Donor lymphocyte infusion Prophylactic T cell infusion after T cell-depleted bone marrow transplantation in patients with refractory lymphoma

Transcription

1 (2002) 29, Nature Publishing Group All rights reserved /02 $ Donor lymphocyte infusion Prophylactic T cell infusion after T cell-depleted bone marrow transplantation in patients with refractory lymphoma C-K Lee, M de Magalhaes-Silverman, RJ Hohl, M Hayashi, J Buatti 2,BCWen 5, A Schlueter 3, RG Strauss 4 and RD Gingrich Department of Internal Medicine, Division of Hematology, Oncology, Blood and Marrow Transplantation, University of Iowa, College of Medicine, Iowa City, IA, USA; 2 Division of Radiation Oncology, Department of Radiology, University of Iowa, College of Medicine, Iowa City, IA, USA; 3 Department of Pathology, University of Iowa, College of Medicine, Iowa City, IA, USA; 4 Department of Pathology and Pediatrics, University of Iowa, College of Medicine, Iowa City, IA, USA; and 5 Department of Radiation Oncology, University of Miami, Miami, FL, USA Summary: Fifty-two patients with refractory lymphoma were prospectively treated with prophylactic T lymphocyte infusion after T cell-depleted allogeneic bone marrow transplantation, to induce graft-versus-lymphoma effect. Thirty-three patients had related donors; 9 had unrelated donors. After transplantation with marrow that had CD3 + cells/kg, T cells up to CD3 + cells/kg were given over 3 months provided grade II acute graft-versus-host disease (GVHD) was not seen. The cumulative incidence of grades II IV acute GVHD was 69%. Twenty of 32 evaluable patients (63%) developed chronic GVHD. Ten patients (9%) died of GVHD. The Kaplan Meier 5- year overall survival of all patients was 34%. On multivariate analyses, chronic GVHD was significant for relapse (hazard ratio of.7, P 0.05), and for overall survival (hazard ratio.4, P 0.00). Chemosensitivity was significant for relapse only on univariate analysis. Patients who developed chronic GVHD had 4 years median survival, compared with 9 months in patients without chronic GVHD, P The study shows that patients with chronic GVHD have superior survivals, most probably related to a graft-versuslymphoma effect, which could be modulated by prophylactic T cell infusion. (2002) 29, DOI: 0.038/sj/bmt/ Keywords: graft-versus-lymphoma effect; DLI; refractory lymphoma High-dose chemotherapy with autologous stem cell transplantation (ASCT) can salvage patients with aggressive non-hodgkin s lymphoma (NHL) in relapse, or poor-risk Hodgkin s disease. 3 Sensitivity to standard-dose salvage chemotherapy has been the most important variable for survival post transplant. 4 7 However, since the effect of the chemotherapy dose depends on the intrinsic sensitivity of the tumor, with an insensitive tumor unlikely to respond to a higher dose, the increase in the dose intensity used for ASCT has not been successful for the majority of patients resistant to secondary or salvage therapy. Allogeneic stem cell transplantation (allo-sct) has been used for the treatment of patients with lymphoma, resulting in equivalent survival rates, especially when the stem cell compartment was involved by lymphoma or when the risk of relapse after ASCT was high. 8 3 A graft-versus-lymphoma effect was suggested by the lower relapse and progression rates that made up for the high mortality due to transplant complications of allo-sct. 8 0,4 In 994, we initiated a prospective study in an attempt to improve on the graft-versus-tumor effect in T cell-depleted marrow transplantation through delayed sequential infusions of T cells, 5 based on reports that delayed infusion enhanced the graft-versus-leukemia effect without GVHD. 6,7 Although our initial results revealed relapsefree survival of patients with leukemia to be most dependent on chronic GVHD and grades II IV acute GVHD, the significance of GVHD in disease progression in lymphoma was not seen, largely because of the insufficient number of patients. We now describe our 6-year study for patients with lymphoma who underwent prophylactic T cell infusions after T cell-depleted allogeneic bone marrow transplantation. We focus on the relative significance of both chemosensitivity and GVHD, since chemosensitivity has been recognized as one of the important prognostic factors after allo-sct. 9,4,8 The current study shows the patients with lymphoma who developed chronic GVHD had better survival rates and a reduced probability of relapse, which could be modulated by T cell infusion after T celldepleted bone marrow transplantation. Correspondence: Dr C-K Lee, Myeloma and Transplant Research Center, University of Arkansas for Medical Sciences, Slot 776, 430 West Markham, Little Rock, Arkansas 72205, USA Received 30 August 200; accepted 9 January 2002

