Discussion

The pathological diagnosis of FTLD encompasses a heterogeneous group of neurodegenerative disorders predominantly affecting the frontal and anterior temporal lobes. While it manifests with an insidious onset and progression in all affected individuals, there are three clinical syndromes: semantic dementia, progressive non-fluent aphasia and bvFTD.3 Semantic dementia and progressive non-fluent aphasia both involve an initial decline in linguistic ability: patients with semantic dementia show deficits in object naming and word comprehension, whereas those with progressive non-fluent aphasia have impaired language production. bvFTD on the other hand is typified by early executive dysfunction and behavioural changes of disinhibition, apathy, loss of empathy or sympathy, motor stereotypies and hyperorality, and generally a fairly symmetrical pattern of FTLD.4 This patient had all of the typical clinical features of bvFTD, although he also developed a language component later.

The presence of behavioural features alone, however, is insufficient for a diagnosis of probable bvFTD. Some patients meet symptomatic criteria but have no evidence of FTLD on imaging and instead follow a benign course, known as ‘behavioural variant FTD phenocopy syndrome’.5 These individuals may have an underlying psychiatric disorder causing behavioural features that mimic bvFTD, sometimes making it difficult to distinguish between the two diagnoses. Indeed, apathy and emotional withdrawal were initially attributed to depression in our patient by his general practitioner. Repetitive and compulsive behaviours, however, would be unusual for major depression. Delusions and euphoria, also sometimes observed in bvFTD, may be wrongly diagnosed as late-onset schizophrenia.6

Given the marked cognitive deficits evident in this case, we also considered other causes of young-onset dementia. For example, an Alzheimer’s pathology can sometimes be distinguished clinically from FTLD by the presence of more prominent deficits in memory, visuospatial tasks and limb praxis, with relatively intact social behaviour and more generalised brain atrophy on imaging.4 However, it is now also recognised that there is a ‘frontal, dysexecutive’ variant of Alzheimer’s disease which can manifest similarly to bvFTD, with frontal atrophy on neuroimaging.7 Up to 40% of patients with FTLD may also develop features of motor neurone disease.8 However, this was not the case in our patient, who did not develop pyramidal or lower motor neurone signs over a 6-year follow-up.

In addition to syndromic classification, FTLD can also be divided into three histopathological subtypes, based on the nature and pattern of abnormal protein deposition within the brain: tau (FTLD-tau), TDP-43 (FTLD-TDP) or FUS (FTLD-FUS).9 In this patient, there was FUS (an RNA/DNA binding protein) pathology within the frontal and temporal cortex on autopsy. This is a rare finding, present in only 5%–10% of patients with FTLD.10 11 It is possible to correlate different pathologies with particular clinical presentations.12 Semantic dementia is almost always associated with TDP-43 proteinopathy, while progressive non-fluent dysphasia can be associated with tau or TDP-43. bvFTD, however, has been linked to all three histopathological subtypes, unless accompanied by motor neurone disease, in which case it is consistently linked with TDP-43 proteinopathy. Thus, it is often difficult to distinguish between pathologies on the basis of clinical features alone.12

Retrospective studies correlating pathology on autopsy with clinical features exhibited in life suggest there might be some features that point to a diagnosis of FTLD-FUS. Typically, bvFTD presents in the sixth decade of life, but in patients with FUS pathology onset is considerably younger, often in the fourth decade.13 Other pathologies are also possible in younger patients though.14 Hyperorality and dietary change in bvFTD are associated with both FUS and tau pathologies, but typically not TDP-43.15

There was no family history of neuropsychiatric illness in our case. Of FTD cases 60% present sporadically (although it is important to bear in mind the possibility of gene mutations in individuals with cryptic family histories).16 The remainder show an autosomal dominant pattern of inheritance, of which 60% are caused by mutations in MAPT, C9orf72 and GRN.17 In recent years, mutations in particular genes, including MAPT and C9orf72, have been correlated with tau and TDP-43 proteinopathies, respectively. However, no mutations have yet been linked with the FTLD-FUS histological subtype (which causes behavioural change without motor neurone signs), including in the FUS gene itself (FUS was not sequenced in our case).18 Instead, FUS mutations, which are always accompanied by FUS protein inclusions, are linked to a specific form of amyotrophic lateral sclerosis (ALS-FUS).19 However, despite shared FUS pathology, cases of ALS-FUS appear to have a different underlying pathogenic mechanism from sporadic FTLD-FUS.

This case is an example of neuropsychiatric and cognitive disturbance in a young patient who was diagnosed with bvFTD with underlying rare FTLD-FUS pathology. Clinical features and investigations pointed to a diagnosis of probable bvFTD. Young onset accompanied by hyperorality and dietary changes can occur with several underlying pathologies, but recent work suggests these are often present in patients with FTLD-FUS pathology.