CD2 ؉ T lymphocytes obtained from either the donor of bone marrow stromal cells (BMSCs) or a third party were cultured in mixed lymphocyte reactions (MLRs) with either allogeneic dendritic cells (DCs) or peripheral blood lymphocytes (PBLs). When autologous or allogeneic BMSCs were added back to T cells stimulated by DCs or PBLs, a significant and dosedependent reduction of T-cell proliferation, ranging from 60% ؎ 5% to 98% ؎ 1%, was evident. Similarly, addition of BMSCs to T cells stimulated by polyclonal activators resulted in a 65% ؎ 5% (P ‫؍‬ .0001) suppression of proliferation. BMSCinduced T-cell suppression was still evident when BMSCs were added in culture as late as 5 days after starting of MLRs. BMSC-inhibited T lymphocytes were not apoptotic and efficiently proliferated on restimulation. BMSCs significantly suppressed both CD4 ؉ and CD8 ؉ T cells (65% ؎ 5%, [P ‫؍‬ .0005] and 75% ؎ 15% [P ‫؍‬ .0005], respectively). Transwell experiments, in which cell-cell contact between BMSCs and effector cells was prevented, resulted in a significant inhibition of T-lymphocyte proliferation, suggesting that soluble factors were involved in this phenomenon. By using neutralizing monoclonal antibodies, transforming growth factor ␤1 and hepatocyte growth factor were identified as the mediators of BMSC effects. In conclusion, our data demonstrate that (1) autologous or allogeneic BMSCs strongly suppress T-lymphocyte proliferation, (2) this phenomenon that is triggered by both cellular as well as nonspecific mitogenic stimuli has no immunologic restriction, and (3) T-cell inhibition is not due to induction of apoptosis and is likely due to the production of soluble factors. (Blood. 2002;99:3838-3843)

The safety and efficacy of reduced-intensity conditioning (RIC) followed by allogeneic stem cell transplantation (SCT) for relapsed lymphomas remains unresolved. We conducted a prospective, multicentered, phase II trial. A total of 170 relapsed/refractory lymphomas received a RIC regimen followed by SCT from sibling donors. The primary study end point was non-relapse mortality (NRM). Histologies were nonHodgkin's lymphomas (NHL) (indolent (LG-NHL), n ¼ 63; aggressive (HG-NHL), n ¼ 61; mantle cell lymphoma (MCL), n ¼ 14) and Hodgkin's disease (HD, n ¼ 32). Median follow-up was 33 months (range, 12-82). The results show that frequencies were as follows: cumulative NRM at 3 years, 14%; acute and chronic graft-versus-host disease (GVHD) 35 and 52%, respectively; 3-year overall survival (OS), 69% for LG-NHL, 69% for HG-NHL, 45% for MCL and 32% for HD (P ¼ 0.058); and 3-year relapse incidence, 29, 31, 35 and 81%, respectively (Po0.001). Relapse risk differed significantly at 3 years between follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL) (14 versus 46%, P ¼ 0.04). Molecular remission occurred in 94 and 40% (P ¼ 0.002) of patients with FL and CLL, respectively. On multivariate analysis, OS was influenced by chemorefractory disease (hazard ratio (HR) ¼ 3.6), diagnosis of HD (HR ¼ 3.5), and acute GVHD (HR ¼ 5.9). RIC allogeneic SCT is a feasible and effective salvage strategy in both indolent and aggressive NHL

No abstract