Redefining the multidimensional clinical phenotypes of frontotemporal lobar degeneration syndromes

The syndromes caused by frontotemporal lobar degeneration (FTLD) have highly heterogenous and overlapping clinical features. There has been great progress in the refinement of clinical diagnostic criteria in the last decade, but we propose that a better understanding of aetiology, pathophysiology and symptomatic treatments can arise from a transdiagnostic approach to clinical phenotype and brain morphometry. In a cross-sectional epidemiological study, we examined 310 patients with a syndrome likely to be caused by frontotemporal lobar degeneration, including behavioural variant frontotemporal dementia (bvFTD), the non-fluent (nfvPPA), semantic (svPPA) variants of primary progressive aphasia, progressive supranuclear palsy (PSP) and corticobasal syndrome (CBS). We also included patients with logopenic primary progressive aphasia (lvPPA) and those who met criteria for PPA but not one of the three subtypes. To date, forty-nine patients have a neuropathological diagnosis. A principal component analysis identified symptom dimensions that broadly recapitulated the core features of the main clinical syndromes. However, the subject-specific scores on these dimensions showed considerable overlap across the diagnostic groups. Sixty-two percent of participants had phenotypic features that met the diagnostic criteria for more than one syndrome. Behavioural disturbance was prevalent in all groups. Forty-four percent of patients with CBS had PSP-like features and thirty percent of patients with PSP had CBS-like features. Many patients with PSP and CBS had language impairments consistent with nfvPPA while patients with bvFTD often had semantic impairments. Using multivariate source-based morphometry on a subset of patients (n=133), we identified patterns of co-varying brain atrophy that were represented across the diagnostic groups. Canonical correlation analysis of clinical and imaging components found three key brain-behaviour relationships that revealed a continuous spectrum across the cohort rather than discrete diagnostic entities. In the forty-six patients with longitudinal follow up (mean 3.6 years) syndromic overlap increased with time. Together, these results show that syndromes associated with FTLD do not form discrete mutually exclusive categories from their clinical features or structural brain changes, but instead exist in a multidimensional spectrum. Patients often manifest diagnostic features of multiple disorders and deficits in behaviour, movement and language domains are not confined to specific diagnostic groups. It is important to recognise individual differences in clinical phenotype, both for clinical management and to understand pathogenic mechanisms. We suggest that the adoption of a transdiagnostic approach to the spectrum of FTLD syndromes provides a useful framework with which to understand disease progression, heterogeneity and treatment.


Neuropathological validation of the study cohort
The forty-nine patients with a neuropathological diagnosis had a similar age (t (309) =-0.7 p=0.48) and gender (X 2 =0.52 p=0.47) distribution compared to whole study population but on average were assessed later in their illness (t (309) =2.67, p=0.008). Twenty-one patients had a full dataset of clinical phenotype, neuropathological diagnosis and imaging. Of the 49 cases with post-mortem data, all patients with a clinical diagnosis of PSP had PSP pathology (n=14). Most patients with svPPA had FTLD-TDP-43 Type C (n=3) but one had Pick's disease. bvFTD was associated with FTLD-tau (Picks' n=1, PSP n=1) or TDP43 (n=5). Three patients with bvFTD had concurrent signs of motor neurone disease during the course of their illness, all with TDP43 pathology. Patients with CBS (n=19) had mainly CBD (n=6), Alzheimer's disease (n=8), multiple system atrophy pathology (n=2, despite normal autonomic function tests), FTLD-TDP43 (n=1) or PSP pathology (n=1). One CBS patient had micrometastatic renal cell carcinoma and paraneoplastic cerebellar degeneration. This patient had been treated for renal cell carcinoma but had no evidence of metastatic disease for over 3 years until a lung metastasis was diagnosed shortly before he died. He had no detectable serum autoimmune or paraneoplastic antibodies during the investigation of his corticobasal syndrome, and ataxia was not a prominent clinical feature despite the cerebellar degeneration. Of the four patients with nfvPPA, three had FTLD-tau and one had Alzheimer's disease. The one patient with lvPPA had Alzheimer's Disease pathology.
The number of patients with a neuropathological diagnosis was too small to test the classification accuracy of the syndrome dimensions or imaging components. This may be possible in future, as more patients in the study cohort undergo neuropathological assessment. We have included descriptive plots of the distribution of neuropathology in the syndrome dimensions, imaging components and canonical correlations. Any conclusions are tentative, due to the low numbers. Further research with larger sample size and cross validated classifier models is required to test the predictive accuracy of the syndrome dimensions for pathology.
Two of the syndrome dimensions have good clinicopathological associations (Section 5). Positive scores on syndrome dimension three was associated with FTLD-tau/PSP pathology. The majority of these patients had a clinical diagnosis of PSP, but one patient with bvFTD and a high score on syndrome domain 3 had PSP pathology. Interestingly, some patients with a clinical diagnosis of CBS or nfvPPA had high scores on domain 3 but CBD pathology at post-mortem. This PSP-like clinical phenotype associated with CBD pathology is well recognised (Alexander et al. 2013). All patients with Alzheimer's disease pathology had positive scores on syndrome dimension 5.
Volume loss in the brainstem was associated with FTLD-tau/PSP pathology. Basal ganglia atrophy (imaging component 13 in appendix 6) was associated with FTLD-TDP43 or FTLD-tau but not AD. No other imaging components were associated with specific neuropathology (Section 6). For example, all FTLD pathological subtypes were associated with volume loss in the frontal lobe (components 1 and 2). Parietal atrophy, reflected by low scores on imaging component 8, is typically associated with Alzheimer's disease pathology. Both cases of Alzheimer's disease had low scores, but so did some cases with FTLD-tau and FTLD-TDP43. Basal ganglia atrophy (imaging component 13 in appendix 6) was associated with FTLD-TDP43 and FTLD-tau but not AD.
Finally, we looked at the distribution of neuropathology within the three canonical correlation components, which show the multivariate relationships between clinical and imaging features (Section 7). In the first canonical correlation, positive scores were associated with FTLD-tau-PSP and negative scores with FTLD-TDP43 or Alzheimer's disease. Positive scores on the second canonical correlation were associated with FTLD-TDP43, and the three cases with AD pathology all had negative scores. The third canonical correlation did not clearly separate pathological subtypes.