Semantic dementia: linking loss of brain and brawn

The term frontotemporal dementia (FTD) refers to a closely related group of conditions characterized by slowly progressive selective deficits of behavioural, verbal, and/or motor functions, sparing other higher neural activities. The two best characterized syndromes are (i) behavioural disturbances that correspond to the pattern recognized by clinicians as frontal dementia, and (ii) a non-fluent progressive aphasia. Pathological examination shows neuronal death restricted to specific areas, leading to severe atrophy mostly restricted to frontal and temporal regions, contrasting with well preserved neighbouring gyri. In almost all cases, affected neurons display peculiar inclusion bodies of a finite number of varieties, the morphology and staining properties of which dictate pathological classification of FTD. Approximately 40–50% of FTD cases express the microtubule associated protein tau in the inclusion bodies (tau+), and are categorized as Pick's disease, corticobasal degeneration (CBD), progressive supranuclear palsy (PSP) or neurofibrillary tangle dementia. Most of the remaining 50–60% of cases (tau−) have Inclusions, which are Tau and Synuclein Negative, but intensely labelled by antibodies to Ubiquitin, known as ITSNU, ubiquitin-only inclusions or most commonly motor neuron disease-type inclusions, with the acronym FTD–MND used for the disease (Lowe and Rossor 2003;Rosso et al., 2003). The term frontotemporal lobar degeneration is used in rare tau− cases where no inclusion bodies are found. The field has been cursed with devilish terminology, characterized by the use of a single term to denote both the entire group and one of its varieties or a clinical syndrome and a pathological substrate. This ambiguous terminology has probably contributed to the muddy thinking characterized by inappropriate mingling of symptoms and inclusions. The early concept that a particular clinical variant of FTD was inextricably linked to a single pathological substrate was based on the limited syndromic expression afforded by small numbers of cases, as poignantly illustrated in the nosological history of CBD. This was gradually abandoned as it became clear that any of the clinical syndromes could be attributed to any one of the pathological substrates. However, the probability of syndrome–substrate association differs for each variety. The history of semantic dementia can be traced to 1975, when Elizabeth Warrington utilized the new concepts of episodic memory (relating to events) and semantic memory (general and conceptual knowledge) to describe three patients with selective impairment of the latter (Warrington, 1975). The initial focus was on psychological rather than neurological aspects, specifically on understanding the organization of meanings systems in the brain. In some patients, the discrepancy between the impairment of knowledge for living things in comparison with that of inanimate objects was particularly striking. Semantic dementia, the progressive disease these patients suffer, was later recognized as a part of the spectrum of FTD. Although some of the early consensus conferences on FTD listed semantic dementia as a separate entity because of its scarce behavioural disturbances and fluent language (Neary et al., 1998), there has been a recent tendency to consider semantic dementia as simply a subset of primary progressive aphasia, as reflected in the report of the Work Group on frontotemporal dementia and Pick's disease (McKhann et al., 2001). Clinicians confronted with the history of a patient repeatedly asking a bewildered relative, ‘what is a … ?’, will naturally attribute the problem to deficits in the processing of language. Patients who are visual artists or possess sketching skills provide an insight into the true nature of their deficits. The accurate graphic representation they convey of any object or person they are allowed to visualize contrasts dramatically with the nonsensical drawings of everyday items they are asked to imagine: trucks move on legs, and cats sport wings (Fig. 1). What these patients lack is the knowledge of the intrinsic properties of things, e.g. that wooden chairs will burn, float and are unlikely to be located on top of cars. In contrast, their mental maps, ability to remember recent events, computing and planning capabilities are unimpaired. Their conversation appears normal until probed. Some of their abnormal behaviours represent manifestations of the underlying cognitive deficit. Thus, the parent offering liqueurs among the refreshments at a children's birthday party, the cook who tries to make chicken marmalade or the hostess who includes tomatoes in the fruit bowl act strangely because the salient, defining features of the individual items are not accessible to them. However, the drawing tasks described above clearly reveal perseveration, and patients are unable to generate new simple four bar designs or lists of words; signs shared with the behavioural presentation of FTD. But until now, the pathological substrate of semantic dementia remained to be established. Drawings at command of a cat (left, chat) and a chicken (right, poule) by a gifted francophone artist with semantic dementia. Notice the mingling of features of mammals and birds, with loss of differentiating characteristics. Confirming inklings from previously published reports of a few cases, the paper presented by Davies et al. (2005) in this issue utilizes a sufficiently ample series to determine that the most common pathological substrate (13 of 18 autopsies) of semantic dementia is FTD–MND, or as the authors aptly suggest motor neuron disease inclusion dementia (MNDID). Clinicopathological studies reveal that FTD–MND patients present with behavioural disturbances in 4 of 5 cases and with primary progressive aphasia in only 1 of 5 cases. In contrast, behavioural abnormalities and language deficits are equally represented among the tau+ varieties of FTD. Moreover, tau+ pathologies, particularly Pick's disease, are the most common substrate of primary progressive aphasia (Hodges et al., 2004; Kertesz et al., 2005). Devising a satisfactory classification scheme for FTD is predicated on correlating the clinical syndrome with the underlying pathological substrate. If one can make inferences regarding the clinical classification of an FTD subtype based on its pathological substrate, the findings in this study indicate that semantic dementia is not related to primary progressive aphasia, but instead to progressive frontal dementia (the most common underlying pathology of which is FTD–MND). The paper makes another important contribution by demonstrating ubiquitin-only inclusions restricted to the inferior olivary nucleus, a region where they are not usually sought, in two cases, one of them with severe degeneration of corticospinal tracts. This probably represents the coup de grace for the lubberly frontotemporal lobar degeneration, also known as dementia lacking distinct histopathological features, an entity rapidly disappearing as the skilled use of ubiquitin antibodies spreads in diagnostic laboratories. The hypothesis may be proffered that MNDID simply reflects an alternative CNS localization of the same pathological process responsible for amyotrophic lateral sclerosis. The remaining five cases in this series were divided into classical Pick's disease (three cases) and Alzheimer's disease (two cases, one with additional white matter pathology), certainly in keeping with the idea of multiple pathological substrates underlying a particular clinical syndrome. It is likely that future research supported by pathological and genetic studies will be able to split the syndrome of semantic dementia into subgroups matching the different substrates. In a parallel field, recent seminal work has been able to delineate three subtypes of primary progressive aphasia with different verbal, anatomic and genetic features, recognizing one variety—logopenic progressive aphasia—with a high apolipoprotein E epsilon4 gene frequency suggestive of a link to Alzheimer's disease (Gorno-Tempini et al., 2004). The Davies et al. study represents a valuable contribution to our understanding of the pathological processes underlying semantic dementia, revealing a possible link between semantic dementia and motor neuron disease. If correct, advances in diagnostic biomarkers and treatment of this more common disorder will promptly benefit the majority of patients with semantic dementia and the behavioural variant of FTD. Although these are exciting prospects, we should not overlook the unique opportunity patients with semantic dementia offer to study the organization of knowledge in the human brain.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.