Prophylaxis regimens for GVHD: systematic review and meta-analysis

Bone Marrow Transplantation (2009) 43, 643–653 & 2009 Macmillan Publishers Limited All rights reserved 0268-3369/09 $32.00

www.nature.com/bmt

ORIGINAL ARTICLE

Prophylaxis regimens for GVHD: systematic review and meta-analysis R Ram1,2,4, A Gafter-Gvili1,2,4, M Yeshurun1,2, M Paul2,3, P Raanani1,2 and O Shpilberg1,2 1

Institute of Hematology, Davidoff Center Rabin Medical Center, Beilinson Hospital, Petah-Tiqva, Israel; 2Sackler School of Medicine, Tel Aviv University, Ramat Aviv, Israel and 3Infectious Disease Unit, Rabin Medical Center, Beilinson Hospital, Petah-Tiqva, Israel

Opinions are divided regarding the best prophylactic regimen for GVHD. The aim of this study was to evaluate potential survival benefit of different prophylactic regimens for acute GVHD (aGVHD). We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) including patients undergoing Allo-SCT. We included trials that assessed the addition of MTX, compared CsA and tacrolimus and evaluated the addition of steroids. Outcomes assessed were all-cause mortality (ACM) at the longest follow-up, aGVHD, chronic GVHD, TRM, relapse rate and regimen-specific adverse events. Relative risks (RRs) with 95% confidence intervals (CIs) were estimated and pooled. The regimen of MTX–CsA vs CsA alone (four trials) yielded no statistically significant difference in ACM (RR ¼ 0.84 (0.61–1.14)), but a significant decrease in aGVHD (RR ¼ 0.52 (0.39–0.7)). There was no difference in ACM for the comparison of MTX–CsA and MTX– tacrolimus (three trials); however, MTX–tacrolimus was superior to MTX–CsA in the reduction of aGVHD (RR ¼ 0.62 (0.52–0.75)) and severe aGVHD (RR ¼ 0.67 (0.47–0.95)). The addition of steroids did not affect the outcomes (four trials). We conclude that MTX–CsA and MTX–tacrolimus are both acceptable alternatives for GVHD prophylaxis, although MTX–tacrolimus may be superior in terms of aGVHD reduction. Bone Marrow Transplantation (2009) 43, 643–653; doi:10.1038/bmt.2008.373; published online 10 November 2008 Keywords: GVHD; prophylaxis; MTX; tacrolimus

Introduction GVHD is one of the major causes of death in patients undergoing allogeneic hematopoietic SCT (Allo-HSCT). Despite prophylactic measures, the incidence of acute GVHD (aGVHD) is estimated to be 40–60% among

Correspondence: Dr P Raanani, Institute of Hematology, Rabin Medical Center, Beilinson Hospital, Petah-Tiqva 49100, Israel. E-mail: [email protected] 4 These authors equally contributed to this work Received 2 July 2008; revised 19 September 2008; accepted 25 September 2008; published online 10 November 2008

patients receiving transplants from HLA-identical sibling donors and reaches 75% in patients receiving HLAmatched unrelated (MUD) transplants.1 The incidence of chronic GVHD (cGVHD) ranges between 40 and 70% and is one of the leading causes of death in Allo-HSCT survivors, estimated to cause 20–25% of late deaths.2,3 Other major causes are relapse and infection, which are estimated to cause around 30% and 10% of deaths, respectively.3 The tight association between the degree of GVHD and TRM caused vigorous attempts to find strategies to decrease its severity.4 Several approaches have been practiced and are currently used as part of transplantation protocols. Presently, GVHD prophylaxis is based on a calcineurin inhibitor, such as CsA or tacrolimus and a short course of MTX. Other pharmacologic options include corticosteroids, mycophenolate mofetil and sirolimus in various combinations.5–8 A regimen based on MTX with a calcineurin inhibitor is considered common practice and is recommended by the EBMT WP-PD (European group for BMT-working party pediatric diseases), IBFM-SG (the international BerlinFrankfurt-Munster (BFM) study group)—a subcommittee for BMT and by the ECOG (eastern cooperative oncology group). We undertook this systematic review of randomized controlled trials (RCTs) to assemble all evidence available regarding the currently recommended regimens to elucidate whether there is a superior prophylaxis regimen, mainly in terms of all-cause mortality (ACM).

Materials and methods Data sources We conducted a comprehensive search strategy to identify both published and unpublished trials, with no restriction on language or study years. Relevant trials were identified by searching The Cochrane Library (Issue 2, 2008), PubMed (January 1966 and onwards), LILACS and CANCERLIT. We searched the following conference proceedings (from 2002): the American Society of Hematology, the American Society of Clinical Oncology and the European Hematology Association. We also searched the IBMTR (international BMT registry), EBMT and conferences of Experimental Hematology (http://www.iseh.org/

GVHD prophylaxis: meta-analysis R Ram et al

644

i4a/pages/index.cfm) for relevant abstracts. The following trial databases were searched for ongoing and unpublished trials: Current Controlled Trials in the metaRegister of controlled clinical trials (http://www.controlled-trials.com/) and the National Institutes of Health Clinical Trials Registry (http://clinicaltrials.gov/). The references of all identified studies were inspected for more trials. Additionally, the first or corresponding author of each included trial was contacted for information regarding unpublished trials or complementary information on their own trial.

Study selection We included only RCTs. We also included trials assessing patients with hematological diseases (both malignant and nonmalignant) undergoing allogeneic BMT and SCT. We chose to include the following comparisons of regimens that are currently considered common practice for GVHD prophylaxis:9 MTX-containing regimens vs the same regimens without MTX, MTX–tacrolimus regimen vs MTX– CsA regimen and steroid-containing regimens vs the same regimens without steroids. We included trials that reported the outcomes up to 5 years following transplantation. We chose a follow-up of 5 years because this time span encompasses deaths from different causes (aGVHD, cGVHD, other causes of TRM and relapse-related mortality). We used the following search term for all electronic databases: ‘graft versus host disease’ or ‘graft versus host’ or ‘GVHD.’ For PubMed, we added the Cochrane highly sensitive search term for identification of clinical trials.10 Two reviewers independently inspected each reference identified by the search and applied inclusion criteria. For possibly relevant articles, or in cases of disagreement between the two reviewers, the full article was obtained and inspected independently by the two reviewers. Data extraction and quality assessment Two reviewers independently extracted data from the included trials (RR and AG). In the event of any disagreement between the two reviewers, a third reviewer extracted the data (MP). Data extraction was discussed, decisions were documented and, where necessary, the authors of the trials were contacted for clarification and complementary information on their trials. Two reviewers independently assessed trials for methodological quality (RR and AG). We individually assessed the following methodological quality measures: allocation concealment, generation of the allocation sequence and blinding. We graded allocation concealment and generation as adequate, unclear or inadequate.10 Outcomes The primary outcome measure was ACM at the end of the study period. When long-term follow-up was reported, mortality data at the longest available time point were used for analysis. Secondary outcomes included the number of patients with grade II–IV aGVHD and severe aGVHD (grade III–IV) at 100 days, the number of patients with cGVHD at the end of study, TRM, number of patients who relapsed during the study period and number of patients Bone Marrow Transplantation

with infections (bacterial, fungal and cytomegalovirus). Adverse events including renal failure and veno-occlusive disease were also assessed.

