Transplantation of unrelated placental blood cells in children with high-risk sickle cell disease

The lack of healthy HLA-identical sibs limits the use of allogeneic hematopoietic cell transplantation in children with high-risk sickle cell disease (SCD). We evaluated unrelated placental blood cell transplantation (UPBCT) after a preparative regimen of busulfan, cyclophosphamide and antithymocyte globulin in three children with SCD who had cerebrovascular accidents (CVAs) and did not have HLA-matched sib donors. The placental blood cell units were matched with the recipients at four of six HLA-A, HLA-B and HLA-DRB1 antigens. Neutrophil levels above 0.5 x 10(9)/l occurred at 23, 38 and 42 days after UPBCT, and platelet levels above 50 x 10(9)/l without transfusions occurred at 62, 81 and 121 days after UPBCT. All patients developed acute graft-versus-host disease (GVHD; two grade II, one grade III), and one developed extensive chronic GVHD. One patient had graft failure and autologous hematopoietic recovery. Two patients have complete donor hematopoietic chimerism without detectable hemoglobin S or symptoms of SCD at 40 and 61 months, respectively, after UPBCT. These observations demonstrate the feasibility of UPBCT in children with SCD. Further studies of UPBCT for SCD are needed but, because of risks of procedure-related morbidity and graft rejection, should be restricted to pediatric patients with high-risk manifestations of SCD.

Stable mixed hematopoietic chimerism after bone marrow transplantation for sickle cell anemia

A multicenter investigation of allogeneic bone marrow transplantation for children with sickle cell disease was conducted that included 27 European and North American transplant centers. Fifty-nine patients who ranged in age from 3.3 to 15.9 years (median, 10.1 years) received HLA-identical sibling marrow allografts between September 1991 and April 2000. Fifty-five patients survive, and 50 survive free from sickle cell disease, with a median follow-up of 42.2 months (range, 11.8 to 115 months) after transplantation. Of the 50 patients with successful allografts, 13 developed stable mixed donor-host hematopoietic chimerism. The level of donor chimerism, measured > or =6 months after transplantation in peripheral blood, varied between 90% and 99% in 8 patients. Five additional patients had a lower proportion of donor cells (range, 11% to 74%). Among these 5 patients, hemoglobin levels varied between 11.2 and 14.2 g/dL (median, 11.3 g/dL; mean, 12.0 g/dL). In patients who had donors with a normal hemoglobin genotype (Hb), the Hb S fractions were 0%, 0%, and 7%, corresponding to donor chimerism levels of 67%, 74%, and 11%, respectively. Among patients who had donors with sickle trait, the Hb S fractions were 36% and 37%, corresponding to donor chimerism levels of 25% and 60%, respectively. Thus, allograft recipients with stable mixed chimerism had Rb S levels similar to donor levels, and only 1 patient required a red blood cell transfusion beyond 90 days posttransplantation. None of the patients have experienced painful events or other clinical complications related to sickle cell disease after transplantation. These observations strongly suggest that patients with sickle cell disease who develop persistent mixed hematopoietic chimerism after transplantation experience a significant ameliorative effect.

Allogeneic hematopoietic cell transplantation for inborn metabolic diseases

Clinical experience for more than two decades has shown that allogeneic HCT may benefit some but not all patients with inherited metabolic diseases. The HCT procedure is most effective in presymptomatic patients and those with indolent forms of storage diseases but is ineffective in those with overt neurological symptoms or aggressive neonatal or infantile forms. HCT alone does not correct skeletal dysplasia in MPSs and may not prevent development or progression of the peripheral neuropathy in sphingolipidoses and ALD. Decisions regarding HCT in patients with storage diseases should be made by investigators knowledgeable about these diseases, with judicious use of laboratory and clinical resources necessary to reach the best therapeutic decision for the individual patient.

Viability of cryopreserved BM progenitor cells stored for more than a decade

BACKGROUND

PBPC or BM is increasingly being harvested in remission for possible use in the event of relapse. Although the value of this approach has not been demonstrated, the long-term storage of progenitor cell components has become commonplace in many facilities.

METHODS

We used multi-parameter flow cytometry to determine the viability of 11 long-term cryopreserved BM components (mean = 11.8 years) in liquid phase nitrogen. The components, prepared for autotransplantation but deaccessioned after confirming patient death, were carefully thawed, washed, and assayed immediately. The flow cytometry assay was performed according to the ISHAGE protocol, modified by the addition of 7AAD for analysis of progenitor viability (CD45+ CD34+ 7AAD-) and total leukocyte viability (CD45+ 7AAD-). In addition, total viability was assessed by fluorescence microscopy using acridine orange dye exclusion; granulocyte-monocyte colony-forming units (CFU-GM) were measured after 14 days culture.

RESULTS

Leukocyte viability by flow cytometry and fluorescence microscopy agreed well (r2 = 0.55, slope = 0.83, P < 0.0005 by linear regression). CFU-GM did not correlate with CD34% or any of the viability parameters. Compared with short-term stored (mean = 33 days) PBPC assayed at infusion, long-term stored BM had a comparable percentage of CD34+ cells, comparable CFU-GM activity, increased CD34 viability, but decreased total cell viability, the latter most likely due to an increased proportion of differentiated myeloid cells.

DISCUSSION

The results indicate that BM products can be cryopreserved for more than a decade without apparent loss of progenitor activity, as measured by these laboratory surrogates. This agrees with clinical anecdotes describing successful engraftment with long-term stored BM, and argues that expiration dates cannot be set for cryopreserved hematopoietic stem-cell components stored in liquid phase nitrogen.

Generation of alloreactive anti-leukemic cytotoxic T lymphocytes with attenuated GVHD properties from haploidentical parents in childhood acute lymphoblastic leukemia

We sought to generate alloreactive leukemia-reactive cytotoxic T lymphocytes from haploidentical parents by culture of peripheral blood mononuclear cells in vitro with irradiated leukemic blasts and IL-2 from children with ALL. After 21 days in culture the mean cytotoxicity of haploidentical lymphocytes against ALL blasts was 38% (n = 11). Cultured parental CTL were not leukemia specific but alloreactive as evidenced by equivalent cytotoxicity against stimulating ALL blasts and ConA-stimulated blasts from the other parent. The cultures yielded primarily CD8(+) T cells (59%). Irradiation of CTL limited proliferation by 96% but had no short-term effects on leukemia reactive cytotoxicity, suggesting a means to limit GVHD potential in vivo. One patient was treated for relapse of ALL post-haploidentical transplant with CTL generated from the original donor. A total of nine infusions were given: the first three were irradiated, while the last six were not due to disease progression. The patient experienced clearance of peripheral blasts, and despite concomitant infusion of IL-2 with the last three CTL infusions, did not experience immediate GVHD reactions. We conclude that ALL blasts are sufficiently immunostimulatory to generate in vitro CTL with provision of exogenous IL-2, and that these CTL could exert an anti-leukemia effect in vivo.

Phase 3 study comparing methotrexate and tacrolimus with methotrexate and cyclosporine for prophylaxis of acute graft-versus-host disease after marrow transplantation from unrelated donors

After the transplantation of unmodified marrow from human leukocyte antigen-matched unrelated donors receiving cyclosporine (CSP) and methotrexate (MTX), the incidence of acute graft-versus-host disease (GVHD) is greater than 75%. Tacrolimus is a macrolide compound that, in previous preclinical and clinical studies, was effective in combination with MTX for the prevention of acute GVHD. Between March 1995 and September 1996, 180 patients were randomized in a phase 3, open-label, multicenter study to determine whether tacrolimus combined with a short course of MTX (n = 90), more than CSP and a short course of MTX (n = 90), would reduce the incidence of acute GVHD after marrow transplantation from unrelated donors. There was a significant trend toward decreased severity of acute GVHD across all grades (P =.005). Based on the Kaplan-Meier estimate, the probability of grade II-IV acute GVHD in the tacrolimus group (56%) was significantly lower than in the CSP group (74%; P =.0002). Use of glucocorticoids for the management of GVHD was significantly lower with tacrolimus than with CSP (65% vs 81%, respectively; P =. 019). The number of patients requiring dialysis in the first 100 days was similar (tacrolimus, 9; CSP, 8). Overall and relapse-free survival rates for the tacrolimus and CSP arms at 2 years was 54% versus 50% (P =.46) and 47% versus 42% (P =.58), respectively. The combination of tacrolimus and MTX after unrelated donor marrow transplantation significantly decreased the risk for acute GVHD than did the combination of CSP and MTX, with no significant increase in toxicity, infections, or leukemia relapse.

Bone marrow transplantation for infantile ceramidase deficiency (Farber disease)

Infantile ceramidase deficiency (Farber disease) is an uncommon, progressive lysosomal storage disease characterized by multiple ceramide-containing nodules (lipogranulomata) in the subcutaneous tissue and upper aerodigestive tract, painful periarticular swelling, psychomotor retardation, and varying degrees of ocular, pulmonary or hepatic involvement. Management of Farber disease has been limited to symptomatic supportive care, and few affected infants survive beyond 5 years of age. We performed an allogeneic bone marrow transplant (BMT) from an HLA-identical heterozygous sister in a 9.5-month-old female with minimally symptomatic Farber disease who received a pre-transplant regimen of busulfan and cyclophosphamide. Ceramidase activity in peripheral blood leukocytes increased from 6% before transplant to 44% (donor heterozygote level) by 6 weeks after BMT. By 2 months after transplant, the patient's subcutaneous lipogranulomata, pain on joint motion, and hoarseness had resolved. Despite modest gains in cognitive and language development, hypotonia and delayed motor skills persisted. Gradual loss of circulating donor cells with autologous hematopoietic recovery occurred; VNTR analyses showed 50% donor DNA in peripheral blood cells at 8.5 months after BMT and only 1% at 21 months after transplant. Interestingly, leukocyte ceramidase activity consistently remained in the heterozygous range despite attrition of donor cells in peripheral blood. This novel observation indicates ongoing hydrolase production by non-circulating donor cells, possibly in the mononuclear phagocytic system, and uptake by recipient leukocytes. Although lipogranulomata and hoarseness did not recur, the patient's neurological and neurocognitive status progressively declined. She died 28 months after BMT (age 37.5 months) with pulmonary insufficiency caused by recurrent aspiration pneumonias. Allogeneic BMT improves the peripheral manifestations of infantile ceramidase deficiency, but may not prevent the progressive neurological deterioration, even when carried out in minimally symptomatic patients.

