Background:Autologous stem cell transplantation and the development of new agents with potent anti-tumor activity have considerably improved the survival of multiple myeloma (MM) patients. However, there is still a high risk of relapse mainly due to the inability of these agents to cure and eliminate definitively the MM cells. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potentially curative treatment particularly for patients with high risk factors but its use is still controversial. We recently demonstrated in a prospective study that long-term outcome in MM patients was better with auto/RIC-allo as compared with auto-HSCT alone (Gahrton et al. , Blood 2013). The debate is still ongoing concerning the best time to propose allogeneic transplantation for MM patients; while there is agreement in the scientific community to perform it preferentially within a clinical trial, the majority of patients in Europe are still treated outside of a clinical trial.The objective of this study is to describe the context of use of allo-HSCT for MM in Europe within EBMT centers over more than two decades with the evaluation of different outcomes.Material and methods:We included in this study a total of 7333 patients who received allo-HSCT between January 1990 and December 2012, 4539 (62%) males and 2794 (38%) females with a median age at allo-HSCT of 51 years (range: 18-78), of which 4726 (64%) have been transplanted after year 2004. We identified 6 groups in this population, patients who received allo-HSCT upfront after induction therapy (Group 1, N=1934), patients who received allo-HSCT maximum 8 months after single auto-HSCT (Group 2, N=1997), patients who received allo-HSCT later than 8 months after single auto-HSCT (Group 3, N=1588), patients who received allo-HSCT maximum 8 months after double auto-HSCT (Group 4, N=588), patients who received allo-HSCT later than 8 months after double auto-HSCT (Group 5, N=930), and finally patients who received allo-HSCT after having received at least 3 different HSCT (Group 6, N=296). The different patient, disease and transplantation characteristics according to each group are detailed in the table.Results:The upfront use of allo-HSCT was seen to decrease after year 2000 to represent 12% of allo-HSCT performed in 2012 while the peak of allo-HSCT use directly after 1 auto-HSCT (within 8 months) was around year 2004 to attend 19% of usage in 2012. Remarkably, allo-HSCT was moreover used at the highest rate during the last years later than 8 months after single auto-HSCT which could be translated in a context of relapse post- first auto-HSCT to reach 33% of usage in 2012. The usage according to groups 4, 5 and 6 was 14%, 17% and 5% in 2012 respectively (Figure).The median overall survival (OS) for groups 1, 2, 3, 4, 5 and 6 was 33, 69, 25, 25, 23 and 15 months respectively with a 5 years OS probability of 39%, 53%, 33%, 31%, 29% and 23% respectively; the median progression-free survival (PFS) was 45, 39, 21, 22, 21 and 13 months respectively with a 5 years PFS probability of 43%, 42%, 26%, 28%, 24%, and 15% respectively. The cumulative incidence of transplant-related mortality for the 6 groups at 3 years was 36%, 20%, 32%, 33%, 36% and 35% respectively. The use of reduced intensity conditioning was associated with a significantly better OS only in group 2 compared to myeloablative conditioning with a median OS of 76 months versus 45 months (p=0.002) respectively. Year of transplantation before or after 2004 did not influence on outcomes.Conclusion:This large retrospective study shows different ways of using allo-HSCT for MM in Europe over a defined time period, it describes the different expected outcomes in each case. Allo-HSCT has not found its place yet in the treatment course of MM patients and the debate should continue based on advantages and disadvantages showed for each group of patients. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.