 95% CI 1.05 to 2.79; five RCTs).
Results of other sensitivity analyses were reported, but did not substantially alter the findings of the analyses.
There was no evidence of publication bias.