Data synthesis and analysis Dichotomous data were analyzed by calculating the relative risk (RR) for each trial with 95% confidence intervals (CIs; review manager version 4.2). For the main outcome, ACM, we performed an intention-to-treat analysis in which we included all known events in both nominator and denominator, even if excluded from the trial’s original analysis. We assessed heterogeneity of trial results by calculating a w2-test of heterogeneity and the I2 measure of inconsistency. The Mantel–Haenszel fixed effect model was used for pooling trial results and is shown unless statistically significant heterogeneity was found (P ¼ o0.10 or I2 ¼ 450%), in which case a random effects model is presented. Correlations were assessed using the Pearson’s correlation coefficient with a two-tailed test of significance (SPSS v.14).

Results The search yielded 747 potentially relevant trials of which 36 were considered for further investigation. Of these, 20 studies were excluded for various reasons (Figure 1).7,11–29 Thirteen trials conducted between the years 1985 and 2007, and randomizing 1439 patients, fulfilled the inclusion criteria. Six trials assessed the addition of MTX to a calcineurin inhibitor (five trials assessed MTX–CsA vs CsA and one trial assessed MTX–tacrolimus vs tacrolimus).6,30–34 Five trials assessed the addition of corticosteroids to various regimens.5,6,35–37 Three trials compared tacrolimus and CsA.38–40 We also included three publications of longterm follow-up for the above trials.41–43 All trials included patients undergoing transplantation from a matched related donor (MRD) or matched unrelated donor (MUD) following the myeloablative conditioning. All patients had a malignant hematological disease. The mean duration of follow-up among all studies was 45 months (range: 24–96). Data regarding demographics, host and donor characteristics, transplantation protocol and post-transplantation data are summarized in Table 1. Seven trials were of adequate methodological quality, in terms of allocation generation and allocation concealment, and none were blinded (Table 1). Among all trials, regardless of the regimen assessed, the risk for mortality at the end of the study was significantly correlated with the risks for severe aGVHD and aGVHD (R ¼ 0.65, P ¼ 0.001, 12 study arms and R ¼ 0.60, P ¼ p0.001, 14 study arms). Data regarding infections were scarce and were analyzed only for the comparison of the addition of steroids to other regimens.

The addition of MTX to calcineurin inhibitors ACM. Overall, there was no statistically significant difference in ACM at the end of the study period (RR ¼ 0.84 (95% CI ¼ 0.65–1.08), RRo1 favor the MTX containing regimens, five trials (Figure 2a)).

GVHD prophylaxis: meta-analysis R Ram et al

645 Potentially relevant studies identified and screened for retrieval (n=747)

Excluded due to nonrandom design (n =711) RCTs retrieved for more detailed evaluation (n =36)

RCTs excluded due to (n =20): Trials comparing MTX vs. CsA monotherapy [n =6 (11-16)] Trials evaluating different durations and dosages of CsA [n =5 (17-21)] Trials evaluating different durations and dosages of MTX [n =1 (22)] Trials evaluating the addition of CsA to various regimens [n =4 (23-26)] Trial comparing between mycophenolate and other regimens [n =1 (7)] Trials comparing between corticosteroids and MTX [n =1(27)] Trials evaluating the addition of ATG to other regiments [n =1 (28)] Trials evaluating chimerism pattern between patient treated with MTX-CsA and CsA [n =1 (29)]

Publications included in the metaanalysis (n=16) RCTs included in the meta-analysis (n=13) Figure 1 Trial flow according to QUOROM (quality of reporting meta-analysis).

This was true when CsA–MTX was compared with CsA (RR ¼ 0.84 (95% CI ¼ 0.61–1.14), four trials) and for the single comparison of CsA–MTX with CsA (RR ¼ 0.83 (95% CI ¼ 0.52–1.33)). Secondary outcomes. There was a significant reduction in aGVHD in the MTX-containing regimens compared with the non-MTX-containing regimens (RR ¼ 0.49 (95% CI ¼ 0.38–0.65), six trials), with an NNT of 4 (95% CI ¼ 3—7); (Figure 2b). The four trials that compared CsA–MTX and CsA alone showed a similar significant reduction (RR ¼ 0.52 (95% CI ¼ 0.39–0.7)). There was no difference in severe aGVHD between the two arms, (RR ¼ 0.74 (95% CI ¼ 0.08–7.09), I2 ¼ 77%, random effects model). However, this was derived from only two trials reporting opposite results. There were no statistically significance differences in cGVHD (RR ¼ 0.84 (95% CI ¼ 0.7–1.02), five trials); TRM

(RR ¼ 0.75 (95% CI ¼ 0.54–1.05), five trials) or relapse rate (RR ¼ 1.37 (95% CI ¼ 0.47–3.97), I2 ¼ 73%, random effects model, three trials).

Tacrolimus-containing regimens vs CsA-containing regimens ACM. There was no difference in ACM between tacrolimus-containing regimens and CsA-containing regimens (RR ¼ 1.10 (95% CI ¼ 0.93–1.3), three trials, Figure 3). In one study with unclear randomization methods, study arms were unbalanced with respect to age, baseline disease, and pre-transplant disease status.40 After excluding this study from the analysis, there was still no difference in ACM between the two arms (RR ¼ 0.95 (95% CI ¼ 0.74–1.23), two trials). For this outcome, we conducted a subgroup analysis according to donor type. Data were too scarce to analyze in the MRD group. There was no difference in Bone Marrow Transplantation

646

Bone Marrow Transplantation

Table 1

Characteristics of included trials Number of Age in years, Base line diseases randomized mean±s.d. or (number of patients) patients median (range) Arm Arm 1 2

GVHD prophylaxis, dose and schedule

Hiraoka et al.38

Tacrolimus–MTX steroids; tacrolimus 0.075 mg p.o.twice a day or 0.05 mg/kg/day cont i.v. for 24 h CsA–MTX CsA (determined by each medical institution)

