Biol Blood Marrow Transplant 25 (2019) 73 85 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Comparative Analysis of Calcineurin Inhibitor Based Methotrexate and Mycophenolate Mofetil Containing Regimens for Prevention of Graft-versus-Host Disease after Reduced-Intensity Conditioning Allogeneic Transplantation D4X XSaurabh ChhabraD15X X , D6X XYing LiuD2,37X X , D8X XMichael T. HemmerD29X X , D10X XLuciano CostaD41X X , D12X XJoseph A. PidalaD13X5X , D14X XDaniel R. CourielD15X6X , D16X XAmin M. AlousiD717X X , D18X XNavneet S. MajhailD19X8X , D20X XRobert K. StuartD921X X , D2X XDennis KimD23X10 X , D24X XOlle RingdenD25X11 X , D26X XAlvaro Urbano-IspizuaD12 27X X , D28X XAyman SaadD429X X , D30X XBipin N. SavaniD13 31X X , D32X XBrenda CooperD3X14 X , D34X XDavid I. MarksD35X15 X , D36X XGerard SocieD16 37X X , D38X XHarry C. SchoutenD17 39X X , D40X XHelene SchoemansD18 41X X , D42X XHisham Abdel-AzimD43X19 X , 20 21 22 23 24 D4X XJean YaredD45X X , D46X XJean-Yves CahnD47X X , D48X XJohn WagnerD49X X , D50X XJoseph H. AntinD51X X , D52X XLeo F. VerdonckD53X X , D54X XLeslie LehmannD5X23 X , D56X XMahmoud D. AljurfD25 57X X , D58X XMargaret L. MacMillanD59X26 X , D60X XMark R. LitzowD27 61X X , D62X XMelhem M. SolhD63X X , D64X XMuna QayedD65X X , D6X XPeiman HemattiD67X X , D68X XRammurti T. KambleD31 28 29 30 69X X , D70X XRavi VijD3271X X , D72X XRobert J. HayashiD73X33 X , D74X XRobert P. GaleD34 75X X , D76X XRodrigo MartinoD35 7X X , D78X XSachiko SeoD36 79X X , D80X XShahrukh K. HashmiD25,27 81X X , D82X XTaiga NishihoriD83X X , D84X XTakanori TeshimaD85X X , D86X XUsama GergisD87X X , D8X XYoshihiro InamotoD89X X , D90X XStephen R. SpellmanD40 5 37 38 39 91X X , D92X XMukta AroraD93X41, X *, D94X XBetty K. HamiltonD895X X 1 Division of Hematology/Oncology, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin 2 Center for International Blood and Marrow Transplant Research, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin 3 Institute for Health and Society, Department of Biostatistics, Medical College of Wisconsin, Milwaukee, Wisconsin 4 Division of Hematology/Oncology, Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama 5 Department of Blood and Marrow Transplantation, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida 6 Division of Hematology and Hematologic Malignancies, Utah Blood and Marrow Transplant Program, Salt Lake City, Utah 7 Department of Stem Cell Transplantation, Division of Cancer Medicine, University of Texas MD Anderson Cancer Center, Houston, Texas 8 Blood & Marrow Transplant Program, Cleveland Clinic Taussig Cancer Institute, Cleveland, Ohio 9 Division of Hematology/Oncology, Department of Medicine, Medical University of South Carolina, Charleston, South Carolina 10 Department of Medical Oncology & Hematology, Princess Margaret Cancer Center, Toronto, Canada 11 Division of Therapeutic Immunology, Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden 12 Department of Hematology, IDIBAPS, Institute of Research Josep Carreras, Hospital Clínic, University of Barcelona, Barcelona, Spain 13 Division of Hematology/Oncology, Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee 14 Department of Medicine-Hematology and Oncology, University Hospitals Case Medical Center, Cleveland, Ohio 15 Adult Bone Marrow Transplant, University Hospitals Bristol NHS Trust, Bristol, United Kingdom 16 Department of Hematology, Ho ^pital Saint Louis, Paris, France 17 Department of Hematology, Academische Ziekenhuis, Maastricht, the Netherlands 18 Department of Hematology, 26-Department of Pediatrics, University Hospital of Leuven, Leuven, Belgium 19 Division of Hematology, Oncology and Blood & Marrow Transplantation, Children's Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, California 20 Blood & Marrow Transplantation Program, Division of Hematology/Oncology, Department of Medicine, Greenebaum Cancer Center, University of Maryland, Baltimore, Maryland 21 Department of Hematology, CHU Grenoble Alpes, Grenoble, France 22 Department of Medical Oncology, Thomas Jefferson University, Philadelphia, Pennsylvania 23 Division of Hematologic Malignancies, Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts 24 Departmentt of Hematology/Oncology, Isala Clinic, Zwolle, the Netherlands 25 Department of Oncology, King Faisal Specialist Hospital Center & Research, Riyadh, Saudi Arabia 26 University of Minnesota Blood and Marrow Transplant Program, Minneapolis, Minnesota 27 Division of Hematology and Transplant Center, Mayo Clinic Rochester, Rochester, Minnesota 28 Blood and Marrow Transplant Group of Georgia, Northside Hospital, Atlanta, Georgia 29 Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia 30 Division of Hematology/Oncology/Bone Marrow Transplantation, Department of Medicine, University of Wisconsin Hospital and Clinics, Madison, Wisconsin 31 Division of Hematology and Oncology, Center for Cell and Gene Therapy, Baylor College of Medicine, Houston, Texas 32 Division of Hematology and Oncology, Washington University School of Medicine, St. Louis, Missouri * Correspondence and reprint requests: Mukta Arora, MD, Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minne- sota Medical Center, Minneapolis, MN Phone: 612-625-8942 E-mail address: schhabra@mcw.edu (S. Chhabra), arora005@umn.edu (M. Arora). https://doi.org/10.1016/j.bbmt.2018.08.018 1083-8791/© 2018 American Society for Blood and Marrow Transplantation. 74 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 33 Division of Pediatric Hematology/Oncology, Department of Pediatrics, Washington University School of Medicine in St. Louis, St. Louis, Missouri 34 Hematology Research Centre, Division of Experimental Medicine, Department of Medicine, Imperial College London, London, United Kingdom 35 Divison of Clinical Hematology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain 36 Department of Hematology & Oncology, National Cancer Research Center East, Chiba, Japan 37 Department of Hematology Kyushu University Hospital, Sapporo, Japan 38 Hematolgic Malignancies & Bone Marrow Transplant, Department of Medical Oncology, New York Presbyterian Hospital/Weill Cornell Medical Center, New York, New York 39 Hematopoietic Stem Cell Transplantation, National Cancer Center Hospital, Tokyo, Japan 40 Center for International Blood and Marrow Transplant Research, National Marrow Donor Program/Be The Match, Minneapolis, Minnesota 41 Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minnesota Medical Center, Minneapolis, Minnesota Article history: A B S T R A C T Received 7 June 2018 The combination of a calcineurin inhibitor (CNI) such as tacrolimus (TAC) or cyclosporine (CYSP) with methotrex- Accepted 9 August 2018 ate (MTX) or with mycophenolate mofetil (MMF) has been commonly used for graft-versus-host disease (GVHD) prophylaxis after reduced-intensity conditioning (RIC) allogeneic hematopoietic cell transplantation (alloHCT), Keywords: but there are limited data comparing efficacy of the 2 regimens. We evaluated 1564 adult patients who under- Calcineurin inhibitor Tacrolimus went RIC alloHCT for acute myelogenous leukemia (AML) and acute lymphoblastic leukemia (ALL), chronic mye- Cyclosporine logenous leukemia (CML), and myelodysplastic syndrome (MDS) from 2000 to 2013 using HLA-identical sibling Methotrexate (matched related donor [MRD]) or unrelated donor (URD) peripheral blood graft and received CYSP or TAC with Mycophenolate mofetil Graft-versus-host disease MTX or MMF for GVHD prophylaxis. Primary outcomes of the study were acute and chronic GVHD and overall sur- prophylaxis vival (OS). The study divided the patient population into 4 cohorts based on regimen: MMF-TAC, MMF-CYSP, Reduced-intensity conditioning MTX-TAC, and MTX-CYSP. In the URD group, MMF-CYSP was associated with increased risk of grade II to IV acute Allogeneic hematopoietic cell GVHD (relative risk [RR], 1.78; P < .001) and grade III to IV acute GVHD (RR, 1.93; P = .006) compared with MTX- transplantation TAC. In the URD group, use of MMF-TAC (versus MTX-TAC) lead to higher nonrelapse mortality. (hazard ratio, 1.48; P = .008). In either group, no there was no difference in chronic GVHD, disease-free survival, and OS among the GVHD prophylaxis regimens. For RIC alloHCT using MRD, there are no differences