Background Steroids are commonly used to treat graft versus host disease (GVHD) in allogeneic hematopoietic cell transplant (allo-HCT). Although higher doses of steroids are effective in treating severe GVHD, it may consequently result in an increased risk of infection, negatively impacting treatment-related mortality (TRM). In the post-transplantation cyclophosphamide (PTCy) setting, lower GVHD occurrence was appreciated compared with non-PTCy allo-HCT with some concerns for increased risk of infection. However, few reports discussed the impact of steroid use on PTCy allo-HCT outcomes. Our study investigated the impact of steroids on patient outcomes who received non-PTCy and PTCy allo-HCT. Methods This study included patients who received allo-HCT from September 2017 to December 2021 at our center. Patients who either received cord blood transplant, did not achieve engraftment, or were receiving continuous steroids at stem cell infusion were excluded. Any dose of systemic steroids used was captured and evaluated for 1 year after allo-HCT from medical records. We used semi-landmark analysis to illustrate TRM and overall survival (OS) curves. We considered steroid use as a time-dependent covariate. Results We analyzed 266 patients who underwent PTCy (n=126) and non-PTCy allo-HCT (n=140). Patient characteristics are shown in Figure 1. There was a higher incidence of steroid use in the non-PTCy group than the PTCy group (74% vs. 52%, p<0.01). The median day of steroid initiation was later in the PTCy group than in the non-PTCy group (Day 87 and 59, respectively, p=0.07). The steroid group had a significantly higher incidence of TRM (p<0.01) in PTCy group and inferior OS in both groups (non-PTCy; p=0.05, PTCy; p<0.01) than the non-steroid group (Figure 2A). In multivariate analysis, steroid use was a risk factor of TRM (HR 9.34, p<0.01) and overall mortality (HR 6.84, p<0.01) in the PTCy and non-PTCy groups (data not shown). In PTCy group patients who received steroids, 64% were started ≥1 mg/kg/day. The ≥1mg/kg/day group had a more significant impact on TRM (vs. no-use steroid group: <1mg/kg/day, p<0.05; ≥1mg/kg/day, p<0.01) and OS (vs. no use steroid group: <1mg/kg/day, p=0.6; ≥1mg/kg/day, p<0.01) than the <1mg/kg/day group (Figure 2B). There were 14 transplant-related deaths in the PTCy with steroid use group, including 8 from infections and 1 from GVHD (Figure 3). Conclusion Our results showed that steroid use, especially starting dose ≥1mg/kg/day, was a risk factor for TRM and overall mortality in PTCy patients. In those PTCy patients who used steroids, the cause of death was mainly infection but not GVHD. This may suggest that a lower starting dose of steroid (<1mg/kg/day) with other steroid-sparing agents could be beneficial in mitigating the risk of infections and controlling GVHD in PTCy setting. Further studies are warranted considering the duration of steroid use and GVHD severity.