66

Tacrolimus–MTX FK-506 0.03 mg/kg i.v. from days 1 to 180

90

Nash et al.39

Ratanatharathorn et al.40

Tacrolimus–MTX Tacrolimus 0.03 mg/kg/day i.v. from day 1 to day 80

165

Lee et al.30

Mrsic et al.31

Storb et al.41

CsA–MTX MTX 15 mg/m2 i.v. on day 1 and 10 mg/m2 on days 3, 6 and 11 CsA

40

CsA–MTX MTX 15 mg/m2 i.v. on day 1 and 10 mg/m2 on days 3, 6 and 11 CsA

37

CsA–MTX MTX 15 mg/m2 i.v. on day 1 and 10 mg/m2 on days 3, 6 and 11

43

B, B

N

BMT

Myeloablative

24

B, B

N

MRD MDS (7), ALL (18), AML (45), CML (51), NHL (13), HD (2), MM (14), CLL (6), other (9) MRD MDS (13), ALL (14), AML (44), CML (57), NHL (16), HD (2), MM (9), CLL (4), other (5)

BMT

Myeloablative

24

B, B

N

35 (15–48)

AML (17), ALL (9), CML (12), MDS (2)

MRD

BMT

Myeloablative

36

B, B

N

32 (15–49)

AML (19), ALL (8), CML (10), MDS (3)

27 (5–43)

AML (16) ALL (13), CML (8)

MRD

BMT

Myeloablative

34

A, A

N

29 (6–42)

AML (14) ALL (11), CML (14) MRD

BMT

Myeloablative

48

A, A

N

NR

35 (12–54)

40 (17–61)

164

CsA–MTX CsA 3 mg/kg/day from day 1 to day 180

30

34 (13–61)

90

CsA–MTX CsA (3 mg/kg i.v. from day1 to day 180)

40

39

Duration of Allocation, Blinding follow-up generation (months) and allocation concealment

NR

40 (16–63)

NR

Leukemia CR1 (34), leukemia other (16), aplastic anemia (2), MDS (9), other (5) Leukemia, CR1 (34), leukemia other (16), aplastic anemia (4), MDS (5), other (6) AML CR1 (6) AML other (15) ALL CR1 (3) ALL other (9) CML CS (37) CML ACC (5) MDS (12), other (3) AML CR1 (4) AML other (14) ALL CR1 (4) ALL other (4) CML CS (40) CML ACC (7) MDS (10), other (7)

AML (17), CML (26)

MRD (32) MUD (30) Other (4)

Source of Conditioning stem cells regimen (BMT/SCT) (myeloablative/ nonmyeloablative)

BMT

NR

65

Type of donor (number of patients)

MRD (30) MUD (32) Other (3)

MRD (75) OTHER (15)

MRD (74) OTHER (16)

GVHD prophylaxis: meta-analysis R Ram et al

Study, year (reference)

Table 1

Continued

Study, year (reference)

GVHD prophylaxis, dose and schedule

Number of Age in years, Base line diseases randomized mean±s.d. or (number of patients) median patients (range) Arm Arm 1 2

CsA Zikos et al.33

Nash et al.4

Atkinson et al.35

CsA–MTX MTX 10 mg/m2 i.v. on day 1 and 8 mg/m2 on days 3, 6 and 11 CsA

32

Tacrolimus–MTX MTX 10 mg/m2 i.v. on day 1 and 8 mg/m2 on days 3, 6 and 11 Tacrolimus Tacrolimus– methylprednisolone (0.5 mg/kg from day 7, 1 mg/kg from day 15, tapered from day 28 and stopped on day 72)

7

CsA–MTX MTX 10 mg/m2 i.v. on day 1 and 8 mg/m2 on days 3 and 6 CsA—steroids

75

CsA–MTX–methyprednisolone i.v. 1 mg/kg on alternate days from day 0 till day 30

21

28

6

74

Ruutu et al.5

CsA–MTX–prednison (p.o. 0.5 mg/kg from day 14, 1 mg/ kg on day 21 and tapering down from day 35 and stopped on day 110) CsA–MTX

53

CsA–MTX–prednisone (1 mg/ kg i.v from day 0 to day 22, 0.5 mg/kg from day 22 to day 35 and stopped)

73

AML CR1 (26), AML, other (6)

MRD

BMT

Myeloablative

84

B, B

N

29 (14–43)

AML CR1 (23), AML other (5) ALL (7) MDS (2) Ly (8) CML (1) (all in CR1)

MRD

BMT

Myeloablative

NR

A, A

N

28 (2–48)

CML (34), AML (25) and ALL (14)

MRD

BMT

Myeloablative

70

A, A

N

32 (1–50)

CML (33), AML (25), ALL (13)

36 (15–52)

ALL CR1 (1) ALL other MRD (3) AML CR1 (2) AML other (6) CML CR1 (5) CML other (4) ALL CR1 (2) ALL other (0) AML CR1 (4) AML other (5) CML ch (5) CML other (4)

BMT

Myeloablative

30

B, B

N

MRD

BMT

Myeloablative

96

A, A

N

MRD and haploidentical donor

BMT

Myeloablative

24

A, A

N

NR NR

32 (15–50)

42 (18–54)

55

41 (17–52) 32 (5–53)

AML (31), CML (27), MDS (8), AA (7)

647

Bone Marrow Transplantation

Storb et al.36

20

Duration of Allocation, Blinding generation follow-up (months) and allocation concealment

AML (21), CML (29)

30 (11–43)

NR

5

CsA–MTX

NR

Conditioning Source of regimen stem cells (BMT/SCT) (myeloablative/ nonmyeloablative)

GVHD prophylaxis: meta-analysis R Ram et al

Chao et al.34

50

Type of donor (number of patients)

GVHD prophylaxis: meta-analysis R Ram et al

Bone Marrow Transplantation

62 CsA–methylperdnisolone (0.5 mg/kg from day 7, 1 mg/kg from day 15, titrated down from day 29 and stopped on day 72) CsA Deeg et al.37

Abbreviations: A ¼ adequate; AA ¼ aplastic anemia; AP ¼ accelerated phase; B ¼ unknown; CP ¼ chronic phase; CR ¼ complete remission; HR ¼ high risk; IT ¼ intrathecal; Lym ¼ lymphoma; MDS ¼ myelodysplastic syndrome; MRD ¼ matched related donor; MUD ¼ mached unrelated donor; N ¼ none; NR ¼ not reported; SR ¼ standard risk.

