INTRODUCTION

• Anterior temporal lobe (ATL) degeneration is a hallmark of several primary progressive aphasia syndromes.
• In particular, ATL atrophy is characteristic of frontotemporal lobar degeneration due to TDP-43, type C (FTLD-TDP type C, or TDP-C).
• However, Pick’s disease (PiD), a subtype of FTLD-tau, can also affect the ATL early, along with the inferior frontal gyrus (IFG).
• We examined differences in ATL atrophy between TDP-C and PiD and tested whether combined ATL and IFG thinning could distinguish TDP-C from PiD pathology.

 

METHODS

• Twenty-eight right-handed PPA cases (PiD, N = 11; TDP-C, N = 17) and 35 age-matched controls underwent structural MRI at their initial visit to the Northwestern PPA Program.
• T1-weighted images were acquired on a Siemens 3T scanner and processed with FreeSurfer to extract cortical thickness.
• Two global regions of interest (ROIs) were defined using the Human Connectome Project (HCP) atlas:
• Left anterior temporal lobe (ATL; 9 parcels)
• Inferior frontal gyrus (IFG; 6 parcels)
• Cortical thickness values were converted to z-scores relative to the control mean for each ROI.
• Logistic regression models evaluated whether mean ATL and IFG thickness predicted pathology, first at the global ROI level and then at the sub-ROI (parcel) level.
• A composite measure of ATL and IFG thinning was compared with models including each ROI separately using Akaike Information Criterion (AIC).

 

RESULTS

• ROI-wise analyses revealed a dissociation in regional vulnerability (Figure 2).
• ATL cortical thickness patterns were similar between TDP-C and PiD, with TGd and TGv as the most atrophied regions.
• IFG cortical thickness patterns differed between TDP-C and PiD, with mean values for TDP-C regions almost all above −1.5 SD except for area 47s.
• Logistic regressions showed ATL thinning predicted TDP-C (β = −1.95, p = 0.006), and IFG thinning predicted PiD (β = 1.59, p = 0.018).
• Individual ROI analyses showed:
• ATL thinning in TGd, STSda, STGa, TGv, PeEC, TE2a, STSva, and TE1a predicted TDP-C.
• IFG thinning in FOP4, FOP5, and area 44 predicted PiD (FDR-corrected p < 0.05).
• A model including ATL and IFG as independent predictors provided the best fit and resulted in near-complete separation of the pathology groups.

 

CONCLUSION

• TDP-C and PiD exhibit distinct regional atrophy patterns, with TDP-C showing predominant ATL thinning and PiD greater IFG involvement.
• Despite early ATL degeneration, IFG atrophy alone remained the best predictor of underlying PiD.
• Future studies will assess additional imaging modalities (e.g., white matter tracts) to further characterize ATL vulnerability across these pathologies.