Impact of bone marrow transplantation for symptomatic sickle cell disease: an interim report. Multicenter investigation of bone marrow transplantation for sickle cell disease

Fifty children who had symptomatic sickle cell disease received matched sibling marrow allografts between September 1991 and March 1999, with Kaplan-Meier probabilities of survival and event-free survival of 94% and 84%, respectively. Twenty-six patients (16 male, 10 female) had at least 2 years of follow-up after transplantation and were evaluated for late effects of transplantation and for its impact on sickle cell-related central nervous system (CNS) and pulmonary disease. Patients ranged between 3.3 and 14.0 (median, 9. 4) years of age and had a median follow-up of 57.9 (range 38-95) months after transplantation. Among 22 of 26 patients who had stable donor engraftment, complications related to sickle cell disease resolved, and none experienced further episodes of pain, stroke, or acute chest syndrome. All 10 engrafted patients with a prior history of stroke had stable or improved cerebral magnetic resonance imaging results. Pulmonary function tests were stable in 22 of the 26 patients, worse in two, and not studied in two. Seven of eight patients transplanted for recurrent acute chest syndrome had stable pulmonary function. Linear growth measured by median height standard deviation score improved from -0.7 before transplantation to -0.2 after transplantation. An adverse effect of busulfan conditioning on ovarian function was demonstrated in five of seven evaluable females who are currently at least 13 years of age. None of the four males tested had elevated serum gonadotropin levels. These data confirm that allogenic bone marrow transplantation establishes normal erythropoiesis and is associated with improved growth and stable CNS imaging and pulmonary function in most patients. (Blood. 2000;95:1918-1924)

CD40 ligand upregulates expression of the IL-3 receptor and stimulates proliferation of B-lineage acute lymphoblastic leukemia cells in the presence of IL-3

The proliferative response of B cell precursor acute lymphoblastic leukemia (BCP-ALL) cells to IL-3 is dependent on the expression of functional IL-3 receptors (IL-3R). Here we report that CD40 ligand (CD40L) in the presence of recombinant IL-3 increased proliferation of BCP-ALL cells by upregulating expression of IL-3R. Upregulation of IL-3R in BCP-ALL cells was observed as early as 1 h after treatment with CD40L, and a 50- to 500-fold increase of IL-3R expression after 24 h was detected in all 12 cases studied. Moreover, expression of receptors for IL-7 (IL-7R) and stem cell factor (SCF-R, c-Kit) was also induced by CD40L in the majority of BCP-ALL cases examined; however, levels of induction were low compared to those for IL-3R. To test the functional activity of upregulated receptors for IL-3, SCF and IL-7, we evaluated the proliferation and growth of BCP-ALL cells cultured in serum-free media with CD40L plus these factors. When CD40L was added with either a single cytokine (IL-3, SCF and IL-7) or their combinations, cell proliferation was significantly increased as detected by DNA synthesis assay. Combinations of CD40L plus IL-3 and either SCF or IL-7 were able to support long-term growth of BCP-ALL cells for at least 8 weeks in three of the seven cases studied. Immunophenotyping and gene rearrangement studies indicated that cells in long-term cultures were monoclonal and retained their original phenotypes. The leukemic cells remained primarily dependent on the presence of IL-3 and its receptor for long-term growth, as shown by selective withdrawal of growth factors or antibody blockade of receptors. These results suggest an important role for CD40L in upregulating expression of IL-3R on BCP-ALL cells and enabling these cells to proliferate in long-term cultures in the presence of IL-3 and either SCF or IL-7.

Cytokine upregulation of the antigen presenting function of acute myeloid leukemia cells

Acute myeloid leukemia (AML) cells are malignant counterparts of normal myeloid pathway progenitors. Myeloid progenitors differentiate into professional antigen presenting cells (APC) under the essential influence of GM-CSF along with additional cytokines. Twelve cases of human AML were tested for ability to be differentiated toward a professional APC phenotype in short-term culture with addition of GM-CSF and the following recombinant proteins: TNFalpha, IL-4, CD40 ligand, Flt3 ligand and SCF. Significant upregulation of CD80 (B7-1) and enhancement of alloantigen presentation was seen with the addition of GM-CSF and TNFalpha alone or with additional cytokines. The combination of GM-CSF and TNFalpha, either alone or in combination with an additional cytokine, resulted in enhancing alloantigen presentation by at least two-fold over the media control group in 10/12 patients studied, and resulted in CD80 expression of greater than 15% in 11/12 patients studied. In AML cultures with GM-CSF and TNFalpha, coexpression of CD80 and either CD34 or an aberrant surface marker (CD56) was seen. In one case, sorted CD80, cells retained a characteristic cytogenetic marker and CD34 expression, proving their derivation from an AML precursor. These studies verify other reports of in vitro differentiation of human AML precursors into enhanced APC, suggesting that this phenomenon could be utilized for immunotherapy strategies aimed at enhancing presentation of leukemia antigens to T cells.

High dose chemotherapy without hematopoietic cell support for the treatment of refractory lymphoma

Conventional dose combination chemotherapy for patients with relapsed or refractory lymphoma is rarely curative. High dose chemotherapy followed by hematopoietic progenitor cell transplant (HPCT) has a clearly defined role in patients who have first relapsed after standard CHOP chemotherapy for lymphoma. However, the role of HPCT is less well defined for patients with chemo-resistant, or chemo-refractory disease. Sixteen patients (15 Non-Hodgkin's, 1 Hodgkin's Disease) were entered into a phase II study to determine if a dose intensive induction regimen in heavily pre-treated refractory lymphoma patients could permit further consolidation with HPCT. The primary endpoints were survival, response, toxicity, and resource utilization. The regimen consisted of continuous infusion etoposide 1 or 2 gm/m2/72 hours, idarubicin 12 mg/m2/d for 3 days followed by cytarabine 2 gm/m2/72 hours on days 8, 9, and 10 (VIC). Fifteen patients were evaluable for objective response. The overall response rate was 53% with 7 patients achieving a partial response and 1 patient achieving a complete response. Of the 8 responders, 6 patients subsequently received high dose chemotherapy followed by HPCT (4 autologous, 2 allogeneic). The median survival was 176 days for the non-responders contrasted with 722 days for the responders. The average duration of hospitalization was 38 days. Toxicity was mainfest primarily as mucositis with a median grade of 3 among the first 13 patients, and a median grade of 2 in three subsequent patients who received an etoposide dose of 1 gm/m2/72 hours. All patients had an episode of neutropenic fever and 5 patients developed clinically significant pneumonitis during therapy. The VIC regimen is active in the treatment of chemo-refractory lymphoma with clinically significant differences in survival for patients who respond to therapy. Further modifications to the regimen could include the addition of a topoisomerase I inhibitor for synergy with etoposide, and using VIC as part of a tandem transplant regimen where response to VIC would allow further therapy with a myeloablative induction followed by HPCT.

Incidence and prognostic significance of MDM2 oncoprotein overexpression in relapsed childhood acute lymphoblastic leukemia

MDM2 overexpression by pediatric ALL cells at initial diagnosis has been linked to poor response to therapy. In the present study, we evaluated the incidence of MDM2 overexpression by ALL cells from pediatric patients at first relapse and compared MDM2 protein levels with in vitro response to adriamycin and with duration of initial complete remission (CR1). Since an important role of MDM2 in enhancing cell proliferation and survival appears to be inhibition of p53 activity, we also evaluated the status of p53 in these patients' leukemic cells. MDM2 protein levels were determined by Western blot analysis of leukemic bone marrow cells obtained from 42 patients with B cell precursor (BCP) ALL who relapsed during or following therapy on standard POG ALL protocols. Twelve of 42 (29%) cases have MDM2 levels >/=10-fold higher than those detected in normal bone marrow mononuclear (NMMC) cells, which express relatively low levels of protein. Thirty cases (71%) expressed MDM2 at levels <10-fold those in NMMC, including 24 MDM2-negative cases (57%). P53 mutations were detected by single-strand conformation polymorphism analysis in two cases. Overexpression of MDM2 (>/=10-fold) was significantly correlated with adriamycin resistance and decreased duration of CR1. Eight of 12 (75%) overexpressers showed high levels of in vitro resistance to adriamycin, compared to four of 30 (13%) non-overexpressers (P < 0.005). The median CR1 for MDM2 overexpressers was 20.5 months (range: 3-75 months) compared to 41 months (range: 8-98 months) for non-overexpressers (P < 0.01). Four of 42 patients failed to achieve CR following re-induction: leukemic cells from three of these patients either overexpressed MDM2 or contained a mutant p53. These results indicate that overexpression of MDM2 plays a significant role in refractory pediatric ALL and is associated with early relapse, adriamycin resistance, and failure to respond to re-induction therapy. Leukemia (2000) 14, 61-67.

