Biol Blood Marrow Transplant 26 (2020) 14591468 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Risk Factors for Graft-versus-Host Disease in Haploidentical Hematopoietic Cell Transplantation Using Post-Transplant Cyclophosphamide Annie Im1,1, Armin Rashidi2,1, Tao Wang3,4, Michael Hemmer3, Margaret L. MacMillan5, Joseph Pidala6, Madan Jagasia7, Steven Pavletic8, Navneet S. Majhail9, Daniel Weisdorf2, Hisham Abdel-Azim10, Vaibhav Agrawal11, A. Samer Al-Homsi12, Mahmoud Aljurf13, Medhat Askar14, Jeffery J. Auletta15, Asad Bashey16, Amer Beitinjaneh17, Vijaya Raj Bhatt18, Michael Byrne7, Jean-Yves Cahn19, Mitchell Cairo20, Paul Castillo21, Jan Cerny22, Saurabh Chhabra3, Hannah Choe23, Stefan Ciurea24, Andrew Daly25, Miguel Angel Diaz Perez26, Nosha Farhadfar27, Shahinaz M. Gadalla28, Robert Gale29, Siddhartha Ganguly30, Usama Gergis31, Rabi Hanna32, Peiman Hematti33, Roger Herzig34, Gerhard C. Hildebrandt35, Deepesh P. Lad36, Catherine Lee37, Leslie Lehmann38, Lazaros Lekakis17, Rammurti T. Kamble39, Mohamed A. Kharfan-Dabaja40, Pooja Khandelwal41,42, Rodrigo Martino43, Hemant S. Murthy40, Taiga Nishihori44, Tracey A. O’Brien45, Richard F. Olsson46,47, Sagar S. Patel48, Miguel-Angel Perales49, Tim Prestidge50, Muna Qayed51, Rizwan Romee52, He  le ne Schoemans53, Sachiko Seo54, Akshay Sharma55, Melhem Solh , Roger Strair , Takanori Teshima58, Alvaro Urbano-Ispizua59, 56 57 Marjolein Van der Poel60, Ravi Vij61, John L. Wagner62, Basem William63, Baldeep Wirk64, Jean A. Yared65, Steve R. Spellman3, Mukta Arora2, Betty K. Hamilton9,1,* 1 University of Pittsburgh/UPMC Hillman Cancer Center, Pittsburgh, Pennsylvania 2 Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minnesota, Minneapolis, Minnesota 3 Department of Medicine, Medical College of Wisconsin, CIBMTRÒ (Center for International Blood and Marrow Transplant Research), Milwaukee, Wisconsin 4 Division of Biostatistics, Institute for Health and Equity, Medical College of Wisconsin, Milwaukee, Wisconsin 5 Blood and Marrow Transplant Program, Department of Pediatrics, University of Minnesota, Minneapolis, Minnesota 6 H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida 7 Vanderbilt University Medical Center, Nashville, Tennessee 8 Immune Deficiency Cellular Therapy Program, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland 9 Blood & Marrow Transplant Program, Cleveland Clinic Taussig Cancer Institute, Cleveland, Ohio 10 Division of Hematology, Oncology and Blood & Marrow Transplantation, Children's Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, California 11 Division of Hematology-Oncology, Indiana University School of Medicine, Indianapolis, Indiana 12 New York University Langone Medical Center, New York, New York 13 Department of Oncology, King Faisal Specialist Hospital Center & Research, Riyadh, Saudi Arabia 14 Department of Pathology and Laboratory Medicine, Baylor University Medical Center, Dallas, Texas 15 Blood and Marrow Transplant Program and Host Defense Program, Divisions of Hematology/Oncology/Bone Marrow Transplant and Infectious Diseases, Nationwide Children's Hospital, Columbus, Ohio 16 Blood and Marrow Transplant Program at Northside Hospital, Atlanta, Georgia 17 Department of Hematology and Oncology, University of Miami, Miami, Florida 18 The Fred and Pamela Buffett Cancer Center, University of Nebraska Medical Center, Omaha, Nebraska 19 Department of Hematology, CHU Grenoble Alpes, Grenoble, France 20 Division of Pediatric Hematology, Oncology and Stem Cell Transplantation, Department of Pediatrics, New York Medical College, Valhalla, New York 21 UF Health Shands Children's Hospital, Gainesville, Florida 22 Division of Hematology/Oncology, Department of Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts 23 James Comprehensive Cancer Center, The Ohio State University Wexner Medical Center, Columbus, Ohio 24 The University of Texas MD Anderson Cancer Center, Houston, Texas Financial disclosure: See Acknowledgments on page 1467. *Correspondence and reprint requests: Betty Ky Hamilton, MD, Blood and Marrow Transplant Program, Taussig Cancer Institute, Cleveland Clinic, 9500 Euclid Ave- nue CA 60, Cleveland, OH 44195. E-mail address: hamiltb2@ccf.org (B.K. Hamilton). 1 *These authors contributed equally to this manuscript https://doi.org/10.1016/j.bbmt.2020.05.001 1083-8791/© 2020 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. All rights reserved. 1460 A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 25 Tom Baker Cancer Center, Calgary, Alberta, Canada 26 Department of Hematology/Oncology, Hospital Infantil Universitario Nino Jesus, Madrid, Spain 27 Division of Hematology/Oncology, University of Florida College of Medicine, Gainesville, Florida 28 Division of Cancer Epidemiology & Genetics, NIH-NCI Clinical Genetics Branch, Rockville, Maryland 29 Hematology Research Centre, Division of Experimental Medicine, Department of Medicine, Imperial College London, London, United Kingdom 30 Division of Hematological Malignancy and Cellular Therapeutics, University of Kansas Health System, Kansas City, Kansas 31 Department of Medical Oncology, Thomas Jefferson University, Philadelphia, Pennsylvania 32 Cleveland Clinic Foundation, Cleveland, Ohio 33 Division of Hematology/Oncology/Bone Marrow Transplantation, Department of Medicine, University of Wisconsin, Madison, Wisconsin 34 University of Kentucky Chandler Medical Center, Louisville, Kentucky 35 Markey Cancer Center, University of Kentucky, Lexington, Kentucky 36 Department of Internal Medicine, Postgraduate Institute of Medical Education and Research, Chandigarh, India 37 Utah Blood and Marrow Transplant Program at Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah 38 Dana-Farber Cancer Institute/Boston Children’s Hospital, Boston, Massachusetts 39 Division of Hematology and Oncology, Center for Cell and Gene Therapy, Baylor College of Medicine, Houston, Texas 40 Division of Hematology-Oncology, Blood and Marrow Transplantation Program, Mayo Clinic, Jacksonville, Florida 41 Division of Bone Marrow Transplant and Immune Deficiency, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio 42 Department of Pediatrics, University of Cincinnati, Cincinnati, Ohio 43 Divison of Clinical Hematology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain 44 Department of Blood & Marrow Transplant and Cellular Immunotherapy (BMT CI), Moffitt Cancer Center, Tampa, Florida 45 Blood & Marrow Transplant Program, Kids Cancer Centre, Sydney Children's Hospital, Sydney, Australia 46 Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden 47 Centre for Clinical Research Sormland, Uppsala University, Uppsala, Sweden 48 Blood and Marrow Transplant Program, University of Utah, Salt Lake City, Utah 49 Adult Bone Marrow Transplant Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York 50 Blood and Cancer Centre, Starship