Impact of New Drugs on the Long-Term Follow-Up of Upfront Tandem Autograft-Allograft in Multiple Myeloma.

Before the introduction of "new drugs," we designed a trial in which 162 newly diagnosed myeloma patients were biologically randomized to receive either an autologous stem cell transplant (auto-SCT) followed by a nonmyeloablative allogeneic stem cell transplant (allo-SCT) or a double auto-SCT. Fifty-eight patients in the allo-SCT arm and 46 in the double auto-SCT arm completed the assigned treatment. At a median follow-up of 12.3 years from allo-SCT and 12.1 years from second auto-SCT, median overall survival (OS) was 11.4 in the allo-SCT arm and 3.9 years in the auto-SCT -arm (P = .007), whereas event-free survival was 3.6 and 1.5 years (P < .001), respectively. A subset of allo-SCT patients showed persistent molecular remission. Two-year cumulative incidence of chronic graft-versus-host disease was 67.2%. At 5 years, 39% of these patients were alive, disease-free, and off immunosuppression; 36.6% had relapsed and 12.2% were still on immunosuppression. Thirty-three of 58 patients (allo-SCT arm) and 39 of 46 (auto-SCT arm) relapsed at least once and were rescued with new drugs. In the allo-SCT arm, 2 patients in biochemical relapse did not reach clinical criteria for treatment. Overall 28 (90%) were treated with new drugs and 14 (45%) received donor lymphocyte infusions (DLIs). In 28 of 31 patients (90%) DLIs were given with new drugs. Median OS from first relapse was 7.5 years in the allo-SCT arm and 2 years in the auto-SCT arm (P = .01). Patients who received DLI showed significantly longer OS (hazard ratio, .38; P = .042) as compared with auto-SCT patients. This difference was slightly lower when only allo-SCT patients who did not receive DLIs were considered (hazard ratio, .56; P = .154). In summary, long-term disease-free survival and survival outcomes after treating relapse with new drugs with or without DLIs were better in allo-SCT patients.

Post-relapse survival after haploidentical transplantation vs matched-related or matched-unrelated hematopoietic cell transplantation.

Relapse remains a major cause of mortality among patients receiving allogeneic hematopoietic cell transplantation (HCT). The impact of donor type on post-relapse survival (PRS) has not been widely examined. We compared the survival outcomes for patients relapsing after haploidentical donor transplantation (HIDT) using post-transplant cyclophosphamide with those relapsing after matched-related donor transplantation (MRDT) or matched-unrelated donor transplantation (MUDT) at our institution. Two hundred and thirty-seven consecutive HCT recipients with relapse occurring after HIDT (N=48), MUDT (N=87) and MRDT (N=102) were included in this analysis. Median age was 49 years (19-77 years) and the median time to relapse was 156 days (12-2465) after HCT. HIDT recipients had similar median time to relapse (5.8 vs 4.8 vs 5.5 months, P=0.638) compared with MUDT and MRDT, respectively. One-year PRS was worse among HIDT recipients compared with MRDT and MUDT (17% vs 46% vs 40%, P<0.05). In a multivariate analysis, time to relapse (<3 vs >3 months post transplant), no use of donor lymphocyte infusion (DLI) following relapse, higher Dana Farber disease risk index and HCT comorbidity index scores at the time of transplant and delayed platelet engraftment post transplant were all predictive of worse PRS. This analysis shows that 1-year PRS is inferior among HIDT when compared with MRDT or MUDT. Lower use of DLI after HIDT may have contributed to this inferior survival.

Prolonged overall survival in gastric cancer patients after adoptive immunotherapy.

