Predicting Steroid Refractroy Acute Gastrointestinal (GI) Acute Graft Versus Host Disease (AGVHD) by Assessment of Crypt Loss at Diagnosis.

Acute graft versus host disease accounts for most of the treatment failures after an allogeneic stem cell transplantation. The prognosis of patients with steroid refractory acute graft versus host disease is poor with conventional treatment. Early identification of patients with steroid refractory acute graft versus host disease may improve the transplant outcomes. Crypt loss has been associated with the severity of disease in patients with inflammatory bowel disease. It is also found in patients with acute gastrointestinal graft versus host disease with unknown clinical significance. In an attempt to evaluate the significance of crypt loss in patients with acute gastrointestinal graft versus host disease, we conducted a retrospective study on 22 consecutive post allogeneic stem cell transplant patients in our institution that underwent colonoscopy and colonic biopsy. In this study, crypt loss was graded and correlated with clinical parameters of disease severity including stool volume, endoscopic appearance, response to graft versus host disease treatment and one year mortality attributed to graft versus host disease. All the biopsies were reviewed by one of the co-authors (Shriram Jakate, M.D.). Twenty two consecutive patients who were treated for gastrointestinal graft versus host disease following stem cell transplantation were studied. Crypt loss was present in 16/22 cases and in 10/22 cases crypt loss was determined to be severe by the presence of contiguous areas of crypt loss. In those with severe crypt loss 9/10 patients had daily stool volumes in excess of 1000 ml/day while only 3/6 in those with minimal crypt loss had this level of severe diarrhea. Patients with severe crypt loss were more likely to have a pathologic appearance at endoscopy and to require second-line therapy (i.e. steroid refractory acute graft versus host disease). The diarrhea of those with severe crypt loss was less likely to resolve even though it was more aggressively treated. Of the 10 patients with severe crypt loss, 5 (50%) died of graft versus host disease related causes. Conversely, only one of 12 patients (8%) with mild or no crypt loss had a death attributable to acute graft versus host disease. Our study shows that severe crypt loss predicts for severity of acute gastrointestinal graft versus host disease, response to treatment and more importantly steroid refractoriness. In addition, crypt loss severity is associated with an increased risk of death related to acute graft versus host disease. We conclude that the assessment of crypt loss at diagnosis may serve as a tool to identify patients with steroid refractory acute gastrointestinal graft versus host disease and could be used to select patients for therapeutic trial in acute graft versus host disease.