Our objective was to evaluate the CSF phospho-Tau181/total-Tau (p/t-Tau) ratio to distinguish between the two main forms of frontotemporal lobar degeneration (FTLD): FTLD with TDP-43 (FTLD-TDP) and FTLD with Tau inclusions (FTLD-Tau). CSF p/t-Tau ratio was examined in 79 FTLD patients with predictable neuropathology, i.e. Tau (affected by progressive supranuclear palsy or carriers of mutations within the MAPT gene) or TDP-43 (carriers of mutations within granulin, C9orf72, TARDBP genes or affected by FTD with motor neuron disease). FTLD patients were randomly grouped in a training cohort (n = 39) to assess the best CSF p/t-Tau cut-off score according to ROC analysis, and a validation cohort (n = 40) to evaluate accuracy values of the identified marker. Results showed that, in the training cohort, we found a significantly reduced CSF p/t-Tau ratio in FTLD-TDP relative to FTLD-Tau. ROC analysis for p/t-Tau ratio was 0.873 and cut-off score of 0.136 allowed to differentiate FTLD-TDP and FTLD-Tau with 81.8% sensitivity and 88.2% specificity, respectively. Analysis in the validation cohort showed CSF p/t-Tau ratio < 0.136 to distinguish FTLD-TDP from FTLD-Tau with 83.3% specificity and 63.6% sensitivity, respectively. The positive predictive value of detecting TDP neuropathology was 82.4%. In conclusion, a reduced CSF p/t-Tau ratio represents a viable biomarker to correctly identify TDP pathology in FTLD.
Keywords: Amyloid beta-peptides/cerebrospinal fluid; biological markers/cerebrospinal fluid; frontotemporal lobar degeneration/cerebrospinal fluid; peptide fragments/cerebrospinal fluid; tau proteins/cerebrospinal fluid.