60

36 (0.9–57)

ALL (15) AML (9) CML (2) lym (21) MDS (11), other (2)

MRD ALL (13) AML (9) CML (2) lym (20) MDS (15), other (3) 39 (12–59)

AML (32), CML (30), MDS (6), AA (6) 28 (1–55) 74 CsA–MTX

GVHD prophylaxis, dose and schedule Study, year (reference)

Table 1

Continued

Number of Age in years, Base line diseases randomized mean±s.d. or (number of patients) patients median (range) Arm Arm 1 2

Type of donor (number of patients)

BMT

Myeloablative

Source of Conditioning stem cells regimen (BMT/SCT) (myeloablative/ nonmyeloablative)

41

A, A

N

Duration of Allocation, Blinding follow-up generation (months) and allocation concealment

648

ACM between the two arms for MUD patients (RR ¼ 0.86 (95% CI ¼ 0.65–1.13), two trials). Secondary outcomes. There was a significant reduction of aGVHD with MTX–tacrolimus compared with MTX–CsA (RR ¼ 0.62 (95% CI ¼ 0.52–0.75), I2 ¼ 55.3%, random effects model, three trials, NNT ¼ 5 (95% CI ¼ 4–9), Figure 3). Similarly, there was a significant reduction in severe aGVHD with MTX–tacrolimus compared with MTX–CsA (RR ¼ 0.67 (95% CI ¼ 0.47– 0.95), three trials). Sensitivity analysis excluding the above-described study40 did not affect results. Subgroup analysis revealed the same significant reduction in aGVHD with MTX–tacrolimus for the MUD patient subgroup (RR ¼ 0.65 (95% CI ¼ 0.51–0.82), two trials). There was no change in the incidence of cGVHD (RR ¼ 1.05 (95% CI ¼ 0.87–1.25), three trials; Figure 3), TRM (RR ¼ 1.00 (95% CI ¼ 0.64–1.57), I2 ¼ 69.7%, random effects model) and relapse rate (RR ¼ 1.26 (95% CI ¼ 0.92–1.71)). Sensitivity analysis did not affect results. As for adverse events, there was a significantly higher incidence of acute renal failure with MTX–tacrolimus (RR ¼ 1.20 (95% CI ¼ 1.031.39)), Figure 3.

The addition of corticosteroids to other regimens ACM. There was no difference in ACM at the end of the study period between patients receiving or not receiving corticosteroids for GVHD prophylaxis (RR ¼ 0.95 (95% CI ¼ 0.8–1.14), four trials). Secondary outcomes. There was no difference in aGVHD or severe aGVHD in the steroid-containing arm as compared with no steroids (RR ¼ 0.83 (95% CI ¼ 0.57– 1.21), I2 ¼ 52%, random effects model, five trials) and (RR ¼ 0.87 (95% CI ¼ 0.61–1.24), four trials), respectively. There was no difference in the risk for cGVHD between the two arms (RR ¼ 1.15 (95% CI ¼ 0.72–1.83), random effects model, I2 ¼ 70%, four trials). There was no change in the risk for bacterial and fungal documented infections (RR ¼ 0.85 (95% CI ¼ 0.55– 1.32) and RR ¼ 1.20 (95% CI ¼ 0.56–2.6), three trials, respectively). There was no difference in relapse rate or in the TRM between the two arms (RR ¼ 0.86 (95% CI ¼ 0.612–1.19), four trials) and (RR ¼ 1.01 (95% CI ¼ 0.71–1.42), three trials), respectively.

Discussion Our systematic review of prophylaxis regimens for patients undergoing Allo-HSCT yielded two main clinically significant results. The first is no superiority of the different prophylaxis regimens in terms of ACM. The second is the superiority of CsA–MTX over CsA and the advantage of MTX–tacrolimus over MTX–CsA in terms of aGVHD reduction. Our first comparison of calcineurin inhibitor MTX vs calcineurin inhibitor showed no significant difference in

GVHD prophylaxis: meta-analysis R Ram et al

649 Study or Subgroup

With MTX Without MTX Total Events Total Events

Weight

Risk ratio M-H, Fixed, 95% Cl

6.0% 24.5% 26.6% 11.5% 68.6%

2.20 [0.84, 5.76] 0.65 [0.39, 1.10] 0.73 [0.44, 1.20] 0.78 [0.35, 1.74] 0.84 [0.61, 1.14]

74 74

31.4% 31.4%

0.83 [0.52, 1.33] 0.83 [0.52, 1.33]

231

100.0%

0.84 [0.65, 1.08]

MTX+CsA vs. CsA Lee, BMT 2004 11 40 5 40 Mrsic 1990 13 37 21 39 Storb, Blood 1989 15 43 24 50 Zikos, Blood 1998 8 32 9 28 Subtotal (95% Cl) 152 157 Total events 47 59 Heterogeneity: Chi2 = 5.08, df = 3 (P = 0.17); I2 = 41%

Risk ratio M-H, Fixed, 95% Cl

Test for overall effect: Z = 1.12 (P = 0.26) MTX+CsA+steroids vs. CsA+steroids Chao, NEJM 1993 22 75 26 Subtotal (95% Cl) 75 Total events 22 26 Heterogeneity: Not applicable Test for overall effect: Z = 0.76 (P = 0.45) Total (95% Cl) Total events

227 69

85

Heterogeneity: Chi2 = 5.08, df = 4 (P = 0.28); I2 = 21% 0.1

Test for overall effect: Z = 1.35 (P = 0.18)

Study or Subgroup

With MTX Without MTX Events Total Events Total Weight

MTX+CsA vs. CsA Lee, BMT 2004 8 40 8 40 7.7% Mrsic 1990 10 37 30 39 28.0% Strob, Blood 1989 14 43 30 50 26.5% Zikos, Blood 1998 11 32 17 28 17.4% Subtotal (95% Cl) 152 157 79.5% Total events 43 85 Heterogeneity: Chi2 = 4.16, df = 3 (P = 0.24); I2 = 28% Test for overall effect: Z = 4.39 (P < 0.0001) MTX+CsA+steroids vs. CsA+steroids Chao, NEJM 1993 7 75 17 Subtotal (95% Cl) 75 Total events 7 17 Heterogeneity: Not applicable Test for overall effect: Z = 2.15 (P = 0.03) MTX+tacrolimus vs. tacrolimus Nash, Blood 1995 1 7 4 Subtotal (95% Cl) 7 Total events 1 4 Heterogeneity: Not applicable Test for overall effect: Z = 1.59 (P = 0.11) Total (95% Cl) Total events

234 51

Risk Ratio M-H, Fixed, 95% Cl

0.2 0.5 Favours MTX arm

1

2 5 Favours nonMTX arm

10

Risk Ratio M-H, Fixed, 95% Cl

1.00 [0.42, 2.40] 0.35 [0.20, 0.61] 0.54 [0.33, 0.88] 0.57 [0.32, 1.00] 0.52 [0.39, 0.70]

