High-Dose Rituximab in the Conditioning Regimen Before Allogeneic Stem Cell Transplantation Reduces the Incidence of Acute Gvhd in B-Cell Lymphomas

Reduced-intensity conditioning (RIC) followed by allogeneic stem cell transplantation (alloSCT) is an effective salvage therapy for relapsed lymphomas. The present GITMO study is a prospective multicenter phase II trial for patients affected by relapsed CD20 positive lymphomas. Compared with the previous thiotepa/fludarabine/cyclophosphamide GITMO protocol (Leukemia 2007), the thiotepa dose is increased, and high-dose Rituximab is included in the regimen to improve the outcome and possibly modulate the incidence of acute GVHD.

Primary end-point was 1-year progression-free survival; secondary endpoints were non-relapse mortality and incidence of acute and chronic GVHD.

Fifty-seven patients (pts) were enrolled so far in the study and 49 are evaluable for analysis. Treatment plan consisted of high-dose R (500 mg/ms on day -6) followed by thiotepa (12 mg/kg), fludarabine (60 mg/kg) and cyclophosphamide (60 mg/kg). Graft-versus-host disease (GVHD) prophylaxis included cyclosporine and mini-methotrexate; ATG (7.5 mk/kg) was only added for pts allografted from one antigen mismatched sibling or unrelated donors. Histopathological subtypes included 24 aggressive (HG) (n= 17 diffuse large B-cell lymphomas, n= 7 mantle cell lymphomas) and 25 indolent lymphomas (LG) (n= 13 follicular lymphomas, n= 12 small lymphocytic/chronic lymphocytic leukemia). Patients were allografted from related siblings (SIB) (n= 32 matched, n=1 one single mismatched) or unrelated donors (UD) (n=11 matched, n=5 mismatched). All the pts had chemosensitive disease (n=20, 41% in complete remission) and 26 (53%) came from a failed autoSCT.

Results: At a median follow-up of 13 months (range, 5–44 months), 36 pts are alive [n=27 (75%) in CR] and 13 died from any cause [n=6 for non-relapse mortality (NRM), n=7 for disease progression]. All the patients engrafted (94% had full donor chimerism at 3 months). The cumulative incidence (CI) of NRM was 13% at 1 year: 9% vs 19% for SIB and MUD (p=0.3), and 9% versus 16% for for LG and HG (p=0.3), respectively. In total only 11 of 49 pts had acute GVHD (n=8 grade II, n=3 grade III) with an estimated CI of 21% at 100 days. In the previous GITMO study the incidence was 35% with SIB only. Forty pts are evaluable for chronic GVHD with an estimated CI of 41% and 47% at 1 and 2 year, respectively (n=11 limited, n=3 extensive). Infections after engraftment requiring hospitalization or intravenous treatment were evaluable in 46 pts (n=3 excluded for early death). The overall incidence of infections was 58% (n=27) including 5 pts experienced sepsis and 10 pts pneumonia. Preliminary data on immune-reconstitution at 1 year showed: 1) low number of circulating B cells (median CD19+/ul: 129/ul) with an expansion of naive cells (IgD+, CD27-); 2) the median value of IgM was 89 mg/dl whereas IgG and IgA remained at low levels.

The CI of relapse was 26% and 37% at 1 year and 2 years, respectively. In the indolent and aggressive groups, OS estimates at 2 years were 79% (95%CI, 52%-91%) and 61% (95 CI, 38%-77%) and PFS estimates were 53% (95%CI, 23%-76%) and 48% (95% CI, 27%-66%), respectively.

The present data suggest that the administration of high-dose R is feasible and causes an unexpected reduction of the incidence of acute GVHD without increasing the NRM and the incidence of severe infections complications. Complete data evaluating the effects of R on immune reconstitution are ongoing.

No relevant conflicts of interest to declare.