in outcomes based on GVHD prophylaxis. However, with URD RIC alloHCT, MMF-CYSP was inferior to MTX-based regimens for acute GVHD prevention, but all the regimens were equivalent in terms of chronic GVHD and OS. Prospective studies, tar- geting URD recipients are needed to confirm these results. © 2018 American Society for Blood and Marrow Transplantation. INTRODUCTION Currently, at most transplant centers, GVHD prophylaxis is Although the development of reduced-intensity condi- largely based on CNI (CYSP or TAC) in combination with short- tioning (RIC) or nonmyeloablative conditioning has allowed course MTX or MMF [14,15]. There is, however, significant var- patients who are ineligible for myeloablative conditioning iability among centers in GVHD prophylaxis regimens used. In (MAC) allogeneic hematopoietic cell transplantation RIC alloHCT, MMF is widely used instead of MTX in combina- (alloHCT) to have access to this potentially curative ther- tion with a calcineurin inhibitor for GVHD prophylaxis[15], apy, nonrelapse mortality (NRM) remains a significant given the advantages of earlier engraftment and less mucositis. obstacle to its success [1]. The tight association between A recent Center for International Blood and Marrow Transplan- graft-versus-host disease (GVHD) and NRM has led to tation Research (CIBMTR) analysis [16] demonstrated signifi- attempts to devise GVHD prevention strategies to decrease cantly worse overall survival (OS), NRM, and aGVHD and its incidence and severity [2]. For the past 3 decades, the chronic GVHD (cGVHD) with MMF compared with MTX after regimen pioneered by the Seattle group combining a calci- RIC alloHCT from unrelated donors (URD). With the 4 regimens neurin inhibitor (CNI) with methotrexate (MTX) has been of MTX-CYSP, MMF-CYSP, MTX-TAC and MMF-TAC being used the most widely adopted for GVHD prevention [3]. Cyclo- frequently as the current standard to prevent GVHD after RIC sporine (CYSP) in combination with a short course of MTX alloHCT, an important and unanswered question is whether 1 has been widely used since the late 1980s [4,5]. In more of them is superior to others in preventing GVHD. We aimed to recent years, tacrolimus (TAC) has emerged as an alterna- describe and evaluate the comparative efficacy of the 4 com- tive to CYSP for GVHD prophylaxis [6]. CYSP and TAC share monly used regimens in a large cohort of patients using the a final common pathway of inhibition of IL-2 mediated T CIBMTR database. cell expansion and cytotoxicity [7]. Randomized trials have shown that post-transplantation TAC-MTX is associated METHODS Data Sources with decreased acute GVHD (aGVHD) compared with CYSP- The CIBMTR is a combined research program of the Medical College of MTX in patients with a matched sibling donor (matched Wisconsin and the National Marrow Donor Program, which consists of a vol- related donor [MRD]) [8] or matched unrelated donor untary network of more than 450 transplantation centers worldwide that (URD) in the myeloablative setting [9]. There are, however, contribute detailed data on consecutive allogeneic and autologous transplan- several caveats with the use of MTX, mainly delayed tations to a centralized statistical center. Observational studies conducted by the CIBMTR are performed in compliance with all applicable federal regula- hematopoietic engraftment, increased oral mucositis, and tions pertaining to the protection of human research participants. Protected gastrointestinal toxicity, as well as pulmonary and renal health information issued in the performance of such research is collected toxicity[2,10-12]. In patients undergoing RIC alloHCT, and maintained in the CIBMTR's capacity as a Public Health Authority under reducing procedure-related toxicities may be of critical the Health Insurance Portability and Accountability Act Privacy Rule. importance, as these patients usually have greater comor- Patients bidities[1]. Therefore, to reduce MTX-associated toxicities, The study population included adult patients with acute myelogenous mycophenolate mofetil (MMF) has been investigated as a leukemia (AML), acute lymphoblastic leukemia (ALL), chronic myelogenous replacement for MTX in RIC regimens in recent years [13]. leukemia (CML), and myelodysplastic syndrome (MDS) who underwent a S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 75 first RIC alloHCT between 2000 and 2013 from an HLA-identical sibling or an problem. On the other hand, for the URD group, to detect a hazard ratio of 1.5 in 8/8- or 7/8-matched unrelated donor (HLA-A, -B, -C, -DRB1) [17] and received 1 of the 6 pairwise comparisons among treatments with significance level .05, a combination of CNI (CYSP or TAC) and either MTX or MMF for prophylaxis the power ranges from 51% to 90%. against GVHD. Haploidentical related donor and cord blood transplants were excluded. All patients received a peripheral blood graft. Patients receiving ex vivo T cell depletion and bone marrow grafts were excluded given their small RESULTS numbers. Patient, Disease, and Transplantation Characteristics In the MRD group (n = 690), patient, disease, and trans- Study Endpoints plantation characteristics showed important differences The primary objective of the study was to evaluate and compare the risks (Table 1). The median age at alloHCT was as low as 53 years of aGVHD, cGVHD, and overall mortality with each of the 4 GVHD prophylaxis in MTX-CYSP and as high as 61 years in MMF-TAC cohort (P regimens in RIC alloHCT patients receiving peripheral blood graft. The pri- mary endpoints of the study, therefore, were grade II to IV and III to IV < .001). A significantly lower proportion of MTX-CYSP aGVHD, cGVHD, and OS. OS was defined as the time from alloHCT to death cohort patients had KPS of <90% (19%), whereas TAC-based from any cause or until last follow-up. Death from any cause was considered cohorts had higher proportions of patients with KPS <90% an event. Surviving patients were censored at last follow-up. Secondary end- (45% to 52%; P < .001). D96XAML X was the most common alloHCT points included absolute neutrophil count recovery, platelet recovery, dis- ease-free survival (DFS), relapse and NRM. Patients were censored at indication (48% to 69%) in the MRD cohorts, followed by subsequent transplant or date of last follow up. DFS was defined as time from MDS (24% to 45%; P < .001). Donor-recipient CMV serostatus alloHCT to either relapse/progression or death from any cause. Patients alive proportions were heterogeneous; for example, 11% of MTX- were censored at the time of relapse/progression or last follow-up, whichever CYSP patients were donor-recipient seronegative, compared came first. NRM was defined as death from any cause in continuous remission and was summarized by cumulative incidence estimate with relapse as com- with 19% to 25% in the other 3 cohorts (P < .001). The com- peting risk. Relapse was defined as molecular, cytogenetic, or morphologic bination of ATG with alkylator (Bu, melphalan [Mel]), nucle- evidence of disease recurrence. Relapse was summarized by cumulative inci- oside analog (fludarabine [Flu]), or TBI-based conditioning dence estimate with NRM as the competing risk. For relapse and NRM, regimen was used in approximately a quarter of all 4 patients in continuous complete remission were censored at last follow-up. cohorts of MRD group. While in the 2000 to 2004 period, aGVHD and cGVHD were defined by the standard criteria[18,19]. For GVHD, death without the event was considered a competing risk. All patients 43% of MTX-CYSP and 50% of MMF-CYSP patients received received RIC/nonmyeloablative conditioning, which was defined as total alloHCT, and in the most recent 2009 to 2013 period, 17% body irradiation (TBI) 5 Gy (single dose) or 8 Gy (fractionated), or busulfan and 29% of the respective CYSP cohorts had alloHCT. In con- (Bu) 8 mg/kg (orally) or 6.4 mg/kg (intravenously) or melphalan <150 mg/ trast, 13% of MTX-TAC and 19% of MMF-TAC received m2 [20]. alloHCT in 2000 to 2004, and in the 2009 to 2013 period, 57% and 56% patients in the respective TAC cohorts under- Statistical Analysis This is a retrospective cohort study describing and comparing outcomes went alloHCT (P < .001). The median follow-up of survivors after RIC alloHCT using