Allogeneic Bone Marrow Transplantation From Unrelated Donors in Multiple Myeloma: A Study from the Italian Bone Marrow Transplantation Donor Registry

Abstract 2009

To evaluate the role of allografting from unrelated donors in the treatment of myeloma we conducted a retrospective study through the Italian Bone Marrow Transplantation Donor Registry. Overall, from 2000 through 2009, 196 myeloma patients, median age 51 years (32–67), for a total of 199 allografts, were transplanted from an unrelated donor at 34 Centers in Italy. Fifty-two (28.1%%), 69 (37.3%), and 64 (34.6%) patients were prepared for transplant with a myeloablative, a reduced-intensity and a non-myeloablative conditioning respectively. Patient characteristics of the 3 cohorts are reported in Table 1.

Cumulative incidence of acute grade II-IV graft-versus-host-disease (GVHD) was 46.4% whereas chronic GVHD was 45.1%. There was no difference in GVHD incidence among the 3 cohorts defined by type of conditioning. Complete and partial remissions in patients who survived at least 3 months post-transplant were 27% and 28% for an overall response rate of 55%. At a median follow up of 32 (0–118) months post-transplant, in the entire study population, median OS from diagnosis was 70.6 months while OS and EFS from the allograft were 18.9 and 14.9 months. Overall, the cumulative incidence of transplant related mortality (TRM) was 29.6% at 1 year and 32.4% at 5 years post-transplant. OS from diagnosis and EFS from transplant were 70.6 and 28.2 months; 66.8 and 9.1 months; and 111.9 and 22.4 months in patients who respectively underwent a myeloablative, a reduced-intensity and a non-myeloablative transplant. One-year and 5-year TRM was 33.3% and 35.7%, 32.2% and 34.4%, and 22.1% and 26.5% respectively. Univariate and multivariate analyses, assessed by multivariate Cox proportional hazards models, were performed for the following variables: number of previous chemotherapy lines (1,2 vs. ≥3), disease status at transplant, HLA- matched antigens (10/10 vs. 9/10 vs. ≤8/10), recipient/donor gender combinations, hematopoietic cell source (peripheral blood vs. bone marrow), conditioning (myeloablative vs. reduced-intensity vs. non-myeloablative), use of anti-thymoglobulin in the conditioning, acute GVHD, chronic GVHD, best response post-transplant, year of transplant (2000–02 vs. 2003–05vs. 2006–09). By univariate analysis, lower number of chemotherapy lines before the allograft, disease status at transplant, a fully HLA-identical donor, the use of peripheral hematopoietic cells rather than bone marrow were statistically significant variables for better OS whereas disease status at transplant, a fully HLA-identical donor, chronic GVHD (either limited or extensive) were statistically significant for better EFS. However, by multivariate analysis, only the development of chronic GVHD (HR 0.50; p<0.001) and a better response post-transplant (HR 2.11; p<0.03) were significantly associated with longer OS whereas chronic GVHD was the only variable associated with better EFS (HR 0.32; p<0.001). Acute GVHD was associated with both poorer OS (HR 2.35; p<0.001)) and EFS (HR 3.19; p<0.001). In conclusion, with a degree of caution given the retrospective nature of this study, there appears to be a strong association between both limited and extensive chronic GVHD and graft-vs.-myeloma effects. However, long term disease control remains an issue independent of the conditioning employed. Prospective trials may allow to define which patient category may most benefit from an unrelated donor allograft.