Biol Blood Marrow Transplant 25 (2019) 1875 1883 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Impact of T Cell Dose on Outcome of T Cell-Replete HLA-Matched Allogeneic Peripheral Blood Stem Cell Transplantation Ayman Saad1, Lawrence Lamb2, Tao Wang3,4, Michael T. Hemmer4, Stephen Spellman5, Daniel Couriel6, Amin Alousi7, Joseph Pidala8, Hisham Abdel-Azim9, Vaibhav Agrawal10, Mahmoud Aljurf11, Amer M. Beitinjaneh12, Vijaya Raj Bhatt13, David Buchbinder14, Michael Byrne15, Jean-Yves Cahn16, Mitchell Cairo17, Paul Castillo18, Saurabh Chhabra19, Miguel Angel Diaz20, Shatha Farhan21, Yngvar Floisand22, Hadar A. Frangoul23, Shahinaz M. Gadalla24, James Gajewski25, Robert Peter Gale26, Manish Gandhi27, Usama Gergis28, Betty Ky Hamilton29, Peiman Hematti30, Gerhard C. Hildebrandt31, Rammurti T. Kamble32, Abraham S. Kanate33, Pooja Khandelwal34, Aleksandr Lazaryan8, Margaret MacMillan35, David I. Marks36, Rodrigo Martino37, Parinda A. Mehta34, Taiga Nishihori8, Richard F. Olsson38,39, Sagar S. Patel40, Muna Qayed41, Hemalatha G. Rangarajan42, Ran Reshef43, Olle Ringden44, Bipin N. Savani15, Harry C. Schouten45, Kirk R. Schultz46, Sachiko Seo47, Brian C. Shaffer48, Melhem Solh49, Takanori Teshima50, Alvaro Urbano-Ispizua51, Leo F. Verdonck52, Ravi Vij53, Edmund K. Waller54, Basem William1, Baldeep Wirk55, Jean A. Yared56, Lolie C. Yu57, Mukta Arora58,*, Shahrukh Hashmi11,59 1 Division of Hematology, The Ohio State University, Columbus, Ohio 2 University of Alabama at Birmingham, Birmingham, Alabama 3 Division of Biostatistics, Institute for Health and Equity, Medical College of Wisconsin, Milwaukee, Wisconsin 4 Center for International Blood and Marrow Transplant Research, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin 5 Center for International Blood and Marrow Transplant Research, National Marrow Donor Program/Be the Match, Minneapolis, Minnesota 6 Utah Blood and Marrow Transplant Program, Salt Lake City, Utah 7 Department of Stem Cell Transplantation, Division of Cancer Medicine, The University of Texas M.D. Anderson Cancer Center, Houston, Texas 8 Department of Blood and Marrow Transplantation, H. Lee Moffitt Cancer Center, Tampa, Florida 9 Division of Hematology, Oncology and Blood and Marrow Transplantation, Children’s Hospital of Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, California 10 Division of Hematology-Oncology, Indiana University School of Medicine, Indianapolis, Indiana 11 Oncology Center, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia 12 University of Miami, Miami, Florida 13 The Fred and Pamela Buffett Cancer Center, University of Nebraska Medical Center, Omaha, Nebraska 14 Division of Pediatric Hematology, Children’s Hospital of Orange County, Orange, California 15 Division of Hematology/Oncology, Vanderbilt University Medical Center, Nashville, Tennessee 16 Department of Hematology, CHU Grenoble Alpes, Grenoble, France 17 Division of Pediatric Hematology, Oncology, and Stem Cell Transplantation, Department of Pediatrics, New York Medical College, New York, New York 18 UF Health Shands Children’s Hospital, Gainesville, Florida 19 Division of Hematology/Oncology, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin 20 Department of Hematology/Oncology, Hospital Infantil Universitario Nino Jesus, Madrid, Spain 21 Henry Ford Hospital Bone Marrow Transplant Program, Detroit, Michigan 22 The National Hospital, Oslo, Denmark 23 Children’s Hospital at TriStar Centennial and Sarah Cannon Research Institute, Nashville, Tennessee 24 Division of Cancer Epidemiology & Genetics, Clinical Genetics Branch, National Cancer Institute, Rockville, Maryland 25 Consultant at Lu Daopei Hospital, Beijing, China 26 Hematology Research Center, Division of Experimental Medicine, Department of Medicine, Imperial College London, London, United Kingdom 27 Division of Transfusion Medicine, Mayo Clinic, Rochester, Minnesota 28 Hematologic Malignancies & Bone Marrow Transplant, Department of Medical Oncology, New York Presbyterian Hospital/Weill Cornell Medical Center, New York, New York 29 Blood & Marrow Transplant Program, Cleveland Clinic Taussig Cancer Institute, Cleveland, Ohio 30 Division of Hematology/Oncology/Bone Marrow Transplantation, Department of Medicine, University of Wisconsin Hospital and Clinics, Madison, Wisconsin Financial disclosure: See Acknowledgments on page 1882. * Correspondence and reprint requests: Mukta Arora, MD, University of Minnesota Medical Center, Division of Hematology, Oncology and Transplantation, Minne- apolis, 420 Delaware St, SE Minneapolis, MN 55455. E-mail address: arora005@umn.edu (M. Arora). https://doi.org/10.1016/j.bbmt.2019.05.007 1083-8791/© 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. 1876 A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 31 Markey Cancer Center, University of Kentucky, Lexington, Kentucky 32 Division of Hematology and Oncology, Center for Cell and Gene Therapy, Baylor College of Medicine, Houston, Texas 33 Osborn Hematopoietic Malignancy and Transplantation Program, West Virginia University, Morgantown, West Virginia 34 Division of Bone Marrow Transplant and Immune Deficiency, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio 35 University of Minnesota Blood and Marrow Transplant Program, Pediatrics, Minneapolis, Minnesota 36 Adult Bone Marrow Transplant, University Hospitals Bristol NHS Trust, Bristol, United Kingdom 37 Division of Clinical Hematology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain 38 Department of Laboratory Medicine, Karolinska Institute, Stockholm, Sweden 39 Centre for Clinical Research Sormland, Uppsala University, Uppsala, Sweden 40 Blood and Marrow Transplant Program, Cleveland Clinic Foundation, Cleveland, Ohio 41 Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia 42 Department of Pediatric Hematology, Oncology, Blood and Marrow Transplantation, Nationwide Children’s Hospital, Columbus, Ohio 43 Blood and Marrow Transplantation Program and Columbia Center for Translational Immunobiology, Columbia University Medical Center, New York, New York 44 Translational Cell Therapy Research, Karolinska Institute, Stockholm, Sweden 45 Department of Hematology, Academische Ziekenhuis, Maastricht, Netherlands 46 Department of Pediatric Hematology, Oncology and Bone Marrow Transplant, British Columbia’s Children’s Hospital, The University of British Columbia, Vancouver, Britich Columbia, Canada 47 Department of Hematology and Oncology, Dokkyo Medical