Quality-adjusted survival after treatment for acute myeloid leukemia in childhood: A Q-TWiST analysis of the Pediatric Oncology Group Study 8821

PURPOSE

To describe quality-of-life considerations in post-remission therapies for children with acute myelogenous leukemia.

PATIENTS AND METHODS

A quality-adjusted survival analysis, using the quality-adjusted time without symptoms or toxicity (Q-TWiST) method, was applied to Pediatric Oncology Group Trial 8821, which compared randomized assignment with intensive consolidation chemotherapy (CC) or autologous bone marrow transplantation (ABMT). Nonrandomized assignment to allogeneic bone marrow transplantation (allo BMT) on the basis of availability of a matched related donor was also evaluated. A 25-patient cohort provided data for modeling chronic graft-versus-host disease. The Q-TWiST analysis was performed based on the intent-to-treat principle.

RESULTS

As previously reported, the 3-year event-free survival was not significantly different between the randomized arms (CC v ABMT). At a median follow-up of 5 years (of the censoring distribution), the CC group had less time in toxicity (TOX) and more time without symptoms or toxicity (TWiST), relapse-free time, and alive time than patients assigned to ABMT (none of these were statistically significant). Compared with the CC group, allo BMT patients spent more time in TOX (P <.001), more time in TWiST (P =.06), and had more relapse-free time (P =.03) and time alive (P =.07). Allo BMT was superior to ABMT with greater time in TWiST (P =.02), relapse-free time (P =.01), and time alive P =.002).

CONCLUSION

The Q-TWiST analysis is a powerful decision aid in choosing among alternative therapies. Prospective information on patient preferences will facilitate future trials evaluating treatment outcomes. Refinements in the Q-TWiST method could be included to further enhance the power of this patient care decision-making tool.

In utero transplantation of human fetal haemopoietic cells in NOD/SCID mice

We have previously demonstrated that high levels of allogeneic, donor-derived mouse haemopoietic progenitor cells engraft following in utero transplantation in NOD/SCID mice. To evaluate whether the fetal NOD/SCID haemopoietic microenvironment supports the growth and development of human fetal haemopoietic progenitor cells, we injected fetal liver mononuclear cells (FL) or fetal bone marrow (FBM) derived CD34+ cells into NOD/SCID mice on day 13/14 of gestation. At 8 weeks of age 12% of FBM recipients and 10% of FL recipients were found to have been successfully engrafted with CD45+ human cells. CD45+ cells were present in the BM of all chimaeric animals; 5/6 recipients showed engraftment of the spleen, and 4/6 recipients had circulating human cells in the peripheral blood (PB). The highest levels of donor cells were found in the BM, with up to 15% of the nucleated cells expressing human specific antigens. Multilineage human haemopoietic engraftment, including B cells (CD19), myelomonocytic cells (CD13/33) and haemopoietic progenitor cells (CD34), was detected in the BM of chimaeric mice. In contrast, no human CD3+ cells were detected in any of the tissues evaluated. When the absolute number of engrafted human cells in the PB, BM and spleens of chimaeric mice was determined, a mean 16-fold expansion of human donor cells was observed. Although multilineage engraftment occurs in these fetal recipients, both the frequency and the levels of engraftment are lower than those previously reported when human cells are transplanted into adult NOD/SCID recipients.

Sensitivity to Fas-mediated apoptosis in pediatric acute lymphoblastic leukemia is associated with a mutant p53 phenotype and absence of Bcl-2 expression

Fas (APO-1/CD95) is a cell-surface protein that can mediate apoptosis upon specific ligand or antibody binding. The Bcl-2 protein may function as a modulator of Fas-induced apoptosis by blocking a downstream activation step, and Bcl-2 expression in acute lymphoblastic leukemia (ALL) cells appears to depend partly on expression of a wild-type (wt) p53 tumor suppressor gene (Findley et al, Blood 1997; 89: 2986). We therefore investigated the relationship between sensitivity to Fas-mediated apoptosis and (1) Fas expression, (2) p53 status, and (3) Bcl-2 protein levels in pediatric ALL cell lines and primary leukemic cells. Cell lines included 21 B cell precursor (BCP)-ALL and four T-ALL lines; in five cases, cryopreserved primary leukemic cells from which these lines were established were also examined. Additionally, we evaluated the effect of anti-Fas monoclonal antibody on the activation of protease CPP32 and induction of apoptosis in these lines. By SSCP analysis and DNA sequencing, we detected p53 mutations (mt) in eight out of 25 ALL cell lines (exon-7, codon 248 n=6; exon-8, codon 273, n=2). The expression of Fas and Bcl-2 was examined by immunofluorescence staining and quantified as the number of molecules of equivalent soluble fluorochrome (MESF). Elevated levels of Fas were expressed in all six lines with a mutation of p53 in codon 248 (1500 to 10800 MESF). Although Fas was detectable in seven of the 17 lines with wt-p53, expression was lower (150-900 MESF) compared with mt-p53+ lines. Bcl-2 was expressed in 10 of the 25 lines. Most (9/10) wt-p53+ lines expressed Bcl-2, whereas only one of eight mt-p53+ lines and no p53-null lines expressed this protein. Treatment of Fas-positive lines with anti-Fas monoclonal antibody (200 ng/ml) for 6 h induced activation of CPP32 and apoptosis in eight of 13 Fas+ lines. Sensitivity to Fas-mediated apoptosis was associated with a mt-p53 phenotype and absence of Bcl-2 expression. Six of eight Fas+/Fas-sensitive (S) lines were mt-53+/Bcl-2-, whereas only two Fas+/Fas-S lines were wt-p53+/Bcl-2+; both of these latter lines expressed low levels of Bcl-2 compared to Fas-resistant lines. In contrast, four of five Fas+/Fas-resistant (R) lines were wt-p53+/Bcl-2+; the exception was p53-null/Bcl-2- but expressed a low level of Fas (150 MESF). Activation of the cysteine protease CPP32 and cleavage of its substrate poly(ADP-ribose)polymerase (PARP) was also detected in Fas-S but not Fas-R lines. We obtained similar results from both the primary leukemic cells and the corresponding cell lines in five cases: overexpression of Fas and Fas-sensitivity were present in mt-p53+/Bcl-2- but not wt-p53+/Bcl-2+ cells. These results suggest that some pediatric ALL cells expressing mt-p53+ may be sensitive to Fas-mediated apoptosis due to high levels of Fas expression and lack of Bcl-2, and further suggest that molecular methods of activating Fas may be useful for therapy of refractory ALL with the Fas+/mt-p53+ phenotype.

Lymphoid reconstitution after autologous PBSC transplantation with FACS-sorted CD34+ hematopoietic progenitors

T-cell and B-cell reconstitution was studied in nine patients who received fluorescence activated cell sorter (FACS)-sorted autologous CD34+ hematopoietic progenitor cells (HPC). The mean numbers of T cells (CD3+), B cells (CD19+) and CD34+ HPC administered to each patient were .004, .002, and 1.8 x 10(6) cells/kg, respectively. After high-dose myeloablative chemotherapy (busulfan, cyclophosphamide, etoposide) CD34+) HPC were infused and lymphoid reconstitution was monitored using flow cytometry and reverse transcriptase-polymerase chain reaction (RT-PCR) amplification of VDJ T-cell receptor (TcR) sequences. Restoration of normal numbers of peripheral blood T cells and B cells among recipients of FACS-sorted CD34+ HPC was delayed compared to recipients of non-T-cell-depleted PBSC autografts. In both patient groups, the circulating T cells were primarily CD4-, CD8+, alphabeta TcR+, and CD45RO+, CD45RA- during the first 2 months after transplant. Subsequent increases in the frequency of CD45RA+ CD45RO- T cells occurred at 2 to 3 months after transplant, suggesting maturation of CD34+ hematopoietic progenitors to "naive" T cells. Analysis of the TcR repertoire after hematopoietic reconstitution demonstrated decreased diversity of Vbeta TcR expression associated with global decreases in the absolute number of total peripheral blood T cells and most Vbeta TcR+ subsets. Three of nine recipients of FACS-sorted CD34+ HPC demonstrated significant increases in the percentage of gammadelta+ peripheral T cells and CD5+ B cells at 3 to 9 weeks after transplantation, and all patients had transient oligoclonal expansions of T cells expressing specific Vbeta TcR. Transplantation with highly purified CD34+ HPC results in reduced diversity of the peripheral T-cell repertoire during the early post-transplant period compared with patients receiving unmanipulated or MoAb-depleted transplants.

Pediatric acute myelogenous leukemia cells express IL-6 receptors and are sensitive to a recombinant IL6-Pseudomonas exotoxin

We have studied IL-6 receptor (IL-6R) expression on AML cells from 15 pediatric patients by immunocytochemistry/flow cytometry, reverse-transcription polymerase chain reaction, and Scatchard analysis. High-affinity IL-6R were detected on leukemic cells from 12 (80%) patients. Binding sites per cell ranged from 140 to 3580 (median 920; mean 1240), with dissociation constants of 0.26 to 0.71 nM. We therefore assessed the in vitro sensitivity of IL-6R+ AML cells to treatment with a recombinant IL6-Pseudomonas exotoxin fusion protein (IL6-PE4E), using the XTT cytotoxicity assay. Leukemic cells from eight patients had ID50 values (concentration of IL6-PE4E producing a 50% decrease in cell viability) of <1000 ng/ml (median, 87 ng/ml; mean, 262 ng/ml). Sensitivity to IL6-PE4E correlated significantly with receptor number. Normal bone marrow mononuclear cells had undetectable IL6-R expression (<20 receptors/cell) and were relatively resistant to IL6-PE4E. To test the efficacy of IL6-PE4E for ex vivo purging in an autologous stem cell transplantation setting, we incubated primary IL-6R+ AML cells with 10(3) ng/ml IL6-PE4E for 24 h, followed by inoculation into SCID mice. Mice receiving treated cells showed no leukemic engraftment, while all mice receiving untreated or control-treated cells developed leukemia with a median presymptomatic interval of 55 days. In recipients of IL6-PE4E treated cells, no evidence of occult leukemia was detected by PCR analysis of blood and bone marrow cells at 185 days postinoculation. These data suggest that IL-6R are expressed on leukemic cells from a substantial percentage of pediatric AML patients. Furthermore, leukemic cells expressing high numbers of IL6-R may be sensitive to IL6-PE4E in an ex vivo purging protocol.

