Impact of Complete Response on Survival with Either Autologous Stem Cell Transplantation or Conventional Chemotherapy: Results of a Pooled Analysis of 5 Phase III Trials in Newly Diagnosed Multiple Myeloma Patients

Patients and Methods Data from NDMM patients enrolled in 5 phase III Italian trials were analysed. Three trials included patients eligible for autologous stem cell transplantation (ASCT): RV-MM-209 (melphalan-prednisone-lenalidomide [MPR] vs high-dose melphalan [Mel200] and ASCT, followed by lenalidomide maintenance vs no maintenance), RV-MM-EMN-441 (cyclophosphamide-lenalidomide-dexamethasone vs Mel200-ASCT, followed by lenalidomide [R] versus lenalidomide-prednisone [RP] maintenance) and  MM-BO2005 (bortezomib-thalidomide-dexamethasone vs thalidomide-dexamethasone as induction/consolidation, followed by dexamethasone maintenance). The two remaining studies included elderly patients ineligible for ASCT: GIMEMA-MM0305 (bortezomib-melphalan-prednisone-thalidomide followed by bortezomib-thalidomide maintenance vs bortezomib-melphalan-prednisone) and EMN01 (MPR vs cyclophosphamide-prednisone-lenalidomide vs lenalidomide-dexamethasone, followed by R vs RP maintenance). The primary objective of the study was the evaluation of the impact of CR on overall-survival (OS) and progression-free survival (PFS) and its relationship with age (young vs elderly patients) and type of treatment (ASCT vs conventional chemotherapy [CC])). Univariate and multivariate analyses of OS and PFS, including ISS, ASCT and type of novel agents used as induction treatment, were performed. Response was treated as a time-dependent variable. A landmark analysis was performed.

Results 2439 NDMM patients were evaluated; the best response was available in 2359 patients. 656 patients achieved a CR or better, whereas 1353 patients achieved a very good partial response (VGPR) or a partial response (PR), and were included in the analysis. After a median follow-up of 44 months,  the 5-year OS was 75% in CR patients as compared with 60% in VGPR/PR patients (HR 0.49, p<0.001), and 5-year PFS was 44% and 22% (HR 0.44, p<0.001), respectively.

Among CR patients, 383 were treated with ASCT and 273 with CC. A trend towards a better 5-year OS was reported in the ASCT group as compared with the CC group (79% vs 69%; HR 0.6,  p=0.09; Figure 1). A significant PFS advantage was observed among CR patients treated with ASCT in comparison with those who received CC (median, 59 vs 47 months; HR 0.54, p=0.008; Figure 2). No significant differences were observed between young and elderly CR patients treated with CC in terms of 5-year PFS (43% vs 41%; HR 0.9, p=0.5) and 5-year OS (73% vs 69%; HR 1.07, p=0.8).

In the multivariate analysis, ASCT confirmed to be an independent predictor of prolonged PFS in CR patients, with a trend towards longer OS, in comparison with CC.

Conclusions

ASCT induced deeper CR that translated into prolonged PFS and OS as compared with CC. No differences were noticed between young and elderly patients achieving a CR with CC.