University, Tochigi, Japan 48 Memorial Sloan Kettering Cancer Center, New York, New York 49 The Blood and Marrow Transplant Group of Georgia, Northside Hospital, Atlanta, Georgia 50 Hokkaido University Hospital, Sapporo, Japan 51 Department of Hematology, Hospital Clinic, University of Barcelona, IDIBAPS, and Josep Carreras Institute of Research, Barcelona, Spain 52 Department of Hematology/Oncology, Isala Clinic, Zwolle, The Netherlands 53 Division of Hematology and Oncology, Washington University School of Medicine, St Louis, Missouri 54 Department of Hematology and Medical Oncology, Winship Cancer Institute, Emory University, Atlanta, Georgia 55 Division of Bone Marrow Transplant, Seattle Cancer Care Alliance, Seattle, Washington 56 Blood & Marrow Transplantation Program, Division of Hematology/Oncology, Department of Medicine, Greenebaum Comprehensive Cancer Center, University of Maryland, Baltimore, Maryland 57 Division of Hematology/Oncology and HSCT, Center for Cancer and Blood Disorders, Children’s Hospital/Louisiana State University Medical Center, New Orleans, Louisiana 58 Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minnesota Medical Center, Minneapolis, Minnesota 59 Department of Internal Medicine, Mayo Clinic, Rochester, Minnesota Article history: A B S T R A C T Received 11 March 2019 Data on whether the T cell dose of allogeneic peripheral blood stem cell (PBSC) products influences transplanta- Accepted 3 May 2019 tion outcomes are conflicting. Using the Center for International Blood and Marrow Transplant Research database, we identified 2736 adult patients who underwent first allogeneic PBSC transplantation for acute leukemia or mye- Key Words: lodysplastic syndrome between 2008 and 2014 using an HLA-matched sibling donor (MSD) or an 8/8-matched T cell unrelated donor (MUD). We excluded ex vivo and in vivo T cell-depleted transplantations. Correlative analysis GVHD dose was performed between CD3+ T cell dose and the risk of graft-versus-host-disease (GVHD), relapse, nonrelapse Allogeneic mortality (NRM), disease-free survival (DFS), and overall survival (OS). Using maximum likelihood estimation, we Transplantation identified CD3+ T cell dose cutoff that separated the risk of acute GVHD (aGVHD) grade II-IV in both the MSD and MUD groups. A CD3+ T cell dose cutoff of 14 £ 107 cells/kg identified MSD/low CD3+ (n = 223) and MSD/high CD3+ (n = 1214), and a dose of 15 £ 107 cells/kg identified MUD/low CD3+ (n = 197) and MUD/high CD3+ (n = 1102). On univariate analysis, the MSD/high CD3+ group had a higher cumulative incidence of day +100 aGVHD grade II-IV compared with the MSD/low CD3+ group (33% versus 25%; P = .009). There were no differences between the 2 groups in engraftment rate, risk of aGVHD grade III-IV or chronic GVHD (cGVHD), NRM, relapse, DFS, or OS. The MUD/high CD3+ group had a higher cumulative incidence of day +100 aGVHD grade II-IV compared with the MUD/low CD3+ group (49% versus 41%; P = .04). There were no differences between the 2 groups in engraftment rate, risk of severe aGVHD or cGVHD, NRM, relapse, DFS, or OS. Multivariate analysis of the MSD and MUD groups failed to show an association between CD3+ T cell dose and the risk of either aGVHD grade II-IV (P = .10 and .07, respectively) or cGVHD (P = .80 and .30, respectively). Subanalysis of CD4+ T cells, CD8+ T cells, and CD4+/CD8+ ratio failed to identify cutoff values predictive of transplantation outcomes; however, using the log-rank test, the sample size was suboptimal for identifying a difference at this cutoff cell dose. In this registry study, the CD3+ T cell dose of PBSC products did not influence the risk of aGVHD or cGVHD or other transplantation outcomes when using an MSD or an 8/8-matched MUD. Subset analyses of CD4+ and CD8+ T cell doses were not possible given our small sample size. © 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. INTRODUCTION (GVHD) [3]. Among these, the most well-studied cells are Allogeneic hematopoietic cell transplantation (HCT) per- CD3+T cells. formed for hematologic malignancies relies on both the con- Although CD3+T cells can exert a strong GVT effect [4], the ditioning regimen and immunotherapy exploiting the graft- risk of acute GVHD (aGVHD) also rises with increasing dose, as versus-tumor (GVT) effect, which is derived primarily from demonstrated in both observational and prospective studies donor immune effector cells [1,2]. A complex interplay [5,6]. T cell-depleted (TCD) allogeneic HCT has led to a between the immune effector cells, including antigen-pre- decreased risk of GVHD but at the expense of an increased risk senting cells, CD3+ cells, CD4+ T cells, CD8+ T cells, regulatory of relapse, as demonstrated by trials of both ex vivo [7] and in T cells (Tregs), and natural killer (NK) cells, is responsible for vivo depletion [8]. The higher risk of GVHD in peripheral blood both the GVT effect and the graft-versus-host disease stem cell (PBSC) grafts compared with bone marrow (BM) A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 1877 grafts is apparent from both observational studies [9] and clini- of primary disease progression/relapse, with the latter event considered a cal trials [10], as PBSCs are known to carry 10 to 15 times the competing risk. aGVHD and cGVHD were graded using standard criteria [21,22] Neutrophil recovery was defined as the first of 3 successive days with quantity of CD3+T cells as BM [11]. Thus, numerous attempts absolute neutrophil count 500/mL after a post-transplantation nadir. Plate- have been made to separate out the GVT effect from GVHD, let recovery was defined as the first of 3 successive days with a platelet count including the use of CD34+T cell selection [12], naïve T cell 20,000/mL without transfusion support for at least 7 days. Data were cen- depletion [13], post-transplantation cyclophosphamide [14], sored for mortality events before neutrophil recovery. microtransplantation [15], and NK cell graft engineering. Few single-center studies have evaluated the role of CD3+T cell Statistical Analysis The primary objective of the study was to correlate the graft T cell dose dose with respect to both relapse and GVHD outcomes post- with the incidence and severity of aGVHD and cGVHD, OS, DFS, relapse, and HCT; however, these studies varied significantly in terms of NRM following PBSC HCT with MSD and 8/8-matched MRD HCT. In a subset selection criteria, with no consensus on an optimal CD3+T cell