Cancers after stem cell transplantation in children: a cost of cure?

There have been many advances in the field of stem cell transplantation over the past few years. Newer preparative regimens and better supportive care measures are allowing more children to survive and lead long lives after stem cell transplantation. We are only now beginning to see what the cost of this success may be in regard to the long-term effects of this treatment. Most long-term follow up studies have been reported in the adult literature and there is very little information on the pediatric population. Children who undergo stem cell transplantation will have a long period of observation in which new and different complications can occur. This article looks at the current literature on secondary malignancies and lymphoproliferative disorders after stem cell transplantation in children, with particular focus on risk factors, possible strategies for prevention, and treatment.

Sustained multilineage engraftment of allogeneic hematopoietic stem cells in NOD/SCID mice after in utero transplantation

Substantial barriers exist to the engraftment of hematopoietic cells in mice after in utero transplantation. Although high levels of donor-derived hematopoiesis have been reported in SCID mice, the majority of chimeric recipients exhibit decreasing levels of donor cells over time. To directly test whether the natural killer cell and macrophage activity of the recipients represents a barrier to sustained engraftment, fetal NOD/SCID mice were injected on day 13.5 of gestation with an enriched congenic hematopoietic progenitor cell population. Forty-four percent of pups showed the presence of Ly5.1+ donor cells 4 weeks after transplantation. The mean number of donor-derived nucleated cells in the peripheral blood (PB) was 30%. Although the majority of circulating donor cells were lymphocytes, up to 15% expressed myelomonocytic markers. Serial PB samples from individual mice indicated that the percentage of circulating donor cells increased from 17% to 55% between 4 and 24 weeks. At 6 months posttransplantation, an increased frequency of multilineage, donor-derived cells was also observed in the bone marrow (BM) and the spleen of chimeric recipients. The engraftment of pluripotent hematopoietic stem cells was evaluated by transplanting BM from chimeric mice into irradiated congenic recipients. Irradiated secondary recipients also exhibited multilineage donor-derived hematopoiesis in the PB, BM, and spleen for up to 6 months. These results show that the in utero transplantation of lineage-depleted BM cells into NOD/SCID recipients produces a high frequency of sustained engraftment of allogeneic hematopoietic stem cells.

Expression of interleukin-6 receptors by pediatric acute lymphoblastic leukemia cells with the t(4;11) translocation: a possible target for therapy with recombinant IL6-Pseudomonas exotoxin

We have detected expression of interleukin-6 receptors (IL-6R) by primary leukemic cells from three of six patients with t(4;11)+ ALL. Scatchard analysis revealed from 960 to 2100 high-affinity IL-6R/cell on these cells (median, 1560; mean, 1540). All three IL-6R+ cases also expressed CD33, which was not expressed on IL-6R-negative cases. To determine if these receptors could serve as a target for a recombinant ligand-toxin, we examined the sensitivity of primary IL-6R+ ALL cells to a recombinant IL6-Pseudomonas exotoxin (IL6-PE4E) fusion protein, in which the toxicity and specificity of the chimeric toxin was enhanced by substitution of four glutamine residues for naturally occurring amino acids in PE domain I. Primary cells from IL-6R+ cases were sensitive to IL6-PE4E in a 48-h cytotoxicity assay, with ID50 values (concentrations causing 50% decrease in viability) ranging from 23 ng/ml to 92 ng/ml (median, 61; mean, 58). Furthermore, incubation of these cells with 10(3) ng/ml IL6-toxin for 24 h prevented their subsequent engraftment in SCID mice. Thus, IL6-PE4E may be useful for ex vivo purging of IL-6R+ leukemic cells in an autologous bone marrow transplantation setting and possibly for therapy of residual, chemotherapy-resistant disease.

Expression and regulation of Bcl-2, Bcl-xl, and Bax correlate with p53 status and sensitivity to apoptosis in childhood acute lymphoblastic leukemia

Bcl-2 and its homologue, Bcl-x1, encode membrane-associated proteins that protect neoplastic cells from DNA damage-induced apoptosis, whereas Bax is a Bcl-2 antagonist that promotes cell death. In the present study, we examined the expression and regulation of these genes at both the mRNA and protein level in 22 pediatric acute lymphoblastic leukemia (ALL) cell lines, as well as their sensitivity to apoptosis after exposure to ionizing radiation (IR). Eleven of 22 lines expressed wild-type (wt) p53, 4 expressed mutant p53, and 7 did not express p53 (p53-null). Nine of 22 (41%) lines expressed Bcl-2; of these, 8 were wt-p53+ and 1 expressed mutant p53. Bcl-2 was not expressed in any p53-null lines. In contrast, all 22 lines were positive for Bcl-x1 and Bax, although expression level varied. Treatment with IR (10 Gy) induced both downregulation of Bcl-2 and upregulation of Bax at 2 to 5 hours post-IR in 5 of 8 (63%) wt-p53+ lines, leading to apoptosis. Conversely, lines that failed to both downregulate Bcl-2 and upregulate Bax after IR were resistant to apoptosis. Although levels of Bcl-x1 expression varied among the 22 lines, high levels of Bcl-x1 were observed in 5 of 7 (71%) p53- lines. There were no obvious changes in the expression of Bcl-x1 in these lines after IR. However, among the p53-null lines, resistance to IR was observed only in those expressing high levels of Bcl-x1. These results suggest that expression of Bcl-2 but not Bcl-x1 is p53-dependent and that IR-induced downregulation of Bcl-2 and upregulation of Bax occur in most wt-p53+ lines and are associated with radiosensitivity. Furthermore, high-level expression of Bcl-x1 occurs predominantly in p53-null lines and is associated with resistance to IR-induced apoptosis in these lines, indicating differential expression and regulation of Bcl-2 and Bcl-x1 in pediatric ALL.

Steroid-refractory cutaneous graft-versus-host disease after transplantation of haploidentical parental CD34+ cells in children with Down's syndrome and recurrent acute leukemia

PURPOSE

This study evaluated the feasibility of performing haploidentical CD34+ selected transplants for children with Down's syndrome (DS) and recurrent leukemia.

PATIENTS AND METHODS

Within a cohort of 15 children, two patients had DS. Transplantation of CD34+ cells from haploidentical parents was performed after the children were conditioned with fractionated total body irradiation, cyclophosphamide, and antithymocyte globulin (ATG). Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine and a short course of methotrexate.

RESULTS

The preparative regimen was well tolerated, and engraftment of polymorphonuclear cells and platelets took place promptly (by day 20) in both patients with DS. However, both patients with DS experienced severe grade IV GVHD that was limited to the skin and was refractory to salvage with high-dose methylprednisolone therapy. In one patient, GVHD responded to second-line salvage therapy with ATG, but the patient died on day 234 from leukemic relapse. The second patient had GVHD that did not respond to ATG and died of multisystem organ failure and refractory GVHD on day 44. Two of two DS patients had steroid refractory severe acute GVHD of the skin, while only one of 11 evaluated and identically treated non-DS patients had severe GVHD (p < 0.05).

CONCLUSION

These observations in patients who underwent mismatched bone marrow transplantation suggests that patients with DS have an increased risk of severe acute GVHD of the skin in this context.

Incidence and clinical significance of CDKN2/MTS1/P16ink4A and MTS2/P15ink4B gene deletions in childhood acute lymphoblastic leukemia

We have examined the incidence and clinical significance of deletions of two candidate tumor suppressor genes, CDKN2/MTS1/p16ink4A and MTS2/p15ink4B, in pediatric acute lymphoblastic leukemia (ALL). Gene deletion was evaluated in leukemic bone marrow (BM) cells obtained at diagnosis from 105 pediatric ALL patients: 83 with B-cell precursor (BCP-ALL) ALL and 22 with T-ALL. CDKN2/p16 deletion was seen in 23 of the 83 (28%) BCP-ALL and 15 of the 22 (68%) T-ALL cases. A virtually identical pattern of MTS2/p15 deletion was detected in these patients: p15 was deleted in 37 of 38 cases with p16 deletion, and p15 was not deleted in any p16-positive specimens. P16/p15 deletions were significantly related to poor prognosis factors including age under 1 year (P < 0.001), initial white cell counts greater than 50 x 10(9) per liter (P < .001), and T cell phenotype P < .005). Analysis of all 105 patients revealed that the 5-year disease-free survival rate was 68% for patients without p16/p15 deletions and 35% for those with p16/p15 deletions (P < .005). The association between gene deletion at initial diagnosis and unfavorable outcome suggests that loss of these genes is clinically significant and indicates a need for prospective studies of p16/p15 deletion in pediatric ALL patients.

Autologous bone marrow transplantation versus intensive consolidation chemotherapy for acute myeloid leukemia in childhood. Pediatric Oncology Group

BACKGROUND

The value of autologous bone marrow transplantation in the treatment of children with acute myeloid leukemia (AML) is unknown. We compared autologous bone marrow transplantation with intensive consolidation chemotherapy as treatments for children with AML in first remission.

