PRESENILE- VS LATE-ONSET BEHAVIORAL FRONTO-TEMPORAL LOBAR DEGENERATION (BVFTLD): CLINICAL, RADIOLOGICAL, AND BIOLOGICAL COMPARISON

Abstract

Late-onset behavioral fronto-temporal lobar degeneration (bvFTLD, >65 years) are clinically and radiologically different from presenile-onset cases (Baborie, Arch Neurol. 2012). BvFTLD diagnostic criteria show low sensitivity for late-onset cases, and Alzheimer’s disease (AD) is the main misdiagnosis. Cerebrospinal fluid (CSF) biomarkers could help in this differential diagnosis (Marelli, DADM 2015), although co-pathology cannot be excluded in older subjects. At present, no data are available about CSF biomarkers in late- vs presenile-onset bvFTLD. A retrospective study (2007-2015) on 41 subjects with possible or probable bvFTLD (Rascovsky, Brain 2011); 17/41 (41%) had onset after and 24/41 (59%) before 65 years. We compared clinical, radiological, and CSF data at the initial valuation. Late-onset bvFTLD (mean age at onset: 70±3 years; probable bvFTLD=59%) and presenile-onset bvFTLD (mean age at onset 59±5 years; probable bvFTLD=71%) had comparable mean disease duration at initial examination (3±3 years) and mean follow up duration (5±3 vs 5±4 years). MRI examination confirmed more frequent hippocampal atrophy (47% vs 21%) and less lobar atrophy (18% vs 58%) in late-onset bvFTLD; this was clinically correlated to more frequent hippocampal memory deficit (53% vs 12.5%). TEP-FDG or SPECT-HMPAO imaging detected focal hypometabolism/hypoperfusion in 53% vs 62.5% of the patients. Unexpectedly, no differences were found in CSF abeta1-42 (1097±276 vs 1034±236), T-tau (289±214 vs 231±100), P-tau (40±17 vs 37±15), and IATI index (2±0.6 vs 2±0.6); the ratio T-tau/abeta1-42 (0.26±0.16 vs 0.23±0.10) and P-tau/abeta1-42 (0.04±0.01 vs 0.04±0.01) were not different, either. In each group three subjects had high T-tau or P-Tau and one subject low abeta1-42. Late-onset bvFTLD is not rare in clinical practice; we confirmed more hippocampal and global atrophy compared to presenile-onset bvFTLD, clinically correlated to more frequent hippocampal memory loss. TEP-FDG and SPECT-HMPAO focal alterations were equally represented. Since no differences were found in CSF biomarkers, the differences between late- and presenile-onset bvFTLD could not be explained by co-existing AD pathology in older patients. They could instead be due to hippocampal sclerosis, known to be associated to FTLD. These data suggest the potential interest of CSF and metabolic markers in the differential diagnosis of late-onset bv-FTLD vs AD.