Children's Hospital, Auckland, New Zealand 51 Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia 52 Dana Farber Cancer Institute, Boston, Massachusetts 53 Department of Hematology, University Hospitals Leuven and KU Leuven, Leuven, Belgium 54 Department of Hematology and Oncology, Dokkyo Medical University, Tochigi, Japan 55 Department of Bone Marrow Transplantation and Cellular Therapy, St. Jude Children's Research Hospital, Memphis, Tennessee 56 The Blood and Marrow Transplant Group of Georgia, Northside Hospital, Atlanta, Georgia 57 Rutgers Cancer Institute of New Jersey, Rutgers University, Brunswick, New Jersey 58 Hokkaido University Hospital, Sapporo, Japan 59 Department of Hematology, Hospital Clinic, University of Barcelona, IDIBAPS, and Institute of Research Josep Carreras, Barcelona, Spain 60 Maastricht University Medical Center, Maastricht, the Netherlands 61 Division of Hematology and Oncology, Washington University School of Medicine, St. Louis, Missouri 62 Department of Medical Oncology, Thomas Jefferson University, Philadelphia, Pennsylvania 63 Division of Hematology, The Ohio State University, Columbus, Ohio 64 Bone Marrow Transplant Program, Penn State Cancer Institute, Hershey, Pennsylvania 65 Blood & Marrow Transplantation Program, Division of Hematology/Oncology, Department of Medicine, Greenebaum Comprehensive Cancer Center, University of Maryland, Baltimore, Maryland Article history: A B S T R A C T Received 8 January 2020 Post-transplant cyclophosphamide (PTCy) has significantly increased the successful use of haploidentical donors with a Accepted 4 May 2020 relatively low incidence of graft-versus-host disease (GVHD). Given its increasing use, we sought to determine risk factors for GVHD after haploidentical hematopoietic cell transplantation (haplo-HCT) using PTCy. Data from the Center for Inter- national Blood and Marrow Transplant Research on adult patients with acute myeloid leukemia, acute lymphoblastic leu- kemia, myelodysplastic syndrome, or chronic myeloid leukemia who underwent PTCy-based haplo-HCT (2013 to 2016) were analyzed and categorized into 4 groups based on myeloablative (MA) or reduced-intensity conditioning (RIC) and bone marrow (BM) or peripheral blood (PB) graft source. In total, 646 patients were identified (MA-BM = 79, MA- PB = 183, RIC-BM = 192, RIC-PB = 192). The incidence of grade 2 to 4 acute GVHD at 6 months was highest in MA-PB (44%), followed by RIC-PB (36%), MA-BM (36%), and RIC-BM (30%) (P = .002). The incidence of chronic GVHD at 1 year was 40%, 34%, 24%, and 20%, respectively (P < .001). In multivariable analysis, there was no impact of stem cell source or condi- tioning regimen on grade 2 to 4 acute GVHD; however, older donor age (30 to 49 versus <29 years) was significantly asso- ciated with higher rates of grade 2 to 4 acute GVHD (hazard ratio [HR], 1.53; 95% confidence interval [CI], 1.11 to 2.12; P = .01). In contrast, PB compared to BM as a stem cell source was a significant risk factor for the development of chronic GVHD (HR, 1.70; 95% CI, 1.11 to 2.62; P = .01) in the RIC setting. There were no differences in relapse or overall survival between groups. Donor age and graft source are risk factors for acute and chronic GVHD, respectively, after PTCy-based haplo-HCT. Our results indicate that in RIC haplo-HCT, the risk of chronic GVHD is higher with PB stem cells, without any difference in relapse or overall survival. © 2020 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. All rights reserved. haploidentical HCT, approaching those in the matched For adults with advanced hematologic malignancies, related and unrelated donor setting [4,5]. Studies evaluat- haploidentical donors allow patients without a matched ing T cell replete haploidentical HCT using PTCy have dem- related or unrelated donor the opportunity to proceed with onstrated rates of acute GVHD in the range of 14% to 41% a potentially curative allogeneic hematopoietic cell trans- and chronic GVHD rates of 0% to 31%, compared to histori- plant (HCT) [1-3]. The use of post-transplant cyclophospha- cal rates of 20% to 80% and 30% to 70%, respectively, in mide (PTCy) as graft-versus-host disease (GVHD) matched related and matched unrelated donor transplants prophylaxis has significantly improved outcomes for [6-8]. Most of these halpotransplants were done using A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 1461 reduced-intensity conditioning (RIC) regimens with bone centers with NRM as a competing event. NRM was defined as death without marrow (BM) as a graft source [3-5,9-21]. evidence of disease recurrence, and relapse was considered a competing event. For relapse and NRM, patients alive in continuous complete remission Peripheral blood (PB) grafts have also been increasingly were censored at last follow-up. For GVHD, death without the event was con- used in the haploidentical setting, and there have been 2 retro- sidered a competing event. HLA matching was defined as described previ- spective analyses comparing BM and PB. A small study by ously [30]. O’Donnell et al. [14] demonstrated no differences in acute GVHD at 100 days, chronic GVHD at 2 years, or overall survival Statistical Methods (OS) at 2 years between BM (33%, 23%, 58%, respectively) and Patient, disease, and transplant-related variables for donor types were PB (40%, 19%, 66%, respectively). However, the rate of relapse compared using chi-square statistics for categorical variables and the Krus- was higher with BM compared to PB (19% versus 49%). In con- kal-Wallis test for continuous variables. Univariate analysis with Gray’s test and log-rank test was used for cumulative incidence and survival, respec- trast, a large Center for International Blood and Marrow Trans- tively. Variables tested included patient and sex, Karnofsky performance plant Research (CIBMTR) analysis by Bashey et al. [9] score, hematopoietic cell transplantation comorbidity index (HCT-CI) [31], demonstrated a significantly lower incidence of grade 2 to 4 DRI, time from diagnosis to HCT, donor age and relation (parent, offspring, acute GVHD (hazard ratio [HR], 0.45; P < .001) and chronic sibling), donor/recipient sex, donor/recipient cytomegalovirus status, use of TBI, and year of transplant. Patients were categorized into 4 groups based on GVHD (HR, 0.35; P < .001) in BM compared to PB. Relapse was conditioning intensity (myeloablative or reduced intensity) and graft source also higher with BM compared to PB (HR, 1.49; P = .009) specif- (BM or PB). Graft source and conditioning intensity were included in all mod- ically in patients with leukemia, but there were no differences els. The primary comparisons were RIC-BM versus RIC-PB and MAC-BM ver- in OS or nonrelapse mortality (NRM). sus MAC-PB. Multivariable models were built using the Cox proportional hazards Although the incidences of GVHD