METHODS

We induced remission with one course of daunorubicin, cytarabine, and thioguanine, followed by one course of high-dose cytarabine (3 g per square meter of body-surface area for six doses). Patients in remission after the second course of induction therapy were eligible for randomization. Between June 1988 and March 1993, 552 of 649 enrolled patients who could be evaluated (85 percent) entered remission. A total of 209 patients were not eligible for randomization; of the remaining 343 patients, 232 were randomly assigned to receive six courses of intensive chemotherapy (117 patients) or autologous transplantation (115 patients). Of the original 649 patients, 189, including 21 with Down's syndrome, were nonrandomly assigned to receive intensive chemotherapy.

RESULTS

The rates of event-free survival and overall survival for the entire group at three years were 34 +/- 2.5 percent and 42 +/- 2.6 percent, respectively. For patients who were randomly assigned to one of the two treatment groups, the mean (+/- SE) rates of event-free survival three years after randomization were not significantly different in the two groups when examined by intention-to-treat analysis: 36 +/- 5.8 percent for the intensive-chemotherapy group as compared with 38 +/- 6.4 percent for the autologous-transplantation group; and the relative risk of treatment failure for the chemotherapy group as compared with the autologous-transplantation group was 0.81 (P = 0.20 by the log rank test; 95 percent confidence interval, 0.58 to 1.12). Overall survival at three years followed a similar pattern. There was a lower relapse rate (31 percent vs. 58 percent, P < 0.001) but a higher rate of treatment-related mortality (15 percent vs. 2.7 percent, P = 0.005) in the group treated with autologous transplantation than in the intensive-chemotherapy group. The event-free survival at three years for the nonrandomized intensive-chemotherapy group was 39 +/- 5.1 percent, and for a contemporaneous group of patients each of whom received a histocompatible bone marrow transplant from a sibling, it was 52 +/- 8.0 percent.

CONCLUSIONS

Treatment of children with AML in first remission with either autologous bone marrow transplantation or intensive chemotherapy prolongs event-free survival equally.

Engraftment potential of different sources of human hematopoietic progenitor cells in BNX Mice

Human hematopoietic progenitor cells (HPCs) from mobilized peripheral blood mononuclear cells (PBMCs), adult bone marrow (ABM), and fetal bone marrow (FBM) were evaluated for their ability to produce multilineage human hematopoietic engraftment in vivo. Sublethally irradiated BNX (beige/nude/xid) mice were injected with either unfractionated cells or CD34+ cells purified from these sources. The presence of human cells in the mouse PB, BM, and spleen was evaluated by flow cytometry at either 6 to 8 weeks or 6 months postinjection. Recipients with > or = 1% human cells in any of these tissues were considered chimeric. Of 26 mice injected with FBM, 4 showed up to 73% human cells in the BM or spleen at 6 months. The phenotypes of these cells included CD13/33+ myelomonocytic cells (38%), CD19+ B cells (67%), and CD34+ progenitor cells (28%). In contrast, ABM gave rise to a mean of 5% human cells in the PB in 2 of 42 (4%) recipients at 6 to 8 weeks. These circulating human cells were predominantly CD3+, whereas CD13/33+ and CD34+ cells were detected in the BM for up to 6 months. A total of 18% of mice injected with PBMCs showed a mean of 36% human cells in the PB. Both the BM and spleens of PBMC-injected mice contained CD3+ cells in a proportion similar to that observed in the PB. These CD3+ cells were phenotypically mature CD4+,CD8- or CD4-,CD8+ T cells and coexpressed a variety of Vbeta T-cell receptor (TCR) genes. The percentage of CD3+ cells in the circulation of chimeric recipients injected with either FBM, ABM, or PBMCs correlated well with the input CD3+ cell dose for each of these HPC sources (r = .99). The high levels of engraftment of CD3+ cells in recipients of PBMCs and the long-term multilineage engraftment of FBM recipients have important implications for developing strategies to study the regulation of these human cells in vivo.

Homozygous deletions of the CDKN2 (MTS1/p16ink4) gene in cell lines established from children with acute lymphoblastic leukemia

Homozygous deletions of the CDKN2 (MTS1/p16ink4) gene have been found at high frequency in cell lines derived from a variety of adult solid tumors. In order to investigate the status of the CDKN2 gene in cell lines established from childhood acute lymphoblastic leukemia (ALL), we surveyed 25 lines representing the major pediatric ALL phenotypes for the presence of this gene by Southern blot analysis. Homozygous deletions of all or part of the CDKN2 gene were detected in 21 (84%) cell lines, including 11 of 14 (79%) early-pre-B-ALL, four of five (80%) pre-B-ALL, and six of six T-ALL lines. CDN2 mRNA was detected by Northern blotting in each of the four lines containing an intact CDKN2 gene. These data suggest an important role for CDKN2 deletion in the cause and/or progression of pediatric ALL.

Overexpression of the MDM2 gene by childhood acute lymphoblastic leukemia cells expressing the wild-type p53 gene

The wild-type (wt) p53 tumor suppressor gene is commonly inactivated in human malignancies, either by mutations or by loss of expression. An additional proposed mechanism for inactivation of wt-p53 is amplification of the murine double minute 2 (MDM2) gene and overexpression of the MDM2 protein, which binds to p53 and eliminates its tumor suppressor function. To investigate a potential role for MDM2 in the inactivation of wt-p53 in pediatric acute lymphoblastic leukemia (ALL), we examined the expression of MDM2 and p53, as well as the occurrence of p53 mutations and possible amplification of the MDM2 gene, in 19 pediatric ALL cell lines and one pediatric acute myelogenous leukemia (AML) line. Although we did not find significant amplification of the MDM2 gene in any of the leukemic lines, we detected overexpression of MDM2 in all 10 lines that expressed wt-p53. Of the 10 lines without overexpression of the MDM2 gene, six (including the AML line) did not express p53, and four expressed mutant p53 with single point mutations in exons 7 and 8. To determine whether primary leukemic cells showed a similar correlation, we analyzed the original cryopreserved leukemic bone marrow cells from seven patients from whom cell lines were established. We obtained similar results from both the primary leukemic cells and the corresponding cell lines: overexpression of MDM2 was present in primary cells that expressed wt-p53 but not in cells that lacked expression of wt-p53. These findings suggest an important role for MDM2 in the pathogenesis of pediatric ALL in which leukemic cells express wt-p53.

The effects of bone marrow transplantation on pulmonary function in children

The effects of current methods of bone marrow transplantation (BMT) on pulmonary function in children have not been extensively studied. We reviewed serial pulmonary function tests (PFTs) in 25 children (median age, 9 yr; range, 4-15) who received allogeneic (n = 14) or autologous (n = 11) BMT for neoplastic diseases at The Johns Hopkins Hospital. The PFTs were obtained before BMT and at 6 months (early) and 15 months (late) after transplant. In all but 6 patients, PFTs were normal before BMT. A mild transient decline in carbon monoxide diffusing capacity (DLCO) was observed early after BMT but returned to baseline levels in the late post-BMT period. A trend towards worsening of PFTs with increasing age of patients was observed. The presence of graft-versus-host disease (GVHD) and pretransplant seropositivity for cytomegalovirus (CMV) were associated with significant decrements in several measurements of pulmonary function in the early post-BMT period. Patients given bulsulfan-containing preparative regimens tended to have less impairment of PFTs than those given other regimens using other combination high-dose chemotherapeutic agents or total body irradiation. These findings suggest that abnormalities in PFTs are common in the first months after BMT in pediatric patients but are not consistently associated with impairment of lung function when studied 15 months post-transplant.

Cyclosporine-induced graft-versus-host disease after autologous bone marrow transplantation for acute myeloid leukemia

Rodents given cyclosporine (CSP) for several weeks after autologous or syngeneic bone marrow transplantation develop a syndrome that mimics allogeneic graft-versus-host disease (GVHD). Autologous GVHD has also been reported after administration of CSP in patients who have received autologous bone marrow transplantation (ABMT) with untreated marrow for lymphoma or acute myeloid leukemia (AML). Our study was designed to determine whether CSP administration is associated with appearance of autologous GVHD in patients with AML receiving ABMT with 4-hydroperoxycyclophosphamide (4HC)-purged marrow and whether there was a dose-dependent effect of CSP on development of the syndrome. Thirty-three patients with AML (18 in first remission [CR1], 10 in CR2, and 5 in CR3) received intravenous CSP, beginning on the day of ABMT, after a preparative regimen of busulfan and cyclophosphamide and ABMT with 4HC-treated marrow. Skin biopsies were obtained weekly after ABMT or on appearance of rash and were graded for GVH changes. In the first phase of this study, groups of patients received CSP dosages of either 1 mg/kg/day (7 patients), 2.5 mg/kg/day (8 patients), or 3.75 mg/kg/day (6 patients) for 28 days. Sixteen of the 21 patients (76%) developed cutaneous histopathologic grade 2 GVHD at a median of 34 days (range, 14-49) after ABMT, and cutaneous manifestation were present at time of positive biopsy in 11 of the 16 patients. There was no apparent difference in frequency, time to onset, or duration of GVHD among the three CSP dosage groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Hematopoietic cell transplantation in the twitcher mouse. The effects of pretransplant conditioning with graded doses of busulfan