analysis of subjects with available CD4+ and CD8+T cell dose data, we also dose cutoff value [16-19]. In a recent large registry study, in tested for an association between graft T cell subset dose and ratio of CD4+/ CD8+T cell doses and these transplantation outcomes in univariate analysis HCTs using unrelated donors, higher CD3+ and CD34+T cell due only to smaller sample size. T cell dose cutoff values were determined doses were significantly associated with an increased risk of using a maximum likelihood method based on a Cox proportional hazards grade III-IV aGVHD (hazard ratio [HR], 3.6; 95% confidence model for the aGVHD grade II-IV endpoint. interval [CI], 1.45 to 9.96; P = .006 and 2.65 (95% CI, 1.07 to Categorical data are summarized as frequency, and continuous data are summarized as median and range. Probabilities of DFS and OS were calcu- 6.57); P = .04, respectively) [20]. Because the aforementioned lated as described previously [23]. Cumulative incidences of aGVHD grade II- studies used different types of donors, different diseases, and IV, aGVHD grade III-IV, cGVHD, NRM, relapse/progression, platelet recovery, different conditioning regimens, the optimum cutoff CD3+T and hematopoietic cell recovery were calculated to accommodate for com- cell dose that can potentially avoid GVHD while still promoting peting risks [24]. Associations among patient-, disease-, and transplantation- the GVT effect are unknown. related variables and outcomes of interest were evaluated using Cox propor- tional hazards regression. All the clinical variables were tested for the affir- We hypothesized that there exists a T cell dose range that mation of the proportional hazards assumption. Factors violating the promotes GVT, whereas levels above this range carry a higher proportional hazards assumption were adjusted through stratification, and risk of both severe aGVHD and cGVHD with subsequent then a stepwise forward model selection procedure was used to select increased nonrelapse mortality (NRM). adjusted clinical variables for each outcome, with a threshold of .05 to be entered into and be retained in the model. Interactions between T cell dose and the adjusted clinical variables were examined, and no significant interac- METHODS Data Sources tions were detected. Center effect was adjusted as a random factor for all out- The Center for International Blood and Marrow Transplant Research comes [25]. The significance level of .01 was used for the overall effects of (CIBMTR) is a working group of more than 420 transplantation centers world- factors followed by Bonferroni adjustment for pairwise comparisons to wide that contribute detailed data on HCT to a statistical center at the Medical account for multiple testing. All statistical analyses were performed using College of Wisconsin. Participating centers are required to report all trans- SAS version 9.4 (SAS Institute, Cary, NC) plantations consecutively; patients are followed longitudinally, and compli- ance is monitored by onsite audits. Computerized checks for discrepancies, RESULTS physician reviews of submitted data, and onsite audits of participating cen- ters ensure data quality. Observational studies conducted by the CIBMTR are Baseline Characteristics performed in compliance with all applicable federal regulations pertaining to We identified 2736 adult patients who met the selection the protection of human research participants. Protected health information criteria described above. Regimen intensity, classified as mye- used in the performance of such research is collected and maintained in the loablative (MAC), reduced-intensity (RIC), or nonmyeloablative CIBMTR’s capacity as a Public Health Authority under the HIPAA Privacy Rule. The Institutional Review Boards of the Medical College of Wisconsin and the (NMA), was defined as described previously [26]. Using a Cox National Marrow Donor Program approved this study. proportional hazards model, we determined the cutoff value The CIBMTR collects data at 2 levels: Transplant Essential Data (TED) and for CD3+T cell dose and separated each group (MSD and MUD) Comprehensive Report Form (CRF) data. TED-level data include disease type, into low risk and high risk of grade II-IV aGVHD: 14 £ 107 age, gender, pre-HCT disease stage and chemotherapy responsiveness, date of diagnosis, graft type (BM and/or PBSCs), conditioning regimen, post-trans- cells/kg for MSD and 15 £ 107 cells/kg for MUD. Then the plantation disease progression and survival, development of new malignancy, patients were divided into 4 groups based on the donor type and cause of death. All CIBMTR centers contribute TED data. More detailed (MSD or MUD) and T cell dose cutoff value: MSD/low CD3+ disease and pretransplantation and post-transplantation clinical information (n = 223), MSD/high CD3+ (n = 1214), MUD/low CD3+ (n = 197), are collected on a subset of registered patients selected for CRF data via a weighted randomization scheme. TED- and CRF-level data are collected pre- and MUD/high CD3+ (n = 1102). The median CD3+T cell dose transplantation, at 100 days and 6 months post-HCT. and annually thereafter was 11 £ 107 in the MSD/low group, 29 £ 107in the MSD/high or until death. Data for the present analysis were retrieved from CIBMTR group, 10 £ 107in the MUD/low group, and 28£ 107in the (TED and CRF) report forms. MUD/high group. The MSD and MUD groups were analyzed separately. Baseline patient-, disease- and transplantation- Patients related characteristics are summarized in Tables 1 and 2. We analyzed data of adult patients (age 18 years) who underwent first allogeneic HCT for acute myelogenous leukemia (AML), acute lymphoblastic leukemia (ALL), or myelodysplastic syndrome (MDS) between 2008 and 2014 MSD Groups with PBSC grafts using an HLA-identical sibling donor (MSD) or an 8/8- matched unrelated donor (MUD) matched at the allele level at HLA-A, -B, -C, Univariate analysis showed a day +100 cumulative inci- and -DRB1. We limited the disease types to AML, ALL, and MDS, hypothesiz- dence of aGVHD grade II-IV of 25% (95% CI [confidence inter- ing that these patients have a comparable risk of relapse and susceptibility to val], 19% to 31%) in the MSD/low CD3+group and 33% (95% CI, the GVT effect. We excluded ex vivo (TCD and CD34+-selected grafts) and in 30% to 36%) in the MSD/high CD3+group (P = .009). However, vivo TCD (antithymocyte globulin or alemtuzumab) HCT. All patients had available data on CD3+T cell