after haploidentical HCT model. All variables were tested to affirm the assumption of proportional haz- have been described, there are little data describing risk factors ards, and no variables violated the proportional hazards assumption. A step- for GVHD in this setting. The aim of this study is to describe the wise model building procedure was used to select the adjusted factors for each incidence, characteristics, and risk factors for acute and chronic outcome with a threshold of 0.05 for both entry and stay in the model. The “center” effect was adjusted for all endpoints. Two-way interactions between GVHD in adult patients with hematologic malignancies who “donor type” and the adjusted clinical variables in the models were tested, and underwent a PTCy-based haploidentical HCT from 2013 to 2016. no significant interactions were detected. A threshold Pvalue of .05 for the pri- mary endpoint of chronic GVHD and of .01 otherwise was used for significance MATERIALS AND METHODS of the main testing variable. A threshold of Pvalue of .01/2 = .005 (or .05/ Data Source 2 = .025 for the primary endpoint of chronic GVHD) was recommended for the This was a retrospective analysis using data from the CIBMTR. The CIBMTR significance of pairwise comparisons when the main variable was significant. is a combined research program of the Medical College of Wisconsin and the The impact of chronic GVHD on relapse was analyzed by treating chronic National Marrow Donor Program. The CIBMTR comprises a voluntary network GVHD as a time-dependent covariate in the model for relapse. Results were of more than 420 transplantation centers worldwide that contribute data on expressed as a hazard ratio with 95% confidence intervals. SAS software ver- consecutive allogeneic and autologous HCTs to a centralized statistical center. sion 9.4 (SAS Institute, Cary, NC) was used in the analysis. Observational studies conducted by the CIBMTR are performed in compliance with all applicable federal regulations pertaining to the protection of human research participants. Protected health information used in the performance of RESULTS such research is collected and maintained in the capacity of the CIBMTR as a public health authority under Health Insurance Portability and Accountability Patients, Disease, and Transplant Characteristics Act regulations. The CIBMTR collects data at 2 levels: Transplant-Essential Data Table 1 shows the characteristics of the study population (TED) and Comprehensive Report Form (CRF). The TED-level data are an inter- (n = 646) grouped by graft source and conditioning intensity. nationally accepted standard of data that contain key variables for all consecu- As expected, the median age of patients undergoing MAC con- tive transplant recipients in the United States. When a transplant is registered with the CIBMTR, a subset of patients is selected for the CRF level of data col- ditioning (45 and 42 years for MAC-BM and MAC-PB, respec- lection. The CRF-level data capture additional data related to the patient, dis- tively) was younger than those undergoing RIC (60 and ease, and transplant. Thus, a greater number of patients contribute to TED- 62 years for RIC-BM and RIC-PB, respectively). More patients level data compared to CRF data. Additional details regarding the CIBMTR reg- receiving PB grafts had a higher HCT-CI (score 3) than those istry have been previously described [22]. receiving BM: 52% MAC-PB compared to 43% MAC-BM and Patients 62% RIC-PB compared to 42% RIC-BM. Likewise, more patients Eligible patients were 18 years with acute myeloid leukemia, acute with a low HCT-CI (score 0) received BM (25% MAC-BM com- lymphoblastic leukemia, chronic myeloid leukemia, and myelodysplastic syn- pared to 12% MAC-PB and 24% RIC-BM compared to 14% RIC- drome. All patients underwent HCT from a haploidentical donor (defined in PB). As expected, patients receiving PB had overall higher the CIBMTR registry as 1 or more antigen-level mismatch among HLA-A, -B, CD34 and CD3 cell doses compared to BM. Forty-seven percent -C, and -DRB1) using BM or PB as a graft source and GVHD prophylaxis with PTCy, tacrolimus, and mycophenolate mofetil; other PTCy-based prophylaxis in the MAC groups received TBI as part of conditioning, with a regimens were excluded to limit heterogeneity and due to low numbers of median dose of 1200 (range, 550 to 1350) while 89% in the RIC these regimens. Conditioning regimens included myeloablative conditioning groups received TBI, with a median dose of 200 (range, 200 to (MAC) or RIC, with or without total body irradiation (TBI) [23,24]. An RIC regi- 400). Median follow-up was longer in the RIC groups com- men was defined as (a) TBI dose of 500 cGy as a single fraction or  800 cGy if fractionated, (b) <9 mg/kg oral busulfan or intravenous equivalent, or (c) pared to the MAC groups: 24 (range, 2 to 53) and 21 (range, 6 <140 mg/m2 melphalan. Transplants using ex vivo T cell depletion, anti-thy- to 50) months in RIC-BM and RIC-PB, respectively, versus 13 mocyte globulin (ATG), or alemtuzumab were excluded. (range, 4 to 49) and 14 (range, 6 to 50) months in MAC-BM and MAC-PB, respectively. Study Endpoints and Definitions The primary endpoints of this study were incidence of grade 2 to 4 acute GVHD and chronic GVHD. Secondary endpoints included grade 3 to 4 acute GVHD and Engraftment GVHD, relapse, OS, NRM, and the composite endpoints of GVHD relapse-free survival (GRFS) (including survival without grade 3 to 4 acute GVHD, chronic In univariate analysis, the incidence of grade 2 to 4 acute GVHD requiring systemic treatment, relapse, or death) and chronic GVHD GVHD at 100 days was highest in MAC-PB at 46% (95% CI, 39% relapse-free survival (CRFS) (including survival without chronic GVHD to 54%), followed by RIC-PB at 36% (95% CI, 29% to 43%), MAC- requiring systemic treatment, relapse, or death) [25]. GVHD was graded BM at 33% (95% CI, 23% to 44%), and RIC-BM at 27% (95% CI, according to consensus criteria [7,26,27]. Disease status was categorized into early, intermediate, and advanced [28], and the revised Disease Risk Index 21% to 33%) (P = .002). A similar pattern was observed for (DRI) was used to categorize patients into low, intermediate, and high/very chronic GVHD, where the incidence at 1 year was highest in high groups [29]. Relapse was defined by hematologic criteria by submitting MAC-PB at 40% (95% CI, 32% to 47%), followed by RIC-PB at 34% 1462 A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 Table 1 Baseline Characteristics Variable MAC-BM (n = 79) MAC-PB (n = 183) RIC-BM (n = 192) RIC-PB (n = 192) P Value Recipient age at transplant, yr <.0001* Median (range) 45 (18-72) 42 (18-71) 60 (18-77) 62 (18-76) <.0001y 18-39 35 (44) 78 (43) 36 (19) 21 (11) 40-59 24 (30) 77 (42) 56 (29) 55 (28) 60+ 20 (26) 28 (16) 100 (52) 116 (60) Recipient sex .82y Male 