Non-myeloablative stem cell transplantation is an effective strategy to reduce the incidence of acute (a) GVHD, a major risk factor for the development of chronic (c) GVHD. However, in NST there is evidence of at least a similar incidence of cGVHD compared with myeloablative transplants despite the lower incidence of aGVHD.

Objective: To evaluate the incidence, natural history and outcome of cGVHD following NST in patients (pts) with Non-Hodgkin’s Lymphoma.

Methods: All pts were treated with NST with Fludarabine/Cyclophosphamide/Rituximab conditioning regimen (I. Khouri. Blood 2001), and tacrolimus/methotrexate for GVHD prophylaxis. Risk factors for the occurrence of de novo cGVHD were evaluated using Cox regression analysis.

Results: 78 pts were treated: follicular lymphoma (n=47; 60%), diffuse large cell (n=16; 21%), and mantel cell lymphoma (n=15; 19%). Median age was 53 years (21–68). Donors were matched sibling (69/78; 88%), other related (n=3; 4%) and matched unrelated (n=6; 8%). Stem cell source was peripheral blood for related transplants and bone marrow for unrelated donor transplants. Median number of prior chemotherapy regimens was 3 (1–10). All pts but 5 (6%) had chemo-sensitive disease prior to transplant. Median follow up among survivors was 35 months (3–70). Grade I–IV aGVHD occurred in 25 pts (32%), II–IV in 15 (19%), and III–IV in 5 pts. There were 30 cases of cGVHD (3-yr cumulative incidence of 46%; 95% CI 35–60) including 20 de novo (3 yr cumulative incidence of 39%; 95% CI 27–56) and 10 with relapsing/progressive disease. Median time of onset of de novo GVHD was 277 days (126–798). Overall survival since the diagnosis of cGVHD was 92% (95% CI 54–99) in the novo group compared with 43% (95% CI 11–72) in the relapsing/progressive group. cGVHD was the primary cause for the only death in the de novo group and for 4 of 5 deaths in the relapsing/progressive group. Patients’ characteristics (age, gender, donor-recipient histocompatibility, diagnosis, prior autologous transplantation, number of prior chemotherapy regimens, disease status prior to transplant (CR/PR), B2microglobulin, LDH, IPI, and patient/donor CMV serostatus), graft characteristics (CD34 and CD3 positive cells dose) and post-transplant course (CMV reactivation within 100 days post NST, chimerism at day 30 and day 90 post NST, time to tacrolimus taper, and use of DLI) were not significantly associated with the occurrence of de novo cGVHD.

Conclusions: Despite a reduction in aGVHD, the incidence of cGVHD was comparable to that described after myeloablative regimens. Furthermore, there was a high incidence of de novo cGVHD, but a low cGVHD-related mortality. The lack of conclusive risk factors, including those traditionally associated with acute GVHD, emphasizes the potentially different pathophysiology and epidemiology of these 2 diseases.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author