High Alpha-1 Antitrypsin Clearance Predicts Severity of Gut Graft Versus Host Disease (GVHD) In Children

Abstract 1242

The clinical evaluation and management of gut GVHD is a significant challenge in pediatric hematopoietic stem cell transplantation (HSCT). It is often difficult to obtain pathological evidence to confirm the GVHD diagnosis and/or determine response to treatment. The severity of the disease itself may not just be related to the associated classic symptoms. Our objective herein was to prospectively study the effect of gut GVHD on protein losing enteropathy (PLE) as measured by alpha-1 antitrypsin clearance (AATC) in stools. Thirteen patients who were diagnosed with gut GVHD by clinical criteria were recruited; 5 patients were excluded as 4 had gut GVHD ruled out by biopsy and 1 was unable to complete study due to stool collection issues. Therefore, 8 patients; 6 males and 2 females, were studied. The median age was 9.5 years (range 6–17). Diagnoses included ALL (4), AML (2), Lymphoma (1) and Adrenoleukodystrophy (1). Donor types were; 7 unrelated (BM n=4 and cord n=3) one matched related BM donor. All patients received cyclosporine for GVHD prophylaxis, in addition 4 patients also received methotrexate and 3 others received prednisone as the second agent. All patients had negative stool EM for viruses and cultures for C.difficile on their first collection. Two 24 hour stool collections were taken from each patient at a mean of 8.5d (range 7–13d) apart. Patients’ gut GVHD staging is summarized in Table 1.

At the time of 1^st collection, 6 patients had ≥ stage II gut GVHD and at 2^nd collection 4 patients had ≥stage II gut GVHD and 4 collections were of non-diarrheal stool. A total of 7 stool collections were taken at a diagnosis of ≥stage III gut GVHD. Mean alpha-1 antitrypsin clearance (AATC) from all 16 collections was 143 mls/day (range 3–561), levels >22 mls/day indicate the diagnosis of protein-losing enteropathy (PLE). Nine of the collections (56%) showed PLE with a mean AATC of 247mls/day (range 31–561). Six out of the 7 samples from patients with ≥stage III gut GVHD were positive for PLE. Five collections with stool volumes >30 ml/kg/day were positive for PLE. Stool volumes were significantly higher on second collection (Wilcoxon signed ranks test, p = 0.003) consistent with the second stool collections being significantly more likely to have an elevated AATC, therefore more severe PLE over time (p<0.001). We conclude that a highly significant positive correlation exists between the severity of PLE and the stage of gut GVHD, particularly obvious in patients with stage 3–4 GVHD (ANOVA, p=0.015). See Table 2.

When patients were grouped as stage 0–2 GVHD vs. stage 3–4 GVHD, those with stage 3–4 GVHD had significantly higher AATC values (p =0.02). Despite the small number of patients recruited, this study emphasizes the need to consider PLE as a useful aspect of the clinical picture. We suggest that in order to see a response to therapy and therefore a decrease in AATC, clinicians should not repeat stool collections before 2 weeks from the initiation of therapy. In light of the significant morbidity and mortality associated with grade 3 and 4 gut GVHD, and as an important therapeutic decision for these patients, one may consider escalating GVHD therapy if a patient's AATC is rising.