Formula predicts survival among adults with graft-versus-host disease
Click Here to Manage Email Alerts
A simple formula predicted OS and nonrelapse mortality among adults with acute graft-versus-host disease who underwent allogeneic stem-cell transplantation with reduced-intensity conditioning, according to retrospective study results.
“The present clinical application of [this formula] is an easy-to-use prediction of acute GVHD mortality in patients who have undergone reduced-intensity conditioning allogeneic stem-cell transplantation, raising awareness of the endothelial nature of the problem,” Thomas Luft, MD, chief physician in the department of medicine at University Hospital Heidelberg in Germany, and colleagues wrote. “The [formula] provides a tool for studies with these therapeutic approaches and could be the future basis for development of risk-adapted GVHD treatment strategies.”
Most individuals who develop GVHD after allogeneic stem-cell transplantation improve with corticosteroid treatment.
However, some patients are steroid resistant, and mortality for this group is high. Early identification of this subset could allow for use of alternative treatments.
Preclinical studies linked endothelial cell dysfunction to GVHD pathogenesis and refractoriness, and several endothelium-related factors have been connected to GVHD outcomes, according to study background.
Luft and colleagues sought to determine the importance of endothelial biology for GVHD pathophysiology.
They analyzed a training cohort and three validation cohorts of patients with acute GVHD who underwent allogeneic stem-cell transplantation at one of four medical centers in Germany and the United States. All patients underwent GVHD prophylaxis.
The researchers constructed a formula called the Endothelial Activation and Stress Index (EASIX), calculated by the following formula: lactate dehydrogenase (U/L)×creatinine (mg/dL)/thrombocytes (109 cells per L). Investigators assessed whether EASIX could help predict death among patients who developed acute GVHD after allogeneic stem-cell transplantation.
The training cohort consisted of 311 patients with acute GVHD, of whom 239 (77%) received reduced-intensity conditioning. The validation cohorts comprised 403 patients, of whom 346 (85.8%) received reduced-intensity conditioning.
In the training cohort, univariate analysis showed EASIX-GVHD strongly predicted OS (HR per onefold increase = 1.16; 95% CI, 1.12–1.2) and nonrelapse mortality (HR = 1.19; 95% CI, 1.15-1.25) as measured from onset of acute GVHD.
In the subset of patients (n = 72) who underwent myeloablative conditioning, EASIX-GVHD did not predict OS. Researchers attributed this to thrombocytopenia at GVHD onset (73×109 cells per L [interquartile range (IQR), 29.75-180] for myeloablative conditioning vs. 160×109 cells per L [IQR, 90-250.5] for reduced-intensity conditioning; P < .0001).
However, among patients who received reduced-intensity conditioning, univariate analysis and multivariate Cox models showed EASIX-GVHD to be a strong predictor of OS (HR for twofold change = 1.23, 95% CI 1.13–1.34) and nonrelapse mortality (HR for twofold change = 1.24; 1.12–1.38).
Model validation for prediction of OS and nonrelapse mortality by EASIX-GVHD appeared successful in two independent cohorts of adults who underwent reduced-intensity conditioning (n = 314), as well as in a cohort that consisted primarily of pediatric patients (n = 89).
“A prospective study for further validation of our results is underway,” the researchers wrote. “To address another possible clinical application, we are currently performing retrospective analyses of EASIX taken before start of conditioning as a predictor of allogeneic stem-cell transplantation outcome.” – by Melinda Stevens
Disclosures: The researchers report no relevant financial disclosures.