Sixty-five RCTs (15,110 elderly patients) were included in the review. The median number of patients was 123. Only six RCTs included more than 500 patients. The included studies assessed 64 direct comparisons between 42 different induction regimens. Overall 7,447 (49.3%) of patients achieved complete remission.
Meta-analysis of 14 direct comparisons was possible; all other direct comparisons were represented by a single trial. Eleven of these meta-analyses showed statistically significant heterogeneity between trial effects. Fourteen of the 64 direct comparisons showed a statistically significant benefit in complete remission (full details given in the paper).
Network meta-analysis indicated a benefit in complete remission for the addition of all trans-retinoic acids (OR 1.93, 95% CI 1.06 to 3.49) or lomustine (OR 1.76, 95% CI 1.08 to 2.88) to idarubicin plus cytarabine. Statistically significantly fewer patients achieved complete remission with no treatment, clofarabine, daunorubicin plus topotecan and two different schedules of gemtuzumab compared with daunorubicin plus cytarabine at standard doses (OR ranged from 0.01 to 0.15; none of the confidence intervals approached 1.00).
There were no statistically significant differences between treatment regimens in induction deaths; incidence of these ranged from 0% to 40%. There were no statistically significant differences between treatments in median disease-free survival (range zero to 23 months) or median overall survival (range 1.1 to 17 months). There was an increase in overall survival with date of trial publications (from 1.8 months in 1976 to 17 months in 2010). There were no statistically significant differences in myelotoxicity or time to hospital discharge.