2 66 Patients and methods Table GVHD risk groups Patients Patients were accrued between July 994 and December Patients who had NHL or HD were eligible if they had primary refractory disease; relapsed disease with evidence of disease in the stem cell compartment that was not cleared by salvage therapy; less than partial remission of relapsed disease by the initial salvage therapy; mantle cell lymphoma at relapse; precursor B-lymphoblastic leukemia/lymphoma at relapse. The institutional review board of the College of Medicine, the University of Iowa, approved the study and all patients or their guardians gave written informed consent at the time of study entry. Eligibility, evaluated within 2 weeks of transplantation, included age between 7 and 60 years; diffusion capacity 55%, FEV.5 l, or PaO 2 60 mmhg; left ventricular ejection fraction 45%; serum creatinine 2.5 times normal; bilirubin or transaminases 3 times normal not related to tumor; no active viral hepatitis; negativity for human immunodeficiency virus; no active, uncontrolled invasive fungal infection; no evidence of psychiatric disturbances with objective incompetence. Patients were staged according to the Ann Arbor system, including bone marrow biopsy, whole body CT scans and brain CT scans. Biopsy samples of the original diagnosis and repeat study were reviewed by hematopathologists at the University of Iowa Hospitals and Clinics. Positron emission tomograms or gallium scans were repeated if they had been positive prior to the study. Histocompatibility testing and donor selection Tissue typing was done as previously described. 5 A standard two-stage microcytotoxicity assay was used to test class I HLA antigens. Class II HLA antigens were assessed by either a microcytotoxicity assay (n = 6) or allele typing of DRB with sequence-specific oligonucleotide hybridization (n = 46). When identical at HLA-A, B, and D/DRB loci, donor and patient pairs were considered matched. Patients with an HLA-identical sibling or HLA phenotypically matched, related donor were eligible up to 60 years of age. Patients with an unrelated donor, mismatched donor at one antigen or donor with phenotypically matched HLA antigens in the GVHD direction (homozygous recipient) were eligible up to 55 years of age. Ablative regimen The TBI-BVAC regimen was administered as previously described: 5 total body irradiation of 200 cgy in six fractions with lung shielding to a total dose of 800 cgy on days 8, 7, 6; BCNU 300 mg/m 2 i.v. on day 5; etoposide 600 mg/m 2 /day on days 3, 2; arabinosyl cytosine 2 g/m 2 every 2 h for four doses on days 4 3, 2; cyclophosphamide 3.6 g/m 2 i.v. on day 2. T cell-depleted allogeneic marrow graft was infused 40 h after cyclophosphamide. Risk group Low-risk Intermediate-risk High-risk HLA status, source of marrow and age Matched siblings and 35 years Matched siblings and years Matched unrelated and 50 years Matched siblings and 5 60 years Matched unrelated and 5 55 years Phenotypically matched related and 60 years Any mismatched recipients GVHD prophylaxis and therapy GVHD prophylaxis was done by ex vivo depletion of marrow T cells, and cyclosporine. For the first eight patients, T cells were depleted by an anti-cd2 IgM mouse monoclonal antibody (CT-2) to the E-rosette receptor on T lymphocytes and new-born rabbit serum as a source of complement. 9 For the remaining 44 patients, T cells were depleted by an anti-cd3 IgG 2 monoclonal antibody (Muromonab-CD3; Ortho Biotec, Raritan, NJ, USA) and new-born rabbit serum. 5 Mononuclear cells were recovered at a mean of /kg. The marrow CD3 + cells were reduced by a median of 2.4 logs to /kg. Other T cell depletion data were published previously. 5 A comparative analysis of T cell depletion did not reveal differences between these two methods. 20 Prophylaxis with intravenous cyclosporine at 3 mg/kg/day was started on day and later changed to the oral route if tolerated, to maintain trough levels of whole blood between 50 and 350 ng/ml. Patients who did not develop GVHD or whose GVHD became well controlled started to taper the cyclosporine at 4 weeks from the last T cell infusion, at a rate of mg/kg every 2 weeks. Corticosteroids were added for initial therapy of acute GVHD of grade II and chronic GVHD. Prophylactic T cell infusions Patients were stratified into low-, intermediate- and highrisk groups for GVHD based on the HLA status, source of marrow, and age (Table ). T cells for infusion were prepared as previously described. 5 Blood T cells to be infused were flow cytometrically quantified for CD3 + mononuclear cells in the donor product. As described in Table 2, patients Table 2 Scheme of T cell infusion Infusion Low-risk Intermediate-risk High-risk First Day 0 a Day 0 a ANC 500 3d a /kg b /kg b 0 5 /kg b Second 4th 6th wk a 4th 6th wk a 6th 8th wk a /kg b /kg b /kg b Third 0th 2th wk a 0th 2th wk a 2th 4th wk a 0 6 /kg b /kg b /kg b ANC = absolute neutrophil count/ l. a Time of infusion in relation to transplantation. b Number of T cells.

