Work

A Three Biomarker Panel at Days 7 and 14 Can Predict Development of Grade II-IV Acute Graft-Versus-Host Disease

Year: 2011

Type: article

Abstract: We have previously identified five biomarker proteins that have diagnostic and prognostic value for acute graft-versus-host disease (GVHD) (Blood 113:273-278, Sci Transl Med 2:50-57). To determine whether biomarkers can predict GVHD before the appearance of clinical symptoms, we measured the three most informative biomarkers (IL2Rα, TNFR1, elafin) by sequential ELISA in patient samples prospectively obtained prior to conditioning (pre-HCT), on day+7, and day+14 after hematopoietic cell transplant (HCT) from 513 unrelated (URD) HCT patients. We randomly divided the patients into training (N = 342) and validation (N = 171) data sets that were balanced for (i) full intensity conditioning, (ii) early onset of grade II-IV GVHD (< day+21), and (iii) HCT performed after 2005. We used the training data set to simulate biomarker values for a hypothetical 50,000 patients using the following assumptions: (1) the incidence of GVHD by day 100 is 55%; (2) the median day of GVHD onset after full intensity URD HCT is day 21, with 10% developing GVHD prior to day+7, and 3% after day+56. We used logistic regression to assign each of the 50,000 patients to low risk and high risk of developing grade II-IV GVHD based upon the biomarker levels pre-HCT and at day+7. We measured all three biomarkers again in samples taken at day+14 in all patients who were categorized as low risk and who did not develop GVHD by day+14. We determined with another logistic regression model which low risk patients had now moved to the high risk category. In the training data set, the combination of these two analyses correctly predicted the continued absence of grade II-IV GVHD (i.e. specificity) in 77% (95% CI: 71%-83%) and correctly predicted the eventual development of GVHD (i.e. sensitivity) in 50% (95% CI: 42%-58%) of patients. In the validation data set, these tests gave a specificity of 75% (95% CI: 67%-83%) and a sensitivity of 57% (95% CI: 44%-69%), (Table 1, first row).Table 1Sensitivity and Specificity Corresponding to Various Values of Predicted Probability at Days +7 and +14 Applied to Validation Data SetCombined Specificity for both cutpointsCombined Sensitivity for both cutpointsCorrect Prediction of GVHD by Day 5675%57%65%74%48%60%73%49%60%71%58%64%70%57%63% Open table in a new tab The median time between a high risk determination and onset of grade II-IV GVHD was 14 days. These cutpoints also led to the greatest correct prediction of GVHD facilitating the design of a trial to test preemptive treatment of GVHD using these panels. In conclusion, measurement of a three-biomarker panel pre-HCT, at day+7, and day+14 predicts grade II-IV GVHD with good specificity (75%) and fair sensitivity (57%).

Source: Transplantation and Cellular Therapy

DOI: 10.1016/j.bbmt.2010.12.048