Single <i>versus</i> tandem high‐dose melphalan followed by autologous blood stem cell transplantation in multiple myeloma: long‐term results from the phase <scp>III GMMG</scp>‐<scp>HD</scp>2 trial

K. Elias(Heidelberg University), Axel Benner(German Cancer Research Center), Uta Bertsch(Heidelberg University), Peter Brossart(University Hospital Bonn), Annette Hänel(Klinikum Chemnitz), Volker Kunzmann(Universitätsklinikum Würzburg), Ralph Naumann(University Hospital Carl Gustav Carus), Kai Neben(Klinikum Mittelbaden), Gerlinde Egerer(Heidelberg University), Anthony D. Ho(Heidelberg University), Jens Hillengaß(Heidelberg University), Marc S. Raab(Heidelberg University), Andreas Neubauer(Universitätsklinikum Gießen und Marburg), Astrid Peyn(Klinikum Bremen-Mitte), Yon‐Dschun Ko(St. Marien-Hospital Bonn), Norma Peter(Carl-Thiem-Klinikum Cottbus), Christof Scheid(University Hospital Cologne), Hartmut Goldschmidt(Heidelberg University)
British Journal of Haematology
March 17, 2016
Cited by 66

Abstract

The prospective, randomized phase III trial GMMG-HD2 aimed at demonstrating non-inferiority of single (Arm A) versus tandem (Arm B) high-dose melphalan followed by autologous transplantation (HDM/ASCT) with regard to 2-year event-free survival (EFS) in newly-diagnosed multiple myeloma (MM) and included 358 evaluable patients [Intention-to-treat population, (ITT), single/tandem HDM/ASCT: n = 177/181]. After a median follow-up of more than 11 years, non-inferiority of single versus tandem HDM/ASCT was demonstrated using the planned non-inferiority threshold of 15% of the 2-year EFS rate. Neither EFS (P = 0·53) nor overall survival (OS) (P = 0·33) differences were observed in the ITT population. In the tandem arm, 26% (n = 47/181) of patients refused a second HDM/ASCT due to non-medical reasons. A per-protocol (PP) analysis, including patients who received the intervention (single/tandem HDM/ASCT: n = 156/93) and patients who did not receive a second HDM/ASCT due to medical reasons (12%, n = 22/181), did not yield differences in EFS (P = 0·61) or OS (P = 0·16). In the ITT and PP set of the tandem arm, the rates of complete responses increased from first to second HDM/ASCT (both P = 0·04). Ten-year OS for the entire ITT was 34% (95% confidence interval: 29-40%). OS after first relapse was significantly shortened in the tandem arm (P = 0·04). In this study single HDM/ASCT was non-inferior to tandem HDM/ASCT in MM.


Related Papers