18F-FDG PET scanning in graft-versus-host disease after allogeneic stem cell transplantation

Background: diagnosis and onset prediction of Graft-versus-host disease (GvHD) after hematopoietic stem cell transplantation (HSCT) as well as therapy monitoring is still an open issue. Aim: to evaluate the role of FDG PET for the assessment of GvHD in patients after HSCT. Methods: we retrospectively studied patients with malignant lymphoma and multiple myeloma who underwent FDG PET between 2004 and 2009. We included patients who underwent HSCT and performed PET scan before transplant and at least three follow up PET scans. Overall 36 patients (8 non-Hodgkin lymphoma, 20 Hodgkin's disease, 8 multiple myeloma) were included. GvHD diagnosis was based on clinical data and, when available, on biopsy results. PET scans were evaluated by two experienced nuclear medicine physicians unaware of clinical data: findings of scans acquired at the time of GvHD clinical manifestation were compared with scans obtained before transplant and with scans during clinical remission, as well as with PET scans carried out in patients without GvHD. Results: overall 25/36 patients developed GvHD (22 acute, 9 chronic) while 11 remained GvHD clinically-free during the time of follow up. In 8/25 patients one or more PET scans were performed during clinical GvHD, and PET was positive in 8/8 cases. Pathological FDG uptake was evident at intestinal and/or liver and/or mouth level. Intestinal GvHD was clinically observed in 6/8 patients. PET corresponding images showed intense uptake in the bowel in comparison to the scans obtained in the same patients in the GvHD clinically-free period. A strong decrease in bowel FDG uptake was observed in patients responding to therapy. In 2 cases in which liver GvHD was pathologically confirmed, PET was positive in one case. Among 6 patients with mouth GvHD, only 1 showed intense FDG uptake as compared to scans obtained in the GvHD clinically-free group. Among the 11 patients who did not develop GvHD clinical manifestation, only one showed a faint intestinal FDG uptake. Overall considering sites of documented GvHD, PET was true positive in all cases presenting intestinal GvHD (6), false negative at liver in 1/2 cases, false negative at mouth in 5/6 cases and false positive in 1 patient not presenting GvHD. Conclusion: our preliminary data show that together with clinical findings, FDG PET could be useful in diagnosing and monitoring GvHD, especially at bowel level.