Bone marrow transplantation for chronic myeloid leukemia (CML) from unrelated and sibling donors: single center experience. T Lamparelli1, MT Van Lint1, ...
Bone Marrow Transplantation, (1997) 20, 1057–1062 1997 Stockton Press All rights reserved 0268–3369/97 $12.00
Bone marrow transplantation for chronic myeloid leukemia (CML) from unrelated and sibling donors: single center experience T Lamparelli1, MT Van Lint1, F Gualandi1, D Occhini1, M Barbanti 2, N Sacchi2, G Ficai2, C Ghinatti2 , GB Ferrara3, L Delfino3, S Pozzi3, A Morabito3, P Zikos1, V Vitale4, R Corvo4, F Frassoni and A Bacigalupo1 1
Divisione Ematologia II, Ospedale San Martino, Genova; 2 Italian Bone Marrow Transplant Registry (IBMDR) Ospedale Galliera; and 3Servizio Immunogenetica, 4 Servizio Radioterapia, Istituto Tumori, Genova, Italy
Summary: This is a report on 60 consecutive patients with chronic myeloid leukemia (CML) who received an allogeneic bone marrow transplant (BMT) in this Unit. Donors were HLA-identical siblings (SIB) (n 5 36) or unrelated donors (MUD) (n 5 24) matched by serology for HLA A and B and by molecular biology for HLA DR. All patients were prepared with cyclophosphamide 120 mg/kg and fractionated total body irradiation 10– 12 Gy. GVHD prophylaxis consisted of cyclosporin A (CsA) starting on day 27 and short-course methotrexate. Bone marrow was unmanipulated in all cases. Cytomegalovirus prophylaxis consisted of acyclovir for SIBs and foscarnet for MUDs. When compared to SIB transplants, MUD patients were younger (29 vs 36 years; P 5 0.002), had younger donors (31 vs 39; P 5 0.001), had a longer interval between diagnosis and BMT (1459 vs 263 days; P , 0.001) and received a smaller number of nucleated cells at transplant (3.3 vs 4.4 3 108/kg; P 5 0.003). More MUDs had advanced disease (50 vs 17%, P 5 0.005). The median day to 0.5 3 109/l neutrophils was similar in both groups (18 days for SIBs vs 17 days for MUDs; P 5 0.06); the median platelet count on days 130, 150, 1100 was significantly (P , 0.01) higher in SIB than in MUD patients (122 vs 38, 113 vs 50 and 97 vs 45 3 109/l, respectively). Acute GVHD was scored as absent-mild, moderate, or severe, in 36, 58 and 6% of SIBs vs 25, 42 and 33% in MUD patients (P 5 0.01). Chronic GVHD was comparable (P 5 0.1). The actuarial risk of CMV antigenemia at 1 year was 60% in both groups. There were six deaths in SIB patients (two leukemia, two infections, one GVHD, one pneumonitis) and four deaths in MUD patients (three acute GVHD and one infection). Fifty patients survive with a median follow-up of 656 days for SIBs and 485 for MUDs. The actuarial 3-year transplant-related mortality is 12% in SIBs and 17% in MUDs (P 5 0.5); the actuarial relapse is 18% in SIBs vs 6% in MUDs (P 5 0.4) and 3-year survival 78% in SIBs vs 82% in MUDs (P 5 0.7). This study suggests
Correspondence: Dr A Bacigalupo, Divisione Ematologia 2, Ospedale San Martino, 16132 Genova, Italy Received 7 April 1997; accepted 1 August 1997
that survival of CML patients after marrow transplantation from unrelated or sibling donors is currently similar, provided the former are well matched. The increased incidence of GVHD in MUD patients is possibly compensated by a lower risk of relapse. Keywords: chronic myeloid leukemia; unrelated donor transplants
