Analysis of risk factor for the development of GVHD after T cell-depleted allogeneic BMT: Effect of HLA disparity, ABO incompatibility, and method of T-cell depletion

Carolyn A. Keever-Taylor, Christopher Bredeson, Fausto R. Loberiza, James T. Casper, Colleen Lawton, Douglas Rizzo, William H. Burns, David A. Margolis, David H. Vesole, Mary Horowitz, Mei Jie Zhang, Mark Juckett, William R. Drobyski

Research output: Contribution to journalArticlepeer-review

54 Scopus citations


Multivariate analysis was performed to determine the independent factors affecting the risk of acute GVHD (aGVHD) grades II to IV and extensive chronic GVHD (cGVHD) and the rate of survival in 481 recipients of T cell-depleted (TCD) marrow allografts who received transplants at a single center between 1991 and 2000. All patients received grafts partially depleted of CD3+ T cells by complement-mediated lysis using 2 narrow-specificity monoclonal antibodies (MoAbs), T10B9.1A-31 (n = 400) or Muromonab-Orthoclone OKT3 (n = 81). Factors considered in the analysis included patient/donor sex, age, cytomegalovirus (CMV) status, and ABO blood group along with T-cell dose, disease and disease status, donor relationship, HLA antigen (Ag) mismatch (MM), growth-factor use, anti-thymocyte globulin use, year of transplantation, and the MoAb used for TCD. The results showed an association of HLA MM with an increased relative risk (RR) of aGVHD for recipients of grafts from related donors that were ≥2 Ag MM (n = 73, RR = 2.09, P = .005), matched unrelated (UR) donors (n = 130, RR = 1.98, P = .004), and ≥2 Ag MM UR donors (n = 34, RR = 2.68, P = .003) compared with the baseline matched-sibling group (n = 121). No increased risk of aGVHD was seen for 0 to 1 Ag MM family donors (n = 24) or 1 Ag MM UR donors (n = 99). aGVHD risk was increased with minor, but not major or major-minor, ABO disparity (RR = 2.0, P = .003) compared with that of ABO-identical pairs. We found less effective TCD and resultant higher T-cell dose for recipients of grafts that were T cell depleted using OKT3. However, the use of OKT3 and not the T-cell dose was associated with increased aGVHD risk (RR of 1.84, P = .001). Increased risk of extensive cGVHD was associated with patient age of >20 years (RR = 2.2, P < .0001) and with CMV status (positive patient/negative donor, RR = 1.9, P = .002). Decreased survival was associated with older age (>20 years), a ≥2 Ag MM related donor, a 1 or ≥2 Ag MM UR donor, risk group, and a CMV-positive patient/-negative donor pair. There was no difference in survival for 0 to 1 Ag MM related or matched UR donors compared with the baseline group. These data indicate that there are quantitative as well as potential qualitative differences in outcome depending on the TCD method. Expected and unexpected risk factors for GVHD and survival were associated with partial TCD. Our data support the consideration of ABO match in donor selection, the preferential selection of CMV-positive donors for CMV-positive recipients, and the acceptance of 1 but not ≥2 Ag HLA MM donors.

Original languageEnglish (US)
Pages (from-to)620-630
Number of pages11
JournalBiology of Blood and Marrow Transplantation
Issue number11
StatePublished - 2001
Externally publishedYes


  • ABO compatibility
  • Bone marrow transplantation
  • CMV serostatus
  • Graft-versus-host disease
  • T-cell depletion


Dive into the research topics of 'Analysis of risk factor for the development of GVHD after T cell-depleted allogeneic BMT: Effect of HLA disparity, ABO incompatibility, and method of T-cell depletion'. Together they form a unique fingerprint.

Cite this