Indirect Comparison of the Efficacy of Melphalan-Prednisone-Bortezomib Relative to Melphalan-Prednisone-Thalidomide and Melphalan-Prednisone for the First Line Treatment of Multiple Myeloma

Melphalan-prednisone (MP) combination has been considered a standard of care for front line treatment of multiple myeloma in patients non eligible for transplant. Melphalan-prednisone-bortezomib (MPV) combination has been approved in the United States in patients non eligible for high-dose chemotherapy (HD-C) and has recently received a positive opinion from the CHMP in Europe. Melphalan-prednisone-thalidomide (MPT) was approved in Europe in patients >65 or not eligible for HD-C. There is no head-to-head trial directly comparing MPV to MPT. The objective of the current study was to compare the efficacy of MPV to MP and MPT as first line treatment of multiple myeloma in patients non eligible for transplant. Six randomized placebo controlled trials investigating the efficacy of MPT (5) and MPV (1) relative to MP were identified with a systematic literature review. The endpoints of interest were overall survival (OS), progression free survival (PFS) and overall and complete response. Relative efficacy estimates of MPT versus MP as obtained from the MPT-MP trials were combined with meta-analysis techniques and simultaneously indirectly compared with the relative efficacy of MPV versus MP from the MPV-MP trial (VISTA). This adjusted indirect comparison was performed with Bayesian fixed and random effects models. As compared to frequentist approach, Bayesian meta-analysis offers a more informative summary of the likely value of efficacy after observing the data and allows for direct probabilistic inferences. Of the three interventions compared, there was an 81% probability that MPV was the most efficacious intervention in terms of overall response and a >99% probability in terms of complete response. With MPV a patient was two times more likely to show a complete response than with MPT (Relative Risk=2.15; 95%Credible Interval (CrI): 0.99–4.45). Both MPV and MPT showed greater OS than MP (HR=0.61; 95%CrI: 0.42–0.88 and HR=0.61; 95%CrI: 0.47–0.78 respectively); the indirect comparison showed similar efficacy in terms of OS between MPV and MPT (MPV vs MPT: Hazard Ratio=1.00; 95%CrI 0.64–1.57). Both MPV and MPT also displayed greater PFS than MP (MPV versus MP: HR=0.61; 95%CrI 0.49–0.76 and MPT versus MP HR=0.51; 95%CrI 0.41–0.63 respectively) and showed similar efficacy (MPV vs MPT: HR=1.19; 95%CrI: 0.87–1.63). In this study, both MPV and MPT are more efficacious than MP in terms of response, PFS and OS. MPV is expected to result in a greater complete and overall response than MPT. No difference in OS or PFS was displayed. Further analyses will need to be undertaken once evidence base data is more mature.