Acute Leukaemia

Pediatric marrow transplantation for acute leukemia using unrelated donors and T-replete or -depleted grafts: a case-matched analysis


A retrospective, case-matched analysis of the short-term toxicity, risk of GVHD and relapse as well as outcome in pediatric unrelated marrow transplantation was conducted by comparing recipients of T-replete and -depleted grafts in a two-center setting. Both groups contained 30 patients with acute leukemia matched by age at transplant, gender, primary diagnosis and disease status. Acute (90% vs 53%) and chronic (48% vs 0%) GVHD were more common among recipients of T-replete grafts. No significant differences in graft rejection/failure or viral infections were encountered between the two groups. Relapses were more prevalent (37% vs 15%) among recipients of T-depleted grafts. Outcome (EFS) was similar in the two groups. Consequently, in the analysis of transplant outcome, the higher risk of procedure-related, toxic complications among pediatric recipients of T-replete marrow grafts appears to be balanced by an increased risk of relapse among the recipients of T-depleted grafts. Bone Marrow Transplantation (2000) 25, 395–399.


There has been a considerable increase in the use of volunteer unrelated donors for bone marrow transplantation during the past few years and the problems associated with the use of unrelated donors have been well described.12345678

One of the main complications associated with this approach has been the substantial increase in the incidence of clinically significant GVHD compared with transplants performed using a matched, sibling donor.1 This has led to the use of T cell-depleted grafts resulting in a decrease in the incidence of acute GVHD. However, with effective depletion of T cells the risk of non-engraftment/graft failure and virus-associated problems as well as relapse of the malignant disease post transplant is increased.45678

Data comparing the use of sibling donors and alternative donors have been published by several groups,179 but case-matched comparisons between recipients of T cell-replete and T cell-depleted grafts have not been reported.

The aim of our study was to retrospectively assess short-term (ie prior to day 100 post BMT), procedure-related toxicity associated with pediatric marrow transplantation from unrelated donors among recipients of T-replete and -depleted grafts in a case-matched, two-center setting. We also assessed the incidence of chronic GVHD, the risk of leukemia relapse as well as mortality in the two groups.

Patients and methods

The patients receiving a T-replete graft (n = 30) were transplanted consecutively in the Division of Hematology-Oncology and Stem Cell Transplantation, Hospital for Children and Adolescents, University of Helsinki between March 1993 and October 1998 (T-replete group). Those with a T-depleted graft (n = 30) were transplanted in the Bone Marrow Transplantation Unit, Bristol Royal Hospital for Sick Children between March 1993 and December 1997 (T-depleted group).

Patients in the two groups were matched case by case with age at transplant, sex, primary diagnosis and disease status. Only recipients of unrelated marrow grafts from donors fully matched for HLA-A, -B and -DR1 were included in the study.

Demographics and diagnoses

There were 15 boys and 15 girls in both groups and the mean age (range) at transplant was 7.3 (0.4–17) and 7.4 (0.2–16) years in the T-replete and -depleted groups, respectively. In both groups there were seven patients with ALL in 1CR, 13 with ALL in 2CR, four with ALL in 3CR and six with AML in 2CR. Among the patients in 2CR, six had relapsed while on therapy in both groups and the rest of the patients within a mean of 9 months (range 4–14 months) and 10 months (range 4–30 months) following cessation of therapy in T-replete and -depleted groups, respectively.

Tissue typing and donors

In both groups of patients HLA compatibility was assessed by conventional serological typing for class I and high- resolution DNA typing using sequence-specific oligo- nucleotide probes for class II antigens.

In 20/30 cases in the T-replete group and 21/30 cases in the T-depleted group the donor was not matched for gender with the recipient. In 9/30 and 16/30 cases a male recipient was transplanted from a female donor, respectively.

CMV status

In the T-replete group 26/30 recipients and in the T-depleted group 16/30 were at risk of CMV reactivation.


In the T-replete group 6/30 were conditioned with cyclophosphamide (60 mg/kg/day for 2 days). One patient received busulphan 4 mg/kg/day for 4 days in conjunction with ifosfamide and etoposide. Fifteen of 30 received high-dose ARA-C (3 g/m2/dose every 12 h, a total of 12 doses, total dose 36 g/m2) and 8/30 were conditioned with melphalan (140 mg/m2 on day 1 and 70 mg/m2 on day 2).