dose; however, some patients were missing data there was no significant between-group difference in the risk on CD4+and/or CD8+ T cell dose. of aGVHD grade III-IV (P= .40) or in the risk of cGVHD at 2 years, NRM, relapse, DFS, and OS. There was also no difference in the Study Endpoints day +100 engraftment rate between the 2 groups. For overall survival (OS), death from any cause was considered an event, In multivariate analysis, CD3+T cell dose did not influence and surviving patients were censored at last contact. For disease-free survival (DFS), either progression/relapse or death from any cause was considered an aGVHD (grades II-IV and III-IV) (Table 3), cGVHD, relapse, event, and patients alive without evidence of disease relapse/progression NRM, DFS, or OS (Supplemental Table 1). However, the risk of were censored at last follow-up. NRM was defined as death without evidence aGVHD grade II-IV was higher with any donor-recipient sex 1878 A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 Table 1 Table 1 (Continued) Characteristics of Adult Patients Undergoing First Allogeneic HCT for AML, ALL, and MDS Between 2008 and 2014 with PBSCs from an HLA-Identical Sibling Characteristic CD3+ T Cell Dose Donor with Valid CD3+ Cell Dose Data, as Reported to the CIBMTR <14 £ 107 14 £ 107 ALL 33 224 Characteristic CD3+ T Cell Dose Intermediate 17 (52) 167 (75) <14 £ 107 14 £ 107 High/very high 16 (48) 57 (25) Number of patients 223 1214 MDS 53 350 Number of centers 58 95 Intermediate 29 (55) 194 (55) Recipient age, yr High/very high 12 (23) 81 (23) Median (range) 51 (18-71) 54 (18-78) Missing 12 (23) 75 (21) 18-29, n (%) 20 (9) 110 (9) Time from diagnosis to HCT, mo 30-39, n (%) 28 (13) 130 (11) Median (range) 6 (1-156) 5 (<1-279) 40-49, n (%) 55 (25) 232 (19) <6, n (%) 121 (54) 695 (57) 50-59, n (%) 74 (33) 419 (35) 6-<12, n (%) 52 (23) 257 (21) 60+, n (%) 46 (21) 323 (27) 12, n (%) 50 (22) 262 (22) Recipient sex, n (%) CD3+ cell dose, £ 107/kg, median 11 (3-14) 29 (14-113) Male 123 (55) 694 (57) (range) Female 100 (45) 520 (43) CD4+ cell dose, £ 107/kg, quartiles Recipient race, n (%) Median (range) 8 (3-169) 19 (<1-180) Caucasian 176 (79) 1057 (87) <10.6, n (%) 73 (33) 30 (2) Non-Caucasian 37 (17) 117 (10) 10.6-16.79, n (%) 4 (2) 99 (8) Missing 10 (4) 40 (3) 16.8-28.79, n (%) 0 103 (8) Body mass index 28.8, n (%) 12 (5) 90 (7) Median (range) 29 (18-62) 27 (15-56) Missing, n (%) 134 (60) 892 (73) Underweight (<18.5), n (%) 2 (<1) 25 (2) CD8+ cell dose, x 107/kg, quartiles Normal (18.5-<25), n (%) 52 (23) 366 (30) Median (range) 4 (<1-59) 8 (<1-253) Overweight (25-<30), n (%) 67 (30) 433 (36) <4.52 61 (27) 43 (4) Obese (30), n (%) 101 (45) 390 (32) 4.52-7.179 16 (7) 87 (7) Missing, n (%) 1 (<1) 0 7.18-12.769 2 (<1) 102 (8) KPS score, n (%) 12.77 11 (5) 92 (8) <90 92 (41) 478 (39) Missing 133 (60) 890 (73) 90-100 125 (56) 718 (59) CD34+ cell dose, £ 106/kg Missing 6 (3) 18 (1) Median (range) 5 (<1-22) 6 (<1-28) Sorror comorbidity index, n (%) <2, n (%) 36 (16) 53 (4) 0-1 101 (45) 560 (46) 2-<4, n (%) 54 (24) 211 (17) 2-3 68 (30) 382 (31) 4-<8, n (%) 101 (45) 624 (51) 4+ 51 (23) 261 (21) 8, n (%) 26 (12) 314 (26) Missing 3 (1) 11 (<1) Missing, n (%) 6 (3) 12 (<1) Disease, n (%) CD4+/CD8+ cell dose ratio, quartiles AML 137 (61) 640 (53) Median (range) 2 (<1-9) 2 (<1-13) ALL 33 (15) 224 (18) <1.53, n (%) 26 (12) 78 (6) MDS 53 (24) 350 (29) 1.53-2.189, n (%) 20 (9) 82 (7) Disease status, n (%) 2.19-3.149, n (%) 23 (10) 79 (7) AML 137 640 3.15, n (%) 19 (9) 84 (7) Early 87 (64) 386 (60) Missing, n (%) 135 (61) 891 (73) Intermediate 21 (15) 107 (17) Donor/recipient sex match, n (%) Advanced 29 (21) 147 (23) Female/female 48 (22) 255 (21) ALL 33 224 Female/male 55 (25) 312 (26) Early 17 (52) 167 (75) Male/female 52 (23) 265 (22) Intermediate 7 (21) 37 (17) Male/male 68 (30) 382 (31) Advanced 9 (27) 20 (9) Donor/recipient CMV serostatus MDS 53 350 match, n (%) Early 35 (66) 228 (65) -/- 56 (25) 276 (23) Intermediate 17 (32) 104 (30) -/+ 52 (23) 304 (25) Advanced 1 (2) 18 (5) +/- 26 (12) 141 (12) Revised DRI, n (%) +/+ 84 (38) 477 (39) AML 137 640 Missing 5 (2) 16 (1) Low 8 (6) 41 (6) Donor/recipient ABO match, n (%) Intermediate 90 (66) 355 (55) Matched 105 (47) 565 (47) High/very high 25 (18) 135 (21) Minor mismatch 26 (12) 136 (11) Missing 14 (10) 109 (17) (continued) (continued) A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 1879 Table 1 (Continued) Table 2 Characteristics of Adult Patients Undergoing First Allogeneic HCT for AML, ALL, Characteristic CD3+ T Cell Dose and MDS Between 2008 and 2014 with PBSCs from an 8/8-matched MUD with <14 £ 107 14 £ 107 Valid CD3+ Cell Dose Data, as Reported to the CIBMTR Major mismatch 22 (10) 149 (12) Characteristic CD3+ T Cell Dose Bidirectional mismatch 7 (3) 36 (3) <15 £ 107 15 £ 107 Missing 63 (28) 328 (27) Number of patients 197 1102 Conditioning regimen intensity, n (%) Number of centers 55 80 MAC 167 (75) 840 (69) Age, yr RIC/NMA 56 (25) 374 (31) Median (range) 55 (19-76) 56 (18-78) Conditioning regimen, MAC, n (%) 18-29, n (%) 18 (9) 111 (10) Bu + Cy § others 52 (31) 242 (29) 30-39, n (%) 25 (13) 120 (11) TBI + Cy 48 (29) 275 (33) 40-49, n (%) 32 (16) 182 (17) Bu + Flu 40 (24) 206 (25) 50-59, n (%) 61 (31) 264 (24) TBI + ETOP 10 (6) 71 (8) 60+, n (%) 61 (31) 425 (39) Others 17 (10) 46 (5) Recipient sex, n (%) Conditioning regimen, RIC/NMA, n Male 123 (62) 629 (57) (%) Female 74 (38) 473 (43) Bu + Flu 20 (36) 156 (42) Recipient race, n (%) Flu + Mel 23 (41) 104 (28) Caucasian 190 (96) 1022 (93) TBI + Flu 2 (4) 64 (17) Non-Caucasian 7 (4) 61 (6) Flu + others 10 (18) 41 (11) Missing 0 19 (2) Others 1 (2) 9 (2) Body mass index TBI used in conditioning regimen, n (%) Median (range) 29 (19-52) 28 (8-62) Yes 81 (36) 461 (38) Underweight (<18.5), n (%) 0 21 (2) No 142 (64) 753 (62) Normal (18.5-<25), n (%) 45 (23) 309 (28) GVHD prophylaxis, n (%) Overweight (25-<30), n (%) 68 (35) 418 (38) CsA + MTX § others 9 (4) 123 (10) Obese (30), n (%) 84 (43) 354 (32) Tac + MTX § others 161 (72) 716 (59) KPS score, n (%) CsA + MMF § others 13 (6) 92 (8) <90 81 (41) 428 (39) Tac + MMF § others 18 (8) 145 (12) 90-100 113 (57) 662 (60) Others 22 (10) 138 (11) Missing 3 (2) 12 (1) Year of transplantation, n (%) Sorror comorbidity index, n (%) 2008-2010 110 (49) 555 (46) 0-1 64 (32) 473 (43) 2011-2014 113 (51) 659 (54) 2-3 61 (31) 357 (32) Follow-up of survivors, mo, median 47 (3-101) 49 (3-107) 4+ 70 (36) 264 (24) (range) Missing 2 (1) 8 (<1) CMV, cytomegalovirus, Bu, busulfan; Cy, cyclophosphamide; TBI, total body Disease, n (%) irradiation; Flu, fludarabine; ETOP, etoposide; CsA, cyclophosphamide; MTX, AML 116 (59) 619 (56) methotrexate; MMF, mycophenolate mofetil; Tac, tacrolimus. ALL 22 (11) 142 (13) MDS 59 (30) 341 (31) Disease status, n (%) mismatch (P= .02 and .009 for female to male and male to AML 116 619 female, respectively). The risk of severe aGVHD grade III-IV Early 70 (60) 351 (57) was worse in patients with a Karnofsky Performance Status Intermediate 20 (17) 124 (20) (KPS) score <90 relative to those with KPS score of 90 to 100 Advanced 26 (22) 141 (23) (P = .005). The risk of cGVHD was increased in patients age Missing 0 3 (<1) >29 years (overall P = .006), in transplants from a female donor (P< .002), and in transplantations performed before 2011 ALL 22 142 (P = .01). DFS was worse in patients age 60 years (P = .01), Early 12 (55) 91 (64) patients with a high/very high Disease Risk Index (DRI) (P < Intermediate 5 (23) 31 (22) .0001), and patients with a lower KPS score (P < .0001). OS Advanced 5 (23) 20 (14) was worse in patients with a high/very high DRI (P = .0001), MDS 59 341 those with a lower KPS score (P < .0001), and those with a Early 41 (69) 233 (68) Hematopoietic Stem Cell Transplantation Comorbidity Index Advanced 16 (27) 92 (27) (HCT-CI) >3 (P = .003). NRM was worse in patients with MDS Missing 2 (3) 16 (5) (P = .002), a lower KPS (P = .007), and HCT-CI >3 (P = .0006). Revised DRI, n (%) The relapse risk was higher in patients with advanced disease AML 116 619 before HCT (P= .0007) and lower KPS score (P= .002). Low 9 (8) 39 (6) Data from subset analyses of CD4+ T cells, CD8+ T cells, and Intermediate 64 (55) 352 (57) CD4+/CD8+ ratio were available for only a limited number of High/very high 26 (22) 129 (21) patients. No significant associations of these variables were Missing 17 (15) 99 (16) detected for aGVHD, cGVHD, NRM, relapse, DFS, or OS. (continued) 1880 A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 Table 2 (Continued) Table 2 (Continued) Characteristic CD3+ T Cell Dose Characteristic CD3+ T Cell Dose <15 £ 107 15 £ 107 <15 £ 107 15 £ 107 ALL 22 142 -/+ 71 (36) 409 (37) Intermediate 12 (55) 91 (64) +/- 15 (8) 116 (11) High/very high 10 (45) 51 (36) +/+ 42 (21) 265 (24) MDS 59 341 Missing 4 (2) 12 (1) Intermediate 33 (56) 210 (62) Donor/recipient ABO match, n (%) High/very high 11 (19) 73 (21) Matched 60 (30) 397 (36) Missing 15 (25) 58 (17) Minor mismatch 42 (21) 216 (20) Time from diagnosis to HCT, mo Major mismatch 34 (17) 168 (15) Median (range) 6 (2-156) 6 (<1-297) Bidirectional mismatch 5 (3) 66 (6) <6, n (%) 94 (48) 505 (46) Missing 56 (28) 255 (23) 6-<12, n (%) 55 (28) 292 (26) Conditioning regimen intensity, n (%) 12, n (%) 47 (24) 305 (28) MAC 134 (68) 668 (61) Missing, n (%) 1 (<1) 0 RIC/NMA 63 (32) 434 (39) CD3+ cell dose, £ 107/kg, median 10 (3-14) 28 (14-113) Conditioning regimen, MAC, n (%) (range) Bu + Cy § others 50 (37) 211 (32) CD4+ cell dose, £ 107/kg, quartiles TBI + Cy 33 (25) 157 (24) Median (range) 6 (2-57) 18 (<1-190) Bu + Flu 29 (22) 203 (30) <9.6, n (%) 63 (32) 20 (2) TBI + ETOP 6 (4) 29 (4) 9.6-14.89, n (%) 3 (2) 80 (7) Others 16 (12) 68 (10) 14.9-23.39, n (%) 0 81 (7) Conditioning regimen, RIC/NMA, n (%) 23.4, n (%) 6 (3) 77 (7) Bu + Flu 27 (43) 117 (27) Missing, n (%) 125 (63) 844 (77) Flu + Mel 20 (32) 145 (33) CD8+ cell dose, £ 107/kg TBI + Flu 8 (13) 100 (23) Median (range) 4 (<1-30) 10 (<1-145) Flu + others 6 (10) 43 (10) <5.19, n (%) 58 (29) 24 (2) Others 2 (3) 29 (7) 5.19-8.519, n (%) 8 (4) 76 (7) TBI used in conditioning regimen, n (%) 8.52-14.439, n (%) 1 (<1) 81 (7) Yes 57 (29) 391 (35) 14.44, n (%) 5 (3) 78 (7) No 140 (71) 711 (65) Missing, n (%) 125 (63) 843 (76) GVHD prophylaxis, n (%) CD34+ cell dose, £ 106/kg CsA + MTX § others 6 (3) 41 (4) Median (range) 5 (<1-24) 7 (1-30) Tac + MTX § others 130 (66) 611 (55) <2, n (%) 10 (5) 11 (<1) CsA + MMF § others 11 (6) 97 (9) 2-<4, n (%) 40 (20) 98 (9) Tac + MMF § others 30 (15) 192 (17) 4-<8, n (%) 117 (59) 511 (46) Others 20 (10) 161 (15) 8, n (%) 27 (14) 455 (41) Year of transplantation, n (%) Missing, n (%) 3 (2) 27 (2) 2008-2010 69 (35) 482 (44) CD4+/CD8+ cell dose ratio 2011-2014 128 (65) 620 (56) Median (range) 2 (<1-6) 2 (<1-19) Follow-up of survivors, mo, median 37 (21-96) 48 (3-102) <1.31, n (%) 19 (10) 62 (6) (range) 1.31-1.649, n (%) 14 (7) 68 (6) 1.65-2.259, n (%) 19 (10) 65 (6) 2.26, n (%) 20 (10) 62 (6) Missing, n (%) 125 (63) 845 (77) Likewise, CD34+ cell dose was not significantly associated with MUD age, yr any of the transplantation outcomes. Median (range) 30 (18-60) 28 (18-61) 18-32, n (%) 116 (59) 692 (63) MUD Groups 33-49, n (%) 59 (30) 296 (27) Univariate analysis showed a cumulative incidence of 50+, n (%) 14 (7) 63 (6) aGVHD grade II-IV at day +100 of 41% (95% CI, 35% to 48%) in Missing, n (%) 8 (4) 51 (5) the MUD/low CD3+group and 49% (95% CI, 46% to 52%) in the Donor/recipient sex match, n (%) MUD/high CD3+group (P= .04). However, there was no Female/female 15 (8) 163 (15) between-group difference in the risk of aGVHD grade III-IV Female/male 19 (10) 174 (16) (P= .90). Likewise, the risks of cGVHD at 2 years, NRM, relapse, Male/female 59 (30) 310 (28) DFS, and OS were not statistically different between the Male/male 104 (53) 455 (41) 2 groups. There also was no difference in the day +100 engraft- Donor/recipient CMV serostatus ment rate. match, n (%) In multivariate analysis, CD3+T cell dose did not influence -/- 65 (33) 300 (27) the risk of aGVHD grade II-IV and III-IV (Table 4), cGVHD, (continued) relapse, NRM, DFS, or OS (Supplemental Table 2). However, the risk of aGVHD grade II-IV was higher in patients who received A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 1881 an MAC regimen (P = .02). The risk of severe aGVHD grade III-IV P Value P Value was elevated in underweight patients (P= .01), and with older .66 .55 donors (age >32 years) (P= .01). The risk of cGVHD was lower in patients with ALL (P= .003), and those who underwent HCT OS OS .96 (.78-1.17) .94 (.77-1.15) HR (95% CI) after 2010 (P = .0003). DFS was worse with older donors (age HR (95% CI) >50 years) (P= .0001) and those with a high/very high DRI (P < .0003). Worse OS was associated with older patients (age 50 1 years) (P = .008), older donors (age 50 years) (P = .0001), high/ 1 very high DRI (P= .0005), and lower KPS (P < .009). NRM was P Value P Value worse with older donors (age >50 