3 were eligible for incremental infusions up to three times, provided they had not had grade II of acute GVHD and had no serious transplant complications, such as disseminated infection or organ failure. Supportive care Supportive care was administered as previously described. 5 Antibacterial, antifungal, and antiviral prophylaxes were carried out according to the institutional protocols, including broad-spectrum antibiotics and amphotericin-b during neutropenia, and bactrim and acyclovir post transplant. Ganciclovir was used for cytomegalovirus (CMV) prophylaxis during the ablative phase and after engraftment, for 6 months. CMV reactivations were monitored weekly for at least 6 months by CMV DNA hybrid capture on buffy coat samples. Granulocyte colony-stimulating factor was administered at 300 g beginning on day +7 and continued until engraftment. Response assessment Response was assessed on the basis of radiologic, and pathologic criteria. 2 All patients underwent bone marrow biopsies and CT scans at regular intervals during the first year post transplant. Thereafter, studies were performed if clinically indicated by suggestive abnormalities or symptoms. Diagnosis of progressive disease and relapse required confirmation by tissue biopsy (n = 5), or autopsy (n = 2). Statistical analysis Patients were studied prospectively for response, GVHD, relapse and death. Acute and chronic GVHD were graded according to the published criteria. 22,23 Variables for analyses were chemosensitivity, infusion of T cells, maximum grades of both acute and chronic GVHD, disease progression (progressive disease + relapse), death, overall survival and event-free survival. Events were defined as relapse or death from any cause. Treatment-related mortality (TRM) included any death, not associated with GVHD or disease, within 00 days post transplant. Survival analysis was done by the product-limit estimate of Kaplan Meier. Day 00 post transplant was used as the landmark point for analysis related to chronic GVHD. Comparisons of the probability of relapse and survival were evaluated by the log rank test. For multivariate analyses, variables with a P of 0. on univariate analysis were put into a stepwise multivariate regression by the Cox proportional hazard model. Results Patient characteristics (Table 3) Fifty-two patients were accrued; 35 males and 7 females. Median age was 44 years (range 22 to 58 years). Thirtyone patients had siblings as donors and two had HLA phenotypically matched related donors. Nineteen patients received unrelated grafts. Six underwent mismatched trans- Table 3 Disease characteristics Disease Total Pr/Sr/Scc Diffuse large B cell lymphoma 5 5/7/3 Follicular center lymphoma (grade ) 7 /3/3 Follicular center lymphoma (grade 2) 8 2/4/2 Follicular center lymphoma (grade 3) 4 /2/ Follicular center lymphoma, diffuse 3 /2/0 (predominantly small cell) Primary mediastinal large B cell 2 0/2/0 lymphoma Mantle cell lymphoma 4 0/3/ Precursor B lymphoblastic 4 0/3/ leukemia/lymphoma High-grade B cell lymphoma, Burkitt- 2 //0 like Nodular sclerosis Hodgkin s disease 3 2//0 Pr/Sr/Scc = Primary refractory/chemosensitive relapse/contaminated stem cell compartment. plantation. Ten patients were at low-risk for GVHD; 22 intermediate-risk; 20 high-risk. Median Karnofsky score was 60% (range 30% to 80%). Twenty-eight patients showed partial remission (six CR; 22 PR) by salvage chemotherapy before the transplant. T cell infusion and GVHD Thirty-eight patients received at least one T cell infusion. Of these, 7 patients received two and five patients had three infusions. Fourteen patients received no T cells: developed de novo grade II acute GVHD and three died of TRM, before the first infusion. Of note, all patients with de novo acute GVHD were high-risk patients. Acute GVHD of grade II occurred in 4 of 38 after the first, nine of 7 patients after the second and two of five patients after the third T cell infusion. The overall incidence of grades II IV acute GVHD was 69 3%, with grade II the most common (35 3%). Nine patients (7 0%) died of acute GVHD or related infectious complications. At day 00, 32 patients were alive. Twenty of these patients developed chronic GVHD, with 0 limited and 0 extensive chronic GVHD, respectively. One patient died of an infectious complication of the extensive chronic GVHD, resulting in an overall mortality from GVHD of 9 %. Causes of death Thirty-four patients died, including 7 patients of TRM, 0 of GVHD-related causes and seven of recurrent disease. All relapsed patients died. The most common cause of TRM was idiopathic pneumonia syndrome (n = 9), followed by infection (n = 6) and post-transplant hemolytic uremic syndrome (n = 2). Six of nine with idiopathic pneumonia syndrome had received prior mediastinal irradiation. Four related and six unrelated recipients died of GVHD. There were no cases of graft failure. Response, relapse and survival Of the 32 patients who survived the first 00 days, 2 achieved complete remission, eight partial remission, and three 67