MTX+CNI (TAC versus CYSP) versus MMF+CNI (TAC ranged from 48 to 59 months in the MRD cohorts. versus CYSP) as GVHD prophylaxis. To understand the impact of prophylaxis In the URD group (n = 874), pretransplant variables were regimen on outcomes after alloHCT, the patient population was divided into similar among the 4 cohorts, with some exceptions 4 cohorts depending on CNI used and whether it was combined with MMF or MTX: MMF-TAC, MMF-CYSP, MTX-TAC, and MTX-CYSP. Furthermore, 2 sepa- (Table 2). AML (54% to 60%) and MDS (26% to 35%) were rate analyses were performed: 1 for the group with MRD and the other for the 2 most common indications for alloHCT in the URD the URD group. group (P = .003). In the URD group, 70% to 85% patients in The outcomes studied were aGVHD (grade II to IV, grade III to IV) and the 4 cohorts were fully matched (8/8-) and 15% to 30% cGVHD, OS, DFS, relapse, and NRM. Categorical variables were summarized as frequency counts and percentages and compared among GVHD prophylaxis were matched at 7/8 loci with their donors (P = .001). ATG cohorts using the chi-square test. Continuous variables were summarized as the was used in 41% of each of the 2 MMF cohorts, but at a median and range and compared using the Mann-Whitney test. Probabilities of higher frequency in the MTX cohorts (62% of MTX-CYSP OS and DFS were calculated using Kaplan-Meier estimator and compared among and 54% of MTX-TAC cohorts) of the URD group (P < .001). the cohorts using the log-rank test. Probabilities of NRM, relapse and cGVHD were calculated by cumulative incidence function accounting for competing The median follow-up of survivors ranged from 49 to 61 risks. Comparisons of cumulative incidence across time cohorts were performed months in the URD cohorts. In the earliest period of 2000 via Gray's test. Multivariate Cox proportional hazards regression models for all to 2004, 17% and 18% of MTX-CYSP and MMF-CYSP cohorts the endpoints (aGVHD, cGVHD, OS, DFS, relapse, NRM, graft failure) were used and 4% and 11% of MTX-TAC and MMF-TAC cohorts, respec- to compare the treatment groups. The assumption of proportional hazards for tively, received alloHCT (P < .001). The proportions of each factor in the Cox model was tested using time-dependent covariates. There is no variable violating the proportional hazard assumption in this study. Step- alloHCT recipients in the most recent 2009 to 2014 were wise selection was used to identify significant covariates that influenced out- 28% and 38% in the CYSP cohorts and 58% and 43% in the comes to be included in the final model to get the adjusted treatment effects, TAC cohorts, respectively. the variables we considered in the variable selection included patient-related (age, sex, Karnofsky Performance Status [KPS], ethnicity, HCT-comorbidity index, disease-related [disease, disease status at alloHCT], donor-related [donor age Acute GVHD (URD only), donor-recipient sex match and cytomegalovirus [CMV] match) and Univariate analysis demonstrated that in the MRD group, transplantation-related (year of transplant, in vivo T cell depletion using antithy- the cumulative incidences of grade II to IV aGVHD at day 100 mocyte globulin [ATG] and use of TBI in the conditioning) variables. Statistical significance of the main effects was tested with the .01 level accounting for mul- post-transplant were 27% (95% CI, 21% to 33%) in the MTX- tiple comparisons across the endpoints. Potential interactions between the main CYSP cohort and 39% (95% CI, 30% to 48%) in the MMF-CYSP effect (GVHD prophylaxis) and significant adjusting were tested, and there are cohort (Table 3). In the MTX-TAC and MMF-TAC cohorts of the no significant interactions at .01 level. Adjusted survival curves and cumulative MRD group, however, the incidences were 21% (95% CI, 17% to incidence curves were generated stratified on the treatment groups and weighted averages of covariate values using the pooled sample proportion as 26%) and 29% (95% CI, 18% to 40%), respectively. Univariate the weight function. These adjusted curves represent likelihood of outcomes in analysis also showed that the cumulative incidences of grade populations with similar prognostic factors. The following power analyses were III to IV aGVHD at day 100 were 8% (95% CI, 5% to 12%), 18% conducted for the main outcome grade II to IV aGVHD, given the current patient (95% CI, 12% to 26%), 8% (95% CI, 5% to 11%), and 14% (95% CI, and event number in each GVHD prophylaxis cohort of the 2 groups. For the 7% to 24%) in the MTX-CYSP, MMF-CYSP, MTX-TAC, and MMF- MRD group, to detect a hazard ratio (HR) of 1.5 in 1 of the 6 pairwise compari- sons among treatments with significance level .05, the power ranges from 27% TAC cohorts of the MRD group, respectively (Table 3). Multi- to 53% with Bonferroni adjustment used to adjust the multiple comparison variate analysis did not show any significant difference in the 76 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 Table 1 Characteristics of adult patients receiving their first RIC alloHCT for AML, ALL, CML, and MDS with a peripheral blood stem cell graft from an MRD and treated with TAC/CYSP + MTX/MMF § ATG from 2000 to 2013 MTX-CYSP MMF-CYSP MTX-TAC MMF-TAC P Value Patients 220 110 296 64 Centers 49 30 58 23 Patient related Age at transplant, yr 53 (20-71) 58 (20-76) 60 (19-77) 61 (19-73) <.001* Age at transplant <.001* 18-29 yr 10 (5) 3 (3) 5 (2) 4 (6) 30-39 yr 34 (15) 4 (4) 14 (5) 3 (5) 40-49 yr 42 (19) 13 (12) 26 (9) 3 (5) 50-64 yr 108 (49) 59 (54) 184 (62) 37 (58) 65+ yr 26 (12) 31 (28) 67 (23) 17 (27) Ethnicity <.001* Caucasian 120 (54) 94 (85) 264 (89) 57 (89) Non-Caucasian 72 (33) 9 (8) 26 (9) 7 (11) Sorror comorbidity index <.001* before 2007 170 (77) 68 (61) 93 (31) 21 (33) 0-1 25 (11) 17 (15) 87 (29) 10 (16) 2+ 20 (9) 24 (22) 113 (38) 33 (52) KPS <.001* <90 43 (19) 40 (36) 154 (52) 29 (45) 90 171 (78) 69 (63) 138 (47) 35 (55) Disease related Disease <.001* AML 111 (50) 55 (50) 203 (69) 31 (48) ALL 15 (6) 10 (9) 8 (3) 3 (5) CML 43 (20) 8 (7) 11 (4) 1 (2) MDS 52 (24) 37 (34) 74 (25) 29 (45) Disease status at transplant* <.001* Early 109 (50) 62 (56) 189 (64) 37 (58) Intermediate 64 (29) 14 (13) 40 (14) 8 (13) Advanced 44 (20) 29 (26) 57 (19) 17 (27) Donor related Donor-recipient gender match .11 M/M 70 (32) 35 (32) 92 (31) 28 (44) M/F 64 (29) 25 (22) 59 (20) 10 (16) F/M 42 (19) 28 (25) 85 (29) 15 (23) F/F 44 (20) 23 (21) 60 (20) 11 (17) Donor-recipient CMV status <.001* +/+ 138 (63) 42 (38) 124 (42) 30 (47) +/ 25 (11) 13 (12) 30 (10) 5 (8) /+ 27 (12) 27 (25) 78 (26) 15 (23) / 24 (11) 28 (25) 56 (19) 14 (22) Transplant related Conditioning regimen and ATG <.001* Bu + Flu § others 69 (31) 18 (16) 149 (50) 18 (28) Flu + Mel § others 72 (33) 36 (33) 69 (23) 18 (28) TBI § Cy § Flu § others 26 (12) 30 (27) 17 (6) 12 (19) ATG § Bu § Flu § others 50 (23) 11 (10) 60 (20) 15 (23) ATG § TBI § others 3 (1) 15 (14) 1 (<1) 1 (2) Year of transplant <.001* 2000-2004 110 (50) 47 (43) 39 (13) 12 (19) 2005-2008 73 (33) 31 (28) 88 (30) 16 (25) 2009-2013 37 (17) 32 (29) 169 (57) 36 (56) Follow-up of survivors, mo 54 (2-138) 59 (3-166) 56 (3-154) 48 (23-124) Values are n, median (range), or n (%). M indicates male; F, female. * statistical significance. cumulative incidences of grade II to IV and grade III to IV (95% CI, 10% to 17%), and 21% (95% CI, 16% to 27%) in the MTX- aGVHD among the 4 cohorts in the MRD group (Table 4, CYSP, MMF-CYSP, MTX-TAC, and MMF-TAC cohorts of the Figure 1A). URD group on univariate analysis, respectively (Table 5). Mul- The URD cohorts had a higher cumulative incidence of tivariate analysis of the URD group demonstrated that MMF- grade II to IV aGVHD on day 100 post-alloHCT on univariate CYSP resulted in increased incidence of grade II to IV aGVHD analysis: 32% (95%CI, 22% to 45%) and 53% (95% CI, 45% to compared with MTX-TAC (HR, 1.78; P < .001) (Table 6, 61%) in the MTX-CYSP and MMF-CYSP cohorts, respectively, Figure 1B) and MTX-CYSP (HR, 2.23; P < .001) (not shown in and 37% (95% CI, 32% to 41%) and 47% (95% CI, 41% to 54%) Table 6). A significantly higher incidence of grade III to IV using MTX-TAC and MMF-TAC, respectively (Table 5). The aGVHD was shown in the URD group with the GVHD prophy- cumulative incidences of grade III to IV aGVHD at