46 (58) 108 (59) 121 (63) 114 (59) Female 33 (42) 75 (41) 71 (37) 78 (41) Recipient race .03y Caucasian 57 (72) 116 (63) 138 (72) 139 (72) African American 13 (16) 56 (31) 34 (18) 38 (20) Asian/Pacific Islander 3 (4) 4 (2) 12 (6) 12 (6) Other/missing 6 (8) 7 (3) 8 (4) 3 (1) Karnofsky performance score prior to transplant .002y <90 30 (38) 96 (52) 77 (40) 101 (53) 90-100 43 (54) 85 (46) 112 (58) 87 (45) Missing 6 (8) 2 (1) 3 (2) 4 (2) Sorror HCT-CI <.0001y 0 20 (25) 22 (12) 46 (24) 26 (14) 1-2 25 (32) 64 (35) 62 (32) 43 (22) 3+ 34 (43) 95 (52) 81 (42) 119 (62) Missing 0 2 (1) 3 (2) 4 (2) Disease <.0001y AML 39 (49) 115 (63) 94 (49) 111 (58) ALL 22 (28) 46 (25) 34 (18) 22 (11) CML 4 (5) 10 (5) 12 (6) 3 (2) MDS 14 (18) 12 (7) 52 (27) 56 (29) Disease Risk Index AML 39 115 94 111 .10y Low 2 (3) 11 (6) 7 (4) 7 (4) Intermediate 20 (25) 57 (31) 66 (34) 69 (36) High/very high 15 (19) 44 (24) 18 (9) 29 (15) Missing 2 (3) 3 (2) 3 (2) 6 (3) ALL 22 46 34 22 .49y Intermediate 15 (19) 23 (13) 20 (10) 14 (7) High/very high 7 (9) 23 (13) 14 (7) 8 (4) CML 4 10 12 3 .03y Low 4 (5) 9 (5) 12 (6) 1 (1) Intermediate 0 1 (1) 0 1 (1) High/very high 0 0 0 1 (1) MDS 14 12 52 56 .15y Intermediate 7 (9) 4 (2) 27 (14) 14 (7) High/very high 6 (8) 7 (4) 23 (12) 37 (19) Missing 1 (1) 1 (1) 2 (1) 5 (3) HLA matching .19y Haploidentical (1-antigen mismatch) 5 (6) 8 (4) 6 (3) 10 (5) Haploidentical (2-antigen mismatches) 74 (94) 173 (95) 181 (94) 182 (95) Haploidentical (mismatch number unknown) 0 2 (1) 5 (3) 0 Donor type < .001y Parent donor 11 (14) 37 (20) 14 (7) 9 (5) Offspring donor 30 (38) 58 (32) 120 (63) 124 (65) Sibling donor 34 (43) 88 (48) 56 (29) 54 (28) Missing 4 (5) 0 2 (1) 5 (3) (continued) A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 1463 Table 1 (Continued) Variable MAC-BM (n = 79) MAC-PB (n = 183) RIC-BM (n = 192) RIC-PB (n = 192) P Value Donor age, yr < .001y Median (range) 38 (12-65) 37 (9-68) 37 (8-73) 39 (8-71) .44* <29 25 (32) 52 (28) 55 (29) 43 (22) 30-49 38 (48) 70 (38) 98 (51) 112 (58) 50 12 (15) 60 (33) 35 (18) 34 (18) Missing 4 (5) 1 (1) 4 (2) 3 (2) Time from diagnosis to transplant, mo .13y Median (range) 6 (3-180) 8 (2-144) 9 (<1-291) 8 (2-171) .16* <6 38 (48) 75 (41) 58 (30) 77 (40) 6 to <12 18 (23) 47 (26) 64 (33) 52 (27) 12 23 (29) 61 (33) 70 (36) 63 (33) CD34 cell dose, £ 106/kg <.001y Median (range) 3 (0-8) 5 (0-21) 3 (0-13) 6 (0-20) <.001* <2 28 (35) 4 (2) 49 (26) 8 (4) 2 to <4 31 (39) 24 (13) 91 (47) 20 (10) 4 to <8 17 (22) 96 (52) 37 (19) 107 (56) 8 0 33 (18) 4 (2) 44 (23) Missing 3 (4) 26 (14) 11 (6) 13 (7) CD3 cell dose, £ 106/kg <.001y Median (range) 0 (0-1) 2 (0-11) 0 (0-12) 2 (0-14) <.001* <2 55 (70) 48 (26) 120 (63) 70 (36) 2 to <4 0 54 (30) 0 73 (38) 4 to <8 0 18 (10) 0 12 (6) 8 0 2 (1) 1 (1) 1 (1) Missing 24 (30) 61 (33) 71 (37) 36 (19) Donor-recipient sex mismatch .69y M/M 34 (43) 62 (34) 82 (43) 74 (39) M/F 21 (27) 47 (26) 46 (24) 45 (23) F/M 12 (15) 46 (25) 39 (20) 40 (21) F/F 12 (15) 28 (15) 25 (13) 33 (17) Donor-recipient CMV status .03y / 16 (20) 31 (17) 48 (25) 37 (19) /+ 27 (34) 41 (22) 66 (34) 62 (32) +/ 7 (9) 16 (9) 16 (8) 14 (7) +/+ 29 (37) 90 (49) 61 (32) 78 (41) Missing 0 5 (3) 1 (1) 1 (1) Conditioning regimen NA BU + FLU/CY (MAC) 35 (44) 80 (44) 0 0 TBI § FLU/CY (MAC) 24 (30) 100 (55) 0 0 TBI § FLU/CY (RIC) 0 0 159 (83) 167 (87) FLU + MEL (RIC) 0 0 29 (15) 19 (10) Others 20 (25) 3 (2) 4 (2) 6 (3) TBI dose, cGy Median (range) 1200 (200-1350) 1200 (200-1200) 200 (200-300) 200 (200-400) <.001* Year of transplant < .001y 2013 6 (8) 11 (6) 52 (27) 13 (7) 2014 16 (20) 38 (21) 48 (25) 39 (20) 2015 20 (25) 61 (33) 40 (21) 69 (36) 2016 37 (47) 73 (40) 52 (27) 71 (37) Follow-up of survivors, median (range), mo 13 (4-49) 14 (6-50) 24 (2-53) 21 (6-50) Values are presented as number (%) unless otherwise indicated. AML indicates acute myelogenous leukemia; ALL, acute lymphoblastic leukemia; CML, chronic myelogenous leukemia; MDS, myelodysplastic syndrome; NA, not applicable; BU, busulfan; FLU, fludarabine; CY, cyclophosphamide; MEL, melphalan. * Kruskal-Wallis test. y Pearson chi-square test. 1464 A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 Table 2 Univariate Analysis of Transplant Outcomes by Stem Cell Source and Conditioning Regimen Intensity MAC-BM MAC-PB RIC-BM RIC-PB (n = 79) (n = 183) n = 192) (n = 192) Outcomes N (95% CI) N (95% CI) N (95% CI) N (95% CI) PValue Grade 2-4 acute GVHD at 6 months 36 (25-46) 46 (39-54) 30 (23-36) 36 (30-43) .002 Chronic GVHD at 1 year 24 (15-35) 40 (32-47) 20 (14-26) 34 (27-41) <.001 Grade 3-4 acute GVHD 13 (6-21) 14 (10-20) 5 (3-9) 9 (6-14) .06 Relapse at 1 year 28 (18-39) 26 (20-33) 37 (30-44) 36 (29-43) .16 Nonrelapse mortality at 1 year 14 (7-23) 18 (12-24) 9 (5-13) 18 (13-24) .01 GVHD-free, relapse-free survival at 1 year 38 (28-49) 23 (17-30) 41 (34-49) 25 (19-32) .002 Chronic GVHD-free, relapse-free survival at 1 year 42 (31-53) 26 (20-33) 43 (36-50) 26 (19-32) .002 Overall survival at 1 year 67 (56-77) 64 (57-71) 70 (63-76) 58(51-65) .07 Neutrophil recovery at 100 days* 92 (85-97) 95 (91-98) 94 (90-97) 93 (89-96) .004 Platelet recovery at 100 days 78 (67-86) 85 (79-90) 90 (85-94) 89 (84-93) .10 * Number of graft failures: MAC-BM (n = 1), MAC-PB (n = 4), RIC-BM (n = 4), and RIC-PB (n = 6). Table 3 Table 3 (Continued) Multivariate Analysis of Transplant Outcomes by Stem Cell Source and Condi- tioning Regimen Intensity Outcome Variable HR (95% CI) P Value Primary endpoints Outcome Variable HR (95% CI) P Value RIC-PB vs RIC- 1.23 (0.94-1.61) .12 Primary endpoints BM Acute GVHD Overall survival grades 2-4 MAC-PB vs 1.02 (0.64-1.62) .93 MAC-PB vs 1.10 (0.66-1.82) .73 MAC-BM MAC-BM RIC-PB vs RIC- 1.17 (0.86-1.61) .32 RIC-PB vs RIC- 1.24 (0.84-1.82) .28 BM BM Bold values represent statistically significant values Chronic GVHD MAC-PB vs 1.56 (0.86-2.82) .14 MAC-BM (95% CI, 27% to 41%), MAC-BM at 24% (95% CI, 15% to 35%) and RIC-PB vs 1.70 (1.11-2.62) .01 RIC-BM at 20% (95% CI, 14% to 26%) (P < .001) (Table 2). RIC-BM In multivariable analysis, there was no significant differ- Secondary endpoints ence in grade 2 to 4 acute GVHD or grade 3 to 4 acute GVHD between conditioning and graft source groups (Table 3). Acute GVHD grades 3-4 Adjusted cumulative incidence of grade 2 to 4 acute GVHD MAC-PB vs 0.78 (0.38-1.60) .49 between all 4 conditioning and graft source groups is shown in MAC-BM Figure 1. The only significant factor for grade 2 to 4 acute RIC-PB vs RIC- 1.93 (0.94-3.96) .07 GVHD in multivariable analysis was donor age 30 to 49 versus BM <29 years (HR, 1.53; 95% CI, 1.11 to 2.12, P = .01). For grade 3 Relapse to 4 acute GVHD, older donor age was also the only significant MAC-PB vs 0.95 (0.59-1.52) .82 factor (50 versus <29 years; HR, 3.89; 95% CI, 1.81 to 8.35; MAC-BM P = .0005). Despite the impact of donor age, donor relation was RIC-PB vs RIC- 0.90 (0.65-1.26) .55 not a significant factor