Haploidentical vs autologous hematopoietic stem cell transplantation ...

Bone Marrow Transplantation (2003) 31, 889–895 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00

www.nature.com/bmt

Haploidentical vs autologous hematopoietic stem cell transplantation in patients with acute leukemia beyond first remission S Singhal1,2,3, PJ Henslee-Downey2, R Powles1, KY Chiang2, K Godder2, J Treleaven1, S Kulkarni1, F van Rhee2, B Sirohi1, CR Pinkerton1, S Meller1, B Jovanovic3 and J Mehta1,2,3 1 Royal Marsden Hospital, Surrey, UK; 2Division of Transplantation Medicine, University of South Carolina, South Carolina Cancer Center, Columbia, SC, USA; and 3The Feinberg School of Medicine, Northwestern University, Chicago, IL, USA

Summary: This is a retrospective comparison of partially mismatched related donor transplantation (PMRDT) and autotransplantation (ABMT) in advanced acute leukemia. Patients underwent T-cell-depleted PMRDT (n ¼ 164) or ABMT (n ¼ 131) for acute myeloid leukemia (n ¼ 130) or acute lymphoblastic leukemia (n ¼ 165). Fewer PMRDT patients were in remission (29 vs 85%; Po0.0001). The 5-year cumulative incidence of transplant-related mortality (TRM) was 52% after PMRDT and 16% after ABMT (Po0.0001). The 5-year cumulative incidence of relapse was 32% after PMRDT and 54% after ABMT (P ¼ 0.006). The actuarial unadjusted 5-year disease-free survival (DFS) was 16% after PMRDT and 30% after ABMT. In Cox’s regression analysis, PMRDT (Po0.0001) and age 415 years (P ¼ 0.002) were associated with higher TRM, active disease (P ¼ 0.0021), ABMT (P ¼ 0.0074) and male sex (P ¼ 0.011) with higher relapse, and age 415 years (P ¼ 0.0007) and PMRDT (P ¼ 0.047) with lower DFS. Amongst second remission patients, TRM was higher after PMRDT (P ¼ 0.0003), relapse was higher after ABMT (P ¼ 0.034), and 5-year DFS was comparable (32% ABMT and 25% PMRDT). ABMT, if feasible, may be preferable to PMRDT in advanced acute leukemia patients since lower relapse after PMRDT is offset by higher TRM. If an autograft is not feasible because of nonavailability of autologous cells or very advanced disease, PMRDT is a potential alternative. Bone Marrow Transplantation (2003) 31, 889–895. doi:10.1038/sj.bmt.1704031 Keywords: acute leukemia; autograft; allograft ; HLAmismatched transplant; graft-versus-leukemia

High-dose chemoradiotherapy with hematopoietic stem cell transplantation (HSCT) is standard therapy in advanced acute leukemia, that is, acute leukemia beyond the first complete remission (CR). The ideal stem cell donor is an HLA-identical sibling. For patients lacking an HLA-

identical sibling, potential treatment options include autologous blood or marrow transplantation (ABMT)1–4 or allogeneic HSCT from alternative donors such as HLAmismatched family members (partially mismatched related donors; PMRD), HLA-matched unrelated individuals, and umbilical cord blood.5–10 While autotransplants have a lower risk of transplantrelated mortality (TRM) than allografts under similar clinical circumstances,11 allogeneic transplants have the potential for greater antitumor effect because of the graftversus-leukemia (GVL) phenomenon.11,12 Alternative donor grafts are typically associated with a greater risk of complications and lower long-term survival than HSCT from HLA-identical siblings. The Seattle group13 found that the outcome of unrelateddonor HSCT and ABMT was comparable in advanced leukemia because higher TRM after allogeneic transplantation was counterbalanced by higher relapse after ABMT.13 Comparison of data from the National Marrow Donor Program (NMDP) and the Autologous Blood and Marrow Transplant Registry (ABMTR) showed that unrelateddonor transplantation was comparable to ABMT in first and second CR acute lymphoblastic leukemia (ALL),14 and inferior to ABMT in first and second CR acute myeloid leukemia (AML).15 PMRD transplantation (PMRDT) is a more readily available alternative than HSCT from an unrelated donor. The practice of PMRDT, despite a relatively long history,5,6,9,16–18 has been restricted because of the difficult nature of the procedure. Very high doses of blood-derived T-cell-depleted CD34+ cells have been utilized recently to achieve rapid haploidentical alloengraftment.19,20 While this has not improved the outcome compared to prior reports,9,17,18 encouraging short-term results have stimulated a resurgence of interest in PMRDT. This is the first report evaluating the outcome of patients with advanced acute leukemia undergoing PMRDT or ABMT to compare the relative merits and shortcomings of the two procedures.

