Performance of the multivariate classifier
| ALL subtype . | Sensitivity (%)* . | Specificity (%)† . | Positive predictive value (%)‡ . | Negative predictive value (%)§ . | Accuracy . | |
|---|---|---|---|---|---|---|
| (%)‖ . | P¶ . | |||||
| BCP ALL subtypes | ||||||
| 11q23 (n = 20) | 100 (67-100) | 100 (95-100) | 100 (67-100) | 100 (100-100) | 100 (95-100) | < .001 |
| HeH (n = 104) | 100 (83-100) | 100 (91-100) | 100 (91-100) | 100 (85-100) | 98 (95-100) | N/A |
| t(1;19) (n = 16) | 100 (100-100) | 100 (100-100) | 100 (100-100) | 100 (95-100) | 100 (95-100) | < .001 |
| t(12;21) (n = 73) | 100 (88-100) | 100 (92-100) | 100 (88-100) | 100 (93-100) | 100 (90-100) | < .001 |
| BCP ALL and T-ALL | ||||||
| BCP ALL (n = 341) | 100 (97-100) | 100 (100-100) | 100 (100-100) | 100 (86-100) | 99 (98-100) | N/A |
| T-ALL (n = 60) | 100 (86-100) | 100 (97-100) | 100 (86-100) | 100 (100-100) | 99 (98-100) | < .001 |
| ALL subtype . | Sensitivity (%)* . | Specificity (%)† . | Positive predictive value (%)‡ . | Negative predictive value (%)§ . | Accuracy . | |
|---|---|---|---|---|---|---|
| (%)‖ . | P¶ . | |||||
| BCP ALL subtypes | ||||||
| 11q23 (n = 20) | 100 (67-100) | 100 (95-100) | 100 (67-100) | 100 (100-100) | 100 (95-100) | < .001 |
| HeH (n = 104) | 100 (83-100) | 100 (91-100) | 100 (91-100) | 100 (85-100) | 98 (95-100) | N/A |
| t(1;19) (n = 16) | 100 (100-100) | 100 (100-100) | 100 (100-100) | 100 (95-100) | 100 (95-100) | < .001 |
| t(12;21) (n = 73) | 100 (88-100) | 100 (92-100) | 100 (88-100) | 100 (93-100) | 100 (90-100) | < .001 |
| BCP ALL and T-ALL | ||||||
| BCP ALL (n = 341) | 100 (97-100) | 100 (100-100) | 100 (100-100) | 100 (86-100) | 99 (98-100) | N/A |
| T-ALL (n = 60) | 100 (86-100) | 100 (97-100) | 100 (86-100) | 100 (100-100) | 99 (98-100) | < .001 |
Median values (and 20%-80% percentiles) are given in the table.
TP indicates true positives; FN, false negatives; TN, true negatives; FP, false positives; and N/A, not available (see “Methods”).
(TP)/(TP + FN).
(TN)/(FP + TN).
(TP)/(TP + FP).
(TN)/(TN + FN).
(TP + TN)/(TP + TN + FP + FN).
Probabilities from 1000 permutations.