In Alzheimer’s disease (AD), several functional connectivity (FC) networks have been characterized, particularly altered FC of the posterior cingulate cortex (PCC) that reflect neurodegenerative brain change and increased risk for AD, as inferred from increased cortical amyloid-beta (Aß) accumulation¹. Moreover, recently published data suggest a close relationship between increased brain iron levels, cerebral Aß-load and risk for AD in the elderly^2,4. While FC may reflect compensatory mechanisms active in a context of aging and neurodegenerative brain change³, there is sparse knowledge regarding the respective impacts of iron and Aß related FC on maintenance of cognitive performance in the elderly. Thus, the aim of the current study was to investigate cortical Aß- and cortical iron related functional connectivity in a population of high-aged individuals with normal cognitive-performance ("super-agers") for regional extent and overlap by simultaneous assessment of iron (susceptibility-MRI), Aß-burden (18F-Flutemetamol-PET) and FC of the PCC (fMRI at rest).A study population of 25 healthy high-aged participants ("super-agers": 5 female, 20 male; mean age= 87.6 years, SD= 2.97), was administered an 8 minute 42 second fMRI resting-state scan using a gradient echo EPI sequence (scan plane: axial, slice thickness: 3.6mm, voxel size: 3x3x3mm³, TR: 2.547s, Phase FOV: 1; 3 Tesla SIGNA General Electrics Healthcare combined PET-MR instrument). Functional images were then preprocessed (realigned, slice-time corrected, coregistered, segmented, normalised, smoothed and band-pass filtered for 0.008 Hz and 0.09 Hz) using SPM12 and the CONN-toolbox⁵. Cortical iron load was inferred from quantitative susceptibility mapping (QSM) - MRI, as reconstructed from a 3D multi-echo GRE sequence (TR/TE/ΔTE=40/3/4ms, voxel size=1x1x1mm³, flip angle=15) using the echoes with echo time between 15 and 27ms. Sequentially Laplacian phase unwrapping, V-SHARP for background removal⁶ and an LSQR based approach for dipole inversion⁷ were used to create the QSM image. Cortical Aβ-plaque density was estimated by PET acquisition of 18F-Flutametamol⁸ (85-105 minutes post injection) and reconstructed using time-of-flight reconstruction (voxel size=1.2x1.2x2.78mm³). The PET image was segmented using the atlas created from the T1-weighted image and all PET-values were then referenced to the cerebellar gray matter for calculation of standardized uptake value ratios (SUVR)⁹. Mean QSM measures of the cerebral cortex were used to generate a "high cortical iron" and a "low cortical iron" group by median split of QSM-measures. For mean cortical 18F-Flutametamol-SUVRs a distinct two-peak distribution resulted, allowing for six participants to be categorized as "high cortical Aβ" and 19 as "low cortical Aβ". The CONN-toolbox was used to define distinct functional connectivity patterns of the PCC associated with "high cortical iron" and "high cortical Aβ", respectively (extent threshold FDR-corrected for p<0.05). The "high cortical iron"-network comprised 20112 voxels within 2 clusters; the "high cortical Aβ"-network comprised 10361 voxels within 7 clusters (Figure 1). There was a significant overlap (7568 voxels) between the networks associated with "high cortical iron" and "high cortical Aβ": Brain regions with most pronounced overlap included (voxel-counts indicated for: high iron; high Aβ): Precuneus Cortex (3086; 2672), left Lateral Occipital Cortex superior (2080; 1070), right Lateral Occipital Cortex superior (1396; 377), Angular Gyrus (3086; 544), right Middle Temporal Gyrus posterior (616; 551) right Middle Temporal Gyrus anterior (218; 176) and the right Temporal Pole (291; 83), which constitute important regions of the DMN¹⁰. Linear relationships between mean BOLD fMRI timecourses were calculated from ROIs defined by the respective networks, and a significant degree of correlation resulted, as indicated by adjusted R²=0.59. However, the contrast "high cortical iron" versus "high cortical Aβ", yielded significant differences for a ROI of 3077 voxels, including the planum temporale, precuneus cortex, parahippocampal gyrus, hippocampus and the angular gyrus (extent threshold FDR-corrected for p<0.05), which include DMN connections that may be affected by Aß-burden even at preclinical stages¹¹.Our data demonstrate that ageing-related brain-pathology, as indicated by increased cerebral iron and Aß-plaque load is associated with distinct patterns of functional connectivity in cognitively normal super-agers. While our data demonstrate both strong regional overlap and temporal synchronicity of connectivity patterns present in the “high iron” and “high Aß” groups, there nevertheless appear to be differences regarding recruitment of temporal and hippocampal areas that are heavily involved in memory function and have altered FC to the precuneus in early or asymptomatic stages of AD¹¹. The departures from network synchronicity could illustrate different early stages of pathogenesis or compensatory mechanisms. Additional studies are necessary to further characterize the role of these networks for age related brain change, and their potential use as biomarkers in clinical trials of early pathology-specific therapeutic intervention.