Treatments of Disease Relapse after Allogeneic Stem Cell Transplantation Focusing on Donor Lymphocyte Infusion.
Liu Hong-tao,Liu Dai-hong,Huang Xiao-jun,Andrew Artz,Michael R. Bishop
DOI: https://doi.org/10.3760/cma.j.issn.0366-6999.20130572
IF: 6.133
2013-01-01
Chinese Medical Journal
Abstract:Hematopoietic stem cell transplantation (SCT) is a potentially curative treatment for patients with hematologic malignancies, such as acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS). There has been tremendous progress in the past several decades in allogeneic SCT with better outcomes through improvements in supportive care, expansion of stem cell donor options (HLA-matched unrelated donors (MUD), haploidentical related donors, and cord blood units (CBUs) etc.), and introduction of better tolerated reduced intensity conditioning (RIC) regimens.1-4 T-cell depletion (TCD) is another approach that enhances tolerability by reducing acute and chronic graft versus host disease (GVHD), which results in tremendous transplant related morbidity and mortality. We have employed invivo TCD with alemtuzumab (Campath) for over 10 years at the University of Chicago and confirmed lower rates of acute and chronic GVHD and similar overall survival (OS).5 Recent observational registry studies6 and now even prospective studies have further established that T-cell depletion through either ATG or alemtuzumab enables better GVHD free survival.7 Although relapse in general remains highly problematic after transplantation,8 particularly for those with active disease at SCT, T-cell depletion further increases rate of relapse. The relapse rate for patients with AML undergoing RIC allo-SCT is in the range of about 25% at one year without T cell depletion and in the range of 30%-40% at one year and 40%-50% at 4 or 5 years with T cell depletion.4-6,9,10 There has been very little progress in reducing the incidence of relapse following allogeneic SCT. T-cell depletion may enable patients to successfully undergo transplant with a low incidence of GVHD, setting the platform to allow approaches that harness the immune system to prevent relapse after transplantation. TREATMENT OPTIONS FOR RELAPSED AMLAFTER SCT The current options of management for disease relapse after allo-SCT include withdrawal of immune suppression, chemotherapy, second allogeneic transplant, cytokine and adoptive cell therapy and donor lymphocyte infusion (DLI). Withdrawal of immunosuppression can be performed in all patients irrespective of hematopoietic stem cell source, but the efficacy is very limited, especially in acute leukemia.11 Chemotherapy For relapsed acute leukemia, both conventional chemotherapy and newer biological agents are able to induce significant remission rates, but long-term survival is very rare. The use of a tyrosine kinase inhibitor (TKI), such as dasatinib and nilotinib, in patients with Ph+ acute lymphoblastic leukemia (ALL) or 5-azacytidine for relapsed AML may have particular benefit. Sorafenib and other FLT3 inhibitors, such as AC220, have demonstrated preliminary activity in a small number of patients with relapsed FLT3 mutated AML.12 A retrospective analysis from the Fred Hutchinson Cancer Research Center (FHCRC) demonstrated that around 30% of relapsed AML patients after allo-SCT could enter CR after chemotherapy, but the median disease free survival (DFS) was very short at 9.7 months.13 The response to chemotherapy with/without DLI was highly influenced by time to relapse after allo-SCT, leading to the current practice standard to offer standard chemotherapy, with and without DLI, only in patients who relapse more than 3 to 6 months after allo-SCT.13-15 Our own data revealed similar results, except that almost all patients relapsing after T-cell depletion with alemtuzumab were able to receive treatment, probably in part due to low rates of GVHD and high-tolerability of the regimens,16 providing further support for the concept that T-cell depletion is a tolerable regimen and platform for subsequent pre-emptive therapy. Second allogeneic transplant Patients who have failed the initial allo-SCT have chemorefractory disease, making additional chemotherapy unlikely to be curative. Historically, the role of second allogeneic SCT has been limited by unacceptable relapse rates and high mortality rates, depending on previous therapies, age, and time from first