Proportional hazards general linear model analysis on the leukemic relapse risk
| . | All patients . | P . | Patients surviving free of relapse at day 100 . | P . | Patients with grades II to IV acute GVHD . | P . | Patients with grades II to IV acute GVHD surviving free of relapse at day 100 . | P . |
|---|---|---|---|---|---|---|---|---|
| No. of patients | 266 | — | 238 | — | 187 | — | 168 | — |
| Covariates* | ||||||||
| Disease stage† | 1.6 (1.2-2.1) | < .0005 | 1.5 (1.1-2.0) | < .009 | 1.6 (1.1-2.2) | < .008 | — | NS |
| Acute GVHD† | — | NS | — | NS | — | — | — | — |
| Chronic GVHD† | 0.2 (0.1-0.4) | < .0001 | 0.3 (0.1-0.5) | < .0001 | 0.2 (0.1-0.3) | < .0001 | 0.2 (0.1-0.4) | < .0001 |
| HCMV pp65-antigenemia† | 0.2 (0.1-0.4) | < .0001 | 0.2 (0.1-0.6) | < .0001 | 0.1 (0.04-0.4) | < .0002 | 0.2 (0.1-0.5) | < .002 |
| . | All patients . | P . | Patients surviving free of relapse at day 100 . | P . | Patients with grades II to IV acute GVHD . | P . | Patients with grades II to IV acute GVHD surviving free of relapse at day 100 . | P . |
|---|---|---|---|---|---|---|---|---|
| No. of patients | 266 | — | 238 | — | 187 | — | 168 | — |
| Covariates* | ||||||||
| Disease stage† | 1.6 (1.2-2.1) | < .0005 | 1.5 (1.1-2.0) | < .009 | 1.6 (1.1-2.2) | < .008 | — | NS |
| Acute GVHD† | — | NS | — | NS | — | — | — | — |
| Chronic GVHD† | 0.2 (0.1-0.4) | < .0001 | 0.3 (0.1-0.5) | < .0001 | 0.2 (0.1-0.3) | < .0001 | 0.2 (0.1-0.4) | < .0001 |
| HCMV pp65-antigenemia† | 0.2 (0.1-0.4) | < .0001 | 0.2 (0.1-0.6) | < .0001 | 0.1 (0.04-0.4) | < .0002 | 0.2 (0.1-0.5) | < .002 |
— indicates not applicable; and NS, not significant (P ≥ .01).
Proportional hazards general linear models with forward and backward selection of covariates; time-dependent covariates: time intervals to grades II to IV acute GVHD, chronic GVHD, and HCMV pp65-antigenemia; and further covariates included in all PHGLM analyses: stratified patient age; patient/donor sex match; pretransplantation risk score; donor type; graft source; pretransplantation HCMV donor and recipient serology; and disease genetic group.
Hazard ratio (HR) and 95% CI after adjustment for all significant (P < .01) covariates in the final models.