Allogeneic Hematopoietic Cell Transplant (Allo-HCT) for Advanced Diffuse Large B-Cell Non-Hodgkin's Lymphoma: The University of Michigan Experience

The treatment of advanced refractory DLBCL remains a major challenge. High dose therapy with autologous stem cell transplant (HDT/ASCT) has a limited role only in chemosensitive relapsed disease, with a curative potential of 25-50%. No therapy has proven to be effective long-term for advanced chemo-refractory DLBCL. Although allo-HCT has curative potential, historically, the outcomes remain unsatisfactory due to high treatment related mortality (TRM) after myeloablative conditioning (MAC), high relapse rate after reduced-intensity conditioning (RIC) regimens and the concern for a less potent graft-versus-lymphoma effect (GVL) A total of 97 DLBCL patients (pts) [M64:F33] undergoing allo-HCT at UM between 1996 and 2011 were retrospectively evaluated. The median age was 52 yrs (range, 23-70). Seventy-four pts (76%) received a MAC [CVB in 49 pts (50%), FluBu4 in 13 pts (13%), others in 12 (12%)]; 23 pts received RIC with FluBu2 (24%); 79 pts (81%) were 8/8 HLA-matched. Only 18 pts (19%) had prior HDT/ASCT. Donor sources were 58 related, 6 mismatched (MM), and 39 unrelated, 12 MM. Stem cell sources were peripheral blood in 89%, bone marrow in 10% and cord blood in 1%. GVHD prophylaxis was tacrolimus/ mini-methotrexate in 71 pts (73%), tacrolimus/MMF in 16 pts (17%) and others in 10 pts (10%). Fifty-five pts (57%) had chemo-sensitive disease at transplant; 22 pts (23%) had transformed DLBCL.

With a median follow-up time of 12 months (range, 1.6-194), the cumulative incidences of grade 2-4, grade 3-4 acute GVHD, chronic GVHD, TRM, relapse, OS and PFS were not statistically different between the MAC and RIC cohorts (Table). Relapse rate remained high in both groups. Chemosensitivity, donor source and type of DLBCL had no impact on survival.

In summary, allogeneic HCT for advanced DLCBL is feasible, with acceptable TRM. Although, relapse is a major issue, allo-HCT provides these heavily pretreated pts with a platform to combine the GVL effect with other strategies to maximize the potential of long-term disease-free survival.

MAC (n=74)		RIC (n=23)		p-value
Endpoint	Estimate (95% CI)	Endpoint	Estimate (95% CI)
aGVHD gr 2-4   100 days   180 days	50% (39%-62%) 54% (43%-66%)	aGVHD gr 2-4   100 days   180 days	39% (22%-63%) 48% (29%-70%)	0.49
aGVHD gr 3-4   100 days   180 days	34% (24%-46%) 36% (27%-49%)	aGVHD gr 3-4   100 days   180 days	13% (4%-36%) 22% (9%-46%)	0.12
cGVHD    1 year    2 years	32 %( 22%-47%) 35 %( 24%-49%)	cGVHD    1 year    2 years	38% (18%-70%) 38% (18%-70%)	0.90
TRM    100 days    1 year    2 years	12% (6%-22%) 18% (11%-29%) 23% (15%-35%)	TRM    100 days    1 year    2 years	9% (2%-31%) 22% (9%-46%) 22% (9%-46%)	0.29
Relapse    1 year    2 years	33%(23%-45% 33 %( 23%-45%)	Relapse    1 year    2 years	43% (26%-67%) 43% (26%-67%)	0.84
OS    1 year    2 years	53% (42%-65%) 43% (33%-56%)	OS    1 year    2 years	39% (24%-65%) 35% (20%-61%)	0.29
PFS    1 year    2 years	46% (36%-59%) 40% (31%-53%)	PFS    1 year    2 years	30% (16%-56%) 30% (16%-56%)	0.41