Patients and methods Correspondence: Dr S Singhal, 676 N. St Clair St., Suite 850, Chicago, IL 60611, USA Received 26 July 2002; accepted 28 December 2002

The PMRDT population comprised all consecutive patients with acute leukemia beyond first CR (n ¼ 164)

Haploidentical vs autologous HSCT in patients with acute leukemia S Singhal et al

890

undergoing a first transplant from an HLA-mismatched relative at the Division of Transplantation Medicine of the South Carolina Cancer Center, Columbia, SC, USA between February 1993 and December 1999 using a standard total-body irradiation (TBI)-containing regimen. Patients undergoing PMRDT after having undergone a preceding autograft,21 those with primary refractory disease,22,23 those in first CR, and those with known secondary leukemia were excluded. The ABMT population comprised all consecutive advanced acute leukemia patients (n ¼ 131) undergoing a first autograft at the Leukaemia Unit of the Royal Marsden Hospital, Surrey, UK between April 1984 and November 1999. Patients undergoing a second autograft,24 those in first CR, and those with known secondary leukemia were excluded. All patients and donors provided written informed consent. All research protocols were approved by the respective institutional review boards.

Patient characteristics As Table 1 shows, the age, gender distribution and the type of leukemia were comparable for the groups. Cytogenetic characteristics could not be compared because these data were missing in over half of the patients. Table 2 shows the donor–patient relationship and the number of HLA mismatch vectors: 81% of the patient–donor pairs were mismatched on two or three loci in the graft-versus-host direction and 81% in the host-versus-graft direction. There were important biologic differences between the groups, which were highly significant statistically (Table 1): only 39 (24%) PMRDT recipients were in second CR compared with 89 (68%) ABMT recipients, and 116 (71%) PMRDT recipients had active leukemia (not in CR) at the time of the transplant compared with only 20 (15%) ABMT recipients. However, the patients transplanted in second CR were comparable for the two groups. All patients in CR, by definition, had chemosensitive disease. Among patients with active disease at the time of transplantation, the proportion of patients who had received no salvage chemotherapy after relapse was significantly higher in the ABMT group (15 of 20 ABMT recipients vs 22 of 116 PMRDT recipients; Po0.0001). Among those who had received salvage chemotherapy, 64 of 94 PMRDT patients had refractory disease compared with three of five ABMT patients (P ¼ 0.66).

Conditioning regimens In the PMRDT group, the conditioning regimens included TBI (1332/1410/1500 cGy in six fractions over 3 days), etoposide (20 mg/kg), cytarabine (12 g/m2 over 3 days), cyclophosphamide (100 mg/kg over 2 days), and methylprednisolone (4 g/m2 over 3 days) without (n ¼ 47) or with (n ¼ 117) horse antithymocyte globulin (ATGAM, Upjohn, Kalamazoo, MI, USA) (ATG; 30 mg/kg over 3 days). Posttransplant immunosuppression comprised cyclosporine, methylprednisolone, and ATG. In the ABMT group, the conditioning regimens included melphalan–TBI (n ¼ 85; 110–140 mg/m2 melphalan and 950–1150 cGy TBI), busulfan–cyclophosphamide (n ¼ 17; Bone Marrow Transplantation

