Table III. AUC values, sensitivities, specificities, and cutoff points of the best discriminating parameters for PD versus healthy or diseased controls.
| Parameter | Training set |
Validation set |
|||||||
|---|---|---|---|---|---|---|---|---|---|
| AUC | p | cutoff | Sens% | Spec% | AUC | p | Sens% | Spec% | |
| SPP1 | |||||||||
| PD vs Con | 0.791 | 1E-4 | 4.362 | 90.0 | 56.7 | 0.821 | 8E-7 | 77.5 | 75.0 |
| PD vs AD | 0.948 | 1E-11 | 87.5 | 97.4 | |||||
| LRP1 | |||||||||
| PD vs Con | 0.706 | 0.006 | −2.218 | 70.0 | 70.0 | 0.698 | 0.002 | 80.0 | 57.5 |
| PD vs AD | 0.662 | 0.014 | 70.0 | 63.2 | |||||
| SPP1, LRP1, CSF1R, EPHA4, and TIMP1 | |||||||||
| PD vs Con | 0.873 | 7E-7 | 0.5394 | 76.7 | 80.0 | 0.853 | 5E-8 | 82.5 | 82.5 |
| PD vs AD | 0.990 | 1E-13 | 95.0 | 97.4 | |||||
| 11-peptide panela | |||||||||
| PD vs Con | 0.982 | 1E-10 | 0.6761 | 90.0 | 96.7 | 0.932 | 3E-11 | 85.0 | 92.5 |
| PD vs AD | 1.000 | 3E-14 | 100.0 | 100.0 | |||||
a APLP1, APOB, CP, CSF1R, EPHA4, GPR37, LRP1, SERPINC1-FAT, SERPINC1-TSD, SPP1, and TIMP1. Note that the 11-peptide data needs to be interpreted with caution due to potential overfitting. AD, Alzheimer disease; AUC: area under curve; Con, healthy control; PD, Parkinson disease; Sens: sensitivity; Spec: specificity.