AIM: To assess the efficacy of immunotherapy with expanded activated autologous lymphocytes (EAALs) in gastric cancer. METHODS: An observational study was designed to retrospectively analyze the clinical data of 84 gastric cancer patients, of whom 42 were treated by EAAL immunotherapy plus conventional treatment and another 42 only received conventional treatment (control group). EAALs were obtained by proliferation of peripheral blood mononuclear cells from patients followed by phenotype determination. Clinical data including age, gender, clinical stage, chemotherapeutic regimens, hospitalization, surgical, radiotherapy, and survival data were collected along with EAAL therapy details and side effects. Patients were followed and the relationship between treatment and overall survival (OS) data obtained for the immunotherapy and control groups were compared retrospectively. The safety of EAAL immunotherapy was also evaluated. RESULTS: After in vitro culture and proliferation, the percentages of CD3+, CD3+CD8+, CD8+CD27+, CD8+CD28+, and CD3+CD16+/CD56+ cells increased remarkably (P < 0.05), while the percentages of CD3+CD4+, CD4+CD25+, and CD3-CD16+/CD56+ (natural killer cells) were overtly decreased (P < 0.05); no significant change was observed in CD4+CD25+CD127- cells (P = 0.448). Interestingly, OS in the immunotherapy group was significantly higher than that in the control group, with 27.0 and 13.9 mo obtained for the two groups, respectively (P = 0.028, HR = 0.573, 95%CI: 0.347-0.945). These findings indicated a 42.7% decrease in the risk of death. In addition, we found that clinical stage and application of EAAL immunotherapy were independent prognostic factors for gastric cancer patients. Indeed, the OS in stage IIIc and IV patients that had received surgery was prolonged after EAAL immunotherapy (P < 0.05). Importantly, in vitro induction and proliferation of EAAL were easy and biologically safe. CONCLUSION: Overall, EAAL adoptive immunotherapy might prolong the OS in gastric cancer patients.

Graft-versus-leukemia effects of transplantation and donor lymphocytes.

Allogeneic transplantation of hematopoietic cells is an effective treatment of leukemia, even in advanced stages. Allogeneic lymphocytes produce a strong graft-versus-leukemia (GVL) effect, but the beneficial effect is limited by graft-versus-host disease (GVHD). Depletion of T cells abrogates GVHD and GVL effects. Delayed transfusion of donor lymphocytes into chimeras after T cell-depleted stem cell transplantation produces a GVL effect without necessarily producing GVHD. Chimerism and tolerance provide a platform for immunotherapy using donor lymphocytes. The allogeneic GVL effects vary from one disease to another, the stage of the disease, donor histocompatibility, the degree of chimerism, and additional treatment. Immunosuppressive therapy before donor lymphocyte transfusions may augment the effect as well as concomitant cytokine treatment. Possible target antigens are histocompatibility antigens and tumor-associated antigens. Immune escape of tumor cells and changes in the reactivity of T cells are to be considered. Durable responses may be the result of the elimination of leukemia stem cells or the establishment of a durable immune control on their progeny. Recently, we have learned from adoptive immunotherapy of viral diseases and HLA-haploidentical stem cell transplantation that T-cell memory may be essential for the effective treatment of leukemia and other malignancies.

Inference for current leukemia free survival.

Donor lymphocyte infusion (DLI) for patients who relapse following an allogeneic stem cell transplant has proved remarkably durable. Because of the potential for second remissions with DLI, the current leukemia free survival (CLFS), which is the probability that a patient has not failed the entire course of the treatment, is becoming of interest to clinical investigators. Based on either a multistate Markov model or a linear combination of Kaplan-Meier estimators, we explore regression models for the CLFS. We focus on the two sample problem and we develop confidence bands for the CLFS or for differences in CLFS as well as a Kolmogorov type hypothesis test using a re-sampling technique. We also examine the use of pseudo-values to make inference on the direct effects of covariates on the CLFS function and we develop a score test for the equality of two CLFS. We illustrate these inference methods on a bone marrow transplant dataset.

Detectable minimal residual disease before allogeneic hematopoietic stem cell transplantation predicts extremely poor prognosis in children with acute lymphoblastic leukemia.

BACKGROUND: The level of minimal residual disease (MRD) prior to allogeneic hematopoietic stem cell transplantation (HSCT) has been shown to be an independent prognostic factor for outcome of pediatric patients with high-risk acute lymphoblastic leukemia (ALL). Retrospective studies which used (semi-) quantitation of clone-specific immunoglobulin/T-cell receptor (Ig/TCR) rearrangements have documented the feasibility and practicality of this technique. This approach has also been disputed due to the occurrence of clonal evolution and generally high MRD levels prior to HSCT. PROCEDURE: In our prospective study, MRD before and after HSCT was monitored using quantitative real-time PCR in a cohort of 36 children with ALL consecutively transplanted in our center between VIII/2000 and VII/2004. RESULTS: In 25 of 36 patients, MRD level prior HSCT was assessed. Seventeen patients were classified as MRD-negative and eight were MRD-positive up to 9 x 10(-2). In MRD-positive subgroup, seven events (six relapses) occurred post-transplant in striking contrast to only one relapse in MRD-negative subgroup (event-free survival (EFS) log-rank P < 0.0001). MRD proved to be the only significant prognostic factor in a multivariate analysis (P < 0.0001). Adoptive immunotherapy including donor lymphocyte infusions in patients with adverse dynamics of MRD after HSCT had only limited and/or temporary effect. Clonal evolution did not present a problem precluding MRD monitoring in any of patients suffering a post-transplant relapse. CONCLUSIONS: We show that MRD quantitation using clonal Ig/TCR rearrangements successfully assesses the risk in pediatric ALL patients undergoing allogeneic HSCT. As our ability to treat detectable MRD levels after HSCT is very limited, alternative strategies for MRD-positive patients prior HSCT are necessary.