The effects of congenic hematopoietic cell transplantation (HCT; transplantation of bone marrow and spleen cells) after graded doses of busulfan (BU), a myeloablative but nonimmunosuppressive alkylating agent, were evaluated in the twitcher mouse model of human galactosylceramidase deficiency, a demyelinating sphingolipid storage disease. C57BL/6 twitcher mice (immunophenotype Ly-5.1) were given 10 to 50 mg/kg of BU or total-body irradiation (9.0 Gy) at age nine days and HCT from congenic Ly-5.2 donors 24 hr later. The 30-day post-HCT survival, an indicator of tolerance of the preparative regimen, was at least 83% in twitcher mice given 45 mg/kg or less of BU, was 50% in recipients of 50 mg/kg BU and 75% in TBI-conditioned twitchers. The lifespan of twitcher mice given HCT after 10 or 20 mg/kg of BU was similar to that of untreated twitchers (median survival, 42 days; range, 30-47). In contrast, mice transplanted after 35 to 50 mg/kg of BU had significantly prolonged survival (median, 82 days; range, 56-208) and stabilization of hindlimb paralysis, similar to TBI-conditioned recipients. Post-HCT repopulation by donor Ly-5.2 cells was determined by flow cytometry. Thirty days after HCT, only 11-15% of lymphohematopoietic cells in blood, bone marrow, and spleens were of Ly-5.2 donor origin in twitcher mice transplanted after 10 mg/kg of BU but 60-80% were of Ly-5.2 donor origin in mice transplanted after higher doses of BU. These levels further increased to 70-90% by 90 days after HCT, comparable to that seen after TBI. Levels of galactosylceramidase in livers, spleens, and brains of twitchers transplanted after 35-50 mg/kg of BU or after TBI increased to 30-116% of normal control values by 90 days after HCT. Conditioning for HCT with as little as 35 mg/kg of BU provides minimal peritransplant mortality, rapid and sustained establishment of donor lymphohematopoiesis, replacement of lysosomal hydrolase, and prolonged survival in this murine model of human sphingolipidosis.

Recent developments in experimental bone marrow transplantation

Animal models have been a mainstay for the preclinical evaluation of the principles of bone marrow transplantation. In addition to evaluation of the therapeutic effectiveness of bone marrow transplantation in oncologic, hematologic, immunologic, and genetic diseases, experimental animal models provide insights into the immunobiology of bone marrow transplantation, including engraftment kinetics, tolerance, immune reconstitution, and graft-versus-host and graft-versus-tumor reactions. Many recent developments in experimental bone marrow transplantation resulted from not only the discovery of new spontaneously occurring models of human diseases but also the induction of neoplastic diseases, eg, chronic myelogenous leukemia, by transplantation of syngeneic marrow cells into which foreign genes were introduced and expressed. Murine bone marrow transplantation systems are an exemplary format for the study of clinically relevant basic aspects of hematopoiesis, including identification and elucidation of the biology of hematopoietic stem cells. Finally, the application of molecular biologic techniques in animal bone marrow transplantation models provides the preclinical foundations of gene-insertion therapy by transplantation of syngeneic or autologous marrow-derived stem cells into which functional genes were inserted.

Treatment of recurrent and refractory pediatric solid tumors with high-dose busulfan and cyclophosphamide followed by autologous bone marrow rescue

PURPOSE

The purpose of this study was to determine the toxicities of and responses to high-dose busulfan and cyclophosphamide with autologous bone marrow transplant (ABMT) in patients with recurrent or refractory pediatric solid tumors.

PATIENTS AND METHODS

We treated 18 patients (ages, 2 to 38 years; median, 14) who had tumors that were resistant to conventional chemotherapy and radiotherapy with busulfan 16 mg/kg and cyclophosphamide 200 mg/kg. Seventeen patients received bone marrow purged with 4-hydroperoxycyclophosphamide; one received unpurged marrow.

RESULTS

Despite extensive prior treatment, including radiotherapy in 16 patients, toxicity generally was acceptable. For seven patients with measurable disease, there were three partial responses of 2, 10, and 20 months' duration, three patients with stable disease (SD), and one early, toxic death. Of the 11 patients with no measurable disease at the time of transplantation, one patient with osteosarcoma continues in remission at 57+ months and one third of the patients survived for at least 16 months. Mucositis was the predominant nonhematopoietic toxicity.

CONCLUSION

Although the high-dose busulfan and cyclophosphamide combination showed modest activity, changes in the preparative regimen should be considered to improve the response rate in refractory tumors.

Optimization of busulfan dosage in children undergoing bone marrow transplantation: a pharmacokinetic study of dose escalation

Busulfan (BU) is a widely used myeloablative and antineoplastic agent in clinical bone marrow transplantation (BMT). The lower incidence of BU-associated toxicities and lower therapeutic effectiveness in young children given BU doses based on body weight (ie, 16 mg/kg) is associated with altered pharmacokinetics of BU; the area under the curve (AUC) of BU concentration versus time is significantly less in these patients than those observed in older children and adults. To optimize BU dosage in young BMT recipients, we developed a dosage regimen based on body surface area (BSA) and determined BU pharmacokinetics and BU-associated toxicities. Seven children (median age, 3.9 years, range, 1.1 to 5.7) undergoing allogeneic or autologous BMT for leukemia received 40 mg/m2/dose BU every 6 hours for 16 doses; BU concentrations were measured in the plasma, and AUCs were determined for each patient after the first and 13th doses. Expressed as a function of body weight, the median BU dosage was 26.4 mg/kg (range, 24.3 to 28.2), a 60% increase over the BU dosage based on body weight. Four patients developed mucositis, and one of them also developed nonfatal hepatic veno-occlusive disease (VOD). No patients receiving 40 mg/m2 BU developed neurotoxicity (eg, seizures) or interstitial pneumonitis. Prompt and sustained engraftment was observed in the allogeneic BMT recipients, and late graft failure was not seen. The mean BU AUCs were 1,105 mumol/L.min (range, 790 to 2,080) after the first dose and 1,022 mumol/L.min (range, 632 to 1,860) after the 13th dose of BU, comparable to the AUCs in adults given 16 mg/kg of BU. These studies suggest that, in young children, BSA-based dosing of BU (40 mg/m2) provides drug exposures (AUCs) closer to adult values with acceptable toxicities and may improve therapeutic effects.

Induction of cutaneous graft-versus-host disease by administration of cyclosporine to patients undergoing autologous bone marrow transplantation for acute myeloid leukemia

Cutaneous graft-versus-host disease (GVHD) has been reported after administration of cyclosporine (CSP) after autologous bone marrow transplantation (ABMT) with unpurged marrow in patients with lymphoma. To determine whether GVHD can be induced after ABMT with chemopurged marrow in acute myeloid leukemia (AML), we administered intravenous CSP for 28 days (beginning on the day of ABMT) to 19 patients with AML (12 in first remission [CR1], six in CR2, and one in CR3) who received busulfan (16 mg/kg) and cyclophosphamide (200 mg/kg) and ABMT with 4-hydroperoxycyclophosphamide (4HC)-treated marrow. In this dose-escalation trial, CSP daily doses were 1 mg/kg in seven patients, 2.5 mg/kg in eight patients, or 3.75 mg/kg in four patients. Skin biopsies were obtained weekly after ABMT or on appearance of rash and were graded for GVH changes. Overall, 15 of 19 patients (79%) had cutaneous histopathologic grade 2 GVHD at a median of 33 days (range, 14 to 49) after ABMT; in 10, cutaneous manifestations were present at time of positive biopsy. The frequency, time to onset, and duration of GVHD were similar among the three CSP dosage groups. No patients had hepatic or gastrointestinal dysfunction attributable to GVHD or required specific therapy for GVHD. Positive biopsies for GVHD were seen in seven of eight patients who received full-course, full-dose CSP and 8 of 11 patients who had CSP discontinued or dosage reduced because of renal insufficiency. Three patients (one with positive biopsy) died with ABMT-related complications. Seven patients (four CR1, three CR2) relapsed with AML at a median of 411 days (range, 178 to 549) after ABMT; six of seven had positive biopsies for cutaneous GVHD. Nine patients (seven CR1, one CR2, and one CR3) are alive without relapse at a median of 501+ days (range, 252+ to 811+) after ABMT; eight of nine had cutaneous GVHD. Short-course CSP can induce autologous GVHD in recipients of chemopurged marrow autografts for AML, but randomized prospective trials are needed to determine whether this immunologic reaction is associated with alterations in leukemic relapse rate and disease-free survival after ABMT in AML.

Growth retardation and depigmentation of hair after high-dose busulfan and congenic hematopoietic cell transplantation in mice

Busulfan, a myeloablative but non-immunosuppressive alkylating agent, is used extensively in clinical bone marrow transplantation (BMT), but the effects of high-dose administration have not been previously evaluated in preclinical BMT settings with young murine recipients. We compared the survival and growth of C57BL/6 mice given graded single doses of busulfan (10-100 mg/kg) or total body irradiation (TBI; 900 cGy) at age 9 days and hematopoietic cell transplantation (HCT; transplantation of congenic bone marrow and spleen cells) 24 h later. The 30-day survival was 87-100% in mice transplanted after 10-40 mg/kg busulfan and 79% after TBI, but fell to 54% and 33%, respectively, after 80 mg/kg and 100 mg/kg busulfan, suggesting that this latter dosage range represents the LD50 for single-dose busulfan in young C57BL/6 mice given stem cell rescue. The weights of 10-week-old mice given HCT after lower doses of busulfan ranged from 87% of control at 10 mg/kg to 64-69% of control in mice conditioned with 35-65 mg/kg busulfan or TBI. Impairment of weight gain was most striking (approximately 50% of control) in mice transplanted after 80-100 mg/kg busulfan. Despite retardation of somatic growth, the brain weights of busulfan-conditioned mice remained at least 90% of control, and there were no obvious neuropathological alterations in the brains of these animals. All mice treated with at least 20 mg/kg busulfan or TBI lost hair by 3-4 weeks after transplant.(ABSTRACT TRUNCATED AT 250 WORDS)

