Acta Neuropathol 2005 Nov 13;110(5):501-12. Epub 2005 Oct 13.
Clinical Neuroscience Research Group, University of Manchester, Greater Manchester Neurosciences Centre, Hope Hospital, Stott Lane, Salford M6 8HD, UK.
We have investigated the pathological correlates of dementia in the brains from a consecutive series of 70 patients dying with a clinical diagnosis of frontotemporal lobar degeneration (FTLD). Clinical misdiagnosis rate was low with only 3 patients (4%) failing to show pathological changes consistent with this diagnosis; 1 patient had Alzheimer's disease and 2 had cerebrovascular disease (CVD). In the remaining 67 patients, the most common underlying histological cause was ubiquitin pathology with 24 (36%) cases so affected. In these, ubiquitin-positive inclusions were present in the cerebral cortex as small, rounded or crescent-shaped structures within the cytoplasm of neurones of layer II, together with coiled or curvilinear bodies within neurites, and in the hippocampus as small, solid and more spherical-shaped inclusion bodies within the cytoplasm of dentate gyrus granule cells. In one patient, "cat's eye" or "lentiform" intranuclear ubiquitin inclusions were also present. The second most common histological type was dementia lacking distinctive histology (DLDH), in which neither tau nor ubiquitin inclusions were present, with 16 cases (24%) being affected. Pick-type histology was seen in 14 cases (21%) and tau histological changes associated with frontotemporal dementia (FTD) linked to chromosome 17 (FTDP-17) were present in 11 cases (16%). One case (1%) showed an unusual tau pathology that could not be allocated to any of the other tau groups. Only 1 case (1%) had neuronal intermediate filament inclusion dementia. No cases with ubiquitinated, valosin-containing protein-immunoreactive intranuclear inclusion bodies of the type seen in inclusion body myopathy with Paget's disease of bone and frontotemporal dementia were seen. Clinicopathological correlation showed that any of these histological subtypes can be associated with FTD. However, for FTD with motor neurone disease (FTD+MND), semantic dementia or primary progressive aphasia (PA), the histological profile was either ubiquitin type or DLDH type; Pick-type histology was seen in only 1 case of PA. None of these latter three clinical subtypes was associated with a mutation in tau gene and FTDP-17 type of tau pathology. All cases of progressive apraxia were associated with Pick-type histology. Present data therefore indicate that, although ubiquitin pathology is the most common histological form associated with FTLD, this pathology is not tightly linked with, nor is pathologically diagnostic for, any particular clinical form of the disease, including FTD+MND.