The MAGIC algorithm probability is a validated response biomarker of treatment of acute graft-versus-host disease

• Type of Material: Article
• Authors: Hrishikesh K. Srinagesh, Icahn School of Medicine at Mount Sinai; Umut Ozbek, Icahn School of Medicine at Mount Sinai; Urvi Kapoor, Icahn School of Medicine at Mount Sinai; Francis Ayuk, University Medical Center Hamburg-Eppendorf; Mina Aziz, Icahn School of Medicine at Mount Sinai; Kaitlyn Ben-David, Icahn School of Medicine at Mount Sinai; Hannah K. Choe, Ohio State University; Muna Qayed, Emory University; John E. Levine, Icahn School of Medicine at Mount Sinai; James L. M. Ferrara, Icahn School of Medicine at Mount Sinai
• Language: English
• Date: 2019-12-10
• Publisher: American Society of HEMATOLOGY
• Publication Version: Final Published Version
• Copyright Statement: © 2019 by The American Society of Hematology.
• Final Published Version (URL): http://dx.doi.org/10.1182/bloodadvances.2019000791
• Title of Journal or Parent Work: BLOOD ADVANCES
• Volume: 3
• Issue: 23
• Start Page: 4034
• End Page: 4042
• Grant/Funding Information: This work was supported by National Institutes of Health, National Cancer Institute grants (P01CA03942 and P30CA196521) and a National Institutes of Health, National Center for Advancing Translational Sciences grant (TL1 TR001434).
• Keywords: SURVIVAL; Life Sciences & Biomedicine; RISK; Hematology; END-POINT; SCORE; Science & Technology; THERAPY; ACUTE GVHD
• Research Categories: Health Sciences, Medicine and Surgery

Abstract

The Mount Sinai Acute GVHD International Consortium (MAGIC) algorithm probability (MAP), derived from 2 serum biomarkers, measures damage to crypts in the gastrointestinal tract during graft-versus-host disease (GVHD). We hypothesized that changes in MAP after treatment could validate it as a response biomarker. We prospectively collected serum samples and clinical stages of acute GVHD from 615 patients receiving hematopoietic cell transplantation in 20 centers at initiation of first-line systemic treatment and 4 weeks later. We computed MAPs and clinical responses and compared their abilities to predict 6-month nonrelapse mortality (NRM) in the validation cohort (n = 367). After 4 weeks of treatment, MAPs predicted NRM better than the change in clinical symptoms in all patients and identified 2 groups with significantly different NRM in both clinical responders (40% vs 12%, P < .0001) and nonresponders (65% vs 25%, P < .0001). MAPs successfully reclassified patients for NRM risk within every clinical grade of acute GVHD after 4 weeks of treatment. At the beginning of treatment, patients with a low MAP that rose above the threshold of 0.290 after 4 weeks of treatment had a significant increase in NRM, whereas patients with a high MAP at onset that fell below that threshold after treatment had a striking decrease in NRM that translated into clear differences in overall survival. We conclude that a MAP measured before and after treatment of acute GVHD is a response biomarker that predicts long-term outcomes more accurately than change in clinical symptoms. MAPs have the potential to guide therapy for acute GVHD and may function as a useful end point in clinical trials.