Anxiety appears early in the course of Alzheimer's disease. The study team asked whether anxiety in older adults without cognitive impairment is associated with patterns of AD-related neurodegeneration. Investigators analyzed 1,775 cognitively unimpaired (CU) older adults who had 3T MRI and amyloid-beta PET; anxiety was assessed with the STAI-6. They found that greater atrophy within an established Alzheimer's disease (AD) atrophy network was associated with higher anxiety. The relationship persisted after adjusting for depression, global cognition, overall cortical atrophy, and amyloid burden, according to the abstract. This suggests anxiety relates to neurodegenerative patterns beyond comorbid mood symptoms or amyloid status.
The authors used Atrophy Network Mapping (ANM) to translate individual cortical atrophy into a functional-circuit signature. ANM projects each subject's atrophy map onto a normative functional connectome (the source reports N=1000 for that connectome). From those projections, investigators derived a distributed anxiety network tied to the measured STAI-6 scores. The abstract provides this high-level description; detailed ANM parameters, preprocessing steps, and regional weighting schemes were not provided in the source text.
The derived anxiety network converged spatially with the previously defined AD atrophy network. The study team reported a reported correlation of r=0.74 (p=0.04) between the anxiety network and the AD network. They also compared the anxiety network to a causal anxiety network obtained from lesion and neuromodulation datasets; the reported correlation there was r=0.72 (p=0.02). These reported correlations indicate substantial spatial similarity across independently derived circuit maps. The authors observed that this convergence was present not only in the CU cohort but also across the AD continuum and in a Parkinson's disease sample included for generalizability.
Investigators tested whether the identified network was specific to anxiety rather than indexing depressive symptoms. The abstract states the convergence was specific to anxiety compared with depression. The team further examined generalizability by testing independent cohorts from AD and Parkinson's disease (total n=878 in those cohorts, as reported). The anxiety network's presence across those samples suggests the circuitry relationship may extend beyond classical AD populations, though the source does not enumerate cohort-level demographics or detailed statistical models for those secondary samples.
The authors reported that greater atrophy within the anxiety network was associated with subsequent cognitive decline over a reported mean follow-up of 4.6 years. This longitudinal association links baseline anxiety-related circuit atrophy with later cognitive outcomes in their dataset. The abstract does not present effect sizes, model covariates beyond those listed above, or the specific cognitive metrics used for the longitudinal analyses; those full-method details are not included in the provided source text.
This work is presented as a preprint and has not undergone peer review. The abstract-level source provides key findings but omits numerous methodological specifics that readers and clinicians will need to evaluate clinical translation: for example, the source did not report comprehensive participant demographics, complete ANM processing parameters, full model covariates in each analysis, or exact cognitive endpoints used in the longitudinal models. The authors noted cross-sample replication and specificity checks, yet replication in fully described, peer-reviewed datasets will be necessary. Remaining clinical questions include whether anxiety severity or tempo predicts conversion to symptomatic AD, how these circuit findings interact with tau pathology, and whether interventions targeting the identified circuit alter clinical trajectories—questions that the source did not report.
Competing interests and funding were disclosed in the preprint. Several authors report relationships with industry and research support; the work was supported in part by NIMH training funding and device or stimulation-related organizations listed in the source. Because this is an unrefereed preprint, clinicians should interpret the reported associations cautiously and await peer-reviewed replication and detailed methodological reporting before changing clinical practice.