A Matching-Adjusted Indirect Treatment Comparison of Daratumumab-Bortezomib-Melphalan-Prednisone Versus Lenalidomide-Dexamethasone Continuous, Lenalidomide-Dexamethasone 18 Months, and Melphalan-Prednisone-Thalidomide

• Published in: Blood 2018-11, Vol.132 (Supplement 1), p.3551-3551
• Main Authors: Dimopoulos, Meletios A, Cavo, Michele, Mateos, Maria-Victoria, Facon, Thierry, Heeg, Bart, van Beekhuizen, Sophie, Nair, Sandhya, Pisini, Marta, Lam, Annette, Slavcev, Mary
• Format: Article
• Language: English
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2018-99-113517

▪ Introduction: Daratumumab-bortezomib-melphalan-prednisone (D-VMP) is a novel treatment regimen for newly diagnosed multiple myeloma patients who are ineligible for autologous stem cell transplantation (ASCT). D-VMP is currently being investigated and compared to bortezomib-melphalan-prednisone (VMP) in the ALCYONE trial, and demonstrated a significant benefit for D-VMP over VMP in terms of progression-free survival (PFS) [1]. Recently, a systematic literature review and a network meta-analysis (NMA) were conducted to compare D-VMP against all other available treatments in that indication. However, the reliability of an NMA on overall survival (OS) may be limited due to several factors. First, OS data from the ALCYONE trial is still immature. Second, within the OS network of evidence, four MP trials showed similar PFS, yet large differences were observed in OS for MP. This is an indication of effect modification due to subsequent therapies and thus violation of the similarity assumption. The four MP trials are a bridge to key approved comparators lenalidomide-dexamethasone (Rd) and melphalan-prednisone-thalidomide (MPT), which have been investigated in the FIRST trial. To overcome these challenges faced in the OS NMA, an unanchored matching-adjusted treatment comparison (MAIC) was conducted to assess the relative OS between D-VMP and Rd continuous, Rd 18, and MPT. In addition, the analysis was extended to also include PFS. Methods: Individual patient data from the ALCYONE trial were adjusted to match the aggregated baseline characteristics of the FIRST trial. Patient level OS and PFS data of FIRST trial were generated based on the Guyot algorithm. The considered baseline characteristics were median age, male (%), ECOG status (%), ISS state (%), IgG serum (%), creatinine clearance (%), and high-risk cytogenetic profile. After matching patients, a Cox proportional hazard model was fitted. In both analyses, the null hypothesis of no difference in relative treatment effect in OS and PFS between D-VMP and each of the three treatment arms of the FIRST trial was tested using the log-rank test. To assess the impact of matching patients, a naïve unanchored comparison between D-VMP and FIRST treatments arms was conducted in addition to the MAIC. Results: The results of the MAIC are summarized in Table 1. Significant results are denoted with an asterisk. OS for D-VMP is significantly longer than for both MPT and Rd 18. When comparing OS for D-VMP with Rd continuous,