Table 1

Patient characteristics

n Age (years) >15 years Male Second CR Other Remission Active disease Acute myeloid leukemia Acute lymphoblastic leukemia Second CR patients Age (years) >15 years Male Acute myeloid leukemia Acute lymphoblastic leukemia

Table 2

PMRDT

ABMT

164 1–54 (median 23) 103 (63%) 102 (62%) 39 (24%) 125 (76%) 48 (29%) 116 (71%) 80 (49%)

131 2–73 (median 18) 80 (61%) 80 (61%) 89 (68%) 42 (32%) 111 (85%) 20 (15%) 50 (38%)

84 (51%)

81 (62%)

1–50 (median 23) 21 (54%) 23 (59%) 14 (36%)

2–63 (median 18) 51 (57%) 52 (58%) 33 (37%)

25 (64%)

56 (63%)

P 0.9 0.4 0.9 o0.0001 o0.0001 0.07

0.9 0.7 0.9 0.9

Type of donors and mismatch vectors n

%

Donor Sibling Parent Offspring Other

59 58 31 16

36 35 19 10

Number of vectors mismatched Graft-versus-host 0 1 2 3

5 26 74 59

3 16 45 36

Host-versus-graft 0 1 2 3

5 25 81 53

3 15 49 32

16 mg/kg busulfan and 120–200 mg/kg cyclophosphamide), 200 mg/m2 melphalan as a single agent (n ¼ 9), busulfan– melphalan (n ¼ 8; 16 mg/kg busulfan and 140 mg/m2 melphalan), cyclophosphamide–TBI (n ¼ 3; 120 mg/kg cyclophosphamide and 1050 cGy TBI), and others (n ¼ 9). ABMT patients receiving TBI were irradiated using opposed 60Co sources at a slow dose rate (3–4 cGy/min), whereas PMRDT patients received TBI using a linear accelerator.

Source of stem cells All PMRDT patients received marrow, which was partially depleted of T cells using the monoclonal antibodies OKT3 (n ¼ 116) or T10B9 (n ¼ 48). The number of T cells administered to the PMRDT patients were 0.5– 161.0 104/kg recipient body weight (median 5.63).


Post-transplant supportive therapy As expected, because of the very different nature of the procedures, supportive care varied between the two centers as well as in each center over different time periods depending upon the prevailing practices and ongoing studies. ABMT recipients did not receive growth factors routinely after transplant to hasten myeloid recovery, whereas the majority of PMRDT patients received filgrastim or sargramostim post-transplant. A proportion of ALL patients received maintenance therapy with 6-mercaptopurine and methotrexate for 2 years after ABMT.4 No maintenance chemotherapy was given to AML patients, who were autografted, or to PMRDT recipients.

Statistical analysis All follow-up data are current as of August 2000. The w2 test was used to compare categoric variables and the Wilcoxon rank-sum test was used to compare continuous variables. Cumulative incidence of TRM and relapse was estimated using each type of event as a competing risk for the other.26,27 The significance of differences in TRM and relapse was calculated using the likelihood-ratio statistic for proportional-hazards regression models. The following factors were analyzed in a multivariable Cox’s proportional-hazards regression models for effect on TRM, relapse, disease-free survival (DFS) and overall survival (OS): age (p15 vs 415), sex (male vs female), diagnosis (AML vs ALL), disease status at transplant (CR vs active disease), and the type of transplant (autograft vs PMRDT). The probabilities of DFS and OS were adjusted for other prognostic factors (plotted at the mean of other covariates using a Cox model). Owing to the disparities in the patient characteristics between the two groups (Table 1), mainly the disease stage, the outcome of second CR patients (a relatively homogeneous population; Table 1) from each group was compared separately.