In the T-depleted group all patients were conditioned with cyclophosphamide (60 mg/kg/day for 2 days). They all also received treatment with Campath as previously described.8

All patients in the T-replete group received fractionated TBI (total dose 10–12 Gy divided into 5–6 fractions over 3 days). In the T-depleted group 26/30 patients received fractionated TBI (total dose 14.4 Gy in 8 fractions over 4 days) and four patients received their TBI in a single fraction.

T cell depletion and cell dose

T cell depletion was performed using CAMPATH antibodies as decribed previously.8

The mean nucleated cell dose infused/kg was 5.3 (range 2.0–8.8) × 108 in the T-replete group and 4.5 (1.0–15.0) × 108 in the T-depleted group of patients.


Patients exceeding an ANC level of 500 × 106/l were considered engrafted. Non-engraftment/graft failure was defined as the absence of donor-derived hematopoiesis on day +28 post transplant or later, or as autologous marrow reconstitution.

Diagnosis and treatment of CMV

In both groups the presence of CMV viremia was monitored regularly through testing for CMV antigenemia (both groups) and PCR (T-depleted group). A pre-emptive approach in the treatment of CMV reactivation was employed in both groups.

Prophylactic and supportive care

In the T-replete group the patients were nursed in two-door isolation from the start of conditioning up until an ANC of >300 × 106/l or longer. All except those with AML (6/30) received G-CSF from day +1. No anti-viral or -bacterial prophylaxis was employed but post-engraftment co-trimoxazole/pentamidine was used as PCP prophylaxis. Systemic antifungal prophylaxis with fluconazole or itraconazole was employed. All blood products were filtered and irradiated.

In the T-depleted group all patients were nursed in two-door isolation under positive pressure with filtered air from the day of transplant until an ANC of 1000 × 106/l or longer. All patients received G-CSF from day +10. Prophylactic oral ciprofloxacin was administered from day −3 and acyclovir (oral/i.v.) from day −4. Oral itraconazole from day −8 onwards was employed as antifungal prophylaxis. Co-trimoxazole/pentamidine from day +28 onwards was also employed as PCP prophylaxis. The recipients at risk of CMV reactivation received i.v. immunoglobulin with a dose of 0.2 g/kg on days −1, +13, +34, +55, +74 and +95 as well as high-dose acyclovir (500 mg/m2) daily from day +1 to day +31. All blood products were given irradiated and filtered.

A clean diet was instituted pre-transplant in both groups and continued throughout the immediate post-transplant period. Similar principles of mouthcare were employed in both groups.

Broad-spectrum antibacterial therapy was instituted and blood cultures taken following similar guidelines when fever and/or other clinical signs suggestive of a bacterial infection developed. Again in both groups, intra-venous antifungals were included in the regimen in the case of suspected or proven fungal infection with similar principles. Apart from CMV no attempts were made to pre-emptively diagnose or treat any other viruses in the two groups.


In both groups cyclosporin A was administered to all patients from day −1 with a starting dose of 1.5 mg/kg twice daily. The T-replete group of patients also received i.v. methotrexate on days +1 (15 mg/m2), +3 (10 mg/m2), +6 (10 mg/m2) and +11 (10 mg/m2).

Clinical GVHD was diagnosed following published guidelines10 and prednisolone/methylprednisolone used as a first-line treatment in both groups.


In the statistical evaluation of the results the chi-square-test was employed. Probability of event-free survival was assessed with the Kaplan–Meier method and log-rank test.


The results of engraftment, non-engraftment/graft failure, viral infections and relapses as well as outcome are given in Table 1 and those on GVHD in Table 2.

Table 1  Engraftment, viral infections and outcome in the T-replete and T-depleted groups of URD-transplant recipients
Table 2  GVHD in the T-replete and T-depleted groups of URD graft recipients

In both groups all patients were febrile at some point during their post-transplant course. In 9/30 cases in the T-replete group and 17/30 in the T-depleted group the bacterial infection was verified and the causative organism identified. No blood cultures positive for fungi were encountered in either of the two groups. In only one case in the T-replete group was the suspicion of a fungal infection verified (Fusarium) through post-mortem sampling.

The key differences between the groups emerged in the incidence and severity of GVHD (Table 2) and toxic mortality (Table 1) as well as relapses of leukemia with resulting mortality (Table 1 and Figure 1).

Figure 1

The cumulative number of relapses in the T cell-replete and -depleted groups of patients.

Deaths in the T-replete group prior to day +100 were all treatment-related and occurred in conjunction with non-engraftment/graft failure with severe acute GVHD being one of the causative factors in all four. One of the four died of a disseminated Fusarium infection and another of disseminated adenoviral infection. Two also showed signs of HUS/TTP. Deaths in this group beyond day +100 were due to relapse (four cases) and chronic GVHD (two cases).

In the T-depleted group six patients experienced non-engraftment/graft failure and only one of them was retransplanted. All of the six subsequently relapsed and none of them is a long-term survivor.

In the T-depleted group there were no early treatment-related deaths and 11 out of the 14 deaths were secondary to leukemia relapse.

None of the patients in either of the two groups had documented CMV disease or severe VOD.

In the T-replete group 10 out of 25 (40%) evaluable patients have thus far (mean follow-up 28 months, range 8–75 months) developed signs of limited and two (7%) those of extensive chronic GVHD. In the T-depleted group none have developed signs of chronic GVHD with a mean follow-up of 37 (range 2–67) months.

In the T-replete group four out of the 26 evaluable patients (15%) have relapsed (mean follow-up 23 months, range 2–75 months) while in the T-depleted group 11 out of the 30 patients (37%) have thus far relapsed (mean follow-up 31 months, range 2–67 months) (Figure 1).

Of the patients who relapsed post BMT in the T-replete group two out of the four had been transplanted in 2CR and both had had their primary relapses off-therapy. In the T-depleted group four out of the six had had their primary relapses while on therapy. The other two patients in the T-replete group relapsing post BMT were transplanted in 1CR and 3CR while in the T-depleted group the five were transplanted in 1CR (n = 3) and 3CR (n = 2).

In our two groups of patients their cumulative mortality currently is 33% in the T-replete group (mean follow-up 24 months, range 2–75 months) and 37% in the T-depleted group (mean follow-up 32 months, range 2–67 months) (Figure 2).

Figure 2

Event-free survival in the T cell-replete and -depleted groups of patients.


In this first published, case-matched analysis of pediatric recipients of T cell-replete and T cell-depleted bone marrow grafts from unrelated donors the higher overall incidence of acute GVHD is evident in the T-replete group. Particularly striking is the difference in the involvement of the gut and/or the liver in the acute GVHD. The incidence of grade III–IV acute GVHD in our T-replete group (27%) is in agreement with that (27–37%) previously published.36 The higher incidence of clinically significant acute GVHD in our T-replete group of patients is also reflected in their short-term outcome with a treatment-related mortality of 13% prior to day +100, which, however, compares favorably with that (24–37%) published by other groups.36

The putative effect of differences between the two groups in regimens used for conditioning as well as in supportive care on procedure-related mortality and morbidity remains unsettled in our analysis. The use of melphalan with TBI in particular appears to be associated with an increased risk of toxic complications.11 Furthermore, the use of only serology in HLA typing for class I antigens runs the risk of one of the two groups being less polymorphic than the other. The possible effect of this cannot be evaluated in our analysis.

Concomitant with the higher risk of clinically significant acute GVHD the incidence of limited and extensive chronic GVHD is increased in our group of recipients of T-replete marrow grafts when compared to those of T-depleted grafts.

With all our donor–recipient pairs having a 6/6 antigen HLA-match we did not observe any difference in non-engraftment/graft failure between the recipients of T-replete and -depleted grafts. This is also in agreement with the experience of other centers.34612

Interestingly and despite a somewhat higher number of CMV-positive grafts being given to CMV-negative recipients in the T-deplete group, re-activation appeared more prevalent among the T-replete patients. This could, however, be explained by the omission of acyclovir prophylaxis in this group as well as the higher number of grafts from CMV-negative donors being infused into CMV-positive recipients.

A lower number of functional T cells present in the recipient immediately post transplant may explain the seemingly increased prevalence of non-CMV viral problems encountered in the T-depleted group.

As to the overall outcome in our two groups of patients, among those with T-depleted grafts their long-term outcome is offset by a somewhat higher risk of relapse of their malignant disease. This observation is in agreement with previous reports.678 The relapse rate among the T-depleted patients of our series is well in agreement with that (35.2%) observed among children (n = 80) transplanted in the Bristol Unit using HLA-matched unrelated marrow grafts and recently reported.13

We conclude that our two-center, case-matched comparison of pediatric recipients of T-replete vs those of T-depleted bone marrow grafts clearly demonstrates the effects of T cell depletion on the procedure-related, immediate, post-transplant (prior to day +100) toxicity. At least in the HLA-matched unrelated setting the risk of non-engraftment/graft failure appears not to be higher among recipients of T cell-depleted grafts but the possibility of increased morbidity associated with viruses other than CMV in this group of patients merits further evaluation. The eventual outcome among recipients of T-replete marrow grafts and those of T-depleted grafts is balanced by a higher risk of relapse in the T-depleted group. In the future, the toxic mortality and morbidity in the T-replete group may be decreased through stringent re-evaluation of the conditioning regimens used (ie eliminating those associated with the highest risk of acute GVHD). In the T-depleted group the risk of leukemia relapse may be decreased through meticulous assessment of hematopoietic chimerism, the use of T cell add-backs and less intensive immunosuppression.


  1. 1

    Vettenranta K, Hovi L, Saarinen-Pihkala UM . Short-term toxicity in pediatric marrow transplantation using related and unrelated donors Bone Marrow Transplant 1999 23: 459–462

  2. 2

    Busca A, Anasetti C, Anderson G et al. Unrelated donor or autologous marrow transplantation for treatment of acute leukemia Blood 1994 83: 3077–3084

  3. 3

    Balduzzi A, Gooley T, Anasetti C et al. Unrelated donor marrow transplantation in children Blood 1995 86: 3247–3256

  4. 4

    Casper J, Camitta B, Truitt R et al. Unrelated bone marrow donor transplants for children with leukemia or myelodysplasia Blood 1995 85: 2354–2363

  5. 5

    Chown SR, Marks DI, Cornish JM et al. Unrelated donor bone marrow transplantation in children and young adults with acute myeloid leukaemia in remission Br J Haematol 1997 99: 36–40

  6. 6

    Davies SM, Wagner JE, Shu X-O et al. Unrelated donor bone marrow transplantation for children with acute leukemia J Clin Oncol 1997 15: 557–565

  7. 7

    Hongeng S, Krance RA, Bowman LC et al. Outcomes of transplantation with matched-sibling and unrelated-donor bone marrow in children with leukaemia Lancet 1997 350: 767–771

  8. 8

    Oakhill A, Pamphilon DH, Potter MN et al. Unrelated donor bone marrow transplantation for children with relapsed acute lymphoblastic leukaemia in second complete remission Br J Haematol 1996 94: 574–578

  9. 9

    Winiarski J, Ringden O, Remberger M et al. Bone marrow transplantation in children using unrelated donors at Huddinge Hospital Acta Paediatr 1996 85: 327–335

  10. 10

    Przepiorka D, Weisdorf D, Martin P et al. Consensus conference on acute GVHD Bone Marrow Transplant 1995 15: 825–828

  11. 11

    Vettenranta K, Hovi L, Taskinen M, Saarinen-Pihkala UM . Allograft with unrelated donor accentuates the gastrointestinal toxicity associated with high-dose melphalan and TBI preparative for BMT in children. (submitted for publication)

  12. 12

    Margolis D, Camitta B, Pietryga D et al. Unrelated donor bone marrow transplantation to treat severe aplastic anemia in children and young adults Br J Haematol 1996 94: 65–72

  13. 13

    Green A, Clarke E, Hunt L et al. Children with acute lymphoblastic leukemia who receive T cell depleted HLA mismatched marrow allo-grafts from unrelated donors have an increased incidence of primary graft failure but a similar overall transplant outcome Blood 1999 94: 2236–2246

Download references


The financial support of the Nona and Kullervo Väre Foundation, Finland, is gratefully acknowledged.

Author information

Correspondence to K Vettenranta.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Vettenranta, K., Saarinen-Pihkala, U., Cornish, J. et al. Pediatric marrow transplantation for acute leukemia using unrelated donors and T-replete or -depleted grafts: a case-matched analysis. Bone Marrow Transplant 25, 395–399 (2000).

Download citation


  • pediatric
  • bone marrow
  • transplantation
  • unrelated donors

Further reading