Comparison of hematopoietic cell transplant conditioning regimens for hemophagocytic lymphohistiocytosis disorders

Background Allogeneic hematopoietic cell transplantation for hemophagocytic lymphohistiocytosis (HLH) disorders is associated with substantial morbidity and mortality. Objective The effect of conditioning regimen groups of varying intensity on outcomes after transplantation was examined to identify an optimal regimen or regimens for HLH disorders. Methods We studied 261 patients with HLH disorders transplanted between 2005 and 2018. Risk factors for transplantation outcomes by conditioning regimen groups were studied by Cox regression models. Results Four regimen groups were studied: (1) fludarabine (Flu) and melphalan (Mel) in 123 subjects; (2) Flu, Mel, and thiotepa (TT) in 28 subjects; (3) Flu and busulfan (Bu) in 14 subjects; and (4) Bu and cyclophosphamide (Cy) in 96 subjects. The day 100 incidence of veno-occlusive disease was lower with Flu/Mel (4%) and Flu/Mel/TT (0%) compared to Flu/Bu (14%) and Bu/Cy (22%) (P < .001). The 6-month incidence of viral infections was highest after Flu/Mel (72%) and Flu/Mel/TT (64%) compared to Flu/Bu (39%) and Bu/Cy (38%) (P < .001). Five-year event-free survival (alive and engrafted without additional cell product administration) was lower with Flu/Mel (44%) compared to Flu/Mel/TT (70%), Flu/Bu (79%), and Bu/Cy (61%) (P = .002). The corresponding 5-year overall survival values were 68%, 75%, 86%, and 64%, and did not differ by conditioning regimen (P = .19). Low event-free survival with Flu/Mel is attributed to high graft failure (42%) compared to Flu/Mel/TT (15%), Flu/Bu (7%), and Bu/Cy (18%) (P < .001). Conclusions Given the high rate of graft failure with Flu/Mel and the high rate of veno-occlusive disease with Bu/Cy and Flu/Bu, Flu/Mel/TT may be preferred for HLH disorders. Prospective studies are warranted. Allogeneic hematopoietic cell transplantation for hemophagocytic lymphohistiocytosis (HLH) disorders is associated with substantial morbidity and mortality. The effect of conditioning regimen groups of varying intensity on outcomes after transplantation was examined to identify an optimal regimen or regimens for HLH disorders. We studied 261 patients with HLH disorders transplanted between 2005 and 2018. Risk factors for transplantation outcomes by conditioning regimen groups were studied by Cox regression models. Four regimen groups were studied: (1) fludarabine (Flu) and melphalan (Mel) in 123 subjects; (2) Flu, Mel, and thiotepa (TT) in 28 subjects; (3) Flu and busulfan (Bu) in 14 subjects; and (4) Bu and cyclophosphamide (Cy) in 96 subjects. The day 100 incidence of veno-occlusive disease was lower with Flu/Mel (4%) and Flu/Mel/TT (0%) compared to Flu/Bu (14%) and Bu/Cy (22%) (P < .001). The 6-month incidence of viral infections was highest after Flu/Mel (72%) and Flu/Mel/TT (64%) compared to Flu/Bu (39%) and Bu/Cy (38%) (P < .001). Five-year event-free survival (alive and engrafted without additional cell product administration) was lower with Flu/Mel (44%) compared to Flu/Mel/TT (70%), Flu/Bu (79%), and Bu/Cy (61%) (P = .002). The corresponding 5-year overall survival values were 68%, 75%, 86%, and 64%, and did not differ by conditioning regimen (P = .19). Low event-free survival with Flu/Mel is attributed to high graft failure (42%) compared to Flu/Mel/TT (15%), Flu/Bu (7%), and Bu/Cy (18%) (P < .001). Given the high rate of graft failure with Flu/Mel and the high rate of veno-occlusive disease with Bu/Cy and Flu/Bu, Flu/Mel/TT may be preferred for HLH disorders. Prospective studies are warranted.