Results Transplant-related mortality Death owing to transplant-related causes occurred in 84 PMRDT patients (51%) 1–987 days (median 67) after transplantation. The causes of death were as follows: acute GVHD (n ¼ 7), bacterial infections (n ¼ 5), multiorgan failure (n ¼ 16), viral infections (n ¼ 17), pulmonary toxicity (n ¼ 10), chronic GVHD (n ¼ 3), fungal infections (n ¼ 12), graft rejection or failure (n ¼ 4), Epstein–Barr virus (EBV) lymphoproliferative disease (n ¼ 6), cardiotoxicity (n ¼ 2), CNS toxicity (n ¼ 1), thrombotic microangiopathy (n ¼ 1). All except seven deaths (three EBV, two chronic GVHD,

891 1.0

Cumulative Incidence

Marrow was the predominant source of stem cells for ABMT (n ¼ 114), and a minority received blood-derived stem cells (n ¼ 17; 2 with marrow). In 13 ABMT patients with ALL, marrow was purged ex vivo using the murine anti-CD52 monoclonal antibody Campath-1M.25

0.8 PMRDT: 84 of 164 died of toxicity

0.6

0.4 ABMT: 21 of 131 died of toxicity 0.2

0.0 0

4

8 Years

12

16

Figure 1 Cumulative incidence of TRM (Po0.0001). The curves and comparison are not adjusted for other prognostic factors.

one bacterial infection and one pulmonary toxicity) occurred within 1 year of transplantation. Death owing to transplant-related causes was seen in 21 ABMT patients (16%) 11–2599 days (median 85) after transplantation. The causes of death were as follows: bleeding (n ¼ 1), graft failure (n ¼ 2), infection (n ¼ 3), pulmonary toxicity (n ¼ 10), secondary malignancy (n ¼ 1; day 2559), renal failure (n ¼ 1), cardiomyopathy (n ¼ 1; day 803), liver failure (n ¼ 1), progressive multifocal leukoencephalopathy (n ¼ 1; day 1382). Except for three late deaths as specified, all transplant-related deaths occurred within 1 year of transplantation. As Figure 1 shows, the cumulative incidence of TRM was significantly higher after PMRDT.

Relapse Leukemia recurred in 51 PMRDT patients (31%) 21–1248 days (median 146) following the transplant. Death owing to relapsed disease or consequences of further therapy occurred in 48 patients. The remaining three patients are alive in CR 1071, 1253, and 1789 days after relapse; one each after chemotherapy, an infusion of donor leukocytes, and a second PMRDT. Leukemia recurred in 69 ABMT patients (53%) 0–1891 days (median 122) following the transplant; one patient with active disease who never cleared detectable leukemia was considered as having relapsed on day 0. The next detectable relapse occurred on day 31. Death owing to relapsed disease or consequences of further therapy occurred in 64 patients. Of the remaining five patients, four are alive in CR 732, 2765, 4165, and 4366 days after relapse; two after second autografts and two after chemotherapy. The fifth is receiving reinduction chemotherapy. As Figure 2 shows, the cumulative incidence of relapse was significantly lower after PMRDT.

Survival At the last follow-up, 32 PMRDT patients (20%) were alive 121–2501 days (median 2121) after transplantation; 29 in Bone Marrow Transplantation


1.0

1.0

0.8

0.8

Disease-free survival


892

ABMT: 69 of 131 relapsed

0.6

0.4

0.2

PMRDT: 51 of 164 relapsed

0.6 ABMT: 41 of 131 alive in continuous CR

0.4

0.2 PMRDT: 29 of 164 alive in continuous CR 0.0

0.0 0

4

8

12

0

16

2

4 Years

Years Cumulative incidence of relapse (P ¼ 0.006). The curves and comparison are not adjusted for other prognostic factors.

Figure 2

6

8

Figure 4 DFS by transplant type; adjusted for all other covariates.

1.0


1.0

Overall survival

0.8

0.6 ABMT: 46 of 131 alive 0.4

0.8 PMRDT: 20 of 39 died of toxicity

0.6

0.4 ABMT: 14 of 89 died of toxicity 0.2

0.2 PMRDT: 32 of 164 alive

0.0 0

0.0 0

4 Years

6

4

8

8

Outcome of patients in second CR Among those in second CR, as with the entire group of patients, TRM was significantly higher and relapse significantly lower after PMRDT compared with ABMT (Figures 5 and 6). DFS was comparable among second CR patients after PMRDT and ABMT (Figure 7).