KIR ligands and prediction of relapse after unrelated donor hematopoietic cell transplantation for hematologic malignancy.

Recurrent malignancy remains a significant complication after allogeneic hematopoietic cell transplantation (HCT). Efforts to decrease relapse have included donor lymphocyte infusion to stimulate donor anti-recipient T-cell allorecognition of major and minor histocompatibility differences. Recently, alloreactive effects of donor natural killer cell-mediated inhibitory killer immunoglobulin-like receptor (KIR) recognition of recipient HLA-C and -B ligands have been described. We examined KIR ligand effects on risk of relapse in 1770 patients undergoing myeloablative T-replete HCT from HLA-matched or -mismatched unrelated donors for the treatment of myeloid and lymphoid leukemias. KIR ligands defined by HLA-B and -C genotypes were used to determine donor-recipient ligand incompatibility or recipient lack of KIR ligand. Among HLA-mismatched transplantations, recipient homozygosity for HLA-B or -C KIR epitopes predicted lack of KIR ligand and was associated with a decreased hazard of relapse (hazard ratio, 0.61; 95% confidence interval, .043-0.85; P = .004). Absence of HLA-C group 2 or HLA-Bw4 KIR ligands was associated with lower hazards of relapse (hazard ratio, 0.47; 95% confidence interval, 0.28-0.79, P = .004; hazard ratio, 0.56; 95% confidence interval, 0.33-0.97; P = .04, respectively). The decrease in hazard of relapse in patients with acute myelogenous leukemia was similar to that in patients with chronic myelogenous leukemia and acute lymphoblastic leukemia (P = .95). Recipient homozygosity for HLA-B or -C epitopes that define KIR ligands is likely to be a predictive factor for leukemia relapse after myeloablative HCT from HLA-mismatched unrelated donors. This effect was not observed in HLA-identical unrelated transplants.

Graft-versus-host reactions and the effectiveness of donor lymphocyte infusions.

We retrospectively analyzed 83 consecutive recipients of donor lymphocyte infusions (DLI) after allogeneic transplantation for factors associated with disease response and graft-versus-host disease (GVHD). DLI was highly effective in relapsed chronic phase chronic myeloid leukemia (CML), with 71% of patients achieving durable complete remissions (CR). In relapsed acute myeloid leukemia, DLI led to durable CRs in 31% of patients; the rate was <20% in all other diseases. Achieving full donor chimerism and GVHD were predictive of CR. Grade II or higher acute or chronic GVHD occurred in 36 (43%) patients and contributed to death in 13 (16%). Even more patients, 33 (40%), died of their underlying malignancy, including 10 who developed active GVHD. In relapsed CML, most durable CRs occurred without clinically apparent GVHD, yet all responders achieved full donor chimerism, including 6 with coincident normal host hematopoiesis at the time of DLI. Thus, in CML, potent lymphohematopoietic graft-versus-host reactions occurred even in the absence of clinically apparent GVHD; this confirms the ability to dissociate these processes and argues against a leukemia-specific immunologic effect. DLI clearly has efficacy in the treatment of relapsed disease after allogeneic transplantation. However, with the exception of CML, most patients die of their underlying disease because of insufficient antitumor activity even with active GVHD.

Unrelated donor or partially matched related donor peripheral stem cell transplant with CD34+ selection and CD3+ addback for pediatric patients with leukemias.