day 100 laxis of MMF-CYSP compared with MTX-TAC (HR, 1.93; were 15% (95% CI, 8% to 25%), 21% (95% CI, 15% to 28%), 13% P = .006) (Table 6). S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 77 Table 2 Characteristics of adult patients receiving their first RIC alloHCT for AML, ALL, CML, and MDS from a URD with a peripheral blood stem cell graft and treated with CNI (CYSP/TAC) + MTX/MMF + ATG from 2000 to 2013 MTX-CYSP MMF-CYSP MTX-TAC MMF-TAC P Value Patients 71 153 432 218 Centers 26 35 68 39 Patient related Age at transplant, yr 59 (23-74) 62 (21-76) 61 (18-76) 60 (20-79) .08 Age at transplant .16 18-29 yr 3 (4) 9 (6) 18 (4) 8 (3) 30-39 yr 4 (6) 7 (5) 23 (5) 20 (9) 40-49 yr 9 (13) 12 (8) 36 (8) 23 (11) 50-64 yr 40 (56) 67 (44) 226 (52) 113 (52) 65+ yr 15 (21) 58 (38) 131 (30) 54 (25) Ethnicity <.001* Caucasian 60 (85) 138 (90) 410 (95) 204 (93) Non-Caucasian 6 (8) 10 (7) 21 (5) 12 (6) Missing 5 (7) 5 (3) 1 (<1) 2 (<1) Sorror comorbidity index <.001* before 2007 46 (65) 75 (49) 129 (30) 98 (45) 0-1 14 (20) 29 (19) 108 (25) 40 (18) 2+ 10 (14) 42 (27) 188 (43) 75 (34) Missing 1 (1) 7 (5) 7 (2) 5 (2) KPS at transplant .07 <90 20 (28) 56 (37) 196 (45) 100 (46) 90 47 (66) 93 (61) 223 (52) 112 (51) Missing 5 (6) 4 (3) 13 (3) 6 (3) Disease related Disease .003* AML 39 (55) 92 (60) 248 (57) 120 (55) ALL 3 (4) 12 (8) 9 (2) 15 (7) CML 9 (13) 9 (6) 23 (6) 7 (3) MDS 20 (28) 40 (26) 152 (35) 76 (35) Disease status at transplant* .39 Early 40 (56) 95 (62) 255 (59) 125 (57) Intermediate 9 (13) 16 (10) 71 (16) 33 (14) Advanced 19 (26) 40 (26) 97 (22) 52 (24) Missing 3 (4) 2 (1) 9 (2) 10 (5) Donor-related Donor type .001* Unrelated, 8/8-matched 50 (70) 120 (78) 369 (85) 163 (75) Unrelated, 7/8-matched 21 (30) 33 (22) 63 (15) 55 (25) Unrelated donor age at transplant, yr 33 (20-60) 32 (20-56) 31 (18-61) 33 (18-60) .25 Unrelated donor age at transplant .27 18-29 yr 34 (49) 75 (49) 235 (54) 106 (49) 30-49 yr 25 (35) 66 (43) 152 (35) 94 (43) 50-60 yr 6 (8) 8 (5) 27 (6) 9 (4) Missing 6 (8) 4 (3) 18 (4) 9 (4) Donor-recipient sex match .06 M/M 25 (35) 56 (37) 201 (47) 83 (38) M/F 25 (35) 39 (25) 121 (28) 58 (27) F/M 9 (13) 32 (21) 54 (13) 35 (16) F/F 12 (17) 26 (17) 56 (13) 42 (19) Donor-recipient CMV status .82 +/+ 12 (17) 42 (27) 102 (24) 50 (23) +/ 6 (8) 16 (10) 43 (10) 16 (7) /+ 30 (42) 52 (34) 175 (41) 89 (41) / 22 (31) 38 (25) 101 (23) 57 (26) Missing 1 (1) 5 (3) 12 (3) 6 (3) Transplant related Conditioning regimen and ATG <.001* Bu + Flu § others 11 (15) 13 (9) 121 (28) 30 (14) Flu + Mel § others 13 (18) 20 (13) 67 (16) 48 (22) TBI § Cy § Flu § others 3 (4) 57 (37) 9 (2) 49 (22) ATG § Bu § Flu § others 41 (58) 52 (34) 235 (54) 88 (40) ATG § TBI § others 3 (4) 11 (7) 0 (0) 3 (1) Year of transplant <.001* 2000-2004 12 (17) 28 (18) 17 (4) 24 (11) 2005-2008 39 (55) 67 (44) 164 (38) 100 (46) 2009-2013 20 (28) 58 (38) 251 (58) 94 (43) Follow-up of survivors, mo 60 (4-120) 71 (16-191) 48 (5-126) 61 (12-146) Values are n, median (range), or n (%). * statistical significance. 78 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 Chronic GVHD significant difference in the incidence of cGVHD among the 4 The 1-year cumulative incidences of cGVHD with MTX- URD cohorts on multivariate analysis (Table 6, Figure 2B). CYSP, MMF-CYSP, MTX-TAC, and MMF-TAC were 49% (95% CI, 42% to 56%), 39% (95% CI, 29% to 49%), 34% (95% CI, 29% to 40%), Overall Survival and 34% (95% CI, 23% to 46%), respectively, on univariate analy- The MRD cohort exhibited 2-year OS of 59% (95% CI, 52% to sis of the MRD group (Table 3). Multivariate analysis did not 66%) in the MTX-CYSP cohort, 46% (95% CI, 37% to 56%) in the reveal any significant difference in the incidence of cGVHD MMF-CYSP cohort, 48% (95% CI, 42% to 54%) in the MTX-TAC among the 4 cohorts in MRD group (Table 4, Figure 2A). In this cohort and 47% (95% CI, 35% to 59%) in the MMF-TAC cohort, group, the addition of ATG to Flu/Bu conditioning was associ- on univariate analysis (Table 3). Multivariate analysis was ated with lower cGVHD incidence (HR, .55; P = .001). unrevealing for a statistically significant difference among the Univariate analysis demonstrated 1-year cumulative inci- 4 MRD cohorts (Table 4, Figure 3A). dences of cGVHD with MTX-CYSP, MMF-CYSP, MTX-TAC, and The unadjusted probabilities of 2-year OS were 40% (95% CI, MMF-TAC of 36% (95% CI, 25% to 47%), 50% (95% CI, 42% to 28% to 52%), 45% (95% CI, 37% to 53%), 47% (95% CI, 42% to 51%) 58%), 40% (95% CI, 35% to 45%), and 44% (95% CI, 37% to 50%), and 41% (95% CI, 35% to 48%) in the MTX-CYSP, MMF-CYSP, respectively, in the URD group (Table 5). There was no MTX-TAC and MMF-TAC cohorts of the URD group, Table 3 Univariate analyses (MRD) MTX-CYSP MMF-CYSP MTX-TAC MMF-TAC (n = 220) (n = 110) (n = 296) (n = 64) Outcomes Probability (95% CI) (%) Probability (95% CI) (%) Probability (95% CI) (%) Probability (95% CI) (%) P Value aGVHD II-IV 1 mo 14 (9-19) 20 (13-28) 7 (4-10) 21 (12-32) <.001 2 mo 21 (15-26) 33 (25-42) 15 (11-20) 24 (14-35) .003 100 d 27 (21-33) 39 (30-48) 21 (17-26) 29 (18-40) .01 aGVHD III-IV 1 mo 2 (1-5) 11 (6-18) 2 (1-4) 14 (7-24) .02 2 mo 6 (3-9) 16 (9-23) 6 (4-9) 14 (7-24) .03 100 d 8 (5-12) 18 (12-26) 8 (5-11) 14 (7-24) .04 cGVHD 100 d 6 (3-10) 3 (1-7) 2 (1-4) 3 (0-9) .24 6 mo 30 (24-37) 26 (18-35) 18 (13-22) 18 (9-28) .01 1 yr 49 (42-56) 39 (29-49) 34 (29-40) 34 (23-46) .01 OS 100 d 91 (87-94) 87 (80-93) 93 (90-96) 83 (73-91) .08 6 mo 80 (75-85) 75 (67-83) 78 (73-83) 69 (57-79) .31 1 yr 67 (61-74) 57 (48-66) 60 (54-66) 55 (42-67) .13 2 yr 59 (52-66) 46 (37-56) 48 (42-54) 47 (35-59) .05 DFS 100-d 82 (76-87) 76 (67-83) 76 (71-81) 71 (60-82) .22 6 mo 68 (61-74) 61 (51-70) 61 (55-66) 54 (42-66) .17 1 yr 57 (50-64) 46 (37-56) 50 (45-56) 48 (35-60) .22 2 yr 50 (43-57) 41 (32-50) 41 (36-47) 44 (32-57) .21 Relapse 100 d 10 (6-14) 17 (10-24) 21 (16-25) 14 (7-24) .005 6 mo 18 (13-23) 26 (18-35) 32 (27-38) 32 (21-44) <.001 1 yr 24 (19-30) 34 (25-43) 37 (31-43) 33 (22-46) .007 2 yr 28 (22-35) 36 (27-46) 43 (37-48) 33 (22-46) .01 NRM 100-d 9 (5-13) 7 (3-13) 3 (2-6) 14 (7-24) .04 6 mo 15 (10-20) 13 (7-20) 7 (4-10) 14 (7-24) .02 1 yr 19 (14-24) 20 (13-28) 13 (9-17) 19 (10-30) .15 2 yr 21 (16-27) 23 (15-32) 16 (12-21) 22 (13-33) .29 GRFS 100 d 71 (64-77) 61 (52-70) 69 (63-74) 64 (52-75) .36 6 mo 39 (32-46) 35 (26-44) 42 (37-48) 39 (27-51) .53 1 yr 16 (11-22) 16 (9-24) 22 (18-27) 23 (13-34) .23 2 yr 13 (9-18) 13 (7-20) 15 (11-19) 13 (6-22) .94 CRFS 100 d 79 (73-84) 75 (67-83) 74 (69-79) 70 (59-81) .49 6 mo 56 (49-62) 49 (39-58) 53 (47-58) 50 (37-62) .67 1 yr 40 (33-47) 30 (22-40) 39 (34-45) 40 (28-52) .35 2 yr 35 (29-42) 27 (19-36) 31 (25-36) 34 (23-46) .52 Neutrophil recovery 14 d 36 (30-43) 38 (29-47) 35 (30-41) 57 (45-69) .02 28 d 93 (89-96) 91 (84-96) 96 (94-98) 98 (92-100) .07 Platelet recovery 14 d 14 (9-19) 29 (20-39) 20 (15-25) 22 (13-34) .02 28 d 78 (71-83) 84 (75-91) 82 (77-87) 93 (85-98) .01 D1X X CRFS indicates chronic GVHD- and relapse-free survival. *statistical significance. S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 79 65+ yr 141 (20) 3.3 (1.4-7.7) .006* Table 4 Disease .008* Multivariate analyses in MRD RIC alloHCT AML (reference) 400 (58) 1.0 n (%) Estimate (95% CI) P Value ALL 35 (5) 1.3 (.8-2.0) .28 CML 63 (9) .6 (.3-1.0) .05 MDS 192 (28) .7 (.6-.9) .02 Grade II-IV aGVHD Disease status .01* GVHD prophylaxis .14 Early (reference) 397 (57) 1.0 MTX-TAC (reference) 290 (43) 1.0 Advanced 147 (21) 1.4 (1.1-1.8) .004* MMF-CYSP 109 (16) 1.6 (1.1-2.4) .02 Intermediate 126 (18) 2X X (.6-1.2) .D9 .50 MTX-CYSP 213 (32) 1.3 (.9-1.8) .17 Missing 20 (3) 1.5D3X (X .8-2.7) .16 MMF-TAC 63 (9) 1.2 (.7-2.0) .44 KPS .006* Conditioning regimen / ATG <.001* <90 (reference) 264 (38) 1.0 BU+FLU § others (reference) 248 (37) 1.00 90-100 413 (60) .7 (.6-.8) .002* ATG § BU § FLU § others 134 (20) .9 (.4-1.1) .10 Missing 13 (2) 1.1 (.6-2.2) .78 FLU+LPAM § others 193 (29) 1.7 (1.2-2.4) .001* TBI § CY § FLU § others 100 (15) 1.2 (.8-1.9) .31 DFS KPS .04* GVHD prophylaxis .22 <90 (reference) 258 (38) 1.0 MTX-TAC (reference) 287 (43) 1.0 90-100 404 (60) .7 (.5-.9) .01 MMF-CYSP 108 (16) 1.1 (.9-1.5) .29 Missing 13 (2) .8 (.2-2.5) .69 MTX-CYSP 212 (32) .8 (.7-1.1) .19 MMF-TAC 63 (9) 1.1 (.8-1.5) .71 Grade III-IV aGVHD Disease status .003* GVHD prophylaxis .26 Early (reference) 390 (58) 1.0 