16

Years

OS by transplant type; adjusted for all other covariates.

continuous CR. At the last follow-up, 46 ABMT patients (35%) were alive 146–5444 days (median 2591) after transplantation; 45 in CR (41 continuous CR) and one on reinduction chemotherapy. Figures 3 and 4 show the adjusted probabilities of OS and DFS.

12

Figure 5 TRM among patients in second remission (P ¼ 0.0003).

1.0


Figure 3

2

0.8

0.6

ABMT: 44 of 89 relapsed

0.4

0.2

PMRDT: 7 of 39 relapsed

0.0 0

4

8

12

16

Years

Cox’s regression analysis

Figure 6 Relapse among patients in second remission (P ¼ 0.034).

Table 3 shows the results of the multivariable analysis. PMRDT, despite reducing relapse, affected OS and DFS adversely as a result of higher TRM. Younger age was the strongest favorable variable.

Subgroup outcomes

Bone Marrow Transplantation

Table 4 shows outcome by the type of transplant, age, diagnosis, and the remission status at the time of the


lower DFS and OS than ABMT in all subgroups, and the difference was statistically significant in several subgroups.

1.0

893


0.8

Discussion Our data show that both ABMT and PMRDT have their place in the treatment of advanced acute leukemia. OS and DFS are better after ABMT because of lower TRM. Immunologic GVL reactions after PMRDT result in lower relapse rates, but this advantage is offset by the higher TRM. The subgroup of patients in second CR represents the most comparable population of patients. In this group, DFS was comparable with the two procedures because the higher TRM with PMRDT counterbalanced the higher relapse rate with ABMT. There are obvious limitations of this analysis. The two populations of patients were very different biologically. They were treated in different institutions. The transplantspecific therapy and supportive care therapy were very different – largely because of the widely disparate nature of the procedures under consideration. However, both institutions are highly experienced in their fields, and, as a result, the results reflect biologic differences between the types of transplant rather than center-specific effects. A question like this can only be answered satisfactorily in a prospec-

0.6 ABMT: 31 of 89 alive in continuous CR

0.4 PMRDT: 12 of 39 alive in continuous CR

0.2 0

1

2

3

4

5

6

Years Figure 7 DFS of patients in second remission by transplant type; adjusted for all other covariates.

transplant. PMRDT was associated with a significantly higher TRM than ABMT in all subgroups. PMRDT was associated with a lower relapse incidence than ABMT in all subgroups; but the difference was statistically significant only in some subgroups. PMRDT was associated with a Table 3

Multivariable analysis

Variable

PMRDT Active disease Male >15 years ALL

Table 4

Transplant-related mortality

Relapse


Overall survival

Relative risk (95% CI)

P


P


P


P

ABMT Remission

0.23 (0.13–0.38) 0.84 (0.55–1.29)

o0.0001 0.42

0.55 (0.35–0.85) 0.49 (0.31–0.77)

0.0074 0.0021

0.73 (0.53–0.99) 0.78 (0.57–1.07)

0.047 0.12

0.62 (0.45–0.86) 0.81 (0.59–1.12)

0.0036 0.19

Female p15 years AML

0.99 (0.67–1.47) 0.47 (0.29–0.76) 0.88 (0.58–1.36)

0.98 0.002 0.57

0.60 (0.40–0.89) 0.70 (0.46–1.06) 0.68 (0.45–1.04)

0.011 0.089 0.075

0.77 (0.58–1.01) 0.58 (0.43–0.80) 0.90 (0.67–1.20)

0.06 0.0007 0.46

0.80 (0.61–1.06) 0.55 (0.40–0.76) 0.93 (0.69–1.26)

0.12 0.0002 0.65

Outcome data by type of transplant, disease status at transplant and age Remission