Unmodified peripheral stem cell transplants are associated with an increased risk of extensive chronic GVHD. T depletion may reduce this risk, but the risk of graft failure or relapse may increase. To decrease the risks of both extensive chronic GVHD and graft failure, we added back a defined dose of CD3+ cells to CD34+ selected PSCs. Twenty-four patients were evaluable for outcome analysis. Donors were unrelated (23) or related (1). Conditioning was thiotepa, cyclophosphamide, and total body irradiation. Cyclosporine was used post transplant. Following CD34+ selection, a total of 5 x 10(5)/kg CD3+ cells were infused. Donors were matched for 12 patients. The median CD34+ dose infused was 7.1 x 10(6)/kg. Engraftment occurred in all patients at a median of 14 days (10-19). Twelve patients are alive in remission 15-34 months (median, 25) post PSCT. GVHD occurred in 17 patients, but was >grade II in only 2. Chronic GVHD occurred in 61.5% of evaluable patients, but was limited to skin and perioral cavity. Two patients relapsed, and 10 patients died of non-relapse causes. This study demonstrates that PSCT with CD34+ selection and a defined dose of CD3+ results in prompt engraftment and may limit development of extensive chronic GVHD.

Prospective trial of high-dose chemotherapy followed by infusions of peripheral blood stem cells and dose-escalated donor lymphocytes for relapsed leukemia after allogeneic stem cell transplantation.

To determine whether induction of graft-versus-host disease (GVHD) improves the outcome of acute relapsed leukemia after stem cell transplantation (SCT), we used high-dose cytarabine (ara-C) followed by infusions of donor-derived buffy coats containing peripheral blood stem cells to treat 12 patients with relapsed leukemia. Donor lymphocyte infusion (DLI) was repeated at least twice over a 5-week interval for patients in whom grade II to IV acute GVHD did not develop after the first DLI. Grade II to IV acute GVHD developed in 4 (33%) of the patients. Chronic GVHD developed in 3 patients, 2 of whom had not experienced acute GVHD. Four (67%) of the 6 patients who developed grade II to IV acute and/or chronic GVHD after DLI responded, but none of the other 6 patients responded. Four (33%) of the patients (2 with acute myelogenous leukemia [AML] and 2 with acute lymphoblastic leukemia [ALL]) achieved complete remission lasting longer than 4 months after the first DLI, but 3 of them had relapses in bone sites. Of these 4 patients, 1 patient with AML and 2 with ALL were alive 8 to 27 months after DLI. These findings indicate that high-dose ara-C combined with megadose DLI may produce durable remission of acute leukemia that has relapsed after SCT when GVHD is induced. The low induction rate of GVHD and extramedullary relapse after remission is achieved with DLI are problems yet to be solved.

Growth factors and DLI in adult haploidentical transplant: a three-step pilot study towards patient and disease status adjusted management.

Haploidentical transplant is now established as a procedure of choice for patients who lack a compatible donor. However, they are still referred too late, heavily pretreated, at very advanced stages. We initiated a three-step phase I study trying improve transplant-related mortality, relapse rate, and immunity: G-CSF + DLI, GM-CSF + DLI, patient- and disease-adapted strategy. Thirty-three consecutive leukemia patients, aged 18-55, were investigated (20 very poor risk, 11 poor risk, and 2 better risk). GvH type NK alloreactivity was chosen when possible (18/33) and balanced across the three groups. In the first nine patients, G-CSF was used and escalated prophylactic DLI started at month 1. Thus, G-CSF and 1-3 DLI (10(4) CD3/kg) is safe. It results in faster CD4 recovery and a low rate of infections. However, it was insufficient to induce a GVL effect. In the next 12 patients, GM-CSF was used plus 1 DLI (10(4) CD3/kg) at day 30 unless aGVHD (3 patients). The comparison between the two first groups can be summarized as follows: G-CSF + DLI: TRM at day 100: 0, RR: 6/9, severe aGVHD: 0. GM-CSF + 1 DLI group: RR: 1/12, TRM at day 100: 3, aGVHD > 1: 9/12, price to pay: GVHD resulting in five deaths in total. Step 3 (13 patients) consists of a patient-adapted strategy: no more aspecific DLI (selected anti-CMV and aspergillus DLI planned in all patients); in myeloid disorders with NK alloreactivity: no GF. In the other cases, GM-CSF (at a reduced total dose of 500 mug) is given the follow-up of these 13 patients, although promising is currently short (median 5 months). Overall, TRM at day 100 is 3/29, reflecting the good tolerance of the conditioning in a heavily pretreated population (median age: 43). NRR mortality (8/26) at 1 year is greater in the GM-CSF + DLI group, reflecting the impact of severe aGVHD. We conclude that the third strategy might improve the outcome without exposing patients to unnecessary severe GVHD.