Allogeneic bone marrow transplantation (BMT) is one of the therapeutic options for patients with chronic myeloid leukemia (CML), and is capable of inducing long-term disease-free survival in over half of patients, when performed in the early stages of the disease.1–4 Complications of allogeneic BMT such as poor graft function, immune deficiency, graft versus host disease (GVHD) and cytomegalovirus infections (CMV) are all major causes of failure, particularly in CML patients and lead to a transplant-related mortality (TRM) of 20–50%.3 To investigate possible differences between CML and acute myeloid leukemia (AML), we analyzed 213 patients undergoing unmanipulated allogeneic BMT with the same conditioning regimen, from an HLA-identical sibling: CML patients had significantly slower neutrophil (P = 0.0001) and platelet engraftment (P = 0.002)5 and delayed immune reconstitution (P = 0.0001) as indicated by lower absolute CD4 counts: CML patients had significantly greater TRM as compared to AML patients (39 vs 16%, P = 0.002) (unpublished). We therefore observed a greater TRM associated with delayed graft function in patients with CML, possibly due to the fact that CML patients may be more immune competent when compared to acute leukemias, not having received multiple courses of chemotherapy. In an attempt to increase suppression of the recipient’s immune system, thus favoring graft function, we therefore opened a study using intravenous cyclosporin A (CsA) from day −7 (CsA-7) rather than from day −1 (CsA-1). It was initially designed as a randomized study, but randomization was closed after 14 patients, due to an excess death rate in the CsA-1 arm (4/7 vs 0/7, P = 0.03). One explanation could be moderate in vivo T cell depletion, since patients on the CsA-7 arm had higher CsA serum levels on the day of transplant, when compared to patients on the CsA-1 arm (median 260 vs 125 ng/ml, P = 0.01). We have since proceeded with a single arm study, and every patient
Marrow transplantation for CML T Lamparelli et al
1058
with CML both on the HLA-identical sibling program as well as on the unrelated marrow protocol, has received the CsA-7 program. We now report 60 consecutive CML patients who were allografted on the CsA-7 protocol from an HLA-identical sibling (SIB), or from an unrelated HLA-matched donor (MUD).
loosing cells in the infusion filters. In cases of major ABO incompatibility (A vs 0, B vs 0, AB vs 0, AB vs B and AB vs A) the marrow was not depleted of red cells, but isohemagglutinins were reduced in the patient to a titer of 5) were treated with combined ganciclovir and foscarnet.9 For MUD patients at any level of CMV antigenemia, ganciclovir (5–10 mg/kg/day) was added to foscarnet for 15 days. Statistical analysis
Table 1
Patient clinical data
No. patients Patients Median age M/F Donors Median age M/F Disease Int Dx-BMT 1CP/adv Transplant Nucleated cells × 108/kg
SIB
MUD
36
24
36 (26–53) 25/11
29 (18–46) 12/12
0.002 0.1
Results
39 (17–61) 19/17
31 (24–50) 13/11
0.001 0.7
Patients
263 (95–886) 30/6 4.4 (2.4–8.3)
P value
The x2 and rank sum test were used for statistical analyses, and Kaplan–Meier plots for time-dependent analysis (survival, transplant mortality and relapse).
1495 (574–2910) ,0.0001 12/12 0.005 3.3 (2.6–6)
0.003
Int Dx-BMT = interval diagnosis BMT; SIB = sibling donors; MUD = matched unrelated donors; ICP = 1st chronic phase; adv = more advanced.
As shown in Table 2, MUD patients had more days in hospital (P = 0.03) and more days of fever (P = 0.01). Engraftment Median days to 0.5 × 109/l neutrophils were 17 (MUD) vs 18 (SIBS) P = 0.06. The median number of days with neu-
Marrow transplantation for CML T Lamparelli et al
1059
Engraftment
No. patients Day of engraftment PMN 0.5 × 109 Platelet counts (×109/l) day 30 day 50 day 100 Acute GVHD grade 0–1 grade II grade III–IV Chronic GVHD (No. patients at risk) no limited extensive Days in hospital Days of fever Days of neutropenia Death
