Graft-versus-host disease and outcome in HLA-identical sibling transplantations for chronic myeloid leukemia

Alois Gratwohl, Ronald Brand, Jane Apperley, Anja V. Biezen, Giuseppe Bandini, Agnes Devergie, Anton Schattenberg, Francesco Frassoni, Cesare Guglielmi, Simona Iacobelli, Mauricette Michallet, Hans Jochen Kolb, Tapani Ruutu, Dietger Niederwieser

Research output: Contribution to journalArticlepeer-review


Graft-versus-host disease in its acute (aGvHD) or chronic form (cGvHD) remains the most important posttransplantation factor influencing outcome after allogeneic hematopoietic stem cell transplantation (HSCT). It increases transplantation-related mortality (TRM) but reduces risk of relapse. The net effect of these 2 discordant effects determines survival. In view of current interests to exploit graft-versus-leukemia (GVL) effects, we analyzed 4174 HLA-identical sibling transplantations for chronic myeloid leukemia in first chronic phase, depending on the presence or absence and severity of GvHD with a landmark analysis. During the first 100 days, only aGvHD grades III and IV had an impact on TRM. During the time period day 100 to 3 years increasing severity of aGvHD is associated with increased TRM and decreased relapse incidence (RI) with hazard ratios (HRs) for TRM as follows: grade 0, HR = 1.0; grade I, HR = 1.52 (1.19-1.96); grade II, HR = 2.48 (1.95-3.14); grade III, HR = 5.76 (4.44-7.48); grade IV, HR = 14.7 (10.9-19.9) and likewise for RI: grade I versus 0, HR = 0.94 (0.76-1.16); grade II, HR = 0.60 (0.46-0.77); grade III, HR = 0.48 (0.29-0.81); grade IV, HR = 0.14 (0.02-0.99). Beyond 3 years, TRM and RI are determined by cGvHD. Limited cGvHD reduces RI to the same extent as extensive cGvHD but has no impact on TRM and, hence, results in best survival with an HR = 0.48 (0.32-0.71). aGvHD grade I has the highest likelihood of subsequent limited cGvHD, which results in cumulative incidence estimates of survival at 10 years being best for patients with initial aGvHD grade I: survival at 10 years grade 0 = 59%, I = 63%, II = 56%, III = 26%, IV = not applicable. These data clarify the role of GvHD in posttransplantation outcome. Considerations for long-term outcome are essential when short-term data of interventions on GvHD are analyzed.

Original languageEnglish
Pages (from-to)3877-3886
Number of pages10
Issue number12
Publication statusPublished - Dec 1 2002

ASJC Scopus subject areas

  • Hematology


Dive into the research topics of 'Graft-versus-host disease and outcome in HLA-identical sibling transplantations for chronic myeloid leukemia'. Together they form a unique fingerprint.

Cite this