Fluorodeoxyglucose F 18 for the Assessment of Acute Intestinal Graft-versus-Host Disease and Prediction of Response to Immunosuppressive Therapy.

Graft-versus-host disease (GVHD) is a common complication that increases morbidity and mortality after allogeneic stem cell transplantation (allo-SCT). Fluorodeoxyglucose F 18 (F-FDG)-positron emission tomography (PET) imaging has been demonstrated to be highly informative for evaluating and mapping of intestinal GVHD. To corroborate and extend existing findings and to investigate whether glucose metabolism assessed by F-FDG-PET might be an effective diagnostic tool to predict corticosteroid-refractory acute GVHD and overall survival. In this retrospective analysis, 101 patients with clinically suspected acute intestinal GVHD underwent F-FDG-PET between June 2011 and February 2019. Seventy-four of these patients with clinically and/or histologically proven acute intestinal GVHD as well as positive F-FDG-PET findings were analyzed in detail to assess the predictive value of F-FDG-PET regarding the response to immunosuppressive therapy and survival. Quantitative PET parameters, particularly the maximum standard uptake value (SUVmax), of patients with a fast response (ie, clinical improvement and decreased GVHD activity by at least 1 stage after 1 week of GVHD treatment) or slow/no response (ie, persistent disease activity for more than 1 week or increasing GVHD activity following first-line immunosuppressive therapy) were evaluated. F-FDG-PET detected intestinal GVHD with a sensitivity of 93% (95% confidence interval [CI], 85% to 97%) and specificity of 73% (95% CI, 45% to 91%). Patients with a fast response to immunosuppressive therapy had a mean SUVmax of 13.7 (95% CI, 11.0 to 16.5) compared with 7.6 (95% CI, 7.0 to 8.3; P = .005) observed in patients with prolonged or no response. The median overall survival (OS) was 573.0 days (95% CI, 539.5 to 606.5 days) for patients with fast response versus 255 days (95% CI, 161.0 to 349.0 days; P = .009) for patients with slow or no responses. A SUVmax threshold >8.95 applied to F-FDG-PET performed within 100 days after transplantation identified patients with a median OS of 390 versus 117 days for patients with SUVmax ≤8.95 (P = .036). SUVmax threshold and donor type were independent factors for OS. Our results indicate that F-FDG-PET is highly accurate in identifying patients with acute intestinal GVHD and may predict responses to immunosuppressive therapy as well as survival, particularly when applied within the first 100 days after transplantation. These results provide a strong rationale to integrate PET imaging in future prospective trials evaluating new therapies for acute GVHD.