4 68 stable disease, with an overall response rate of 60 3%. Seven patients had relapse/disease progression, resulting in a 5-year cumulative risk of 23 8% (Figure a). The probability of both 5-year overall and event-free survival was identical at 34 7%, since all relapsed patients died (Figure b). On univariate analysis for relapse, both chemosensitivity and chronic GVHD were significant, with each P The 5-year probability of relapse in patients with chemosensitive disease was 8 7%, compared with 38 3% in chemoresistant patients. Chronic GVHD was associated with a reduced 5-year probability of relapse: 7 4% of patients with vs 39 4% of patients without chronic Probability of survival Probability of survival Cumulative risk of relapse a Years post transplant b Years post transplant c No. of patients alive yr: 20 2 yr: 9 3 yr: Years post transplant Figure (a) Cumulative risk of relapse; (b) overall survival; (c) eventfree survival. Solid line, chronic GVHD (+); dotted line, chronic GVHD ( ). GVHD. Acute GVHD and T cell infusion were not significant for relapse. On multivariate analysis, chronic GVHD was significant for relapse, with a hazard ratio of.7 and P For overall survival on univariate analysis, T cell infusion (P = 0.052), acute GVHD (P 0.05) and chronic GVHD (P 0.00) were significant. These variables were also significant for event-free survival, with P 0.05, 0.05 and 0.00, respectively. Patients who received at least one T cell infusion had a 5-year overall survival of 45 %, compared with 24 9% of patients who did not receive T cells. Likewise, the 5-year overall survival of patients with acute GVHD was 39 8%, compared with 7 % of patients without acute GVHD. The 5-year overall survival of patients with chronic GVHD was 48 6%, whereas it was 7% in patients without chronic GVHD. On multivariate analysis for both overall survival and event-free survival, chronic GVHD was most significant with a hazard ratio of.4 and P 0.00 (Figure c). At present, six patients with diffuse large B cell lymphomas are alive; seven with follicular center lymphoma; two with mantle cell lymphoma; two with precursor lymphoblastic leukemia/lymphoma; one nodular sclerosing Hodgkin s disease. Discussion This study shows that chronic GVHD is associated with better survival rates and a lower risk of relapse. It suggests an additional anti-lymphoma activity independent of ablative therapy, most probably a graft-versus-lymphoma effect, to have been present in the patients who developed chronic GVHD. Pre-transplant chemosensitivity was marginally important for relapse only, indicating limited tumor control by the ablative regimen. The findings require further study with a larger cohort since the population consisted of heterogeneous patients with different disease characteristics, and there was an asymmetric loss of patients due to a high rate of early mortality. The GVHD mortality rate was high because of the inclusion of unrelated and mismatched donor transplants. The high rate of fatal idiopathic pneumonia syndrome must have been associated with prior radiation to the chest area. T cell infusions may also have played a role, in a way similar to the murine model of the syndrome induced by TBI and donor spleen cells. 24 The role of T cell-depleted allogeneic transplantation has not been well defined for patients with lymphoma, except for the reduced rates of severe acute GVHD. In a study using T0B9 murine monoclonal antibody for T cell depletion, Juckett et al 25 reported on 2 patients with aggressive NHL and 6 patients with indolent NHL. With a median follow-up of 4.4 years, 33% of the patients with aggressive NHL and 62% of those with indolent NHL were alive and progression-free. Eight patients died of GVHD or related complications. In a study of 22 patients with NHL (0 high-grade; nine intermediate-grade; three low-grade), CD6 + T cell depletion was used without additional immunosuppression for transplantation from HLA-identical siblings. 26 The incidence of acute GVHD of grades II IV was 23%, with no deaths from GVHD. With a median fol-