transplantation. Second SCT has been repeatedly demonstrated to provide some benefits in a small subset of patients, who have remission more than 12 months after the first SCT, and who are able to achieve complete remission (CR) prior to second SCT.17-19 In a large Center for International Blood and Marrow Transplant Research (CIBMTR) retrospective study of patients with hematologic malignancies undergoing second allogeneic SCT, transplant-related mortality was 30% and the relapse rate was 42%, yielding an overall survival rate of 28% at 5 years post-SCT.19 A recent retrospective registry study was performed by the Acute Leukemia Working Party of The European Group for Blood and Marrow Transplant (EBMT) that analyzed the treatment options, risk factors, and outcomes of adults with relapsed AML after RIC allo-SCT.20 The estimated 2-year OS from relapse after RIC allo-SCT was 14%, which is comparable to the results of AML relapse after standard myeloablative conditioning. Importantly, long-term survival was only achieved in patients who achieved CR by cyto-reductive therapy, followed either by donor lymphocyte infusion or second SCT as consolidation.20 Novel approaches with immune-therapies Innovative immunotherapeutic approaches are currently under investigation. Non-specific ex vivo activation and expansion through co-stimulation of donor T cells have been used safely, with intriguing graft-versus-tumor (GVT) responses.21 Donor derived leukemia-specific cytotoxic T cells could also be used as adoptive immunotherapy.22 Cooper et al23 generated T lymphocytes engineered to express chimeric antigen receptors (CARs) specific for the CD19 molecule that may be able to prevent or treat leukemia relapse in ALL patients. Suppression of negative immune-modulation by anti-CTLA4 antibodies, Ipilimumab, has been explored.24 Vaccine strategies with tumor-specific antigens or modified tumor cells are other promising approaches to generate tumor specific immunity.25,26 These strategies will be effective in the setting of minimal residual disease. DLI FOR RELAPSE AFTER SCT DLI has been proven to induce remissions post-transplant in patients with relapsed hematologic malignancies most consistently observed in patients with CML, and to a lesser extent in acute leukemias.27 General principles of DLI: source, effective cell dose,timing and toxicity Typically, donor lymphocytes are obtained by leukapheresis of un-stimulated peripheral blood (a.k.a., steady-state lymphocytes) which will contain other cell types in addition to CD3+ cells, such as dendritic cells, B cells, monocytic cells, and natural killer (NK) cells, providing a spectrum of allo-reactive cells that might play a role in the GVT effect.28 The optimal CD3 cell dose in DLI is not established. However, some studies have demonstrated that in patients with CML, a dose of 1×107/kg can induce complete donor chimerism and a potent graft versus-leukemia (GVL) effect, in some cases in the absence of clinical GVHD particularly if given at later time points following transplantation.29 On the other hand, Fozza et al30 demonstrated that the incidence of GVHD was not significantly different in patients receiving less than 1×107 CD3+/kg compared with the patients who received doses greater than 1×107 CD3+/kg. The timing of DLI and the interval after DLI required to observe a response to DLI are important factors influencing the effectiveness of this strategy. Patients with relapses occurring more than 6 months post-transplantation have higher chances of responding to DLI.31 In support of the aforementioned data, Choi et al15 found 55% overall survival at one year in patients who were treated for relapse which occurred greater than six months post-transplantation, as opposed to 0% survival at one year in patients treated for relapse that occurred within six months following SCT. Disease response following DLI can be seen between 40 days and up to one year following DLI. Another predictor of DLI success is the tumor burden at the time DLI is administered. Patients with evidence of molecular relapse at the time of DLI have better responses even in malignancies not typically viewed as responsive to DLI, such as ALL, which might support careful screening of patients for detection of molecular relapse31 and offer further support for prophylactic approaches. The major complications with DLI are the development of GVHD and