Not in remission

P

p15 years

>15 years

P

AML

ALL

P

TRM ABMT PMRDT P

19% 58% 0.0001

0% 52% o0.0001

0.99 0.58

16% 44% 0.068

16% 65% o0.0001

0.84 0.0002

20% 62% o0.0001

16% 42% 0.0002

0.6 0.1

Relapse ABMT PMRDT P

52% 20% 0.03

77% 35% 0.13

0.05 0.009

42% 32% 0.09

64% 23% 0.004

0.003 0.21

56% 26% 0.14

55% 37% 0.73

0.7 0.4

DFS ABMT PMRDT P

31% 22% 0.29

23% 13% 0.19

0.41 0.03

42% 23% 0.01

22% 12% 0.03

0.005 0.02

27% 12% 0.02

32% 21% 0.03

0.5 0.3

OS ABMT PMRDT P

34% 28% 0.19

28% 14% 0.04

0.67 0.02

48% 27% 0.01

23% 13% 0.001

0.008 0.002

29% 13% 0.004

36% 23% 0.009

0.5 0.1

The TRM and relapse figures represent cumulative incidence at 5 years, and the OS and DFS figures represent actuarial 5-year probabilities. Bone Marrow Transplantation


894

tive, randomized study. However, a comparison like this would be virtually impossible to perform because of logistic reasons as well as the wide disparity in the potential toxicity of the two transplant approaches. Therefore, we feel that despite the shortcomings of the study, the information obtained does allow some conclusions to be drawn that could guide approach to choosing therapy in patients who need a transplant but do not have an HLA-identical sibling. In this sense, the present analysis is analogous to two recent studies from NMDP and ABMTR.14,15 As expected in high-risk acute leukemia, the long-term survival of these patients was modest. However, there is scope for improvement in the results. The major problem with PMRDT is high TRM, which can be decreased to some extent with advances in transplant technique and supportive care. An obvious possibility is the use of a nonmyeloablative transplant approach,28 although engraftment could be an issue with nonmyeloablative PMRDT. Transplantation relatively early in the course of the disease rather than very late (as done here because patients were usually subjected to PMRDT as a therapy of last resort), may result in lower TRM. However, as seen in the second CR (Figure 5) as well as other (Table 4) subgroups, a difference in TRM is bound to remain between ABMT and PMRDT because of the vastly different biologic nature of the procedures. The major drawback of ABMT is disease recurrence; reducing which may be somewhat more difficult. A joint study from the NMDP and the ABMTR compared ABMT and HSCT from unrelated donors in ALL14 and AML15 in first and second CR. For both diseases and stages, relapse rates were significantly higher after ABMT and TRM significantly higher after allogeneic transplantation. The two procedures yielded equivalent results in patients with ALL in first as well as second CR. In AML, ABMT was significantly superior for first CR (3year DFS probability of 53 vs 41% with unrelated-donor HSCT) as well as second CR (3-year DFS probability of 40 vs 33% with unrelated-donor HSCT) patients. A recent study showed that HSCT from HLA-matched unrelated donors resulted in a better outcome than PMRDT or HSCT from HLA-mismatched unrelated donors in patients with hematologic malignancies.29 The outcome of selected patients allografted from closely HLAmatched unrelated donors is excellent.30,31 The data from these two studies would suggest that in appropriate clinical situations, HSCT from well-matched unrelated donors might be preferable to PMRDT. However, there are two circumstances under which PMRDT may be preferable. One is the lack of a wellmatched unrelated donor. The second is an emergent clinical situation where it may not be possible to await the 8–12 weeks required to identify and obtain cells from an unrelated donor. We have shown that in patients with primary refractory acute leukemia, the time required to set up PMRDT is not much longer than required to organize a transplant from an HLA-matched sibling donor.23 Additionally, identification of biologic factors affecting the outcome of ABMT will also help determine choice of therapy. For example, AML patients with high-risk karyotypes or those failing to attain CR with a single cycle of chemotherapy32 or ALL patients with high-risk disease33