years) (P = .0006). The risk .77 .96 of relapse was worse in patients with a high/very high DRI DFS (P = .0002). Subset analyses of CD4+ T cells, CD8+ T cells, and DFS CD4+/CD8+ ratio was available in only a limited number of .97 (.80-1.18) .99 (.82-1.20) HR (95% CI) HR (95% CI) patients. No significant associations were detected between these variables for aGVHD, cGVHD, NRM, relapse, DFS, or OS. Likewise, CD34+ cell dose was also not significantly associated 1 1 with any of the transplantation outcomes. P Value P Value DISCUSSION .73 .87 Our data show no associations between the CD3+T cell dose NRM of PBSC grafts and the risk of aGVHD, cGVHD, or relapse in our NRM .95 (.71-1.27) .97 (.70-1.36) HR (95% CI) study cohort. Nonetheless, the subgroup analyses suggest cer- HR (95% CI) tain associations that merit further exploration prospectively. Although the univariate analysis showed a correlation between 1 CD3+T cell dose and the risk of aGVHD in both the MSD and 1 MUD groups, the multivariable analysis failed to prove such an P Value P Value association. It is possible that the subgroups selected for multi- .96 .85 variate analysis were not large enough to power a detection in difference in the binary outcome (presence or absence of grade Relapse Relapse 1.01 (.78-1.29) II-IV aGVHD), leading to the possibility of type II error. It is also 1.02 (.81-1.29) HR (95% CI) possible that the variables chosen in the univariate analysis HR (95% CI) did not include some potential risk factors for aGVHD (eg, inadequate information on CD4+, CD8+, and CD56+T cells and 1 1 dendritic cells in PBSC grafts). The only group with an increased risk of aGVHD grade II-IV on multivariate analysis P Value P Value was patients who underwent MUD HCT using an MAC regi- .29 .81 men. This finding is consistent with a CIBMTR study showing cGVHD that RIC regimens were associated with decreased risk of cGVHD .89 (.73-1.10) aGVHD in MUD HCT recipients [27]. .97 (.79-1.21) HR (95% CI) HR (95% CI) Our data contrast with the European Society of Blood and Marrow Transplant (EBMT) study of MUD HCT showing an association between CD3+T cell dose >35 £ 107/kg and 1 1 Multivariate Analysis of the MUD Group Showing the Influence of CD3+ T Cell Dose Multivariate Analysis of the MSD Group Showing the Influence of CD3+ T Cell Dose increased risk of aGVHD [20]. This discrepancy may be attrib- uted to differences in median CD3+T cell doses in PBSC grafts P Value P Value in the 2 studies, as well as in the statistical methodology used .34 .25 aGVHD III-IV aGVHD III-IV to categorize the primary outcome variable. In the EBMT study, CD3+T cell dose was categorized by interquartile range, .85 (.61-1.19) .78 (.51-1.18) HR (95% CI) whereas in the present study, we used a cutoff values of CD3+ HR (95% CI) T cell dose based on the differential risk of aGVHD grade II-IV. Moreover, the EBMT study included TCD allogeneic HCT, 1 1 whereas the present study excluded it. In addition, some of the conditioning regimens used in the EBMT study were not evalu- P Value P Value ated in the present study. Of note, the BMT CTN 0201 trial has .07 .10 also failed to show an association of the T cell dose of the PBSC aGVHD II-IV aGVHD II-IV graft with survival or GVHD in patients with AML or MDS [28]. .81 (.65-1.02) A single-institution study using BM rather than PBSC grafts .79 (.60-1.04) HR (95% CI) HR (95% CI) demonstrated a paradoxical increase in the risk of cGVHD with lower CD3+T cell dose in a subset of patients who received an MAC regimen with busulfan and cyclophosphamide 1 1 (P= .006) [29]. Dose, £ 107/kg Dose, £ 107/kg Owing to our limited sample size, further analysis was not possible in order to detect outcome differences based on T cell CD3+ Cell CD3+ Cell phenotypic subsets: CD4+, CD8+, or CD4+/CD8+ ratio. However, Table 4 > 15  15 Table 3 >14 14 transplantation outcomes may depend on functional T cell sub- sets: naïve T cells, effector T cells, and/or central memory T 1882 A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 cells. In particular, depletion of naïve T cells (either CD4+ or T cell dose has also been shown to be predictive of graft failure CD8+) was associated with a lower risk of cGVHD and a greater with TCD allogeneic HCT [42]. risk of steroid-responsive aGVHD in small phase II study [13]. Tregs (CD4+/CD25+/FOXP3+), another small subset of CD4+ T ACKNOWLEDGMENTS cells, have been shown to ameliorate cGVHD [30]. Unbalanced Financial disclosure: The CIBMTR is supported primarily recovery of Tregs and effector T cells after transplantation also by Public Health Service Grant/Cooperative Agreement has been correlated with an increased risk of cGVHD [31]. 5U24CA076518 from the National Cancer Institute (NCI), the Even though PBSC grafts include coinfusions of both CD34+ National Heart, Lung and Blood Institute (NHLBI) and the and CD3+T cells in HCTs, the dose of CD3+ is not evaluated rou- National Institute of Allergy and Infectious Diseases (NIAID); a tinely in most transplantation centers, because it continues to Grant/Cooperative Agreement 1U24HL138660 from NHLBI and be controversial. Farhan et al [32] retrospectively evaluated NCI; a contract HHSH250201700006C with Health Resources CD3+T cell doses in both MUD and MSD HCTs and found no sig- and Services Administration (HRSA/DHHS); Grants N00014-17- nificant correlation with aGVHD; however, they observed that 1-2388, N00014-17-1-2850 and N00014-18-1-2045 from the OS was significantly affected by a higher CD3+cell dose (mean Office of Naval Research; and grants from Adaptive Biotechnolo- dose 12 £ 107/kg) in their cohort. This CD3+T cell dose differs gies; *Amgen, Inc.; Anonymous donation to the Medical College from the doses in our cohort and the EBMT cohort, perhaps of Wisconsin; Astellas Pharma US; Atara Biotherapeutics, Inc.; contributing to different outcomes. Be the Match Foundation; *bluebird bio, Inc.; *Bristol Myers Although our analysis did not show an impact of CD34+T Squibb Oncology; *Celgene Corporation; *Chimerix, Inc.; cell dose on transplant outcome, it is worth noting that >50% *CytoSen Therapeutics, Inc.; Fred Hutchinson Cancer Research of the patients in our cohort received a CD34+ cell dose of 4 to Center; Gamida Cell Ltd.; Gilead Sciences, Inc.; HistoGenetics, 8 £ 106 cells/kg, and a minority (5% to 10%) received a dose Inc.; Immucor; *Incyte Corporation; Janssen Scientific Affairs, <2 £ 106 cells/kg (Tables 1 and 2). In our opinion, this pre- LLC; *Jazz Pharmaceuticals, Inc.; Karius, Inc.; Karyopharm Thera- cludes an accurate conclusion as to the impact of CD34+ cell peutics, Inc.; *Kite Pharma, Inc.; Medac, GmbH; *Mediware; The dose on transplantation outcomes. Previous studies have eval- Medical College of Wisconsin; *Merck & Co, Inc.; *Mesoblast; uated this question with favorable outcomes with higher MesoScale Diagnostics, Inc.; Millennium, the Takeda Oncology CD34+T cell doses [33-35], although observing a higher risk of Co.; *Miltenyi Biotec, Inc.; Mundipharma EDO; National Marrow cGVHD with CD34+ cell doses >8 £ 106 cells/kg [35,36] or Donor Program; Novartis Pharmaceuticals Corporation; PCORI; >10 £ 106 cells/kg [37]. *Pfizer, Inc; *Pharmacyclics, LLC; PIRCHE AG; *Sanofi Genzyme; Donor age group was identified as a risk factor for the *Seattle Genetics; Shire; Spectrum Pharmaceuticals, Inc.; St. Bal- development of severe aGVHD and for worse DFS (donor age drick’s Foundation; Swedish Orphan Biovitrum, Inc.; *Takeda >50 years) in MUD HCT. The effect of donor age on the clinical Oncology; and University of Minnesota. The views expressed in outcomes is similar that seen in another study [16], which this article do not reflect the official policy or position of the found a correlation between donor age and the CD8+ T cell National Institute of Health, the Department of the Navy, the dose of the PBSC graft. Given the limited availability of CD8+T Department of Defense, Health Resources and Services Adminis- cell dose in the PBSC grafts in our cohort, we could not assess tration (HRSA) or any other agency of the U.S. Government. this association with age. This study was congruent with other *Corporate Members. large studies in terms of results pertaining to well-known risk Conflict of interest statement: There are no conflicts of inter- factors for GVHD, such as older recipient age [38] and a lower est to report. KPS score [39]. As expected, a higher DRI was predictive of a greater risk of relapse in both the MUD and MSD groups [40]. SUPPLEMENTARY DATA A strength of our study is the large sample size of both the Supplementary data related to this article can be found MUD and MSD groups, which allowed us to categorize the entire online at https://doi.org/10.1016/j.bbmt.2019.05.007. cohort into 4 groups based on donor type and CD3+T cell dose in the PBSC graft. Other strengths are the availability of comprehen- REFERENCES sive data on both transplantations (including both MAC and RIC/ 1. Kolb HJ, Schattenberg A, Goldman JM, et al. Graft-versus-leukemia effect NMA regimens) and disease-associated risk factors (in the 3 dis- of donor lymphocyte transfusions in marrow grafted patients. Blood. ease types selected for the study), and the long median follow-up 1995;86:2041–2050. 2. Boyiadzis M, Arora M, Klein JP, et al. Impact of chronic graft-versus-host of 4 years (49 months for MSD, 47 months for MUD). disease on late relapse and survival on 7,489 patients after myeloablative To our knowledge, this is the largest study reported to date allogeneic hematopoietic cell transplantation for leukemia. Clin Cancer addressing the impact of the T cell content of PBSC grafts on trans- Res. 2015;21:2020–2028. 3. Storb R, Gyurkocza B, Storer BE, et al. Graft-versus-host disease and graft- plantation outcomes. In this registry study, the CD3+T cell dose in versus-tumor effects after allogeneic hematopoietic cell transplantation. the PBSC products did not influence the risk of aGVHD or cGVHD J Clin Oncol. 2013;31:1530–1538. or other transplantation outcomes when using MSDs or 8/8 4. Deol A, Lum LG. Role of donor lymphocyte infusions in relapsed hemato- logical malignancies after stem cell transplantation revisited. Cancer Treat MUDs. Prospective studies are needed to determine whether the Rev. 2010;36:528–538. T cell subset (CD4+, CD8+, Tregs, and naïve T cells) contents of the 5. Ranganathan P, Heaphy CE, Costinean S, et al. Regulation of acute graft- allografts have a meaningful influence on transplantation out- versus-host disease by microRNA-155. Blood. 2012;119:4786–4797. 6. Coghill JM, Sarantopoulos S, Moran TP, Murphy WJ, Blazar BR, Serody JS. comes. The results of the ongoing phase II clinical trial using a Effector CD4+ T cells, the cytokines they generate, and GVHD: something standardized CD3+ T cell dose with HLA-matched related PBSC old and something new. Blood. 2011;117:3268–3276. transplantations are awaited (NCT00959140). In addition, in the 7. Soiffer RJ, Lerademacher J, Ho V, et al. Impact of immune modulation with anti-T cell antibodies on the outcome of reduced-intensity allogeneic current era of post-transplantation cyclophosphamide (PTCy) for hematopoietic stem cell transplantation for hematologic malignancies. prevention of GVHD, it may be imperative to assess the impact of Blood. 2011;117:6963–6970. these T cell subsets in haploidentical and HLA-matched HCT. 8. Soiffer RJ, Kim HT, McGuirk J, et al. Prospective, randomized, double-blind, Interestingly, a multicenter study has indeed indicated an phase III clinical trial of anti-T-lymphocyte globulin to assess impact on chronic graft-versus-host disease-free survival in patients undergoing increased risk of cGVHD of all severity with an elevated CD3+ HLA-matched unrelated myeloablative hematopoietic cell transplanta- T cell dose with haploidentical PBSC HCT using PTCy [41]. CD3+ tion. J Clin Oncol. 2017;35:4003–4011. A. Saad et al. / Biol Blood Marrow Transplant 25 (2019) 1875 1883 1883 9. Anasetti C, Logan BR, Lee SJ, et al. Peripheral-blood stem cells versus bone 25. Commenges D, Andersen PK. Score test of homogeneity for survival data. marrow from unrelated donors. N Engl J Med. 2012;367:1487–1496. Lifetime Data Anal. 1995;1:145–156. [discussion: 157-159]. 10. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia reactions 26. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning after bone marrow transplantation. Blood. 1990;75:555–562. regimens: working definitions. Biol Blood Marrow Transplant. 2009;15:1628– 11. Bensinger WI, Martin PJ, Storer B, et al. Transplantation of bone marrow as 1633. compared with peripheral-blood cells from HLA-identical relatives in 27. Jagasia M, Arora M, Flowers ME, et al. Risk factors for acute GVHD and sur- patients with hematologic cancers. N Engl J Med. 2001;344:175–181. vival after hematopoietic cell transplantation. Blood. 2012;119:296–307. 