Table 2.
Nonshared HLA-C allotype* | Acute GVHD† (N = 453/1861) | Nonrelapse mortality† (N = 709/1727) | Overall mortality† (N = 1246/1975) | Relapse† (N = 501/1727) | ||||
---|---|---|---|---|---|---|---|---|
OR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | |
Patient’s mismatch | 1.34 (1.10-1.62) | .003 | 1.22 (1.06-1.39) | .005 | 1.15 (1.03-1.27) | .009 | 1.03 (0.86-1.22) | .76 |
Donor’s mismatch | 1.07 (0.88-1.30) | .49 | 1.15 (1.01-1.31) | .04 | 1.14 (1.03-1.26) | .01 | 0.97 (0.82-1.16) | .74 |
Sum of mismatched allotypes | 1.16 (1.03-1.32) | .02 | 1.15 (1.06-1.25) | .002 | 1.12 (1.05-1.19) | .001 | 1.00 (0.89-1.12) | .96 |
The level of expression of the patient’s mismatch, the donor’s mismatch, and the sum of these mismatched allotypes were each modeled as a continuous linear variable. ORs and HRs are presented as an increase in risk of failure associated with each increase in expression of 100 fluorescence intensity units.
For the shared (matched) allotype, the OR for acute GVHD was 0.90 (95% CI, 0.71-1.13; P = .36), the HR for nonrelapse mortality was 0.85 (95% CI, 0.72-1.00; P = .05), the HR for overall mortality was 0.94 (95% CI, 0.84-1.07; P = .34), and the HR for relapse was 1.10 (95% CI, 0.90-1.34; P = .34).
Numbers denote the number of patients with failure for each of the 4 end points of all patients with clinical data for the given end point.