Background The Mount Sinai Acute GVHD International Consortium (MAGIC) algorithm uses two biomarkers, ST2 and REG3α, measured on day 7 following allogenic stem cell transplant to predict aGVHD and 6-month NRM. This algorithm defines two risk groups denoted Ann Arbor high risk and low risk. The MAGIC algorithm has not been independently validated in the pediatric allogeneic stem cell transplant population. Transplant associated thrombotic microangiopathy (TA-TMA) is also associated with elevated ST2 levels and previous studies have shown TA-TMA is associated with steroid-resistant aGVHD. Incidence of TA-TMA was not evaluated in the previous adult studies using the 2-biomarker algorithm. Methods Biorepository blood samples were collected from 106 consecutive pediatric patients on day 7 following allogenic stem cell transplant between April 2014 and June 2017. ST2 and REG3α were measured via ELISA. The previously described 2-biomarker algorithm was performed on each patient to determine high-risk versus low-risk score for each patient. Outcomes examined included 6-month NRM, aGVHD and TA-TMA. Results Of the 106 patients, 18 patients were categorized as high risk and 88 low risk based on the 2-biomarker algorithm. Six-month NRM occurred in 5/18 high risk patients and 4/88 low risk patients (Fig. 1A). TA-TMA occurred in 7/18 high risk patients and 13/88 low risk patients (Fig. 1B). The 2-biomarker algorithm was not predictive of aGVHD with 5/18 high risk patients and 32/88 low risk patients having aGVHD (Fig 1C). There was no difference in mean REG3α levels in our patients with and without TA-TMA, aGVHD and 6-month NRM (Table 1). Interpretation The MAGIC algorithm does not predict aGVHD in our pediatric population, but rather TA-TMA and 6-month NRM. The algorithm is driven by ST2 levels as there was no significant difference in REG3α levels by outcome. Pediatric specific biomarker algorithms for prediction of aGVHD will need further delineated through future research. One potential explanation for this difference from adult studies is important differences in the population studied. Our population includes a high number of children with hemophagocytic lymphohistiocytosis leading to pre-existing endothelial injury.