Risk of event in ALL patients associated with the ATF5 T1562 allele in multivariable analysis
| Variable . | HSJ test group . | DFCI validation group . | ||||
|---|---|---|---|---|---|---|
| HR . | 95% CI . | P . | HR . | 95% CI . | P . | |
| Sex | 1.6 | 0.8-3.1 | .2 | 1.6 | 1.0-3.8 | .06 |
| Age, y | 1.4 | 0.5-3.8 | .5 | 2.1 | 1.2-3.7 | .01 |
| WBCs, × 109/L | 1.5 | 0.5-4.3 | .4 | 1.9 | 0.8-4.7 | .1 |
| Cell type | 0.3 | 0.03-2.0 | .2 | 0.8 | 0.2-2.7 | .7 |
| Risk | 0.6 | 0.3-1.5 | .3 | 0.8 | 0.3-2.1 | .6 |
| Treatment protocol | 1.4 | 1.0-1.9 | .04 | 1.7 | 0.9-3.3 | .1 |
| ATF5 T1562 carrier | 2.3 | 1.1-4.7 | .03 | 2.2 | 1.0-4.9 | .05 |
| Variable . | HSJ test group . | DFCI validation group . | ||||
|---|---|---|---|---|---|---|
| HR . | 95% CI . | P . | HR . | 95% CI . | P . | |
| Sex | 1.6 | 0.8-3.1 | .2 | 1.6 | 1.0-3.8 | .06 |
| Age, y | 1.4 | 0.5-3.8 | .5 | 2.1 | 1.2-3.7 | .01 |
| WBCs, × 109/L | 1.5 | 0.5-4.3 | .4 | 1.9 | 0.8-4.7 | .1 |
| Cell type | 0.3 | 0.03-2.0 | .2 | 0.8 | 0.2-2.7 | .7 |
| Risk | 0.6 | 0.3-1.5 | .3 | 0.8 | 0.3-2.1 | .6 |
| Treatment protocol | 1.4 | 1.0-1.9 | .04 | 1.7 | 0.9-3.3 | .1 |
| ATF5 T1562 carrier | 2.3 | 1.1-4.7 | .03 | 2.2 | 1.0-4.9 | .05 |
The Cox regression model was used to estimate the risk of the ATF5 genotype in the presence of the prognostic factors shown in Table 1. HR and 95% CI are indicated. The analyses were limited to the patients who received E coli asparaginase. There was no difference in prognostic factor distribution between the E coli group and all patients in either test or validation set. For the analysis, the absence of the T1562 allele and categories typically associated with lower risk of relapse were considered as references (age ≥ 1 and < 9.99 years, WBC count < 50 × 109/L, B-cell type, standard risk group, female sex, and most recent treatment protocol).