cytopenias; marrow aplasia has also been seen but is quite rare. Acute GVHD develops in up to 40%-60% of patients who receive DLI. The development of GVHD does not always correlate with GVT activity.32 The time interval between SCT and DLI therapy appears to influence the likelihood of developing GVHD. A small dose of 1×105 T cells/kg can induce GVHD if administered on the day of transplant,33 yet a dose of 1×107 T cells/kg can be given at 12 months post-transplant without GVHD development.29 Other factors which make GVHD more likely to occur include donor sex mismatch (female donor to male recipient), advanced patient age and mismatch at the minor histo-compatibility antigen level.34 Aplasia is now a relatively infrequent complication of DLI. It is often transient, but in some cases may require hematopoietic stem cell rescue. It was reported historically in 15%-20% of treated CML patients with an associated mortality rate of about 5%. Aplasia is more common in hematological relapse of CML, possibly due to poor donor myeloid reserve, and is rarely reported in patients with exclusively cytogenetic or molecular relapse35,36 or in those treated for low levels of recipient mixed chimerism. Overall, DLI is an effective form of immunotherapy in patients with CML who relapse following SCT, with remission rates of approximately 80%. The results in patients with acute leukemias and MDS are disappointing with remission rates in 15%-25% of patients, and often the responses are not durable. Strategies to avoid DLI-associated toxicity and improveoutcome of DLI Escalated dose regimen (EDR) Administration of DLI as a single bolus of cells collected from a single leukapheresis containing variable numbers of CD3+ T cells is referred to as a bulk dose regimen (BDR) and this approach is associated with a high incidence of GVHD.35-37 The EDR approach is fundamentally different in that the DLI product is quantitated for CD3+, CD4+ and CD8+ T-cell numbers and is then administered in multiple small aliquots with a dose escalation over time. In this way, the minimum cell dose needed to achieve disease remission is administered and with more modest cell doses, the likelihood of GVHD may be reduced.29 One study in CML comparing BDR and EDR approaches demonstrated equivalent remission rates with both schedules, but a significantly lower incidence of GVHD in the EDR cohort.36 It is critical when using the EDR schedule to wait for an adequate interval between DLI doses to allow for assessment of response and toxicity. The optimum intervals between doses are yet to be defined, but Dazzi et al36 reported that shorter intervals (rather than total cell dose) leads to a higher incidence of GVHD. Manipulation of DLI products In order to decrease the incidence of GVHD, many methods have been attempted including depletion of alloreactive T cells by co-incubation of donor lymphocytes with allogeneic recipient stimulator cells followed by targeting with immunotoxin-conjugated antibodies specific for cell-surface activation markers.38,39 CD8+ T cells are thought to be the primary mediators of GVHD in humans while CD4+ T cells are reported to contribute more to the GVT effect.40 For this reason, a number of groups have explored CD8+ T-cell depletion as a strategy to reduce the incidence of GVHD. CD8+ T-cell depletion of the stem cell graft has been reported to reduce the risk of GVHD without a parallel increase in relapse rates in several studies.41-43 Recently, the potential inhibitory role of T regulatory cells within the donor lymphocyte infusion has been studied. Maury et al44 presented interesting data to demonstrate that ex vivo depletion of T regulatory cells from the donor lymphocytes prior to donor lymphocyte infusion improves the graft versus tumor effect of donor lymphocyte after allo-SCT. Among the 17 patients who had no clinical response to the standard donor lymphocyte infusion, two patients developed GVHD and long-term remission of their malignancy after single T regulatory depleted DLI; and additional four out of four patients who received T regulatory depleted DLI after lymphodepletion chemotherapy developed GVHD that was associated with a partial or complete and durable remission. The preliminary results from the study showed depletion of T regulatory cells from donor lymphocytes could be a safe and effective way to overcome allo-reactivity resistant to