MTX-TAC (reference) 291 (43) 1.0 Advanced 146 (22) 1.4 (1.1-1.8) .002* MMF-CYSP 109 (16) 1.6 (.9-2.8) .15 Intermediate 116 (17) .9 (.7-1.2) .55 MTX-CYSP 213 (31) .9 (.5-1.6) .76 Missing 18 (3) 1.6 (.9-2.7) .08 MMF-TAC 63 (9) 1.5 (.8-3.1) .23 Donor-recipient CMV match .006* Conditioning regimen / ATG .03* / (reference) 118 (18) 1.0 Bu+Flu § others (reference) 248 (37) 1.0 +/+ 325 (48) 1.1 (.9-1.5) .38 ATG § Bu § Flu § others 134 (20) .6 (.3-1.3) .25 +/ 69 (10) 1.3 (.9-1.8) .20 Flu+Mel § others 193 (29) 1.7 (1.0-2.8) .04 /+ 144 (21) 1.2 (.9-1.7) .20 TBI § Cy § Flu § others 101 (15) 1.2 (.6-2.4) .53 Missing 14 (2) 3.1 (1.7-5.6) <.001* KPS .01* cGVHD <90 (reference) 259 (39) 1.0 GVHD prophylaxis .42 90-100 399 (60) .7 (.6-.9) .004* MTX-TAC (reference) 287 (43) 1.0 Missing 12 (2) 1.0 (.5-2.1) .97 MMF-CYSP 103 (15) 1.1 (.7-1.7) .58 MTX-CYSP 210 (32) 1.3 (.9-1.8) .13 NRM MMF-TAC 63 (9) 1.3 (.8-2.0) .25 GVHD prophylaxis .54 Conditioning regimen / ATG .006* MTX-TAC (reference) 287 (43) 1.0 Bu+Flu § others (reference) 246 (37) 1.0 MMF-CYSP 108 (16) 1.3 (.8-2.1) .23 ATG § Bu § Flu § others 134 (20) .5 (.4-.8) .001* MTX-CYSP 212 (32) 1.3 (.9-2.0) .19 Flu+Mel § others 186 (28) 1.0 (.7-1.4) .95 MMF-TAC 63 (9) 1.2 (.7-2.1) .51 TBI § Cy § Flu § others 97 (15) 1.1 (.7-1.7) .66 Conditioning regimen / ATG .02* Disease status .04* Bu+Flu § others (reference) 246 (37) 1.0 Early (reference) 385 (58) 1.0 ATG § Bu § Flu § others 131 (20) 1.2 (.7-2.0) .47 Advanced 142 (21) 1.1 (.8-1.4) .74 Flu+Mel § others 188 (28) 1.9 (1.3-2.8) .002* Intermediate 117 (18) 1.2 (.8-1.6) .37 TBI § Cy § Flu § others 105 (16) 1.5 (.9-2.6) .10 Missing 19 (3) 2.5 (1.3-4.7) .005* Disease status .03* Donor-recipient CMV match .02* Early (reference) 390 (58) 1.0 / (reference) 113 (17) 1.0 Advanced 146 (22) 1.2 (.8-1.7) .32 +/+ 321 (48) 1.1 (.7-1.5) .74 Intermediate 116 (17) .6 (.4-1.0) .07 +/ 71 (11) .6 (.4-1.0) .05 Missing 18 (3) 2.2 (1.0-4.7) .04 /+ 144 (22) .7 (.5-1.1) .11 Donor-recipient sex match .002* Missing 14 (2) 1.8 (.7-4.5) .22 M-M (reference) 219 (33) 1.0 Donor-recipient sex match .01* F-F 135 (20) .9 (.6-1.5) .77 M-M (reference) 219 (33) 1.0 F-M 164 (24) 1.8 (1.2-2.7) .003* F-F 131 (20) 1.4 (1.0-2.0) .06 M-F 152 (23) 1.0 (.6-1.6) .98 F-M 161 (24) 1.7 (1.2-2.4) .002* Sorror comorbidity index .03* M-F 152 (23) 1.5 (1.1-2.0) .02 0-1 (reference) 134 (20) 1.0 Ethnicity .005* 2 187 (28) 2.3 (1.3-4.2) .003* Caucasian (reference) 512 (77) 1.0 Missing 10 (1) 2.4 (.7-8.3) .18 Missing 43 (6) .9 (.5-1.6) .78 N/A before 2007 339 (51) 2.0 (1.2-3.5) .01 Non-Caucasian 108 (16) 1.7 (1.2-2.4) .001* Relapse OS GVHD prophylaxis .22 GVHD prophylaxis .69 MTX-TAC (reference) 287 (43) 1.0 MTX-TAC (reference) 296 (43) 1.0 MMF-CYSP 108 (16) 1.1 (.7-1.6) .70 MMF-CYSP 110 (16) 1.2 (.9-1.5) .26 MTX-CYSP 212 (32) .7 (.5-1.0) .07 MTX-CYSP 220 (32) 1.0 (.8-1.3) .77 MMF-TAC 63 (9) .9 (.6-1.5) .80 MMF-TAC 64 (9) 1.1 (.8-1.6) .52 Conditioning regimen / ATG <.001* Age at transplant .002* Bu+Flu § others (reference) 246 (37) 1.0 18-29 yr (reference) 22 (3) 1.0 ATG § Bu § Flu § others 131 (20) 1.1 (.8-1.5) .65 30-39 yr 55 (8) 1.6 (.6-4.0) .35 Flu+Mel § others 188 (28) .4 (.3-.6) <.001* 40-49 yr 84 (12) 1.9 (.8-4.7) .13 TBI § Cy § Flu § others 105 (16) .9 (.6-1.3) .55 50-64 yr 388 (56) 2.8 (1.2-6.4) .01 Disease .03* (continued) (continued) 80 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 AML (reference) 393 (59) 1.0 22% to 35%) and 36% (95% CI, 27% to 46%) in the MTX-CYSP and ALL 34 (5) .9 (.5-1.8) .85 CML 59 (9) 1.0 (.6-1.7) .96 MMF-CYSP cohorts, respectively, and 43% (95% CI, 37% to 48%) MDS 184 (27) .6 (.5-.8) .003* and 33% (95% CI, 22% to 46%) in the MTX-TAC and MMF-TAC Disease status <.001* cohorts, respectively (Table 3). The risk of relapse was not Early (reference) 390 (58) 1.0 shown to be significantly different among any of the 4 cohorts Advanced 146 (22) 2.1 (1.5-2.8) <.001* Intermediate 116 (17) 1.2 (.8-1.9) .28 in the MRD group on multivariate analysis (Table 4). Missing 18 (3) 1.9 (.9-4.2) .11 In the URD group, the 2-year cumulative incidences of Donor-recipient CMV match .02* relapse were 34% (95% CI, 23% to 46%), 27% (95% CI, 21% to / (reference) 118 (18) 1.0 35%), 40% (95% CI, 35% to 44%), and 31% (95% CI, 25% to 37%) in +/+ 325 (48) 1.2 (.9-1.8) .23 the MTX-CYSP, MMF-CYSP, MTX-TAC, and MMF-TAC cohorts, +/ 69 (10) 1.4 (.9-2.2) .15 /+ 144 (21) 1.3 (.9-1.9) .21 respectively, on univariate analysis (Table 5). On multivariate Missing 14 (2) 3.3 (1.6-6.7) <.001* analysis, URD patients receiving MMF-CYSP had a significantly Sorror comorbidity index .003* lower risk of relapse, compared with those receiving MTX-TAC 0-1 (reference) 134 (20) 1.0 (HR, .53; P < .001) (Table 6). 2 187 (28) 1.0 (.7-1.4) .94 Missing 10 (1) .4 (.1-1.4) .18 N/A before 2007 339 (51) .6 (.4-.8) .001* Nonrelapse Mortality * statistical significance. On univariate analysis, the cumulative incidences of NRM at 2 years post-alloHCT in the MRD group were 21% (95% CI, 16% to 27%), 23% (95% CI, 15% to 32%), 16% (95% CI, 12% to 21%), and respectively (Table 5). Multivariate analysis did not show any 22% (95% CI, 13% to 33%) in the MTX-CYSP, MMF-CYSP, significant difference among the cohorts of URD group (Table 6, MTX-TAC, and MMF-TAC cohorts, respectively (Table 3). Multi- Figure 3B). variate analysis did not show any significant difference in NRM among the 4 GVHD cohorts of MRD group (Table 4). Disease-Free Survival The cumulative incidences of NRM at 2 years in the URD Univariate analysis demonstrated 2-year DFS of in the MRD group were 31% (95% CI, 20% to 43%), 31% (95% CI, 24% to 39%), group of 50% (95% CI, 43% to 57%) in the MTX-CYSP cohort, 41% 23% (95% CI, 19% to 27%), and 37% (95% CI, 31% to 43%) in the (95% CI, 32% to 50%) in the MMF-CYSP cohort, 41% (95% CI, 36% MTX-CYSP, MMF-CYSP, MTX-TAC, and MMF-TAC cohorts, to 47%) with MTX-TAC, and 44% (95% CI, 32% to 57%) with respectively (Table 5). Multivariate analysis of the URD group MMF-TAC (Table 3). There was no significant difference in DFS demonstrated that compared with MTX-TAC, MMF-TAC was among any of the GVHD cohorts in the MRD group on multi- associated with increased risk of NRM (HR, 1.48; P = .008) variate analysis (Table 4). (Table 6), notwithstanding the faster neutrophil recovery In the URD group, DFS at 2 years was 36% (95% CI, 25% to observed with MMF-TAC compared with other regimens (P < 48%) in the MTX-CYSP cohort, 41% (95% CI, 33% to 49%) in the .001) (Table 5). MMF-CYSP cohort, 38% (95% CI, 33% to 42%) in the MTX-TAC cohort, and 33% (95% CI, 27% to 40%) in the MMF-TAC cohort DISCUSSION (Table 5). Multivariate analysis did not show any significant The combinations of CNI with MTX or MMF for prevention difference in DFS among any of the URD cohorts (Table 6). of aGVHD and cGVHD after alloHCT have been accepted as the current standard,[6,8,9,18] although conflicting reports of out- Relapse comes with MMF- and MTX-containing CNI-based regimens Univariate analysis revealed that in the MRD group, the have been noted. Despite the lack of prospective comparative cumulative incidences of relapse at 2 years were 28% (95% CI, data between MMF- and MTX-based regimens in the RIC Figure 1. (A) Adjusted curves for cumulative incidence of grade II-IV acute GVHD in MRD recipients of RIC alloHCT using 1 of the 4 GVHD prophylaxis regimens: TAC- MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. (B) Adjusted curves for cumulative incidence of grade II-IV aGVHD in URD RIC alloHCT patients on 1 of the 4 GVHD pro- phylaxis regimens: TAC-MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 81 Table 5 Univariate analyses (URD) MTX-CYSP MMF-CYSP MTX-TAC MMF-TAC (n = 71) (n = 153) (n = 432) (n = 218) Outcomes Probability (95% CI) (%) Probability (95%CI) (%) Probability (95% CI) (%) Probability (95% CI) (%) P Value aGVHD II-IV 1 mo 15 (8-25) 38 (30-45) 18 (15-22) 26 (21-32) <.001 2 mo 31 (21-42) 51 (43-59) 33 (29-37) 41 (35-48) <.001 100 d 32 (22-44) 53 (45-61) 37 (32-41) 47 (41-54) <.001 aGVHD III-IV 1 mo 8 (3-16) 16 (10-22) 8 (6-11) 14 (10-19) .03 2 mo 15 (8-25) 21 (15-28) 12 (9-16) 19 (14-25) .02 100 d 15 (8-25) 21 (15-28) 13 (10-17) 21 (16-27) .06 cGVHD 100 d 3 (0-8) 10 (6-15) 4 (3-6) 2 (1-5) .03 6 mo 19 (10-29) 36 (28-43) 22 (18-26) 26 (21-33) .01 1 yr 36 (25-47) 50 (42-58) 40 (35-45) 44 (37-50) .11 OS 100 d 77 (67-86) 83 (77-89) 87 (83-90) 81 (75-86) .12 6 mo 66 (54-76) 69 (62-76) 78 (74-82) 67 (61-73) .006 1 yr 54 (42-66) 57 (49-65) 60 (55-65) 49 (42-56) .06 2 yr 40 (28-52) 45 (37-53) 47 (42-51) 41 (35-48) .49 DFS 100 d 72 (60-82) 73 (66-80) 74 (69-78) 70 (64-76) .78 6 mo 59 (47-71) 60 (52-68) 61 (57-66) 56 (49-62) .59 1 yr 50 (38-62) 50 (42-58) 48 (43-53) 41 (34-47) .21 2 yr 36 (25-48) 41 (33-49) 38 (33-42) 33 (27-40) .47 Relapse 100 d 16 (9-26) 15 (10-21) 17 (14-21) 15 (11-20) .85 6 mo 26 (16-37) 19 (13-26) 26 (22-31) 24 (18-30) .29 1 yr 30 (20-42) 23 (17-30) 35 (30-39) 28 (22-34) .03 2 yr 34 (23-46) 27 (21-35) 40 (35-44) 31 (25-37) .02 NRM 100 d 12 (5-21) 12 (7-17) 9 (6-12) 15 (10-20) .16 6 mo 15 (7-25) 21 (15-28) 13 (10-16) 21 (16-27) .02 1 yr 20 (11-30) 27 (20-34) 17 (14-21) 31 (25-37) <.001 2 yr 30 (19-42) 31 (24-39) 23 (19-27) 36 (29-42) .005 GRFS 100 d 63 (52-74) 52 (44-60) 63 (58-67) 58 (51-65) .12 6 mo 41 (30-52) 31 (24-38) 38 (34-43) 33 (26-39) .20 1 yr 24 (15-34) 14 (9-21) 16 (13-20) 11 (8-16) .10 2 yr 13 (6-22) 10 (5-15) 11 (8-14) 7 (4-11) .29 CRFS 100 d 66 (55-77) 67 (60-75) 73 (68-77) 70 (63-76) .52 6 mo 49 (37-61) 47 (40-55) 55 (50-59) 47 (41-54) .22 1 yr 37 (27-49) 35 (27-42) 37 (32-42) 31 (25-38) .50 2 yr 21 (12-32) 26 (20-34) 28 (24-33) 22 (17-28) .31 Neutrophil recovery 14 days 46 (35-58) 32 (25-40) 35 (31-40) 69 (62-75) <.001 28 days 97 (92-100) 96 (92-99) 96 (94-98) 95 (92-98) .91 Platelet recovery 14 days 17 (9-27) 19 (13-27) 19 (13-27) 18 (13-24) .61 28 days 73 (62-84) 82 (75-89) 83 (79-87) 80 (74-85) .38 *statistical significance. setting, MMF-CNI has been an established regimen after RIC significant difference in the grade III to IV aGVHD risk. All 4 alloHCT. While some studies have compared the CNIs (TAC ver- regimens resulted in similar cGVHD incidence after URD sus CYSP) and others have compared MTX- and MMF-contain- alloHCT. MTX-TAC, in addition, was associated with 32% lower ing GVHD prophylaxis, no study to date has investigated all 4 NRM risk compared with MMF-TAC. Furthermore, MTX-TAC regimens concomitantly to compare the outcomes after RIC was associated with 88% increase in relapse risk relative to alloHCT, and, therefore, there has been no convincing evidence MMF-CYSP but did not meet statistical significance when com- to date supporting the use of a particular regimen in the RIC pared with MMF-TAC. Higher relapse risk observed with MTX- setting. TAC did not translate into worse DFS and OS as the analysis We have made several important observations in this anal- revealed no significant difference among any of the cohorts in ysis. No single GVHD prevention regimen is superior, the lim- the URD group. No significant interaction was found between ited power notwithstanding, to detect differences in aGVHD GVHD prophylaxis and the conditioning, but patients receiving and survival outcomes in the MRD group. In those with URD, Flu/Mel conditioning had significantly higher risk of grade II to however, MTX-TAC performed better than MMF-CYSP and IV (HR, 1.75; P < .001) and grade III to IV (HR, 2.71; P < .001) resulted in 44% risk reduction in grade II to IV and 48% risk aGVHD compared with the Flu/Bu regimen (Table 6). The risks reduction in grade III to IV aGVHD. Furthermore, MTX-CYSP of cGVHD and NRM with Flu/Mel conditioning were also resulted in 48% reduction in the incidence of grade II to IV increased but did not meet statistical significance (P = .03 and aGVHD relative to MMF-CYSP, but did not show statistically .05, respectively). 82 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 65+ yr 258 (29) 1.1 (.7-1.7) .55 Table 6 Disease status .004* Multivariate analyses in URD RIC alloHCT Early (reference) 515 (59) 1.0 n (%) Estimate (95% CI) P Value Advanced 208 (24) 1.2 (1.0-1.5) .02 Intermediate 127 (14) 1.0 (.8-1.3) .82 Grade II-IV aGVHD Missing 24 (3) 2.0 (1.3-3.1) .002* GVHD prophylaxis <.001* Donor type MTX-TAC (reference) 429 (49) 1.0 7/8-matched URD (reference) 172 (20) 1.0 MMF-CYSP 152 (17) 1.8 (1.3-2.4) <.001* 8/8-matched URD 702 (80) .7 (.6-.8) <.001* MTX-CYSP 71 (8) .8 (.5-1.2) .29 KPS .002* MMF-TAC 216 (25) 1.3 (1.0-1.7) .03 <90 (reference) 372 (43) 1.0 Conditioning regimen / ATG .002* 90-100 475 (54) .8 (.6-.9) .001* Bu+Flu § others (reference) 173 (20) 1.0 Missing 27 (3) 1.1 (.7-1.6) .71 ATG § Bu § Flu § others 414 (48) 1.1 (.8-1.4) .61 Sorror comorbidity index .05* Flu+Mel § others 147 (17) 1.7 (1.3-2.4) <.001* 0-1 (reference) 191 (22) 1.0 TBI § Cy § Flu § others 134 (15) 1.2 (.8-1.7) .37 2 315 (36) 1.1 (.9-1.4) .43 Disease status .01* Missing 20 (2) .5 (.3-1.1) .09 Early (reference) 512 (59) 1.0 N/A before 2007 348 (40) 1.2 (1.0-1.5) .08 Advanced 206 (24) 1.1 (.9-1.5) .25 Intermediate 127 (15) .7 (.5-1.0) .03 DFS Missing 23 (3) 1.6 (.9-2.7) .08 GVHD prophylaxis .68 Donor type MTX-TAC (reference) 423 (49) 1.0 7/8-matched URD (reference) 172 (20) 1.0 MMF-CYSP 153 (18) .9 (.7-1.1) .28 8/8-matched URD 696 (80) .7 (.5-.9) .002* MTX-CYSP 67 (8) 1.0 (.7-1.4) .94 MMF-TAC 216 (25) 1.0 (.8-1.2) .82 Grade III-IV aGVHD Age at transplant .04* GVHD prophylaxis .01* 18-29 yr (reference) 38 (4) 1.0 MTX-TAC (reference) 428 (49) 1.0 30-39 yr 52 (6) .5 (.3-.9) .02 MMF-CYSP 152 (18) 1.9 (1.2-3.1) .006* 40-49 yr 78 (9) .8 (.5-1.2) .24 MTX-CYSP 71 (8) .8 (.4-1.6) .53 50-64 yr 439 (51) .9 (.6-1.3) .59 MMF-TAC 214 (257) 1.5 (1.0-2.3) .03 65+ yr 252 (29) 1.0 (.6-1.4) .88 Conditioning regimen / ATG <.001* Disease status .003* Bu+Flu § others (reference) 173 (20) 1.0 Early (reference) 511 (59) 1.0 ATG § Bu § Flu § others 412 (48) 1.1 (.7-1.7) .78 Advanced 204 (24) 1.4 (1.1-1.6) <.001* Flu+Mel § others 146 (17) 2.7 (1.6-4.4) <.001* Intermediate 121 (14) 1.1 (.9-1.4) .46 TBI § Cy § Flu § others 134 (15) .9 (.5-1.6) .66 Missing 23 (3) 1.7 (1.0-2.6) .03 Disease .02* Donor type AML (reference) 495 (57) 1.0 7/8-matched URD (reference) 170 (20) 1.0 ALL 39 (4) .4 (.1-1.2) .10 8/8-matched URD 689 (80) .8 (.6-.9) .006* CML 48 (5) 2.1 (1.1-4.0) .03 KPS .005* MDS 283 (33) 1.4 (1.0-2.0) .08 <90 (reference) 361 (42) 1.0 Disease status .04* 90-100 471 (55) .8 (.7-.9) .002* Early (reference) 511 (59) 1.0 Missing 27 (3) 1.1 (.7-1.6) .70 Advanced 204 (24) 1.2 (.8-1.7) .46 Intermediate 127 (15) .7 (.4-1.3) .25 NRM Missing 23 (3) 2.3 (1.2-4.5) .01 GVHD prophylaxis .05 Donor type MTX-TAC (reference) 423 (49) 1.0 7/8-matched URD (reference) 172 (20) 1.0 MMF-CYSP 153 (18) 1.4 (1.0-1.9) .06 8/8-matched URD 693 (80) .5 (.4-.7) <.001* MTX-CYSP 67 (8) 1.3 (.8-2.0) .26 MMF-TAC 216 (25) 1.5 (1.1-2.0) .008* Chronic GVHD Age at transplant .003* GVHD prophylaxis .35 18-29 yr (reference) 38 (4) 1.0 MTX-TAC (reference) 428 (49) 1.0 30-39 yr 52 (6) .6 (.3-1.3) .20 MMF-CYSP 152 (17) 1.3 (.9-1.7) .10 40-49 yr 78 (9) .8 (.4-1.7) .60 MTX-CYSP 70 (8) 1.0 (.7-1.5) .96 50-64 yr 439 (51) 1.1 (.6-2.2) .69 MMF-TAC 217 (25) 1.2 (.9-1.5) .18 65+ yr 252 (29) 1.5 (.8-3.0) .20 Conditioning regimen / ATG <.001* Conditioning regimen / ATG .008* Bu+Flu § others (reference) 174 (20) 1.0 Bu+Flu § others (reference) 172 (20) 1.0 ATG § Bu § Flu § others 413 (48) .7 (.6-1.0) .03 ATG § Bu § Flu § others 410 (48) .9 (.6-1.2) .43 Flu+Mel § others 146 (17) 1.4 (1.0-1.9) .03 Flu+Mel § others 143 (17) 1.4 (1.0-2.1) .05 TBI § Cy § Flu § others 134 (15) 1.4 (1.0-2.0) .04 TBI § Cy § Flu § others 134 (16) .8 (.5-1.2) .35 Year of transplant .03* Disease status .02* 2000-2004 (reference) 80 (9) 1.0 Early (reference) 511 (59) 1.0 2005-2008 366 (42) .7 (.5-1.0) .03 Advanced 204 (24) .9 (.7-1.2) .68 2009-2013 421 (49) .6 (.4-.9) .007* Intermediate 121 (14) .8 (.6-1.2) .34 Missing 23 (3) 2.2 (1.3-3.9) .006* OS Donor type <.001* GVHD prophylaxis .82 7/8-matched URD (reference) 170 (20) 1.0 MTX-TAC (reference) 432 (49) 1.0 8/8-matched URD 689 (80) .5 (.4-.7) <.001* MMF-CYSP 153 (17) 1.0 (.8-1.2) .97 Donor-recipient CMV match .04* MTX-CYSP 71 (8) 1.1 (.8-1.5) .49 / (reference) 215 (25) 1.0 MMF-TAC 218 (25) 1.1 (.9-1.3) .47 +/+ 201 (23) 1.2 (.8-1.7) .30 Age at transplant .008* +/ 80 (9) 1.7 (1.1-2.6) .01 18-29 yr (reference) 38 (4) 1.0 /+ 340 (40) 1.3 (1.0-1.8) .05 30-39 yr 54 (6) .6 (.3-1.0) .03 Missing 23 (3) 2.2 (1.1-4.2) .02 40-49 yr 80 (9) .8 (.5-1.3) .37 KPS .007* 50-64 yr 444 (51) .9 (.6-1.4) .67 <90 (reference) 361 (42) 1.0 (continued) (continued) S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 