Growth in children after bone marrow transplantation: busulfan plus cyclophosphamide versus cyclophosphamide plus total body irradiation

Growth was assessed during the first and second years following bone marrow transplantation (BMT) in 47 children treated by either busulfan plus cyclophosphamide (BU/CY) (n = 24) or cyclophosphamide plus fractionated total body irradiation (CY/TBI) (n = 23). Before transplant, the median height was only 0.2 SD below age- and sex-adjusted means (range, -2.5 to +3.0). Height was greater than 2.0 SD below normal in only three patients (6%). The pretransplant heights were comparable in the BU/CY and CY/TBI groups (-0.1 v -0.6 SD, P = .35). Following transplant, median 1- and 2-year heights were 0.7 and 0.9 SD below normal, respectively. Growth rates were 2.2 SD and 1.4 SD below normal during the first and second years, respectively. Growth rates were greater than 2.0 SD below normal in 24 of 47 (51%) at 1 year and in 12 of 31 (39%) at 2 years after transplant. Growth rates in patients treated with BU/CY were comparable to those treated with CY/TBI during both years: -2.5 versus -1.7 SD during the first year (P = .19, Wilcoxon), and -1.5 versus -1.1 SD during the second year (P = .61). Growth rates during the second year correlated with growth rates during the first year (r = .36, P = .046). Growth rates during the first year were lower in patients who had been given prior cranial irradiation, those who were near pubertal age at the time of transplant, and those who were transplanted for a disease other than acute lymphoblastic leukemia (ALL). During the second year, poor rates of growth were associated only with the use of corticosteroids after transplant.

Lymphoid reconstitution after transplantation of congenic hematopoietic cells in busulfan-treated mice

The effects of pretransplant conditioning with high-dose busulfan, a myeloablative but nonimmunosuppressive alkylating agent, on reconstitution of lymphoid tissues by donor cells after bone marrow transplantation (BMT) has not been extensively examined. We used flow cytometric analyses to study the kinetics and extent of lymphocyte repopulation in C57BL/6 mice (immunophenotype Ly-5.2) given graded doses of busulfan (10 to 100 mg/kg) or total body irradiation (TBI; 900 rad) and hematopoietic cell transplantation (HCT; transplantation of bone marrow and spleen cells) from congenic Ly-5.1 donors. Mice transplanted after 10 mg/kg of busulfan had slow and incomplete lymphoid engraftment; only 6% to 11% of lymphocytes in the peripheral blood, lymph nodes, and spleen were positive for Ly-5.1 at 30 days after transplant, slightly increased to 13% to 20% at 60 days, and stabilized at 40% to 46% by 180 days after HCT. Higher doses of busulfan (20 to 100 mg/kg) provided dose-dependent congenic lymphoid reconstitution. Thirty days after HCT, the range of Ly-5.1 cells in blood, lymph nodes, and spleen of Ly-5.2 recipient mice was 43% to 54% after 20 mg/kg of busulfan, 66% to 71% after 50 to 80 mg/kg, and 77% to 85% after 100 mg/kg. Sixty days after transplant, lymphoid chimerism increased to 57% to 68% in 20 mg/kg recipients, 72% to 79% after 35 mg/kg, and 75% to 90% in animals given 50 mg/kg or greater, as seen in radiation chimeras. Despite slower early reconstitution after lower doses of busulfan, donor lymphocytes exceeded 90% to 95% by 90 to 120 days after HCT in all mice given at least 20 mg/kg. Even though busulfan lacks directly immunosuppressive properties, virtually complete sustained lymphoid reconstitution by transplanted congenic donor stem cells occurs after its administration. These observations suggest that pretreatment with busulfan may be effective in gene therapy strategies that involve infusion of autologous marrow cells into which functional genes have been inserted.

Pediatric bone marrow transplantation

Bone marrow transplantation is a high-visibility, high-technology discipline with a growing list of potentially curative applications in neoplastic, hematologic, immunologic, and genetic diseases of children. The clinical problems experienced by children who have received bone marrow grafts involve pediatricians in both general and subspecialty practice and require a working knowledge of the applied immunobiology of bone marrow transplantation. As the transplantation procedure has evolved from a research tool to an established therapeutic modality, collaboration between basic scientists and clinicians has led to a greater understanding of the pathophysiology of the immunobiologic events that occur after allogeneic bone marrow transplantation and, thus, to improvements in the clinical care of transplant recipients. As its applications in pediatric medicine grow, bone marrow transplantation presents additional challenges for the future: expansion of the allogeneic donor pool, use of more effective or aggressive preparative regimens and autologous bone marrow-purging strategies, more effective prevention and treatment of graft-versus-host disease, and development of gene-replacement therapy by infusion of modified autologous bone marrow cells.

Analysis of factors predicting speed of hematologic recovery after transplantation with 4-hydroperoxycyclophosphamide-purged autologous bone marrow grafts

We previously described the predictive value of graft colony-forming units granulocyte macrophage (CFU-GM) content after 4-hydroperoxycyclophosphamide (4-HC) purging for the duration of aplasia after autologous bone marrow transplantation. Despite the uniform 4-HC concentration, we observed heterogeneity in CFU-GM survival and the kinetics of engraftment. We have now analysed patient and graft characteristics for 154 patients undergoing autologous transplantation with 4-HC purged grafts to further define this heterogeneity. Patients transplanted for the treatment of malignant lymphoma reached a peripheral blood granulocyte count of greater than 0.5 x 10(9)/l (median, 20 versus 40 days; p less than 0.001) and platelet transfusion independence (median, 30 versus 70 days; p less than 0.001) significantly faster than patients transplanted for acute non-lymphoblastic leukemia. Other diagnostic groups were intermediate. These differences were independent of graft CFU-GM content. Multiple other patient and graft factors including patient age, peripheral blood counts on day of harvest, and amounts of other hematopoietic progenitors also predicted the kinetics of engraftment in univariate and multivariate analysis. Cytomegalovirus infection during the aplastic period predicted a delay in granulocyte (p = 0.024) but not platelet recovery (p = 0.174). This analysis demonstrates that multiple patient, graft, and post-transplant factors predict the engraftment capacity of autografts, and the kinetics of engraftment with 4-HC purged grafts. The multiple predictive factors explain a significant portion of the variability in engraftment kinetics observed after transplantation with 4-HC purged autografts.

Predictive value of colony-forming unit assays for engraftment and leukemia-free survival after transplantation of chemopurged syngeneic bone marrow in rats

We evaluated the efficacy of in vitro clonogenic assays for acute myeloid leukemia (AML) (CFU-Leuk) and granulocyte-macrophage progenitor cells derived from normal bone marrow (BM) (CFU-GM) to predict hematopoietic engraftment, median survival time (MST) and leukemia-free survival (LFS) in LBN rats that received injections of untreated or drug-treated AML and/or normal BM cells. Injection of untreated AML cells resulted in a log-linear relationship between AML cell dose and time of death from leukemia; LBN rats given 10(6) cells died with AML (MST, 24 days; range, 19-28) after injection. A minimum of 0.5-1.0 X 10(6) untreated normal BM cells was needed to insure satisfactory hematopoietic reconstitution in at least 50% of lethally irradiated LBN rats. After ex vivo incubation with graded concentrations of 4-hydroperoxycyclophosphamide (4HC) or bleomycin (BLEO), LBN AML or normal BM cells were cultured for CFU-Leuk or CFU-GM and injected into untreated or lethally irradiated syngeneic recipients. Over a variety of drug concentrations (4HC, 3-30 micrograms/ml; BLEO, 100-10,000 mU/ml) and cell doses (10(6)-10(7)/animal) examined, the log-kill estimates derived from in vitro CFU-Leuk assays correlated with the observed MST or LFS. Recovery of greater than 1% CFU-GM from 4HC- or BLEO-treated suspensions of normal BM was associated with satisfactory engraftment in lethally irradiated LBN rats. Clonogenic assays also predicted for engraftment and LFS in animals that received mixtures of AML and normal BM cells (1:10) treated with 4HC and/or BLEO. We conclude that CFU-Leuk and CFU-GM assays are useful screening techniques to develop and evaluate strategies for ex vivo purging with chemotherapeutic agents in this preclinical model of autologous marrow transplantation for AML.

Hematopoietic cell transplantation after administration of high-dose busulfan in murine globoid cell leukodystrophy (the twitcher mouse)

We evaluated the effectiveness of administration of high-dose busulfan (BU), an alkylating agent with substantial myeloablative but negligible immunosuppressive properties, on the engraftment of congenic normal bone marrow and spleen cells (hematopoietic cell transplantation; HCT) in murine galactosylceramidase deficiency (the twitcher mouse), a model of human sphingolipid storage disease. Untreated mice died with extensive demyelination and failure to thrive at a median age of 40 d (range, 28-47; n = 51). The life-span of twitcher mice given HCT at age 10 d after 100 mg/kg BU was significantly prolonged (median, 94 d; range, 55-140; n = 17); these animals did not develop the hindlimb paralysis seen in untreated twitchers. Histologic examination of twitcher sciatic nerves after HCT showed remyelinated areas and decreased globoid cell inclusions. Animals given HCT after conditioning with BU had both lymphoid and hematopoietic repopulation with donor cells by 90 d after HCT, as documented by presence of donor glucose phosphate isomerase-1A activity in blood, bone marrow, spleen, and lymph nodes. After BU and HCT, galactosylceramidase activity in livers and spleens of twitcher mice reached 45 and 80% of control, respectively; only modest elevations were observed in kidneys and lymph nodes. Hydrolase activity rose to 20% in the brains and exceeded control values in the sciatic nerves of transplanted twitcher mice, indicating entry of at least some donor-derived cells and/or hydrolase across the blood-brain and blood-nerve barriers after BU and HCT.(ABSTRACT TRUNCATED AT 250 WORDS)