74 74

16.4% 16.4%

0.41 [0.18, 0.92] 0.41 [0.18, 0.92]

6 6

4.1% 4.1%

0.21 [0.03, 1.43] 0.21 [0.03, 1.43]

237

100.0%

0.49 [0.38, 0.65]

106

Heterogeneity: Chi2 = 5.26, df = 5 (P = 0.39); I2 = 5% Test for overall effect: Z = 5.13 (P < 0.00001)

0.1

0.2 0.5 Favours MTX arm

1

2 5 10 Favours nonMTX arm

Figure 2 (a) All-cause mortality and (b) acute GVHD for trials comparing the addition of MTX to other (mostly CsA) regimens. Studies are identified by the name of first author and year of publication and sorted by their weight. Relative risks are pooled using the fixed effect model and shown on a logarithmic scale of 0.1–10.

Bone Marrow Transplantation

GVHD prophylaxis: meta-analysis R Ram et al

650 Study or Subgroup

Tacrolimus containing Events Total

CsA containing Events Total

All cause mortality Hiraoka, BMT 2001 24 66 23 Nash, Blood 2000 41 90 45 Ratanatharathorn1998 89 165 71 Subtotal (95% Cl) 321 Total events 154 139 Heterogeneity: Chi2 = 2.72, df = 2 (P = 0.26); I2 = 26% Test for overall effect: Z = 1.12 (P = 0.26) Acute GVHD Hiraoka, BMT 2001 12 66 31 Nash, Blood 2000 46 90 63 Ratanatharathorn1998 43 165 67 Subtotal (95% Cl) 321 Total events 101 161 Heterogeneity: Chi2 = 4.47, df = 2 (P = 0.11); I2 = 55% Test for overall effect: Z = 4.89 (P < 0.00001) Severe Acute GVHD Hiraoka, BMT 2001 6 66 14 Nash, Blood 2000 16 90 23 Ratanatharathorn1998 22 165 28 Subtotal (95% Cl) 321 Total events 44 65 Heterogeneity: Chi2 = 1.38, df = 2 (P = 0.50); I2 = 0% Test for overall effect: Z = 2.22 (P = 0.03) Chronic GVHD Hiraoka, BMT 2001 26 56 26 Nash, Blood 2000 43 69 38 Ratanatharathorn1998 50 112 54 Subtotal (95% Cl) 237 Total events 119 Heterogeneity: Chi2 = 0.19, df = 2 (P = 0.91); I2 = 0% Test for overall effect: Z = 0.49 (P = 0.62) Acute renal failure Nash, Blood 2000 50 90 36 Ratanatharathorn1998 111 165 98 Subtotal (95% Cl) 255 Total events 161 134 Heterogeneity: Chi2 = 1.40, df = 1 (P = 0.24); I2 = 28% Test for overall effect: Z = 2.38 (P = 0.02)

Risk ratio M-H, Fixed, 95% Cl

65 90 164 319

1.03 [0.65, 1.62] 0.91 [0.67, 1.24] 1.25 [1.00, 1.56] 1.10 [0.93, 1.30]

65 90 164 319

0.38 [0.22, 0.68] 0.73 [0.57, 0.93] 0.64 [0.46, 0.88] 0.62 [0.52, 0.75]

65 90 164 319

0.42 [0.17, 1.03] 0.70 [0.39, 1.23] 0.78 [0.47, 1.31] 0.67 [0.47, 0.95]

55 63 132 250 118

0.98 [0.66, 1.46] 1.03 [0.79, 1.36] 1.09 [0.82, 1.46] 1.05 [0.87, 1.25]

90 164 254

1.39 [1.02, 1.90] 1.13 [0.95, 1.33] 1.20 [1.03, 1.39]

Risk ratio M-H, Fixed, 95% Cl

0.1 0.2 0.5 1 2 5 10 Favours Tacrolimus Favours CsA Figure 3

All-cause mortality, acute GVHD, severe acute GVHD, chronic GVHD and acute renal failure for trials comparing tacrolimus-based regimens to CsA-based regimens. Studies are identified by the name of first author and year of publication and sorted by their weight. Relative risks for the subcategories are pooled using the fixed effect model, on a logarithmic scale of 0.1–10.

terms of ACM. However, there was a significant reduction of aGVHD with MTX, with an NNT of 4 (95% CI ¼ 3–7). There were no significant differences between the two arms in the other outcomes that we examined. Our second comparison of tacrolimus vs CsA (both added to MTX) showed no difference in ACM between the two arms. Nevertheless, there was a significant reduction of aGVHD and severe aGVHD in the MTX–tacrolimus arm. Bone Marrow Transplantation

There were no significant differences in the other outcomes, although patients treated with MTX–tacrolimus had an increased risk for developing acute renal failure. It is noteworthy that one of the studies included in the metaanalysis was unbalanced with respect to baseline risk factors for mortality.40,44 When we eliminated this study for sensitivity analysis, results were not affected. Similar to our results, the superiority of tacrolimus-based regimens in

