Risk factors associated with outcomes after NIH chronic GVHD
| . | Hazard ratio (95% CI) . | P . |
|---|---|---|
| Nonrelapse mortality* | ||
| Platelet count less than 100 000/μL at onset | 2.38 (1.6-3.5) | < .001 |
| Prednisone dose category at onset, per group | 1.74 (1.4-2.2) | < .001 |
| Total serum bilirubin concentration more than 2 mg/dL at onset | 2.10 (1.4-3.2) | .001 |
| Patient age, per 10 years | 1.33 (1.2-1.5) | < .001 |
| Acute GVHD before day 100, vs grades 0-I | .008 | |
| Grade II | 1.19 (0.6-2.2) | |
| Grades III-IV | 2.15 (1.1-4.2) | |
| Antecedent late acute GVHD | 2.45 (1.4-4.2) | .002 |
| No. of HLA-mismatched loci, per locus | 1.31 (1.1-1.6) | .01 |
| Female donor for male recipient | 1.49 (1.0-2.2) | .04 |
| Discontinued systemic treatment† | ||
| Antecedent late acute GVHD | 0.35 (0.2-0.6) | < .001 |
| G-CSF–mobilized blood as source of stem cells | 0.57 (0.4-0.8) | .001 |
| Total serum bilirubin concentration more than 2 mg/dL at onset | 0.62 (0.4-1.0) | .04 |
| Prednisone dose category at onset, per group | 0.63 (0.5-0.8) | .001 |
| Female donor for male recipient | 0.66 (0.5-0.9) | .006 |
| No. of sites involved at onset (except skin) | 0.67 (0.6-0.8) | < .001 |
| Donor age, per 10 years | 1.13 (1.0-1.3) | .02 |
| No. of HLA-mismatched loci, per locus | 0.84 (0.7-1.0) | .04 |
| . | Hazard ratio (95% CI) . | P . |
|---|---|---|
| Nonrelapse mortality* | ||
| Platelet count less than 100 000/μL at onset | 2.38 (1.6-3.5) | < .001 |
| Prednisone dose category at onset, per group | 1.74 (1.4-2.2) | < .001 |
| Total serum bilirubin concentration more than 2 mg/dL at onset | 2.10 (1.4-3.2) | .001 |
| Patient age, per 10 years | 1.33 (1.2-1.5) | < .001 |
| Acute GVHD before day 100, vs grades 0-I | .008 | |
| Grade II | 1.19 (0.6-2.2) | |
| Grades III-IV | 2.15 (1.1-4.2) | |
| Antecedent late acute GVHD | 2.45 (1.4-4.2) | .002 |
| No. of HLA-mismatched loci, per locus | 1.31 (1.1-1.6) | .01 |
| Female donor for male recipient | 1.49 (1.0-2.2) | .04 |
| Discontinued systemic treatment† | ||
| Antecedent late acute GVHD | 0.35 (0.2-0.6) | < .001 |
| G-CSF–mobilized blood as source of stem cells | 0.57 (0.4-0.8) | .001 |
| Total serum bilirubin concentration more than 2 mg/dL at onset | 0.62 (0.4-1.0) | .04 |
| Prednisone dose category at onset, per group | 0.63 (0.5-0.8) | .001 |
| Female donor for male recipient | 0.66 (0.5-0.9) | .006 |
| No. of sites involved at onset (except skin) | 0.67 (0.6-0.8) | < .001 |
| Donor age, per 10 years | 1.13 (1.0-1.3) | .02 |
| No. of HLA-mismatched loci, per locus | 0.84 (0.7-1.0) | .04 |
In this model, recurrent malignancy was treated as a competing risk. Follow-up was censored for patients who had recurrent malignancy before death. P values are derived from the likelihood ratio test.
Hazard ratios less than 1.0 indicate a lower rate of discontinued treatment, resulting in prolonged systemic treatment in the presence of the indicated risk factor compared with its absence. In this model, death and recurrent malignancy were treated as competing risks. Follow-up was censored for patients who died or had recurrent malignancy before systemic treatment was discontinued. P values are derived from the likelihood ratio test.