83 90-100 471 (55) .7 (.5-.9) .002* MMF arm (19% versus 4%; P = .03), predominantly in MAC Missing 27 (3) .9 (.5-1.7) .82 alloHCT using URD. A meta-analysis of the previously men- Relapse tioned 3 randomized trials by the Cochrane Collaboration GVHD prophylaxis .006* found no differences in the rates of aGVHD and cGVHD among MTX-TAC (reference) 423 (49) 1.0 the different regimens [24]. There was no evidence for a signif- MMF-CYSP 153 (18) .5 (.4-.7) <.001* MTX-CYSP 67 (8) .8 (.5-1.2) .35 icant difference between MMF and MTX for the incidence of MMF-TAC 216 (25) .7 (.5-1.0) .03 aGVHD and cGVHD, neutrophil engraftment, incidence of Conditioning regimen / ATG .003* relapse, NRM, and OS. The results are also in accord with those Bu+Flu § others (reference) 172 (20) 1.0 of a meta-analysis of 11 studies [2] including 1076 patients (a ATG § Bu § Flu § others 410 (48) .8 (.6-1.1) .26 mix of MAC and RIC alloHCT recipients) that determined Flu+Mel § others 143 (17) .5 (.3-.7) <.001* TBI § Cy § Flu § others 134 (16) 1.0 (.7-1.5) .99 greater incidence of grade III to IV aGVHD in MMF recipients Disease .006* (HR, 1.6; 95% CI, 1.2 to 2.3). The increased risk of severe aGVHD AML (reference) 492 (57) 1.0 with MMF was limited to the patients with URD and was not ALL 37 (4) 1.1 (.7-1.9) .66 evident after MRD alloHCT. The 3 prospective trials had rela- CML 48 (6) .6 (.3-1.1) .09 MDS 282 (33) .6 (.5-.9) .002* tively small sample sizes and only included patients receiving Disease status <.001* MAC and therefore, their findings cannot be applied to RIC Early (reference) 511 (59) 1.0 patients. It is noteworthy that none of the previously men- Advanced 204 (24) 1.8 (1.4-2.3) <.001* tioned studies including the meta-analyses demonstrated any Intermediate 121 (14) 1.2 (.9-1.7) .17 significant differences in the relapse risk between MTX- and Missing 23 (3) 1.2 (.5-2.8) .64 Sex .03 MMF-based GVHD regimens, unlike reported by our study. We Male (reference) 487 (57) 1.0 can only speculate that there are unknown variables and con- Female 372 (43) 1.3 (1.0-1.6) .03 founders, in addition to the competing risk of low NRM that * statistical significance. contributed to the increased relapse risk and neutralized any possible survival advantage MTX-based regimens could have had in URD group. Previously, 3 randomized studies had compared outcomes Eapen et al. [16] compared outcomes between bone mar- of alloHCT after CYSP [21,22] or TAC [23] combined with MTX row and peripheral blood grafts for RIC alloHCT for patients or MMF in the myeloablative setting. One study enrolled with AML, MDS and non-Hodgkin lymphoma using URD in 88 alloHCT from URD [21], another study from MRD [22], and a US transplant centers (2000 to 2008) and reported no differen- third study included both [23]. None of the studies showed a ces in outcomes between the 2 graft sources [], but patients statistically significant difference in the cumulative incidence receiving MMF (versus MTX) had an increased risk of grade II of aGVHD among the regimens. Bolwell et al. [22] reported the to IV and III to IV aGVHD, cGVHD, NRM, and worse OS. Patients randomized study comparing MMF-CYSP and MTX-CYSP with ALL and those receiving TBI were excluded in this study. (n = 40) after marrow transplantation using MRD []. No differ- Despite the differences in the primary objectives and patient ence was observed in the incidence of GVHD or survival. Per- populations between the 2 studies, the results of our analysis kins et al. [23] reported the results of a randomized phase II are in concordance with Eapen et al.’s [16] study as we show study comparing MMF-TAC and MTX-TAC after alloHCT from MMF-containing regimens are associated with worse interme- MRD and URD (n = 89) []. Patients in the MMF cohort were less diate outcomes without impact on OS in the URD setting. This likely to experience severe mucositis, and the cumulative inci- is an important study because it examines not only the efficacy dence of grade II to IV aGVHD was similar. However, the cumu- of MTX or MMF, but also the added impact of TAC and CYSP in lative incidence of grade III to IV aGVHD was higher in the ensuring post-alloHCT outcomes. To compare only MTX and Figure 2. (A). Adjusted curves for cumulative incidence of aGVHD in MRD RIC alloHCT patients on 1 of the 4 GVHD prophylaxis regimens: TAC-MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. (B) Adjusted curves for cumulative incidence of aGVHD in URD RIC alloHCT patients receiving 1 of the 4 GVHD prophylaxis regimens: TAC-MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. 84 S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 Figure 3. (A). Adjusted curves for OS in MRD RIC alloHCT patients receiving 1 of the 4 GVHD prophylaxis regimens: TAC-MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. (B) Adjusted curves for OS in URD RIC alloHCT patients receiving 1 of the 4 GVHD prophylaxis regimens: TAC-MTX, CYSP-MTX, CYSP-MMF, and TAC-MMF. MMF would be assuming that the 2 CNIs, TAC and CYSP, have limitation in having variable therapeutic target blood level no difference in efficacy and can be used interchangeably and ranges for TAC and CYSP at different centers. Furthermore, the study findings do not support this assumption. we examined the cumulative incidences of cGVHD of any Owing to the retrospective nature, the findings of the study grade reported to the database and did not specifically eval- need to be interpreted with caution. We acknowledge the dif- uate the risk of moderate-to-severe or organ-specific cGVHD ferences in patient, disease, and transplant characteristics in the cohorts. among the cohorts in both donor groups, especially the small It is also important to note in the study the trade-off sample size in certain cohorts (MMF-TAC cohort in the MRD between low NRM and higher relapse risk with MTX-TAC com- group and MTX-CYSP in the URD group), differences in the pro- pared with MMF-CYSP, resulting in no difference in OS. For portions of ATG recipients in the cohorts of the URD group and this reason, it would be worth considering a future prospective the fact that the inclusion of ATG in the conditioning makes for study in the URD patient population using the composite end- a heterogeneous study population. These differences were point such as GVHD- and relapse-free survival (GRFS) that addressed by performing a controlled analysis that accounted assesses all significant and relevant endpoints [27]. The events for all the characteristics and any center effects. We also exam- for GRFS include grade III to IV aGVHD, systemic therapy- ined the study population for differences in the outcomes of requiring cGVHD, relapse, or death. A similar composite end- grade II to IV and grade III to IV aGVHD and cGVHD after point that has been in vogue is cGVHD- and relapse-free sur- excluding ATG recipients: univariate analysis showed vival, which includes survival without development of cumulative incidence of grade II to IV aGVHD was highest cGVHD, disease relapse or progression and death [28]. Interest- with MMF-CYSP and lowest with MTX-TAC in both the ingly, the analysis for both GRFS and cGVHD- and relapse-free MRD and URD groups (Supplementary Tables 1A and 1B), survival did not reveal any significant differences among the but no significant differences in the incidence of grade III MRD and URD cohorts (Tables 3 and 5). to IV aGVHD were observed in either group. cGVHD was In summary, in this observational study, we described the observed more frequently with CYSP regimens than TAC in outcomes after RIC alloHCT using the 4 CNI-based regimens. the MRD group, and similarly, in the URD group, MMF- This differentiates the study in that we considered the 2 drugs CYSP had higher incidence of cGVHD compared with TAC- of each prophylactic regimen as a unique combination, which based regimens. enabled comparisons among the 4 regimens. This analysis Despite carefully considering multiple potentially signifi- demonstrated equivalent outcomes in those with MRD using cant variables, the effect of unrecognized biases and residual either of the 4 CNI-based combinations and inferior efficacy of confounding in the analysis cannot be ruled out. For MMF-based approach with regard to grade II to IV and III to IV