for acute GVHD. There was a signifi- BM cantly higher risk for chronic GVHD in the RIC-PB group com- Nonrelapse pared to RIC-BM (HR, 1.70; 95% CI, 1.11 to 2.62; P = .01). There mortality MAC-PB vs 1.31 (0.62-2.78) .48 Acute GVHD, Grades II-IV MAC-BM 100 p=0.07 Adjusted Cumulative Incidence, % RIC-PB vs 2.06 (1.15-3.68) .01 MA-BM (n=79) RIC-BM 80 MA-PB (n=180) GVHD-free, RIC-BM (n=191) relapse-free RIC-PB (n=191) 60 survival MAC-PB vs 1.32 (0.94-1.85) .11 40 MAC-BM RIC-PB vs RIC- 1.31 (1.02-1.69) .03 20 BM Chronic GVHD- 0 free, 0 2 4 6 8 10 12 relapse-free Months survival MAC-PB vs 1.36 (0.94-1.98) .10 Figure 1. Incidence of acute GVHD grades 2 to 4 by conditioning intensity and MAC-BM graft source. (continued) A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 1465 Chronic GVHD Relapse 100 p=0.009 100 p=0.14 Adjusted Cumulative Incidence, % Adjusted Cumulative Incidence, % MA-BM (n=79) MA-BM (n=79) 80 MA-PB (n=181) 80 MA-PB (n=183) RIC-BM (n=191) RIC-BM (n=192) RIC-PB (n=190) RIC-PB (n=192) 60 60 40 40 20 20 0 0 0 6 12 18 24 30 0 1 2 3 Months Years Figure 2. Incidence of chronic GVHD by conditioning intensity and graft Figure 3. Incidence of relapse by conditioning intensity and graft source. source. 1 to 105). There was no difference in platelet recovery at was no difference in chronic GVHD between MAC-PB and 100 days between groups (Table 2). Given the small numbers, MAC-BM (Table 3). Adjusted cumulative incidence curves for there was no notable trend regarding graft failure: 6 in RIC-PB, chronic GVHD are shown in Figure 2. There were also no differ- 4 in RIC-BM, 4 in MAC-PB, and 1 in MAC-BM. ences in chronic GVHD severity among any of the groups; overall, 59% were categorized as mild, 28% were moderate, Relapse and Survival Outcomes and 13% were severe (Table 4). Relapse at 1 year was similar among the groups (Table 2; To further determine the significance of graft source on adjusted cumulative incidence shown in Figure 3). In multivar- chronic GVHD, subset analyses of MAC (n = 262) and RIC iable analysis, revised DRI high/very high versus low (HR, 2.11; (n = 384) cohorts were performed. This confirmed an increased 95% CI, 1.21 to 3.68; P = .008) was the only significant factor incidence of chronic GVHD with PB compared to BM in both MAC (HR, 1.81; 95% CI, 1.00 to 3.28; P = .05) and RIC (HR, 1.72; Non-Relapse Mortality 95% CI, 1.10 to 2.70; P = .02) groups, but this met the predeter- 100 p=0.0292 Adjusted Cumulative Incidence, % mined significance level only in the RIC group. Due to the MA-BM (n=78) known differences in age between MAC and RIC groups, an 80 MA-PB (n=176) age-adjusted analysis for chronic GVHD was also performed RIC-BM (n=181) and demonstrated similar results (data not shown). We further RIC-PB (n=185) 60 analyzed chronic GVHD in a larger cohort of CIBMTR patients receiving a haploidentical transplant (n = 1401) using TED 40 level data and confirmed that in the RIC setting, PB was associ- ated with higher rates of chronic GVHD (P = .0015) (Table 5). 20 Regarding hematopoietic recovery, neutrophil recovery at 0 28 days was highest in MAC-PB (95%), followed by RIC-BM 0 1 2 3 (94%), RIC-PB (93%), and MAC-BM (92%) (P = .004). Time to neu- Years trophil recovery was comparable across the groups: 19 days in MAC-BM (range, 5 to 125), 16 days in MAC-PB (range, 1 to 90), Figure 4. Incidence of nonrelapse mortality by conditioning intensity and 18 days in RIC-BM (range, 2 to 48), and 17 days in RIC-PB (range, graft source. Table 4 Chronic GVHD Severity by Stem Cell Source and Conditioning Regimen Characteristic MAC-BM (n = 19), No. (%) MAC-PB (n = 70), No. (%) RIC-BM (n = 43), No. (%) RIC-PB (n = 65), No. (%) P Value Chronic GVHD severity .44 Mild 13 (68) 33 (47) 26 (60) 44 (68) Moderate 5 (26) 25 (36) 12 (28) 13 (20) Severe 1 (5) 11 (16) 5 (12) 8 (12) Missing 1 (1) Table 5 Multivariate Analysis of Chronic GVHD by Stem Cell Source and Conditioning Regimen Intensity Based on TED-Level Data (n = 1401) Outcome Variable HR (95% CI) P Value Chronic GVHD MAC-PB vs MAC-BM 1.44 (0.96-2.16) .08 RIC-PB vs RIC-BM 1.75 (1.24-2.47) .0015 Bold values represent statistically significant values 1466 A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 GVHD-Free, Relapse-Free Survival OS at 1 year in univariate analysis was similar among all groups (Table 2). In multivariable analysis, patient age 50 to 59 100 p=0.0467 MA-BM (n=78) (HR, 1.73; 95% CI, 1.12 to 2.68; P = .013) or 60 years (HR, 1.84; 95% CI, 1.20 to 2.82; P = .005) versus 18 to 29 years, donor age Adjusted Probability, % 80 MA-PB (n=173) RIC-BM (n=183) 50 versus <29 years (HR, 1.77; 95% CI, 1.24 to 2.52; P = .002), RIC-PB (n=186) and revised DRI high/very high versus low (HR, 2.68; 95% CI, 1.46 60 to 4.90; P = .001) were significant factors contributing to OS. 40 Relapse was the leading cause of death in all groups (Table 6). 20 DISCUSSION 0 This study demonstrates that donor age and graft source 0 1 2 3 are important risk factors for acute and chronic GVHD, respec- Years tively, after PTCy-based haploidentical HCT. Similar to previous Figure 5. GVHD-free, relapse-free survival by conditioning intensity and graft findings in both matched [32-34] and alternative [35] donor source. settings, PB grafts in RIC haplotransplant were found to be sig- nificantly associated with chronic GVHD of any severity. Although the impact of specific components of transplant is for relapse. We further analyzed the impact of chronic GVHD historically studied as distinct entities [36], we report a signifi- on relapse and found no significant associations between cant association of graft source and conditioning (RIC-PB) on chronic GVHD and relapse in the entire cohort (HR, 1.00; 95% chronic GVHD. CI, 0.70 to 1.44; P = .99) or in a subset analysis of MAC (P = .80) Previous studies have reported incidences of acute and and RIC (P = .75) cohorts. chronic GVHD after PTCy-based haploidentical HCT with vary- NRM at 1 year was highest in RIC-PB at 18% (95% CI, 13% to ing combinations of stem cell sources and conditioning regi- 24%) and MAC-PB at 18% (95% CI, 12% to 24%), followed by 14% mens, but only 2 prior studies have compared BM and PB (95% CI, 7% to 23%) in MAC-BM and 9% (95% CI, 5% to 13%) in grafts. O’Donnell et al. [14] reported no difference in GVHD or RIC-BM (P = .01) (Table 2). In multivariable analysis, there was OS but lower relapse with PB, whereas Bashey et al. [9] dem- a higher risk of NRM in RIC-PB compared to RIC-BM (HR, 2.06; onstrated higher incidences of acute and chronic GVHD, as 95% CI, 1.15 to 3.68; P = .01). There was no difference in NRM well as lower relapse with PB, without a difference in OS. Our between MAC-PB and MAC-BM. Adjusted cumulative inci- study confirms the finding of increased chronic GVHD with PB dence of NRM between groups (P = .0292) is shown in Figure 4. grafts but only in the