Donor chimerism does not predict response to donor lymphocyte infusion for relapsed chronic myelogenous leukemia after allogeneic hematopoietic cell transplantation.

We studied the effect of donor chimerism level on the outcome of donor lymphocyte infusion (DLI) therapy in 42 patients with persistent or relapsed hematologic malignancies after non-T cell-depleted allogeneic hematopoietic cell transplantation. Seventy-five percent of chronic myelogenous leukemia (CML) and 39% of non-CML patients entered remission after DLI therapy. Remission and survival rates were similar for CML patients irrespective of their pre-DLI donor chimerism levels; however, remission occurred sooner in patients with > or =10% pre-DLI donor chimerism. None of the non-CML patients with <10% pre-DLI donor chimerism and 47% of those with > or =10% pre-DLI donor chimerism attained remission. The 2-year survival rates after DLI were 75% for CML and 17% for non-CML patients. We conclude that a low level of donor marrow chimerism is not an adverse prognostic factor for response to DLI in CML patients, but for non-CML patients it may confer worse outcomes. Better methods to augment the response to DLI for patients with hematologic malignancies other than CML that recur after allogeneic hematopoietic cell transplantation are needed, whereas for relapsed CML patients, combination therapies including imatinib mesylate or other promising antileukemic agents may provide outcomes superior to those with DLI alone.

CC chemokine receptor 2 expression in donor cells serves an essential role in graft-versus-host-disease.

The complete repertoire of cellular and molecular determinants that influence graft-vs-host disease (GVHD) is not known. Using a well-established murine model of GVHD (B6-->bm12 mice), we sought to elucidate the role of the donor non-T cell compartment and molecular determinants therein in the pathogenesis of GVHD. In this model the acute GVHD-inducing effects of purified B6 wild-type (wt) CD4(+) T cells was inhibited by wt non-T cells in a dose-dependent manner. Paradoxically, unlike the chronic GVHD phenotype observed in bm12 mice transplanted with B6wt unfractionated splenocytes, bm12 recipients of B6ccr2-null unfractionated splenocytes developed acute GVHD and died of IFN-gamma-mediated bone marrow aplasia. This switch from chronic to acute GVHD was associated with increased target organ infiltration of activated CD4(+) T cells as well as enhanced expression of Th1/Th2 cytokines, chemokines, and the antiapoptotic factor bfl1. In vitro, ccr2(-/-) CD4(+) T cells in unfractionated splenocytes underwent significantly less activation-induced cell death than B6wt CD4(+) T cells, providing another potential mechanistic basis along with enhanced expression of bfl1 for the increased numbers of activated T cells in target organs of B6ccr2(-/-) splenocyte-->bm12 mice. Collectively, these findings have important clinical implications, as they implicate the donor non-T cell compartment as a critical regulator of GVHD and suggest that ccr2 expression in this cellular compartment may be an important molecular determinant of activation-induced cell death and GVHD pathogenesis.

Donor lymphocyte infusion for relapsed chronic myelogenous leukemia: prognostic relevance of the initial cell dose.

Donor lymphocyte infusion (DLI) can produce durable remissions in patients with chronic myelogenous leukemia (CML) who have a relapse after an allogeneic stem cell transplantation. However, the best modality to administer DLI is still unclear. The effect of the initial cell dose (ICD; ie, mononuclear cells x 10(8)/kg received in the first instance) on outcome was retrospectively analyzed in 298 of 344 patients treated with DLI at 51 centers. Patients were classified into 3 groups according to the ICD: 98 in group A ( 2.0). Additional infusions were given to 62%, 20%, and 5% of patients in groups A, B, and C, respectively. A lower ICD was associated with less graft-versus-host disease (GVHD; A, 26%; B, 53%; C, 62%; P <.001), less myelosuppression (A, 10%; B, 23%; C, 24%; P =.01), and similar response rate (A, 78%; B, 73%; C, 70%; P =.48). Nonadjusted estimates of 3-year survival, failure-free survival, and DLI-related mortality were 84%, 66%, and 5% respectively, in group A; 63%, 57%, and 20% in group B; and 58%, 45%, and 22% in group C. Outcome analysis was adjusted for patient age, donor type, sex of donor, sex mismatch, disease phase at transplantation, T-cell depletion, interval from transplantation to DLI, GVHD prior to relapse, relapse type, and date of DLI. After adjustment, lower ICD was associated with less GVHD, less myelosuppression, same response rate, better survival, better failure-free survival, and less DLI-related mortality. Our results suggest that the first DLI dose should not exceed 0.2 x 10(8) mononuclear cells/kg.