GVHD prophylaxis: meta-analysis R Ram et al

651

MUD transplant has also been observed in a large retrospective study,45 in which tacrolimus was found to decrease GVHD, but contrary to our study, to improve the overall survival. This superiority was not shown in MRD transplants.45 In the available RCTs, tacrolimus was superior in terms of GVHD reduction in MUD, with no significant impact on overall survival. Both tacrolimus and CsA are calcineurin inhibitors. They are biologically different as tacrolimus is associated with more pronounced inhibition of the production of IL-2 by T-lymphocytes, a major mediator in the pathogenesis of GVHD.1,46 This might be an explanation for its advantage over CsA in the reduction of GVHD. Although limited by their sample size, RCTs constitute a better platform to assess the overall survival, adjusting for risk factors that cannot be foreseen or adjusted for in observational studies. Our third comparison of the addition of corticosteroids did not show a benefit for additional corticosteroids. Several meta-analyses examined survival and reduction of GVHD in patients undergoing Allo-HSCT. The addition of immunoglobulins47 and hematopoietic growth factors48 failed to show superiority in terms of ACM and GVHD. Different myeloablative conditioning regimens conferred similar results regarding ACM and incidence of GVHD.49 There was also no change in ACM when allogeneic PBSC transplantation was compared with allogeneic BMT.50 However, PBSCTs (PBSC transplants) were found to have a survival benefit over BMTs when four instead of three doses of MTX were used for GVHD prophylaxis.51 All-cause mortality in patients undergoing Allo-HSCT results from several causes, mainly GVHD, relapse and infections.3,52 There is a complex relationship between these factors. A limited stage of aGVHD is associated with improved survival,53 whereas severe aGVHD reduces survival among patients transplanted for malignant hematological diseases. cGVHD is associated with a decreased relapse rate, presumably by its GVL effect. Relapse by itself is a major contributor for reduced survival. Thus, GVHD and relapse are opposing outcomes. Using the data available from all RCTs, we showed that aGVHD, mainly severe aGVHD, was significantly correlated with ACM at the end of study period. However, despite the reduction in aGVHD by using prophylaxis, this did not translate into a significant reduction in ACM for all comparisons. This may be explained in several ways: first, GVHD prophylaxis has never been shown to reduce mortality in patients who received prophylaxis compared with patients who did not receive it, although this was evaluated only in three small nonrandomized studies.54–56 Second, supportive and early directed treatments for established GVHD are beneficial. In fact, TRM in patients who respond to first-line therapy for aGVHD is reduced compared with patients who do not respond.9,57 The major limitation of this systematic review is the paucity of RCTs on GVHD prophylaxis. All our results claiming no difference are based on comparisons of few small studies; neither the primary studies nor the metaanalysis are powered to refute an existing difference. Our results should be read in light of this limitation. Furthermore, we could not address the role of newer drugs and strategies, such as sirolimus and mycophenolate mofetil as

well as different strategies for T-cell depletion, such as antithymocyte globulin, as part of the prophylaxis regimens for GVHD. Some of our analyses were heterogeneous. This may be explained by different populations of patients. However, pooling of results is legitimate because when the analysis did yield a statistically significant result, all studies showed a variable degree of benefit for the intervention rather than divergent results. We could not conduct subgroup analysis according to donor status (MRD or MUD transplants) for the primary outcome of ACM because of sparse data in many of the trials. This subgroup analysis is especially important for evaluating the influence of the various regimens on relapse rate and ACM in the MRD group of patients, who are prone to relapse and may not benefit from enhanced GVHD prophylaxis. Furthermore, our conclusions may not be applicable to transplant following nonmyeloablative conditioning, as timing and presentation of GVHD occurrence are different.

Implications for practice and for research. Our review showed no survival benefit for the various regimens. In terms of aGVHD reduction, the regimen of MTX– CsA is superior to CsA alone and is a good option for GVHD prophylaxis in Allo-HSCT. MTX–tacrolimus may be superior to MTX–CsA in aGVHD reduction. There is probably no role for the addition of corticosteroids to any of the regimens. Well-controlled randomized trials that will further assess the feasibility of MTX–tacrolimus in different settings of hematological transplants are needed. Clinicians can use our findings to tailor therapy for individual patients, depending on the desired balance between GVHD and graft vs tumor effect. Newer pharmacological and biological methods should be pursued to decrease GVHD while maintaining the graft vs tumor effect. All-cause mortality should be assessed as the primary end point in these future trials, because of the divergent effects of prophylaxis on GVHD and relapse rate of the primary disease.

Acknowledgements We thank the authors who responded to our letters and supplied additional data on their trials.

References 1 Davies JK, Lowdell MW. New advances in acute graft-versushost disease prophylaxis. Transfus Med 2003; 13: 387–397. 2 Lee SJ, Klein JP, Barrett AJ, Ringden O, Antin JH, Cahn JY et al. Severity of chronic graft-versus-host disease: association with treatment-related mortality and relapse. Blood 2002; 100: 406–414. 3 Bhatia S, Francisco L, Carter A, Sun CL, Baker KS, Gurney JG et al. Late mortality after allogeneic hematopoietic cell transplantation and functional status of long-term survivors: report from the Bone Marrow Transplant Survivor Study. Blood 2007; 110: 3784–3792. Bone Marrow Transplantation

GVHD prophylaxis: meta-analysis R Ram et al

652 4 Lazarus HM, Vogelsang GB, Rowe JM. Prevention and treatment of acute graft-versus-host disease: the old and the new. A report from the Eastern Cooperative Oncology Group (ECOG). Bone Marrow Transplant 1997; 19: 577–600. 5 Ruutu T, Volin L, Parkkali T, Juvonen E, Elonen E. Cyclosporine, methotrexate, and methylprednisolone compared with cyclosporine and methotrexate for the prevention of graft-versus-host disease in bone marrow transplantation from HLA-identical sibling donor: a prospective randomized study. Blood 2000; 96: 2391–2398. 6 Nash RA, Etzioni R, Storb R, Furlong T, Gooley T, Anasetti C et al. Tacrolimus (FK506) alone or in combination with methotrexate or methylprednisolone for the prevention of acute graft-versus-host disease after marrow transplantation from HLA-matched siblings: a single-center study. Blood 1995; 85: 3746–3753. 7 Bolwell B, Sobecks R, Pohlman B, Andresen S, Rybicki L, Kuczkowski E et al. A prospective randomized trial comparing cyclosporine and short course methotrexate with cyclosporine and mycophenolate mofetil for GVHD prophylaxis in myeloablative allogeneic bone marrow transplantation. Bone Marrow Transplant 2004; 34: 621–625. 8 Furlong T, Kiem HP, Appelbaum FR, Carpenter PA, Deeg HJ, Doney K et al. Sirolimus in combination with cyclosporine or tacrolimus plus methotrexate for prevention of graft-versus-host disease following hematopoietic cell transplantation from unrelated donors. Biol Blood Marrow Transplant 2008; 14: 531–537. 9 Bacigalupo A. Management of acute graft-versus-host disease. Br J Haematol 2007; 137: 87–98. 10 Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 2006. www.cochrane.org. 11 Storb R, Martin P, Deeg HJ, Sanders JE, Pepe M, Singer J et al. Long-term follow-up of three controlled trials comparing cyclosporine versus methotrexate for graft-versus-host disease prevention in patients given marrow grafts for leukemia. Blood 1992; 79: 3091–3092. 12 Torres A, Martinez F, Gomez P, Herrera C, Rojas R, GomezVillagran JL et al. Cyclosporin A versus methotrexate, followed by rescue with folinic acid as prophylaxis of acute graft-versus-host disease after bone marrow transplantation. Blut 1989; 58: 63–68. 13 Backman L, Ringdon O, Tollemar J, Lannqvist B. An increased risk of relapse in cyclosporin-treated compared with methotrexate-treated patients: long-term follow-up of a randomized trial. Bone Marrow Transplant 1988; 3: 463–471. 14 Atkinson K, Biggs JC, Concannon A, Dodds A, Downs K, Ashby M. A prospective randomised trial of cyclosporin versus methotrexate after HLA-identical sibling marrow transplantation for patients with acute leukemia in first remission: analysis 2.5 years after last patient entry. Aust NZJ Med 1988; 18: 594–599. 15 Ringden O, Backman L, Lo¨nnqvist B, Heimdahl A, Lindholm A, Bolme P et al. A randomized trial comparing use of cyclosporin and methotrexate for graft-versus-host disease prophylaxis in bone marrow transplant recipients with haematological malignancies. Bone Marrow Transplant 1986; 1: 41–51. 16 Irle C, Deeg HJ, Buckner CD, Kennedy M, Clift R, Storb R et al. Marrow transplantation for leukemia following fractionated total body irradiation. A comparative trial of methotrexate and cyclosporine. Leuk Res 1985; 9: 1255–1261. 17 Kansu E, Gooley T, Flowers ME, Anasetti C, Deeg HJ, Nash RA et al. Administration of cyclosporine for 24 months compared with 6 months for prevention of chronic graftversus-host disease: a prospective randomized clinical trial. Blood 2001; 98: 3868–3870. Bone Marrow Transplantation

