Conditioning Regimens and Outcomes after Allogeneic Hematopoietic Cell Transplant for Hyperinflammatory Inborn Errors of Immunity

Introduction: Inborn errors of immunity such as hemophagocytic lymphohistiocytosis (HLH) and chronic granulomatous disease (CGD) are characterized by hyperinflammation. Hematopoietic cell transplantation (HCT) in the setting of hyperinflammation leads to high morbidity and mortality. Consequently, there is increasing use of less intense conditioning regimens, which can increase risk of mixed chimerism or graft failure. We sought to study the effect of common regimens on outcomes after HCT using data reported to the Center for International Blood and Marrow Transplant Research.

Methods: 365 patients aged <21 years with HLH (n=263) and CGD (n=102) were transplanted in the US between 2005-2018. Included are recipients of HLA-matched sibling (n=58; 16%) and HLA-matched (n=149; 41%) and mismatched unrelated (n=158; 43%) donor HCT. The analysis considered 3 conditioning regimen intensity groups: 1) fully myeloablative conditioning with busulfan (Bu; median dose 16 mg/kg [IQR 13-17]), cyclophosphamide (Cy) ± anti-thymocyte globulin (ATG) or alemtuzumab, n=142; 2) reduced intensity conditioning consisting of fludarabine (Flu), melphalan (Mel; 140mg/m² [60%], 100 mg/m² [40%]) ± alemtuzumab or ATG, n=131; and 3) reduced toxicity myeloablative conditioning consisting of either Flu, Mel (140mg/m² [75%], 100 mg/m² [25%]), and thiotepa (TT; 8 mg/kg or 10 mg/kg), or Flu, Bu (12mg/kg, IQR 9-15) ± alemtuzumab or ATG, n=92. The cumulative incidence rates of veno-occlusive disease (VOD) and infections were calculated. The probabilities of overall survival and event-free survival were calculated using Kaplan-Meier estimator. For event-free survival, an event was defined as the first occurrence of any of the following: primary graft failure, secondary graft failure, cellular product intervention for mixed chimerism, donor chimerism <5%, second transplant, or death. The Fine and Gray method for acute and chronic GVHD and Cox regression analysis for event-free and overall survival were used to determine factors affecting outcomes.

Results: Patient demographics were similar across the three treatment groups. Patients with HLH were more likely to receive the Flu/Mel regimen. Although unrelated donor HCTs were predominant across the treatment groups, cord blood graft was more common in the Bu/Cy group. Conditioning regimens changed over the study period with most Flu/Mel/TT and Flu/Bu regimens used after 2010. Consequently, outcomes were censored 2-years post-HCT to account for differences in follow-up. The day-100 incidence of VOD was higher with Bu/Cy (18%) compared to Flu/Mel (4%) and Flu/Mel/TT or Flu/Bu (7%) regimens (p<0.001). The 6-month incidence of bacterial infection was higher after Bu/Cy (50%) and Flu/Mel (58%) compared to Flu/Mel/TT or Flu/Bu (43%) regimens (p=0.013). Viral infections were higher in Flu/Mel group (72%) compared to Bu/Cy (44%) and Flu/Mel/TT or Flu/Bu (56%), p<0.001. There were no differences in overall survival (Figure 1A), but event-free survival (Figure 1B) was lowest with the Flu/Mel regimen, after adjusting for donor type (Table 1). Compared to matched sibling, survival was lower with matched (HR 2.41, p=0.05) and mismatched (HR 2.89, p=0.01) unrelated donor HCT. Chronic GVHD but not grade II-IV acute GVHD was lower with Flu/Mel regimen. Table 2 shows the results of multivariate analysis for HLH disorders and findings consistent with the main analysis.

Conclusion: The data does not support the use of a reduced intensity Flu/Mel regimen for hyperinflammatory inborn errors of immunity. Although we did not observe differences in event-free survival between Bu/Cy and Flu/Mel/TT or Flu/Bu regimens, lower incidences of VOD and bacterial infections favor Flu/Mel/TT or Flu/Bu regimens.