The performance of r-PET TLG in distinguishing R/R lymphoma from benign histology for different lymphoma subtypes
. | All cases N = 271 . | Large B-cell lymphoma N = 88 . | Indolent NHL N = 101 . | HL N = 44 . |
---|---|---|---|---|
ROC AUC | 0.798 | 0.877 | 0.722 | 0.730 |
P | <.001 | <.001 | .002 | .002 |
Using optimal cut-off of the whole group: ≥245 | ||||
Sensitivity, specificity | 63%, 86% | 69%, 92% | 64%, 60% | 39%, 100% |
PPV, NPV | 97%, 30% | 94%, 46% | 92%, 24% | 100%, 41% |
Using optimal cut-off of the individual subgroups | ≥220 | ≥470 | ≥160 | |
Sensitivity, specificity | 71%, 92% | 54%, 80% | 52%, 92% | |
PPV, NPV | 92%, 44% | 94%, 22% | 94%, 43% |
. | All cases N = 271 . | Large B-cell lymphoma N = 88 . | Indolent NHL N = 101 . | HL N = 44 . |
---|---|---|---|---|
ROC AUC | 0.798 | 0.877 | 0.722 | 0.730 |
P | <.001 | <.001 | .002 | .002 |
Using optimal cut-off of the whole group: ≥245 | ||||
Sensitivity, specificity | 63%, 86% | 69%, 92% | 64%, 60% | 39%, 100% |
PPV, NPV | 97%, 30% | 94%, 46% | 92%, 24% | 100%, 41% |
Using optimal cut-off of the individual subgroups | ≥220 | ≥470 | ≥160 | |
Sensitivity, specificity | 71%, 92% | 54%, 80% | 52%, 92% | |
PPV, NPV | 92%, 44% | 94%, 22% | 94%, 43% |