18 Locatelli F, Zecca M, Rondelli R, Bonetti F, Dini G, Prete A et al. Graft versus host disease prophylaxis with low-dose cyclosporine-A reduces the risk of relapse in children with acute leukemia given HLA-identical sibling bone marrow transplantation: results of a randomized trial. Blood 2000; 95: 1572–1579. 19 Pession A, Locatelli F, Zecca M, Rondelli R, Prete A, Bonetti F et al. Cyclosporine-A as GVHD prophylaxis in allogeneic BMT for childhood acute leukemia. AIEOP-BMT group. Bone Marrow Transplant 1998; 21: S50–S52. 20 Storb R, Leisenring W, Anasetti C, Appelbaum FR, Deeg HJ, Doney K et al. Methotrexate and cyclosporine for graft-vs.host disease prevention: what length of therapy with cyclosporine? Biol Blood Marrow Transplant 1997; 3: 194–201. 21 Bacigalupo A, Van Lint MT, Occhini D, Gualandi F, Lamparelli T, Sogno G et al. Increased risk of leukemia relapse with high-dose cyclosporine A after allogeneic marrow transplantation for acute leukemia. Blood 1991; 77: 1423–1428. 22 Sullivan KM, Storb R, Buckner CD, Fefer A, Fisher L, Weiden PL et al. Graft-versus-host disease as adoptive immunotherapy in patients with advanced hematologic neoplasms. N Engl J Med 1989; 320: 828–834. 23 Sorror ML, Leisenring W, Deeg HJ, Martin PJ, Storb R. Twenty-year follow-up in patients with aplastic anemia given marrow grafts from HLA-identical siblings and randomized to receive methotrexate/cyclosporine or methotrexate alone for prevention of graft-versus-host disease. Biol Blood Marrow Transplant 2005; 11: 567–568. 24 Storb R, Leisenring W, Deeg HJ, Anasetti C, Appelbaum F, Bensinger W et al. Long-term follow-up of a randomized trial of graft-versus-host disease prevention by methotrexate/ cyclosporine versus methotrexate alone in patients given marrow grafts for severe aplastic anemia. Blood 1994; 83: 2749–2750. 25 Ringden O. Decreased transplant-related complications and improved leukemia-free survival in adults receiving methotrexate combined with cyclosporin compared with either agent alone for prevention of graft-versus-host disease. Advisory Committee of the International Bone Marrow Transplant Registry. Transplant Proc 1993; 25: 1241–1242. 26 Storb R, Deeg HJ, Pepe M, Doney K, Appelbaum F, Beatty P et al. Graft-versus-host disease prevention by methotrexate combined with cyclosporin compared to methotrexate alone in patients given marrow grafts for severe aplastic anaemia: longterm follow-up of a controlled trial. Br J Haematol 1989; 72: 567–572. 27 Gondo H, Harada M, Taniguchi S, Akashi K, Hayashi S, Teshim T et al. Cyclosporine combined with methylprednisolone or methotrexate in prophylaxis of moderate to severe acute graft-versus-host disease. Bone Marrow Transplant 1993; 12: 437–441. 28 Bacigalupo A, Lamparelli T, Bruzzi P, Guidi S, Alessandrino PE, di Bartolomeo P et al. Antithymocyte globulin for graftversus-host disease prophylaxis in transplants from unrelated donors: 2 randomized studies from Gruppo Italiano Trapianti Midollo Osseo (GITMO). Blood 2001; 98: 2942–2947. 29 McCann S, Passweg J, Bacigalupo A, Locasciulli A, Locatelli F, Ryan J et al The influence of cyclosporin alone, or cyclosporin and methotrexate, on the incidence of mixed haematopoietic chimaerism following allogeneic sibling bone marrow transplantation for severe aplastic anaemia. Bone Marrow Transplant 2007; 39: 109–114. 30 Lee KH, Choi SJ, Lee JH, Kim S, Seol M, Lee YS et al. Cyclosporine alone vs cyclosporine plus methotrexate for posttransplant immunosuppression after HLA-identical sibling bone marrow transplantation: a randomized prospective study. Bone Marrow Transplant 2004; 34: 627–636.

