Multivariate models for associations between HLA-DPB1 matching status according to allele or functional matching in unrelated donor hematopoietic cell transplantation
| Outcome . | HLA-DPB1 . | |||||
|---|---|---|---|---|---|---|
| Allele* . | Functional TCE3* . | Functional SNP proxy† . | ||||
| Match (n = 1216)‡ . | P value . | NP (n = 1654)‡ . | P value . | High (n = 481)‡ . | P value . | |
| OS | 0.96 (0.87-1.06) | .40 | 1.15 (1.05-1.25) | .002 | 1.13 (0.95-1.35) | .16 |
| aGVHD grades 3-4 | 0.84 (0.69-1.03) | .09 | 1.31 (1.11-1.54) | .002 | 1.50 (1.12-2.01) | .007 |
| Relapse | 1.34 (1.17-1.54) | <.0001 | 0.89 (0.77-1.02) | .10 | 0.89 (0.68-1.17) | .40 |
| Outcome . | HLA-DPB1 . | |||||
|---|---|---|---|---|---|---|
| Allele* . | Functional TCE3* . | Functional SNP proxy† . | ||||
| Match (n = 1216)‡ . | P value . | NP (n = 1654)‡ . | P value . | High (n = 481)‡ . | P value . | |
| OS | 0.96 (0.87-1.06) | .40 | 1.15 (1.05-1.25) | .002 | 1.13 (0.95-1.35) | .16 |
| aGVHD grades 3-4 | 0.84 (0.69-1.03) | .09 | 1.31 (1.11-1.54) | .002 | 1.50 (1.12-2.01) | .007 |
| Relapse | 1.34 (1.17-1.54) | <.0001 | 0.89 (0.77-1.02) | .10 | 0.89 (0.68-1.17) | .40 |
Bold type indicates statistical significance.
aGVHD, acute graft-versus-host disease; NP, nonpermissively HLA-DPB1–mismatched transplantations; OS, overall survival; SNP, single-nucleotide polymorphism; TCE3, T-cell epitope 3-group model.
Data are from Fleischhauer et al.12 Baseline is the 8/8 TCE3 permissively HLA-DPB1–mismatched transplantations (n = 2539) against the HLA-DPB1 allele–matched (Match) transplantations or against the TCE3 nonpermissively HLA-DPB1–mismatched transplantations. Models were adjusted for disease severity, patient age, patient/donor sex and cytomegalovirus status, hematopoietic stem cell source, use of T-cell depletion, year of transplantation, conditioning regimen, and donor registry (Japanese Registry vs others).
Data are from Petersdorf et al.13 Baseline is the 8/8 transplants with a single HLA-DPB1 mismatch in the graft-versus-host direction, where the mismatched allele in the donor carried the SNP proxy A and the mismatched allele in the recipient also carried the SNP proxy A (n = 413), against the same type of pairing but where the mismatched allele in the recipient carried the SNP proxy G. Models were adjusted for disease severity, patient age, patient/donor sex and cytomegalovirus status, hematopoietic stem cell source, use of T-cell depletion, year of transplantation, and conditioning regimen.
Shown are the relative risk values followed by the 95% confidence intervals in parentheses.