12. Pasquini MC, Devine S, Mendizabal A, et al. Comparative outcomes of 28. Waller EK, Logan BR, Harris WA, et al. Improved survival after transplanta- donor graft CD34+ selection and immune suppressive therapy as graft- tion of more donor plasmacytoid dendritic or naive T cells from unrelated- versus-host disease prophylaxis for patients with acute myeloid leukemia donor marrow grafts: results from BMTCTN 0201. J Clin Oncol. 2014;32: in complete remission undergoing HLA-matched sibling allogeneic 2365–2372. hematopoietic cell transplantation. J Clin Oncol. 2012;30:3194–3201. 29. Patel SS, Rybicki LA, Corrigan D, et al. Effect of bone marrow CD34+ cells and T 13. Bleakley M, Heimfeld S, Loeb KR, et al. Outcomes of acute leukemia cell subsets on clinical outcomes after myeloablative allogeneic hematopoietic patients transplanted with naive T cell-depleted stem cell grafts. J Clin cell transplantation. Bone Marrow Transplant. 2019;54:775–781. Invest. 2015;125:2677–2689. 30. McDonald-Hyman C, Flynn R, Panoskaltsis-Mortari A, et al. Therapeu- 14. Mielcarek M, Furlong T, O'Donnell PV, et al. Posttransplantation cyclo- tic regulatory T cell adoptive transfer ameliorates established murine phosphamide for prevention of graft-versus-host disease after HLA- chronic GVHD in a CXCR5-dependent manner. Blood. 2016;128:1013– matched mobilized blood cell transplantation. Blood. 2016;127:1502– 1017. 1508. 31. Alho AC, Kim HT, Chammas MJ, et al. Unbalanced recovery of regulatory 15. Guo M, Hu KX, Liu GX, et al. HLA-mismatched stem-cell microtransplanta- and effector T cells after allogeneic stem cell transplantation contributes tion as postremission therapy for acute myeloid leukemia: long-term fol- to chronic GVHD. Blood. 2016;127:646–657. low-up. J Clin Oncol. 2012;30:4084–4090. 32. Farhan S, Mckinnon R, Fortney C, Divine G, Janakiraman N. CD3 cell dose 16. Reshef R, Huffman AP, Gao A, et al. High graft CD8 cell dose predicts and outcome after allogeneic stem cell transplantation. Biol Blood Marrow improved survival and enables better donor selection in allogeneic stem- Transplant. 2010;16:S279. cell transplantation with reduced-intensity conditioning. J Clin Oncol. 33. To€ rle n J, Ringde n O, Le Rademacher J, et al. Low CD34 dose is associated 2015;33:2392–2398. with poor survival after reduced-intensity conditioning allogeneic trans- 17. Pastore D, Delia M, Mestice A, et al. CD3+/Tregs ratio in donor grafts is plantation for acute myeloid leukemia and myelodysplastic syndrome. linked to acute graft-versus-host disease and immunologic recovery after Biol Blood Marrow Transplant. 2014;20:1418–1425. allogeneic peripheral blood stem cell transplantation. Biol Blood Marrow 34. Yamamoto C, Ogawa H, Fukuda T, et al. Impact of a low CD34+ cell dose on Transplant. 2012;18:887–893. allogeneic peripheral blood stem cell transplantation. Biol Blood Marrow 18. Saad A, Almubarak M, Kanate A, et al. Balancing acute graft versus host Transplant. 2018;24:708–716. disease (aGVHD) and survival after peripheral allogeneic stem cell trans- 35. Zaucha JM, Gooley T, Bensinger WI, et al. CD34 cell dose in granulocyte plantation (SCT) in hematological malignancies: a potential for graft engi- colony-stimulating factor-mobilized peripheral blood mononuclear cell neering. Presented at 49th Annual Meeting of the American Society of grafts affects engraftment kinetics and development of extensive chronic Hematology. December 8-11, 2007; Atlanta, GA. graft-versus-host disease after human leukocyte antigen-identical sibling 19. Nakamura R, Bahceci E, Read EJ, et al. Transplant dose of CD34(+) and CD3(+) transplantation. Blood. 2001;98:3221–3227. cells predicts outcome in patients with haematological malignancies under- 36. Mohty M, Bilger K, Jourdan E, et al. Higher doses of CD34+ peripheral blood going T cell-depleted peripheral blood stem cell transplants with delayed stem cells are associated with increased mortality from chronic graft- donor lymphocyte add-back. Br J Haematol. 2001;115:95–104. versus-host disease after allogeneic HLA-identical sibling transplantation. 20. Czerw T, Labopin M, Schmid C, et al. High CD3+ and CD34+ peripheral blood Leukemia. 2003;17:869–875. stem cell grafts content is associated with increased risk of graft-versus-host 37. Heimfeld S. Bone marrow transplantation: how important is CD34 cell dose in disease without beneficial effect on disease control after reduced-intensity HLA-identical stem cell transplantation? Leukemia. 2003;17:856–858. conditioning allogeneic transplantation from matched unrelated donors for 38. Hahn T, McCarthy Jr PL, Zhang MJ, et al. Risk factors for acute graft-versus- acute myeloid leukemia: an analysis from the Acute Leukemia Working Party host disease after human leukocyte antigen-identical sibling transplants of the European Society for Blood and Marrow Transplantation. Oncotarget. for adults with leukemia. J Clin Oncol. 2008;26:5728–5734. 2016;7:27255–27266. 39. Arora M, Klein JP, Weisdorf DJ, et al. Chronic GVHD risk score: a Center for 21. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on International Blood and Marrow Transplant Research analysis. Blood. Acute GVHD Grading. Bone Marrow Transplant. 1995;15:825–828. 2011;117:6714–6720. 22. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health 40. Armand P, Kim HT, Logan BR, et al. Validation and refinement of the Dis- consensus development project on criteria for clinical trials in chronic ease Risk Index for allogeneic stem cell transplantation. Blood. 2014;123: graft-versus-host disease: I. Diagnosis and staging working group report. 3664–3671. Biol Blood Marrow Transplant. 2005;11:945–956. 41. Mussetti A, De Philippis C, Carniti C, et al. CD3+ graft cell count influence on 23. Zhang X, Loberiza FR, Klein JP, Zhang MJ. A SAS macro for estimation of chronic GVHD in haploidentical allogeneic transplantation using post-trans- direct adjusted survival curves based on a stratified Cox regression model. plant cyclophosphamide. Bone Marrow Transplant. 2018;53:1522–1531. Comput Methods Programs Biomed. 2007;88:95–101. 42. Urbano-Ispizua A, Rozman C, Pimentel P, et al. The number of donor CD3 24. Zhang X, Zhang MJ. SAS macros for estimation of direct adjusted cumula- (+) cells is the most important factor for graft failure after allogeneic trans- tive incidence curves under proportional subdistribution hazards models. plantation of CD34(+) selected cells from peripheral blood from HLA- Comput Methods Programs Biomed. 2011;101:87–93. identical siblings. Blood. 2001;97:383–387.