GVHD prophylaxis: meta-analysis R Ram et al

653 31 Mrsic M, Labar B, Bogdanic V, Nemet D, Pavletic Z, Plavsic F et al. Combination of cyclosporin and methotrexate for prophylaxis of acute graft-versus-host disease after allogeneic bone marrow transplantation for leukemia. Bone Marrow Transplant 1990; 6: 137–141. 32 Storb R, Deeg HJ, Whitehead J, Appelbaum F, Beatty P, Bensinger W et al. Methotrexate and cyclosporine compared with cyclosporine alone for prophylaxis of acute graft versus host disease after marrow transplantation for leukemia. N Engl J Med 1986; 314: 729–735. 33 Zikos P, Van Lint MT, Frassoni F, Lamparelli T, Gualandi F, Occhini D et al. Low transplant mortality in allogeneic bone marrow transplantation for acute myeloid leukemia: a randomized study of low-dose cyclosporin versus low-dose cyclosporin and low-dose methotrexate. Blood 1998; 91: 3503–3508. 34 Chao NJ, Schmidt GM, Niland JC, Amylon MD, Dagis AC, Long GD et al. Cyclosporine, methotrexate, and prednisone compared with cyclosporine and prednisone for prophylaxis of acute graft-versus-host disease. N Engl J Med 1993; 329: 1225–1230. 35 Atkinson K, Biggs J, Concannon A, Dodds A, Young S, Wilson F et al. A prospective randomised trial of cyclosporin and methotrexate versus cyclosporin, methotrexate and prednisolone for prevention of graft-versus-host disease after HLA-identical sibling marrow transplantation for haematological malignancy. Aust NZJ Med 1991; 21: 850–856. 36 Storb R, Pepe M, Anasetti C, Appelbaum FR, Beatty P, Doney K et al. What role for prednisone in prevention of acute graft-versus-host disease in patients undergoing marrow transplants? Blood 1990; 76: 1037–1045. 37 Deeg HJ, Lin D, Leisenring W, Boeckh M, Anasetti C, Appelbaum FR et al. Cyclosporine or cyclosporine plus methylprednisolone for prophylaxis of graft-versus-host disease: a prospective, randomized trial. Blood 1997; 89: 3880–3887. 38 Hiraoka A, Ohashi Y, Okamoto S, Moriyama Y, Nagao T, Kodera Y et al. Japanese FK506 BMT (Bone Marrow Transplantation) Study Group. Phase III study comparing tacrolimus (FK506) with cyclosporine for graft-versus-host disease prophylaxis after allogeneic bone marrow transplantation. Bone Marrow Transplant 2001; 28: 181–185. 39 Nash RA, Antin JH, Karanes C, Fay JW, Avalos BR, Yeager AM et al. Phase 3 study comparing methotrexate and tacrolimus with methotrexate and cyclosporine for prophylaxis of acute graft-versus-host disease after marrow transplantation from unrelated donors. Blood 2000; 96: 2062–2068. 40 Ratanatharathorn V, Nash RA, Przepiorka D, Devine SM, Klein JL, Weisdorf D et al. Phase III study comparing methotrexate and tacrolimus (prograf, FK506) with methotrexate and cyclosporine for graft-versus-host disease prophylaxis after HLA-identical sibling bone marrow transplantation. Blood 1998; 92: 2303–2314. 41 Storb R, Deeg HJ, Pepe M, Appelbaum F, Anasetti C, Beatty P et al. Methotrexate and cyclosporine versus cyclosporine alone for prophylaxis of graft-versus-host disease in patients given HLA-identical marrow grafts for leukemia: long-term follow-up of a controlled trial. Blood 1989; 73: 1729–1734. 42 Ross M, Schmidt GM, Niland JC, Amylon MD, Dagis AC, Long GD et al. Cyclosporine, methotrexate, and prednisone compared with cyclosporine and prednisone for prevention of acute graft-vs.-host disease: effect on chronic graft-vs.-host disease and long-term survival. Biol Blood Marrow Transplant 1999; 5: 285–291. 43 Sayer HG, Longton G, Bowden R, Pepe M, Storb R. Increased risk of infection in marrow transplant patients receiving

44

45

46 47

48

49

50

51

52

53

54

55

56

57

methylprednisolone for graft-versus-host disease prevention. Blood 1994; 84: 1328–1332. Horowitz MM, Przepiorka D, Bartels P, Buell DN, Zhang MJ, Fitzsimmons WE et al. Tacrolimus vs. cyclosporine immunosuppression: results in advanced-stage disease compared with historical controls treated exclusively with cyclosporine. Biol Blood Marrow Transplant 1999; 5: 180–186. Yanada M, Emi N, Naoe T, Sakamaki H, Takahashi S, Hirabayashi N et al. Tacrolimus instead of cyclosporine used for prophylaxis against graft-versus-host disease improves outcome after hematopoietic stem cell transplantation from unrelated donors, but not from HLA-identical sibling donors: a nationwide survey conducted in Japan. Bone Marrow Transplant 2004; 34: 331–337. Busuttil RW, Lake JR. Role of tacrolimus in the evolution of liver transplantation. Transplantation 2004; 77: S44–S51. Raanani P, Gafter-Gvili A, Paul M, Ben-Bassat I, Leibovici L, Shpilberg L. Immunoglobulin prophylaxis in hematopoietic stem cell transplantation—systematic review and metaAnalysis. J Clin Oncol 2008 Accepted (in press). Dekker A, Bulley S, Beyene J, Dupuis LL, Doyle JJ, Sung L. Meta-analysis of randomized controlled trials of prophylactic granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor after autologous and allogeneic stem cell transplantation. J Clin Oncol 2006; 24: 5207–5215. Hartman AR, Williams SF, Dillon JJ. Survival, disease-free survival and adverse effects of conditioning for allogeneic bone marrow transplantation with busulfan/cyclophosphamide vs total body irradiation: a meta-analysis. Bone Marrow Transplant 1998; 22: 439–443. Group Stem Cell Trialists’ Collaborative. Allogeneic peripheral blood stem cell compared with bone marrow transplantation in the management of hematologic malignancies: an individual patient data meta-analysis of nine randomized trials. J Clin Oncol 2005; 23: 5074–5087. Bensinger W. Individual patient data meta-analysis of allogeneic peripheral blood stem cell transplant vs bone marrow transplant in the management of hematological malignancies: indirect assessment of the effect of day 11 methotrexate administration. Bone Marrow Transplant 2006; 38: 539–546. Gratwohl A, Brand R, Frassoni F, Rocha V, Niederwieser D, Reusser P et al. Cause of death after allogeneic haematopoietic stem cell transplantation (HSCT) in early leukaemias: an EBMT analysis of lethal infectious complications and changes over calendar time. Bone Marrow Transplant 2005; 36: 757–769. Nevo S, Enger C, Swan V, Wojno KJ, Fuller AK, Altomonte V et al. Acute bleeding after allogeneic bone marrow transplantation: association with graft versus host disease and effect on survival. Transplantation 1999; 67: 681. Lazarus HM, Coccia PF, Herzig RH. Incidence of acute graft-versus-host disease with and without methotrexate prophylaxis in bone marrow transplant patients. Blood 1984; 64: 215–220. Elfeinbein G, Goedert T, Graham-Pole J. Is prophylaxis against acute graft-versus-host disease necessary if treatment is effective and survival not impared. Proc Am Soc Clin Oncol 1986; 5: 643 (abstract). Sullivan KM, Deeg HJ, Sanders J. Hyperacute graft-versushost disease in patients not given immunosupression after allogeneic marrow transplant. Blood 1986; 67: 1172–1175. Bacigalupo A, Oneto R, Lamparelli T, Gualandi F, Bregante S, Raiola AM et al. Pre-emptive therapy of acute graft-versushost disease: a pilot study with antithymocyte globulin (ATG). Bone Marrow Transplant 2001; 28: 1093–1096.

Bone Marrow Transplantation