induce GVT effects, which might offer a rational therapeutic approach for cancer immunotherapy. Other manipulations under study include engineered tumor-reactive T cells expressing either HLA-restricted, heterodimeric TCRs or CARs that recognize native cell-surface antigens. Second generation CARs often comprise an antibody binding motif and a CD28-CD3 dual signaling receptor which facilitates T-cell activation and expansion following stimulation.45 On the other hand, Porter et al21 demonstrated that infusion of "ex vivo" activated donor lymphocytes (using anti-CD3 and anti-CD28 coated beads) in patients with a range of hematologic malignancies led to responses where conventional DLI had been disappointing. A total of 17 patients were evaluated and eight achieved CR. The incidence of GVHD in this cohort compared favorably with that of conventional DLI. Infusion of G-CSF-mobilized peripheral blood progenitorcells Infusion of G-CSF mobilized peripheral blood progenitor cells (G-CSF mobilized DLI (G-DLI)) instead of steadystate donor lymphocytes was also studied to prevent disease relapse after allo-SCT.15,46,47 Huang et al46 reported that infusion of G-CSF-mobilized peripheral blood progenitor cells produced superior disease-free survival than infusion of traditional donor lymphocytes with no difference in the incidence of severe GVHD. In order to decrease the incidence of GVHD from G-CSF mobilized DLI, short term prophylactic immunosuppression with cyclosporine or methotrexate for 2-4 weeks was given to 54 patients before G-DLI. Using immunosuppressive agents for 2-4 weeks reduced G-DLI-associated acute GVHD without influencing relapse and survival.48 Choi et al15 demonstrated that cyto-reduction chemotherapy followed by G-CSF-mobilized peripheral blood progenitor cell infusion was capable of inducing durable remissions in a proportion of relapsed AML patients with relatively long post-SCT remission duration. A recent retrospective analysis of 67 patients who received G-CSF-mobilized DLI or traditional DLI for the treatment of relapse of malignant diseases or poor donor engraftment post transplant revealed similar rates of GVHD and improvement in donor engraftment, suggesting that G-CSF mobilized DLI may be substituted for conventional donor lymphocytes when such cells are available.47 Infusion of G-CSF mobilized DLI has also been used by the researchers at the Institute of Hematology, Peking University to treat disease relapse after mismatched and haploidentical stem cell transplant.49 DLI preceded by chemotherapy especially in acuteleukemia Use of chemotherapy appears to improve the results of DLI. Response rates vary from 10% to 60%, with higher response rates than those reported for DLI alone.14,50 The EBMT reported a retrospective analysis of 399 patients with AML in first hematologic relapse after transplant and found that in the DLI subgroup, having fewer blasts in the BM (<35%), female sex, presence of favorable cytogenetics, and CR at the time of DLI were covariates associated with improved survival.50 The benefit of chemotherapy prior to DLI is suggested here by the 2-year survival >50% for patients that received DLI in CR. Patients with recurrent AML after allogeneic transplantation have been treated with chemotherapy prior to DLI. Chemotherapy was administered because of rapidly progressive disease or in an attempt to debulk patients prior to DLI. Levine et al14 conducted a prospective trial of chemotherapy and G-CSF-stimulated DLI. Patients who emerged from the chemotherapy and DLI in complete remission had a 2-year overall survival rate of 41%. Studies from Japan and Korea treating relapsed acute leukemia with chemotherapy followed by DLI reported an overall complete response rate of 33%, with a 31% estimated overall survival at 24 months.15,51 Lymphodepletion prior to DLI An earlier study in CML patients demonstrated that lymphodepletion, achieved with cyclophosphamide and fludarabine, would promote in vivo expansion of the infused donor lymphocytes, enhancing their immunologic effects possibly due to suppression of regulatory T cells. However, patients developed significantly more grades II to IV (60% vs. 24%, P=0.01) and grades III to IV GVHD (47% vs. 14%, P=0.01) with greater GVHD lethality, preventing the routine use of lymphodepletion prior to DLI.52 The same group at the University of Minnesota recently demonstrated that lymphodepleting chemotherapy followed by DLI could achieve durable complete remission with tolerable GVHD in patients with AML, or MDS/MPD. These results demonstrate that DLI after lymphodepleting chemotherapy results in frequent CRs, and the lower DLI dose regimen improved the tolerability of this therapeutic approach, with modest rates of severe GVHD.53 Similar results were obtained from a study in Europe.54 The best combination of lymphodepletion regimens, and DLI doses should be further determined in a larger cohort of patients before this approach will be more widely accepted. PROPHYLACTIC DLI AFTER SCT Because DLI seems to be most effective for patients with minimal residual disease, the role of prophylactic DLI (pDLI) for high risk patients after SCT has been explored, especially in the setting of T-cell depleted SCT with increased relapse rates. Several studies have examined the utility of pDLI to minimize tumor recurrence in the myeloablative55-60 and in the RIC setting.61-66 Prophylactic DLI in myeloablative T cell depleted SCT In the myeloablative setting, the largest series were reported by Montero et al.60 One hundred and thirty-eight patients with hematologic malignancies received myeloablative T cell depleted peripheral blood stem cell transplant from an HLA-identical sibling donor. One hundred and twelve patients with GVHD grade <2 received 1 or 2 donor lymphocyte infusions of (10-50) ×106 CD3+ cells/kg between days 45 and 100. Overall survival (OS), relapsefree survival, relapse, and transplant-related mortality (TRM) were 58%, 46%, 40%, and 20% respectively, after a median follow-up of 4 years. Fifty-three (39%) and 21 (15%) patients developed Grade II-IV and Grade III-IV acute GVHD respectively. Forty-two (36%) had limited and 29 (25%) had extensive chronic GVHD. In multivariate analysis, disease risk was an independent factor for OS and relapse, day-30 lymphocyte count for OS and TRM, and chronic GVHD for OS and relapse. PBSCT with early T cell add back leads to comparable rates of chronic GVHD compared with T cell replete PBSCT. However, the chronic GVHD after T cell add back was associated with less mortality and retains a protective effect in terms of relapse, at least in the standard-risk patients.60 The patients reported by Schaap received planned DLI at a median of 22 weeks (range 12-40 weeks) post-SCT, provided the post-SCT immunosuppression was discontinued for at least two months without evidence of active chronic GVHD and no history of acute GVHD above Grade I.58 Similarly, Nakamura et al57 reported on patients who had planned non-mobilized cryo-preserved DLI of 10×106 CD3/kg at day +45 and a second infusion of 50×106 CD3/kg at day +100. Alternatively, Lee et al59 and Ferra et al56 selected DLI dosing based on the risk of GVHD and/or relapse risk. In summary, the risk of relapse ranged from 18% to 69% with TRM occurring in 6%-52% of cases. This translated into a disease free survival (DFS) that exceeded 40% at two years.55-60 Schaap et al58 compared outcomes of patients receiving DLI with those patients not receiving DLI, and relapse rates were lower resulting in improved LFS. Furthermore, the incidence of acute and chronic GVHD and the risk of TRM do not seem to differ from expected outcomes after conventional transplants without DLI. Prophylactic DLI in non-myeloablative or RIC T celldepleted SCT Several studies have analyzed outcomes of pDLI after RIC.61-66 Barge et al61 reported on 11 patients, who were given planned DLI at six months after RIC MRD SCT with in vitro TCD. The DLI dose, given as unselected mononuclear cells, was based on disease status. Patients with relapse or progression at six months received (10-100) ×106 MNC/kg plus IFN-α compared with only 10×106 MNC/kg for patients with stable disease or mixed chimerism. For the 11 patients receiving DLI, five responded (CR=3; PR=2) and one patient had stable disease. Acute GVHD developed in six patients, chronic GVHD in four patients and GVHD accounted for the death of one patient. In the report from de Lima et al,62 12 patients with anticipated life expectancy of less than six months received fludarabine-melphalan conditioning and sibling PBSC allogeneic SCT for active hematologic malignancies, including AML (n=4), myelodysplasia (MDS, n=1), ALL (n=3), CML (n=3) and MM (n=1). All patients were scheduled to receive non-mobilized DLI at