Development of a sample bank and clinical database for the retrospective analysis of unrelated bone marrow transplants : a pilot study of 138 transplants using RSCA for high resolution HLA matching
The Anthony Nolan Register provides donors for allogeneic stem cell transplantation for haematological disorders. A clinical database and sample bank were established for the ongoing analysis of transplants from Register donors. Clinical data and blood samples were collected from donors and patients transplanted in the UK from 1996 onwards. DNA was extracted from all samples received, and used for the detection of HLA mismatches at six loci (HLA-A, -B, -C, -DRB1, -DQB1, and -DPB1) by RSCA, and HA-1 by SSP. 138 donor/patient pairs were selected for a pilot study. HLA matching between patient and donor, along with the diagnosis of the patient, donor and patient age, CMV status and gender, and the T cell depletion status of the transplant were analysed for their effect on transplant outcome. The outcome variables studied included overall survival, disease free survival, transplant related mortality, relapse incidence, and the occurrence of acute and chronic GvHD. Patients with AML had decreased overall and disease free survival, and patients with CML had increased risk of relapse. Transplant from a same sex donor increased the risk of death or relapse, and male patients receiving stem cells from female donors showed reduced overall survival. Developing acute GvHD increased the risk of transplant related mortality. A mismatch at HLA class I and at class II was associated with unfavourable outcome for all survival variables. The inclusion of HLA-DPB1 matching did not alter the effects seen when other mismatches were present. Matching at the allele level for all loci gave an increased risk of relapse compared with patients mismatched with their donor for HLA-DPB1 only, indicating that a mismatch for HLA-DPB1 alone may protect from relapse. The data from this study was used to calculate the numbers required for a definitive study, for the optimisation of donor selection from the Register.