RIC setting. Despite the incidence of The other significant factors for NRM were older patient age chronic GVHD at 1 year being highest in the MAC-PB cohort, 60 versus 18 to 29 years (HR, 2.31; 95% CI, 1.18 to 4.54; there was no difference confirmed in multivariable analysis in P = .015) and higher HCT-CI 3 versus 0 (HR, 3.32; 95% CI, 1.50 the myeloablative setting. While GRFS and CFRS were inferior to 7.35; P = .003). in the PB groups in both RIC and MAC groups, there was no dif- GRFS and CRFS at 1 year were lowest in the PB groups for ference in risk of relapse or OS between any of the groups. both RIC (25% and 26% respectively, compared to 41% and 43% These results suggest that PTCy does not fully negate the risk in BM groups) and MAC (23% and 26%, compared to 38% and of chronic GVHD using PB in the RIC setting. Moreover, no 42%) (Table 2), but in multivariable analysis, these outcomes other factors emerged as significant risk factors for develop- were not significantly impacted by graft source or conditioning ment of chronic GVHD after haploidentical HCT with PTCy. regimen (adjusted probability, Figure 5). Donor age 50 versus Notably, the higher risk of chronic GVHD with the use of PB <29 years (HR, 1.43; 95% CI, 1.08 to 1.88; P = .012) and Karnof- was not offset by a lower relapse rate or difference in overall sky performance score 90 to 100 versus <90 (HR, 0.74; 95% CI, survival. The ability to detect these differences in relapse and 0.61 to 0.90; P = .003) were significant factors for GRFS, survival is likely to be limited by power. There was also no whereas patient age 50 to 59 versus 18 to 29 years (HR, 1.53; increase in graft failure with BM compared to PB grafts, with a 95% CI, 1.09 to 1.96; P = .013) and patient race non-Caucasian similar time to neutrophil engraftment in all groups. versus Caucasian (HR, 1.33; 95% CI, 1.07 to 1.65; P = .010) were The effect of both graft source and conditioning was also significant factors for CRFS. confirmed in our analyses of a larger cohort of CIBMTR patients Table 6 Causes of Death MAC-BM(n = 79) MAC-PB(n = 183) RIC-BM(n = 192) RIC-PB(n = 192) No. of dead patients 32 75 84 99 Cause of death, No. (%) Primary disease 16 (50) 35 (47) 60 (71) 45 (45) Graft failure 1 (3) 1 (1) 2 (2) 0 GVHD 1 (3) 7 (9) 4 (5) 3 (3) Infection 3 (9) 6 (8) 3 (4) 2 (2) IPn/ARD 6 (19) 12 (16) 8 (10) 16 (16) Organ failure 2 (6) 6 (8) 3 (4) 17 (17) Secondary malignancy 1 (3) 8 (11) 3 (4) 14 (14) Others 2 (6) 0 1 (1) 2 (2) IPn/ARDS is interstitial pneumonitis and acute respiratory distress syndrome A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 1467 using TED-level data receiving haplotransplant. Thus, although Prospective evaluation of PB versus BM using PTCy platforms conditioning intensity is usually determined by a patient’s age, is needed to confirm these findings. comorbidities, and other nonmodifiable factors, graft source is a potentially modifiable variable. ACKNOWLEDGMENTS It is also notable that overall incidence of grade 3 to 4 acute Annie Im, Armin Rashidi, and Betty K. Hamilton contributed GVHD was low, especially in the RIC setting, a similar finding to equally to this study. what has been recently reported by McCurdy et al. [37], suggest- The CIBMTR is supported primarily by Public Health Service ing that PTCy immunomodulation decreases the risk of severe grant/cooperative agreement U24CA076518 with the National acute GVHD without reduction of grade 2 GVHD, which was sig- Cancer Institute (NCI), the National Heart, Lung, and Blood Insti- nificantly associated with higher progression-free survival. In tute (NHLBI), and the National Institute of Allergy and Infectious contrast to this recent analysis, we demonstrate older donor age Diseases (NIAID); grant/cooperative agreement U24HL138660 as a significant risk factor for both grade 2 to 4 and grade 3 to 4 with NHLBI and NCI; grants OT3HL147741, R21HL140314, and acute GVHD, as opposed to donor relation, which did not have an U01HL128568 from the NHLBI; contract HHSH250201700006C impact. We were unable to evaluate cell doses for this analysis. with Health Resources and Services Administration (HRSA); Of note, the distribution of donor age between RIC and MAC was grants N00014-18-1-2888 and N00014-17-1-2850 from the similar. This may have implications on donor selection when Office of Naval Research; subaward from prime contract award there is more than 1 haploidentical donor available (eg, older sib- SC1MC31881-01-00 with HRSA; subawards from prime grant ling versus younger offspring). Other studies have also shown awards R01HL131731 and R01HL126589 from NHLBI; donor age to be a factor in outcomes after haploidentical trans- subawards from prime grant awards 5P01CA111412, plant, and our findings support this as a recommended consider- 5R01HL129472, R01CA152108, 1R01HL131731, 1U01AI126612, ation in donor selection [38-40]. and 1R01CA231141 from the National Institutes of Health; and There are several limitations that should be considered in commercial funds from Actinium Pharmaceuticals, Inc.; Adap- interpreting these results. First, this was a retrospective analy- tive Biotechnologies; Allovir, Inc.; Amgen, Inc.; anonymous sis based on data submitted to a registry, and thus factors that donation to the Medical College of Wisconsin; Anthem, Inc.; led to decisions regarding stem cell source and conditioning Astellas Pharma US; Atara Biotherapeutics, Inc.; BARDA; Be the regimen cannot be determined. We did restrict the analysis to Match Foundation; bluebird bio, Inc.; Boston Children’s Hospi- select conditioning and prophylaxis regimens to limit hetero- tal; Bristol Myers Squibb Co.; Celgene Corp.; Children’s Hospital geneity. In addition, although the use of the registry data of Los Angeles; Chimerix, Inc.; City of Hope Medical Center; CSL allowed for a large study population, it is possible that we did Behring; CytoSen Therapeutics, Inc.; Daiichi Sankyo Co., Ltd.; not have the power to elicit other risk factors contributing to Dana Farber Cancer Institute; Enterprise Science and Comput- GVHD or significant relationships in the observed trends. The ing, Inc.; Fred Hutchinson Cancer Research Center; Gamida-Cell, power to detect significant differences in chronic GVHD in our Ltd.; Genzyme; Gilead Sciences, Inc.; GlaxoSmithKline (GSK); subset analysis of MAC and RIC groups was just 48% and 66%, HistoGenetics, Inc.; Immucor; Incyte Corporation; Janssen Bio- respectively, and this may be why we were unable to detect tech, Inc.; Janssen Pharmaceuticals, Inc.; Janssen Research & significant differences in graft source in the MAC setting for Development, LLC; Janssen Scientific Affairs, LLC; Japan GVHD. Nevertheless, our findings of increased chronic GVHD Hematopoietic Cell Transplantation Data Center; Jazz Pharma- in the RIC groups were further confirmed by a larger cohort