Study Design
For this retrospective, registry study, we gathered data from the Florida Pediatric Bone Marrow Transplant and Cell Therapy Consortium (FPBCC). The objectives of FPBCC are to identify pediatric HCT best practices and improve survival of children receiving HCT in Florida. FBPCC participation involves monthly virtual meetings, participation in quality improvement projects, sharing data for retrospective trials, and developing consortium-wide prospective clinical trials. FPBCC was founded in 2018 and comprises 5 of the 6 pediatric transplant programs in the state of Florida, USA. All FPBCC participating centers signed memoranda of understanding and data use agreements and obtained institution-specific IRB approvals for this retrospective data analysis. All patients have signed the informed consent for data reporting. Transplant centers report detailed data on consecutive hematopoietic cell transplantations to the statistical center of the Center for International Blood and Marrow Transplant Research (CIBMTR), and compliance with this reporting is monitored by on-site audits. Data from participating centers were downloaded from the enhanced data back to center platform of CIBMTR and forwarded to the FPBCC statistical center at the University of Florida (Gainesville, FL, USA), where data were combined into a single set and analyzed.
We describe characteristics of pediatric HCT recipients, including gender, age, race, performance score, diagnosis, type of transplant, and number of pre-transplant comorbidities. Transplant characteristics consist of type of donor, stem cell source, Human Leukocyte Antigen (HLA) match, conditioning regimen, and regimen intensity. Outcomes data include length of survival, cause of death, and incidence and grade of GVHD. Our data set included information on timing and use of second transplant; however, reasoning (e.g. early or late graft rejection) for second transplant was not available and the dataset did not contain details of donor lymphocyte infusions or details of GVHD prophylaxis other than the use of serotherapy. Thus patients who required the 2nd or subsequent transplant (N=21) and those who rejected the transplant but did not undergo a subsequent transplant (N=3) were not included in this manuscript. Patients for whom the center did not provide information on GVHD (N=3) were considered not-evaluable for GVHD and excluded as well. Myeloablative regimens were defined as those using one or more of the following: total body irradiation >500 cGy for a single dose or >800 cGy for fractionate, busulfan >7.2 mg/kg iv, melphalan >150 mg/m2, or thiotepa ≥10 mg/kg. Reduced intensity regimens were defined as those using melphalan ≤150 mg/m2, thiotepa <10 mg/kg, TBI >200 cGy and ≤500 cGy as a single dose or ≤800 cGy fractionated, busulfan ≤7.2 mg/kg. Non-myeloablative regimens used any dose of ATG, fludarabine, cyclophosphamide, or TBI≤200 cGy. (7) Fully HLA-matched unrelated bone marrow (BM) or peripheral blood (PB) donors were matched at A, B, C, and DRB1 antigens by high-resolution typing. For umbilical cord blood (UCB) donors, full match was defined as 6/6 HLA-antigen match (A, B, DRB1) by either low or high resolution. A related donor was considered haploidentical if ≥ 2 different antigens (A, B, C, or DRB1) were mismatched. CIBMTR gathers and reports the highest grade of acute GVHD by one year post transplant following criteria published by Przepiorka et al. (8) Although currently CIBMTR gathers data on individual organ involvement with cGVHD based on NIH Consensus Criteria, 2014, the data available through our platform contained only information on extent of cGVHD by one year post transplant.
cGVHD was reported as limited or extensive based on definition by Shulman et al. (9) Limited cGVHD includes only localized skin involvement and/or liver dysfunction, while any other organ involvement is considered extensive.