days +30, +60 and +90. Six patients received DLI. Of these, four patients achieved a CR, only one of whom was in CR at 14 months after SCT. The other three patients died due to TRM or due to relapse. Schmid et al64 studied 75 patients with high-risk AML or MDS who received non-myeloablative conditioning with fludarabine, cytarabine, and low-dose total body irradiation. Patients were scheduled to receive prophylactic DLI after 120 days if there was no evidence of GVHD and they were off immunosuppression medications. Of the 75 patients enrolled, only 12 patients were able to receive prophylactic DLI due to early relapse, GVHD and other transplantrelated complications, demonstrating the challenges of this approach. Another study used prophylactic CD8-depleted DLI after reduced-intensity transplant.65 In that study, 11 of 23 patients were able to receive DLI. Patients receiving CD8-depleted DLI demonstrated accelerated immune reconstitution and minimal GVHD. Mixed chimerism is common after reduced-intensity regimens and may be associated with higher risks of relapse following transplantation. DLI has been used successfully in some patients with acute leukemia in order to facilitate conversion to full donor chimerism after reduced-intensity transplantation, and seems to lower the relapse risk, albeit with a significant risk of acute GVHD.63,66 Two recent publications provide additional support for the feasibility and efficacy of pDLI after TCD allo-SCT.67,68 In a prospective study, Liga et al67 reported on 15 out of 56 consecutive patients who underwent allo-SCT with an alemtuzumab-containing regimen (10-20 mg) and received prophylactic DLI at a median of 162 days posttransplantation. Seven received standard myeloablative conditioning while eight received reduced-toxicity myeloablative conditioning for age >55 years old or significant co-morbidities. Six of the eight patients (75%) who received pDLI with mixed chimerism converted to stable, complete donor chimerism and none of the patients who received pDLI relapsed. Forty-seven percent of the DLI recipients developed Grade III-IV GVHD and all four deaths were due to GVHD-related causes. This suggests that while low-dose pDLI after conditioning with alemtuzumab could decrease the risk of relapse, this approach is associated with a relatively high incidence of severe GVHD. On the other hand, Krishnamurphy et al68 reported promising results of pDLI in AML and MDS patient in a retrospective study. They evaluated the efficacy and toxicity of pDLI in 62 patients after TCD (alemtuzumab at 100 mg or ATG at 6 mg/kg) RIC allo-SCT for AML or MDS. Despite 52% of patients receiving pDLI within 6 months post-SCT, the 5-year incidence of GVHD was only 31%. The overall survival of 80% and event-free survival of 65% at five years in the recipients of pDLI support the use of pDLI to safely promote durable remission after TCD RIC allo-SCT for AML or MDS patients. Prophylactic G-CSF-mobilized DLI Researchers at the Institute of Hematology of Peking University also studied the efficacy of prophylactic G-CSF-mobilized DLI in patients after allo-SCT and mismatched/haploidentical stem cell transplant. In a retrospective study, 33 patients with advanced leukemia received G-DLI and short-term immunosuppressive agents after HLA-matched sibling hematopoietic stem cell transplantation. With an 18-month median follow-up, 16 patients had diseasefree survival, and overall survival at one year was 69.0%, demonstrating the safety and efficacy of this approach.69 A similar strategy was also found to increase survival of patients with advanced-stage acute leukemia who receive HLA-mismatched/haploidentical SCT without a significant increase of the incidence of GVHD.70 In a recent published prospective study, Yan et al71 demonstrated that MRD directed preemptive G-DLI, with or without low dose IL-2, significantly decreased the risk of relapse and improved disease free survival in standard-risk acute leukemia patients after allogeneic SCT. Rationale for repetitive DLI A recent study demonstrated that patients with AML or CML in remission following SCT exhibited significant numbers of CD8+ T cells that recognize epitopes derived from leukemia-associated antigens. However, these cells failed to proliferate, release cytokines, or de-granulate in response to antigen-specific stimuli. The use of IL-15 or high-dose IL-2, elimination