LAUSR

Regenerating islet-derived protein 3 alpha (REG3A) is a C-type lectin constitutively expressed and apically secreted by Paneth cells at high levels, with powerful bactericidal activity against gram-positive bacteria of the gut [1]. By limiting the number of mucosal adherent bacteria, REG3A separates gut microbiota and epithelium, promoting intestinal tolerance to microbial antigens and reducing bacterial translocation [2]. Circulating REG3A has been used as a biomarker of gastrointestinal GVHD, and has been used to predict response to therapy and nonrelapse mortality (NRM) [3–5]. A recent study reported a novel association between a single-nucleotide polymorphism (SNP rs7588571; minor allele A with frequency 0.49) in the putative promotor region of REG3A and extensive chronic GVHD (ecGVHD) [6]. In that study, performed in 119 evaluable adult Japanese allogeneic HCT recipients, a higher incidence of ecGVHD was observed in patients with the non-GG genotype than in those with the GG genotype (odds ratio in multivariable analysis: 2.6; 95% confidence interval [95% CI]: 1.1–6.0; P= 0.029). No association was found between this SNP and acute GVHD (overall and grade II–IV). All patients received a bone marrow graft from an HLA-matched sibling and GVHD prophylaxis with cyclosporine and methotrexate. The median age was ~37 years and conditioning was myeloablative (MA) in >95% of transplants. Our goal in the present study was to test the reproducibility of the observed association between rs7588571 and ecGVHD using samples from two Blood and Marrow Transplant Clinical Trials Network (BMT CTN) protocols 0201 and 0901. BMT CTN-0201 was a phase 3 randomized multicenter clinical trial (2004–2009, 551 patients) comparing bone marrow (BM) vs. peripheral blood (PB) as graft source in HCT from unrelated donors, and with 2-year overall survival as the primary endpoint [7]. The results demonstrated similar rates of acute GVHD between the groups, but higher rates of cGVHD among recipients of PB grafts (2-year cGVHD: 53% vs. 41% in PB vs. BM groups, respectively). BMT CTN-0901 was a phase 3 randomized multicenter clinical trial (2011–2014, 272 patients, matched sibling or unrelated donors) comparing myeloablative (MA) vs. reduced-intensity (RI) conditioning in patients with acute myeloid leukemia or myelodysplastic syndromes, and with the primary endpoint of 18-month overall survival [8]. The graft source was PB in 92% of patients and BM in 8%. The results demonstrated higher rates of acute and chronic GVHD among MA patients (1.5-year cGVHD: 66% vs. 37% in myeloablative vs. RI, respectively). From BMT CTN-0201, we excluded patients with one or more HLA locus mismatch (HLA-A, B, C, or DRB1; These authors contributed equally: Armin Rashidi, Ryan Shanley.