T/N and T/C Mismatch
Tau-Neurodegeneration/Cognition Mismatch Framework
The tau-neurodegeneration (T-N) mismatch framework, pioneered by researchers at the University of Pennsylvania including Sandhitsu Das, Xueying Lyu, and David Wolk, was developed to address the profound clinical and structural heterogeneity seen in Alzheimer’s disease (AD). While tau neurofibrillary tangles are long established as a primary driver of downstream neurodegeneration, individuals frequently exhibit varying degrees of brain atrophy that deviate significantly from what their focal tau burden would predict. To quantify this variability, the UPenn team introduced a T-N mismatch metric calculated from the regional residuals of a normative model mapping in vivo tau burden (measured via PET) to cortical thickness (measured via MRI) (Das et al., 2021). By clustering these continuous deviances, they identified distinct patient phenotypes characterized by either “higher-than-expected” or “lower-than-expected” neurodegeneration. Crucially, these spatial T-N mismatch patterns have been shown to reflect the latent influence of non-AD co-pathologies—such as TDP-43 or vascular white matter disease—as well as individual resilience and age-related vulnerability factors, rather than tau toxicity alone (Lyu et al., 2025).
Building directly upon these structural findings, the team (including Chris Brown and Wolk) recently extended the paradigm to evaluate tau-clinical (or tau-cognition) mismatch in order to better capture individualized cognitive trajectories (Brown et al., 2025). By comparing a patient’s actual clinical performance against the expected trajectory for their specific tau burden (using both PET and highly accessible plasma p-tau217 biomarkers), the researchers categorized individuals into canonical, “resilient” (better cognition than expected for their tau levels), and “vulnerable” (worse cognition than expected) groups. They demonstrated that the vulnerable cohort experienced much earlier, faster cognitive decline and harbored significantly higher rates of mixed pathologies, including an α-synuclein and TDP-43 imaging signature. Collectively, these multi-modal mismatch frameworks provide a powerful, scalable approach to determining patient resilience, isolating the effects of common co-pathologies, and fundamentally improving individual prognosis in the era of disease-modifying therapies.