Prophylactic donor lymphocyte infusions after moderately ablative chemotherapy and stem cell transplantation for hematological malignancies: high remission rate among poor prognosis patients at the expense of graft-versus-host disease.

We investigated the use of 'prophylactic' donor lymphocyte infusions (DLI) containing 1 x 107 CD3+ cells, given at 30, 60 and 90 days post-allogeneic blood and marrow transplantation (BMT), following conditioning with fludarabine 30 mg/m(2)/4 days and melphalan 70 mg/m(2)/2 days. GVHD prophylaxis consisted of cyclosporin A (CsA) 2 mg/kg daily with early tapering by day 60. Our goals were the rapid achievement of chimerism and disease control, providing an immunological platform for DLIs to treat refractory patients with hematological malignancies. Twelve heavily pre-treated patients with life expectancy less than 6 months were studied; none were in remission. Diagnoses were AML (n = 4), MDS (n = 1), ALL (n = 3), CML (n = 3) and multiple myeloma (n = 1). Response rate was 75%. Three patients are alive at a median of 450 days (range, 450-540). Two patients are in remission of CML in blast crisis and AML for more than 14 months. Median survival is 116 days (range, 25-648). Six patients received 12 DLIs; three patients developed acute GVHD after the first infusion and were excluded from further DLIs, but no GVHD occurred among patients receiving subsequent DLIs. One patient with CML in blast crisis went into CR after the first DLI. The overall incidence of acute GVHD was 70%. Primary causes of death were infections (n = 3), acute GVHD (n = 3), chronic GVHD (n = 1) and disease relapse (n = 2). We observed high response and chimerism rates at the expense of an excessive incidence of GVHD. DLI given at day +30 post BMT caused GVHD in 50% of the patients, and its role in this setting remains unclear.

[Allogeneic stem cell transplantation].

Over the last two decades, four major therapeutic approaches have dramatically changed the prognosis in chronic myelogenous leukemia(CML). Those include allogeneic stem cell transplantation, interferon-alpha based regimen, donor-leukocyte infusions, and the revolutionary BCR/ABL tyrosine kinase inhibitor such as STI571. Each modality has exploited and targeted different aspects of CML biology, and is associated with different risk-benefit ratios. In this section, we update the results of both related and unrelated donor transplantation, donor lymphocyte infusions, and non-myeloablative stem cell transplantation in CML in comparison with the other treatment modalities.

Long-Term follow-up of recipients of CD8-depleted donor lymphocyte infusions for the treatment of chronic myelogenous leukemia relapsing after allogeneic progenitor cell transplantation.

Donor lymphocyte infusions (DLIs) are an effective treatment for relapsed Chronic myeloid leukemia (CML) after allogeneic transplantation but are limited by the occurrence of GVHD. CD8+ T lymphocytes are involved in the pathogenesis of GVHD but may not be essential for the graft-versus-leukemia (GVL) effect in CML. We have treated 26 CML patients with posttransplantation relapse with CD8-depleted DLI. Thirteen of 15 patients (87%) who relapsed in early-phase CML achieved complete cytogenetic response, but only 1 of 11 who relapsed in advanced-phase disease achieved complete response. Acute GVHD occurred in 2 patients (8%), and extensive chronic GVHD occurred in 2 patients (11%). Treatment-related mortality was 11.5%. Responses were durable; with a median follow-up of 4.2 years (1-7.5 years), only 1 responding patient relapsed (7%). CD8-depleted DLI was equally effective and safe after unrelated donor transplants and sibling transplants. Cytogenetic clonal evolution at the time of DLI was not predictive of treatment failure unless associated with hematologic criteria for disease acceleration. CD8 depletion is an effective method to separate GVL from GVHD for posttransplantation relapsed CML. This strategy is associated with durable complete remissions and a low rate of complications and therefore merits further investigation in larger-scale comparative trials.