Myeloablative T cell-depleted alloSCT with early sequential prophylactic donor lymphocyte infusion is an efficient and safe post-remission treatment for adult ALL

Bone Marrow Transplantation (2014) 49, 287–291 & 2014 Macmillan Publishers Limited All rights reserved 0268-3369/14 www.nature.com/bmt ORIGINAL ARTICLE Myeloablative T cell-depleted alloSCT with early sequential prophylactic donor lymphocyte infusion is an efficient and safe post-remission treatment for adult ALL This article has been corrected since Online Publication and a corrigendum has also been published M Eefting1, CJM Halkes1, LC de Wreede2, CM van Pelt1, S Kersting1, EWA Marijt1, PA von dem Borne1, R Willemze1, H Veelken1 and JHF Falkenburg1 The prognosis of adult patients with ALL remains unsatisfactory. AlloSCT is associated with a beneficial GVL response mediated by donor T cells. However, GVHD results in substantial mortality and long-term morbidity. T-cell depletion (TCD) of the graft reduces the severity of GVHD, but is associated with an increased relapse rate after alloSCT. Therefore, early sequential donor lymphocyte infusion (DLI) is likely to be necessary for a successful GVL reaction. Twenty-five adult ALL patients (10 Ph þ ALL) were eligible for early DLI after initial disease control with myeloablative TCD-alloSCT in first CR (CR1), if active GVHD was absent at 3–6 months after alloSCT. Patients with a sibling donor or an unrelated donor were scheduled for 3.0  106 CD3 þ cells/kg or 1.5  106 CD3 þ cells/kg, respectively, at 6 months after alloSCT. Three patients died before evaluation (one early relapse). Five patients had active GVHD. Fourteen of the remaining seventeen patients received DLI (median time-to-DLI: 185 days). Overall, only 17% required long-term systemic immunosuppression for GVHD. With a median follow-up after TCD-alloSCT of 50 months, 2-year survival probability was 68% (95% confidence interval (CI) 49–87%). In conclusion, myeloablative TCD-alloSCT with early sequential DLI is an efficient and safe post-remission treatment for adult ALL patients in CR1. Bone Marrow Transplantation (2014) 49, 287–291; doi:10.1038/bmt.2013.111; published online 12 August 2013 Keywords: alloSCT; T-cell depletion; ALL; donor lymphocyte infusion; DLI INTRODUCTION The prognosis of adult patients with ALL remains unsatisfactory. Despite high CR rates after induction chemotherapy, 5-year OS ranges between 40 and 60% with standard consolidation/ intensification and maintenance treatment including hematopoietic SCT.1,2 AlloSCT is associated with a beneficial antileukemia effect as compared with autologous SCT or chemotherapy maintenance for adult patients with ALL.3–7 This beneficial effect is due to donor T cells mediating a GVL response, thereby preventing a relapse. However, GVHD and infectious complications result in a substantial non-relapse mortality (NRM) and alloSCT-associated morbidity. Therefore, recommendation of alloSCT from non-sibling donors has remained controversial for patients without high-risk ALL in first CR (CR1).8 T-cell depletion (TCD) of the stem-cell graft reduces the severity of GVHD, but is associated with an increased relapse rate after alloSCT, especially if patients are transplanted in second CR or with refractory disease.2,9,10 Therefore, the concept of prophylactic donor lymphocyte infusion (DLI) has been developed.11 DLI not only decreases relapse rate in ALL patients who are at high risk for relapse, but can also induce molecular remissions in patients with detectable minimal residual disease.11,12 TCD-alloSCT with postponed administration of DLI is associated with a decreased severity of GVHD.11,13 We describe here the results of a treatment strategy with myeloablative conditioning before TCD-alloSCT in CR1 for efficient medium-term leukemia control, followed by early sequential prophylactic DLI to establish a GVL response for definitive prevention of relapse. We postulated that this strategy should limit the incidence of early relapses in conjunction with a low incidence of severe GVHD. MATERIALS AND METHODS Study population All patients who underwent myeloablative TCD-alloSCT for ALL in CR1 at Leiden University Medical Center, between January 2003 and June 2011 were included in this study. All patients gave informed consent. High-risk ALL was defined by high leukocyte count at diagnosis (430  109/L in B-ALL and 4100  109/L in T-ALL), failure to achieve CR after prephase and first induction therapy, and/or unfavorable karyotype at diagnosis, that is, t(9;22), t(4;11) and other 11q23 abnormalities, hypodiploidy or complex abnormalities (X5, excluding hyperdiploidy). Data were analyzed as of October 2012. Transplantation procedure and follow-up The conditioning regimen consisted of CY 60 mg/kg i.v. at days  6 and  5 and TBI in all patients. Patients with a matched sibling donor received 9 Gy TBI on day  1. Patients with a mismatched sibling or an unrelated donor received 9 Gy TBI on day  8 or  7, 30 mg alemtuzumab i.v. divided between days  6 and  5, and cyclosporine 3 mg/kg i.v. (and corresponding oral dose after engraftment) as GVHD prophylaxis from day 1 Department of Hematology, Leiden University Medical Center, Leiden, The Netherlands and 2Department of Medical Statistics and Bioinformatics, Leiden University Medical Center, Leiden, The Netherlands. Correspondence: Professor JHF Falkenburg, Department of Hematology, Leiden University Medical Center, C2-R, Albinusdreef 2, 2333 ZA Leiden, The Netherlands. E-mail: Received 9 January 2013; revised 27 May 2013; accepted 22 June 2013; published online 12 August 2013 Myeloablative T cell-depleted alloSCT with DLI for ALL M Eefting et al 288  1 until day þ 60. In vitro TCD was performed by adding 20 mg of alemtuzumab to the stem-cell product.9 CD34 þ cell dose was determined by flow cytometry. Starting in 2008, patients with Ph þ ALL received 400 mg imatinib per day after engraftment until at least 3 months after DLI. The day of granulocyte engraftment was defined as the first of 3 consecutive days of absolute granulocyte counts 40.5  109/L. GVHD was graded according to modified Glucksberg and Shulman criteria.13,14 BM chimerism was determined every 3 months during the first 2 years after alloSCT, and additionally at 6 weeks after DLI. In sex-matched patient– donor pairs, chimerism was determined on total BM leukocytes, and mononuclear cell or granulocyte fractions after ficoll separation, using a short-tandem-repeat-PCR-based protocol.15 In sex-mismatched patient– donor pairs, FISH analysis was performed on unseparated BM leukocytes until 2007. Thereafter, chimerism analysis was performed analogous to sex-matched pairs. Mixed chimerism was defined as X1% patient cells in any of these cell fractions. Relapse was defined as reappearance of X5% blasts in BM by morphology (hematological relapse), and/or by reappearance of a molecular marker, that is BCR-ABL (molecular relapse). NRM was defined as death in continuous CR. OS was defined as time from alloSCT until death from any cause. Study end points were incidence of acute GVHD, NRM, relapse and OS. All time intervals were calcu (...truncated)