Risk of hematologic relapse among patients on the higher-risk arm, based on combined GSTM1 nonnull and TYMS genotypes, by multivariate analysis
Feature* . | Coefficient . | SE . | Hazard ratio (95% CI) . | P† . |
|---|---|---|---|---|
| Combined GSTM1 and TYMS (GSTM1 nonnull + TYMS 3/3 vs others) | 2.67 | 0.905 | 14.5 (2.46-85.2) | .0031 |
| Leukocyte count (fewer than 100 × 109/L vs at least 100 × 109/L) | -1.95 | 0.964 | 0.14 (0.02-0.94) | .043 |
| t(9;22)/BCR-ABL (present vs absent) | 0.38 | 0.906 | 1.47 (0.25-8.66) | .67 |
| t(4;11)/MLL-AF4 (present vs absent) | 3.13 | 0.985 | 22.8 (1.20-157) | .0015 |
| Day 19 marrow (at least 5% blasts vs less than 5% blasts) | 1.07 | 0.551 | 2.92 (0.99-8.61) | .052 |
| Minimal residual disease on remission date (at least 0.01% vs less than 0.01%) | 2.75 | 1.00 | 15.7 (2.20-112) | .0060 |
Feature* . | Coefficient . | SE . | Hazard ratio (95% CI) . | P† . |
|---|---|---|---|---|
| Combined GSTM1 and TYMS (GSTM1 nonnull + TYMS 3/3 vs others) | 2.67 | 0.905 | 14.5 (2.46-85.2) | .0031 |
| Leukocyte count (fewer than 100 × 109/L vs at least 100 × 109/L) | -1.95 | 0.964 | 0.14 (0.02-0.94) | .043 |
| t(9;22)/BCR-ABL (present vs absent) | 0.38 | 0.906 | 1.47 (0.25-8.66) | .67 |
| t(4;11)/MLL-AF4 (present vs absent) | 3.13 | 0.985 | 22.8 (1.20-157) | .0015 |
| Day 19 marrow (at least 5% blasts vs less than 5% blasts) | 1.07 | 0.551 | 2.92 (0.99-8.61) | .052 |
| Minimal residual disease on remission date (at least 0.01% vs less than 0.01%) | 2.75 | 1.00 | 15.7 (2.20-112) | .0060 |
This analysis was performed in the 64 patients with all features evaluable; because all patients in the higher-risk group who relapsed had an unfavorable DNA index, it was not included as a potential predictive feature.
SE indicates standard error for the coefficient estimate; CI, confidence interval.
Hazard of relapse for the first feature compared to those with the second feature.
P value was determined by Fine and Gray estimator.