fare poorly with ABMT, and thus may be better served by an alternative-donor transplant. Another important factor that has not been considered in any of the studies comparing ABMT to alternatives13–15 is the cost, which is far higher for high-risk allografts (on an average, at least twice as much as autografts for the first month alone). High-risk allografts are also associated with greater morbidity for a longer period of time. These data as well as those from the NMDP-ABMTR collaboration14,15 suggest that the potential benefit from autotransplantation may be underrated in clinical practice. With all the limitations of the data, it may be possible to suggest some general practical guidelines. For patients with acute leukemia in second CR, if an autograft is feasible (for example, good marrow function and disease which is not very high risk), it should be the procedure of choice. However, autografting is not always possible in patients with more advanced disease. For example, patients who are in advanced CR (beyond second CR) and have poor marrow function or those who are in refractory relapse will not be candidates for autotransplantation. In these patients, an alternative donor transplant (PMRDT or unrelated-donor HSCT) is the only choice. It is fair to say that a significant proportion of patients who underwent PMRDT in this study in relapse or in advanced remission would not have been eligible for ABMT. Indeed, we have shown that it is possible to salvage patients relapsing after an autograft with PMRDT.21 Thus, the appropriate hierarchy of potential transplant procedures in a patient with advanced acute leukemia who requires high-dose therapy but lacks an HLA-identical sibling should be an autograft (if this is feasible; vide supra), HSCT from a nine or 10 allele-matched unrelated donor, PMRDT, and finally, HLA-mismatched unrelateddonor HSCT or umbilical cord blood transplantation.

References 1 Mehta J, Powles RL. Autologous blood and marrow transplantation. In: Whittaker JA, Holmes JA (eds). Leukaemia and Associated Diseases. Blackwell Science: Oxford, 1998, pp 455–481. 2 Chopra R, Goldstone AH, McMillan AK et al. Successful treatment of acute myeloid leukemia beyond first remission with autologous bone marrow transplantation using busulfan/ cyclophosphamide and unpurged marrow: the British autograft group experience. J Clin Oncol 1991; 9: 1840–1847. 3 Petersen FB, Lynch MHE, Clift RA et al. Autologous marrow transplantation for patients with acute myeloid leukemia in untreated first relapse or in second complete remission. J Clin Oncol 1993; 11: 1353–1360. 4 Mehta J, Powles R, Singhal S et al. Melphalan–total body irradiation and autologous bone marrow transplantation for adult acute leukemia beyond first remission. Bone Marrow Transplant 1996; 18: 119–123. 5 Powles RL, Morgenstern GR, Kay HE et al. Mismatched family donors for bone-marrow transplantation as treatment for acute leukaemia. Lancet 1983; 1: 612–615. 6 Anasetti C, Amos D, Beatty PG et al. Effect of HLA compatibility on engraftment of bone marrow transplants in patients with leukemia or lymphoma. N Engl J Med 1989; 320: 197–204.


