Profiling Tissue and Biofluid miR-155-5p, miR-155^*, and miR-146a-5p Expression in Graft vs. Host Disease

Introduction: Acute graft vs. host disease (aGvHD) is a frequent complication following allogeneic haematopoeitic transplantation (HSCT). Despite recent advances, there are no universally accepted biomarkers to determine development of aGvHD. MicroRNAs miR-146a and miR-155 have been previously associated with aGvHD and show promise as clinically translatable biomarkers. In this study, we performed comprehensive expression profiling of miR-146a, miR-155, and miR-155^* expression in aGvHD target tissue and biofluids and relate expression to post-HSCT outcomes. Materials and Methods: MicroRNA expression was assessed by qRT-PCR in gastrointestinal (n = 31) and skin (n = 31) biopsies as well as serum (exploratory cohort n = 34, verification cohort n = 81, diagnostic cohort n = 65) and urine (exploratory cohort n = 30, verification cohort n = 56, diagnostic cohort n = 20) biofluids, including extracellular vesicle (EV) cohorts (serum EV n = 15, urine EV n = 30). Expression was related to aGvHD incidence, severity and overall survival. Results: In GI samples, expression of miR-155 (p = 0.03) and miR-146a (p = 0.03) was higher at aGvHD onset compared to patients with no GvHD. In skin biopsies, expression of miR-155 (p = 0.004) was upregulated in aGvHD patients compared to normal control skin. Expression of miR-146a was higher in aGvHD compared to no aGvHD biopsies (p = 0.002). In serum, miR-155 (p = 0.03) and miR-146a (p = 0.02) expression was higher at day 14 (D14), while in urine expression was elevated at D7 post-HSCT in patients who developed aGvHD compared to those disease-free. This was verified in an independent serum (miR-155 p = 0.005, miR-146a p = 0.003) and urine (miR-155 p = 0.02, miR-146a p = 0.04) cohort, where both microRNAs were also associated with aGvHD by ROC analysis. In serum and urine samples taken at the time of aGvHD symptoms, expression of miR-155 and miR-146a was also elevated (serum miR-155 p = 0.03, miR-146a p < 0.001; urine miR-155 p = 0.02, miR-146a p = 0.02). In contrast, miR-146a and miR-155 were downregulated at D14 in serum EVs and at D7 in urine EVs in patients who developed aGvHD compared to those that remained disease-free, in both an exploratory (serum miR-155 p = 0.02, miR-146a p = 0.06; urine miR-155 p = 0.02, miR-146a p = 0.07) and an independent cohort (serum miR-155 p = 0.01, miR-146a p = 0.02). Conclusions: These results further support a role for miR-155 and miR-146a as non-invasive, clinically relevant biomarkers for aGvHD. However, the link between their involvement in generalized inflammation and in specific pathophysiology requires further investigation at a systemic level.