5 low-up of 30 months, 54% of patients were alive without disease. The major cause of failure was eight cases of relapse. Counter-flow centrifugation was used for T cell depletion in another study of the 5 patients with poorrisk NHL, using HLA matched siblings as donor. 27 After a median follow-up of 36 months, 0 patients were alive and in complete remission. None of them died of acute GVHD and one died of chronic GVHD. A graft-versus-lymphoma effect has not been unequivocally substantiated despite several studies. In a study of patients with relapsed HD and aggressive NHL, evidence of a graft-versus-lymphoma effect was suggested by a lower probability of relapse in chemosensitive patients who underwent allo-sct, compared with similar patients who had ASCT: 8% vs 46%, P = A prospective comparative study of 35 patients with ASCT vs 3 patients with allo-sct showed a higher probability of disease progression in the autologous group than in the allogeneic group: 69% vs 20%, P = In a comparative, casecontrolled study of the European Bone Marrow Transplant Group of 202 patients with NHL, a lower relapse rate was associated with chronic GVHD, compared to that seen in patients with no chronic GVHD: 0% vs 35%, P = Another case-controlled study of 94 patients with HD showed no advantage of allo-sct over ASCT due to its high mortality. 0 A 0-year retrospective study of 73 patients with aggressive NHL who were treated with allo- SCT showed no differences in relapse rate according to GVHD. 2 A recent study of allo-sct for patients with mantle cell lymphoma suggested the presence of a graftversus-lymphoma effect, based on the presumptive association of remission with GVHD and the time-dependent change in the bcl- polymerase chain reaction assay. 28 The present study suggests there may not be a distinct graft-versus-lymphoma effect separate from GVHD. Although T cell infusion was seen as important for survival on univariate analysis, this significance was lost on multivariate analysis that selected chronic GVHD instead. It may be that a quantifiable graft-versus-lymphoma effect is not seen with the current scheme of T cell infusion. Furthermore, alloreactivity based on disparity of minor HLA antigens may be more important than number of T cells. The findings are similar to the previous reports of delayed donor lymphocyte infusion where GVHD was necessary for a survival benefit in leukemic patients. 8,29 The graft-versuslymphoma effect may be modulated by T cell infusion by means of its capacity to induce GVHD. Further studies with T cells designed to induce moderate and tolerable GVHD, including non-myeloablative transplantation and delayed lymphocyte infusion, should be pursued for patients with refractory lymphoma who are not eligible for autologous transplantation. References Philip T, Guglielmi C, Hagenbeek A et al. Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-hodgkin s lymphoma. New Engl J Med 995; 333: Horning SJ, Chao NJ, Negrin RS et al. High-dose therapy and autologous hematopoietic progenitor cell transplantation for recurrent or refractory Hodgkin s disease: analysis of the Stanford University results and prognostic indices. Blood 997; 89: Nademanee A, O Donnell MR, Snyder DS et al. High-dose chemotherapy with or without total body irradiation followed by autologous bone marrow and/or peripheral blood stem cell transplantation for patients with relapsed and refractory Hodgkin s disease: results in 85 patients with analysis of prognostic factors. Blood 995; 85: Burns LJ, Daniels KA, McGlave PB et al. Autologous stem cell transplantation for refractory and relapsed Hodgkin s disease: factors predictive of prolonged survival. Bone Marrow Transplant 995; 6: Crump M, Smith AM, Brandwein J et al. High-dose etoposide and melphalan, and autologous bone marrow transplantation for patients with advanced Hodgkin s disease: importance of disease status at transplant. J Clin Oncol 993; : Philip T, Armitage JO, Spitzer G et al. High-dose therapy and autologous bone marrow transplantation after failure of conventional chemotherapy in adults with intermediate-grade or high-grade non-hodgkin s lymphoma. New Engl J Med 987; 36: Appelbaum FR, Sullivan KM, Buckner CD et al. Treatment of malignant lymphoma in 00 patients with chemotherapy, total body irradiation, and marrow transplantation. J Clin Oncol 987; 5: Chopra R, Goldstone AH, Pearce R et