. Distinct Subtypes of Behavioral Variant Frontotemporal Dementia Based on Patterns of Network Degeneration. JAMA Neurol. 2016 Sep 1;73(9):1078-88. PubMed.


Please login to recommend the paper.


  1. This is an interesting study using principal component analysis (PCA) on 18 regions of interest data derived from MRI scans of 90 patients meeting FTD Consortium consensus criteria for bvFTD. From this analysis four clusters were derived, each conforming to a particular anatomical pattern of atrophy. A differential involvement of salience and semantic appraisal networks in the patterns of atrophy formed the basis for characterizing these subgroups. However, given that the bvFTD group comprised patients with a multiplicity of underlying disorders, ranging from pure cortical degenerations like Pick’s disease to cortico-subcortical degenerations such as corticobasal degeneration, it is not altogether surprising that different atrophy patterns would emerge from PCA.

    However, what would have been really interesting, and important, about the study would have come from the neuropathological and genetic correlations, had any of the four subgroups predicted underlying histology or associated with any one genetic mutation. In this way, underlying disease mechanics might have been predicted in bvFTD in life with obvious therapeutic value. Instead, mutations such as C9ORF72 expansions were spread across all four subgroups, and tau and TDP-43 histologies were similarly represented within each subgroup. While it is clear that other forms of FTLD such as semantic dementia, bvFTD with motor neurone disease, and primary progressive non-fluent aphasia show tight histological or genetic linkages, this unfortunately is not the case for bvFTD, and histological predictions for this particular FTLD subgroup, employing “biomarker analysis,” remain elusive. Unfortunately, this much-needed added value to the study was not forthcoming.

    View all comments by David M.A. Mann
  2. This paper is carefully modest in its claims, as one would expect from this very expert group of clinical researchers who understand well the complexities of FTD. The identification of a significant subset (36 percent) of bvFTD patients with primarily subcortical atrophy is a potentially important finding that needs replication. Not only may it be important for prognostication (slow progression), it might also help stratify patients for trials of putative disease-modifying agents. We may try this approach on our Lewy body dementia patients who also show considerable phenotypic variation, which we believe to be at least in part due to different neuroanatomic lesion patterns.

    View all comments by Ian McKeith

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Brain Atrophy Patterns Support Distinct Subtypes of Frontotemporal Dementia