A systematic search of MEDLINE, EMBASE, BioSciences Information Service, and the Cochrane Library databases was conducted from initiation to September 2016. Abstracts published by international congresses (2014-2016) and bibliographies of pertinent systematic reviews and meta-analyses were also searched. Eligible studies consisted of randomized controlled trials (RCTs) or long-term follow-up studies with >1 treatment arm assessing the efficacy or safety of MM therapies. An NMA was conducted by using Bayesian fixed effect mixed-treatment comparisons. Outcomes considered were hazard ratios for PFS and odds ratios for overall response rate (ORR).
In total, 108 articles reporting 27 RCTs were included in the NMA. Data formed 2 evidence networks: RCTs with DRd and RCTs with DVd. Primary analysis of PFS found that DRd and DVd had a higher probability of being the best treatments (probability, 0.997 and 0.999, respectively) and had the lowest risk of progression or death than other treatments approved by the US Food and Drug Administration for the treatment of MM. Results from sensitivity analyses using time to progression as a proxy for missing PFS data were consistent. DRd and DVd also showed improved ORR compared with other treatments. Subgroup analyses of PFS in patients treated with only 1 prior therapy were like the results of the primary analyses.
This NMA provides comparative efficacy for MM treatments not studied in head-to-head RCTs. The NMA suggests that, compared with other approved MM treatments in the United States, DRd and DVd have a higher probability of providing the longest PFS in patients who have received at least 1 prior therapy and in patients who have received only 1 prior therapy.