al. Autologous versus allogeneic bone marrow transplantation for non-hodgkin s lymphoma: a case-controlled analysis of the European Bone Marrow Transplant Group Registry data. J Clin Oncol 992; 0: Ratanatharathorn V, Uberti J, Karanes C et al. Prospective comparative trial of autologous versus allogeneic bone marrow transplantation in patients with non-hodgkin s lymphoma. Blood 994; 84: Milpied N, Fielding AK, Pearce RM et al. Allogeneic bone marrow transplant is not better than autologous transplant for patients with relapsed Hodgkin s disease. European Group for Blood and. J Clin Oncol 996; 4: van Besien K, Thall P, Korbling M et al. Allogeneic transplantation for recurrent or refractory non-hodgkin s lymphoma with poor prognostic features after conditioning with thiotepa, busulfan, and cyclophosphamide: experience in 44 consecutive patients. Biol Blood Marrow Transplant 997; 3: Dhedin N, Giraudier S, Gaulard P et al. Allogeneic bone marrow transplantation in aggressive non-hodgkin s lymphoma (excluding Burkitt and lymphoblastic lymphoma): a series of 73 patients from the SFGM database. Societ Francaise de Greffe de Moelle. Br J Haematol 999; 07: van Besien K, Sobocinski KA, Rowlings PA et al. Allogeneic bone marrow transplantation for low-grade lymphoma. Blood 998; 92: Jones RJ, Ambinder RF, Piantadosi S et al. Evidence of graftversus lymphoma effect associated with allogeneic bone marrow transplantation. Blood 99; 77: Lee CK, Gingrich RD, demagalhaes-silverman M et al. Prophylactic infusion of T cells for T cell-depleted allogeneic bone marrow transplantation. Biol Blood Marrow Transplant 999; 5: Johnson BD, Drobyski WR, Truitt RL. Delayed infusion of normal donor cells after MHC-matched bone marrow transplantation provides an antileukemia reaction without graft versus host disease. Bone Marrow Transplant 993; : Johnson BD, Truitt RL. Delayed infusion of immunocompet- 69

6 620 ent donor cells after bone marrow transplantation breaks graft versus host tolerance and allows for persistent anti-leukemic reactivity without severe graft versus host disease. Blood 995; 85: Juckett M, Rowlings P, Hessner M et al. T cell-depleted allogeneic bone marrow transplantation for high-risk non- Hodgkin s lymphoma: clinical and molecular follow-up. Bone Marrow Transplant 998; 2: Trigg ME, Petersen A, Erickson C et al. Depletion of T cells from bone marrow for allogeneic transplantation: method for treatment of bone marrow in bulk. Exp Hematol 986; 4: Lee CK, Gingrich RD, Harman GS et al. Anti-CD3 (OKT3) T cell depletion and add-back: comparative analysis with anti- CD2 (CT2) T cell depletion and add-back in allogeneic bone marrow transplantation. Blood 995; 86: 570 (Abstr. 2266). 2 Cheson BD, Horning SJ, Coiffier B et al. Report of an international workshop to standardize response criteria for non- Hodgkin s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 999; 7: Deeg HJ, Cottler-Fox M. Clinical spectrum and pathophysiology of acute graft-vs.-host disease. In: SJ Burakoff, HJ Deeg, J Ferrara, K Atkinson (eds). Graft-vs.-Host Disease. Marcel Dekker, Inc: New York, NY, 990, pp Glucksberg H, Storb R, Fefer A et al. Clinical manifestations of graft versus host disease in human recipients of marrow from HLA matched sibling donors. Transplantation 974; 8: Shankar G, Bryson S, Jennings CD et al. Idiopathic pneumonia syndrome after allogeneic bone marrow transplantation in mice. Role of pretransplant radiation conditioning. Am J Respir Cell Mol Biol 999; 20: Juckett M, Rowlings P, Hessner M et al. T cell-depleted allogeneic bone marrow transplantation for high-risk non- Hodgkin s lymphoma: clinical and molecular follow-up. Bone Marrow Transplant 998; 2: Soiffer RJ, Freedman AS, Neuberg D et al. CD6+ T celldepleted allogeneic bone marrow transplantation for non- Hodgkin s lymphoma. Bone Marrow Transplant 998; 2: Mandigers CM, Raemaekers JM, Schattenberg AV et al. Allogeneic bone marrow transplantation with T-cell-depleted marrow grafts for patients with poor-risk relapsed low-grade non-hodgkin s lymphoma. Br J Haematol 998; 00: Khouri IF, Lee MS, Romaguera J et al. Allogeneic hematopoietic transplantation for mantle-cell lymphoma: molecular remissions and evidence of graft-versus-malignancy. Ann Oncol 999; 0: Naparstek E, Or R, Nagler A et al. T-cell-depleted allogeneic bone marrow transplantation for acute leukaemia using Campath- antibodies and post-transplant administration of donor s peripheral blood lymphocytes for prevention of relapse. Br J Haematol 995; 89: