We report results in 41 consecutive patients with chronic lymphocytic leukemia (CLL) who underwent allogeneic hematopoietic cell transplantation (HCT) after fludarabine, melphalan, and alemtuzumab conditioning. Donors were 24 HLA-matched siblings and 17 unrelated volunteers, 4 of them mismatched with recipients. All but 3 patients had initial hematologic recovery, but 5 more patients had secondary graft failure. Median intervals to neutrophil (greater than 0.5 × 109/L) and platelet (greater than 20 × 109/L) recovery were 14 days (range, 9-30 days) and 11 days (range, 8-45 days), respectively. Eleven (27%) patients had relapses and received escalated donor lymphocyte infusions, but only 3 of them had sustained responses. Acute and chronic graft-versus-host disease (GVHD) was observed in 17 (41%) and 13 (33%) patients, respectively. Seventeen (41%) patients have died, 5 of progressive disease. The 2-year overall survival and transplantation-related mortality (TRM) rates were 51% (95% confidence interval [CI], 33%-69%) and 26% (95% CI, 14%-46%), respectively. The alemtuzumabbased regimen was feasible and effective in patients with CLL with a relatively low rate of GVHD. However, TRM remains relatively high as a result of a variety of viral and fungal infections. Studies are ongoing to address the efficacy of reduced doses of alemtuzumab in this group of immunosuppressed patients.

Introduction

The course of chronic lymphocytic leukemia (CLL) is extremely variable. Although some patients have indolent disease without any need for therapeutic interventions, others succumb rapidly despite intensive treatment.1  Approximately one third of CLL patients are younger than 60 years, and 10% to 15% are younger than 50 years. In young adults, CLL has no major distinctive features, the prognostic factors are the same as those in older patients, and patients with advanced disease have a median survival of 24 to 72 months with conventional chemotherapy.2-4  In particular, CLL patients who have failed to respond to fludarabine or any other purine analog have the poorest prognosis, with a median survival of approximately 8 months and a 1-year survival rate of approximately 40%.5  Autologous hematopoietic cell transplantation (HCT) is feasible in young patients with CLL and results in long-lasting clinical and molecular remissions.6  The transplantation-related mortality (TRM) rate is less than 10%, with complete response rates of approximately 80%; a risk-matched case-control study revealed a prolongation of survival compared with conventional therapy.7  However, these results have never been confirmed in prospective randomized studies, and it is becoming evident that most patients attaining complete remissions after autograft will eventually experience relapse.8 

Prolonged remissions have been achieved with conventional allogeneic HCT.9  Unfortunately, this procedure is only suitable in a small proportion of young patients with a compatible donor and is hampered by a TRM rate of approximately 50%.10  After early attrition because of TRM, the survival curves approach a plateau, and the achievement of a molecular remission is more likely than after autologous transplantation, probably reflecting a graft-versusleukemia effect.11  To reduce the TRM rate, several reduced-intensity conditioning regimens have been developed. The UK Collaborative Group protocol incorporates anti-CD52 monoclonal antibody (alemtuzumab; Campath-1H) to deplete recipient and incoming donor T cells, providing durable engraftment while significantly reducing the risk for graft-versus-host disease (GVHD) in HLA-matched siblings and unrelated donors.12,13  Furthermore, CD52 is also expressed at high density in CLL cells. Indeed, alemtuzumab has proven activity in patients with advanced CLL, even if refractory to fludarabine.14,15  Unfortunately, the use of alemtuzumab also delays post-HCT immune reconstitution, increasing the risk for infective complications and potentially impairing the graft-versus-leukemia effect. This reduced antitumor activity may require the early administration of sequential donor lymphocyte infusion (DLI) after transplantation.16-18 

Based on the encouraging results obtained with an alemtuzumabbased reduced intensity regimen in low-grade non-Hodgkin19,20  and Hodgkin lymphoma,21  the aim of this study was to evaluate the feasibility and efficacy of the same regimen in patients with advanced CLL.

Patients and methods

Study patients

The study protocol was approved by the ethics committee of each participant center, and all patients and donors gave written informed consent. We prospectively collected data from 41 consecutive patients with advanced CLL who underwent allogeneic HCT using a fludarabine-melphalan-alemtuzumab conditioning regimen. All patients fulfilled standard morphologic and immunophenotypic criteria for CLL.22  Fluorescence in situ hybridization (FISH) cytogenetic studies and analysis of the mutation status of the immunoglobulin gene were only available in a minority of patients and were not analyzed in this study. Patients were excluded if they were older than 70 years, the left ventricular ejection fraction was less than 40%, creatinine clearance was less than 30 mL/min per 1.73 m2, serum bilirubin level was greater than 34 μM, or liver transaminases were more than 3 times the upper limit. Two patients who experienced high-grade (Richter) transformation before transplantation were included in the study.

Conditioning regimen, GVHD prophylaxis, and stem cell source

The conditioning regimen consisted of alemtuzumab 20 mg/d from days -8 to -4 (total dose, 100 mg), fludarabine 30 mg/m2 daily from days -7to -3 (total dose, 150 mg/m2), and melphalan 140 mg/m2 on day -2. Fourteen (34%) patients each received a total dose of 60 mg (6 patients), 50 mg (7 patients), and 40 mg (1 patient) alemtuzumab on days -2 and -1 as part of an ongoing deescalation trial. Cyclosporine was administered as an intravenous infusion over 1 hour at 3 mg/kg daily starting on day -1, with a target level of 150 to 200 ng/mL. In the absence of GVHD, cyclosporine was tapered from 3 months after transplantation. Acute and chronic GVHD were graded according to standard criteria.23  On day 0, patients received peripheral blood stem cells (n = 35), bone marrow (n = 5), or both (n = 1) from HLA-matched siblings (n = 24) or unrelated donors (n = 17). Patients and their donors were matched for HLA-A, -B, -C, -DRB1, and -DQB1 by intermediate- or high-resolution DNA-typing techniques, as appropriate. Four unrelated donor/recipient pairs had single-allele mismatches.

Analysis of donor chimerism and donor lymphocyte infusions

Whole blood and lineage-specific chimerism (in T-cell and myeloid lineages) were assessed by means of polymerase chain reaction (PCR) analysis of informative minisatellite regions (short tandem repeat loci) or FISH for X and Y chromosomes in the event of sex-mismatched transplantations. Mixed chimerism was defined as the presence of more than 5% and less than 95% CD3+ lymphocytes from the donor. Patients who had mixed chimerism or residual disease 6 months after transplantation were eligible to receive DLI if there was no evidence of active GVHD. Escalating doses of CD3+ lymphocytes were administered starting at a dose of 1 × 106 T cells/kg. Increasing doses were administered at 3-month intervals (3 × 106, 1 × 107,3 × 107, and 1 × 108 T cells/kg) in the absence of GVHD if mixed chimerism persisted or if there was no evidence of disease response. The doses administered for disease progression depended on the disease type and the interval since transplantation.

Supportive care

All patients received standard nursing and supportive care protocols. Infection prophylaxis varied across all participating centers but always included antifungals (fluconazole or itraconazole), acyclovir and cotrimoxazole or its alternatives (dapsone or pentamidine). Patients with persistent fever during neutropenia were treated with broad-spectrum antibiotics. Additional agents (glycopeptides, aminoglycosides) were added as clinically indicated. Amphotericin B was given to patients with unexplained fever that persisted beyond 96 hours. Patients with persistent fever were also screened for other atypical infections (eg, respiratory syncytial virus [RSV], influenza, parainfluenza, adenovirus, Epstein-Barr virus [EBV], Toxoplasma gondii) as clinically indicated. Patients at risk for cytomegalovirus (CMV) reactivation (CMV-seropositive recipients or those who received grafts from CMV-seropositive donors) were monitored weekly by quantitative PCR. When the CMV PCR was positive, the test was repeated and preemptive treatment with ganciclovir was started if the result was confirmed. Blood products were universally depleted of white blood cells and irradiated. CMV-seronegative recipients received CMV-seronegative blood products.

Definitions and statistical analysis

Disease response was assessed using National Cancer Institute Working Group (NCI-WG) criteria.24  Briefly, complete remission (CR) was defined as absence of lymphadenopathy, hepatosplenomegaly, and constitutional symptoms, less than 30% lymphocytes in the bone marrow, and a normal full blood count. Partial remission (PR) was defined as 50% reduction in peripheral blood lymphocyte count, lymphadenopathy, and hepatosplenomegaly with a normal full blood count (or 50% improvement over baseline). Where possible, patients in CR underwent multiparameter flow cytometry analysis and PCR for immunoglobulin heavy-chain rearrangement performed on bone marrow or peripheral blood samples every 3 to 6 months after transplantation.

Patients who achieved a significant response (CR or PR by NCI-WG criteria24 ) to the last chemotherapy regimen before transplantation were considered to have chemosensitive disease. Conversely, patients failing to respond to the last chemotherapy regimen before transplantation were considered to have chemorefractory disease. Fludarabine refractoriness was defined as failure to achieve PR or CR to the last fludarabine-based regimen administered.

Time-to-event outcomes with competing risks (ie, TRM and relapse rates) were estimated by cumulative-incidence curves. Comparison of cumulative-incidence curves was performed by the Lunn-McNeill approach in which Cox regression is applied to competing risks. Overall survival (OS) and progression-free survival (PFS) were estimated by the Kaplan-Meier method and compared using the log-rank test. Several factors were analyzed for their association with OS, PFS, and TRM, including age (55 and younger vs older than 55), donor type (sibling vs unrelated), status at transplantation (PD vs CR + PR), previous autologous transplantation, previous fludarabine, alemtuzumab, or rituximab therapy, number of previous chemotherapy regimens (less than 3 vs 3 or more), alemtuzumab dose in the conditioning regimen (less than 100 vs 100 mg), and fludarabine refractoriness. We also evaluated the so-called “center effect” by splitting the patients in 2 groups: those treated at centers that contributed 8 or more patients and those treated at centers that contributed less than 8 patients. Subsequently, multivariate analysis according to the Cox proportional hazards regression model was used to explore the independent effect of variables that showed a significant influence on OS or PFS by univariate analysis. In all statistical calculations, P values less than .05 were considered significant.

Results

Patient characteristics

Age at transplantation ranged from 37 to 67 years (median, 54 years). At the time of transplantation, 7 (17%) patients had chemorefractory and 34 (83%) patients had chemosensitive disease (Table 1). Eleven (31%) patients were refractory to fludarabine at the time of transplantation; for 3 (8%) others, it was stopped because of autoimmune hemolytic anemia (AIHA, 2 patients) and immune thrombocytopenic purpura (ITP, 1 patient). Another 2 patients had fludarabine-independent immune cytopenias before HCT (1 AIHA, 1 ITP). Pretransplantation characteristics of patients refractory to fludarabine are shown in Table 2.

Table 1.

Baseline characteristics of patients


Characteristic

Value
Median age, y (range)   54 (37-67)  
Immune cytopenias before HCT, no. of patients (%)   5 (12)  
Median previous chemotherapy regimens, no. (range)  3 (1-6)  
   Fludarabine, no. of patients (%)   36 (88)  
   Corticosteroids, no. of patients (%)   25 (61)  
   Cyclophosphamide, no. of patients (%)   25 (61)  
   Chlorambucil, no. of patients (%)   23 (56)  
   Anthracyclines, no. of patients (%)   21 (52)  
   Alemtuzumab, no. of patients (%)   9 (22)  
   Rituximab, no. of patients (%)   9 (22)  
   Autologous HCT, no. of patients (%)   11 (27)  
Status at transplantation, no. of patients (%)  
   Complete remission   5 (12)  
   Partial remission   29 (70)  
   Progressive disease   7 (17)  
Median time from diagnosis to HCT, mo (range)   54 (10-164)  
Type of donor, no. of patients (%)  
   Matched related   24 (58)  
   Matched unrelated   13 (32)  
   Mismatched unrelated   4 (10)  
Median CD34+ cell count, × 106/kg (range)   5.0 (0.8-16.97)  
Stem cell source, no. of patients (%)  
   Peripheral blood stem cells   35 (85)  
   Bone marrow   5 (12)  
   Both   1 (3)  
CMV serostatus, no. of patients (%)  
   Donor or recipient positive   28 (68)  
   Both negative   13 (32)  
Median time to follow-up, mo (range)
 
15.1 (0.2-62.5)
 

Characteristic

Value
Median age, y (range)   54 (37-67)  
Immune cytopenias before HCT, no. of patients (%)   5 (12)  
Median previous chemotherapy regimens, no. (range)  3 (1-6)  
   Fludarabine, no. of patients (%)   36 (88)  
   Corticosteroids, no. of patients (%)   25 (61)  
   Cyclophosphamide, no. of patients (%)   25 (61)  
   Chlorambucil, no. of patients (%)   23 (56)  
   Anthracyclines, no. of patients (%)   21 (52)  
   Alemtuzumab, no. of patients (%)   9 (22)  
   Rituximab, no. of patients (%)   9 (22)  
   Autologous HCT, no. of patients (%)   11 (27)  
Status at transplantation, no. of patients (%)  
   Complete remission   5 (12)  
   Partial remission   29 (70)  
   Progressive disease   7 (17)  
Median time from diagnosis to HCT, mo (range)   54 (10-164)  
Type of donor, no. of patients (%)  
   Matched related   24 (58)  
   Matched unrelated   13 (32)  
   Mismatched unrelated   4 (10)  
Median CD34+ cell count, × 106/kg (range)   5.0 (0.8-16.97)  
Stem cell source, no. of patients (%)  
   Peripheral blood stem cells   35 (85)  
   Bone marrow   5 (12)  
   Both   1 (3)  
CMV serostatus, no. of patients (%)  
   Donor or recipient positive   28 (68)  
   Both negative   13 (32)  
Median time to follow-up, mo (range)
 
15.1 (0.2-62.5)
 
Table 2.

Pretransplantation characteristics and outcomes of patients refractory to fludarabine therapy


Age, y

Salvage chemotherapy

Donor

Status at HCT

Best response after HCT

Outcome
50   CHOP (Richter transformation)   MR   PD   CR   Relapsed shortly after HCT. Unresponsive to DLI. Died of PD.  
51   Alemtuzumab, CHOP   mMU   PR   CR   Delayed graft rejection. Alive in clinical CR, but with increased proportion of CD5/CD19+ cells.  
56   CHOP, alemtuzumab,   mMU   PR   PR   Delayed graft failure. Died of RSV pneumonia after top-up transplantation.  
47   Alemtuzumab   MR   PR   PR   Lack of platelet engraftment. Died of PTLD after top-up transplantation.  
52   Rituximab   MR   PR   PR   Ganciclovir-induced graft failure. Alive in PR (persistent thrombocytopenia).  
62   Rituximab   MR   PR   CR   Relapse 5 mo after HCT. Alive with PD.  
57   None   MU   PD   CR   Alive in CR.  
51   CHOP, alemtuzumab, rituximab   MU   PD   PD   Never responded to any therapy. Died of PD.  
56   Alemtuzumab   MU   PR   CR   Relapse 3 mo after HCT. Complete response to DLI. Alive in CR.  
59   HDMP, alemtuzumab   mMU   PR   NA   Died of neutropenic sepsis on d 12.  
67
 
Rituximab
 
MR
 
PD
 
CR
 
Died of fungal cerebral abscess 3.5 mo after HCT.
 

Age, y

Salvage chemotherapy

Donor

Status at HCT

Best response after HCT

Outcome
50   CHOP (Richter transformation)   MR   PD   CR   Relapsed shortly after HCT. Unresponsive to DLI. Died of PD.  
51   Alemtuzumab, CHOP   mMU   PR   CR   Delayed graft rejection. Alive in clinical CR, but with increased proportion of CD5/CD19+ cells.  
56   CHOP, alemtuzumab,   mMU   PR   PR   Delayed graft failure. Died of RSV pneumonia after top-up transplantation.  
47   Alemtuzumab   MR   PR   PR   Lack of platelet engraftment. Died of PTLD after top-up transplantation.  
52   Rituximab   MR   PR   PR   Ganciclovir-induced graft failure. Alive in PR (persistent thrombocytopenia).  
62   Rituximab   MR   PR   CR   Relapse 5 mo after HCT. Alive with PD.  
57   None   MU   PD   CR   Alive in CR.  
51   CHOP, alemtuzumab, rituximab   MU   PD   PD   Never responded to any therapy. Died of PD.  
56   Alemtuzumab   MU   PR   CR   Relapse 3 mo after HCT. Complete response to DLI. Alive in CR.  
59   HDMP, alemtuzumab   mMU   PR   NA   Died of neutropenic sepsis on d 12.  
67
 
Rituximab
 
MR
 
PD
 
CR
 
Died of fungal cerebral abscess 3.5 mo after HCT.
 

CHOP indicates cyclophosphamide-doxorubicin-vincristine-prednisone; HDMP, high-dose methyl-prednisolone; MR, matched related; PD, progressive disease; CR, complete remission; DLI, donor lymphocyte infusions; mMU, mismatched unrelated; PR, partial remission; RSV, respiratory syncytial virus; and PTLD, posttransplantation lymphoproliferative disorder; MU, matched unrelated; NA, not assessed.

Eleven (27%) patients received allografts after autologous transplantation failed. These patients were similar in age (median, 55 years; range, 40-66 years) and number of previous chemotherapy regimens (median, 4 regimens; range, 2-5 regimens) as the whole cohort. Three of these 11 (27%) patients had chemorefractory disease at the time of transplantation, and 8 (73%) of them received their graft from an unrelated donor.

Engraftment and donor chimerism

The median number of CD34+ cells infused was 5.0 × 106/kg recipient body weight (range, 0.8-16.97 × 106/kg). Median intervals to neutrophil (more than 0.5 × 109/L) and platelet (more than 20 × 109/L) recovery were 14 days (range, 9-30 days) and 11 days (range, 8-45 days), respectively. All patients had initial hematopoietic recovery, except 2 patients who died 6 and 12 days after HCT, respectively, and a third patient with refractory ITP before HCT who never achieved platelet engraftment and required a second stem cell infusion.

Five (12%) more patients had secondary graft failure at a median of 4 months after HCT, 2 of them after HLA-mismatched unrelated transplantation. All 5 patients received more than 4.0 × 106 CD34+ cells/kg recipient body weight, and 3 (60%) of them achieved full donor chimerism before graft failure occurred. Two patients gradually lost their grafts as assessed by peripheral blood chimerism, and 1 of them eventually had a relapse. He is alive with stage A disease and does not require any treatment. The second patient with autologous reconstitution is in CR by NCI-WG criteria, but with a significant proportion of CD5/CD19-coexpressing cells in the peripheral blood. The third patient had persistent anemia after ABO-mismatched transplantation that responded to a second HCT after ATG. However, he eventually died of respiratory syncytial virus pneumonia and disseminated adenovirus infection. The fourth patient developed CMV hepatitis early after HCT and lost his graft because of ganciclovir-induced myelotoxicity. He subsequently underwent second T-cell-depleted HCT (alemtuzumab in vitro), which was successful, and is currently alive and well. The fifth patient had delayed graft failure 6 months after HCT and died of neutropenic sepsis. There was no significant association between graft failure and number of previous chemotherapy regimens, previous autologous transplantation, or previous fludarabine or alemtuzumab therapy. Interestingly, there was a trend toward a higher incidence of graft failure among fludarabine-refractory patients (36%) compared with fludarabine-sensitive patients (8%) (P = .057, Fisher exact test).

Chimerism data were available for all surviving patients who did not have disease progression immediately after HCT. At 6 months after HCT, 97% of all evaluable patients had evidence of donor cell engraftment, but only 22 (67%) of 33 had 100% donor whole-blood chimerism. Of all patients with mixed chimerism, 9 underwent DLI and 5 (56%) of them subsequently achieved full donor chimerism. The other 4 patients never achieved full donor chimerism, and 3 of them eventually relapses, but they all remain alive with stable disease (Table 3).

Table 3.

Donor lymphocyte infusions


Pt

Donor

Indication

After HCT chemotherapy

CD3+ cell dose

Response

GVHD

Outcome (cause of death)

Follow-up, mo
1   MR   MC   —   1 × 106  Yes   Yes   Alive, CR   16+  
2   MR   MC   —   1 × 106  NA   No   Alive, CR   10+  
3   MR   MC   —   1 × 106  Yes   No   Alive, CR   10+  
4   MR   MC   —   1 × 106  Yes   No   Alive, CR   48+  
5   MR   MC, then CRel   —   1 × 106, 3 × 106, 1 × 107, 3 × 107  No   No   Alive, PD   63+  
6   MR   MC, then CRel   Fludarabine + chlorambucil   1 × 106, 3 × 106, 3 × 107  No   Yes   Alive, PD   29+  
7   MR   MC + MRel   —   1 × 106, 3 × 106, 1 × 107, 3 × 107, 1 × 108  No   No   Alive, PD   61+  
8   MU   MC + CRel   —   1 × 106  Yes   Yes   Died (GVHD)   22  
9   MR   MC + RD   —   1 × 106, 3 × 106, 1 × 107, 3 × 107  Yes   No   Alive, CR   62+  
10   MR   CRel   CHOP   1 × 107, 3 × 107, 1 × 108  No   No   Alive, PD   16+  
11   MR   CRel   CHOP   1 × 107, 3 × 107  Yes   Yes   Alive, CR   40+  
12   MR   CRel   COP   1 × 108  No   Yes   Died (PD)   13  
13   MU   CRel   CMOP, rituximab, alemtuzumab   1 × 106, 1 × 107  No   No   Died (pneumonia)   16  
14   MR   CRel   Alemtuzumab, chlorambucil   1 × 106, 1 × 107  No   No   Died (PD)   13  
15   MU   CRel   —   3 × 106, 1 × 107, 3 × 107  Yes   Yes   Alive, CR   51+  
16   MU   CRel   —   1 × 107  No   No   Died (PD)   21  
17   mMU   RD   —   1 × 106, 1 × 107  Yes   Yes   Died (GVHD)   5  
18
 
MU
 
RD
 

 
1 × 107, 3 × 107
 
Yes
 
Yes
 
Alive, PR
 
21+
 

Pt

Donor

Indication

After HCT chemotherapy

CD3+ cell dose

Response

GVHD

Outcome (cause of death)

Follow-up, mo
1   MR   MC   —   1 × 106  Yes   Yes   Alive, CR   16+  
2   MR   MC   —   1 × 106  NA   No   Alive, CR   10+  
3   MR   MC   —   1 × 106  Yes   No   Alive, CR   10+  
4   MR   MC   —   1 × 106  Yes   No   Alive, CR   48+  
5   MR   MC, then CRel   —   1 × 106, 3 × 106, 1 × 107, 3 × 107  No   No   Alive, PD   63+  
6   MR   MC, then CRel   Fludarabine + chlorambucil   1 × 106, 3 × 106, 3 × 107  No   Yes   Alive, PD   29+  
7   MR   MC + MRel   —   1 × 106, 3 × 106, 1 × 107, 3 × 107, 1 × 108  No   No   Alive, PD   61+  
8   MU   MC + CRel   —   1 × 106  Yes   Yes   Died (GVHD)   22  
9   MR   MC + RD   —   1 × 106, 3 × 106, 1 × 107, 3 × 107  Yes   No   Alive, CR   62+  
10   MR   CRel   CHOP   1 × 107, 3 × 107, 1 × 108  No   No   Alive, PD   16+  
11   MR   CRel   CHOP   1 × 107, 3 × 107  Yes   Yes   Alive, CR   40+  
12   MR   CRel   COP   1 × 108  No   Yes   Died (PD)   13  
13   MU   CRel   CMOP, rituximab, alemtuzumab   1 × 106, 1 × 107  No   No   Died (pneumonia)   16  
14   MR   CRel   Alemtuzumab, chlorambucil   1 × 106, 1 × 107  No   No   Died (PD)   13  
15   MU   CRel   —   3 × 106, 1 × 107, 3 × 107  Yes   Yes   Alive, CR   51+  
16   MU   CRel   —   1 × 107  No   No   Died (PD)   21  
17   mMU   RD   —   1 × 106, 1 × 107  Yes   Yes   Died (GVHD)   5  
18
 
MU
 
RD
 

 
1 × 107, 3 × 107
 
Yes
 
Yes
 
Alive, PR
 
21+
 

Pt indicates patient number; HCT, hematopoietic cell transplantation; MR, matched related; MC, mixed chimerism; CR, complete remission; NA, not available; CRel, clinical relapse; PD, progressive disease; MRel, molecular relapse; MU, matched unrelated; GVHD, graft-versus-host disease; RD, residual disease; CHOP, cyclophosphamide-doxorubicin-vincristine-prednisone; COP, cyclophosphamide-vincristine-prednisone; CMOP, cyclophosphamide-mitoxantrone-vincristine-prednisone; mMU, mismatched unrelated; PR, partial remission.

Graft-versus-host disease

Acute GVHD (aGVHD) was observed in 17 (41%) patients, but only 4 (10%) patients developed grade III or IV aGVHD. Twelve (29%) patients developed aGVHD after transplantation, and 5 (12%) patients developed it after DLI. No statistical difference was observed between unrelated and sibling donor recipients (29% vs 29%, P > .999; Fisher exact test).

Thirty-nine (95%) patients survived until at least day 100 and were assessable for chronic GVHD (cGVHD). The incidence of cGVHD was also low; limited (n = 11) or extensive (n = 2) cGVHD developed in 13 (33%) patients after transplantation or DLI. There was a trend toward a higher incidence of cGVHD in unrelated compared to sibling donor recipients (53% vs 21%, P = .079, Fisher exact test).

Two patients with grade III and IV aGVHD and both patients with extensive cGVHD died of infections, 2 of them of pulmonary aspergillosis, resulting in a GVHD-related mortality rate of 10%.

Infections

Infections were the most common cause of death and morbidity in our cohort of patients. Nine (22%) patients died of infectious complications, 4 of them GVHD related. These included pulmonary aspergillosis (2 patients), bacterial pneumonia (2 patients), neutropenic sepsis (2 patients), RSV pneumonia (1 patient), EBV-related posttransplantation lymphoproliferative disease (1 patient), and fungal cerebral abscess (1 patient). Other documented infections were herpes simplex virus oral infections (4 patients), adenovirus viremia (2 patients), gastroenteritis (1 patient), herpes zoster (2 patients), influenzae upper respiratory tract infection (1 patient), parainfluenzae upper respiratory tract infection (1 patient), EBV viremia with no evidence of lymphoma (1 patient), and Nocardia asteroides cerebral abscess (1 patient). Six patients had documented septicemia caused by the following organisms: Pseudomonas aeruginosa (2 patients), Staphylococcus aureus (2 patients), Enterobacter spp, and Listeria monocytogenes. Three patients had recurrent bacterial lower respiratory tract infections. Nineteen (68%) of 28 patients at risk had CMV reactivation, but only 2 progressed to CMV disease (hepatitis and colitis) despite prompt ganciclovir therapy. One patient with CMV disease did not respond to ganciclovir and was successfully treated with foscarnet and cidofovir.

Disease response, DLI, and survival

After a median observation period of 15.1 months (range, 0.2-62.5 months), 24 patients were alive and 17 patients had died, 5 of progressive disease and 12 of transplant-related complications. Causes of death were infections (9 patients), acute respiratory distress syndrome (1 patient), myocardial infarction (1 patient), and unknown (1 patient). Two patients died within 3 months of transplantation and were not assessable for disease response.

Six (86%) of 7 patients who underwent transplantation for chemorefractory disease achieved CR or PR, which was sustained in 3 (43%) of them. One patient did not respond to the procedure and died, 1 patient died of a fungal cerebral abscess, and 2 patients had relapses shortly after transplantation. Both patients received DLIs, but neither responded and both died of progressive disease.

For patients with chemosensitive disease, the overall response rate (ORR = CR + PR) was 100%; only 3 of these patients remained in PR because of persistent cytopenia after HCT. Nine (28%) patients had relapses at a median of 5 months after HCT (range, 2.1-39.4 months). After relapse, all patients received DLIs, but these were effective in only 3 (33%) of them. Interestingly, 4 of 6 patients who did not respond to DLI are alive and well; 2 died of progressive disease. In 3 of these 6 unresponsive patients, disease relapse was heralded by mixed chimerism (see “Engraftment and donor chimerism”), and, even though DLIs were commenced at that stage, full donor chimerism was never achieved and relapse could not be avoided (Table 3).

Figure 1.

Relapse, TRM, and survival. (Top) Cumulative incidence curves of relapse and TRM. (Bottom) Kaplan-Meier plots of OS and PFS.

Figure 1.

Relapse, TRM, and survival. (Top) Cumulative incidence curves of relapse and TRM. (Bottom) Kaplan-Meier plots of OS and PFS.

The probabilities of day-100 and 2-year TRM were 5% (1%-19%) and 26% (14%-46%), respectively, and the 2-year relapse risk was 29% (17%-49%) (Figure 1). Age was the only factor with a significant impact on the TRM (hazard ratio, 3.52 [95% CI, 1.06-11.77]; P = .041). Other pretransplantation factors— such as previous autologous transplantation; previous therapy with fludarabine, alemtuzumab, or rituximab; number of previous chemotherapy regimens (less than 3 vs 3 or more), alemtuzumab dose in the conditioning regimen (less than 100 vs 100 mg), or fludarabine refractoriness—had no significant effect on TRM.

Kaplan-Meier estimates of OS and PFS are also shown in Figure 1. Estimated OS and PFS rates at 2 years were 51% (95% CI, 33%-69%) and 45% (95% CI, 27%-62%), respectively. Importantly, donor type (sibling vs unrelated) did not have a significant impact on survival.

Three factors turned out to have a significant effect on OS by univariate analysis: older age (P = .0381), fludarabine refractoriness (P = .0437), and center effect (P = .0131), but none of these variables remained significant in multivariate regression analysis. Older age (P = .0176), center effect (P = .0044), and fludarabine refractoriness (P = .0015) also had a significant impact on PFS by univariate analysis. Fludarabine refractoriness (HR, 5.05 [95% CI, 1.67-15.28]; P = .004) and center effect (HR, 3.14 [95% CI, 1.20-8.25]; P = .020) were included in the Cox regression equation. However, this center effect was difficult to gauge because both groups were unbalanced in terms of previous chemotherapy and donor type. Indeed, 90% of patients who underwent transplantation in the “less experienced” centers had 3 or more previous chemotherapy lines compared with 52% of patients who underwent transplantation in the “more experienced” centers (P = .016, Fisher exact test). In addition, 63% of patients in the first group received an unrelated allograft compared with 23% of patients in the latter group (P = .012, Fisher exact test).

Discussion

Modern salvage therapies, including alemtuzumab and rituximab, are promising in young patients with advanced CLL,15,25  but it is uncertain whether they will lead to prolonged survival. In the past few years, more data have become available with regard to reduced-intensity allogeneic transplantation in CLL.26-29  The aim of these studies is to deliver the advantages of the graft-versus-leukemia effect without the TRM associated with conventional HCT.

A reduced-intensity protocol comprising fludarabine and cyclophosphamide with or without rituximab was used at the MD Anderson Cancer Center in 17 patients with HLA-identical sibling donors.26  The incidence of cGVHD was 60%, with 5 patients dying of cGVHD or GVHD-related infections. The 2-year TRM rate was 22%, but the response rate was high (94%), even though the effect of posttransplantation rituximab is impossible to evaluate. The estimated 2-year OS and PFS rates were 80% and 60%, respectively, with no clear evidence of a plateau.20  The Cooperative German Transplant Study Group reported on the outcomes of 30 patients conditioned with fludarabine, busulfan, and ATG.27  Fifteen had sibling and 15 had unrelated donors. The TRM rate was 15% at 2 years, and aGVHD was relatively common despite in vivo T-cell depletion, particularly in patients with unrelated donors. The incidence of cGVHD was 75%, and 40% of patients attained CR after HCT. Only a minority of patients was eligible for DLI because of the high GVHD rates. In addition, only 1 of 6 patients had a durable response to DLI. The 2-year OS and PFS rates were 72% and 67%, respectively. The Seattle group28  has recently reported its experience with low-dose total body irradiation with or without fludarabine therapy in 64 patients with CLL. The 2-year TRM rate was 22%, and extensive cGVHD developed in 50% of patients. Responses were observed in 67% of patients with measurable disease before HCT. The 2-year relapse, OS, and PFS rates were 26%, 60%, and 52%, respectively. Unrelated donor HCT induced lower relapse rates than sibling HCT, which suggests a powerful graft-versus-leukemia effect. However, patients who experienced disease progression generally did not respond to DLI; only 1 PR was documented in 6 patients treated (17% response rate). Finally, 77 patients with CLL were identified in a recent European Group for Blood and Marrow Transplantation (EBMT) survey.29  Unfortunately, the interpretation is complicated by the diversity of conditioning regimens. The 1-year TRM rate was 18%, and the 2-year EFS and OS rates were 56% and 72%, respectively. The 2-year probability of progression or relapse was 31%. Achievement of CR (in 69% of patients) was significantly associated with cGVHD. Furthermore, objective responses were obtained in 4 (33%) of 12 patients who received DLI for insufficient disease control.

Our alemtuzumab-based regimen was feasible and effectively reduced the incidence of GVHD in sibling and unrelated donor recipients compared with other published regimens. Indeed, the overall incidence of aGVHD was 41% (12% after DLI), with an incidence of grades 3 to 4 GVHD of only 10%. The cGVHD rate was also low (33%), and only 2 patients acquired extensive cGVHD, both after DLI. The GVHD-related mortality rate was only 10%, which compares favorably with all previous reports. Trilineage engraftment was achieved in all but 1 patient. Unfortunately, graft rejection occurred in 5 other patients at a median of 4 months after HCT. Four of these patients received a second stem cell infusion. In 2 of them, it was conditioned with ATG according to institutional protocol. Engraftment was successful in all 4 patients, and none acquired GVHD. Both patients who received ATG died shortly afterward of EBV-induced posttransplantation lymphoproliferative disease (PTLD) and RSV pneumonia, respectively. ATG has been frequently used in patients with these conditions to enhance engraftment and to prevent GVHD,30  but it might have been unnecessary in this subgroup of CLL patients previously conditioned with alemtuzumab. Indeed, the risk for PTLD depends greatly on the method of T-cell depletion, which is considerably higher when specific T-cell antibodies (ATG) rather than T- and B-cell antibodies (alemtuzumab) are used.31 

Furthermore, the antitumor activity of the conditioning regimen was impressive, with 100% of chemosensitive and 86% of chemorefractory patients initially attaining CR or PR after HCT. This is not surprising because alkylating agents, purine analogs, and monoclonal antibodies constitute the mainstay of treatment for advanced-stage CLL. Four patients had active immune cytopenia at the time of transplantation, 2 of which were induced by fludarabine therapy, but this was not an impediment, nor did the cytopenias worsen after HCT. Unfortunately, these responses were durable in only 74% and 43% of chemosensitive and chemorefractory patients, respectively. The 2-year relapse risk rate was 29%, with no difference between sibling and unrelated donors. Furthermore, DLI was effective when used for reversing mixed chimerism but not for relapsed/progressive disease. Our study revealed that only 3 (27%) of 11 patients who received DLI for relapsed/progressive disease actually responded to the procedure (Table 3). This observation mirrors the results obtained by 3 different groups; DLI achieved an ORR of only 14% to 17% in CLL patients with relapsed/progressive disease after HCT.16,27,28  The MD Anderson group,26  on the other hand, has reported a very good response rate to DLI in a subset of 7 patients with progressive disease (ORR, 86%), but interpretation of this report is complicated by the fact that rituximab was concomitantly administered in 5 patients.

Interestingly, not all clinical relapses were ominous, in particular in those patients with chemosensitive disease before transplantation. Indeed, 4 of these patients are alive 14, 23, 24, and 44 months after relapse, and 2 have not required any treatment. A recent study from Boston has addressed the antitumor effect of T cell-depleted, reduced-intensity transplantation in patients whose grafts were rejected.32  In their series, 9 (41%) of 22 patients who lost donor chimerism ultimately achieved disease control, and 7 of these 9 patients were alive at 2.5 to 5.5 years after HCT. They postulated that the host-versus-graft immune response associated with graft rejection might have the potential to promote antitumor responses in this group of patients.32 

The 2-year TRM rate was significantly higher than in our previous report on reduced-intensity transplantation in patients with low-grade lymphoma (26% vs 11%),19  primarily because of high incidences of severe bacterial, viral, and fungal infections. Unlike patients with follicular lymphoma, who usually die of refractory disease or transformation to high-grade disease,33  it is widely acknowledged that infections account for up to 50% of all CLL-related deaths.34  This susceptibility is disease and therapy related35  and is secondary to multiple factors, including hypogammaglobulinemia, defective T- and NK-cell function, neutropenia, and deficient complement activity. Furthermore, alemtuzumab depletes incoming T cells and, therefore, delays the immune reconstitution after HCT. Indeed, the median time to CD4+ T-cell recovery (more than 0.2 × 109/L) with this protocol is 9 months, which is more delayed than with other myeloablative or nonmyeloablative regimens.17,36  In this study, CMV reactivation was a common event (68%), though only 2 patients acquired CMV disease despite preemptive treatment.36  Other particularly problematic viruses were adenovirus, EBV, and RSV—each had a significant impact on our patients' morbidity and mortality.37,38  Among the other opportunistic infections we observed were 2 cases of fatal pulmonary aspergillosis and 1 case of cerebral nocardiosis in 3 patients receiving corticosteroids because of cGVHD, highlighting the importance of appropriate anti-infective prophylaxis in such patients.

Unexpectedly, the graft failure rate was significantly higher (15%) than previously reported for patients with Hodgkin and non-Hodgkin lymphoma conditioned with the same regimen.19,21  The limited number of patients in our study prevents us from reaching any definitive conclusion, but we speculate that this is disease related. In particular, the deficient complement activity described in CLL patients might have reduced the cytotoxic effect of alemtuzumab, leaving residual host T cells capable of rejecting the graft.39  Furthermore, there is some evidence that sustained allogeneic engraftment is facilitated by host dendritic cells and other ancillary marrow elements,40  which are severely defective in CLL patients.41  Interestingly, a recent study42  on unrelated transplantation for CLL using myeloablative conditioning regimens revealed a similar incidence (18%) of graft failure.

Fludarabine refractoriness emerged as a clear prognostic factor in this group of transplant patients and had a significant impact on OS and PFS. Indeed, fludarabine refractory patients had 1-year and 2-year survival rates of 54% (24%-84%) and 31% (1%-61%), respectively, which is only marginally better than the results obtained with chemotherapy alone.5  Alternative therapies or transplantation schemes might be necessary in this group of patients with poor prognosis. We also observed a possible center effect, but this is difficult to assess because of unbalanced pretransplantation characteristics.

In summary, our alemtuzumab-based reduced intensity protocol is feasible and effective in CLL patients at poor risk, and it has an impressive response rate and a relatively low GVHD rate. However, the TRM rate remains relatively high because of a variety of viral and fungal infections. Ongoing studies are trying to address the efficacy of reduced doses of alemtuzumab in this group of immunosuppressed patients.

Prepublished online as Blood First Edition Paper, October 20, 2005; DOI 10.1182/blood-2005-08-3372.

Conceived by D.W.M. and J.D. and written principally by J.D., K.T., N.R., J.E., W.S., G.C., S.D., R.L., R.C., and S.M. contributed patients and have made contributions to the manuscript. S.M. devised the conditioning regimen. D.I.M. is Chair of the British Society of Blood and Marrow Transplantation (BSBMT) Research Committee and contributed to the manuscript.

The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 U.S.C. section 1734.

We thank Drs Richard Szydlo, Ronald Brand, and Rodrigo Martino for their statistical advice and all referring physicians and nurses for their dedicated care of these patients.