instance, the dose of MTX and the dose and schedule of MMF aGVHD and NRM in those with URD. Moreover, the analysis in the regimens are variable among the transplant centers. did not suggest using a particular regimen in URD alloHCT Different dosing protocols for short-course MTX and differ- recipients using RIC and peripheral blood graft, based on the ent doses and duration of MMF adopted by the transplant lack of significant differences in OS, even though aGVHD risk centers were not captured in the database. It has been dem- was significantly improved with MTX-CNI regimens and there onstrated that higher trough levels of MMF attributed to may be a trend for improved 1-year GRFS in the URD group intensified dosing are correlated with a decreased incidence with MTX-CNI than with MMF-CNI. Finally, a prospective ran- of severe GVHD after umbilical cord blood transplantation domized controlled trial of URD RIC alloHCT recipients is [25,26]. Moreover, the proportion of patients that did not needed to evaluate these GVHD prophylaxis regimens with receive all 4 doses of MTX due to severe oropharyngeal uniform dosing schedules and target pharmacokinetic ranges mucositis is not known. We cannot confirm that oral (and and using novel endpoints such as GRFS to confirm the find- not intravenous) formulations of TAC and CYSP were used ings of this study. The results of such a trial may also inform for all RIC alloHCT in the study. We also recognize the the ideal partner for GVHD prevention strategies such as S. Chhabra et al. / Biol Blood Marrow Transplant 25 (2019) 73 85 85 post-transplant cyclophosphamide and other novel agents in high-dose cytotoxic therapy with graft-versus-tumor effects. Blood. the future clinical trials. 2001;97:3390–3400. 14 Ostronoff F, Ostronoff M, Souto-Maior AP, et al. Prospective trial of myco- phenolate mofetil-cyclosporine A prophylaxis for acute GVHD after G-CSF stimulated allogeneic bone marrow transplantation with HLA-identical SUPPLEMENTARY DATA sibling donors in patients with severe aplastic anemia and hematological Supplementary data associated with this article can be malignancies. Clin Transplant. 2009;23:33–38. found in the online version at doi:10.1016/j.bbmt.2018.08.018. 15 Ruutu T, van Biezen A, Hertenstein B, et al. Prophylaxis and treatment of GVHD after allogeneic haematopoietic SCT: a survey of centre strategies by the European Group for Blood and Marrow Transplantation. Bone Marrow REFERENCES Transplant. 2012;47:1459–1464. 1 Pinana JL, Valcarcel D, Fernandez-Aviles F, et al. MTX or mycophenolate 16 Eapen M, Logan BR, Horowitz MM, et al. Bone marrow or peripheral blood mofetil with CsA as GVHD prophylaxis after reduced-intensity condition- for reduced-intensity conditioning unrelated donor transplantation. J Clin ing PBSCT from HLA-identical siblings. Bone Marrow Transplant. Oncol. 2015;33:364–369. 2010;45:1449–1456. 17 Weisdorf D, Spellman S, Haagenson M, et al. Classification of HLA-matching 2 Ram R, Yeshurun M, Vidal L, et al. Mycophenolate mofetil versus metho- for retrospective analysis of unrelated donor transplantation: revised defini- trexate for the prevention of graft-versus-host-disease systematic review tions to predict survival. Biol Blood Marrow Transplant. 2008;14:748–758. and meta-analysis. Leuk Res. 2014;38:352–360. 18 Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on 3 Alyea EP. Graft-versus-host disease prevention: corticosteroids revisited. J Acute GVHD Grading. Bone Marrow Transplant. 1995;15:825–828. Clin Oncol. 2016;34:1836–1837. 19 Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health 4 Storb R, Deeg HJ, Pepe M, et al. Methotrexate and cyclosporine versus consensus development project on criteria for clinical trials in chronic cyclosporine alone for prophylaxis of graft-versus-host disease in patients graft-versus-host disease: I. Diagnosis and staging working group report. given HLA-identical marrow grafts for leukemia: long-term follow-up of a Biol Blood Marrow Transplant. 2005;11:945–956. controlled trial. Blood. 1989;73:1729–1734. 20 Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning 5 Storb R, Deeg HJ, Whitehead J, et al. Methotrexate and cyclosporine com- regimens: working definitions. Biol Blood Marrow Transplant. pared with cyclosporine alone for prophylaxis of acute graft versus host 2009;15:1628–1633. disease after marrow transplantation for leukemia. N Engl J Med. 21 Kiehl MG, Schafer-Eckart K, Kroger M, et al. Mycophenolate mofetil for the 1986;314:729–735. prophylaxis of acute graft-versus-host disease in stem cell transplant 6 Nieto Y, Patton N, Hawkins T, et al. Tacrolimus and mycophenolate mofetil recipients. Transplant Proc. 2002;34:2922–2924. after nonmyeloablative matched-sibling donor allogeneic stem-cell trans- 22 Bolwell B, Sobecks R, Pohlman B, et al. A prospective randomized trial com- plantations conditioned with fludarabine and low-dose total body irradia- paring cyclosporine and short course methotrexate with cyclosporine and tion. Biol Blood Marrow Transplant. 2006;12:217–225. mycophenolate mofetil for GVHD prophylaxis in myeloablative allogeneic 7 Storb R, Antin JH, Cutler C. Should methotrexate plus calcineurin inhibitors bone marrow transplantation. Bone Marrow Transplant. 2004;34:621–625. be considered standard of care for prophylaxis of acute graft-versus-host 23 Perkins J, Field T, Kim J, et al. A randomized phase II trial comparing tacroli- disease? Biol Blood Marrow Transplant. 2010;16:S18–S27. mus and mycophenolate mofetil to tacrolimus and methotrexate for acute 8 Ratanatharathorn V, Nash RA, Przepiorka D, et al. Phase III study comparing graft-versus-host disease prophylaxis. Biol Blood Marrow Transplant. methotrexate and tacrolimus (prograf, FK506) with methotrexate and 2010;16:937–947. cyclosporine for graft-versus-host disease prophylaxis after HLA-identical 24 Kharfan-Dabaja M, Mhaskar R, Reljic T, et al. Mycophenolate mofetil versus sibling bone marrow transplantation. Blood. 1998;92:2303–2314. methotrexate for prevention of graft-versus-host disease in people receiv- 9 Nash RA, Antin JH, Karanes C, et al. Phase 3 study comparing methotrexate ing allogeneic hematopoietic stem cell transplantation. Cochrane Database and tacrolimus with methotrexate and cyclosporine for prophylaxis of Syst Rev. 2014;2014 CD010280. acute graft-versus-host disease after marrow transplantation from unre- 25 Bejanyan N, Rogosheske J, DeFor T, et al. Higher dose of mycophenolate lated donors. Blood. 2000;96:2062–2068. mofetil reduces acute graft-versus-host disease in reduced-intensity con- 10 Neumann F, Graef T, Tapprich C, et al. Cyclosporine A and mycophenolate ditioning double umbilical cord blood transplantation. Biol Blood Marrow mofetil vs cyclosporine A and methotrexate for graft-versus-host disease Transplant. 2015;21:926–933. prophylaxis after stem cell transplantation from HLA-identical siblings. 26 Harnicar S, Ponce DM, Hilden P, et al. Intensified mycophenolate mofetil Bone Marrow Transplant. 2005;35:1089–1093. dosing and higher mycophenolic acid trough levels reduce severe acute 11 Cutler C, Li S, Kim HT, et al. Mucositis after allogeneic hematopoietic stem graft-versus-host disease after double-unit cord blood transplantation. Biol cell transplantation: a cohort study of methotrexate- and non-methotrex- Blood Marrow Transplant. 2015;21:920–925. ate-containing graft-versus-host disease prophylaxis regimens. Biol Blood 27 Holtan SG, DeFor TE, Lazaryan A, et al. Composite end point of graft-versus- Marrow Transplant. 2005;11:383–388. host disease-free, relapse-free survival after allogeneic hematopoietic cell 12 Ho VT, Weller E, Lee SJ, et al. Prognostic factors for early severe pulmonary transplantation. Blood. 2015;125:1333–1338. complications after hematopoietic stem cell transplantation. Biol Blood 28 Pasquini MC, Logan B, Jones RJ, et al. Blood and Marrow Transplant Clinical Marrow Transplant. 2001;7:223–229. Trials Network report on the Development of Novel Endpoints and Selec- 13 McSweeney PA, Niederwieser D, Shizuru JA, et al. Hematopoietic cell trans- tion of Promising Approaches for Graft-versus-Host Disease Prevention plantation in older patients with hematologic malignancies: replacing Trials. Biol Blood Marrow Transplant. 2018;24:1274–1280.