Systemic Steroid Treatment of Grade I Acute GVHD Increases Steroid-Refractory GVHD and NRM

In a collaborative study from the Mount Sinai-led MAGIC consortium study, with collaborators from major transplant and hematology centers in the US, Germany, Italy, Japan, and Thailand, the authors show that treating grade I acute graft-versus-host disease (GVHD) with up-front systemic corticosteroids may cause more harm than benefit. Although early systemic steroids reduced progression to grade II–IV GVHD, they did not reduce severe grade III–IV GVHD, chronic GVHD, need for second-line treatment, or improve overall survival. Instead, up-front steroids were associated with more steroid-refractory GVHD, infectious deaths, and nonrelapse mortality (NRM), supporting current recommendations to reserve systemic steroids for grade II–IV disease and manage low-risk grade I GVHD with topical therapy or observation.

Despite consensus guidance against systemic treatment for grade I GVHD, many clinicians use systemic steroids early to prevent progression to severe or steroid-refractory disease. To test whether this real-world practice improves outcomes, the authors retrospectively analyzed 1,143 pediatric and adult patients with grade I acute GVHD from 24 MAGIC transplant centers in North America, Europe, and Asia. Patients underwent first allogeneic hematopoietic cell transplantation between 2014 and 2022 and were compared according to treatment strategy: up-front systemic steroids before progression versus watch-and-wait/topical therapy until progression. Serum suppressor of tumorigenicity 2 (ST2) and regenerating islet-derived protein 3α (REG3α) were also measured in available samples to calculate MAGIC/Ann Arbor biomarker risk.

Of 1,143 patients, 591 (51.7%) received up-front systemic steroids and 552 (48.3%) followed watch-and-wait. Only 184/552 (33.3%) in the watch-and-wait group later required systemic steroids, meaning two-thirds avoided systemic exposure. Six-month steroid-refractory GVHD was 19.5% with up-front steroids versus 6.2% with watch-and-wait (P<0.001). One-year infectious death was 4.7% vs 1.3% (P=0.001), and NRM was 12.6% vs 6.9% (P=0.002). In multivariable analysis, up-front steroids remained associated with higher NRM (HR 1.63; 95% CI 1.06–2.50; P=0.026). These findings place grade I acute GVHD within a broader precision-treatment framework, showing that biomarker-supported restraint may be safer than pre-emptive immunosuppression for most clinically low-risk patients.

Reference:

Akahoshi Y, Katsivelos N, Louloudis IE, et al. Systemic steroid treatment of grade I acute GVHD increases steroid-refractory GVHD and NRM. Blood Immunol Cell Ther. 2026;2(2):100044. http:doi.org/10.1016/j.bict.2026.100044