1
Rozman C, Montserrat E. Chronic lymphocytic leukemia.
N Engl J Med
.
1995
;
333
:
1052
-1057.
2
Keating MJ, O'Brien S, Lerner S, et al. Long-term follow-up of patients with chronic lymphocytic leukemia (CLL) receiving fludarabine regimens as initial therapy.
Blood
.
1998
;
92
:
1165
-1171.
3
Mauro FR, Foa R, Giannarelli D, et al. Clinical characteristics and outcome of young chronic lymphocytic leukemia patients: a single institution study of 204 cases.
Blood
.
1999
;
94
:
448
-454.
4
Paneesha S, Milligan DW. Stem cell transplantation for chronic lymphocytic leukaemia.
Br J Haematol
.
2005
;
128
:
145
-152.
5
Keating MJ, O'Brien S, Kontoyiannis D, et al. Results of first salvage therapy for patients refractory to a fludarabine regimen in chronic lymphocytic leukemia.
Leuk Lymphoma
.
2002
;
43
:
1755
-1762.
6
Dreger P, Montserrat E. Autologous and allogeneic stem cell transplantation for chronic lymphocytic leukemia.
Leukemia
.
2002
;
16
:
985
-992.
7
Dreger P, Stilgenbauer S, Benner A, et al. The prognostic impact of autologous stem cell transplantation in patients with chronic lymphocytic leukemia: a risk-matched analysis based on the VH gene mutational status.
Blood
.
2004
;
103
:
2850
-2858.
8
Milligan DW, Fernandes S, Dasgupta R, et al. Results of the MRC pilot study show autografting for younger patients with chronic lymphocytic leukemia is safe and achieves a high percentage of molecular responses.
Blood
.
2005
;
105
:
397
-404.
9
Khouri IF, Keating MJ, Saliba RM, Champlin RE. Long-term follow-up of patients with CLL treated with allogeneic hematopoietic transplantation.
Cytotherapy
.
2002
;
4
:
217
-221.
10
Michallet M, Archimbaud E, Bandini G, et al. HLA-identical sibling bone marrow transplantation in younger patients with chronic lymphocytic leukemia: European Group for Blood and Marrow Transplantation and the International Bone Marrow Transplant Registry.
Ann Intern Med
.
1996
;
124
:
311
-315.
11
Esteve J, Villamor N, Colomer D, et al. Stem cell transplantation for chronic lymphocytic leukemia: different outcome after autologous and allogeneic transplantation and correlation with minimal residual disease status.
Leukemia
.
2001
;
15
:
445
-451.
12
Kottaridis PD, Milligan D, Chopra R, et al. In vivo Campath-1H prevents graft-versus-host disease following nonmyeloablative stem cell transplantation.
Blood
.
2000
;
96
:
2419
-2425.
13
Chakraverty R, Peggs K, Chopra R, et al. Limiting transplantation-related mortality following unrelated donor stem cell transplantation by using a nonmyeloablative conditioning regimen.
Blood
.
2002
;
99
:
1071
-1078.
14
Keating MJ, Flinn I, Jain V, et al. Therapeutic role of alemtuzumab (Campath-1H) in patients who have failed fludarabine: results of a large international study.
Blood
.
2002
;
99
:
3554
-3561.
15
Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy is associated with prolonged survival.
J Clin Oncol
.
2005
;
23
:
2971
-2979.
16
Marks DI, Lush R, Cavenagh J, et al. The toxicity and efficacy of donor lymphocyte infusions given after reduced-intensity conditioning allogeneic stem cell transplantation.
Blood
.
2002
;
100
:
3108
-3114.
17
Morris EC, Rebello P, Thomson KJ, et al. Pharmacokinetics of alemtuzumab used for in vivo and in vitro T-cell depletion in allogeneic transplantations: relevance for early adoptive immunotherapy and infectious complications.
Blood
.
2003
;
102
:
404
-406.
18
Peggs KS, Thomson K, Hart DP, et al. Dose escalated donor lymphocyte infusions following reduced intensity transplantation: toxicity, chimerism, and disease responses.
Blood
.
2004
;
103
:
1548
-1556.
19
Morris EC, Thomson K, Craddock C, et al. Outcomes after alemtuzumab-containing reduced-intensity allogeneic transplantation regimen for relapsed and refractory non-Hodgkin's lymphoma.
Blood
.
2004
;
104
:
3865
-3871.
20
Peggs KS, Mackinnon S, Linch DC. The role of allogeneic transplantation in non-Hodgkin's lymphoma.
Br J Haematol
.
2005
;
128
:
153
-168.
21
Peggs KS, Hunter A, Chopra R, et al. Clinical evidence of a graft-versus-Hodgkin's-lymphoma effect after reduced-intensity allogeneic transplantation.
Lancet
.
2005
;
365
:
1934
-1941.
22
Jaffe ES, Harris NL, Stein H, et al, eds.
Pathology and Genetics of Tumours of the Haematopoietic and Lymphoid Tissues: World Health Classification of Tumours
. Lyon, France: IARC Press;
2001
.
23
Przepiorka D, Weisdorf D, Martin P, et al. Consensus conference of acute GVHD grading.
Bone Marrow Transplant
.
1995
;
15
:
825
-828.
24
Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-sponsored Working Group guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis and treatment.
Blood
.
1996
:
87
:
4990
-4997.
25
Wierda W, O'Brien S, Wen S, et al. Chemoimmunotherapy with fludarabine, cyclophosphamide, and rituximab for relapsed and refractory chronic lymphocytic leukemia.
J Clin Oncol
.
2005
;
23
:
4070
-4078.
26
Khouri IF, Lee MS, Saliba RM, et al. Nonablative allogeneic stem cell transplantation for chronic lymphocytic leukemia: impact of rituximab on immunomodulation and survival.
Exp Hematol
.
2004
;
32
:
28
-35.
27
Schetelig J, Thiede C, Bornhäuser M, et al. Evidence of a graft-versus-leukemia effect in chronic lymphocytic leukemia after reduced-intensity conditioning and allogeneic stem-cell transplantation: the Cooperative German Transplant Study Group.
J Clin Oncol
.
2003
;
21
:
2747
-2753.
28
Sorror ML, Maris MB, Sandmaier BM, et al. Hematopoietic cell transplantation after nonmyeloablative conditioning for advanced chronic lymphocytic leukemia.
J Clin Oncol
.
2005
;
23
:
3819
-3829.
29
Dreger P, Brand R, Hansz J, et al. Treatment-related mortality and graft-versus-leukemia activity after allogeneic stem cell transplantation for chronic lymphocytic leukemia using intensity-reduced conditioning.
Leukemia
.
2003
;
17
:
841
-848.
30
Remberger M, Ringden O, Ljungman P, et al. Booster marrow or blood cells for graft failure after allogeneic bone marrow transplantation.
Bone Marrow Transplant
.
1998
;
22
:
73
-78.
31
Curtis RE, Travis LB, Rowlings PA, et al. Risk of lymphoproliferative disorders after bone marrow transplantation: a multi-institutional study.
Blood
.
1999
;
94
:
2208
-2216.
32
Dey BR, McAfee S, Colby C, et al. Anti-tumour response despite loss of donor chimaerism in patients treated with non-myeloablative conditioning and allogeneic stem cell transplantation.
Br J Haematol
.
2005
;
128
:
351
-359.
33
Johnson PW, Rohatiner AZ, Whelan JS, et al. Patterns of survival in patients with recurrent follicular lymphoma: a 20-year study from a single center.
J Clin Oncol
.
1995
;
13
:
140
-147.
34
Oscier D, Fegan C, Hillmen P, et al. Guidelines on the diagnosis and management of chronic lymphocytic leukaemia.
Br J Haematol
.
2004
;
125
:
294
-317.
35
Hensel M, Kornacker M, Yammeni S, Egerer G, Ho AD. Disease activity and pretreatment, rather than hypogammaglobulinaemia, are major risk factors for infectious complications in patients with chronic lymphocytic leukaemia.
Br J Haematol
.
2003
;
122
:
600
-606.
36
Chakrabarti S, Mackinnon S, Chopra R, et al. High incidence of cytomegalovirus infection after nonmyeloablative stem cell transplantation: potential role of Campath-1H in delaying immune reconstitution.
Blood
.
2002
;
99
:
4357
-4363.
37
Chakrabarti S, Avivi I, Mackinnon S, et al. Respiratory virus infections in transplant recipients after reduced-intensity conditioning with Campath-1H: high incidence but low mortality.
Br J Haematol
.
2002
;
119
:
1125
-1132.
38
Avivi I, Chakrabarti S, Milligan DW, et al. Incidence and outcome of adenovirus disease in transplant recipients after reduced-intensity conditioning with alemtuzumab.
Biol Blood Marrow Transplant
.
2004
;
10
:
186
-194.
39
Schlesinger M, Broman I, Lugassy G. The complement system is defective in chronic lymphocytic leukemia patients and their healthy relatives.
Leukemia
.
1996
;
10
:
1509
-1513.
40
Fugier-Vivier IJ, Rezzoug F, Huang F, et al. Plasmacytoid precursor dendritic cells facilitate allogeneic hematopoietic stem cell engraftment.
J Exp Med
.
2005
;
201
:
373
-383.
41
Orsini E, Guarini A, Chiaretti S, Mauro FR, Foa R. The circulating dendritic cell compartment in patients with chronic lymphocytic leukemia is severely defective and unable to stimulate an effective T-cell response.
Cancer Res
.
2003
;
63
:
4497
-4506.
42
Pavletic SZ, Khouri IF, Haagenson M, et al. Unrelated donor transplantation for B-cell chronic lymphocytic leukemia after using myeloablative conditioning: results from the Center for International Blood and Marrow Transplant Research.
J Clin Oncol
.
2005
;
23
:
5788
-5794.