Effect of donor HTLV-I serostatus on transplantation outcomes
| Outcome . | Univariable analysis . | Multivariable analysis . | |||
|---|---|---|---|---|---|
| Number* . | Hazard ratio (95% CI) . | P . | Hazard ratio (95% CI) . | P . | |
| Overall survival† | |||||
| Donor HTLV-I antibody positive | 43/68 | 1.00 | Reference | 1.00 | Reference |
| Donor HTLV-I antibody negative | 52/88 | 0.90 (0.60-1.35) | .603 | 0.83 (0.54-1.28) | .395 |
| Treatment-related mortality‡ | |||||
| Donor HTLV-I antibody positive | 20/64 | 1.00 | Reference | ||
| Donor HTLV-I antibody negative | 37/86 | 1.51 (0.88-2.58) | .133 | ||
| Disease-associated mortality§ | |||||
| Donor HTLV-I antibody positive | 19/64 | 1.00 | Reference | 1.00 | Reference |
| Donor HTLV-I antibody negative | 13/86 | 0.44 (0.22-0.89) | .022 | 0.43 (0.21-0.90) | .026 |
| Outcome . | Univariable analysis . | Multivariable analysis . | |||
|---|---|---|---|---|---|
| Number* . | Hazard ratio (95% CI) . | P . | Hazard ratio (95% CI) . | P . | |
| Overall survival† | |||||
| Donor HTLV-I antibody positive | 43/68 | 1.00 | Reference | 1.00 | Reference |
| Donor HTLV-I antibody negative | 52/88 | 0.90 (0.60-1.35) | .603 | 0.83 (0.54-1.28) | .395 |
| Treatment-related mortality‡ | |||||
| Donor HTLV-I antibody positive | 20/64 | 1.00 | Reference | ||
| Donor HTLV-I antibody negative | 37/86 | 1.51 (0.88-2.58) | .133 | ||
| Disease-associated mortality§ | |||||
| Donor HTLV-I antibody positive | 19/64 | 1.00 | Reference | 1.00 | Reference |
| Donor HTLV-I antibody negative | 13/86 | 0.44 (0.22-0.89) | .022 | 0.43 (0.21-0.90) | .026 |
CI indicates confidence interval; and HTLV, human T-cell leukemia virus.
Number of events/number of evaluable patients.
Other variables considered in the multivariable analysis were disease status before transplantation, type of GVHD prophylaxis, and type of graft source. Variables significantly associated with overall survival were disease status before transplantation and type of GVHD prophylaxis: not in complete remission versus complete remission (hazard ratio, 1.95; 95% CI, 1.17-3.24, P = .010); tacrolimus- versus cyclosporine-based (hazard ratio, 4.22; 95% CI, 1.58-11.26, P = .004).
Multivariable analysis was not performed because no variable was significantly associated with treatment-related mortality by univariable analysis.
Other variables considered in the multivariable analysis were disease status before transplantation, type of GVHD prophylaxis, and type of graft source. The only variable significantly associated with disease-associated mortality was disease status before transplantation: not in complete remission versus complete remission (hazard ratio, 2.88; 95% CI, 1.01-8.24, P = .049).