895 7 Kernan NA, Bartsch G, Ash RC et al. Analysis of 462 transplantations from unrelated donors facilitated by the National Marrow Donor Program. N Engl J Med 1993; 328: 593–602. 8 Gluckman E, Rocha V, Boyer-Chammard A et al. Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N Engl J Med 1997; 337: 373–381. 9 Henslee-Downey PJ, Abhyankar SH, Parrish RS et al. Use of partially mismatched related donors extends access to allogeneic marrow transplant. Blood 1997; 89: 3864–3872. 10 Mehta J, Singhal S, Chiang KY et al. Haploidentical BMT for acute leukemia: single-center experience of 220 patients. Blood 2000; 96(Suppl. 1): 840a. 11 Mehta J, Tricot G, Jagannath S et al. Salvage autologous or allogeneic transplantation for multiple myeloma refractory to or relapsing after a first-line autograft? Bone Marrow Transplant 1998; 21: 887–892. 12 Mehta J. Graft-versus-leukemia reactions in clinical bone marrow transplantation. Leu Lymphoma 1993; 10: 427–432. 13 Busca A, Anasetti C, Anderson G et al. Unrelated donor or autologous marrow transplantation for treatment of acute leukemia. Blood 1994; 83: 3077–3084. 14 Weisdorf D, Bishop M, Dharan B et al. Autologous vs allogeneic unrelated donor transplantation for acute lymphoblastic leukemia: comparative toxicity and outcomes. Biol Blood Marrow Transplant 2002; 8: 213–220. 15 Lazarus HM, Perez WS, Weisdorf D et al. Autologous vs unrelated donor (URD) transplantation for acute myeloid leukemia (AML) in first or second remission (CR1 or CR2). Blood 2000; 96(Suppl. 1): Abstract #1781. 16 Soiffer RJ, Mauch P, Tarbell NJ et al. Total lymphoid irradiation to prevent graft rejection in recipients of HLA nonidentical T cell-depleted allogeneic marrow. Bone Marrow Transplant 1991; 7: 23–33. 17 Henslee-Downey PJ. Mismatched bone marrow transplantation. Curr Opin Oncol 1995; 7: 115–121. 18 Henslee-Downey PJ, Parrish RS, MacDonald JS et al. Combined in vitro and in vivo T lymphocyte depletion for the control of graft-versus-host disease following haploidentical marrow transplant. Transplantation 1996; 61: 738–745. 19 Aversa F, Tabilio A, Velardi A et al. Treatment of high-risk acute leukemia with T-cell-depleted stem cells from related donors with one fully mismatched HLA haplotype. N Engl J Med 1998; 339: 1186–1193. 20 Handgretinger R, Schumm M, Lang P et al. Transplantation of megadoses of purified haploidentical stem cells. Ann NY Acad Sci 1999; 872: 351–361. 21 Godder KT, Mehta J, Chiang KY et al. Partially mismatched related donor bone marrow transplantation as salvage for

22

23

24

25

26

27

28

29

30

31

32

33

patients with AML who failed autologous stem cell transplant. Bone Marrow Transplant 2001; 28: 1031–1036. Chiang KY, van Rhee F, Godder K et al. Allogeneic bone marrow transplantation from partially mismatched related donors as therapy for primary induction failure acute myeloid leukemia. Bone Marrow Transplant 2001; 27: 507–510. Singhal S, Powles R, Henslee-Downey PJ et al. Allogeneic transplantation from HLA-matched sibling or partially HLAmismatched related donors for primary refractory acute leukemia. Bone Marrow Transplant 2002; 29: 291–295. Mehta J, Powles R, Treleaven J et al. Outcome of acute leukemia relapsing after bone marrow transplantation: utility of second transplants and adoptive immunotherapy. Bone Marrow Transplant 1997; 19: 709–719. Mehta J, Powles R, Treleaven J et al. Autologous transplantation with CD52 monoclonal antibody-purged marrow for acute lymphoblastic leukemia: long-term follow-up. Leu Lymphoma 1997; 25: 479–486. Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18: 695–706. Klein JP, Rizzo JD, Zhang MJ, Keiding N. Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part I: unadjusted analysis. Bone Marrow Transplant 2001; 28: 909–915. 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. Drobyski WR, Klein J, Flomenberg N et al. Superior survival associated with transplantation of matched unrelated vs oneantigen-mismatched unrelated or highly human leukocyte antigen-disparate haploidentical family donor marrow grafts for the treatment of hematologic malignancies: establishing a treatment algorithm for recipients of alternative donor grafts. Blood 2002; 99: 806–814. Petersdorf EW, Gooley TA, Anasetti C et al. Optimizing outcome after unrelated marrow transplantation by comprehensive matching of HLA class I and II alleles in the donor and recipient. Blood 1998; 92: 3515–3520. Hansen JA, Gooley TA, Martin PJ et al. Bone marrow transplants from unrelated donors for patients with chronic myeloid leukemia. N Engl J Med 1998; 338: 962–968. Mehta J, Powles R, Singhal S et al. Autologous bone marrow transplantation for acute myeloid leukemia in first remission: identification of modifiable prognostic factors. Bone Marrow Transplant 1995; 16: 499–506. Powles R, Sirohi B, Treleaven J et al. The role of posttransplant maintenance chemotherapy in improving the outcome of autotransplantation in adult acute lymphoblastic leukemia. Blood 2002; 100: 1641–1647.
