A Novel Grading System of Lower Gastrointestinal Acute Graft-Versus-Host Disease At Disease Onset Predicts Response to Therapy and Non-Relapse Mortality

Plasma REG3α concentrations, serum albumin concentrations, clinical stage, and histologic grade as defined by Lerner (Transplant Proc; 6: 367–71) were all available at GVHD diagnosis for 140 patients experiencing lower GI GVHD from 3 HCT centers (129 from the University of Michigan and 11 from Regensburg, Germany, and Kyushu, Japan). Patients with lower GI GVHD who did not respond to GVHD therapy at 4 weeks had REG3α concentrations over 3-fold higher at onset than those who experienced at least a partial response (p<0.001). Patients with lower GI GVHD who had onset plasma REG3α concentrations above the median (>151 ng/ml) had increased 1-year NRM compared to those who had concentrations below the median (≤151 ng/ml; 59% versus 34%, respectively; p<0.001; Figure 1A). NRM was modeled with cumulative incidence regression methods as described by Fine and Gray (J Am Stat Assoc; 94: 496–509); probability of response was modeled with logistic regression. We performed an independent model analysis and found all 4 risk factors independently predicted lack of response to GVHD therapy four weeks following treatment and predicted 1-year NRM after adjustment for known risk factors of donor type, degree of HLA match, conditioning intensity, age and baseline disease severity (odds ratios: 4.8, 3.9, 18.9, and 2.5, respectively; Table 1). When lack of response to therapy and NRM were modeled simultaneously on all four parameters, all but albumin remained statistically significant when adjusting for the aforementioned risk factors. High REG3α concentration and histologic grade displayed the highest odds ratios for no response to treatment (5.7 and 16.7, respectively), while advanced clinical stage and histologic grade provided the highest hazard ratio for NRM (3.1 and 2.9, respectively). When only advanced clinical stage and severe histologic grade were considered, NRM was 71% (Figure 1B), but the inclusion of high REG3α concentration produced a significantly greater NRM of 86% for patients with all three risk factors (p<0.001; Figure 1C). Patients with increasing numbers of high risk parameters had increasing rates of NRM. Thus all three high risk parameters provided important prognostic information prior to the initiation of therapy rather than at the time of maximum grade of GVHD. If the prognostic value of REG3α is confirmed in additional patients, we believe the integration of clinical stage, histologic grade and REG3α plasma concentrations into a single grading system will permit better risk stratification and rapid identification of those patients with severe GI damage in whom standard treatment is likely to be insufficient.