ceuticals, Inc.; Karius, Inc.; Karyopharm Therapeutics, Inc.; Kite, using TED-level data. This was one of the largest analyses of a Gilead Company; Kyowa Kirin; Magenta Therapeutics; Mayo this type, allowing the ability to evaluate conditioning regimen Clinic and Foundation Rochester; Medac GmbH; Mediware; and graft source as specific risk factors for GVHD after haploi- Memorial Sloan Kettering Cancer Center; Merck & Company, dentical HCT. Another limitation includes the duration of fol- Inc.; Mesoblast; MesoScale Diagnostics, Inc.; Millennium, the low-up for the MAC groups, which was only 13 and 14 Takeda Oncology Co.; Miltenyi Biotec, Inc.; Mundipharma EDO; months, compared to the RIC groups, which had 24 and 21 National Marrow Donor Program; Novartis Oncology; Novartis months of follow-up. Notably, the follow-up was similar Pharmaceuticals Corporation; Omeros Corporation; Oncoim- between the BM and the PB groups, and thus the subset analy- mune, Inc.; OptumHealth; Orca Biosystems, Inc.; PCORI; Pfizer, ses within the MAC and the RIC groups were not impacted by Inc.; Phamacyclics, LLC; PIRCHE AG; Regeneron Pharmaceuti- different follow-up periods. Although most relapse and cals, Inc.; REGiMMUNE Corp.; Sanofi Genzyme; Seattle Genetics; chronic GVHD occur within 1 year of transplant, longer follow- Shire; Sobi, Inc.; Spectrum Pharmaceuticals, Inc.; St. Baldrick's up, particularly in the MAC groups, is necessary, especially Foundation; Swedish Orphan Biovitrum, Inc.; Takeda Oncology; given that there was not a difference in chronic GVHD between The Medical College of Wisconsin; University of Minnesota; the MAC groups. In addition, we excluded patients (n = 24) University of Pittsburgh; University of Texas-MD Anderson; who received in vivo T cell depletion with ATG or alemtuzu- University of Wisconsin, Madison; Viracor Eurofins; and Xeni- mab, as it is known that these agents decrease the incidence of kos BV. The views expressed in this article do not reflect the chronic GVHD [41,42]. Patients who received both PTCy and in official policy or position of the National Institutes of Health, the vivo T cell depletion may potentially have had lower risks of Department of the Navy, the Department of Defense, HRSA, or chronic GVHD, although this was overall a small number of any other agency of the US government. patients. Finally, we included only patients with acute myeloid Conflict of interest statement: There are no conflicts of inter- leukemia, acute lymphoblastic leukemia, chronic myeloid leu- est to report. kemia, and myelodysplastic syndrome, and thus our findings Authorship statement: are applicable to only this patient population. A.I., A.R., and B.K.H. designed the study, analyzed results, In sum, our results show that PB stem cells contribute to an and wrote the manuscript; T.W. and M.H. provided statistical increased risk of chronic GVHD and NRM in RIC PTCy-based analysis and support; S.S. and M.A. critically reviewed design haploidentical HCT and that other outcomes such as relapse of study, provided data analysis, and reviewed the manuscript; and OS are similar. These results aid in decision making M.M., J.P., and M.J. provided input into interpretation of data regarding graft source in RIC PTCy-based haploidentical HCT. analysis and critically reviewed the manuscript; S.P., N.M., and 1468 A. Im et al. / Biol Blood Marrow Transplant 26 (2020) 14591468 D.W. reviewed design of study, provided data analysis, and leukemia: a retrospective single-center review. Biol Blood Marrow Trans- critically reviewed the manuscript; GVHD Working Committee plant. 2016;22(9):1696-1701. 20. Kanate AS, Mussetti A, Kharfan-Dabaja MA, et al. Reduced-intensity trans- members provided review of study design, analysis, and plantation for lymphomas using haploidentical related donors vs HLA- review of manuscript. matched unrelated donors. Blood. 2016;127(7):938-947. 21. Bashey A, Zhang X, Jackson K, et al. Comparison of outcomes of hematopoietic cell transplants from T-replete haploidentical donors REFERENCES using post-transplantation cyclophosphamide with 10 of 10 HLA-A, 1. O'Donnell PV, Luznik L, Jones RJ, et al. Nonmyeloablative bone marrow -B, -C, -DRB1, and -DQB1 allele-matched unrelated donors and HLA- transplantation from partially HLA-mismatched related donors using identical sibling donors: a multivariable analysis including Disease posttransplantation cyclophosphamide. Biol Blood Marrow Transplant. Risk Index. Biol Blood Marrow Transplant. 2016;22(1):125-133. 2002;8(7):377-386. 22. Horowitz M.The role of registries in facilitating clinical research in BMT: 2. Luznik L, O'Donnell PV, Symons HJ, et al. HLA-haploidentical bone marrow examples from the Center for International Blood and Marrow Transplant transplantation for hematologic malignancies using nonmyeloablative Research. Bone Marrow Transplant. 2008;42(suppl1):S1-S2. conditioning and high-dose, posttransplantation cyclophosphamide. Biol 23. Giralt S, Ballen K, Rizzo D, et al. Reduced-intensity conditioning regimen Blood Marrow Transplant. 2008;14(6):641-650. workshop: defining the dose spectrum. Report of a workshop convened 3. Brunstein CG, Fuchs EJ, Carter SL, et al. Alternative donor transplantation by the center for international blood and marrow transplant research. Biol after reduced intensity conditioning: results of parallel phase 2 trials using Blood Marrow Transplant.2009;15(3):367-369. partially HLA-mismatched related bone marrow or unrelated double 24. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning umbilical cord blood grafts. Blood. 2011;118(2):282-288. regimens: working definitions. Biol Blood Marrow Transplant. 2009;15 4. Bashey A, Zhang X, Sizemore CA, et al. T-cell-replete HLA-haploidentical (12):1628-1633. hematopoietic transplantation for hematologic malignancies using post- 25. Holtan SG, DeFor TE, Lazaryan A, et al. Composite end point of graft-ver- transplantation cyclophosphamide results in outcomes equivalent to sus-host disease-free, relapse-free survival after allogeneic hematopoietic those of contemporaneous HLA-matched related and unrelated donor cell transplantation. Blood. 2015;125(8):1333-1338. transplantation. J Clin Oncol. 2013;31(10):1310-1316. 26. Przepiorka D, Weisdorf D, Martin P, et al. 1994Consensus conference on 5. Ciurea SO, Zhang MJ, Bacigalupo AA, et al. Haploidentical transplant with acute GVHD grading. Bone Marrow Transplant.1995;15(6):825-828. posttransplant cyclophosphamide vs matched unrelated donor transplant 27. Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host for acute myeloid leukemia. Blood. 2015;126(8):1033-1040. syndrome in man: a long-term clinicopathologic study of 20 Seattle 6. Martin PJ, Rizzo JD, Wingard JR, et al. First- and second-line systemic treat- patients. Am J Med. 1980;69(2):204-217. ment of acute graft-versus-host disease: recommendations of the Ameri- 28. Armand P, Gibson CJ, Cutler C, et al. A disease risk index for patients under- can Society of Blood and Marrow Transplantation. Biol Blood Marrow going allogeneic stem cell transplantation. Blood. 