of CD4+ regulatory T cells, and blockade of PD-L all failed to rescue the activation of these CD8+ T cells in in vitro assays. The mechanism for CD8+ unresponsiveness after SCT seemed to be replicative senescence.72 Multiple sources of chronic stimulation following SCT may contribute to potential T-cell senescence, including GVHD, GVL activity, infection, persistent stimulation of T cells by residual leukemia cells, and slow reconstitution of CD4+ T cells after SCT during homeostatic proliferation to repopulate the T-cell pool. In addition, population dynamics of the T-cell pool after transplant may be influenced by the intensity of the conditioning regimen before transplant, T-cell dose within the graft, immunosuppressive therapy, and the use of donor lymphocyte infusions.72 Donor T cell tolerance or rapid senescence after allo-SCT may contribute to post-transplant relapse. Thus, preserving and maintaining a competent and reactive pool of CD8+ T-cell precursors after allo-SCT by repetitive DLI once a patient achieves remission, may prove efficacious in preventing relapse. RATIONALE TO CONDUCT PROPHYLACTICDOSE-ESCALATION DLI TO PREVENT RELAPSEIN HEMATOLOGIC MALIGNANCIES AFTER TCELL DEPLETED ALLO-SCT Relapse remains a major issue in non-myeloablative T cell depleted allo-SCT, and DLI is a clinically available and established treatment that is most effective for minimal residual disease. Therefore, the role of dose-escalation prophylactic DLI for patients in remission to prevent relapse will be assessed in a pilot study. The study will focus on patients with high risk hematologic malignancies undergoing allo-SCT, including leukemia not in remission, lymphoma not achieving a partial response, or other diseases under poor control. We reviewed the University of Chicago transplant database for patients with high-risk AML or MDS who underwent a T-cell depleted allogeneic transplant. Of the 145 patients with AML or MDS transplanted with active disease, relapse occurred at a median of 128 days (24 to 2364 days), with PFS of 137 days (4 to 3272 days) and OS of 214 days (4 to 3434 days). Patients had received either fludarabine/melphalan/alemtuzumab (flu/mel/campath) or clofarabine/melphalan/alemtuzumab (clo/mel/campath) conditioning regimens (unpublished data). From our prior published data, after flu/mel/campath conditioning, patients with high-risk disease had a 39% probability of disease recurrence, a 39% probability of treatment-related mortality, and a 25% probability of progression-free survival at one year after stem cell transplant.73 In our clo/mel/campath protocol database,74 35 out of 72 patients accrued had high risk, active disease, and had a one year PFS of 31% which is similar to the 25% one year PFS seen with flu/mel/campath. Additionally, the one year relapse rate was 29% for the whole cohort, making the one year relapse rate for the high risk patients comparable to one year relapse rate of 39% in the high risk patients after flu/mel/campath. The cumulative probability of Grade II-IV acute GVHD was 33% in a flu/mel/campath and 22% in a clo/mel/campath study at one year, and one year TRMs were 33% and 26%, respectively.73,74 Published data using fludarabine+bulsufan+alemtuzumab or ATG has also demonstrated comparable survival and relapse results in high risk hematologic malignancies.75,76 However, this novel approach still requires feasibility testing of each step of the process, including enrollment of high-risk patients, early withdrawal of immune suppression, and the ability to escalate DLIs. The clinical protocol has been approved by the IRB at the University of Chicago, and we are actively enrolling patients to the study at the time of submission. CONCLUSIONS Relapse remains the major failure after allogeneic stem cell transplantation. Prevention of disease relapse after allo-SCT has been a challenging research area. While new strategies and novel treatments have been under investigation, DLI has proven effective to treat or prevent disease relapse after allo-SCT, especially if DLI-induced GVHD can be ameliorated through decreased cell dose, escalating dose administration, and short term immunosuppression with DLI. The very low incidence of acute and chronic GVHD after in vivo T cell depletion provides an excellent platform to test different strategies of DLI to prevent relapse after allo-SCT with tolerable toxicities.