Density-gradient separation of autologous bone marrow grafts before ex vivo purging with 4-hydroperoxycyclophosphamide

We found in previous studies that the cytotoxicity of 4-hydroperoxycyclophosphamide (4-HC) toward normal and malignant cells in autologous bone marrow grafts was modulated by the variable erythrocyte content of the incubation mixture. This affected both the kinetics of engraftment and the probability of leukemic relapse. To decrease this variability in 4-HC cytotoxicity, we initiated a trial of density-gradient separation of the autografts to remove erythrocytes and other mature blood cells before the 4-HC treatment. A series of 64 consecutive patients was studied. Light-density (sp. gr. much less than 1.077 g/ml) cells were isolated in a 2-step procedure by Ficoll-diatrizoate separation of previously isolated buffy-coat cells. Overall cell recovery after buffy-coat and gradient separations was 28.9 +/- 11.7% ( +/- SD). These cells were then treated with 60 micrograms/ml of 4-HC. The erythrocyte content of the incubation mixtures averaged 0.8%, and did not predict percent CFU-GM survival after the 4-HC incubation. The variability in CFU-GM survival after the purge was less than for previous patients who received buffy-coat cells (with erythrocyte contents ranging from 4-22%) treated with 100 micrograms/ml of 4-HC, indicating that a more uniform purge was achieved. Although the selected dose of 4-HC (60 micrograms/ml) yielded a lower CFU-GM survival, the kinetics of engraftment did not significantly differ between the recipients of light-density or buffy-coat cells. No patient required use of a 'backup' graft because of engraftment failure.

Saposins (sphingolipid activator proteins) in the twitcher mutant mouse

The twitcher mutant mouse, the animal model of Krabbe disease (human globoid cell leukodystrophy), is characterized by apparent deficiency of galactosylceramide beta-galactosidase activity. Saposin A and C, the heat-stable small sphingolipid activator glycoproteins, stimulate the activity of galactosylceramide beta-galactosidase as well as glucosylceramide beta-glucoside. The role of these saposins in the twitcher mutation was investigated. Boiled supernatant fractions, which contained saposins, were prepared from homogenates of twitcher brain, liver, kidney, and spleen. These preparations showed an almost identical effect on the activity of purified glucosylceramide beta-glucosidase (measured by hydrolysis of 4-methylumbelliferyl-beta-glucoside) with similar preparations from control tissues. The effect on the activity of galactosylceramide beta-galactosidase as well as 4-methylumbelliferyl-beta-glucoside beta-glucosidase in the twitcher brain and liver homogenates by authentic saposin A and C was similar to that in control tissues. These results suggest that the twitcher mutation does not affect the concentrations of saposin A or C or their interaction with galactosylceramide beta-galactosidase.

Allogeneic bone marrow transplantation for patients with high-risk acute lymphoblastic leukemia

Seventy-four consecutive patients with high-risk acute lymphoblastic leukemia (ALL) were given cyclophosphamide (CY; 50 mg/kg on each of 4 days) plus total body irradiation (TBI; 300 rad on each of 4 days) followed by a human leukocyte antigen (HLA)-identical allogeneic bone marrow transplant (BMT). Eighteen patients in first complete remission (CR1), 36 in CR2, 16 in CR3, and four in CR4 were transplanted. Patients in CR1 were transplanted 1 to 8 months (median, 3 months) after attaining CR. All 18 patients in CR1 had one or more poor risk factors: age more than 18 (N = 17), initial leukocyte count greater than or equal to 20,000 (N = 11), Ph 1 chromosome (N = 2), delay in attaining CR more than 6 weeks (N = 8), or extramedullary disease (N = 1). Of those transplanted in CR2, 72% had relapsed on therapy. The 5-year event-free survival (EFS) rates for patients transplanted in CR1, CR2, and CR3 are 42%, 43%, and 25%, respectively, at median follow-up times of 57, 54, and 72 months, respectively. Children aged less than 18 years transplanted in CR2 have a 5-year EFS rate of 54%. All CR4 patients died early after transplant. The actuarial probability of relapse is 20%, 26%, and 48% for those transplanted in CR1, CR2, and CR3, respectively. Although there was substantial transplant-associated mortality, it decreased over the decade of the study (P = .01). This study indicates that BMT offers an attractive alternative to postremission chemotherapy in patients in CR1 with poor prognostic factors and in patients in second remission.

Bone marrow transplantation in lysosomal storage diseases: a prelude to gene-insertion therapy

Allogeneic bone marrow transplantation is a therapeutic option for enzyme replacement in children with lysosomal storage diseases, for which no current conventional treatments are curative. Although marrow allografting is costly, intensive, and not risk free, strides are being made in the management and prevention of procedure-related complications.

Ex vivo chemopurging of autologous bone marrow with 4-hydroperoxycyclophosphamide to eliminate occult leukemic cells. Laboratory and clinical observations

The application of autologous bone marrow transplantation (ABMT) in treating acute leukemias in children has been limited by the presence of residual occult viable leukemic cells in the marrow cell suspension. One approach to this problem is the ex vivo treatment ("purging") of the autograft to eradicate these tumor cells yet spare the normal lymphohematopoietic stem cells. Initial studies of acute myeloid leukemia (AML) in a rodent model demonstrated that incubation with 4-hydroperoxycyclophosphamide (4HC), a congener of cyclophosphamide and an active alkylating agent in aqueous solution, could effectively eliminate viable AML cells from marrow cell suspensions without apparent toxicity to normal stem cells. We have conducted clinical trials of ABMT with 4HC-treated marrow in children with acute leukemia in remission; marrow was collected, treated ex vivo with 4HC (100 micrograms/ml), and cryopreserved in liquid nitrogen until reinfusion. Children received pre-ABMT conditioning with either high-dose cyclophosphamide and total body irradiation (CY-TBI) for acute lymphocytic leukemia (ALL) or high-dose busulfan and cyclophosphamide (BU-CY) for AML. Of nine children who underwent ABMT with 4HC-treated marrow for ALL in second complete remission (CR2), all relapsed (eight in the marrow, one in the central nervous system) at a median of 5 months (range, 2-17) after ABMT and all have died with relapsed ALL or as a consequence of its treatment. Twenty-nine children with AML (five in CR1, 24 in CR2) received autografts with chemopurged marrow at a median remission duration of 3 months (range, 2-15). Three patients died from sepsis during aplasia; 10 children (one in CR1 and nine in CR2) relapsed with AML at a median of 7 months (range, 2-23) after ABMT, for an actuarial relapse rate of 47%. Sixteen patients with AML (four in CR1, 12 in CR2) are in unmaintained remission at a median of 16 months (range, 6-102) after ABMT, for an actuarial disease-free survival of 49%. Although ABMT with 4HC-treated marrow appears to have a limited role in the treatment of children with ALL who lack a suitable related donor, the results in AML are encouraging and compare favorably with both syngeneic and allogeneic BMT in similar groups of patients.

A subline of the Brown Norway myeloid leukemia in the Lewis x Brown Norway rat: in vivo growth characteristics and development of an in vitro clonogenic assay

A subline of Brown Norway (BN) acute myelocytic leukemia (AML) which can be propagated in suspension culture (designated IPC-81) is described. Injection into Lewis x BN F1 hybrid (LBN) rats resulted in a log-linear correlation between tumor cell dose and time till death from the onset of leukemia even after multiple (greater than 16) passages in vitro. An in vitro clonogenic assay for IPC-81 colony formation (CFU-leuk) was developed with excellent cloning efficiency (55-82%). Colonies grew without the addition of specific growth factors; syngeneic spleen-conditioned medium inhibited CFU-leuk by 40%, but co-culture with untreated normal LBN rat bone marrow cells had no effect on CFU-leuk. CFU-leuk could be detected in the bone marrow 7 to 10 days before morphologic detection of leukemia in injected animals. This cell line should prove useful in the preclinical evaluation of new strategies for treating AML and evaluating new bone marrow purging methods.

Efficacy of ex vivo purging for autologous bone marrow transplantation in the treatment of acute nonlymphoblastic leukemia

We used an in vitro measure of drug activity to predict the efficacy of ex vivo purging of leukemic cells from autologous bone marrow grafts. We previously found that the myeloid progenitor cell (CFU-GM) content of the marrow grafts after ex vivo purging with 4-hydroperoxycyclophosphamide (4-HC) correlates with time to hematologic recovery after autologous bone marrow transplantation in patients with acute nonlymphoblastic leukemia. We observed that variable red blood cell concentration of the bone marrow incubation mixture results in differential cytotoxic activity of 4-HC. The CFU-GM content of the graft after the ex vivo treatment is a measure of this 4-HC activity. We analyzed the disease-free survival of 45 patients with acute nonlymphoblastic leukemia undergoing autologous bone marrow transplantation with 4-HC purged grafts. Patients who relapsed after transplantation had 4.2 +/- 1.1% of graft CFU-GM surviving the ex vivo purge, compared with 1.1 +/- 0.4% for patients who achieved a sustained remission (P = .06). Twenty-three patients with a CFU-GM content after 4-HC purging of greater than 1% of the pretreatment value had an actuarial disease-free survival of 12%, compared to 36% for 22 patients with a less than or equal to 1% CFU-GM content after purging (P = .006). Therefore, percent CFU-GM survival as a measure of 4-HC cytotoxicity identified a group of patients with insufficient purging. Although no randomized clinical trials have documented the need for ex vivo purging, our results suggest that effective bone marrow purging is important for the optimal application of autologous transplantation in the treatment of acute nonlymphoblastic leukemia.