2012;120(4):905-913. Transplant. 2012;18(8):1150-1163. 29. Armand P, Kim HT, Logan BR, et al. Validation and refinement of the Dis- 7. Jagasia MH, Greinix HT, Arora M, et al. National Institutes of Health con- ease Risk Index for allogeneic stem cell transplantation. Blood. 2014;123 sensus development project on criteria for clinical trials in chronic graft- (23):3664-3671. versus-host disease: I. The 2014 Diagnosis and Staging Working Group 30. Weisdorf D, Spellman S, Haagenson M, et al. Classification of HLA-match- report. Biol Blood Marrow Transplant. 2015;21(3):389-401.e1. ing for retrospective analysis of unrelated donor transplantation: revised 8. Zeiser R, Blazar BR.Pathophysiology of chronic graft-versus-host disease definitions to predict survival. Biol Blood Marrow Transplant. 2008;14 and therapeutic targets. N Engl J Med. 2017;377(26):2565-2579. (7):748-758. 9. Bashey A, Zhang MJ, McCurdy SR, et al. Mobilized peripheral blood stem 31. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation cells versus unstimulated bone marrow as a graft source for T-cell-replete (HCT)-specific comorbidity index: a new tool for risk assessment before haploidentical donor transplantation using post-transplant cyclophospha- allogeneic HCT. Blood. 2005;106(8):2912-2919. mide. J Clin Oncol. 2017;35(26):3002-3009. 32. Bensinger WI, Martin PJ, Storer B, et al. Transplantation of bone marrow as 10. Castagna L, Crocchiolo R, Furst S, et al. Bone marrow compared with compared with peripheral-blood cells from HLA-identical relatives in peripheral blood stem cells for haploidentical transplantation with a non- patients with hematologic cancers. N Engl J Med. 2001;344(3):175-181. myeloablative conditioning regimen and post-transplantation cyclophos- 33. Flowers ME, Parker PM, Johnston LJ, et al. Comparison of chronic graft-ver- phamide. Biol Blood Marrow Transplant. 2014;20(5):724-729. sus-host disease after transplantation of peripheral blood stem cells ver- 11. Di Stasi A, Milton DR, Poon LM, et al. Similar transplantation outcomes for sus bone marrow in allogeneic recipients: long-term follow-up of a acute myeloid leukemia and myelodysplastic syndrome patients with hap- randomized trial. Blood. 2002;100(2):415-419. loidentical versus 10/10 human leukocyte antigen-matched unrelated and 34. Schmitz N, Eapen M, Horowitz MM, et al. Long-term outcome of patients related donors. Biol Blood Marrow Transplant. 2014;20(12):1975-1981. given transplants of mobilized blood or bone marrow: a report from the 12. Kasamon YL, Bolanos-Meade J, Prince GT, et al. Outcomes of nonmyeloa- International Bone Marrow Transplant Registry and the European Group blative HLA-haploidentical blood or marrow transplantation with high- for Blood and Marrow Transplantation. Blood. 2006;108(13):4288-4290. dose post-transplantation cyclophosphamide in older adults. J Clin Oncol. 35. Anasetti C, Logan BR, Lee SJ, et al. Peripheral-blood stem cells versus bone 2015;33(28):3152-3161. marrow from unrelated donors. N Engl J Med. 2012;367(16):1487-1496. 13. McCurdy SR, Kanakry JA, Showel MM, et al. Risk-stratified outcomes of 36. Jagasia M, Arora M, Flowers ME, et al. Risk factors for acute GVHD and sur- nonmyeloablative HLA-haploidentical BMT with high-dose posttransplan- vival after hematopoietic cell transplantation. Blood. 2012;119(1):296-307. tation cyclophosphamide. Blood. 2015;125(19):3024-3031. 37. McCurdy SR, Kanakry CG, Tsai HL, et al. Grade II acute graft-versus-host 14. O'Donnell PV, Eapen M, Horowitz MM, et al. Comparable outcomes with disease and higher nucleated cell graft dose improve progression-free sur- marrow or peripheral blood as stem cell sources for hematopoietic cell vival after HLA-haploidentical transplant with post-transplant cyclophos- transplantation from haploidentical donors after non-ablative conditioning: phamide. Biol Blood Marrow Transplant. 2018;24(2):343-352. a matched-pair analysis. Bone Marrow Transplant. 2016;51(12):1599-1601. 38. Canaani J, Savani BN, Labopin M, et al. Donor age determines outcome in 15. Raiola AM, Dominietto A, di Grazia C, et al. Unmanipulated haploidentical acute leukemia patients over 40 undergoing haploidentical hematopoietic transplants compared with other alternative donors and matched sibling cell transplantation. Am J Hematol. 2018;93(2):246-253. grafts. Biol Blood Marrow Transplant. 2014;20(10):1573-1579. 39. Solomon SR, Sizemore CA, Sanacore M, et al. Haploidentical transplanta- 16. Blaise D, Furst S, Crocchiolo R, et al. Haploidentical T cell-replete trans- tion using T cell replete peripheral blood stem cells and myeloablative plantation with post-transplantation cyclophosphamide for patients in or conditioning in patients with high-risk hematologic malignancies who above the sixth decade of age compared with allogeneic hematopoietic lack conventional donors is well tolerated and produces excellent relapse- stem cell transplantation from an human leukocyte antigen-matched free survival: results of a prospective phase II trial. Biol Blood Marrow related or unrelated donor. Biol Blood Marrow Transplant. 2016;22 Transplant. 2012;18(12):1859-1866. (1):119-124. 40. Karam E, Laporte J, Solomon SR, et al. Who is a better donor for recip- 17. Gaballa S, Ge I, El Fakih R, et al. Results of a 2-arm, phase 2 clinical trial ients of allogeneic hematopoietic cell transplantation: a young HLA- using post-transplantation cyclophosphamide for the prevention of graft- mismatched haploidentical relative or an older fully HLA-matched versus-host disease in haploidentical donor and mismatched unrelated sibling or unrelated donor?Biol Blood Marrow Transplant. 2019;25 donor hematopoietic stem cell transplantation. Cancer. 2016;122 (10):2054-2060. (21):3316-3326. 41. Shah AJ, Kapoor N, Crooks GM, et al. The effects of Campath 1H upon graft- 18. Ghosh N, Karmali R, Rocha V, et al. Reduced-intensity transplantation for versus-host disease, infection, relapse, and immune reconstitution in lymphomas using haploidentical related donors versus HLA-matched sib- recipients of pediatric unrelated transplants. Biol Blood Marrow Trans- ling donors: a Center for International Blood and Marrow Transplant plant. 2007;13(5):584-593. Research Analysis. J Clin Oncol. 2016;34(26):3141-3149. 42. Kroger N, Solano C, Wolschke C, et al. Antilymphocyte globulin for preven- 19. Rashidi A, DiPersio JF, Westervelt P, et al. Comparison of outcomes after tion of chronic graft-versus-host disease. N Engl J Med. 2016;374(1): peripheral blood haploidentical versus matched unrelated donor alloge- 43-53. neic hematopoietic cell transplantation in patients with acute myeloid