Pretransplant comorbidities predict severity of acute graft-versus-host disease and subsequent mortality

• Published in: Blood Vol. 124; no. 2; pp. 287 - 295
• Main Authors: Sorror, Mohamed L., Martin, Paul J., Storb, Rainer F., Bhatia, Smita, Maziarz, Richard T., Pulsipher, Michael A., Maris, Michael B., Davis, Christopher, Deeg, H. Joachim, Lee, Stephanie J., Maloney, David G., Sandmaier, Brenda M., Appelbaum, Frederick R., Gooley, Theodore A.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 10.07.2014; American Society of Hematology
• Subjects: Adolescent; Adult; Aged; Child; Child, Preschool; Comorbidity; Graft vs Host Disease - diagnosis; Graft vs Host Disease - etiology; Graft vs Host Disease - mortality; Hematologic Diseases - diagnosis; Hematologic Diseases - mortality; Hematologic Diseases - therapy; Hematopoietic Stem Cell Transplantation - adverse effects; Hematopoietic Stem Cell Transplantation - statistics & numerical data; Humans; Infant; Infant, Newborn; Middle Aged; Prognosis; Retrospective Studies; Transplantation; Transplantation Conditioning - adverse effects; Transplantation Conditioning - methods; Transplantation Conditioning - mortality; Young Adult
• ISSN: 0006-4971; 1528-0020; 1528-0020
• DOI: 10.1182/blood-2014-01-550566

Whether the hematopoietic cell transplantation comorbidity index (HCT-CI) can provide prognostic information about development of acute graft-versus-host disease (GVHD) and subsequent mortality is unknown. Five institutions contributed information on 2985 patients given human leukocyte antigen-matched grafts to address this question. Proportional hazards models were used to estimate the hazards of acute GVHD and post-GVHD mortality after adjustment for known risk variables. Higher HCT-CI scores predicted increased risk of grades 3 to 4 acute GVHD (P < .0001 and c-statistic of 0.64), and tests of interaction suggested that this association was consistent among different conditioning intensities, donor types, and stem cell sources. Probabilities of grades 3 to 4 GVHD were 13%, 18%, and 24% for HCT-CI risk groups of 0, 1 to 4, and ≥5. The HCT-CI was statistically significantly associated with mortality rates following diagnosis of grade 2 (hazard ratio [HR] = 1.24; P < .0001) or grades 3 to 4 acute GVHD (HR = 1.19; P < .0001). Patients with HCT-CI scores of ≥3 who developed grades 3 to 4 acute GVHD had a 2.63-fold higher risk of mortality than those with scores of 0 to 2 and did not develop acute GVHD. Thus, pretransplant comorbidities are associated with the development and severity of acute GVHD and with post-GVHD mortality. The HCT-CI could be useful in designing trials for GVHD prevention and could inform expectations for GVHD treatment trials. •The HCT-CI stratifies patients into 3 groups for risks of grades 3 to 4 GVHD regardless of conditioning intensity, donor, or graft types.•Comorbidity burden and development of grades 2 to 4 acute GVHD have cumulative effects on mortality rates.